ED 336 904 AUTHOR Berendes, Heinz W., Ed.; and Others TITLE Advances in the Prevention of Low Birthweight. Proceedings from an I

ED 336 904 EC 300 636

AUTHOR Berendes, Heinz W., Ed.; And Others TITLE Advances in the Prevention of Low Birthweight. Proceedings from an International Symposium (Cape Cod, Massachusetts, May 8-11, 1988). INSTITUTION National Center for Education in Maternal and Child Health, Waseington, DC. SPONS AGENCY Health Resources and Services Administration (DHHS/PHS), Rockville, MD. Bureau of Maternal and Child Health and Resources Development.; National Inst. of Child Health and Human Levelopment (NTH), Bethesda, Md. PUB DATE 91 CONTRACT MCU-117007 NOTE 247p. AVAILABLE FROMNational Maternal and Child Health Clearinghouse (NMCHC), 38th and R Streets, N.W., Washington, DC 20037 (single copies free). PUB TYPE Collected Works - Conference Proceedings (021)

EDRS PRICE MFC1/PC10 Plus Postage. DESCRIPTORS *Birth Weight; *Congenital Impairments; Etiology; Infants; Medical Researcn; Neonates; Nutrition; Perinatal Influences; Pregnancy; *Premature Infants; *Prenatal Influences; *Prevention; *Preventive Medicine; Smoking; Social Support Groups; Stress Variables

ABSTRACT This proceedings document contains papers addressing trends, determinants, and interventions in preventing low birthweight. Papers have the following titles and authors: "Trends in Rates of Low Birthweight in the United States" (Mary McCormick); "Evolution of the Preterm Birth Rate in France" (Gerard Breart); "The Etiology and Prevention of Low Birthweight: Current Knowledge and Priorities for Future Research" (Michael Yramer); "Support and Stress during Pregnancy: What Do They Tell Us about Low Birthweight?" (Jeanne Brooks-Gunn); "Results of a Three-Year Prospective Controlled Randomized Trial of Preterm Birth Prevention at the University of Pittsburgh" (Eberhard Mueller-Heubach); "Decrease and Rise in Rates of Prete= Deliveries: Haguenau Prenatal Study, 1971-1988" (Emile Papiernik et al.); "The West Los Angeles Prematurity Prevention Project: A Progress Report" (Calvin Nobel et al.); "The South Carolina Multicentered Randomized Controlled Trial To Reduce Low Birthweig1it,9 (Henry Heins et al.); "A Prematurity Prevention Project in Northwest North Carolina" (Paul Meis et al.); "The Family Workers Project: Evaluation of a Randomized Controlled Trial of a Pregnancy Social Support Service" (Brenda Spencer); "The Social Support and Pregnancy Outcome Study' (Ann Oakley and Lynda Rajan); "Prevention of Preterm Deliveries by Home Visiting System: Results of a French Randomized Controlled Trial" (Beatrice Blondel et al.); "Smoking Interventions during Pregnancy" (Mary Sexton); "Cervical Cerclage: New Evidence from the Medical Research Council/Royal College of Obstetricians and Gynecologists" (Adrian Grant); "Prevention of Intrauterine Growth Retardation with Antiplatelet Therapy" (Serge Uzan); "Does Calcium Supplementation Reduce Pregnancy-Induced Hypertension and Prematurity?" (Jose Villar); "Magnesium Supplementation in Pregnancy: A Double-Blind Study" (Ludwig Spatling); "An Overview of Trials of Social Support during Pregnancy" (Diana Elbourne and Ann Oakley); "Inhibition of Preterm Labor: Is It Worthwhile?" (Marc Keirse). Appendixes provide the symposium program and list of participants. (DB) U.S. DIAARTINEP4Tor EDUCATION Offtoit of Mahon* Ramtarchand liftgonxemott EDUCA11ONAL RESOURCES INFORMATION CENTER {ERIC) 0 his 000urnitot Disn 1110todoced recfnvirCt from the poison oforganaatton onsinating it hirnof change!, have Plum MO tounpfOr reproductson ovaIlty

Pomts of veiny or opinions tatio ri Ms document do not necessinlir reortient attic* OE RI poalt;on or policy

Advana- in the Prewntion qf Low Birthuvight

An Intenzational Sympasil tin

Proceedings

BEST COPY AVAILABLE Advances in the Prevention of Low Birthweight Advances in the Praention of Low Birthtteight Syinpasium May 8-1 I, 1988 Chatham Bais Cape Cod, Massachusetts

Proceedings

Heinz U. Berendes. .11.11.S.. Sanuwl kessel. .11.1.1.. .11.1).11., Sumner }affe..11.1,1., editors

.spfmsiireit by the

Bureau (il. Maternal and Child Health mut 14)sources 1.1erelopnu,nt lealtb Resources WU/ SeiTiCeS Ad1111111.tilnitiO1l

eind the National Institute qiChild Health curd Human Developuwnt Natimud Institutes (y.1 lealtb

(y' the

1'nitecl Slates Public Sen'ice Mpartunent (!l. 1 learn) an(/ Human 4C1Th'es

Pmducol by

,Vaticinal Centern- Education in .11aternal and (Aid Health tl'ilshingt( in. 1).C.

e". 4 Cite as Berendes, H., Kessel, S., and Yaffe, S. (1991). Advances in the Prevention of Low Birthweight: International Svmposium, Washington, D.C.: Natkmal Center for Education in Maternal and Child Health.

Advances in the Prevention of Low Birthweight An International Symposium is not copyright- ed. Readers are free to duplicate and use all or part of the informatkm contained in this publication. In accordance with accepted publishing standards. NCEMCII requests acknowledgment, in print, of any informatkm repa)duced in another publication.

Library of Congress Catalog Card No,: 91-60905

The National Center for Education in Maternal and Child I lealth (NCEMCI I) provides information services, educational materials, and technical assistance to organizations. agencies, and individuals with maternal and child health interests. The Center was established in 1982 at Georgetown I Tniversity, within the Department of Obstetrics and Gynecology. NCEMCH is funded primarily by the LI.S. Department of Health and Human Services, through its Maternal and Child Health Bureau (formerly the Office of Maternal and Child Health. Bureau of Maternal and Child I lealth and Resources Development),

Piimhied and published by: National Center for Education in :Maternal and Child lit:Ali (NCEMCI 38th and R Streets. NW Washington, DC 200i7 (202) 625-8-400

Sil447le copies this publication are available at no cost from: National Nlaternal and Child Ilealth Clearinghouse (NMCHC) 38th and R Streets, N\X' 'Washington. DC 200-S7 (202) 62i-8410 or (703) 821-89Si x 25.4

This publication has been produced by the National Centerfor &location in Maternal and Child Health under its cooperatity agreement (MCI'- I 170071 with the Maternal and Child Health Bureau. Health Resources and Services Administration. Public Health Service, IS. Department 411ailth and Human SerliaN. 5 Table of Contents

Acknowledgments xiii Foreword XV

Section I: 'Rends `Frends in Rates of IJ)w Birthweight in the lnited States .1Iary McCormick, M.D.,Se.D., Boston, Massachnssells 3 Evolution of the Preterm Binh Rate in France Gerard Tht;art, France 13

Section II: Determinants The Etiology and Prevention of Low Birthweight: Current knowledge and Prioritiesfcg Future Research Michael Knimer M.D.. Montreal, Canada )5 Support and Stress During Pregnancy: What Do They Tell l's About LowBirthweight? Jeanne Bmoks-Gunn. PhD.. Princeton. .Ven'ioNey

Section III: hzterventions Results of a Three-Year Pnrspective Controlled Randomized Trial of Preterm Birth Prevention at the ITliVQrsitV of Pittsburgh LheriPard Muellerikulmcb, Piltsbitugh. Pow.9'franiel 61 Decrease and Rise in Rates of Preterm Deliveries: I laguenau Prenatal Study, 19-1-1985 Emile papiemik. (..t al., Paris. I,rance 69 The West Los Angeles Prematurity Prevention Project: A Progress Report Ilobel. M.D.. el al., Los Angeles. California 79 The South Carolina Multicentered Randomized Controlled Trial to Reduce LowBirthweight Hem). LleinslLD., et al.. (..harleston. Swab Carolina 1. A Prematurity Prevention Project in Northwest NorthCarolina Paul Meis. MD., et al., Winston-Salem. NorthOtrolina 101 The Family Wcykers Project: Evaluation ofa Randomized Controlled Trial of a Pregnancy Social Support Service Brenda Spencer PhD., Manchester. !niter! Kingdom 109 The Social Support and Pregnancy Outcome Study Ann Oakley awl lynda Rafitn, London. I 'tilled Kingdom Prevention of Preterm Deliveries by I lome VisitingSystem: Results of a French Randomized Controlled Trial Beatrice Blonde!. M.D.. et al.. Paris. France Smoking Interventions During Pregnancy Akin' Sexton, PhD.. M.P.H., Baltimore. Maryland 153 Cervical Cerclage: New Eviclence From the MedicalResearch CounciVRoyal College of Obstetricians and Gynecologists Adrian Gmnt. 046rd, ()tiled Kingdom 107 Prevention of Intrauterine Growth Retardation with AntiplateletTherapy Serge t`zan. MD.. Paris. France 1-- Does Calcium Supplementation Reduce Pregnancy-InducedI lypertension and Prematurity? Jose Villar. MD.. Bethesda. Mar.vland 187 Vagnesium Supplementation in Pregnancy: A Double-lilindStudy Ludwig Vatling, Herne. West Gernutily 197 An Overview of Trials of Social Support During Pregnancy Diana Elhourne. PhD.. a:id Ann Oakkr. 0.0)rd. l'uitedKingdom kl3 Inhibition of Preterm Labor: Is It Worthwhile% Marc Keirse. Leiden. Netherlawls

Appoulices Symposium Pr()gra in List of Participants '05

10.1 Tables and Figures

Trends in Rates of Low Birthweight in the I Tnited Statc...s (Figure 1.1) Figure 1.1: Proponion of LBW and \IBA Births (By Race)

Statistical Findings: Evolution of Preterm Birth in France (Figure 2.1: Tables 2.1-2.9) rgure 2.1: :Maternity l'nits Invoked in the Swdv 20 Table 2,1: Design of the National Surveys 19 'table2.2:Evolution of Gestational Age tl\ati(Whil Data) 19 Table 2.3: Ev()Iuticm ()t. Birthwcight (Natim;t1 Data} 19 Table Ev olution

Statistical Findings: The Elio! )gy and Prevention of Low Birthweight (Figures 3.1-3.3: Table 3.1) Figure 3.1: Relati)c Importance of Fstahlished Factors with 1)frect Caus,..d Impacts on II 'GR in a Rural Deveh)ping Country with Endemic Malaria ...... Figure 3.2:Relative linponamvt Fstabhshed Factors with 1)irect hnpacts on II "GR in a Developed Cc nintry Where 10 Percent of Women Smoke DuringPregnancy Figure $.3: Rehtive lmpciitthce c)t. Established Factms with Direct Causal Impacts on Pronaturity it) De%clopecf Country Where 10 Percent of Women Smoke During Pregnancy__ ...... Tahlc 3.1:FaLtors Assessed for Independent Cau),a1 Impact on Intrauterine Growth and C. iestatt(mal Duration .3i

Statistical Findings: The Role of Support and Stress During Pregnancy I Figures A.1-1.2e: Table i.1) Figure f.1: Social Support and the Buffering Ifypothesis Figure 1.2: Five Possible N1(Klels tOr Studying .Associati( in, Between Psych( is( icial Factors and Pregnancy ( )utconws Ei,ore i.)11.:Direct Niects Minty!...... , ..... i() Figu e 1.21): Direct OC'ets .11Hdel. bIterrehahms.liiiinig PsivhisoculI Eacti)r ..... Figure i..2c: Direct and Indirect /Weds .11ridel.

bilernitiliHnsIntwg, and Bettirot Psivhomie lid Fat 1(nN and .SticiudenirtgrapInc . .S0 Figure Durtt anti P810.7u..ncial Sociottenupgraphic LimihtualS, Hcalth Ik'bdtlor it) Ilediated Mudd.1:fikThrwigh Health Beharuir

Tahle #.1: Correlates of Cigamte-Smoking in Pregnanc\ Among Women of Central I larlem . . Results of a Three-Ycar Prospective Controllr2d Randomized Trial of Preterm 13irth Preventionat The University of Pittsburgh (Figure 5.1) Figur-L. 5.1:Percentage of Patients with Preterm Pretenn Birth, and Tocolysis During the Three Years of the Study

Statistical Findings: Decrease and Rise in Rates of Preterm Deliveries Haguenau Prenatal Study 1971-1978 (Tables 6.14).8) Figure 0.1:Regression Curve for Wc,inen by Group and Number of Pretenn Deliveries Table 0.1: Pretenn Deliveries and Live Births Table 0.2: Transfers to a Ne( 'natal Intensive Care 1 Table 0.3: Days of ilospitalization Related to Pretenn Delivery Table OA: Neonmal Deaths Po. Thousand Live Births -0 Table 0.5: Rates of Preterm Births, by Risk Factor. by Period Table 0.0: Rates of Pretem Births. When a Risk Factor \Vas Observed or Not Obsen,' 70 Table 0.-": Rates of Iligh-Risk Factors Anumg Pregnant Women by Study Pericid -0 Table 0.8; Change in lkenatal Caw and Pretenn Binh Rate According to Parental Educational Level The West Los Angeles Prematurity Prevention Project: A Progress Report (Tables -7. l-1.2) Table '.1: West I.os Angeles Prematurity Prevention Project Preliminary Data...... 82 Table 7 .2: Psychosocial Risk Factors Associated with Preterm Birth 83

Statistical Findings: The South Carolina Multicentered Randomized Controlled Trial to Reduce Low Birthweight (Figure 8.1; Tables 8.1-8.5)

R I Figure 8.1:The South Carolina Multicentered Contn4lcd tz Ili!to ..i'LLICV LOW Birthweight . Thhle 8.1: Selected Maternal Characteristics 96 Table 8.2: Selected Maternal Characteristics 9- Table 8.3: South Carolina Randomized Controlled Trial Sur\ ce Statist cs t Live Births). Analytical GnIup 98 Table 8.i: Birthweight Outcoines (Live Births) 99 Table 8.5Birthweight ()utcomes (Live Births) 9')

Statistical Findings: A Prematurity Prevention Project in Northwest North Carolina (Figures 941-9.5: Table 941)

Figure 9.1: Rates of LBW and VLBW Births in Northwest North . .. Figure 9.2: Rates of LBW Births Within the Projeus. Rates Among 1'nenrolled Private Patients ...... Figore 9.3: Rates of VLBW Births Within the Project vs Rates Among I'nenrolled Private Patients 1(K) Figure 9.1: Rate!-: of Premature Births t< 38 Weeks Gestation) Within the Project vs Rates Aiming l'nenrolled Private Patients 100 Figure 9.5:Rates of P-LBW Births Within the Pniject vs Rates Among I 'nenrolk.d Private Patients able 9.1; Private Births in Northwest North (;:irolina

Statistia Findings: The Family Workers Project: Evaluation of a Randomized Controlled Trial of a Pregnancy Social Support Service (Figure 10.1; Tables 10.1-107) Figure 10.1: Overview of Trial ...... Table 10.1:Study Population 118 Table 10.2:Reasons for Non-Acceptance 119 Table 10.3:Outccime Information Not Obtained 120 Table 10.4: Mean Birthweight (Singleton Live and Stillbirths) and Length of Gestation

Table 1.5:0 Proponion I.(iw Birthweight. Small for Gestaticinal Age. and Preturm Babies 112112( Table 10.0:Outctnth: of Pregnancy 121 Table 10.7:Categorization ()I t.;statkmal Age. Excluding Terminations for litith Stu. al and Medical Reasons 121 Statistical Findings: The Sodal Support and Pregnancy Outcome Study (Figures 11.1-11.2; Tables I 1.1-11.19) Figure 11.1: Consent. Randomization. and Data CollectionProvediires 133 Figure 11.2; Study Sample :ind Pregnancy Outcome by Assigmnent intoInterventkm ..tnd Comrol 133 Table 11.1:Structure And Content of the SSPO Interventkm 134 Table 11.2:Otinparability of Groups at Reeruittnent to 'Fria! 13.1 Table 11.3:NIcan Birthweight 134 Table 11.4: Low Birthweight 13i Tidily 11.5:Ges:iilion it Delivery 135 Table 11.0; Onset of Libor 135 Table 117: Mean Length of Labor 135 TAIle 11,8: Type of Delivery Table 11.9:Analgesia .Anesthesia in Libor 'raffle 11,10: Maternal Phsieal Ifealth in Pregnancy 130 Table 11.11: Medical C:ire in Pregnancy Table 11.12: Nlaternal Physical Ilcalth Atter Birth 11:ii(6) Table 11.13: Baby's Condition After Birth Table 11.1 t: Baby's Care After Birth 1321.4 Table 11.15: Baby's 1Icalth After Birth 13" 'Lillie1 1.16: Psvcluisocial Outcomes able IAA": Partner's Stipp( wt and IlcIp During Pregnancy. Birth,and Atter Birth 1 3- Tahle 11.18: Attitudes Toward the Intervention 138 l'ahle I 1.19: Value cif Contact with NIidwile in Pregnancy 13S

StAistical Findings: Prevention of Preterm Deliveries1w !Ionic Visiting System (Tables 12.1-12.0) Table 1 2.1:Nlaternal DemographL tnc1 Scwial Characteristics 149 'Fable 12,2:Conn.:it:0(qm Cervical state at Study Entry 149 Table 12.3:Prenatal Care Alter Study Entry Table 12.4:Nlother's View ot Prenatal Care 150 1...11)1e 12.5:Ink Wination CAintractions ;end liedrest Tahle 12.o:tiedit..st and limle

Statistical Findings: Smoking Interventicms DuringPregnancy (Tables 13.1-13.9) 'Lihie 13.1:!:.stimates ut Relative Risk tor Principal It 'C,R Risk Factors 163 1(13 Table 13.2:Smoking Characteristics Contnil and Treatment Groups at Eighth Nli)illht Pregnano-. Table 13.3:Nleasurement ()t. status (4. Newhtirn 13.,f:Restilis of Three Randoini/ed Clinieal Trials 1 161 13.5:Smoking and Gestaticinal Characteristks by it(71's Examining Birthwvight

. 11(6,34 Table I $.0:inter\ ention and Vic k.ation Characteristics ot RCI's of ExaminingBinhw eight ...... Table 13,-:Quit Rates and l3irilvglns 1)V RiniIca 11: and Nlarvland Colton,. .105 ...... 1 (r"; Table 13,8:Quit Rates and Birthw eights tor Blacks and \\'hites inNlaryland Cohort Table 13.9:Additional Randonlifed Clinic:al Trials of Smoking Cessat)n 1 tv;

Statistical Findings: Crvical Crckige: New EvidenceFrom the Medical Research Ctnincil Royal College of Obstetricians and Gynecok)gists(Tables 1.1.1-1,4.5) Table Ill: Iitiii CcRical Cerclage c in C )bstetric Nlanagement----FAidcnie hoin the Rand,imized Trials 1-3 TAN,: j+.2., HIcts Cvrcical Ccrdage t n ( A) 1)cliverv 1h.tOre 33 Wc'eks and ( B) NliNcarriage, ;Ind Neunatai Death Ctimbined--Evidenie trunithe Four Randinni/ed Trials 1- t Second Trimester Nlicarriage e ir Preterm Table 1 t.3:11 it Prophylactic. (:rvical Cerclage Atter Pre ions Delivery fin ) 1)eliery Before 3 Vi.eks and ( HIMiscarriage. Stilibioh, and Neonatal Death._ ...... 1- Tahle 1 ie t PR)phyLalik. Cervical COI:Lige kir Tv. in PR.griani. on(AI h. hytTV 1 M.( v 33 Wecks and Ili) Miscarriage, Stillbinh, and Neonatal Dcath 1-5 Statistical Findings: Cervical Cerclage: New EvidenceFrom the Medical Research Council/ Roya' College of Obstetricians and Gynecologists(Thb les 1-+.1-1.1.5. Continued) A)1(.. 1.1.5:Effects of Prophylactic Cervical Cerdage for Womenat Moderate or Low Risk of Early Dehvery on (A) Delivery Before 33 Weeks and (13) Miscaniage. Stillbirth. andNeonatal Death 175

Statistical Findings: Prevention of Intrauterine C;rowthRetardation with Antiplatelet Therapy (Tables 15.1-15.4) Table Design Characteristics of Five Rev.ewed Studies 183 Table 11.2: Sampie Sizes and Other Criteria (4. Eive ReviewedStudies 153 Table 14;.3: Summary of Results of Three CompletedStudies 1S:t Table ICA: Evolution of Iii(nnechanical M(mitors 18.1

Statistical Findings: Does Calcium Supplementation Reduce Pregnancy-Induced I iypertension andPrematurity? (Figure's 16.1-16.3: Table's 10.1-16.() Figure 10.1: Differences in Adjusted Mr-cid Pressure Nlean Valuesat Term Between Calcium and Placebo Groups (Adjusted for Race and Initial Blood Pressure< 2( Aeeks) .194 Figure 10.2: Differences in Systolic Blood Pressureat Term Between Calcium-Supplemented and the CoiTesponding PLicelto Groups in Two Studies of CalciumSupplementatkin 1/uring Pregnancy.....19:1 Figure 10.3: Study Design. Randomized D(mhle-Blind Controlled ClinicalTrial Table 10.1: Incidence of Pregnancy-Induced I lypertension Data from Calcium Supplementation Studies193 Table 10.2:Reliability of Bloc id Pressure Measurements. Rosario. Argentina (Whites) 193 Table 10.3: Reliability of Bloc )d Pressure Measurements. Johns llopkins Ilospital(Blacks) 193 Table 10,:t:Outcome Variables f( ir the Randomized Clinical Trial ot Calcium Supplementation

/uring Pregnanc 19-1 Table )03:Effect cif Calcium Supplementati(inin 1111 Preeclampsia: A Randomi/ed Double-Blind Contnilled Clinical Trial 19.t T.thle 10.0: Etruct)1 Calcium Supplementatkin on the Ineidence of Prematurityand 1.(itt 13inhtt eight 19a

Nlagnesium SLipplementation in Pregnancy: A I)ouhle-Blind Study(Tables 1-.1-1-.2) Thhk 1-.1Comparison Between Magnesium and Placebo ('Innips Table 1-.2:Ccimpariscin Betwecn NLtgnesium and PLicebo Groups After Exclusion of Patients Who Did Not Take Their Tablets Regularly 2(10

Statistical Findings: Prevention of Preterm Deliveries by IIonic Visiting Systcm (Figures 18.1-18.8; Tables 18.1-18.2) Figure 18.1: Effect of S ocial Support in Pregnancy (asan Explicit co-lniert entioni tin Preterm ' /divert- Rate (< 3- \\ eeks) Figure Effect (CI Social support in Pregnancy (asa Primary Inter\ ention) Pieteriii I/cliverv Rate (< 3Veeks) 221 Figure 18.3: Effect (ifoci.il suppon in Pregnanct (asan Eylio (:(,-InterN entu CI) Extremely Preterit) Dein cry Rate (< 33 Weeks) 'I Figure 18.Effect of St n ial Supp(irt in Pregnancy (asa Pr)mal) Inter\ ention3 ( Extremely Preterm Delivery Rate « 33 weeks) Figure 183: Ettect tit S oct:il StIppcla in Pregnancy (asan Explicit (:(-Intervention) in row Birthweight Rate t< 2:1(H) g) Figure 18.0: Iltect of Social Support in Pregnancy(as a Prnriary Intervention, on It iw Bin Int eight Rate t< 2-400 gi 123 ...... Effect of s cicial Stipp( in it) Pregnancy (as an ExpIk.it Co-InterventicinI on Very 1.(A% Binhv eight Rate (< 1500 g) Figure 18.8: Effect of St )cial Suppon in Pregnancy (asa Pr Mart Interventicm) tin Verv I.( At Birthweight Rale (

Statistical Findings; Inhibition of Preterm Labcr: Is It Worthwhile? (Figures 19.1-19.6; Tables 19.1-19..1) Figure 19.1: Effect of B,etamimetic Drug Treatment in Preterm Lahor on the Incidence of Delivery Within 2+ flours in the 1.f Trials Which Plovided Data on this Outcome '53 Figure 19.2: Effects of Bet:unimetk. Drug Treatments in Preterm Labor: 'Typical Odds Raticr.. With l'heir 25 Percent (:onliiknce Inter-v:0s. ..kcrossi.ri:ds for the V.iri)us Outcomes Studied Figure 19.3: Effect cif Betamimetic Drug Treatment in Preterm Lalxir On the tocidence of Severe Respirahwv Disc wders Including Respiratory Dktress Syndr()nw 25 I Figure 19.-t: Effect of liciamimoic Drug Treatment in Preterm Libor on the Incidence of Perinatal Nlortalitv Not Attributable to I.ethal Congenital Malformations )51 Figure 193: Comparison Between Betainimetic Drugs and Eth:mol for TreatmentiF Preterm Labor: Effects on the Incidence of Delivery Within '18 to -.2 liEmrs After Entrt: into the Trial 2-S5 Figure 19.0: Effects of lndomethacin in Preterm Labor: 'Typical' Odds Ratios. With Their 95 Percent Confidence Intervals. Across Trials for the Various Outcomes Studied 255 Table 19.1: Summary of Appntaches Vsed to Inhibit Preterit) Labor in the Past Years and Their Purported Success Rates in the First English Language Publication Documenting Their I se Table 19.2:Characteristics of Birth and In-Hospital Mortality and Morbidity in Infants Born Alive Bel( we 32 A-CAS (22.1 Days) of Gestatkyn in the Netherlands in 1983 14-,1 Table 19.3: Agents Reported to he 1 sed for the Inhibition ot Preterm Labor in the Liteure Since 1980 )432 Table 19..1;Characteristics cif the Trials Comparing Betatnimetics With Ethainil tor Inhi1,41 in (if Preterm Labor ....252

xi Acknowledgments

TIIE STEERING CONIMITIVE IS INDEIffED TO Till'. sponsorship of the meeting by the Task Force on the Preventionof Low Birthweight of the Secretaty of Health and Human Servicesand the financial support provided by theMaternal and Child Health Bureau. Ilealth Resources and Services Administration. and the National Institute of Child Health and Human Development, National I nst itut es of Healthof the Public Ilealth Service. United States Department of Health and Human Services. We also acknowledge gratefullythe support provided by Ruby Hearn from the Robert Wood Johnson Foundation andRichard lorton from the Nlarch of Dimes. Much of thecreditforobtaining manuscripts from the conference participants, editorial corrections and assemblingof final manuscript drafis belongs to MarthaGordon who also contributed immensely to the planning and arrangementsfor the original CltatIlam conference. Foreword

A MAIM DITERNIINANT OF INFANT NIORTALITY is low birthweight. particularly very low birthweight births. Very low birthweight births (under 1500 grams) account for almo,zt 50 percent Of deaths during thefirst year of life. Neonatal intensive care is very expensive. particularly for the very low birthweight or severely premature baby. and the cost of neonatal care in the tinited States has been estimated to be about S2---S2.5 billion per year. In addition, a significant prop()11ion of surviving small premature babies have substantial neun)gical, behavioral or learning pa )blems as they grow up in society. Prevention of low birthweight. therefore, has become ',Imai( w national objective. and research and interventions designed to achieve a reduction in therisk of low birthweight are of the highest priority. In 1985 Dr. Emile Papiernik organized a meeting entitled "Pievention of Pretenn Births" with the subtitle -New Goals and New Practices

14 Advance.. in the tremnion qf Lute Birlinveight in Prenatal Care- which was held in Evian. effects on the risk of pretem delivery through France. May 19-22, 1985. It brought together calcium supplementation or magnesium supple- scientists from around the world involved in mentation. A concluding discussion attempted to clinical trials or community-based interventions identify directions for future research. aimed at reducing the risk of preterm delivery. There was a spirited exchange alx)ut scien- This meeting was highly successfUl and provid- tific intervention projects in urban populations ed an excellent assessment of the then-ongoing and international collabcnatk)n. Slumld we test effort to prevent preterm births and served as single or multiple interventions, especially in an important forum for exchange among vari- inner-city populations? Dr. Papiernik's interven- ous investigators. It also emphasized the impor- tion strategy in France consisted of multiple tance of clinical triak in perinatal medicine. interventkms. His pnigram also appeared to be In this spirit, a steering eonimittee was most effective in low- or medium-risk women formed in 1987 to plan for a follow-up ilmfer- rather than high-risk women. Overwhelming ence which was entitled -Advances in the social problems in inner-city populations in the Prevention ()FLOW Birthweight- which many of United States May create immense obstacles to us affectionately called Evian II. Members of the evaluation of the Papiernik educatkmal the Steering Committee included Drs. Robert interxention model and therefore socially high- Creasy, Cal Hubei. Woodie Kessel. Irwin risk groups may be inappropriate for testing of Merkatz. Richard Morton. Sunnier Valle and these interventions. In designing social support Heinz Berendes. We wanted to bring together trials what is meant by social support? Patients' researchers conducting clinical trials or commu- percepticni may be different film) investigators' nity-based interventions aimed at reducing the intention. There is urgent need for an improved risk of low hirthweight.It was our belief that risk assessment tool although the current expanding knowledge through this state-of-the- imprecisi(m of risk assessment may be an accu- art review of strategies for preventing loN: rate reflection of our very limited understand- birthweightthe principal determinant of risk ing of the determinants of preterm delivery or for poor survival and or life-Icing intrauterine gn)wth retardaticm. morbiditywill greatly enhance our ability to Considerable attention v.as devoted to a provide quality health services to vulnerable discussicm of methodological issues and end- populations and significantly cc mtrihute to the points (A interventions sudi as biological vari- advancement of matemal and child health in ables and or rroNies for mortality and the I'nited States and worldwide. morbidity. The papers presented in this volume cover We have the sincere hope that this series a broad array of scientific investigaticms which of conferences which Dr. Emile Papiernik initi- includes biological (...xposures, clinical trials of ated in Evian in 1985 has made a significant the effect of social support during pregnancy on ccmtributkm to our knowledge and ability IC ) birth oUtcoine, the Use of antiplatelet therapy to improve the health and well-being of future prcvent preeclampsia. and possible beneficial generatic ms.

h Arewon I 15 ill TreIIII lt, Treridc in Rates qf Low Bilibueight in the United States

NIARIE MCCORMICK, 1\1D,, Sc.1).

INTRoDuclioN Presenting this topic to this audience represents a classic example of the "coals to Newcastle" phenomenon. This phenomenon will become particularly evident as the discussion unfoIdssince it relies so heavily on the work of several here. What I hope to achieve, therefore, is to summarize what can be characterized as -what we already know" hy way of introduction to this conference.

IMPOWIANCE OF Low. Bum WEIGI-IT What do we -know- al)out the importance of low birthweight (LBW)? In the United States, infant mortality rates remain relatively high compared to other industrialized countries.' Most infant deaths now occur in the neonatal period, and the majority of neonatal deaths occur Advances in the Pretention Birtinveight among LB\X1 infants." The persistence of high Birthweight tends to be inure accurately mea- levels of infim; mortality is not because we have sured and less likely to be missing from our failed to achieve dramatic declines in infant mor- vital statistics data, however, and has therefore tality over the past 20 years. These declines, become the tneasure most frequently used in however. reflect our success in reducing the data from the t Inited States. mortality rates anumg small babies; indeed, our Let us also quickly acknowledge that refer- birthweight-specific mortality rates may be ring to birthweight groupings or cut-offs among the best in the world. A recent report reflects a somewhat arbitrary point along a from the U.S. Congress' Offk.e of Technology continuum of risk and mortality. In other Assessment (MAY illustrates this point. Over the words, 2500 g or 1500 g does not represent a past 20 years, the mortality rate among very tiny biolc)gic discontimity, but a cut-c)ff point which infants has declined dramatically, while the pro- has achieved utility as a marker of risk through portion surviving with severe to moderate handi- repeated usage. Because certain groupings or cap remains small. Thus, the proportion of cut-off points have achieved currency and their survivors in reasonable health has increased. meaning is well understood, we will continue In spite of this increase, the persistent to use them in this discussion. minority who survive with appreciable levels of Let us return, however, to pursue the point handicap is of concern. Nh)reover, the cost Of of intrauterine growth a hit further. Low weight care for such tiny infants is substantial, and at birth may result from one of two pro )cesses. remains high even when compared to our seem- either independently or in combination: ( ) ingly exorbitant costs for a -normal- delivery.- shortened duration of gestation, or (2) less Thus, both on a health and on a financial basis, growth than womid be anticipated fon a given low birthweight presents an important pix)blern. length of gestaticm. Among the former, growth (height and weight) is appropriate for gestatk )11- a I age (A( A), but delivery has occurred befc)re 37 weeks of completed gestation. t 'nlike growth MEANING)lIA AV BUM tWl1GIFF retardatkm, the causes of preterm labor and the Likewise, we know that a discussion of factors which increase the risk of it require fur- 10 )w birthweight is really a shortlund notation ther elucidation.' Among the latter, which is for the adequacy of a complex physiologic pro- attributed to intrauterine growth retardation cess: intrauterine fetal gr()wth. At each week of (11'GR), the infants are judged to be small for gestation, there is a distributkin of birthweights. gestational age (SGA) hy being horn less than Using a cut .off point or grouping based cm the established percentile fc)r growth fc)r a given hirtIm eight, then. clearly captures a gr()up that gestational age. A variety of conditions are asso- is heterogeneous for duration of gestation: ciated with An SGA infant may ako be Both birthweight and gestational age are relat- premature. At a given gestatk)nal age. an SGA ed to mortality,- and fc)r many purposes gesta- infant is less likely to survie than an AGA infant tional age is the better indicator of risk. where higher weight confers an advantage. At a 1 s

Trends in Rates (f tow Biabweight in the ()tiled States Mends

given birthweight, however, maturityconfers the tribution with a higher proportion ofthe very advantage to the SGA infant until term.' high-risk births.9 Most births less than 1501 g, both AGA or SGA, are also preterm. These trends suggest that much of thv decrease which we have experienced isin low birthweight due CI IANGES OVER TIME to IUGR. This interpretationis supported by What, then, has the recent U.S. experience other analyses."'" What factors may be invoked been in terms of trends in low birthweight over to explain this trend orlack of trend? time? Unfortunately, the rates of lowbirthweight are not changing very much. The proportion of live births less than 2501 ghas FActI AssoctATED WM !- decreased slowly by a little more than 15 per- TRENDS IN .Low Bun 1WEIGIrr cent in die past 15 years (seefigure 1.1). The proportion of live births less than 1501 g. how- The most striking factor in the U.S. experi- ever, has remained the same orrisen slightly. ence remains the racialdisparity in rates of low Thus. while the prc)portion of LIMbirths has birthweight." The LBW ratefor black declined overall, this population is now weight- Americans is twice that of whites andother ed toward the lower end of the birthweight dis- racial/ethnic groups, and is well above the

Figure 1.1 Proportion of LBW and VLBW Births, By Race

6 -A

4 -

A A

PM7 1(150 1955 196() 1%4 1'170

NtrnA hitt. < 2500 g Whitt. < 2500 g 11111 Non-whitv <.I 500 g Whitt- < I -AM g

Marie McCormick 1 5 Advances in aw Prelyntion cf Low Birthweigbt objective set for 1990. This higher rate of LBW ing higher proportions of births to women char- is paralleled by higher infant mortality rates. acterized by these factors than whites. For both although unlike LBW, the mortality rates are groups, however, the pr()portion of mothers less declining." What accounts for this difference? than 18 years of age and mothers with less than One hypothesis is that there are genetic or 12 years of education (graduatk)n/schooMeaving inherited differences in birthweight distributions. norm for the ITnited States) has decreased: the Support for this argument can be derived from proportion of mothers who are unmarried has studies which document an intergenerationai increased, but this probably does not connote correlation of birthweight. In addition. analy- the same degree of risk now as it did in more ses which compare the full range of birth- permissive times.It' anything, then, the relative weights between ethnic groups reveal that the risk for these variables has declined, which rep- entire birthweight distribution for blacks is shth- resents a trend that is not consistent with the ed downward." Such shifts suggest that standard trend in low birthweight. cut-off points, such as 2500 g. do not connote These factors are not independent. A the same degree of risk for black b::-ths as for woman who initiates childbearing earlis less nonblack births. Indeed, for a given low birth- likely to complete school and to he married. weight. black infants have lower birthweight- Kleinman and Kessel have examined this ques- specific mortality rates than whites. Adjusting for tion for selected states in the years 1973 and birthweight distributions results in higher mortal- 1983.'" When these factors are analyzed in com- ity rates throughout the birthweight distribution bination. the relative differences between blacks for blacks, however, supporting the argument and whites diminish among the high-risk that they are at a disadvantage." While the women. In other words, women who are at high potential genetic contribution of race to birth- risk by virtue of their age and or educational weight and the appropriate statistical techniques attainment have similar pregnancy experiences for adjusting for differences in birthweight distri- regardless of race. Thus, part, but not all. of the bution require further study. disparities in birth- racial disparity can then be attributed to the fact weight among subgroups within the black that black births are niore likely to occur among population using conventional markers of birth- women who fall into the high-risk categories. weight risk suggest room for improvement with- If this is the case, then, what isit about in the current genetic endowment. Itis these maternal age or education which leads tO the disparities among subgroups which we will higher rates of IOW birthweight? Although blacks explore further. are twice as likely tO be adolescent mothers, and Epidemiologic tradition has identified three substantially more likely to he very young moth- sociodemographic characteristics which are ers. the relative risk of a low birthweight birth associated with increased risk of low birth- for young hlack mothers c(nnpared to older weight: (1) young maternal age. (2) low mater- black mothers is less than the relative risk for nal educational attainment. and (3) marital young white mothers compared to older white status.' Blacks remain at a disadvantage by hav- mothers. These findings are consistent with the

2 0 Trends in Rules (y* 1.(m' Birtbeivigbt in the ()tiled States l'rends

increased risk of eVen -low-risk" black mothers for maternity services which stem from our medi- noted earlier. They are also consistent with an in- cal care financing customs: 17 percent of all terpretation that age per se does not confer risk. women of childbearing agelack insurance for Such an interpretation is reinforced by the find- maternity services. Even with private insurance ings of special programs for adolescent mothers coverage for MOSEmedical services, maternity which reduce the risk of adverse outcome, hut care may only bepartially covered or not cov- only to the level of older mothers in their emi- ered at all, since pregnancy is considered an ronment. A similar picture emergeswith educa- elective event not amenable to classical insur- tion.1; Thus, changes in sociodemographic ance actuarial techniques.Another 10 percent of characteristics of mothers account for a relatively women of childbearing agerely on public small percent of the trends in low birthweight. resources, but even thoseeligible for public sup- If education or age per se does not account port may find obtainingprenatal care difficult. for what we see, then perhaps the questk ms Participation may be limited by complex applica- should be rephrased. Is there some factor which tion processes and inadequate fee schedules to is associated with early child bearing. 10 AV educa- cover the usual costsof maternity services: tional attainment, and perhaps black face which Consistent with a relationship between pre- leads to adverse outcome? In other words. are natal care and pregnancy outcoine is the fact that these factors markers for other, truly causal fac- gnmps identified as high risk are morelikely to tors, factors which would pertain toluith whites receive inadequate care.' Mc wet )Ver. the propor- and blacks, but to blacks disproportionately? An tion of women potentially at higher risk for immediate candidate leaps to mind; poverty. adverse outcome who start their care in the first Consideration of poverty as a causal factor trimester has changed little over the pastfew suggests a number of pathways by which the years, which is again consistentwith the lack of risk of low birthweight might be increased. change in pregnancy outcome. As with maternal Since much of the remainder of the conference sociodemographic characteristics, however, elim- will be spent in discussion of many of these ination of those with less than adequate care factors and potential interventions to ameliorate w(mld have only a modest effect on the rate of their effects, 1 will restrict the following discus- low birthweight (a12-1.C('4, reductkm ), Even sion to illustrative examples. c(mibining this with reductions of those with One aspect of poverty is a restriction Of sociodemographic risk results in a decrease of financial resources to obtain needed goods and only 29 percent in the LBW rate for whites and services. A necessary. if not totally sufficient, ser- 30 percent in the rate for blacks.'"Fhus. access vice or set of services hi attaining positive preg- to prenatal care is only partof the problem. nancy oflit (*Mlle S isadequate prenatal care. Constrained financial resciurces may also Inadequate prenatal care is higher anumg low- indicate an inability to obtain other necessities income gnitips and among the high-risk groups for a good pregnancy outcome. One which has previously noted.`' This deficit for mothers in the received attention is adequate nutrition. United States reflects, in pan, gaps in payment Controversy exists as to the proportion of the

Marie McCornikk 21 Advances in the Prevention (y. Low Birthweight

U.S population experiencing hunger or an their pregnancies.'" National data on smoking inability to Obtain the needed daily caloric do non reveal much disparity between the rates intake to sustain weight. Mtal caloric depriva- of black and white wo)men, and black women tion, however, is less likely than inappropriate tend to drink less. The role of illicit drugs is mixture of basic foods and other nutrients. notoriously difficult to assess both due to the Since it is not clear what an "appropriate mix- reluctance of individuals to report drug use and should be. the exact deficits in the diets of the the uncertain content of "street- drugs. A well- poor cannot be readily identified. That there established correlate of poverty is the lack of are deficits is suggested by the data on the employment. Thus, such health-related habits effectiveness of the WIC program (Special would appear, at first blush, not necessarily to Supplemental Food Program for Women, be related to) racial inequities in birthweight. Infants and Children).'" This is not unalloyed It would seem, however, that the questions evidence, however, since this program provides ought to be more sophisticated than that. First, a mixture of services, as well as ft >Oct supple- the lack of paid employment does not rule out ments. The area of appropriate nutrition repre- stressful labor, given the lack of financial sents :in area of further research. resources to invest in labor-saving houselmld Consideration of other aspects of the appliances. More importantly, such factors may effects of poverty reveals additional mecha- confer a different level of risk for women nisms by which it may contribute to adverse whose health may already be connpromnised by pregnancy outcome. Substantial evidence exists long-term exposure to health risks associated to support a relationship between poverty and with disadvantaged environments. In a recent poor health generally. With regard to pregnan- study of the mothers of young infants in central cy outcome. maternal health factors would Harlem. H) percent had been hospitalized in include the (I) relationship of specific health the year since delivery; 12 percent rated their problems as complications of pregnancy, and health as fair or poor; and 1.4 peecent set ved in (2) health practices which might affect bonfi the distressed range on a mental health scale. maternal and child health. Among the latter are Nloreover, soma..( f the risk factors fo)r poor the use of cigarettes, illicit drugs, and excessive pregnancy ciutcome also correlated with poor alcohol, In addition, certain types of physical maternal health subsequent to pregnancy, sug- activity inav increase the risk of pun. pregnan- gesting an interactive effect. cy outcome., ;IS suggested by the data for More current infonnatitm is needed on the women with physically and emonkmally stres.s- health pro)blems and behaviors of women of ful lobs which require long commutes.th In the childbearing age as they might affect their capac- United States as a whole, approximately 30 per- ity for reproduction. Perhaps our very low mater- cent of women smoke and 55 percent drink nal mortality rates have provided a sense of before their pregnancies. hut a pro)portion security which may require further re-examina- become abstinent during pregnancy, so that tion. although black .white disparities in this close to half of U.S. women do neither during health status indicator may have caused somie

22 8 Trends in Rates (y. blet. Birthweight In the I 'tilled States irriuts

concern.' Ve do collect 11ealth clata on women studies have implicated such combinations in the of childbearing age through the Natkmal I lealth 'risk of adverse pregnancy outcome." Interview Survey (our Harlem data parallel national data on black women), the National Natality Survey, and birth certificate analyses, It St 'N1MARN' AND St IGGES11ONS is difficult to establish, however, how the subset Of %%Innen who bear chiklren may differ from the In summary, the km birthweight rate in larger group of women of childbearing age, the United States remains high, with only slow since at any one time only a minorityof women decreases and a predominance of prematuw are experiencing pregnancyand childbirth. births. These high rates reflect the persistence Seeking prenatal care and changing health of socioeconomic disadvantage. especially habits to improve pregnancy outcome require a anumg blacks. The factorsunderlying these certain amount of anticipation orplanning. The trends are complex, and this complexity sug- groups nuist at risk, however, arethe ones least gests that change will not comeeasily. likely to have intended to he pregnant.' These In view of this complexity and lack of findings suggest that, in addition to prenatal ready explanation for the persistence of low care. low-income women mayhave difficulty in binhweight, our own experiences and those of achieving access to family planning services as others suggest some questicms and appr()aches well. It is small wonder that such access may he which, while not exhaustive nor even perhaps difficult to sustain given the fact that reproduc- original, may be useful in the next two days. tive care for low-income women may hespread The first question relates to our understanding across several typesof clinics: family planning of "risk.- The majority of low-inc(mie women, or clinics, gynecology clinics, clinics for sexually black women, or poorly educated women. or transmitted diseases, abortion clinics, and obstet- very young women havenormal hirthweight ric clinics (many with legally dictated separa- ( NW )infants. We tend to treat woinen with tions of facility and stall).- the.se characteristics as if they were homoge- Nh)re( wer. overcoming harriers to care and nec us. and yet each of theselabels masks ccm- modifying unheahhful behaviors requires motiva- siderable heterogeneity. Perhaps we should be tion. energy, and support.' As the earlierdata on considering more -micro- studies on the risk fac- maternal mental health suggest. other aspects of tors within communities and trends in outcomes disadvantaged envir(mments also adversely affect at a more local level with morefocused data to women in this sphere. Again inthis same cohort understand the nature of this heter()geneitv. of young mothers, envir(mmental stress as mea- Secondly, even with established risk fac- sured lw stressful life events is high, hut social tors, we need to be more specificabout the supportas measured by theability to identify mechanisms by which such factors alter risk an individual by nameand relationship to pro- and the potential modifiers of both the effect of vide help in six common situationsis not. the risk factor and interventkms to ameliorate These factors relate to maternal health, and sonny its effect. We are recurrently taken aback by the

Mark, McO)rnik-k C) Advances in the Provntion (y. Lou' Iiinbueight

relatively small contribution of prenatalcare to REFERENCES changes in birthweight. vet the relevance of the I. Wcgman. E. .Nnnual stitninAry individual components of that package of ser- statistics:1956.Poliatrics 80, vices which we call prenatal care to the current 2. Hogue, It.. Buehler, .1. W., Strauss.1..T.. :ind prciblems affecting pregnancy ou;come require Smith. J. C.(19S-1.Overview tit the \ational Infant Mortality stir% cillanee priwet: further examination. Furthermore, examples of Deslgn methods And results.PublicHealth 102.1 20- 1 .38. successful programs are rarely described in suf- SILipirci, S., NIct'airmick.tl.(... Suffield. B.1 I., ficient detail to assess the generalizability of the Krischcr. .1.P...,ind Noss. 1).(195( I)Relevance of interventkm. Likewise, the modifiers which correlates of infant deaths jor significant nImBidity at may contribute to the persistence of a well-rec- me year)1 age..1 moleau ./(qtrnat crfObstdricx mid 0:17recobigr 13o, 303-3-3. ognized risk like cigareue smoking are not well

f. . (.Ongrl'sr. t It'dint)it )gy AsS:sstill`nt. described. For example. in our study, envinni- II t)8-).Avoaata1 alleastne care lote hallairtght mental stress and lack of micial support did not atlaats(.4s1s atul clfivtavaess ( Health iivbilalagr contrilnite directly to birthweight differences. case Study :;,+' Publication No. OTI-11CS-35. Washington,I)C: 1.,s.Congress. ()Mei:ot. but were clearly associated with smoking Tt.'d mok igy Asst.,ssmem, behavkw in pregnancy, Additionally. we need Gold, H. B.. Kenney. A.. and Singh. S.t 198' I. better information on how sociodelllOgraphic Bie.:ed cvcuts and thelu ;mon Fina ;icing and behavioral risk factors translate into biolog- maternity caw in the maws. VA\York: Alan ical pr()cesses affecting fetal growth. inStialte. Finally. we may need to pursue some more Suffield.li., Shapiro. S.. Ntet:orinkk, NI. and complex intervention models.A comparable 1 irnss, 1).t 1982 I.Mortality and morhildity with Intratiterim. growth retanlation, turaal poillatrics problem is the mental retardation seen in chil- to!, trs-9s3. dren who come from disadvantaged back- Williams, H.I... Creas. H.K.. cunningharn. G C., grounds. There :Ire now several well-document- S. F... I'.D., etal. (1952).Fetal ed randomized trials of successful interventicms. .v.rt)w II),ind 1.11111ityIn and their success provides support for expanding )bstetrics mai O'racc-ology -;(),()2., +-(02. national programs for early childhood education. 5. ..raincr,k NI. S.(1)5-)Intrauterine gimili and ges- utional duration deiermin.imsPciliatrics so. These experiences, however. rely on well-devel- ;02-4;11. oped and explored ctmceptuahzations of early t) \aticin.11 Cuntcr Statistics I980).I le,alth childhood development and specific curriculums 'nnedShifts ?9,SY)1)1111S Puhlicatii in \ is aimed at the identified deficiencies of sufficient 1232. Wasl»ngton. DC: Government Printing Office. intensity to achieve the desired results.Itis 10. Kcssel s s I. Bet-Click's. II,N5... and \ugent. hoped that the presentations over the next few I.P. (19s ii.The changing t hA% hirth- days will raise us all to this more sophisticated cight in the Iniick1 I. ./Hurilell (?,1 the .1/neman MedicalAssoc uau ta -5 1. 19"M-1952. level of approach tO this complex problem, 11. Sarpcntield. W. 51., Hogue. ( J.It. Strauss.1..T.. and Smith. .1.C. lt)S-).Dilterenti..s in nctinata1 And pystneiinatal 111(0.1111v hy binh m..cight and gestational agc. Publn.loath Rep orts 1u2. 182-192 24

Trends in Rates (!j* tott.Birthweigbt inthe I lifted States 7Yends

12. Klebanciff. M. A.. and Yip. R.i198").Influence of 23. Istvan,I. C 198(, ). Stress. anxiety and birth out- maternal birth weight on rate of fetal gRiwth and comes:A critical review. of tile CV denee. duration of gest a Hi Ui.parrna/ (4. Pediatri(s 111, BUlleliP/ 100. 331-348, 28--292. 24. Brooks-Ounn. I.. NkCormk.k. M. C.. and Ileagarty,

I. 19881. Prventing infant mortality and mor- 13, Klebantiff .N1 ...... tiratibard. B. L. Kessel. S. S.. and c. Berendes. 11.\\..(1984),LOW birth weight across bidity:1)evelopmental perspectives.American generations.finirnal af the American Medical juunuif cg. Urthypsvcbiatti. S. 2$8-290, Association 252. 27. 1.1. Wilcox. A. .1.. and Russell. I. . 41983).Perinatal mortahty:Stanclardizing for birthweight is biased. Americali.journal (!fhpulemifdtigy 118.85"--64. 15. Insbtute of N1edicine. Committ to studythe Preventkin of Low Birthweight. 1f)85).Proem/lig /Awe Birihweighi,Washington, DC:National Academy Press, 16. Kleinman. .1. C.. and Kessel, S. S.(198").RACI:11 (fir fcrinces In lcm ltili \\ eight:Trends ;ind risk factors..Veu. EngIaad j(wrnal 01. MecheiMe .

17. K(nelchuck. Schwan,. J. H.. .nelerka. M. T.. anti Finison. K. s(198.4),VIC parridpation and pregnancy outcomes: Massachusetts statewide evaluation protect.American fiat rued Pubhc I. 10S6-1092 18. Mamelle. N.. LumanB.. and Laiar. P.t 19s e I. Prematurity .ind (it cupational tUi ity during pregnancy.American journal 01. Fptdomuilf,,10. 119. 309-322. 19. Prager. K.. Manlin. 11.. spieglet. D.. Van \ALL P.. and Placek, P(198+ ).Smoking and drinking behaviors before and during pregnancy of married Inuthers live-born infants and stillborn infants.Public Mika-is 99. 11"-128. 20. McCormick. NI. C.. Brooks-(unn, .1.. Shorter, T.. Holmes. J.It, Wallace. C. Y.. and fleapits, M. C. (198").Ilealth in the Year following delivery among low -Meow wi 'men. Pediatnc Research 21.2.85A. 21. Sachs, B. P. 13n iwn. D. A.. Driscoll. S G.. Schulman, F., Acker, D.. Ransil. B. et al.(198-1.Maternal int irtalitin Nlassachusetts:Trends and pies ention. .\eir kn,Q1and.harrual ,?/.1hAlic-ine 310. 60--(-2.

Aries, N.(198).Fragmenution and reproductise freedom:Federally subsith/ed famil planning set- V iCes.190()-- 1980.American Journal vf Pub/N.- 111.'0h-, 1393-1 395.

Marie McCormick 1 I 25 Erni ution ((the Preterm Birth Rate in Fiance

(FRAR1) BREAM', M.D.

INTRODt (711ON During the 1970s in France, reduction in the preterm birth rate was considered a major objective in perinatal medicine.This objective was included in a perinatal program implemented between 1970 and 1975.4 This program included incentives to increase prenatal care, outpatient clinics for high-risk women, and educational programs. It was I( )l lowed in 1975 by a series of measures, including regulations for the working conditions of pregnant women and implementation of a home visiting system. Since 1981. no new specific programs or actions have beendevekved. The aim of this paper is to trace the evolution of the preterm birth rate in France. The paper will be divided into two parts.The first part corresponds to the evolutionof the Advances in tile Prelention (V. Lou' Birthweight

preterm birth rate between 197 2 and 1981. changed dramatically in France (see table based on national data, and the second corre- both quantitatively and qualitatively. For exam- sponds to the evolution between 1984 and ple. the percentage of women with 7 or more 1986, based on hospital data. prenatal visits increased from 22 percent to 5.5 percent. The temporal relationship between the modification of prenatal care and the decrease in preterm births is not su. tH....c.ent to prove a Ex01.1.110N OF '11IF PREITRM BIRT11 RATE causal relationship. Other factors may have had BEWEEN 1972 AND 1981 an impact on the rate of preterm birth. Among them, the denuTraphic factors are the nu)st Material ai id ined.xxls important. Table 2.5 shows that women who The data h4.re presented are derived from delivered in 1981 were at lower risk for three national surveys conducted in France in preterm hirth than women who delivered in 1972, 1976, and 1981 on representative samples 1972. There were fewer women younger than of birth.' Each sample was obtained br a two- 20 years of age, fewer women with 3 or more stage (maternity unit, and deliveries within tile previous pregnancies, and fewer women with a unit) sampling procedure (see tabk. 2.1). For sh(wt interval since the previous birth. every delivery included in the surx.ey, a ques- To take into account the evolution of these tionnaire was completed. The representative- factors, standardized rates of preterm birth ness of the data was assessed by comparison were computed (see table 2.(). The 1972 popu- with data on all births. The main objective of latkm was used as the reference populatkm. these surveys was to evaluate, at the national The comparison between crude and standard- level, the results of the perinatal program. ized rates shows that changes in denu)graphic factors explained only (me-third of the decrease in the number of preterm births. Reszals Table 2.2 shows that. between 19-2 and 1981, the preterm birth rate decreased from 8.2 Onnitlents percent to 5.6 percent. It should be noted that In adchtion to the changes in demographic this decrease Was observed for births (Kcurring factt )rs. there have been other changes. Not all before 34 weeks gestation. as well as for births of the changes led to a decreased risk in occurring between 3-1 and 36 weeks' gestation. preterm birth. however: some of the changes.

This reduction in the preterm birth rate in fact. lud the opposite result. Therefore. WaS acc(mipanied by a reductk)n in the rate Of changes in maternal characteristics. as evaluat- low birthweight births between 1984 and 1986 ed through the data collected. do not appear (see table 2.31. This reductkm was observed I( n' sufficient to explain the decrease in preterm very low birthweights as well. birth. It is likely that the modificatkm in prena- During this same period, prenatal care tal care (including nonmedical care) had an

iv0" Ii Erellahni (y. tbe Preterin Birth Rate in Emuice ''rends

impact on the overall rate of preterm birth. neonatology. Another factor which may have Since many interventions have been imple- influenced French obstetricians is the trend, for mented at the same time. ranging from fetal growth-retarded letuse.s, toward an increased use of betamimetics to regulations increase in premature termination of pregnancy concerning working conditions, itis very diffi- to prevent long-term handicap. Thosefactors cult to evaluate the efficacy of any one particu- have raised some questions concerning the lar intervention. recent trend in the preterm birth rate in France. It is now accepted' that some of the pro- In order to collect the necessary inf(wmation, a posed interventitms have been overused or survey was conducted aincmg several maternity misused. flowever. in agreement with what is units. The results ore presented here. known about risk factors of preterm birth.' which are multiple and cannot explain all of the cases of preterm birth. itis believed that a Material cuul methods reduction in the ',menu birth rate could not To obtain information on the recent evolu- be obtained through a single intervention. tion of the preterm birth rate, as well as its Instead, a comprehensive program is needed determinants, a questknmaire was sent to 30 which includes several types of interventions. maternity units known to be users of the same among which obstetricians,midwives, and t \ pe of computerized record. These units social workers could choose one (or several) belong to an association called Association des specifically adapted to each v(nnan according rtilisateurs d'1111 Dossier Informatise en to her living and working c(mditions aswell as Perinatoh)gie. Obstetrique et Gyttecologh.'. The to the symptoms she presents. Sincethe purpose of the questionnaire was tocollect known risk factors explain only part of the data for the three more recent Years (198-4, overall rate of prtierm birth.- a pn)gram 1985, and 198( ) on the pretenn birth rate: decrease preterm birth sh(Aild not be limited to birthweight and mortality rates; social, demo- high-risk women. but shoiad be targeted to all graphic. and medical characteristics of the pop- women in the general population. ulation: and policy adopted for the preventicm of preterm birth. Among the 30units approached. only 9 were able t() provide complete data. They are iocated in different parts of Evou "no \ or France (see figure 2.1). and 2W)00 women per THE PIUTIIRNI HIRTI i RM'E SINCE1983 yeor deliver in these units. In recent \Val's. it seems that the viewsof French obstetricians- concerning prevention of preterm birth have changed. This change seems ResitIts linked to two main causes: First, concerns In the inaternity units studied (see table about the safety of very conuncmly used drugs. 2.7). the preterin birth has increased from (0 such as betamimetics: and. sect )nd. progress in percent to percent. According to impor-

Gourd lin31rI 15 0 V Advances in the Prevention ql Lou Birthweight tance of prematurity (< or > 3-1 weeks), or t o dons in the use of betaminietics), as well as induction of labor (artificial or spontaneous), modificatkms in women's attitudes as reflected an increase was observed for spontaneous by the percentage of women starting their pre- preterm delivery no matter what the gestation- natal care, after the first trimester, at the outpa- al age, whereas an increase in induced preterm tient clinics of the units. This change may not delivery was observed only for very low gesta- be due only to changes in women's attitudes, tional age (< 3.4 weeks). Table 2.7 shows that however, but also to changes in general practi- inure than one-quarter of the increase in tkmers' attitudes, who could have delayed the preterm births was directl- related to an referral to the maternity unit, or to socioeco- incwase in preterm induction for termination nomical problems, which could have made it of pregnancy. more difficult to access the maternity units To screen the possible factors which might under study. have led to an increase in spontaneous preterm The modificatkm of the obstetricians' atti- birth, the evolution of sonie risk indicators tudes seems to be recent.' as does the upward (e.g., frequency of women with previous trend in the preterm birth rate. Several of the preterm birth, maternal age under 20 years, maternity units which have experienced a unemployment, use of betamimetics, and first recent increase in the preterm birth rate had prenatal visit at the unit during the first documented a decrease during the 1970s. trimester) was examined during this period. The increase in preterm birth has not been Among the five indicators studied, twothe observed only in French hospitals. Table 2.9 percentage of w(mien receiving I>etainimetics shows thatit has been observed in Sweden, and the percentage of women coming to unit Norway, and Iceland as well:" during the first trimestervaried markedly (see table 2.8). Both frequencies decreased during the period. CoNct.t siox From the data presented here, it can prc)h- Cbmtneills ably be ccmcluded that a portkm of the preterm Even if the increase in preterm births births can be prevented. This statement is observed in some hospitals in France cannot be based on the decrease observed in France dur- considered as representative of the actual trend ing the 19M)s and the recent increase in some at the national level, the dat:i collected showed hospitals. This variation in trend in "spomta- that an increase in the pretenn birth rate has neous- preterm birth suggests the possible been observed at the same time as modifica- effect of scmie environmental factors. These tions in preventive attitudes have been factors could be social, demographic. or medi- observed. Modifications in obstetricians' ;1.10- cal, since the observed change in the preterm tudes have been observed (as demonstrated in birth rate corresponds to changes in preventive the increase in preterm induction and restric- policies (e.g., increase in medical intervention

2 9

/6 Evolution (01.7e Preternt Birth Rate France 7*rends

during the first period, and decrease during REFERENCES subsequent periods) as well as changes in I.Bouyer, J Dreyfus. J.. Guegen, S.. Lazar. P.. and social factors. The type of data presented here Papiernik. E.(19)8).La prt;maturitC'.Paris: does not allow anv conclusions to be drawn INSERM: EnquCie Perinatale 1 laguenau. concerning the factors which are responsible 2. Breart. Goujard, J., Blonde!, B.. Maillard. F., for the decrease or the increase in preterm Chavigny, C., Sureau. C.. and Rumeau-Rouquene, c, ( 1981 ). A ct nnpa rist )1'1 of two policies of antenatal birth. Such conclusions can only be drawn supervision for the prevention of prematurity. from experimental surveys. The results of the linernationallournaltf EpidernicIlogv 2.+1-2.41. randomized controlled trials presented at this 3. Papiernik. E.(1969/.Coefficient Lk. risque symposium, as well 'as presented at the d'aconichement preimiture.Presse Medicate 77(21), Evian meeting," do not lead to clear conclu- . sions concerning which interventions can Rumeau-Rouquent... C.. and Blondel. B. (198=0. The reduce preterm birth and which interventions French perinatal program, In:I. Vallin and A. D. )pez Eds. ). flealtbixdicy, social Ltxdic.y and mortal- cannot. These conflicting and disappointing ity provects (pp. 299-3311. Liege; Ordina. results might be due to the fact that prevention S. Rumeau-Rouquette. (.. du Niazauhrun. C.. and of preterm births can only be obtained through Raharison. (Eds.).(1984).Aaitty en France: Dix multiple interventions, each of them alone ans di:volution.Paris: INSERM. resulting in too few benefits to be observed in 6.Papiernik. E.. Wean. G.. and Spiro, N.(1986). an experimental study. Prevention tie la naissance prOmaturc;e. INSERNI, Another conclusion which can be drawn 138. from the hospital data is that prevention of Sureau. C.. Blt.)t.P.IL. Cairo). Cavaille. F.. and preterm birth cannot be considered as an Germain. G.(1986).Control and management of objective per se, but as a means to reduce parturition. 1NSERM John Lihhey. long-tenn handicaps. in some circumstances, 8. Kramer, M. S. 41987'1, Determinants of low birthweight:Nlethodological assessment and meta-analv- obstetricians will induce preterm terminatim of sis.Bulletin ty. the World Health Organization pregnancy because of fetal distress, and an 663-7'37. increase in preterm birth rates might lead to 9. Blondel. B.. Ringa, V.. and Breart, C.(1989).The better long-term results. Therefore, the overall use of ultrasound evaminations. intrapartum fetal preterm birth rate can no longer be considered heart rate nu)nitoring and betamimetics drugs in France. British Journal of Obstetrics anti a good indicator of perinatal results. Adistinc- Gynae(Mhigy 90. tion must be made between spontaneous and (1957).Births in ihe Nordic countries: Registration induced preterm deliveries, and perinatal pro- (ye the outcome q1" pregnancy 1979- /91S'3.Reykjavik: grams should be evaluated based onlong-term Nomesko Publieatkms. outcome or on indicators which are highly correlated with long-term outcome. Such indicators remain to be defined.

Gerard Bread I 7 3 0 Mends

Statisiical Evolution of Preterm Birth in France

Table 2.1 Table 2.4 Design of the National Surveys Evolution of Prenatal Care (National Data)

1972 1976 1981 1972 1976 1981 Percentage Percentage Percentage Sampling fraction 0.01.37 0.0068 0.0068 Number of prenatal visits Number of units 637 420 .390 < 4 15.3 10.6 3.9 Number of deliveries 11,254 4,685 5,508 4 35.4 29.1 16.5 5-6 27.1 p< 0.01 26.4 p< 0.01 24.7 + 22. 3.3.9 54.9 Table 2.2 iospitalization during pregnancy Evolution of Gestational Age (National Data) 7.3 p< 0.01 13.0 p< 0.01 15.6

1972 1976 1981 Betamimetics Percentage Percentage Percentage 8.9 pc 0.01 14.7

Number of Weeks Care by general practitioner only 1.1 < 34 2.4 1.7 20.8 p< 0.01 8.8 34-36 5.8 5.1 4.4 37-39 .37.0 p< 0.01 34.9 pc0.01 17.9 nc't ailable 40-42 50.8 54.0 5.3.1 43+ 4.0 4.3 3.4 <. :36 8.2 p< 0.01 6.8 p <0.05 5.6 Table 2.5 Changes in Demographic Factors Table 2.3 (National Data) Evolution of Birthweight (National Data) 1972 1976 1981 Percentage Percentage Percentage 1972 1976 1981 PeR entage Percentage Percentage Maternal age < 20 years 9.9 8.5 6.0 p< 0.001 1500 g 0.8 0.7 0.4 > 40 years 2.4 1.1 0.9 1500-1999 g 1.2 0.9 1.0 2000-2499 g 4.2 4,9 3.8 Number of previous pregnanc ies 2500-2999 g 19.1 NS 17.9 p< 0.01 18.0 3 + 18.3 14.1 15.4 p< 0.001 3000-3499 g 41.8 42.3 41.3 3500-3999 g 25.5 26.4 )7.7 Previous birth 4000 g + 7,4 6.9 8.0 < 17 months 21.6 15.4 10.2 p < 0.001 < 2500 g 6.2 NS 6.5 p< 0.01 r

Gerard [Wad /9 31 Advances in the Prevention of Low Birthweight

Poitou Charentes

3 '

20 Statistical Findings: Emlution of Pretenn Blab in France nyncis

Table 2.6 Table 2.9 Evolution of preterm birth rate Preterm Birth Rate (< 36 Weeks) in (Crude and Standardized*Rate, National Data) Nordic Countries

1972 1976 1981 1979 1980 1981 1982 1983 Percentage Percentage Percentage Denmark 4.9 5.0 4.9 4.8 4.8 Crude rate 8.2 6.8 5.6 Finland 5.2 Standardized rate 8.2 NS 7.8 p < 0.05 6.7 Iceland 2.6 2.7 3.0 3.1 3.4 Norway 4.9 4.9 5.1 5.0 r

StilnOaultration in) age.flUlTlik'r ofpalI11.1PTI;nd interv.il Sweden 4.5 3.6 5.8 5.9 6.1 beat, een Source: l'apk.rnik, E., 8re.,art. G., slnd Spita, N. (19861. Prevention de la naissance prentaturee.INSERM, 138. Table 2.7 Evolution of Preterm Birth (Hospital Data)

Year*

1 2 3

< 34 weeks 2.6 2.9 3.3 spontaneous 1.5 1.6 1.8 induced 1.1 1.3 1.5

35.-36 weeks 4.0 4.4 4.4 spontaneous 2.3 2.5 2.7 induced 1 .7 1.9 1.7

*Years 1, 2 and 3 generally represent 1984, 1985, and 1986.

Table 2.8 Evolution of Risk Indicators (Hospital Data)

Year*

1 2 Per ,ntage Percentage Percentage

Betamirnetics Use 21.8 20.5 17.5 1st visit during 1st trinwster 50.3 45.3 44..

(erani Bri.;art 21 3t) Detemiiitra The Etiology and Pretention of Low Birthuvight: iirent Kilowledge and Prioritia for Future Rovarch

MR :11AEI. S. KRAMER, M.D.

NTRODI. ICTION In l)oth developing and developed countries, low birthweight (LBW), that is, birthweight less than 2500 g, is a major risk factor for neonatal mortality and fin subsequent morbidity and impaired functkmal NI-tin-mance. The design of interventions to prevent LBW depends on a sufficient understanding of its etiokTic determinants. Although itis widely acknowledged that causality of LI3W is multifactorial, there have been considerable a mfusion tnCI conta)-- versy about which factorshave independent causal effects, as well as the quantitative importance of those effects. One of the mAjor reasons for this confusionand controversy has been a failure to distingthsh LBW caused by a Athwuces in the Prevention q' Lou' Birthweight short duration of gestation (i.e., premature efforts, both in terms of clinical and public delivery) from LI3W caused by intrauterine health policy and future research priorities, will growt h retardation ( It VTR ). differ in different settings depending on the rel- Prematurity is defined as a gestational age ative incidence of prematurity and It TGR. less than 37 weeks. IUGH. which is also referred Reported discrepancies concerning the eti- to as small for gestational age (SGA) or small for ology of LBW may have explanations other dates (SFD), is usually defined as birthweight than the IUGR -prematurity distinction. Primary less than the 10th percentile for gestational age among these is the matter of association versus for a "standard- reference population. (It is obvi- causation; that is, "markers- of LBW versus true ous from these definitions that babies can be causal determinants. premature or growth retarded without necessari- The issue here is not merely the theoretical ly weighing less than 2500 g. Thus, while itis concern t'or "scientific puritv.- Rather. the asso- true that LBW babies are either premature or ciation/causation distinction is of considerable growth retarded or both, the converse is not practical import for improving public health, necessar..y 1.1e case, particularly in well-nour- since an intervention will succeed in reducing ished populations from devel()ped countries.) LBW only if it affects a true causal factor. Many The distinction between prematurity and of the potential determinants are highly associ- It TGR is of crucial importance because they dif- ated with one another, and adequate control fer considerably not only in etiology, hut also for confounding is thus required to identify and in terms of their prognostic significance and quantify their causal effects. Two possible sta- relative incidence in different settings. tistical explanations for the reported discrepan- For example, premature infants are at des concern the portion of the birthweight or greater risk for developing hyaline membrane gestational age distribution (e.g., middle v. disease, apnea, intracranial hemorrhage, sepsis, l)wer tail) on which a determinant acts and retrolental fibroplasia, and other conditions inadequate statistical power in studies with relating to physiologic immaturity. Several of small sample sizes. these conditions are responsible for the far In an attempt t() help clarify these issues greater neonatal mortality in premature babies. and synthesize the recent literature,Irecently On the other hand, It TGR infants are more undertook a methodologic review and meta- likely to.raveI deficits in growth, and at least analysis of the English and French literature Jme of these deficits appear to be permanent. published between vro and 1984.'' Although As to relative incidence, Villar and Iklizan have several issues have been further clarified over analyzed data from 25 developing and1 I the last several years,I believe the results of developed regions.' In devel()ping countries. that review still provide a reasonable synthesis nu)st LBW births ippear to he due to 11.`( R. of our current state of knowledge about the eti- whereas in developed countries (especially ologic determinants of prematurity and I I 'GR. those with the k)west LBW rate), in()st are due The following section summarizes the methods to prematurity. Thus, the focus of preventive and results of that review. 36

26 The Etiologr (Ind Prelottion (y. Lou. Binbuvight Determinants

CURRENT KNOWLMGE reports. and so on. Each reference idemified by the search was examined for data concerning 1 Summary qf 1970-84 Review or more of the '13 factors. The 1970-84 review was restricted to sin- Methodokigic standards were established a gleton pregnancies occurring in women living at priori for studies of each candidate factor. These sea level without chronic illness, Factors Of included such general aspects of research de,,ign extremely low prevalence (e.g.. maternal rubella as definition of the target population and study infection or uterine malformation) were not con- sample, description of study participation and sidered. because even though such factors will follow-up rates, clear demonstration of the be of great importance to the individual women appropriate temporal sequence between the fac- in whom they occur. they are not responsible tor and outcome, and the use of an experimen- for a significant portion of IUGR and prematurity tal research design (where feasible and ethical). on a population-wide basis. Also excluded from The remaining standards pertained to the poten- consideration were medical complications of tially confounding variables requiring control for pregnancy, such as pregnancy-induced hyper- nonexperimental studies and differed according tension, abruptio placentae. placenta praevia. to the factor under assessment. and premature rupture of the membranes. In my Studies satisfactorily meeting (SM) or par- view. such conditions should he considered as tially meeting (PM) the standards were selected intermediate outcomes of pregnancy. Their for further analysis. SM studies fulfilled the inclusion in a multifactorial model of causaticin majority of the predetermined criteria. PM stud- will inevitably lead to an underestimate of the ies gave some attention to rigorous design and effect of factors whose impact on intrauterine analysis, but fulfilled less than half the criteria. growth or gestational duration may operate For several of the factors, certain standards through one of the conditions. were judged to he of overriding importance in After these restrictions and exclusions, '13 assigning an SM or PM rating. Based on the fact( rs (or groups of factors) were left for assess- studies selected for further analysis, each factor ment. These are listed in table 3.1. The literature was issessed for the existence of an indepen- search began by examining the subject catalogue dent causal effect on birthweight, gestati(mal of the World !leak h Organization library for age. prematurity, and 11'GR. If an independent books and mom)graphs published since1 (r0. A causal effect was judged to be denuinstrated on similar search was carried out for review articles the basis of the combined evidence of the listed in Index .1/edicus over the same time peri- selected studies, and sampling variatkm could od, These were supplemented by a MEDLINE be excluded as an explanation (Le_ p< MS), the computer search fig the years 1982-4;4. Finally. a effect size was extracted from each study, For "snowball- procedure was used, by which each birthweight or gestational age. this may have article or book chapter located was examined for taken the form of the difference in means from further references published since 1970, followed a rand()mized trial (g a matched cohort study, by the references cited in those secondary an adjusted difference derived from an analysis

Micbael S. Kramer 27 3 7 Advances in tbe Prevention of Low Birthuvigbt of covariance, or a regression coefficient from a tors. this total far exceeds 895, since many multiple linear regression analysis. For prematu- reports provided data on several factors. rity and ItTGR, the corresponding effet size Factors with well-established direct or indi- extracted was the relative risk or odds ratio rect (i.e., acting via one or more direct-acting adjusted for potential confounders by using factors) effects on intrauterine growth are listed matching. the Mantel-Haenszel procedure, or in table 3.2. Those with effects on gestational multiple logistic regression analysis. duration are contained in table 3.3. In order to The extracted effect sizes were then illustrate the quantitative importance of the weighted by the study-speci sample sizes to well-established direct determinants, I have yield an overall estimate for each factor. Finally, constructed pie diagrams in which the size of using available data on the prevalence of each the pie slices is roughly proportional to the eti- demonstrated causal factor in different popula- ologic fraction for each of the indicated factors. tion groups, etiologic fractions were calculated Figure 3.1 is the pie diagram for It1GR in a typi- for prematurity and It1GR. The etiologic frac- cal rural developing country in which malaria is tion (EV), which is also known as the popula- moderately endemic but pregnant women do tion attributable risk or attributable risk percent, not smoke. Nonwhite racial origin is probably is defined as the proportion of HIGH or prema- responsible fin a large proportion of IUGR in turity in a populatkni that is attributable to a developing countries with high prevalences of given factor. It is calculated as follows: black or Indian racial origins. The other major factors are poor gestational nutrition. low EF= RR-l) prepregnancy weight, short maternal stature, P(RR-I) + and malaria. Itis important to emphasize that. of the five leading facfiws, three may be mcKlifi- where P is the prevalence of the factor in the able in the short term: Gestational nutrition, population and RR is its corresponding relative prepregnancy weight, and malaria. risk or odds ratio. Figure 3.2 shows the pie diagram for a The literature search using the ccmibined developed country in which 40 percent of the snowball procedure identified a total of 921 w(mien smoke during pregnancy. In this setting. publications. Of these, 895 or 97.2 percent, the most important single factor by far is were successfully located and reviewed. cigarette smoking. This is followed by poor ges- Although a majority of the reports originated tatknial nutrition and low prepregnanc) weight. from developed countries in North America and The three leading factors are all potentially western Europe, a large number also came modifiable, once again, with obvious implica- from developing countries in Latin America, tions for public health intervention. Although India, Africa, and southeast Asia. The total the overall size of the pie is smaller than in the number of factor assessments carried out was developing country setting (i.e., the II1GR rate is 1,566. Even though some of the reports did not lower), a larger proporticm of existing fetal contain original data bearing on any of the fac- growth retardation may be preventable. 3° 28 The Etiokr and Prevention qt. Lore Birthweight Determinants

A pie diagram for the well-established crate sport and exercise may reduce it (at least direct determinants of prematurity in the devel- among well-nourished women indeveloped oped country setting is shown in figure 3.3. Of countries),' " but further studies are required. A the five factors with well-established direct Subcommittee on Dietary Intake and Nutrient causal effects, cigarette smoking and low Supplements During Pregnancy has been con- prepregnancy weight are modifiable but vened by the Food and Nutrition Board, account for a relatively small proportion of pre- National Academy of Sciences. to synthesize mature births. The major message from figure the information available in this area. This com- 3,3 is that the majority of prematurity occurring mittee slumid be publishing its report within in the population remains unexplained. In part. the next several months. [Editor's note: This this large gap in our knowledge reflects the far publication was released in 1990. Itis entitled less intense previous effon in studying gesta- Nutrition During Noma/icy. and is available titmal duration as compared to intrauterine from the National Academy of Sciences] growth. This has been especially true in devel- Another area of intense recent research oping countries. activity concerns identification and close surveillance of women at high risk for prema- Recent Developments. ture delivery. The assessment of high risk is Since this assessment was carried out, a based on multifetal gestation. uterine malforma- number of studies that bear on prevention of tion, inc( )mpetent cervix. or a history of previ- prematurity have been published in developed ous premature birth or second trimester country settings. Although I have undertaken no abortion. Surveillance has used such m(xlalities formal assessment of the literature a ppea ring as frequent pelvic examinations, teachingmoth- since 198-i. I would like to highlight three areas ers to recognize early signs of uterine contrac- in which research has shown promise for pre- tions.and,mostrecently.ambulatory venting prematuritv: (1) Reducticm of maternal tocodymunometry (i.e.. 11(mie numitoring of work and physical activity during pregnancy: (2) uterine contractions using a portable recording identification and close surveillance of women device). The evidence from intervention studies at high risk f(mr premature delivery: and (3) treat- involving these modalities is inconclusive,' 's ment of genital tract infection,/colonizatkm. but some of the results have been quite impres- The available evidence permits no defini- sive and certainly are sufficient to justify further tive conclusions as to whether work and physi- research.It should be emphasized. however. cal activity during pregnancy are beneficial, that even if effective methods can be devel- harmful. or irrelevant for gestat k mal duration. oped for preventing prematurity in women at Recent evidence does s ggest that prolonged high risk, the overall impact on a populaticm- strenuous or stressful v. ork activities, aerobic wide basis will depend on the predictive value endurance-type exercise continued into the of the risk assessment and on the proportion of third trimester, and upright posture may all premature births that occur in women classi- increase the risk of prematurity, and that mod- fied as at high risk.'"

Michael S. Kramer 3a Advances in tlw Prevention (y. Low Birthweight

Finally, sufficient evidence exists concern- assignment remains unpredictable hy either the ing the possible etiokTic role of bacterial 'agi- subject or the investigator. It also tends to result nosis and genital tract infection or cokmization in groups which are similar for both known with such organisms as Ureaplasmaure- and mknown susceptibility factors that could alyticum. akunydia itathomatis, GardnervIla otherwise confound the treatment effect. Thus, vaginalls, species of 13actervides. and other randomization of individual subjects is a plime anaerobes to justify randomized antibiotic trials desideratum for a methodologically sound clini- in women colonized with these organisms in cal trial. For some types of interventions. how- the second and early third trimesters.'" ever, random assignment by individuals can Several such trials are already in progress actually be detrimental because interaction in the United States. including a large multicen- between subjects may lead to consequent "con- ter study supported by NICHD. and the results tamination" of the interventicm, and, hence, a are eagerly awaited. Recent evidence also indi- biased comparison. cates that genital tract infection:colonization in For example. randomizing half of the women with preterm labor is associated with women within a given work setting to a change failure of conventional tocolytic therapy."'- in posture or position during work (e.g., by Further trials should be undertaken of broad- allowing them to sit rather than stand while spectrum antibiotic treatment as an adjunct to performing certain tasks) or to a shortening of routine tocolysis in women with preterm labor the work day is likely to lead to a "spill-over" or rupture of the membranes. effect to the other half of the women working in the same setting who are randomized to the control intervention. The result would be a diminution of the difference in the treatments GENERAL 1\1iMiol)01.(x;ic Ism 'ES received by the to groups of women and the Since a major focus of this symposium is possibility of a false negative result. In such sit- intervention studiesand. particularly. rand(mi- uations, where relatively closed. naturally ized clinical trialsaimed at reducing the rate f(gmed groups are capable of imxlifying the of prematurity li.sd IUGR. Iwill conclude this intervention allocated to individuals within paper with a summary of several key method- those groups. group randomization appears ologic issues that apply generally to such trials. preferable to the randomization of individual These issues are: Individual versus group ran- subjects. domization, blinding and unblinding. U ie effect Unfortunately. group allocation carrit:s of participation on outc()Me. selective subject some hazards mn. inherent in the individual participaticm. and c(mipliance. approach. Since individuals within a group cannot necessarily be regarded as independent Iiidiiicliial11.1NUS Group Ralulowizaliou fiom one an()ther. the effective sample size is Random allocation of treatment to each reduced by the extent of within-gnmp depen- subject maximizes the likelihood that tieatment dence. Ideally, the solution to this problem is 4 0

30 71.7e Etiology enui 'mention'Low Birtinveigbt Determinants to base the statistical analysis on the number of Mothers knowledge of the group to which work settings (e.g.. companies or factories) they have been assigned can then lead to a rather than on the number of individual biased treatment comparison. This is especially women. Because of the large number of work true when the comparison is one of interven- settings required in such studies. however, this tion versus no intervention. Mothers who strategy will often be infeasible. receive the intervention will obviously he A more practicable approach would aware that they are among the "chosen.- and involve the use of a pre-intervemion study to this awareness may create a feeling of special- demonstrate that individuals in two or more ness that can exert a profound (and uncon- different work settings experience similar birth- trolled) placebo effect. weight and gestational age distributions before Although subjects in these types of inter- institution of the intervention. Equivalent pretri- vention studies cannot be kept blind to the al results increase the plausibility that any dif- actual intervention they receive. they can and ferences in outcome that occur when thoSe should (when ethically defensible) be kept work settings are exposed to different interven- blind to the interventions being compared and tions during the actual trial are attributable to to the study hypothesis. Placebo-type control the intervention, rather to potentially c(mf(aind- interventions can be designed in ways to facili- ing differences between the different settings. tate this type of blinding. For example, women The randomized interventkms could even be enrolled in a trial to assess tIk' effect of reduc- introduced in selected settings sequentially mer tion in maternal energy expenditure could be time so that efficacy could be evaluated yia asked to participate in a study of work before and after comparisons. as well as con- changes. Changes in work activities would he current comparisons between interventionand implemented in both intervention and control control work settings. gr()ups. with the former changes f()cusing energy-reducing maneuvers and the latter Blinding and I 'nblinding involving other changes not expected to reduce Blinding of study subjects and observers maternal energy expendiwre. In this example. (i.e., double blinding) is highly desirable in blinding would be relatively easy to maintain. placebo-contmlled interventions involving as long as the two study groupsworked in dis- drugs. Such a design is clearly feasible for such tinct gow,raphical settings. interventions as placelx) controlled antibiotic 1:11ect Participatiwi (al Outoinic trials.Itis obviously infeasible. however. for reduction in physical activity or ambulatory SRIChy participation can change behavic >r. monitoring of uterine contractions, hn the lat- even in the control group. and the behavioral ter types of interventic ns. observers involved in change itself may affect the outcome under assessing gestational age and birthweight can study (this is known as theI lawthorne effect). usually be kept blind. hut the women them- It seems likely that the feeling of spedalness selves cannot. attendant upon participation in an intenention

Mic..hael S. Kramer Advances in the Prevention cf Low Birtinveigbt study might itself havc a beneficial effect on under study, can affect the susceptibility to certain outcomes. Study participation might, for develop IUGR or prematurity. Socioeconomic example, serve as a fOrm of social support and status, age, parity, height, prepregnancy weight, thus mitigate the effect of stress, thereby reduc- cigarette smoking. and alcohol consumption ing the risk of premature delivery. Because the are examples of the types of characteristics that potential magniwde of the benefit of the inter- should be considered. vention may be limited (the ceiling effect), the end result may be either a smaller treatment Compliance difference or no significant difference, and thus Many of the interventions currently under a false negative result concerning the effective- investigation for prevention of low birthweight ness Of the intervention. involve some effort b study women to comply If possible. investigators should attempt to with the assigned intervention. To the extent keep study subjects unaware that they are that women do not comply, differences in out- being studied, or at least unaware of the main come between the intervention and control outcomes being ctnnpared. For most types of groups will be correspondingly reduced. Once interventions, however, keeping study mothers again, this can lead to false negative inferences in the dark will not he ethically defensible. In about the intervention's efficacy. Investigators those cases, the potential for a Hawthorne should routinely include strategies for stimulat- effect should he acknowledged by the study's ing and maintaining subject compliance, as investigators, and inferences should be mitigat- well as measuring its extent, Monitoring of ed accordingly. compliance could involve pill counts or tracer labels detectable in urine for interventions Selective Sulyect Participation involving medication, or onsite observations for In general, participation rates tend to be such interventions as reduction in maternal lower in intervention studies than in observa- work .physical activity or recording of uterine ti(mal studies. Although statistical power can he contractions. maintained by approaching additional Wo men. The Indirl statistical analysis for any of those who agree to participate may be quite these intervention studies slumld be based on different from those who do not. L'nf()rtunately, the intenticm to treat. that is, a comparison of therefore, the study's findings may not he gen- all women (or groups or other units of rand(mi- eralizable to all mothers who are eligible to i7.3' randinnized to the IWo) study groups. receive the interventkm in the real world. not just those who comply. A secondary analy- Investigators should make every effort to sis comparing those subjects with good compli- keep track of, and include in all resulting publi- ance, however, can suggest what the potential cations, both the numbers and relevant charac- biological efficacy of the intervention is and teristics of all women who accept and decline point to efforts required in future trials to participatiom. The characteristics of importance achieve the c()mpliance necessary to (kmum- are those that, independent of the intervention strate overall treatment effectiveness. 4'

MC Etiology cord Pretottion tow Birthuvight Determinants

PRIORMS FOR Ft TURF RIMARC:1 lenges. A randomized trial designshould be Future intervention studiesshould focus used whenever feasible andethical. As I have of human On modifiablefactors of potential quantitative discussed, however, the vagaries importance for which the evidenceof causal behavior, especially during a period aspsycho- impact on prematurity or IUGR isinconclusive. logically sensitive as pregnancy, canaffect trial In developing countries, theemphasis should participation, compliance, blinding,and out- be on preventing HIGH. Based on current come. Thus, astudy's use of this design does knowledge, one promising intervention con- tun necessarilyconfer certainty on its conclu- cerns reductionof strenuous maternal work sions. The planning, execution,and analysis of during pregnancy, particularly in womenwith intervention studies to preventlow birthweight poor prepregnancy orgestational nutrition. provide great scope for creativescience and Improved prenatal care tin termsof timing, fre- high hope fin improved publichealth. quency, or c(mtent)should he another high-priority target. Because a few studieshave shown ACKNOWLEIX;MENT a beneficialeffect of folk acid supplementa- From the Departmentsof Pediatrics and of tion, further randomized trialsin women with Epidemiology andBiostatistics,McGill fokue-deficient diets should also heundertak- University Faculty of Medicine.This work was en. Although theavailable evidence concerning carried out while Dr. Kramer was aNational iron deficiency and ;memiaindicates no signifi- Health and Research Scholarof the National cant effect on intrauterinegrowth, additional Health Research and DevelopmentProgram, trials of iron supplementation inanemic or I lealth and Welfare Canada. iron-deficient women may he necessary torule out such an effect.Other intervention trials REFERENCF.S might involvc supplementationwith other vita- I.McCormick. NI. C.(1)8i).The eontributhin of chikflu)od m( r- calcium or zinc) hirthweight to infant numality and mins or trace elements (e.g., bidity. NewEngiancilaurna/crrtledicine 312.82-90. and reduction in exposure toindoor smoke. 2. Villar. J., and Belizan. J. NI.(1)82). The timing fac- In developed countries,future research tor in the pathophysiolc)gyof the intrauterine grcm-th should focus on preventing prematurity.The retardation syndrome,0/),stetrica/ and Gyne- most promising avenues to pursueat this time CtibigiCal Stinky 37. 499-5(K). Determinants of low birth- include reduction in strenuous orstressful .4. Kramer. M. 5.(1)8'). weight:A methodological assessmentand meta- monitoring maternal work and physical activity, analysis.Bulletin (il Ibe world uvula)(n:iallization of uterine contractions in winnen athigh risk 65,063-73-. for premature delivery, and treatmentof Kramer, NI. S.(1987).Intrauterine growth and ges- Pediatrics80, women withbacterial vaginosis Or colonization tational duration determinants. 502-511. by certain genital tract pathogens. Ifolford, T. R.. and developed coun- Berkowitz. G. S.. Kelsey. In both developing and lierkowit7. R.I.(198.4l. Physical activity and the tries,these researchprioritiespresent risk of sp(mtanyous pretenndclivcry.journal (Y. formidable methodologic andpractical chal- ReproduCtire Medicine 25. S81-S55.

.33 ,,Atichael S. Kramer Advances in the Prevention qf Lone Birthweight

C. tarret. J. C.. and Spellacv, NV. N.(1983). jogging dur- birth by ambulatory assessment of uterine activity:A ing pregnancy:An impuwed outcome?obstetrics randomized study,American lwirnal of Obstetrics and o.'www,logy 61. 705-709. and (iyilecologr 150,536-543.

7.Clapp. J.F., and Dickstein. S.(1984).Endurance 18.lamns.I.1)., Johnson, F.F., O'Shaughnessy, NV.. and exercise and pregnancy outc(nne.Medicine and West, L. C.(1987).A prospective random trial of Science in Sports and Evercise10.556-562. home uterine activity monitoring in pregnanciesat 8. Mainelle. N.Laumon. B.. and Lazar. P.( 198.1). increased risk of preterm labor.American journal (f Prematurity and (x.cupational activity during pregnan- obstetrics and Gynecolow 157. 038-643. cy. American Journal fpidemiology 119. 3(19-322. 19, Main. D. NI.. and Gablv. S. G.(1987).Risk scoring 9.Saurel-Cubizolles. M..L. Kaminski. M., Llado-Arkhipolf. t or preterm labor:Where do we go from here? .1.. du Mazaubrun. C.. Estryn-13ehar, M.. Berthier. C.. Amerawn journal qf Obstetrics and Gynecology 15-. Mouchet. M.. and Kelfa. C. (1985). Pregnancy and its 789-793. outcome anuing hospital personnel according to o(cu- 20. Graven, M. Nelson. H. P.. Deliciiien. Critchlow, pation and working conditions.ournal cif C.. Eschenback. D. A.. and Holmes. K. K.(1986). Ephloniology and COMMUnity Health 39. 129-13.i, Independent associations of bacterial vaginosis and 10. Stengel, B.. Saurcl-Cubizolles. NI, J.. and Kaminski, NI. Chlamydia tnichonuitis infection with adversepreg- (1986).Pregnant immigrant women:Occupational nancy outcome. Journal of tiv American Medical activity. antenatal care and (111ICCM1e.Intern(4ll(Mal zissociation 256. 1899-1993. Journal +)./EpidemPdoj 15. 533-5$9. Lamont. R.F.. Taylor-1i( )hinson, D., Newman, NI., 11. Zuckerman. B. S.. Frank, 1). A.. I hngson. R., Morelock. Wigglesworth, .1.. and Elder, NI. G.(1986).sponun- S.. and Kavne. II. L.(1986). Impact of maternal work e(nis early preterm lalunir associated with abnormal outside the luime during pregnancy oil neonatalout- genital bacterial cc )l on zatic in.British Journal of (ome. Pediatrics:7.45'9-464. Obstetrics and C.,:ynaccology 93.804-8B). 12. Saurel-Cubizolles, NI. J.. and Kaminski, NI.(1987). Berman. S. M., Harrison. li. R.. Boyce. W. T., flaffner. Pregnant women's working conditions and their W. J., Lewis. NI.. and Arthur. J. B.(198-). Low birth- changes during pregnancy:A national study in weight, prematurity. and ilostpartum undometritis: France.Bitish finirnal of Industrial Medallicft. Association with prenatal cervical Mycoplasma hoini- 230-243. nis and chlanlydia tmchomatis infecticms, Journal .71. the American Medica1.-Issociation 25. 1189-119 i. 13. Mamelle. N.. And Munoz. F.( 1987 ).Occupat ion.11 working conditi( ns and preterm birth :A reliable scor- 23. Manius..l.. Krohn. NI. A..111111er. S.L., Stamm. W. F.. ing system.American journal (f Epidemiology 12ft Holmes. K. K.. and Eschenbach. D. A.( 1988 ). 150-152. ReLnionships of vaginal factobacillus species. cervical Chlanirdia trachomatis. and bacterial vaginosis to 14. Grunebaum. .A..Nlink(iff.II.. and Blake. I).(1)8-). pretenn birth. oNtetrics and Gynceidogy 71. 89-95. Pregnancy am( ng ()bstetricians:A (.i)mparison ).4. births before. during. and after residency.American ..perimg, R.S., Newton. E.. and Gibbs, H. S. hurnal (Y. (thstetrics auI 6.17/(Vgihigr I'S-. "9-83 Intraamniotic infection in low birthweight infants. ./(mrual of Infrctious niseaseN I 5. 113-117 15. Herron\I KatzNI.. and Creasy. R.K. ( 1982). Evaluation Of a preterm birth fvcvention prograill:A 25. Potkul. R. K.. Moawas. A. II., and Polito, K. I,.(19$5). preliminary report.Obstetrics and Gyneculugy 59. The ass( nlatiun suhchnkAI infection with preterm 452-456. lain*:The role of C-reactiye protein.American Journal cy'(71i8letrics and Gynecolifrgy J53.6 +2-6 +5. 16. Papiernik. E.. Bouyer. J.. Dreyfus. .1.,( an, 1),, Winisdorfter. Guegen, S. Lecomte. M.. and Li /:1 r. 26. McGregor. .A.. French..1L. Heller. L. B., 'nick!. J. K.. P. (1985).Prevention (if preterin births:A perinatal and Mak()wski. F. L.i 1,N6), Adjunctive ervtlininIVCin study in ilaguenau. France. Poliatrics -0.151-158. treatment tor idiopathic preterm labor:Results of a randomized. double-blinded, placebo-controlled trial. 17. Morrison..1. C.. Martin. J. N.. Martin. R. NX... Gookin. K. American ournal ofCibstetric. and Gynecolugy 14; f, S.. and Wiser. W. 1..(1987).Prevention of preterm 98-103.

4 4 $4 The EtiolQgy and Pretentiwt Lue Birthue0.7t Determinants

Statistical Findings: The Etiology and Preventionof Low Birthweight

Table 3.1 Factors Assessed for independentCausal Impact on intrauterine Growthand Gestational Duration

A. Genetic and ConstitutionalFactors

Maternal Prepregnanc y Weight; Maternal Hemodynamics Infant Sex' Racial/Ethnic Origin* Maternal f leight' Paternal Height and Weight* Additional Genetic' Factors

B. Deniographic and PsyChos00.31Fac-tors

iAaternal Psyc holcigic Uact )rs Maternal Age Socioeconomic Status Marital Status

C Obstetric factors

Intrauterine Growth and Gestational Durationin Prior Pregnancies Parity" Birth or Pregnanc y Interval Sexual Ac tivity Prior Stillbirth or Neonatal Death In Uter() EXpOtillre toDiethylstilbestror'' Pnor induced Abortion Prior Intert Prior Spontarwous Aborticm"

O. Nutritional Factors

Energy Expenditure, Work, arid l'hysk-al Activity Gestational %Veight Gain* Vitamin B, Caloric Intake" Folic A. id and Vitamin B.. Calcium, Phosphorus, and Vitamin D Protein Intake/Status Iron arid Anemia Other Vitamins and Trace Elements Zinc and Copper

E. Maternal Morbidity DuringPregnancy Genital Tract Infection General Morbidity and Episodic Illneso Nialaria* Urinary I rac t !rite( non

ial'ottioi( prepre,gnanc V Wetgill, pahVihilhVight and tteight, f gablishey/ drretdelern»nants or int rair141 grrni h U7I lode imam! .4e, r.n malaria.igarettc smoking, alcohdl con- motyrnal height and 4t 4.4,071 panty. ,nriorI 81V. ,t4t...lationd I%%eight gain a I oric intake, .14(./leral morbiditv, surnptidn,and folio( 0 ( ing t Estahlished indirect determlnants intrauter int, et:rots fir? rniIthit,maternal age and mg ith% oncnrr( status. s vir;ht in idprtmlturd% prror spuntanvoirs abor- aosal 1.1710,1( t mslat dna/ duration in( ludo pIpTi,t).)f, 4; rautors tiith is ell-estalrlistrcst drrec t nt titrro diethvIstillx-strol cpo.urc.and cigarette smoking, matt.rnal age and sin toe( noir/nit status, rat tors v%ith tell-e.tablis"lcd inchrec tcausal rit1pa4 I or? gcst,rt road darati(ur Inc lud(

.1 ichael Kramer 4 5 Advances in the Prevention of Low Binhweight

Figure 3.1 Relative Importance of Established Factors with Direct CausalImpacts on 1UGR In a Rural Developing Country with Endemic Malaria E.1 General morbidity A.6 Small paternal size, other

A.7, C.4 Maternal LBW and prior LBW history A.2 Non-white race

0.1, D.2 Low caloric intake or weight gain

A.3 Short stature A,4 Low prepregnant weight *KO 151212

Figure 3.2 Relative Importance of Established Factors with Direct Causal Impactson IUGR In a Developed Country Where 40 Percent of Women Smoke During Pregnancy

E.1 General morbidity Other

A.7, C.4 Maternal LBW and prior LBW history

A.2 Non-white race

A.3 Short stature F.1 Cigarette smoking

A.1 Female sex

C.1 Primiparity

0.1, D.2 Low caloric intake or weight gain A.4 Low prepregnant weight wt./01151M

36 ,S7atistical Findings: Me Etio/ogy arid Prevention (f Birthweighi Determinants

Figure 3.3 Relative Importance of Established Factors with Direct Causal Impacts onPrematurity in a Developed Country Where 40 Percent of Women Smoke During Pregnancy

F.1 Cigarette smoking

tr I VV VV r C.4, C.5 Prior PT and VV . . spontaneous abortion

s .. OPP 0 NN,,*.N.`.". A.4 Low prepregnantweight s- S K S

C.9 In utero DES exposure

VVVV VVV

Unknown VVV

*WO SS122.

Michael Kramer .3 7 4 Strms ai id Siipport Ditritig Pregnancy What Do They Tell Us Abolit Low Binthueight?

EANNE BROOKti-GuNN, Pita

INTRO1M rnoN The remarkably high incidence of lowbirthweight births in the I Tnited States over the past 25 years remains vety much apuzzle. While timely antenatal care is assodated withbetter birthweight outcomes, the increases in the proportion of women receiving early care inthe last two decades have not resulted in comparable declines in the low birthweight rate. In additkm, race difkrentials in the proportionof low birthweight have not changed duringthis historical period (there is a 2:1 ratio between blacks and other racial and ethnic groups).''' It is estimated that perhaps 40 percentof the valiance in low birthweight births isdue to environmental and psychc)social factms (The Institute of Medicine 1985 reportidentified health behavior, psychosocial andphysical stress, demographicconditions, and toxic expo-

4S Advances in tlye Pretvntion cd Low Birtbweigbt sure as principle nonmedical risk factors). To Five issues are addressed in this chapter: solve the puzzle, then, we need to know which (a) the commonly held belief's about psychoso- environmental and psychosocial eventsare cial contributions to poor pregnancy outcome, contributing to low birthweight; how physio- particularly as they relate to anxiety, stressful logical, enviammental, and psychosocial condi- life events, and social support; (b) the associations each contribute to pregnancy outcome in tion of these psychosocial factors with health a relative sense; and what the links might be behavior; (c) the importance of considering the between environmental events or psychological context in which these health behaviors and states and subsequent physiological changes psychological factors occur, especially with leading to early labor. regard to ethnicity and social class; (d) theuse- The focus in this chapter is on which psy- fulness of different models for studying the chosocial events are important, rather than how effects of health behaviors and psychosocial they actually might initiate early laboror what factt)rs simultaneously; and (e) the implications they contribute over and above medical factors. of both beliefs and conceptual models for the Direct proximate links between psychosocial types of preventative programs that have been factors and actual physiok)gical consequences and might he initiated. have not been studied in great detail.`' Most of the studies conducted to date center on health behaviors whose affects are believed to he cumulative in naure (e.g., smoking, postponing Siiiss AND St xlIAI. SI l'PORT: antenatal care, and eating poorly); psychosocial EFFECTS ON PHR ;NANCY 01 TCONIF. states that are thought to influence the preg- The following assumptiims permeate the nant woman over a relatively long period of popular and clinical literature aboutpregnancy: time (e.g., anxiety, depression, and stress experienced due to negative life events); and envi- Anxiety and emotional distress during preg- ronmental events that are chronic (e.g.. nancy result in poor pregnancy outcomes. poverty, low level of education, and crowded The occurrence of negative life events, and living conditions). Since the human fetus is the stress associated with coping with such believed to be relatively protected (at least pri- events, negatively influence a pregnancy. mates seem to be less influenced by negative Social support acts as a moderatt )r. or'aS environmental conditions than are rodents),"Th continuous exposure to a negative condition is buffer, from the untoward effects of stressful thought to he necessary for a poor pregnancy life experiences and emotional dysfunction. outcome. Indeed, the possible effects of short- Ernotkmal dysfunction and stress have direct term stressors in addition to chronic stressors or effects on pregnancy outcomes ( via some as multiple negative life events on the long-term vet not-well-specified physiological path- mental and physical health of the pc)stpart urn ways) over and above health behaviors. human are just beginning to be studied."'" such as smoking, nutrition, and exercise.

4 40 .Sb.ess and Support During Pregruincy Determinants

Persistent race differences in neonatal out- vidual's initial determination that an event is come are accounte(I for in part bythe fact indeed stressful, thereby helping the individual that blacks arc more likely to live in poverty, cope successfully, so that the event is not per- have lives that are more stressful, have expe- ceived as stressful (see figure 4.1). In addition, riences that are more difficult and anxiety- social support may alter the individual's behav- producing, and have fewer available social ioral or physiological response to an event supports (both f)rmal and informal). actually pe-ceived as stressful. Support may help individuals regulate health behaviors, Results linking obstetric and neonatal Out- lessen their responses to stress, or help solve come to emotional states, negative life events, the problem.".'" and social support will be reviewed separately.' It is important to realize, however, that all Anxiety and Emotional Distress three factors are associated with current notions The idea that emotional upset is associated about stress. The physiological effects of emo- with pc)or pregnancy outcome has a time-hon- tional distress or upset have been document- ored history. Folk wisdom has linked stress in ed."' Often, emotional upset is presumed to pregnancy to abnormalities in neonates as be caused by negative life events and amelio- described in the Old Testament, by Greek rated by social support. physicians, and in the writings from the Middle Stress is often considered to occur when Ages to the present:1- Currently, associations an individual is confronted with an event (a) between emotional states and obstetric and that is perceived as threatening; (b) that pregnancy outcome have been examined, with requires a novel response; (c) that is seen as a number of these studies being reviewed important (i.e.. demands a response); and (d) recently by Istvan and earlier by McDonald.'8' for which the individual does not have an Typically, this literature has focused on anxiety, appropriate coping response available."Stress and, to a lesser extent, on depression, anger, has been associated with physical illness as and hostility. In the 1950s and 1960s, nonstan- well as negative emotional states. Stress is dardized measures of anxiety were often used, believed to influence illness via changes in making itdifficult to make across-study com- neuroendocrine functioning, immune system parisons. In addition, most studies have responses, and health behaviors. Thus, stress focused on the more trait-like (i.e., presumably might influence pregnancy outcome directly (in more stable and enduring) aspects of anxity terms of physiology) or indirectly (in termsof and emotional upset, rather than on the more health behavior). situation-mediated aspects of anxiety (i.e., pre- It has been hypothesized that social sup- sumably less continuous or enduring). This is port ameliorates these effects at different points surprising, given that the folk wisdom usually between the appraisal of an event as stressful mentions emotional distress cau:.,ed by a specif- and a physical illness. As illustrated by Cohen ic situation and that theories such as that of and Willis," social support may alter the indi- Seyle consider short-term stressors as well as

Jeanne Brooks-Gunn 41 Adivnces in the Prevention cf Low Biribueight long-term adaptation to difficult situations. evidence for the commonly held belief, and 10 Finally, some (but not all) studies are prospec- do not. tive in nature. In retrospective studies,itis Turning to neonatal, rather than obstetric quite likely that women with preterm infants outcomes, the folk wisdom is comparable. will reintopret past events or emotional states Indeed, the assumptions about links are sub- as having been more stressful.'" No studies stantiated in the nonprimate literature. In most have followed a sample of women prior to (but not all) studies of nxients. emotional stress conception tluough pregnancy and delivery. (as inferred by environmental conditions such Thus, no baseline data exist with which to as crowding, and exposure to light, heat and compare levels of anxiety during pregnancy. or noise) is associated with poor neonatal out- prepregnancy differences between women who e.nue (i.e.. prematurity and low birthweight). report high and low levels of anxiety during The human literature is less etmvincing. Only their pregnancies. Even studies of changes in looking at prospective studies, three have anxiety throughout the pregnancy are rare. reported positive associations' "and eight With these caveats in mind, what do the have not."-- All of these studies used studies (focusing primarily on the prospective birthweight or Apgar scores as their neonatal ones), tell us about the belief that emotional outcomes. upset affects poor obstetric outcome? The two In brief. the commonly held beliefs about studies conducted in the 1950s-'"-'' did not sup- maternal emotional upset, at least when powt the thesis and indeed, if anything, provid- assessed as anxiety. arc not validated by the ed evidence of the opposite (that is, anxiety current literature. While more support exists for was associated with better obstetric Outcome). links with obstetric than neonatal outcomes. Of the nine studies reviewed lw Istvan- from the most recent studies using standardized the 1960s, seven reported sonne evidence of link measures of emotional distress have not always between anxiety and poor obstetric outcome supported the conclusions reached in the and two) did not.' '' BY the 1970s. when more obstetric outcome literature of the 1960s. standardized scales were being used with greater frequency, of six studies (excluding Stre.s...%/tr/ Evotts those with very small samples), two) provided The occurrence of many life events is confirmatory"and lona had discomfirmakwy believed to constitute a stressor with implica- findings.'1- Two) studies from the 1980s found tions for emotic Ma 1 and physical well-being. no significant direct links.-Of particular inter- Indeed, events such as the illness or death of a est is the study of Beck and colleagues:- who spouse, child, sibling, or parent. as well as the assessed state rather than just trait anxiety and occurrence of many events simultaneously or in found no associations with obstetric outcome a short period of time, have been associated during the third trimester. A measure taken just with many illnesses, either as an etiological or before delivery, ho )wever. was associated with an exacerbating factor.'=" Recent work has doc- length of labor. Thus, of 19 studies. 9 report umented changes in immunological and corti-

Stre.s-s and Sunxin During Pregnano. Determittants sol functioning in the aftermath of negative two reported associations between negative life events." "' In addition, the occurrence of multi- events and prematurity: In brief, the litera- ple life events as well as chronic stress often ture is inconclusive with respect to associations associated with some life events and conditions between negative life events and poor out- has been shown to place individuals at comes.- increased risk for decrements in well-being."''.'" A huge body of literature exists exploring the Social Support conditions under which multiple life events are The availability and use of s(icial Support likely to be detrimental and to buffer the indi- buffers the individual fr(nn the possible delete- vidual ftom untoward effects."''."'' rious effects of negative life events.-" Indeed, Research attempting to demonstrate links some research validates this claim, even though between negative life events and obstetric out- questions have been raised as to whether what come is not as extensive as that on emotional is being measured is a direct effect of life distress, possibly because most of the theoreti- events upon social support or an interaction of cal work on life events is fairly recent. Of the events and support.'" In addition, social support six studies reviewed that looked at obstetric may affect psychological state (and, by infer- outcome, results are tr IA)oking at "situa- ence, physioh)gical state) directly. rather than tional" stresses (perhaps the precursors to neg- thrtmgh interactkm with life events. ative life events). Grimm and Venet found no It is commonly believed that social support associations with obstetric outcomes.' In the protects the pregnant woman or neonate from 1970s. several studies found positive associa- the untoward effects of stressful life events or tions. but only when considered in relation to emoti(mal distress. Perhaps the most often social support.' " The findings from these quoted study is that of Nuckolls. Cassel. and three studies are difficult to interpret, however. Kaplan:2 in which the occurrence of negative and caution sh(mid be taken in considering life events and the availability of social support them as confirmatory.- In addition. two recent were assessed. While neither was associated studies either reported no association or a with obstetric outcome, of the women who had "marginal" one:8" a number of life changes. those with low social In four prospective studies focusing on support were more likely to have obstetric neonatal outcome, an association between neg- pr()blems than those who had high social sup- ative life events and neonatal status (i.e., low port. This study has been widely cited as evi- birthweight or premature contracti(ms and:or dence f()r the importance of social support for deliveries) was fOlind tfltwo:1'''' a mediated buffering the pregnant woman against the pos- effect was found in the third, where life events siNe negative effects of stressful life events. As were associated with a -sense of permanence," Istvan points out, however, the result was which in turn was negatively associated with based on a difference of five cases between the neonatal problems:and no effect was found two groups. and few studies have appeared in in the fourth." In three retrospective studies. the literature that have replicated the original

eanne Brooks-Glitill Advances in the Prevention of Lou' Birtbweight

finding. In one, which was a retrospective women in Harlem who w...re first seen during study, an interaction between life change and their first or second clinic visit, a somewhat dif- complications was found; specifically, high life ferent tack was taken. No differences in the changes and low social support were associat- proportion of low hirthweight births were seen ed with neonatal problems (low life changes for schooling, work, maternal age, or psychoso- and low social support were inexplicably asso- cial factors such as stressful events, social sup- ciated with labor and delivery problems, how- port, and emotional functioning, while smoking even.'" In another, duration of support was and adequacy of care were associated with low associated with neonatal complications, hot h birthweight. Indirect effects were found, how- directly and indirectly.' ever, as smoking status was associated with the women's adequacy of prenatal care, stressful events, sodal support, mental health, and prior adverse pregnancy outcome."4"' Current smok- HEALTH BEI iAVK)R AND lit.FSS ers were less likely to have adequate care in In the discusskm to date, and in almost all the current pregnancy and more likely to have of the studies just reviewed, no consideration experienced :I prior adverse ne(matal outcome has been given to possible associations than nonsmokers. They had experienced fewer between psychosocial factors and women's negative life events than nonsmokers; however, health-related behavior. The implicit premise is they were less likely to he Hying with a hus- that they affect pregnancy outcomes directly, band or boyfriend (see Table +I ). via physiological processes related to stress. An In brief, previously documented associa- alternative possibility is that they play an indi- tions between psychosocial factc)rs and birth- rect role. operating thniugh health behavior, wcight may 1 x mediated by health behaviors. That is, enu)tional distress, social support. and That is, social support and emotional function- negative life events might influence women's ing ma% influence health behavi(Is such as compliance with health regimes, or at least may smoking and prenatal care, which in turn influ- co-occur with p;trticular health behaviors. For ence pregnancy outcome. In acklitkm, a differ- example. in the study by Newton and I lunt. ent mix of psvchosocial factors may he relevant life events were associated with low birth- ft ir different health hellaVit Ks: fur eX:1,111i ile.ill weight, hut this correlation was not significant the a hove-ment lolled study, the psychosocial after contr()Iling for cigarette smoking. In correlates of timing of antenatal-care onset were another study demonstrating a link between life stimewhat different than those f(ir smoking. stress and prematurity, the mothers of preterm infants were more likely to smoke and drink than the mothers of term infants (unfortunately, ATRIBI )Ns w En Ni( rry AN/ ) Pc wERTy the effects of substance use in mediating the original association were not examined). .= Thus far, ethnicity and poverty have mq In a study of 500 disadvantaged minority been ccnisidered, even though both are known

-i4 Stre.ss and Support During Provtancy Ikkytnittants to be associated with low birthweight births al status, and ethnicity (Hispanic and black). and both form the context in which a number Thus, the risk associated with these factors may of psychosocial conditions occur, For example, be more generally due to poverty status. in the occurrence of negative life events is associ- additkm, few or relatively weak associations ated with emotional and physical well-being:'" between birthweight and psychosocial factors In addition, undesirable life events are more were found. This may be because poverty con- prevalent in the economically disadvantaged, fers increased risk for lx)th environmental stress the poorly educated, the female, the young, and pocw pregnancy outcomes," and the unmarried."'' Not only are certain A final approach involves the possibility of groups likely to experience negative events, interactions between sociodemographic factors but these very same groups are more vulnera- and health behaviors. In one study, black Ne to them: that is, they will be more likely to mothers with preterin deliveries had more med- exhibit psychological disturbance when con- ical and social conditions than those without fronted with negative events, as Kessler has preterm deliveries, but the strongest medical suggested.7' predictor, hemocrit values, was associated with Such vulnerability may be due, at least in social risk factors (e.g., being unmarried, a part, to the unavailability of reurces, as teenager, on welfare, or having a low level of demonstrated elegantly in several studies, educati()fl). The authors suggest that combina- where the joint occurrence of many stressful tions of medical and sociodeinographic factors events and low social support are more predic- are most likely to be predictive of preterm tive of psychological distress than either one is delivery and that models including such combi- separately:'-` One study has demonstrated this nations are more likely to reduce the black- interaction in pregnant women: high support white differences than models of socio- and personal control reduced the impact of dem()graphic conditions alone.--. negative life events significantly.Another approach to considefing the importance of race and poverty as possible mediators, or as code- Mo Drts To STil». terminants of psychosocial functk ming, consid- ers high-risk populations.Demographic Psycitc)s( VIAL FACR )RS AN!) I UAL-1i I BEI IAVIORS characteristics, such as race, age, and educa- Itis imp(mant to consider the conceptual tion, and risk factors on a population-wide models that are implicit in the studies conduct- basis, may lose some of their discriminatory ed to date. Indeed, one of the slumcomings of power within low-income groups. That is, the literature is that simple, direct-effects mod- increased risk may be confounded with disad- els are usually tested, even though the 1,he- vantage. For example, in the Harlem study, no 1101IIC11011 under study is bcuer representedby associ!fons were found between birthweight mediated or indirect-effects models. We also and the traditional demographic risk factors need to consider whether relations are best such as maternal status, education, occupation- characterized as linear, cumulative, or threshold.

Jeanne Bruoks-Gunn 45 Advances in the Prevention of Low Birthweig1.7t

At least five models are relevant (see S. The fifth in()del differs from the f(mrth Figure 4.2). in that psychosocial factors are assumed to 1. The firstis considered a simple, direct- affect pregnancy outcome via their influence on effects m( x.e.1, wAere1 the links between social health behavior: no direct effects of psychoso- support, emotional distress, and life events on cial factors are hypothesized. F(n- example. the the one hand and poor pregnancy outcome on larlem study found psyclu)social factors not to the other are examined. Most of the studies be related to low birthweight. but to be associ- conducted to date have relied implicitly on this ated with smoking and timeliness of antenatal model. care, which in turn influenced low birthweight. 2. The second, a somewhat more complex Not only do investigators need to be direct-effects model, assumes that interrelaticms explicit about what m(Klel they are testing and may exist among the psychosocial factors., the believe to best represent the links between Turner and Nollstudy is an exemplar. pregnancy conditions and outcome, but the Interestinply, of the more than 20 studies shape of the hypothesized associatilms must be reviewec that considered both emotional dis- considered. Generally, the illustrations in Figure tress and life events, none considered how the 2. like most of our models, are cast in linear two might he related. terms. At least three other curve shapes must 3. The third model builds upon the second considered--the curvilinear, the interactive, by adding sociodeniographic factors into the aiLl the weighted. With respect to life changes. equation. This is important because. as indicat- a linear associatirm would imply that a cumula- ed earlier, suhgr(mps differ with respect to their tive nu)del may be operating; that is, as addi- experience of social supp(wt, life events, and ti(mal life stressors occur, they increase the en-union:II distress (these conditions not being probability of poor outcome accordingly. Each equally distributed across the population), and life event, therefore. has the Sallie weight. A those subgroups at highest risk for stressful curvilinear association (depending on the events may be the very ones with fewer shape) might suggest that a threslu)ld model resources available to cope with such events. c(mld he operating. That is. a link between life 4. The fourth model extends this mediated changes and poor outcomes would not he seen model to include health behaviors. Health until a critical number of life changes had behaviors and psychosocial factors are hypoth- occurred. An interactive association (adding in esized to interact with (me another and to he life changes separately as well as adding in the associated with poor pregnancy outcomes. An interactions among the changes). if tested for example is the work by Newton and associates, and found, would suggest a more c(miplex. in which cigarette smoking and life events were multilevel model. That is,life changes would associated. with both relating to poor neonatal not be influencing pregnancy outcome in a outcome. When controlling for sinoking, the cumulative fashion. but different events, in associatam between life events and outc( me combination with others. would be in()st pre- disappeared. dictive. This is in essence the type of model

r 46 sirvss am/ Support Ourin,c; Prcgnano. Determittattts used in a few studies reviewed, although these low binhweight, then these strategies are likely were examining links between life eventsand to be effective. Other programs offer social social support, rather than different life support with the expectation that women will events.'"2-' A weighted model would assign become more adept at managing stressful different weights to life changes, making the events: in other words, pregnant women are assumption that some events are more likely to provided resources or skills in the hopes that he stressful than others; these weights would they will learn how to obtain support for them- then be used. either in a cumulative, threshold. selves. becoming less reliant on a caseworker or interactive model (sociological literature sug- or other support person in the health service gests that links between stressful events and system. The services offered by these programs mental well-being are no altered appreciably are often diverse, ranging from information by using weighted instead of unweighted sums about pregnancy, delivery, and child care; of life events). referral services for housing; nutrition; medical care; jobs; child care; welfare; and so on.In a few, specific curriculums have been developed to teach problem-solving skills. altlumgh these IMPLICAIIONS FOR PREVENTION STRAMIES programs have not been used extensively for The types of preventative services offered pregnant women. They have been developed to pregnant women and the subgroupswho more frequently for new mothers, specifically are provided services areinfluenced by beliefs those who are disadvantaged, young, or who about the effects of social support and stress on have a low birthweight infant. the pregnant women. In addition, the ways in If, however. a program developer accepts which stress and social support are thought to the premise that psychosocial factors in and of affect pregnancy will influence how a program themselves are less important than how these is structured. factors affect health behavior, then different strategies. or a mix of strategies, might be Type,!TAT/ iceS adopted. For example, programs attempting to Types of services might be different if the bring women into antenatal health care earlier pn wider thought that psychosocialfactors influ- may find inf(wmation on the psychosocial cor- enced low birthweight directly or indirectly, via relates Of late entry helpful. In an extensive health behavior. Se.:eral programs offer social outreach effort to identify pregnant women not support to pregnant women in the beliefthat enrolled in antenatal care in Harlem. compar- support may buffer the woman fr(nn the unto- isons were made between wc)inen who entered ward effects of stressful life events. In sUrne the Harlem system via mitreach efforts and cases, the implicit gc)al seems toIx. amehorat- those who came into the system through other ing emotional distress: in others, minimizing means. The former group turned out to be the effects of life changes, If interactive or num.. socially isolated and slightly olderthan buffering models are found to be applkaNe to the latter gr(mp.- In a recent paper. discussion

Jeanne Bryoks-Gunit G Advances in the PtvventionLou' /*dime:phi centers on the ways in which this information cal in order to document changes in the deliv- might be used to design future outreach ery of a service or curriculum over time; such efforts." This is particularly important since the changes probably always occur. percentage of black mothers with early antena- Even when a particular group or communi- tal care did not increase in the early 1980s.8R ty has been targeted (as is the case in most pro- The mix of services also may be affected grams), itis important to document which by the ways in which psychosocial factors are women actually received services, orwhat thought to influence low birthweight. .v1ost pro- intensity of services each woman received. In grams provide a mix of services, because (a) principle, most programs provide s:rvices providers believe that multiple services are equally to all women; however, in reality, this is needed to alter behavior; (b) providers are probably rare. Sr.)me women have more prob- unsure what services actually make a difference lems, some women seem to need more support, (given the research findings to date, this is not some women more readily stay in touch with an unreasonable assumption); (c) providers providers, and sone women comply more read- wish to target a variety of outcomes and ily with program requirements. All of these fac- believe that different services are necessary for tors influence on whom the service provider different outcomes; or (d) services often are focuses. How this situation affects outcome or subject to changes in funding levels and per- effk.,q data is not well documented. sonnel complements. While appropriate in terms of offering clients a rich array of services, Pwgrant Outcomes it is impossible to find out what actually makes Assumptions about stress and support a difference when a mix of services is provid- influence what data are gathered. With mediated. Few pregnancy programs allow for separa- ed models, the out("mnes over and above early tion of different program components (even labor, low birthweight, iGR. and Apgar scores when randomized control designs are used, are relevant. We need to expand our outcome typically one group receives the enriched ser- measures, however, to look at subgroups, and vice package and the other does not). to consider long-term effects. In the Elmira Because few -strong" models exist to Project, in which support and educatim were explain how and in what circumstances some- provided to rural disadvantaged pregnant thing like social support would influence women in New York, home visits were associ- women's behavior or emotional state, few pro- ated with reductions in smoking, preterm deliv- grams have developed a specific detailed cur- ery for the smokers, and in lowbirthweight for riculum. Such work is being done for mothers the young adolescents (those under 17 years).' of young children and may be relevant to preg- In addition, nurse-visited women were more nancy programs.'M It is critical to collect likely to attend childbirth class and the W1.: inf(mation on what service prcwiders actually program; they also reported usingsocial sup- do with their clients; process data would be ports more frequently. Such positive outcomes welcome. Parenthetically, process data are criti- may increase the chances ofgood child health

r 48 J Stro.s and Suppoil During Pregnancy Determinants and development following the birth.sq As ENDNOTES another example, long-term effects of a sup- a. Race differentials may be accounted for in partby eth- portive curriculum may be seen well after the nic group vanations in demographic risk factors and program has ended. In a randomizedtrial of health behavior. Although controlling for factors such the efficacy of special and comprehensive pre- as education, age, initiation of prenatal care,smoking, and prior low birthweight births reduces race differen- natal services for teenagers, no effects were tials, it does not totally eliminate them.2."' seen for pregnancy outcomes (asindicated by Apgar scores and low birthweight). Those K A slightly different approach has been taken by Kramer." who has attempted, via meta-analysis, to young women who attended the specialclinic, specify the proportion of variance inlow hirthweight however, were more likely to delay their sec- and premature births accounted for by various envi- ond birth than those who did not. This delay ronmental, psychosocial and medical factors. was associated with notbeing on welfare 17 c. Other psychosocial factors not discussed here may Ix. years later."' Long-termeffects on children (but relevant for an understanding of pregnancy outcome: not mothers) were seen in ProjectRedirection, InfOrmation Neeking, relationships with mother and spouse, and self-definitions associated with mother- a program for teenagemothers.' hood are such factors. Iligh information seeking is ass(wiated with high self-esteem; the former may act to alleviate anxiety and elevate self-esteem in the pregnant women, although this hypothesis has not been CONCIA;SION tested directly.' In addition. relationships with spouse and mother may be particularly salient during In conclusion, the literature on psychoso- pregnancy. both as social support and, in the case of a mother, as a source of information about and role cial influences on pregnancy outcome, while model for parenting:" they are associated with anxiety not particularly strong, provides someclues for as well as perceptions of littlesupport."' Women future research directions. More sophisticated establish definitions of themselves as "mothers" dunng pregnancy. Pregnant women who have self-confidence models could be derived in order to test specif- as mothers and have no difficulty visualizingthem- ically the associations between and within psy- selves as mothers tend to have better postpartum chosocial factors, sociodermgraphic conditions, adjustment; they also may he less anxious. although and health behavior. Links between chronic little research ha.s addres,sed this issue.'" stress and uterine contractions, andthe possi- A retrospective e(miparison of women who had bility that social support could alter these asso- preterm. term, and post-term deliveries suggested that mt)thers with pre-terms had more psychological dis- dations, should be studied further. Efforts to tress with mothers with terms, who in turn had higher "explain" race differentials need to take into scores than mothers with post-terms. The autluns sug- account factors other thansociodemographic gest that women with post-term labors are less sensi- why these differentials are tive to oxyt(win than women with pre-term or term ones, to determine labors.' so persistent. Finally,research designs taking into account mediated effects may be useful in e. Alm)st none of the studies reviewedhere report the magnitude of the associations. Typically, less than 10 the design of prevention programs. percent of the variance in mental and physical health outcomes is accounted for by life events." Possible reasons for the relatively weak associations include the

jeanize Brooks-Gunn 49 Advances in the Preventkm (y. [)e Birthweight

following: the importance of 4.;ontextually specific facts Anderson. 1). K., Rhees. R. W., and Fleming. D. F. surrounding the occurence of a negative life event. (1985).Effects of prenatal stress on differentiation inadequacy of assessnient instruments, and the r()Ie of the sexually dimorphic nucleus of the preoptic chronic stressors rather than specific life events."'" area (SDN-POA) of the rat brain.Brain Rescyirch The last is particularly important in the study of preg- 332, 113-118. nant women. For example. chronic stressors possibly , Stress, anxiety, and birth out- could be a contributing factor to the race diffrentials Istvan.I. (198(,). in low birthweight and prematurity, over and above comes: A critical review of the evidence. factors such as age of mothers. number of children. Ps)'chological Bulletin 10(1, 331-3-19. marital status, and education.' 5. Riopelle. A. J.(1982).Prenatal pri)tein deprivati(m and early behavioral development.In:J. Orbach (Ed./. Neu rivsycht,logy alter lashlev:Filty _years ACKNOWLEGN1EN1' Si 11 CC 1 he publication of Brain Mechanisms and Support was provided by the Bureau of Intelligence.Hillsdale, NJ: Erlbaum. Maternal and Child Ifealth and the National 9 Kessler. R. C.. Price. R. and Wortman. C. B. (198i).Social factors in psychopathology. Annual InstituteofChildflealthand Human Psycliologv 36, 531-5'72. Development. The chapter was written while the 10. PeaJin. 1..I. (1983).Role strains and personal author was a Visiting Scholar at the Russell Sage stress. In: II.B. Kaplan (Ed.). Psrchiisocial stress; Foundation. The support is appreciated. Marie Trenl ...in tbeury and research.Nvw v(Irk, McCormick is to be thanked for her comments. Acadc..rnic Press. 11. Cann( n. W. B.(1939).The wisdom (/ the inhlv. New YiKk: Norton.

REFERENCES 12. Seyle. IL(1982).1 listorV and pres('nt status of the stress concept.In: 1.. Goklberger and S. Brezniti 1. Kessel. S. 5,. Villar. J..Bert...fides,II. W..and (Eds.). Hatuibi.)4 (fstress:777eoretical and clinical Nugent. R. P.(1984).The changing patterns of aspects. New York: Free Press. low birthweight in the l'nited States:19-0-1980. Journal ry' the American Meelk.al sso('iati(?n 24;1. 13. (:ohen, S. and Wills, T. A. I985).Stress, social sup- 1978-1982. port. and the buffering hypothesis.RsTchuliceil Bulletin 95. 310--3-S-. Kleinman..J. C., and Kessel. S.S. (198-').Racial l'svehoh)gical stress anti the differences in low birthweight:Trends and risk 1,azanis. R.S. (196(). coping process. New York: Mc(1raw factors.Neu. England Journal yl Medicine 31-. 7-49-7'53. I.azarus. R. S.. and Launier. R.( vr8).Stress-related transactions between persons and environment.In: 3. Institute of Medicine. Committee to Study the I..A. Pervin and M. Lewis (Eds.). Perspective in Prevention of Low Birthweighl. (198.i ). Preventing b)ii. birtbiecight. Washington. DC: nitenictional psyclu dojo.. NeW YorkPlenum. National Academy Press. 16!louse:J. S.(1981 ). 'tirk streSS svcial supprot Reading. NIA: Addison-Wesley. -4. Fuchs. A. R,and Fuchs. F.(1984). Endocrinology of term and preterm labor,In:F. Fuchs. and P. G. 1. Ellis. IL(1936i. The psychic state during pregnan- Stubbk.field (Eds.). Preterm birth: C(IuseN. prctru- cy.In:II. Ellis (Ed.). Studies in the psychology of lion. (old management. New Yiwk: Macmillan sex (Yol. 2 ).New itwk:Modern Library. (Original work published 190(0 Myers. R. F.. and NIvers, S. E.( 19-9).use of sedative. analgesic, and anesthetic drugs (luring 18NIcDonakl. R.I..(1968). The role of emotional Lu. labor and and delivery:Bane or botm? American torsinobstetric complications. A re iew, journal if Obstetrics and C,:rnecologr 133. 83-10i. Psivt)suinatic Medicinc 30.

50 Stress and Supprnl Ottrint;Prtgnallo' Detertninants

19. Omer. H.. Friedlander, D.. Patti. Z.. and Shekel, I. 31. Gorsuch. R.1... and Key. M. K.(1974).Abnormal- (1986).Life stresses and premature lahc)r:Real con- ities of pregnancy as a function of anxiety and life nection Or artificial findings?Psychosomatic stress. Psychosomatic Medicine 30, 352-362. Medicine 48. 362-369. 32. (;randon. A. J.(1979).Maternal anxiety and obstet- Cramond. (1951).Psvchological aspvcts of uter- ric complications:Journal of Psychosomatic ine dysfuncticm. Lancet 2.1241-1245. Research 23. 109-111 21. Scott, E. Nt. and Thomson. A. NI.(195(,). A psycho- 33. Nlorcos, 1.11.. and Funke-Ferber. J.(1980).Anxiety logicalinvestigation of prinligravidae. IV: and depression in the mother and father and their Psychological factors and the clinical phenomena of relationship to physical conlplications of pregnancy labour. Journal (y. Obstetrics and Gynecology 03. and labour.In:L. Carenza and L. Zichella (Eds.). 502-508 Emotion and reproiducthm: Proceedings (?/. the Serono .s.ymposia (VOL 2011i. London: Academic Press. Grimm. E. R.(1%1). Psychological tension in pregnancy. Psycluisomatic Medkine 23. 52o-52-. 34. Jones. A. C.(19-s).Life change and psychological distress as predictors of pregnancy outcome. 23. Dayids, A.. DeVault. S., and Talmadge, NI.(1%1). Pswbosomatic Medicine -*O. ,+02.--.+12. Psychological study of emotional factors in pregnan- 35. Burstein, E.. Kinch. R. A.. and Stern. L.(197.o. cy:A preliminary report.Psychosomatk Medicine Anxiety pregnancy labor and the neonate. 23, 93-103. American Journal (Thstetlks and Civneco/ogy 118. 24. Dayids. A.. and DeVault. S. (1%2). Maternal anxiety 195-199. during pregnancy and childbirth abnormalities. 36. Edwards. K. 11., and Jones. NI.R. (19-0).Person- Psychosomatic Medkine 404-470. ality changes related to pregnancy and obstetric 25, Kapp, F. T.. Hornstein. S.. and G-.-ahmu. V..(1%3). complications: Proceedings of the 78th A nnual Sotne psychologic factors in prolonged labor due to Convention of the<-11t1CriCatiPsycl)ological inefficient uterine action.Comprebensin, Psychiatry Association 5. 341- 3.+2. 4, 9-18. 37 Beck. N. C.. Siegel. I.. J.. Davkison. N. P.. Kormeier. 26. McDonald. R. L.. Gynther. NI. D.. and Christakos. S.. Breitenstein. A.. and Hall. D. C.(1980). The pre- C.(1963).Relationships between maternal anxiety diction of pregnancy outcome: Maternal preparation and obstetric complications.Psychosomatic anxiety and attitudinal sets.Journal (if Psych(,- Medicine 25, 35--363. somatic Research 21. 3.43-351. 38. Norbeck. J.S.. and Tilden. V.1'. (1983),Life stress Grimm. E. R.. and Venet. W. R. (1%6). The relationship of emotional adjustnwnt and attitudes to the scippcwt and emotional disequilibrium in complications of pregnancy: A prospective multivariate study. COLIN( and Outcome of pregnancy.Psychosomatic Journal (y' Health and Social Behavior 24. 30-46. Medkine 28. 34-49. 39. Ottinger. D. R.. and Simmons. J. E. (1964). Behavior 28. McDonald. K.L.. and Christakos. A. C.(1963). of human net mates and prenatal maternal anxiety. Relationshio of enuiticinal adjustment during preg- Psychological Reports 1 4, 391-394. nancy to obstetric complications. Amerkan Journal (y'obsterrics (no 6:myco1ogy 86. 3,41-348. Standley. K.. Souk.. B.. and Copans. S.(19-9). Dimensimls or prenatal anxiety and their influence McDonald. R. L.(196:3.),PeNonality characteristics on pregnancy outcome.AmericanJournal of in patients with three obstetric complications. obstetrics and c.,:ynecolo,i,o,135. 22-26. Pswhosontatic Medicine 38$-389. .11. Ching. J.. and Newton. N.(1982).A prospective 30. NIcDonald. R. L.. and Parham. K. J.(19641.Relation study of psychological and social factors in pregnan- of emotional changes during pregnancy to obstetric cy related to pretern) and low-birthweight deliveries. complicationsinunmarriedprimigravidas. In.if. Prill and NI. Stauber (Eds.). Ackances in .1111(0CtIll Journal (!f Obstetrics (nut Gynecohig,y 90. psychosomatic obstetrics and gynecolor.J.:New York: 19S-.201. springer,Verlag.

Jeanne 13moks-Gunn 51 Advances in the Prevention of Low Birthuvight

42. Bunny, W., Shapiro, A., Ader, R., Davis. J., Herd, 53. Smilkstein, G., Helsper-Lucas, A., Ashworth, C., A., Kopin, I. J., Krieger, D., Matthysse, S., Montano, D., and Pagel, NI.(1984).Prediction of Stunkard, A., Weissman, M., and Wyatt. R. J. pregnancy complications:An application of the (1982). Panel report on stress and illness.In: G. biopsychosocial model.Social SCienCe and Medicine Eliot and C. Eisdorfer (Eds.), Stress and human 18, 315-321. health,.Analysis and implications (if research. 54. Newton, R. W.. and Hunt, L.P.(1984).Psychosocial New York: Springer. stress in pregnancy and its relation to low hirthweight. 43. Cox, D. J., Taylor, A. G., Nowacek, G., Holley-Wilcox, British Medical journal 288. 1191-1194. P.. Pohl. S. L., and Guthrow, E.(1984). The relation- 55, Boyce. W. T.. Schaeffer, C., and Utti. C.(1985). ship between psychological stress and insulin-depen- Permanence and change:Psychosocial factors in the dent diabetic blood glucose control:Preliminary outcome of adolescent pregnancy. Social Science and investigatü m s.Health Psychology 3, 63-75. Medicine 21. 1279-1287. 44. Kiecolt-Glaser, J. K.. Fisher, I.. D., Ogrocki. P., Stout. 56. Berkowitz, G. S and Kasl, S. V.(1983). The role of J. C., Speicher, C. E., and Glaser, R.(1987).Marital psychosocial factors in spontaneous preterm delivery. quality, marital disruption and immune function. Journal of Psychosomatic Research 27, 283-290. Psychosomatic Akdicine 49,13-34. 57. Newton, R. W., Webster, P. A. C., Binu, R S., Maskrey, 45. Schleifer. S. J.. Keller, S. E., Camerino, M.. Thornton. N.. and Phillips, A. B. (1979).Psychosocial stress in J. C., and Stein, M.(1983). Suppression of lympho- pregnancy and its relation to the onset of premature cyte stimulation f011owing bereavement, journal qf labor.British Aledicalfournal 2, 411-113. the American Afedical Association 250, 374-377. 58, Williams. C. C., Williams, R. A., Griswold, M. J., and 46. Glaser. R., Kiecolt-Glaser, J.K.. Stout, J. C.. Tarr% K. Holmes. T. H.(1975).Pregnancy and life change. L., Speicher, C. E., and Ilolliday, J. E.(1985). Stress- Journal of Psychosomatk Research 19, 123-129. related impairments in cellular immunity.Psychiatry 59. Cassel, J.(1976). The contributOn of the social envi- Research 16, 233-239. ronment to host resistance:The Fourth Wade 47. Dohrenwend. B. S.,and Dohrenwend, B. P. (Eds.). Hampton Frost Lecture.American Journal of (1981).StresVid hfr events and their contests. New hpidemictlogy 101, 107-123. York: Prodist. 00. C(ibh. S.(1970). Social support as a moderator of life distress. Psychosomatic Afedicine 10, 300-31,1. 48. Thoits. P. A.(1984).Coping, social suppm-t. and psychohigical outcomes:The central role of emo- 01. Thoits.1'.A. (1982).Life stress social support and tion.In:P. Shaver (Ed.). Review qf personality and ilsychological vulnerability:Epidemiological consider- social psychology (vol. 5).Beverlylills:Sage, ations. ./ournal gl'Community Psychology 10, 3.41-302. -49. Kaplan. 11. B.(1983).Psychosocia/stress: Tren(Is Iii 62. Berkowitz, G. S.. Holford, T. R.. and Berkowitz. R. L. theon and research. New York: Academic Press, (1982).Effects of cigarette smoking, alcohol, coffee. and tea consumption of preterm delivery.Early 50. Thoits. P. A.(1983).Dimensions of life events that Human Develipment 7. 239-250. influence psycholcigical distress:An evaluation and synthesis of the literature.In: II.I'. Kaplan (Ed.). 03. Brooks-Gunn. J.. Nict.:ormick. M. C.. and Heagartv, Psycliosocial stress:Trends in theory and research. C.(1988).Preventing infant mortality and morbidity: New York: Academic Press. Developmental perspectives...1,71CriCall fOttr?/(11 Orthopsychiatry 58, 288-290, Sarason, I.15,, and Sarason. 15. R.(1984). Social support: Theoty research and applications. The Iiague: (1,1. McCormick. M. C.. Brooks-Gunn. J., Shorter T.. Holmes. J. H.. Wallace. C. Y., and fleagarty. NE C. Martinus (1990).Factors associated with smoking in low- 52. Nuckolls, K. 15.. Cassel..1.. and Kaplan. B. II.(1972). income pregnant women: Relationship to birthweight, Psychosocial assets. life crisis, and the prognosis of stressful life events, social support.health behaviors. pregnancy.American Journal (!l* Epidemiology 95. and mental distress. Journal (y. anical lpidemioloKy 431-+4 I . 43, +11-448.

52 61 Stress (yid Support During Pregnanc.)., Determinants

65. Dohrenwend, B. S., and Dohrenwend B. P.(1974). 78. Olds, D. L., Henderson, C. R.. Jr., Tatelbaum, R., and Stressful 0. events:Their nature and Cifra.c.New Chamberlin. R.(1986). Improving the delivery of pre- York: Wiley. natal care and outcomes of pregnancy: A randomized trial of nurse-home visitation.Pediatrics 77,16-28. 66, Mueller, D., Edwards, D. W.. and Yarvis, R. M. (1977). Stressful life events and psychiatric symptomatology: 79. Clewell, B. C., Brooks-Gunn, J., and Benasich. A. A. Change or undesirability,journal cif Health and (1989).Evaluating child-related outcomes of teenage Social Behavior 18, 307-316. parenting programs. Family Relations 201-209. 07. Ross, C. E., and Mirowski, J.(1979). A comparison of 80. Zigkr, E., and Valentine, J., (Eds.).(1979).Project life event weighting schemes: Change undesirability Head Start:A legacy of the war on poverty.New and effect proportional indices.journal qf Health York: Macmillan Publishing. and Social Behavior 20, 166-177. 81. The Infant Health and Development Program Staff. 68. Brown, G. W., Bhrolchain. NI. N.. and Harris. T. (1990).F.nhancing the outcomes of low birthweight, (1975). Social class and psychiatric disturbance among premature infants: A multisite randomized trial. women in an urban population. Sociolw 9,225-254. journal of the r1nterican Medical Association 263, 3035-3070. 69, Pearlin. L. I., and Johnson, J. S.(1977). Marital status, life strains, and depression.A nzerican Sociohgical 82. limoks-Gunn, J.(1990),Promoting healthy devel- Reileu. 42, 70+-715. opment in young children:What educational interventions work?In:D. E. Rogers and E. Ginzberg 70. Pearlin. L. L. and Lieberman. M. A.(1977),Social (Eds.), Improving the life chances cf children at risk. sources of emotional distress.In:It Simmons (Ed.). Boulder, CO: Westview Press. Researchincommunity and mentalhealth. Greenwich. CT: JAI Press. 83. McCormick. M. C.. Brook.s-Gunn, J., Shorter, T., Holmes, J. IL, Wallace, C. Y.. and Heagarty, M. C. 71. Kessler. R. C.(1979).Stress social status and psycho- (1989). Outreach as casefinding:Its effect on enroll- logical distreSS. journal (f I et -411 h and Social Behal'Ufrr ment in prenatal care. Medical Care 27, 95-102. 20, 259-272. 84. Brooks-Gunn. j.. McCormick, M. C., Gunn, R. W., 72. Kessler. It C.. and Cleary. P. D.(1980).Social class Shorter, T.. Wallace. C. Y.. :md Ife2g:trty, r. Anwrican Sociobwkal and psychological distress. (1989).Outreach as casefinding:The process of i5, 63-78. locating low income pregnant women.Medical 73. Kessler. R. C., and Essex. M.(1982).Marital status Care 27, 95-102. and depression: The role of coping resources.Social 85. Ingram, D. D.. Makuc. D.. and Kleinman. J. C. Forces 61. 484-507. (1986).National and state trends in use of prenatal Liem. R., and Lion. J.(1978).Social class and mental care. 1970-1983. .4mericalt lottrnalqf Pith& Health illness reconsidered: The role of economic stress and 70, 415-t23. s()cial suppcirt. journal cf Heald) and Social Behavior 86. Brooks-Gunn, .1., and I learn. R.(1982).Early inter- 19, 139-156. vention anddevelopmental dysfunction: 75, Turner. It J.. and Noh, S.(1982),Class and psycho- Implicaticms for pediatrics.In:Advances in pedi- logical vulnerability among wcimen: The significance atrics. New York: Yearbook Publishers. Inc. of social support and personal control.journal (f 87. Lazar. I.. and Darlington, R.(1984).Lasting effects Health and Social Behavior 2.i. 2-15. of early education.Monographs of the Society for 76, Lieberman. E.. Ryan. K. .1.. Monson. R. R., and Research in Child Development 47.(1-2. Serial No. Sclux..nbaum. S. C. (1987). Risk factors accounting for 194). Neu, racial differences in the rate of premature birth. 88. Ramey. C. T.. and Campbell, E. A.(1986). The England journal cf Medicine 317,7.43-7-48. Carolina Abecedarian Project: An educational exper- 77. Behrman. R. E. (1987). Premature births among black iment concerning human malleability.In:J. J. women. (Editorial).Neu' England Journal qf ,ttedicine Gallagher and C. T. Ramey (Eds.). The malleability of 317. 763-765. children.Baltimore: Brooks Publishing Co.

*Jeanne Brooks-Gunn 53 A(/vances in the Prevention of Lou' Birthuvigbt

89. Brooks-Gunn. J.C1990).Identifying the vulnerable young child.In:D. F. Rogers and E. Ginzberg ( Eds.). Improving the chances (?,0' children at risk. Boulder. CO: Westview Press. Furstenberg, F. E. Jr.. Brooks-Gunn. J.. and Morgan.

1'. (1987').Adolescent mothers and their children in later lifi.New York:Cambridge University Press. Also Family naming.; PersM'ctil 198-7: 19:142-51

91. Polit. D. F.(1989),Effects of a comprehensive pn)- grant for teenage parents:Five years after Project Redirection.Family Planning Perspectives 21. 16.1-169.

92. NIcCormick.M. C. (1985),The contribution ofI()W birthweight to infant mortality and clnldhood morbidity. A'ew England Journal (y Medkine 312, 82-90.

93. Kramer.NI. S. (198-). Intrauterine growth and gestational duration determinants.Pediatrics 30. 502-511. 94. Deutsch. F. M.. Ruble. D. N.. Fleming. A., Brooks- Gunn. J.. and Stangor. C. (1988). Information-seeking and self-definition during the transition to niother- hood, Journal cf Personality and Social Psychology 55, 420-431.

95,Ruble. D. N.. Brooks-Gunn. J.. Fleming A S Fitzmaurice. G.. Stangor. C.. and Deutsch. F. (1990). Transitiontonmtherhoodand tilt!self: Measurement. salability and change. Journal of Pemoudity aml Social Psychology...18..450-163.

96. Oakley. A.t1980). Womenconfined:Tineards a sociology 4childbirth.New York: Sclux:hen B( >oks.

97'. Belsky. J. (1984), The determinants of parenting: A process model. 07thi Derehonent 55. 83-96. 98. Takagashi. K.. Diamond. F.. Biencarz..I.. et al. (1980). Uterine contractility and oxymevn sensitivity in preterm term and postterm pregnancy. American Journal (y. (thstetrics and 6.yiweo1oto, 136. 99. Rabkin..1. and Struening. F. L (19-6). Life events stress and illness.Science 194. 1013-1020. 1.00.Brown. G. K. (19" O. Meaning measurement and stress of life events. In:B. S. Dohrenwend and B. P. Dohrenwend (Pds.). Stressful life events:Their nature and effects. New 'York: John Wiley and sons.

54 6 3stress too Sumort fluring Povtano' Detennitiatzts

Statistical Fiiidings: Stress and Support During Pregnancy

Table 4.1 Correlates of Cigarette-Smoking in Pregnancy Among Womenof Central Harlem

Smoking Status Quit For Never Pregnancy Current

(N=195) (N=70) (N 179)

Health Habits and Prenatal Care Uses alcohol 9.7 9.0 10.5 Started care on first trimester 39.9 40.8 32.1 Numer of visits > 4 86.1 87.0 78.5 Adequate care 32.6 31.1 18.7*

Stressful Events/Social Support/Mental Health Stressful events < 24.6 22.9 14.0* Lives with husband/boyfriend 35.0 27.6 20.3* Belongs to a church 58.6 41.7 41.7* No situations with support 23.1 18.6 10.6* GHQ > S 25.1 47.1 27.4*

p < 0 OS, ic v%ith appropnote degrees ottreillont Fun?? Mt On tip< h, tit .11., present pubbt ation

Figure 4.1 Social Support and the Buffering Hypothesis Two points at which social support may intertvrevvith the hypothesized causal link between stressful eventsand illness*

So< tal +tipporf may result in rtypprais,11, Soi. ;al Support' inhibition or maladjustivei may prevent stress responses, or tat ilitationnt apprais.11 arliustRe t our er responses

Sr( 1- andior Potential f (ntts, I motionally linked ph% App aisal logic al respr ins(' or lwha% - beha%ior stresstul apprAiseci Prot ess moral adaptation eventisr as stressful

'fro!, 0)17111 N 1(485, page I

55 JeC11111V kc-6.1111 11 64 Advances in the Prevention qf Low I3irthavight

Figure 4.2 Five Possible Models for Studying Associations Between Psychosocial Factors and Pregnancy Outcomes

Figure 4.2d 4.2a Direct and Indirect Effects Model: Psychosocial Direct Effects Model: Simple Factors, Sociodemographic Conditions, and Health Behavior Psychosocial Factors Pregnancy Outcome

4.2b Emotional Distress Direct Effects Model: Interrelations Among Psychosocial Factors Life Events Support

Emotional Distress .

Negative Life Events Pregnancy Outcome !Aealth Behavior*A wr 1

Social Support 1

1 Pregnancy Outcome

Education

Figure 4.2c Age Direct and Indirect Effects Model: Interrelations Among and Between Psychosocial Factors and Number of children

Sociodemographic Conditions Income

Emotional Distress

Negative Life Events Race Social Support #

Education

Age 0.47 Pregnancy CJutcome Number of children

Income

Race

56 Statisi441 Findings: Stress and Support During Pnignancy O Determinants

Figure 4.2e Mediated Model: Effects Through Health Behavior

Emotional Distress

Life Events

Social Support

Education

Age Health Behavior ID-Pregnancy Outcome During Pregnancy Number of children

lnconw A

Race

Jeanne Brouks-Gutin 6 ;

Rau lts of a Three-Year Praspectiu? ControlW Randomized Trial of Pretenn Birth Prevention at the Univetsily of Pittsburgh

EMU LARD Mt ELIER4 1E1.13AC11, M.1).

INTRODITCVION Measures of health care in difkrent societies are difficult to compare;however, itis often stated that infant mortality (momility during the first year of lif) is an important indicator of health care in a nation. According to the latest available statistics, the United t;iates ranks 19th in infant mortality among the nationsof the world, a decline from the rank of 15th held in 1968. Upon examination of the differences in infant mortality in different societies, it is readily apparent that by far the most importantfac- tor is neonatalmortality due to preterm deliveiy. In countries with lower infant mortality rates, fewer neonatesdie as a result of preterm delivery. Thus, the causesand the Adtzinces in the Prowntion qf Lou, Birth u vigb t prevention of preterm birth are of utmost con- were considered to be in a low-risk category. cern to individuals interested in perinatal health Patients who presented for the first time after care. Definition of preterm birth as birth before 28 weeks' gestation were excluded. Similarly, completion of the 36th week is preferable to a patients whose estimated date of delivery was definition by birthweight (<2500 g) because the after the end of the three-year study period latter includes term newborns who are grmth were not enrolled. Patients assigned to the retarded and have lower morbidity and mortali- high-risk group were approached and invited ty. Definition of preterm birth by gestational age to particirate in a prospective randomized con- is often difficult, however, due to uncertainty trolled trial of preterm birth prevention. alxmt the gestational age in some patients Participants signed a consent fc)rm appnwed by Preventiom of preterm birth is the ultimate the Institutional Review Board. Patients who goal of perinatal health care, and any promis- did not participate were classified as refusers if ing approach in this area is readily embraced, they declined participation; they were classified often without careful scientific scrutiny, by as ineligible if they could not be randomized means of prospective randomized trials. before 28 weeks' gestation or if they did not Reports of a reduction in preterm births as a return for prenatal care after randomization result of preterm birth prevention pmgrams in until they were 28 or more weeks' gestation. the United States,France,' and Martinique' Patients were randomly allocated to either a prompted us to evaluate a preterm birth pre- control group with the usual prenatal care or an vention program in an indigent clinic pcpula- intervention group. Two nurses with extensive tion at our institution. In contrast to the obstetrical experience gave individual teaching published reports. we planned to study the to patients in the intervention grcnip regarding potential efficacy of such a prograin in a subtle symptoms and signs of preterm labor. prospective controlled randomized design such as appearance of -menstrual cramps." lower using birth before 36 completed weeks of ges- back pain, -gas pain." suprapubic pressure, thigh tation to define preterm birth. pain, or change in vaginal discharge. The patients were instructed in self-palpation of the uterus and the contrast in feeling between a con- tracted and a relaxed muscle was demonstrated MATERIAL AND MI:11101)5 to them. Patients were told that they should lie During a three-year period, all patients down on their side when they noticed contrac- who registered for prenatal care at Magee- tions and drink a quart of liquid within an hour. Women's Ilospital in Pittsburgh were screened, If contracticms did not subside during this hour, and the scoring system of Greasy et al: was patients were instructed to come to the hospital used to assign risk scores for preterm labor and for evaluation. Weekly cervical examinations birth. A score of 10 or more placed the patient from 20 to 37 weeks' gestation were performed in a high-risk group for pretenn labor and on patients in the interv-...ntion group, with gentle birth, while patients with a score below 10 palpation of cervical effacemem and dilation.

6 ; )

62 A Prospective Contmlled Randomized Trial qf Preterit? Binb Nervation at tbe I .niversity (plias-burgh /Wen VIItions

Rescreening of low-risk patients was done as being at high risk of pretcrm labor (21.9%) between 22 and 26 weeks' gestation. as well as compared to patients scored as being at low updating of the initial risk score at other times risk (7.4%). The majority of patients delivering when new findings were made. preterm (60.4%), however, were in the low-risk Repeated teaching sessions were given by group. Rescreening at 22 to 26 weeks was done the investigator and the two study nurses to the in 2,145 of the patients initially scored at low resident physicians and the nurses involved in risk; only 33 patients (1.5%) became high risk. the care of clinic patients. The physicians and Six additional patients became high risk as a nurses were informed about the instructions result of updating risk scores. given to patients in the intervention group in Comparison of high-risk intervention, con- order to assure a proper response lw medical tn)1, refuser, and ineligible groups demonstrated personnel during communications with patients similar preterm birth rates (22.1%, 22.8%, 19.5%, in this group. and 23,6%, respectively) throughout the study. All data cencerning risk scoring, admis- There was, however, a significant decline in sions for preterm labor, and deliveries were preterm births between the first and second placed on special forms from which they were years of the study (from 13.8% to 9.3%, p < entered into a computer data base to permit 0.001). This decline was maintained during the data analysis upon completion of the study. third year when cc unpared to the second(.!.ar Preterm deliveries were classified :is sponta- (9.3% v. 8,7%, NS). The decline in preterm neous, indicated, or iatmgenic. Datafrom the births from the first to the second year was of Magee-Women's Ilospital Perinatal Computer the same magnitude in the high-risk population Data Base finan eight-month period before ini- (-29.1%) as in the low-risk population (-32.1%). tiation of the study were used to establish a During the first year of the study, the attitude of historic control period. Chi-squared or Fisher resident physicians and nurses involved in the exact tests were used for statistical analysis. care of the clinic population toward the preterm birth prevention program changed from skepti- cism to acceptance and then to genuine interest. Being concerned about the problem of preterm l(zst tus birth. these health care professionals applied A total of 5,i57 patients had initial risk the instructions learned in the multiple teaching scoring fc)r preterm labor, of which 4.595 deliv- sessions tO all patients under their care. ered at 20 or m( re weeks gestatkm. Alx)rtkm In order to verify that the decline in preterm and loss to f()Ilow-up accounted tbr the differ- births during the second and third years of the ence. Risk scores of > 10(high risk) were pre- study was related to the preterm birth preven- sent in 831 of the .)ri patients (18.1%). The tion program. we first examined whether the preterm birth rate (less than 3('completed preterm birth rate during the first year of the weeks) was 10,1 percent for all patients. This study was exceptionally high by comparing it to rate was about 3 times higher in patientsscored the rate during the eight months before the

Eberhard .tfuellerlIeltbach 6.3 70 Adt,ances in the Prevention qf Low Birthweight

study (the historic control period). There was no with preterm labor in the high-risk intervention significant difference in preterm birth rate group and those in the high-risk control group, between the historic control period and the first but significantly more patients in the interven- year of the study (14.1% v. 13.8%, NS). tion group were candidates for tocolysis = To exclude the possibihty of a time-related 0.01). Nevertheless, there was no significant dif- decrease in the preterm birth rate during the ference among patients receiving tocolysis with historic control period and the three years of respect to prolongation of pregnancy to term. the study unrelated to the preterm birth preven- Figure 5.1 illustrates the frequency of preterm tion program, we studied the preterm birth rate labor, preterm birth, and tocolysis in the total in the private patients who delivered at our study population during the three years of the institution and had received their prenatal care study. The decline in the preterm birth rate is by private physicians in their offices in the area. clearly related to the fact that fewer patients Neither these private physicians nor their office presented with preterm labor during the sec- staffs were aware of the content of the preterm ond and third years of the study. The frequency birth prevention program. During the historic of tocolysis was unchanged during the three control period and the 3 years of the study, the years of the study. Thus, it appears that the preterm birth rate in the private patient population (with an average of 7,764 births per year) Table 5.1: Percentage of Patients with Preterm did not change significantly and ranged from Labor, Preterm Birth, and Tocolysis During the 8.0 to 8.6 percent. When the historic control Three Years of the Study period and the first year of the study were com- 12 pared with regard to preterm birth rate in clinic and private patients, the preterm birth rate was 11 found to be significantly lower ( p < 0.001) in the private patients. Due to the decline in 10 preterm births in the study clinic patients during the second and third years of the study, there 9 was no significant difference in the preterm 8

birth rate between clinic study patients and pri- 7 vate patients during these time perk xis. 6 We evaluated the study population (exclud- % of Patients ing patients with indicated preterm deliveries) 5 regarding episodes of preterm labor, use of tocolysis, and prolongaticm of pregnancy to term in patients having received tocolysis. No cse of 3 iatrogenic preterm birth was noted in our study. There was no significant difference 2 between the percentage of patients presenting

7 0 Year l Yew 2 Year 3 64 A Prospectite Contrail& Randomized Trial VPreterm Birth Pretvntion ell the CfPittsburgib Intenentions

teaching and the change in attitude was associ- second and third years of the study irrespective ated with a decrease in the nunther of patients of the risk assignment.° Thus, the health care presenting with preterm labor. Onthe other providers, with their concern about preterm hand, this was not accompanied by adecrease birth, had defeated the prospective controlled in the incidence of tocolysis,indicating that the randomized design of the study by also chang- diagnosis of preterm labor was moreoften ing their care for patients who were not inthe made correctly by patients. high-risk intervention group. A reduction in preterm births maylead to a The initial risk scoring system classified decrease in neonatal mortality in infantsdeliv- '18.1 percent of the patients ashigh-risk ered before 37 weeks' gestation without congen- patients, but only 39.6 percent of the preterm ital anomalies. There was a significantdecrease births occurred in the high-risk group. in neonatal deaths in this categorywhen the his- Conversely, a patient in the high-risk grouphad torical control period was compared tothe sec- a 78.1 percentchance of delivering at term. ond year of the study (from 9.1/1000 to Rescreening and updating of risk scores 2.2/1000. p0.0089) and the third year of the changed the initial risk score in few cases. study (from 9 1/1000 to 2.8/1000, p = 0.0217). Thus, the concept of risk scoring may either be of limited value or specific risk scoring systems need to be developed for specific populations. The results of our study suggest that education- DNA'S:410N al programs applied to entire patientpopula- Preterm birth prevention programs attempt tions may he preferable. to increase the awarenessand recognition of Itis clear that the observed decline in subtle symptoms and signs of pretermlabor preterm birth rate in the studypopulation was among patients andtheir health care providers. not a time-related trendindependent of the As a result of the teaching sessionsin our preterm birth prevention programbecause study. such a change became apparentduring comparison with the preterm birth rate inthe the first year of the study when health care per- private population during the same timeperiod sonnel asked many questions about this pro- did not reveal such a trend. The decline inthe gram, became awareof the subtle symptoms preterm birth rate observed in ourstudy may and signs of preterm labor, andconveyed these not he possible in all patientpopulations. In a to the patients. Patients werefrequently sent to much smaller, similarly designed study of poor, the study nurses for enrollment inthe preterm inner-city black women in Philadelphia,the birth prevention program when they,in fact, investigators were not able to show anydiffer- had been previously randomized intothe con- ence betweenhigh-risk control and interven- trol group. Evaluation of the pretermbirth rates tion patients.- This is in agreementwith our in our study revealed nodifference between findings; however, the study was not large the high-risk interventk)n andcontrol groups, enough to possibly document any ofthe but an overall significant decreaseduring the changes over time which we observed in our

Eberhard Afileller-I ieu bad) 65 7 Advances in the Prevention of Low 131n/weight

large, three-year study. Considering the magni- REFERENCES tude of the problem of preterm birth, with its 1. The National Commission to Prevent Infant Atonality. accompanying neonatal mortality and serious (1988. August).Death btyOre e:The tiligedy long-term morbidity (such as permanent neuro- injant mortalit.y. Washington, DC: Author. logical handicaps, bronchopulm(mary dyspla- 2. 1Ierron N1 Katz. M.. and Creasy. R, K.(1982). and retrolental fibroplasial, any promising Evaluation of a preterm birth prevention program: Preliminary report.obstetrics and Gynecology 59. effort to reduce the preterm birth rate should 452-456. be pursued. Attempts should be made to evalu- 3, Papiernik, E., Bouyer. I.. Dreyfus..L. Collin. D., ate whether a preterm birth prevention pro- Winisdortier. Guegen, S., Leconte. N1., and Lazer. gram can lower preterm birth rates in other (1985). Preventkm of preterm births: A perim1t:t1 populations with the same degree of effective- study in I laguenau. France. Pediatrics 70, 154-158. ness as it did in our indigent clinic population 1. (;()u)on. 11.. Papiernik. E.. and Main, D. (1984). The in Pittsburgh. prevention of preterm delivery thrc)tigh prenatal care: An intervention study in Martinique. International Journal (?1Gynaecology (Ind obstetrics Supported by March (if Dimes Grant 2- 22339-3.+3. 196/C-404 5. Creasy, R. K Gummer. B. A.. and Liggins. C. C. (198(1).System for predicting spontaneous preterm birth.Obstetrics and 6:yneodoAy 55. (92-095.

(). Mueller-I leulLich, E.. Reddick. D., Barnett. B and Bente. R.(1989). Preterm birth prevention: Evaluation of a pnispective controlled rand( wnized trial. .4mericanjournal cy Obstetrics and Gynec(dogy IU. II72-117s. Main. D. NI.. Gabby. S. Cr.. Richardson. D.. and Strong. S.(198,4). Can preterm deliveries be pre- venlecl? Americanicornal f!1( )bstetries and Gynecology 892-898.

8. Institute of Medicine.( 1985 Prewnting low birth (p. 31-38). Washington, DC: National Academy Press.

66 A Prospectne Controlled Randomized Dial qf Prelerm Birth Pretention at the t nivercity ((Pittsburgh Decrease and Rise in Rates of Pretenn Deliveries: Haguenau Prenatal Study, 1971-1988

EMILE PAPIERNIK, M.D. JEAN BOUYER, PHD. JEAN DIU:XI-Vs, M.D. GERARD WINISIX)RFER, M.D. DOMINIQUE COLLIN, M.D.

INTRODI X.:MN A 17-year study on the prevention of preterm deliveries in a delimited region of France has been used as an evaluation tool for the measure of effectivenessof a national policy for improving perinatal outcome (1971-1988). The same tool served as a measureof the recurrence of higher ratesof preterm deliveries after this national policy was no longer followed. This paper will describe the ways in which pregnant women have changedtheir behavior, coming earlier in the course of their pregnancy to the maternity outpatientclinic and more often to seek the best available prenatal care. It

'7 Advances in the Prevention uflow 13irthweight

was observed that the reduction in the rates of information directed at women toencourage bet- preterm births was obtained not in the high-risk ter perinatal follow-up. categories, as was expected, but in the lower This government initiative helped in risk categories. spreading basic concepts and techniques for After cessation of the government drive,we the improvement of prenatal and intrapartum observed how the efforts of pregnantwomen to care, and neonatal resuscitation measures as come early and often to the maternity outpatient well as neonatal care. In addition, the drive clinic were turned down, and how this resulted helped in spreading new ideas, suchas the in the rise in the number of preterm deliveries. prevention of preterm delivery using techniques we proposed. After 1981, a new government tookover PATIENTS AND NIEIIK)Ds and decided not to pursue this established poli- The Ilaguenau Maternity Hospitalserves a cy. As good results had been achieved, the specific limited region in Alsace. France, andwas opinion of the politicians was that the program used as an evaluation tool to measure the effec- had served its purpose and thatno supplemen- tiveness of a policy applied nationwide. The tary efforts were needed to maintain the ongo- principles of preventive measures to reduce the ing measures. Thus, perinatalogycame to be rates of preterm deliveries have been exposed in out of fashkm, and public attention was drawn prevkms papers. They consist of a community- to other aspects of reproduction, specifically to based intervention proposing to allpregnant questions centered around in utero fertilization women a new type of prenatal care. The major in the early 1980s and to AIDS in the late components of the intervention are a risk analy- 1980s. sis for individual factors predictinga preterm In the present study, we measure the deliverv. and several measures aimed at decreas- knowledge and conviction of pregnantwomen ing uterine contractions and preterm deliveries, to use the best accessible prenatal care system which have been described in detail elsewhere.' pnivided free of cost at the Outpatient clinic of The basis for applying this policywas a gov- the Haguenau Maternity Hospital; specifically. ernment program appnwed for five years in the proportion of women consulting in the first 1971. and extended for five additionalrears in trimester (instead of being followed by their 1976. The objectives were to rethice perimital general practitioner) and the number ofprena- accidents leading to death and perinatal-induced tal visits made before 20 weeks. handicapping conditions. The means employed Information about the women and their included financial inducements for the public pregnancies was very carefully gathered using hospitals to help improve their equipment for the best knowledge available by the lastmen- delivery and neonatal care; financial induce- stnial period (LNIP). length of cycle, ultrasound ments for postgraduate training sessions for doc- scan, and pediatric examinatkm of the neonate. tors, midwives, and nurses working in the field; All of the data were computed in die days fol- and financial inducements to speedup public lowing delivery. All babies transferred toa

70 Decrease and Rise in Rates of Preterni 1)eliveries--11aguenau Prenatal Study.1971-1988 1nterivntions neonatal intensive care unit were included in number of days of hospitalization requiredfor the study, as well as Ixthies of women trans- the babies transferred to a Level III center neo- Strasbourg ferred to a referral center. natal intensive care unit (WU) in as well as to theneonatal care unit in the local hospital of Haguenau (see table 6.3). REst.iLis The results were examined for two succes- Reduction in neonatal mortality sive policy application periods.The first Neonatal mortality was reduced in this sequence, from 1971 to 1982, WaSdivided into time sequence, and the decrease inthe number three periods of four years each. Thesecond of preterm babies was a part of thisreduction. sequence will showthe last period of four Table 6.4 describes the total neonatalmortality years, from 1983 to1986. in the three successive timeperiods. A direct standardizaticm was done, as if the pa)portkm Reduction in preterm deliveries of preterm births would not have changedfrom A pn)gressive reduction in pretermdeliver- that of the first period (1971-1974). withweek- ies was observed during the 12-yearperiod specific mortality rates for 1971 to 1978 andfor from 1971 to 1982. The global rate of preterm 1979 to 1982. 11 sAows1 that about half ofthe births defined as less than 37 weeks of gesta- progress can beattributed to an improvement tion from the first day ofthe last menstrual in the care of newborns andthat half of the period was reduced from 1971 to 1974and progress can be explainedby the reduction in from 1979 to 1982 (see table 6.1). the number of pretenn babies. The totalreduc- tion of neonatal deaths wasthen obtained by Reduction in tran. enc better care as well as "better babies." to a neonatal intensive core unit As the number of babies inneed of inten- Reduction in proeml births sive care had decreased, thetransferral to a in women with no defined 1-iskfilctoIN neonatal intensive care unit because of preterm table 6.2). The reduction in preterm births was not birth was significantly reduced (see obtained in high-risk w(mien; that is,those that the propor- It should be noted, however, birth, with a the number of women with a previous preterm tion of transferred babies to of less than preterm births was not reducedbut rather previous stillbirth, of short stature, less than 21 years, of more increased. In the same period, transfersfor average weight, of than 35 years. or with a historyof bleeding in other reasons and, specifically, formalforma- the second or third trimester. Onthe other tion of the newborn were augmented. hand, a significant reduction in pretermbirths Reduction in length cy' hospitalization was observed whenthe women did not fall in required hypreterm newborns the above-mentiomed risk group. Thus.fewer The need for intensive care for preterm preterm births were observedin those women newborns has been quantified according to)the characterized by the absence of adefined risk

Emile Papiernik ci aL 71 Advances in tbe Prevention of low Birtburight

factor; that is, with no previous preterm birth, all women. We used these simple measurements with no previous stillbirth (among pawns as markers of the success of the government women), with a normal stature (above 152 cm), efforts in the prevention of preterm births. with a normal prepregnancy weight (more than Table 6.7 and figure 6.1 show the relation- 48 kg), of between 24 and 34 years of age, and ship between earliness of care and preterm with no history of bleeding in the second or deliveries for four-year periods and for three third trimester (see table 6.5). groups of women defined hy the number of years of school completed (Up to 9 years, Reduction in bigb-riskfactotN 10-12 years, and 13 years and above). The The rate of high-risk factors was reduced in acceptance measures here described are: the observed population. This reduction was A. The mean number of prenatal visits in the observed for previous preterm birth, for age less first two trimesters of pregnancy at the out- than 21 years and more than 35 years, and for patient clinic of the obstetrical department; bleeding in the second or third trimester (see table 6.6). The relationship between this reduc- H. The proportion of women in each group tion in risk factors and the population of preg- enrolled in this outpatient clinic; nant women will be discumed subsequently. C. The proportion of pregnant wc.)men in each group who never had a prenatal visit at this Early obstetrical prenatal care outpatient clinic of the obstetrical depart- and prelyntion oJpretertn delii)eries ment; and Our hypothesis was that a new type of D. The proportion of preterm deliveries (by prenatal care could prevent preterm deliveries. group and by study period). The basic premise was that 7-duction in job- related and home-related excessive physical The new type of prenatal care was accept- efforts for at-risk women would be effective. ed and was requested by all groups of women. That meant that women had to be convinced to with important differences between groups. come early and often to meet the teams of The more informed the women were, the earli- obstetricians and midwives able to inform them er they came in. For the first 12 years of study, and offer specific risk assessments for each from 1971 to 1982, a progressive rise in accep- WOMarl at each prenatal visit. tance of care was noted, with persistent differ- The pregnant women had to be convinced ences between groups, and with a time lag of to come early and often to meet these new pro- about four years between women with 13 years posals. Then the early enrollment in the obstetri- of schooling and women with 10-12 years, and cal outpatient clinic and the number of prenatal a time lag of eight years between women with visits before the end of the 6th month could 13 years of schooling and women with up to serve as accurate measures of the acceptance of nine years. In the meantime, during this period this care proposal and of the effectiveness of the of progressive acceptance of prenatal care, the information given to the general public and to rate of preterm deliveries went down, with per-

77 72 Decrease and Rise in Rates cf Preterm lkiiivries--11aguenati Prenatal Study. 1971- 1988 Intert,entioits sistent differences between groups. and also Disct ISSION with a time lag difference between groups of The observed mathemthical relationship be- the same order as that observed on the behav- tween the characteristics of care and outcome ior of acceptance of new care. After 12 year:, are very similar to adose-effect relationship. This in the fourth period (1983-1986), instead of feature supports the hypothesis of a causal rela- improvement, a new movement of reduction of tionship between this type of care and the the three measures was observed. The pregnant observed effect. This point is important. as our women sought consultation muchlater in the policy has been applied in the whole of France, course of their pregnancies and cameless often and without the opportunity to apply an experi- in the first two trimesters. In addition, more mental design, as we would like, to support a WOMCII came in directly for deliverywithout demonstration of the effectiveness of our policy. being previously followed. During the same On the other hand, we know the effort to pre- period, the rates of preterm deliveries went up vent preterm deliveries has been applied in for the three observed groups of women. France, but has not been applied in Germany, the United kingdom, or the United States. It hap- pens that France is the only country amongthese Mathematical relationship beturen four to have observed a reduction in preterm prenatal care and preterm delitteries births between 1970 and 1980, and, during this A regression analysis was done to measure same period, no reductioncould be demonstrat- the precise relationships between the preceding ed in the three other countries. Obviously, this is characteristics of prenatal care and the outcome not as good a demonstration as a randomized in terms of the proportion of preterm births. trial would have been; however, this difference The regression curve for women by group and cannot be discarded. number of preterm deliveries is shown in figure We did not expect that France, by itself, 6.1, with the equation Y(pr()ortion of preterm could serve as a control. But the cessation of the births) = X (proportion of pregnant women by national policy in the perinatal field in 1981 has school attendance group enrolled in prenatal provided us with this quasi-experimental design. care at the outpatient clinicof the maternity We can take advantage of this to propose a new hospital in the first trimester). argument fm a causal relationship betweenthe 1' = 0.078-0.07.1X new type of care and preventionof preterm r-0.95 deliveries, as the numbers were exactly the p = 0.001 same on the regression curvebackup, A similar relationship was found lvtween the number of prenatal units at the outpatient REFERENCES clinic of the maternity hospital and the number Papicrnik, E.(19(9 }. Coefficient de risquc of preterm births, and a similar correlation was d'accouchcment prCmature.Presse Medic-ale 793--7(Xl. found between the proportion of women not 2.Pvicrnik, F.(1)8,4).Pniposals fur a programmed followed by the obstetrical team and the num- prevention policy of preterm birth.Clinical ber of preterm births. Obstetrics attd 6:rtwcolfxr 27. 61-4-035.

Emile Papiernik et al. 7$ 7:1; Interventions

Statistical Findings: Decrease and Rise in Rates of Preterm Deliveries Haguenau Prenatal Study 1971-1988

Table 6.1 Preterm Deliveries and Live Births

1971-74 1975-78 1979 82 P Value < 32 weeks 5.4 4.0 3.6 0.001 299/5548 192/4787 210/5811

35-36 weeks 2.9 2.1 2.4 N.S. 159/5548 102/4787 142/5811

33-34 weeks 1.1 0.8 01 0.05 59/5548 .39/4787 39/5811

< 33 weeks 1.5 1.0 0.5 0.001 81/5348 51/4787 29/5811

Reduction ot preterm births ihve-brrthi time perio4 and by duration 01 gestation sun e last 1.81P. The observed redut non in imeterm births is al one birth tOr the total preterm figure's, and is one halt tar the most dangerous preterm births des+ than 3 3 weeks,

Table 6.2 Table 6.3 Transfers to a Neonatal Intensive Care Unit Days of Hospitalization Related to Preterm Delivery

1971-74 1975-78 1979-82 1971-.741975-781979-82 P Value

33 weeks 38/81 27/51 12/29 NICLI 425 296 182 0.001 33-36 weeks 15/218 17/141 17/181 Strasbourg

> 37 weeks 58/5548 17/4787 22/5811 Neonatal 4:37 320 2 23 0,001 The numbers at transfered babies tsere signitie antiv rechrced for term pediatric lubies a, st as far the less than ;1 weeks birth babres. The prepit- ward lion at transfers t ompared to preterit) birth was not redue-ed, the recicie - Haguenau turn was only possible by the eftx rease in early preterm births. Numbers of days of StaV in the intensive c-are unit rat Strasbourg reterral center-I or in the neonatal pediatric ward in liagUNTall hospital tar Thu number at dat s needed tar hospitalizaturn related to preterni delit Ivbies born bethre 15 It evils, atter exc fusion or hospitalizations due to en bias significantly reticle ed tor transferee! babies toNK.0 and to the malfarmatran .N the r/h/tor ( 16r 1,000 birthsr lex al pediatric cvard.

Emile Papienlik et al. Adc.ances in the Prevention of Low Birth Weight

Table 6.4 Table 6.6 Neonatal Deaths Per Thousand Live Births Rates of Preterm Births, When a Risk Factor was Observed or Not Observed 1971-741975-781979-82 P Value 1971-741975-781979-82 P Value 0/00 8.3 6.5 3.1 0.001 46/554831/4787 18/5811 Age less than 7.3 6.5 6.0 N.S. 21 years 70/961 52/801 49/806 Standardized 8.3 7.9 5.0 21-35 years 5.4 rates and 48/96 30/76 3.9 3.6 0.001 confidence 217/4172146/3770177/4851 interval > 36 years 6.7 5.5 5.6 N.S 31/461 10/181 10/180 The upper line shows the observed figures of neonatal deaths among live births by study period. The lower line depicts the figures obtained Bleeding in second trimester after standardization, as if the distribution of births by gestional age would not haw changed in the second or third study periods, and with yes 18.2 20.6 14.3 N.S week-specific mortalib rates ut these specific periods. 27/148 20/97 18/126

no 6.2 3.9 3.8 0.001 Table 6.5 120/1931108/2804164/4371 Rates of Preterm Births, by Risk Factor, Bleeding in third trimester by Period yes 17.1 11.8 23.0 N.S 24/140 8/68 10/87 1975-781979-82 1971-74 PValue no 5.3 4.0 3.6 0.001 179/3392159/3967194/5429 Previous 12.7 12.5 12.3 4-5 preterm 52/408 31/248 33/269 Table 6.7 No previous 4.3 3.2 2.9 0.01 Rate of High-Risk Factors Among Pregnant preterm 124/286871/2247 88/3008 Women by Study Period Previous 10.8 13.2 8.9 N.S. 1971-741975-781979-82 P Value still-born 7/65 7/53 7/79 Previous 12.5 10.0 8.3 0.001 No previous 5.3 .3.9 3.6 0.001 preterrn 428/3418259/2604280/3357 still-born 169/3211 95/2442114/3198 Previous 2.0 2.1 2.4 N.S stillbirth 69[3418 54/260481/3357 Height 8.2 7.8 6.3 N.S. Height 4.6 4.9 4.5 N.S < 1.52 m 19/232 1 7/2 19 16/252 152 cm 242/5292236/4807263/5819 Height 5.6 4.0 3.8 0.001 Height 12.9 14.1 13.6 N.S > 1.52 m 272/4871179/4439209/5478 48 cm 688/5.336664/4721790/5818 Age 21 years 17.4 17.1 14.1 0.001 Weight 7.5 5.4 5.3 1014/:311341/4918839/5948 < 48 kg 50/668 35/645 41/777 Age 35 years 8.4 4.0 3.1 0.001 Weight 5.5 4.2 '3.8 0.001 488/5811196/4915187/5948 > 48 kg 245/4489164/3944186/4951) Bleeding 7.1 3.3 2.8 0.001 2nd trimester151/213499/2961 126/4514 Bleeding 3.9 1,7 1.6 0.001 3rd trimester143[363071/4122 88/5572

76 Decrmseand Rise in RatesPreterm Delireries-Ilaguenau Prenatal Study. 1971-1988 Interventions

Table 6.8 Change in Prenatal Care and Preterm Birth Rate According to ParentalEducational Level

School attendance 1971-74 1975-78 1979-82 1983-86 P Value

2794 2975 3806 2984

2 1965 1640 1 725 2109

3 282 377 433 475

0.001 A 1 0.9 1.8 2.3 2.3

2 1.5 3.0 3.5 2.8 0.001

3 2.2 3.8 3.8 3.1 0.001

1 17.7 28.4 38.6 37.5 0.001

2 30.0 49.5 61.6 48.4 0.001

3 47.5 65.5 69.5 55.2 0.001

0.001 1 46.2 23.9 8.6 18.8

2 27 .7 8.8 3.6 20.7 0.001

3 10.6 4.0 0.9 17.3 0.001

5.2 0.001 1 6.8 5.0 4.5

2 5.6 3.5 3.3 4.3 0.01 3.7 2.7 2.5 3.6

N : numbers ot pregnant 1.somen In school 13 : proportion of f ?regnant Komen ernolled proonlum 01 pregnant w ()men not to/ . lowed at tlu. ohstetric al department. attendam 0 group the (nit Athent hnu of the ohstetnt al department. : rates or preterm births by group and by A : mean number of prenatalisits at the out perr(xl. pationt clinic (4 the obstctru departmt.nt.

Figure 6.1 Regression Curve for Women by Group And Numberof Preterm Deliveries (See Page 7 3)

.07 A t'ear 7 74 75- 781t9- 82 83-86F .06 - t {

n 1+ e 13 I mei CO 11 ( 1) .05 - H SLhoohng C 4 .04 .1 F of; Lf .03 - v" 0.078-0.074N .02 - r = .01 - A p < 0 001

.1 0 .20 .30 .40 .50 .60

Emile Papiernik et al 77 S I The West Los Angeles Prematurity Prevention Project: A Progress Report

CALVIN J. I 1013EL,M.D. Row. L.BEMIS,R.N.P. G. Ros,s, MD. J.ROIWRT BRA( ;ONIER, M.D., PH.D. Mc. >RAVE13. 13EAR BRNANT M. MORI

INTROIM1C11ON During the 1970s, the I. Tnited States observed a dramatic reduction inperinatal morbidity and mortality. On the other hand, the incidence of low birthweight (LBW) births during this period decreased only 11 percent (approximately 1% per year).' Thus, the technological advances in obstetrical and neonatal care, which werethought to impact morbidity and mortality, did not significamly influence the incidence of LBW. Since prematurity, a major component of the low birthweight problem, is now the leading cause of poor Advances in the Prevention of Low Birthweigbt

pregnancy outcome, we must begin to devel- DEVELOPMENT 01, A PREVENTION PROJECT op strategies for prevention. We believe that a prematurity prevention Beginning in 1970, we collected baseline program thould include three major components. data from the west area of Los Angeles County to identify pregnancy-related problems. Using Prenatal Care A),stem these data, we developed a risk scoring system, A well-organized prenatal care system is which was global in design, to idvntify preg- an absolute requirement fin a prematurity pre- nancies at risk for neonatal morbidity and mor- vention program. A new pn)gram encomraging tality. Between 1975 and 1979, our experiences a different philosophy and a change in helped us recognize that our global approach provider behavior will not be successful unless to risk assessment needed to be redefined and the system is prepared to provide a different directed toward identifying patients at risk for type of prenatal care from that customarily specific problems, such as preterm birth, provided in America. intrauterine growth retardation, diabetes, and hypertension." Thus, over time we developed a Risk Assessment ,S:rstem strategy of prenatal care based on levels of risk, A task tOrce was established in 1978 by the with special attention given to the prevention World Ifealth Organization to set forth guide- of specific problems. lines for developing risk assessment systems for Guidelines for the implementation of pro- maternal and child health." This task force rec- grams for the preventkm of preterm birth came ommended that risk assessment systems should from two sources: (1) A risk assessment system, be developed from and tested on the population described by Papiernik in 1969, which formed to which they are to be applied prior to begin- the basis for identifying patients at risk and for ning intervention programs. The process is as developing prevention strategies:4 and (2) pre- folknvs: First, a risk assessment system is devd- liminary data subsequently published by oped from rorospectively collected data and its Ilerron et al. in 1982, which described the con- sensitivity and specificity is determined. Next, tent of a program thought to significantly the risk assissment system is applied prospec- reduce the incidence of prematurity.' tively to validate its sensitivity and specificity. These preliminary data generated tremen- The final step is to test the scoring system in a dous interest in the prevention of prematurity prospective type intervention study using a con- in the linited States. Ompling these guidelines trol population to again test the scoring system's and this interest, we developed a project for sensitivity and specificity. Most risk assessment studying prematurity prevention in our patients systems have not been properly designed and at risk. The following provides a progress tested prior to instituting intervention strategies. report on that project. Data 5:ystent An important component of Our prevention project is our ability to manage the large

,r, 1113 t)

80 Me West Los Angeles Prematurity Prevention Prolect: A Progress Report Inten'entions amounts of data gathered fromeach patient. Our review of the literature suggested thatthe The source document for our program is the etiology of prematurity is most likely multifae- POPRAS (Problem Oriented Perinatal Risk torial, yet stress appeared to be a central issue. Assessment System) perinatal record.- Using In 198.1, we had proposed a hypothesis for the the POPRAS record, we have developed an etiology of preterm labor where stress was felt online interactive computerized network to be a predisposing condition.We therefore referred to as PIDS (POPRAS Interactive Data directed our attention toward interventions System). This network links our west Los which reduced stress and interventions which Angeles prenatal clinics with the scientific data promoted the maintenance of pregnancy. This center located at CedarsSinaiMedical Center. approach was in concert with the interventions ourprojectheadquarters.andthe used by Papiernik. HarborUCLA Medical Center. This system pro- For our prematurity prevention project, we vides us with the following: decided to test five treatment protocols for high-risk women. The precise study design Computerized risk assessment: 1. algorithm was previously published.9 The inter- A randomization scheme for assignment of vention protocols selected were bed rest, psy- intervention protocol; chosocial support, and an oral progestin 3. An appointment scheduling system; (Provera) matched with a placebo and a control (education and nutrition only). 4 . Patient tracking and compliance checks;

5.Real-time reports:

6.Online hcispital data management; and DEscumoN OF PROJECI' 7.Rapid turnaround data management and The patients' risk status is determined analysis. using the PIDS program. Patients who have one or more selectedhigh-risk factors ate considered high-risk patients and are randomized into one of Ilk. five interventions.' In addition FtNDING PREvEN*110N PROJECT to standard risk assessment, adetailed psy- This project is funded by the Maternal and chosocial history is taken but not used in the Child Ilealth Branch of the state of California. assessment of risk. All high-risk patients receive education about preterm labor, signs and symptoms. what actions to take if symptoms occur, and aclass Smin DEslo\ detailing what to expect in the hospital. In We were initially confronted with the task addition, all patients receive extensive nutrition of determining what types of interventions counseling and support. Most patientsdeliver should he tested for preventing prc:erm birth. atI larbort*CLA Medical Center and theirdata

liobel et al. 81 S4 Advances in the Provention of Low Birthweight

are entered into the data system by labor and been published in abstract form.0 In a retrospec- delivery room nurses or data entry clerks. tive study of 359 women with preterm deliveries matched with a group of women delivering at term, 4 factors were found to he significantly associated with preterm birth (see table 7.2). In THEPRELLMINARY REsi LTS our prospective study from 1 January 1979 to 30 Preliminary results from 4,034 total deliver- June 1986, only three factors were determined ies occurring between January 1, 1979. and to be significantly associated with preterm birth June 30, 1986, are presented in table 7.1. These (see table 7.2). These preliminary results suggest data were previously published in abstract that there may he certain psychosocial factors form." The importance of these preliminary which could be used to facilitate the identifica- results is to recognize that during the early tion of patients at risk for preterm birth, phase of our study we had an apparent impact on the incidence of preterm deliveries, espe- daily those with very low birthweight infants (< 1500 g). The very low birthweight infants are SUN1N1AW AM)FunRE DIREcnoN the most costly to care for and account for the As of May 1988, the \Vest Area Los Angeles majority of preterm infants who are left with Prematurity Prevention Demonstraticm Project is disabilities or handicaps. well established and has completed its fifth In addition, a preliminary analysis has been year of operation. The strategy of risk assess- carried out to assess the association of psy- ment, randomized application of specific inter- chosocial stress with prematurity. This has also ventions. and computerized monitoring to

Table7.1 West Los Angeles Prematurity Prevention Project Preliminar y Data

Baseline Prematurity Rate (percentage) Study Prematurity Rate (percentage)" (4,034 patients)

< 37 Completed <31 Completed Weeks of Gestation Weeks of Gestation

1979-1982 1982-1983 Control Experimental Control Experimental (8,249 patients) (:3,108 patients) Clink.s Clinics Clinics Clinics

High Risk 11.74 9.92 8.54 5.70f 1,02 0.6P

Low Risk 5.04 4.58 4.77 5.53 0.41 0.40

Oata as vi 6/30 6(7. 4q",, dv reds.? compared %iith I 979-82 baseline. 40" de( reaw compared tt ith Control (slinics. 4 ?",, increaw ( orrparedrth 198244bdr.e,line. 3 3', in( reawd compared tt ith Control Crofts to= .05.

82 The West Los Angeles Prematurity Prerentiwt Pro/ect: A Progress Report Interventions

assess their impact on pretermbirth appears REFERENCES promising for providing new information to 1. Committee toStudy the Prevention of Low facilitate our understanding of the problems of Birthweight. (1985).Preventinglow birthweight.In: Institute of Medicine. Trends in lowbirtbuteight(pp. preterm birth. 95). Washington, DC: National Academy Press. Randomization of high-risk patients to spe- 2. Hobe!, C. J., Hyvarinen, M. A., Okada, D. M. and Oh, cific interventions will be completed as of 31 W. (1973).Prenatal and intrapartum high risk screen- December 1988, and the project will end by ing.I.Prediction ofthe high-risk neonate.American July 1989, at which time all patients entered Journal of Obstetrics and Gynecology117, 1-9. into the study will have delivered. At that time, 3, Hobe!, C. J.(1979).Assessment of the high risk fetus. Clinicsin Obstetrics and Gynaecology6. 367-377. a final analysis of the overalleffect of the pro- 4. Papiernik, E.(1969).Coefficient de risque ject, as well as the effect of each specific inter- d'accouchement premature.Presse Medicate 77, vention, will be determined. In addition, an 793-796. important part of the final analysis will include 5. Herron, M. A., Katz, M.. and Creasy, R. K.(1982). validation of our risk assessment screening tool Evaluation of a preterm birth prevention program: Preliminary repoit Obstetrics and Gynecololgv 59, 452. in the control population to determine whether 6.World Health Organization. (1977).Something forall it has maintained its original sensitivity and and more forthose in greaternevd: A-risk approach" specificity. Since psychosocial risk factors were for integrated maternal and child healthcare.WHO not originally included in our risk assessment Chronicle31,150-151. model, we will test the inclusion of certain of 7. Hobel, C.J.(1982).Development of POPRAS: these factors into our scoring system to deter- Problem oriented perinatal risk assessment system. In:T. R. Harris and T. D. Bahr (Eds.),Me useof mine whether they would improve the sensitiv- computersin perinatal medicine(pp. 117-139). ity and specificity of our original tool. New York: Praeger Scientific.

Table 7.2 Psychosocial Risk Factors Associated with Preterm Birth

Retrospective Study (Preliminary Results)

I.Mother working throughout pregnancy; 2. Patient not living with the baby's father; 3. Perceived problem or crisis near delivery; and 4. Perceived excessive stress during pregnancy.

Prospective Study (Preliminary Results)

I.Mother born in the United States; 2. Patient having had thoughts of hurting self or unborn baby; and 3. Patient having a close family member die during pregnancy.

Calvin Hobel et al. 83 Advances in the Prevention of Low Birthweight

8, Bragonier, J. R., Cushner, 1. M., and Hobel, C. J. (1984). Social and personal factors in the etiology of preterm birth.In:F. Fuchs and P. G. Stubblefield (Eds.),Preterm birth:Causes, prevention, and management (pp. 64-435). New York: MacMillian. 9.Hobel, C. J., and Bemis, R. L.(1986). The west area Los Angeles prematurity prevention demonstration project.In:E. Papiernik and G. Breart (Fds.), Prevention of preterm birth (pp. 205-222). Colloque INSERM, VoL 138. 10. Ross, M., Hobe!, C. J., Bragonier, R. J., and Bear, M. (1986).Prematurity:A simplified risk scoring system. American Journal of Pertnatology 3, 339. 11. Hobel, C., Bragonier, R., Ross, M., Bear, M., Bemis, R., and Mori, B.(1987, March).Significant impact: West Los Angeles premature prevention program. Abstract presented at the Society for Gynecologic Investigation 34th Annual Meeting, Atlanta, GA. 12. Ross, M. G., Hobe!, C. J., Bragonier, J. R Bemis, R., and Bear, M. (1988, February).Psychosocial stress: Association with prematurity.Abstract presented at the Society of Perinatal Obstetricians, Las Vegas, NV.

84 Me West Los Angeles Prenuiturity PreventionPro/ea:A Progress Report The South Carolina Multicentered Randomized Controlled Trial to Reduce Low Birthweight

HENRY C. HEINS, MD., M.P.H. NANCY WEBSTER NANCE, C.N.M., M.S.N. BRIAN 1 MCCARTHY, M.D. CATHY NIELVIN PAM, PH.D. )LIABORAMI G ROL T*

INTROD1'(:FION South Carolina has one of the highest low birthweight rates (LBWR, < 2500 g) and very low birthweight rates (VLBWR, < 1500 g) in the United States. This dubious distinction contributes heavily to the state's excessive infant mortality rate, also one of the highest in the nation.' As other causes of infant mortality decrease, low birthweight assumes greater importance. In the United States, the LBWR is 7 percent and accounts for 80 percent ofthe infant deaths. Very low birthweight rates have not changed significantly inthe last 20 years.' Adt.ances in the Pretvntion (ye Low Birthuvight

These small infants are not only at high risk for Mim 101)5 dying, but also for long-term morbidity, suchas cerebral palsy and other neurological deficits.' Study Population Several investigators have described inter- Women who attended state-fundedprena- ventions to reduce preterm birth (< 37 complet- tal clinics were eligible for randomization if ed wtA:s' gestation) and low birthweight with they had a score of 10 pointsor greater on a inconsistent results.". This paper reports on a scoring system using the risk factors of multicentered randomized controlled trial (RCT) Papiernik," later modified by Creasy.'2 Public of an intervention used in an attemptto reduce health nurses in prenatal and the Special the LBWR in a group ofwomen identified as Supplemental Food Program for Women, being at increased risk for thispoor perinatal Infants and Children (WIG) clinics administered Outcome. the scoring index after being trained in itsuse. Experience 1th two previous South In addition, women who had a low birthweight Carolina programs provided the basis for the infant in their last pregnancy were eligible for hypothesis tested in this study. Oneprogram randomization. All women randomized had to used nurse-midwives in the adolescent clinics he free of known medicalor pregnancy com- of the Medical University of South Carolina plications (i.e., hypertension, diabetes, renal (MUSC) to provide high-quality, comprehensive disease, or multiple pregnancy). Women with prenatal care to adolescents. Thisprogram. these complications were eligible for the although not an RCT. reported a reduction of statewide high-risk medical piligram andwere the LBWR in these young women." The other not rand(imized. If a randomized woman deyel- program used public health nurses to success- oped any of these complications during the fully screen for high-risk patients in the index pregnancy, she continued in the study statewide high-risk perinatal program.'" The and also received treatment from an obstetri- stated hypothesis was that the low birthweight cian for the complications. rate in the intervention group would be less than the LBWR in the control group. 'Random ..11/ocation A list of etnuputer-gcnerated randomized numbers was sealed in sequentially numbered o'reenrille Hospital Svsu,nt: Turn (,ailey, M.1).; kirri Mc Own. LAM: Ray Re.milds. RA; Libby ,s,leelmall. RA; opaque envelopes. Upon identification of an McLeod Regional Medical Center: Steve Adams. .11 IX. eligible patient in the clinic, a central telephone Linda 14neson. kV Marilyn Polony. kV: Beth Turner, number was called and the next envelope was lnirers.ity South Caroling: Ilenr) c. chosen by a lay administrator not involved in ['COM %!.I)Nanc.V Nance, cAllf: lanna llliui.. c:VM: Patricia PcIpie, CAM: Barham Stone. CAM; R(ihntt patient care. Patients were assigned to attend Stone. R.V; Rich/mut Momirial janke either a separate low birthweight prevention M.D.; Barbara France. CAM: .Spartanbutg RoVonal clinic (study group) or to attend the regular Medical C:entcr. Hal Rubel. M.D.: Pat Rubel. R.V, South c`..arolina Depannient vj Health and Environmental high-risk obstetrical clinic (control group), Both (.ontrol.. I.'ienise Ingram. .11.1). groups had access to the WIG program, nutri-

S Me South Carolina Multicentered Randomized Onitrolled Thalto Reduce Low Birthuvight Interventions tionists, public health nurses, and complete palpate for contractions. They were given the funding for prenatal care by the state health phone number of the low hirthweight clinic department. All women received intrapartum nurse and the labor roomand encouraged to care by obstetricians at thefive regional centers call with any problem. The return visits lasted collaborating in the trial. 20-30 minutes, with a review of all the initial teaching done each time. The same nurses or Sample Size nurse-midwives saw the patient at one- to two- From the beginning of the study, it was felt week intervals throughout the piegnancy. A that it would be essential to pay for the prena- gentle cervical exam to monitor possible cervi- tal care of both the program and control cal change was done at each visit, with care patients. Funds were made availablefrom mul- taken not to enter the endocervical canal. tiplesources(e.g.,the jobsBill,the Recommendations to decrease physical activity Department of Health and Environmental were based on theinformation gathered during Control, and the March of Dimes) for support the visits. of approximately 1,600 total patients. The sam- Intensive follow-up was done by the nurs- ple size necessary to detect a change from a es/nurse-midwive.s on women who had mi.ssed LBWR of 13% (the estimated low birthweight appointments. The assistant study directormade rate of the high-risk group) to8% (the state quarterly site visits to each of the centers to LBWR) at p 5...05 with a power of 90 wascalcu- review charts and monitor adherence to the pro- lated to he 1,546 patients. The sample size tocol. Case loads were assessed to ensure that determination to test a similar hypothesis for the time allotted for each patient wasadequate the VLBWR exceeded projected resources. The to maintain the protocol. Inaddition, staff from original sample size of 1,54> was, however. all five regional centers met quarterly todiscuss reduced to 1,435 women due to limited problems and receive feedback on data collec- resources extended over aprolcmged peric)d of tion but without knowledge of perinatal out- time. comes in either the program orcontrol group.

hitervention (msent As soon as possible after randomization The Institutional Review Boar.; of the (within three weeks), the women in the pro- MedicalUniversityof SouthCarolina gram group vvere seen inthe low birthweight determined that no f()rmal consent was neces- pret.vntion clinic. An initial hour-long, one-on- sary for women to enterthe study since there one teaching sessionidentified the women's was no unusual orhazardous risk involved in specific risks with careful analysis of lifestyle prenatal care given in either group. behaviors, including substance abuse, nutrition status, stress, level of activity,and social sup- Interim Anal)wis port. Patients were taught to recognizethe sub- After accumulating 800 pregnancy out- tle signs and symptoms of preterm laborand to comes, an interimanalysis was conducted by a

Henry lkins et al. 89 Advances in the Prevention of Low Birthweight

team of three investigators who were not total of 17 women had multiple gestation, 11 in directly involved in the clinical management of the program and 6 in the control group. Six the trial. All were independent of the study and women were not pregnant, three in each the institutions associated with it. The data group. Of the remaining 1,435 singleton preg- review group met in Atlanta on 26 March 1986, nancies, 34 resulted in abortions (22 program together with the principal investigator of the and 12 control.). There were 13 fetal deaths (3 study and two consultants from the Centers for program and 10 control). Forty-two patients Disease Control. were lost to follow-up (22 program and 20 The analysis concentrated on live births control). All of the above women were exclud- since they represented the denominator of the ed from analysis either because they did not low birthweight rate, which was the measure have a live singleton birth or because the out- used in the hypothesis. The predictive power come of their pregnancy was unknown. All of the screening tool for low birthweight and exclusions were validated by a blinded observ- very low birthweight was two to three times er not connected with the study. The profile of higher than for U.S. population rates. There the exclusions does not vary significantly with was essentially no difference in the low birth- the profile of those women included in the weight rate between the groups. Similarly, the analysis (see table 8.1). mean birthweight did not differ. The proportion A total of 1,346 women with live births and of live births with birthweight less than 1500 g, known outcome, 667 in the program group and *however, was 2.1 percent in the program group 679 in the control group, were included in the versus 4.2 percent in the control group. The reported analysis. The randomization process apparent effect of the program was to shift achieved comparability between groups for births from the very low birthweight to the maternal race, education (in years), marital sta- moderately low birthweight group, with no tus, age (in years), risk scores, gravidity, and apparent shift from the moderately low birth- gestational age at randomization at clinic sites weight group into the normal birthweight (see table 8.2). Nearly one-third of the patients group. The d'ita review group ccmcluded that were less than 16 gestational weeks when ran- there were no reasons for stopping the trial. domized. Randomization of the patients to the encouraged its continuation, and kept their program and control occurred evenly within report confidential from the project staff. each hospital. One site (MUSC) accounted for 30 percent of the patients (see table 8.3). The group birthweight distribution for both the program and the control groups is present- RE.st.urs ed in table 8..4. There were no statistically sig- The trial began on 1 .1uly 1983, and ended nificant differences, either in the very low on 31 October 1987. During this period, 1,458 birthweight rate or in the low binhweight rate, women were randomized, 728 to the program between program and control patients. The and 730 to the control group (see figure 8.1). A odds ratio for program versus control patients

91 try^ 90 The South Carolina Alulticentered Randomized Cwitmlled Trial to Reduce Lou' Birthweight Interventicits for LBW was 1.09 (95% CL, 0.8 1.4), and for and acceptance of low birthweight prevention VLBW was 1.15 (95% CL, 0.7-2.0). programs by social class of thepatient. Race was not considered in the hypothesis. Main et al.,in a randomized controlled hut interest as to whether intervention may trial with inner-city patients, failed to detect any have affected risk groups differently led to an increase in gestational duration, but their study analysis by risk category and race. This post- had a very small sample size (60 patients in hoc analysis suggested that black program each group). They subsequently reported dis- women with a risk score of 10 or morehad a appointment in the lack of sensitivity and statistically significant lower very low birth- specificity in efforts to validate Creasy's scoring weight rate than black control women (see system.r table 8.5). When the risk group""9 was exam- Moore described an educational program ined, no significant difference occurred in ges- for doctors and patients with much more suc- tational age for either the total population or cess in private patients,but little change detect- for any one particular race (see table 8.5). ed in results in public health programsfor less advantaged women.' Meis and Moore reported the etiology of preterm labor in the patient population to be a factor in the success of th,.,ir DISCUSSION prevention program.19 This randomized controlled trial with The value of social support in improving patients in public health clinics demonstratedlit- perinatal outcomes has been the subject of tle change in the low birthweight rate.Effective studies by Oakley' and Spencer.21. Differences interventions have been reported in some in bitthweight were found in South Carolina in patient populations, but have not been replicat- a case control studyof social support using a ed by other investigators. Thus, itis becoming resource mother favoring thiskind of social apparent that low birthweightprevention pro- support." grams involvingrisk assessments which target In a state where the ac.c.css to andavail- specific populations for interventions do not ability of prenatal care for indigent women at benefit all populations to the same degree. It risk for low birthweight outcomes is an increas- appears that the importanceof specific risk fac- ing concern, this study provides important tors for low birthweightand their corresponding insight as to the future direction that research intervention may depend on the population shoukl take. It appears from this RCT that nurs- studied. es/nurse-midwives may have provided prenatal Papiernik first claimed benefit from assess- care to a high-riskpopulation with comparable ment and early intervention inFrancel' with results, rather than better results (as stated in education, cervical check. cerclage, and/or the original hypothesis) when compared to out- tocolysis. Creasy modified Papiernik's risk fac- comes of womenprovided care by obstetri- tors in the pilot study inCalifornia." Papiernile cians in high-risk clinics. Thishypothesis has also demonstrated general improvement .should be tested in a future study. The sample

Henry Heins et al. 9 91 Advahces in the Prevention of Lotv Birthuvight size necessary to test this hypothesis is not pro- dated with mounting further research is mini- hibitive and the study would provide important mal compared to the savings in human and public health information. financial terms if a 50 percent reduction in the The trend of black women in the program VLBWR is potentially achievable. group to do better than black women in the control group is even more encouraging ACKNOWLEIXiMENTh because it suggests that some component of The authors wish to acknowledge the their persistently higher low birthweight rate encouragement and scientific advice of Ian may be more amenable to this intervention Chalmers, Director, National Perinatal Epidem- than their white counterparts. This observation, iology Unit, Oxford, England. We also wish to however, was not in the original hypothesis, express our grititude to the March of Dimes, and needs to be studied specifically by repeat- the South Carolina Department of Health and ing this RCT with that as the stated hypothesis. Environmental Control, and the Robert Wood Possibly the difference in low birthweight/voy Johnson Foundation for their support of this low birthweight by race was related to access report. to quality prenatal care. Greenberg" noted a greater effect of prenatal care in those patients ft.F.FEREN('ES at greatest socioeconomic risk. KorenbroVs reported a trend toward lowering of the low 1. Kleinman, J. (1988). Natality Report 1983- /985 tcumuLitive).11yattsville, N11): National Center Health birthweight and very low birthweight rates with Statistics. a package of quality care. 2. Wegman. M. E. (1987). Annual summary of vital The unchanging rates of low birthweight swtistics - 1986. Pediatrics 80,817-827. preterm births should be recognized as a very 3. Taffel, S. (1980). Factors associated with birth- serious maternal and child health problem. u.eight ('nited States, 19 76 t Pub. No. (1)115). 80- 1915). Hyattsville. MD: National Center for !leak') Despite improvements in infant mortality, the Statistics. United States compotes unfavorably with devel- -+. NicCormick, NI. C. (1985). The contrthution of low oped nations.= Our birthweight-specifk mortal- binhweight to infant mortality and childhood mor- ity (i.e., survival within birthweight groups), bidity. Neu England journal (y. Medicine 312. 82-90. however, ranks with the best in the world Shapiro. S.. McCormick. M. C.. starfield, B. H., Intensive care crisis intervention is seen as ltrischer, J. P. and Bross, 1). (1980), Relevance of correlates of infant deaths for significant morbidity at the pinnacle of our achievement and a measure one year of age. .-tinerican Journal (?I' ohstctrics and of our SUCCCSS. Lost from sight is the fact that C.,:vnecaogy 136. 36$-373. the neonatal intensive care unit (NICU) should 6. l'apiernik-Berkhauer, E. (1969). Coefficient de risque serve as a backup adjunct to prevention. One d'accouchement premature. Presse .ift;dicale 77. should not let the previous strides made in 793-796. technological and expensive treatment over- Johnson. J. W. C.. Austen. R.L.. and .lones. C. S. (1975). Efficacy of 17-alpha hydroxyprogesterone shadow the small hut important steps that can coporate in the prevention of premature labor.Neu, be made toward prevention. The expense asso- England journal cy'Afedicitw293. 675-680,

.)C)

92 Me South Carolina Muiticentered Randomized Controlled Trial to Reduce Low Birthweight Interventions

8.Herron. M., Rati. M.. and Creasy. R. K. (1982). 21. Spencer, H.. Morris, J., and Thomas, H. (1987). South Evaluation of a preterm birth preventkm program. Manchester family worker scheme. Health Obstetrics and Gyrwcology 59, 452-45(, Promotion 2,1.

9.Piechnik. S., and Corbett. M. A. (1985). Reducing kw 22. Heim, H. C., and Nance, N. 'W. (1987). Social sup- birthweight among socioeconomically high risk ado- port in improving perinatal outcome: The resource lescent pregnancies. Journal (?1* Nurse MiduoijiTy 30. mother program. Obstetrics and Gynecology 69. 88. 263-266. 10. Heins, H. C.. Miller. J. M., and Sear. A. (1983). 23. Greenberg, R. S. (1983). The impact of prenatal care Benefits of a statewide high risk perinatal program. in different social groups. Amerkan Journal of Obstetrics and Gynecology 02. 29,4-290. Obstetrics and Gynecology 1.43, 397-801. 11. Papiernik-Berkhauer. E. (1980). Prediction of the 2.4. Korenbrot. C. (1983). Risk reduction in pregnancies preterm baby, clinical Obstetrics and Gynecology 11. of low income women: Comprehensive prenatal care 315-319. through the Oh Access Project. Mobios 34. 12. Creasy. R. K.. Geimmer. B. A., and Liggins. G. C. 25. lichrman, R. E. (1985). Preventing low hirthweight: A (1980). System for predicting spontaneous preterm pediatric perspective. Pediatrics 107.8-H-849. births. Obstetrics and Gyttecology SS, 692-698, 20. Lantos, J. (1987). Sounding board-Baby Doe five 13. Papiernik-I3erkhauer. E. (19(9). Coefficient de risque years later: Implications for childhealth. New craccouchement prematme. Press(' Medicale 77. England _Journal qic Medicine 31, 44-4-440. 793-790. 14. Herron. M., Ratz. M.. and Creasy. R. R. (1982), Evaluation of preterm birth prevention program. Obstetrics and Gynecology 59,452-,50. 15. Papiernik. F... Bouyer, J.. and Yaffee. K.. et al. (1986). Women's acceptance of a perinatal birth prevention program, ilmerican Journal ofObstetrics and GyrwcohR1*155. 939-940. 16. Main. D. M.. Gabbe. S. C ,Richardson. D.. et al. (1985). Can preterm deliveries be prevented? AnteriCcill lottrnal (?I Obstetrics and Gynecology 151. 892-898. 17, Main. D. M., and Gabbe, S. C. (1987). Risk scoring for preterm labor: Where do we go from here? Anler1Catt Journal cy Obstetrics and 6:171ecolugY I. 789-.793. 18. Moore. M. L.. Meis. P. J.. Ernest. J. M., et al. (1986). A regional program to reduce the incidence of preterm births. Journal Perinatology210-22o.

19Meis. P. J.. Ernest. J. M.. Moore. M. L.. et al. (198). Regional program for prevention of premature Nrths in northwestern tic mb Carolina. Amencan jutirnal obstetrics and (,yttecolQqr 157,550-Sio. 20. Oakley. A., Macl'arlarw. A.. and (halmers, I.( 0)82). Social class stress and repro)duction. In: A. R. Rees and 11, Purcell (Eds.). Disease environment. New York: John Wiley and sons Ltd.

Henry Heins et al. 93 1ntementions

Statistical Findings.. The South Carolina Multicentered RandomizedControlled Trial to Reduce Low Birthweight

Figure 8.1 The South Carolina Multicentered Randomized Controlled Trial to Reduce LowBirthweight

Randomized Patients Exclusions 1,458

Multiple Gestations Nut Pregnant 17 6

Program Control Program Control

1 1 3 3

Total Birth,Singletons 1, 35 Alxrtions Fetal Peaths 47

Program (.ontrol 25 22

Total Births, Singletons 1,388

Moved 11 42

Program Control 22 20

Live Births Known Outcomes 1,346

Henry Heins et al. 95 Advances in the Prevention of Low Birthuvight

Table 8.1 Selected Maternal Characteristics South CarolinaRandomizedControlled Trial

Patient Analytical Status Maternal Total Singletons Excluded Analytical Group Characteristic* Number Percentage Number Percentage Number Percentage

Total Participants 1,435 100 89 6.6 1,346 93.4

Race White 759 52.9 41 46.1 718 53.3 Black 648 45.2 34 38.2 714 45.6 Other 6 0.4 0 0.0 6 0.4 Missing 22 1.5 14 15.7 8 0.6

Education (Years) 12+ 110 7.7 7 7.9 103 7.7 12 .392 27.3 19 21.3 .37.3 27.7 <12 887 61.8 47 52.8 840 62.4 Missing 46 3.2 16 18.0 30 2.2

Marital Status Married 633 44.1 33 :37.1 600 44.6 Single 772 53.8 34 38.2 738 54.8 Missing 30 2.1 22 24.7 8 0.6

Age (years) Under 18 249 17.4 4 4.5 245 18.2 18-34 1,069 74.5 32 35.9 1,037 77.1

Over 34 43 3.0 1 1.1 42 3.1 ) Missing 74 5.1 .)r. 58.4 12 1.6

Risk Score 0-9 126 8.8 8 9.0 118 8.8 10-19 1,056 73.6 58 65.2 998 74.1 20-29 176 12.3 9 10.1 167 12.4 30+ 43 3.0 3 .3.4 40 3.0 Missing 34 2.4 11 12.4 1.7

Grayida

1 264 18.4 7 7.9 257 19.1 2-4 915 6.3.8 56 62.9 859 63.8 5+ 231 16.1 1.3 14.6 218 16.2 Missing 25 1.7 13 14.6 1.) 0.9

* No statistically significant difference occurred between program and control for any of thnarameters above.

0 0

6Statistical Findings: South Carolina Multicentered Randomized Controlled Mal to Reduce Lou. Birthweiglit Interventkms

Table 8.2 Selected Maternal Characteristics South Carolina Randomized Clinical Trial

Patient Type Maternal Program (P) Control (C) Characteristic Number Percentage Number Percentage

Total Participants 667 679

Race White 310 46.5 304 44.8 Black 348 52.2 370 54.5 Other 5 0.7 1 0.1 Missing 4 0.6 4 .6

Education (years) 12+ 55 8.2 48 7.1 12 183 27.4 190 28.0 <12 421 63.1 419 61,7 22 3.2 Missing ,8 1.2

Marital Status Married 300 45.0 300 44.4 55.1 Single 364 54.6 374 Missing 3 0.4 5 0.7

Age (years) Under 17 127 19.0 118 17.4 18-19 85 12.7 114 16,8 20-34 427 64.0 411 60.5 35+ 17 2.5 25 3.7 Missing 11 1.6 11 1.6

Risk Score 0-9 61 9.1 5'1 7.5 10-19 490 73.5 508 74.8 20-29 85 12.7 82 12.1 30+ 18 2.7 27 3.2 Missing 13 1.9 10 1.5

Gray ida 123 18.1 1 134 20.1 2-4 411 61.6 448 66.0 5+ 117 17.5 101 16.0 Missing 5 0.7 7 1.0

Henry Heins et al, 97 Advances in the Prevention of Low Birth weight

Table 8.3 South Carolina Randomized Controlled Trial Service Statistics (Live Births), Analytical Group

Year of Total Program Control Recruitment Number Percentage Number Percentage Number Percentage

1,435 100.0 667 100.0 679 100.0 1983 125 9.3 62 8.9 63 8.8 19E4 405 30.1 194 28.0 211 31.0 1985 420 31.2 207 30.5 213 30.8 1986 371 27.6 191 28.0 180 26.3 Missing 25 1.8 12 4.3 11 3.3

Gestational Age at Randomization (Onset ofCare) <12 81 6.0 29 4.3 52 7.7 12-16 247 18.4 114 17.1 133 19.6 17-20 252 18.7 132 19.8 120 17.7 21-24 251 18.6 128 19.2 123 18.1 25-29 313 23.3 188 28.2 125 18.4 >30 124 9.2 63 9.4 61 9.0 Missing 78 5.8 13 1.9 65 9.6

Clinic Site GHS* 180 13.4 95 14.2 85 12.5 MRMC 233 17.3 109 16.3 124 18.3 MUSO 487 36.2 242 36.3 245 36.1 RMH" 190 14.1 93 13.9 97 14.3 SGHI 255 18.9 127 19.0 128 18.9

Missing 1 0.1 1 0.3 0.0

*Greenville Hospital System 'McLeod Regional Medical Center 'Medical University of South Carolina Richland Memorial Hospital 'Spartanburg General Hcspital

98Statistical Findings: South Carolina Multicentewd Randomized Controlled Trial to Reduce Lou Birthuvight Interventions

Table 8.4 Birthweight Outcomes South Carolina Randomized Clinical Trial (Live Births)

Patient Type Birthweight Program (P) Control (C) Group Number Percentage Number Percentage Odd Ratio (95% CL) C vs. P Total VLBW 24 3.6 28 4.1 1.15 (0.7-2.0) LBW 103 15.4 111 16.3 1.09 (0.8-1.4) NBW 540 81.0 540 79.6 Total 667 (100.0) 679 (100.0)

Table 8.5* Birthweight Outcomes South Carolina Randomized Clinical Trial (Live Births)

Patient Type Birthweight Program (P) Control (C) Group Number Percentage Number Percentage Odd Ratio (95% CL) C vs. P White VLBW 11 3.5 8 2.6 0.73 (0.3-1.9) LBW 38 12.3 33 10.8 0.83 (0.5-1.3) NBW 261 84.2 263 86.5 Total 310 100.0 304 100.0 Black VLBW 12 3.4 20 5.4 1.6(0.8-3.3) LBW 64 18.4 76 20.5 1.25 (0.9-1.8) NBW 272 78.2 274 74.1 Total 348 100.0 370 100.0

Risk Group 10-19

White VLBW 6 2.4 5 2.0 0.83 (0.2-2.7) LBW 27 10.8 24 9.6 0.86 (0.5-1.5) 216 86.7 221 88.4 Total 249 100.0 250 100.0 Blaik VLBW 6 2.6 17 6.7 2.85 (1.1-7.3) LBW 40 17.0 46 18.0 1.35 (0.9-2.1) NBW 189 80.2 192 75.3 Total 235 100.0 255 100.0

*Table 8.5 includes only white and black infants.

Hemy Heim et al. 9 9 99 A Prematurity Prevention Project in Northwest North Carolina

PAUL J. MEIS, M.D. PAUL A. Bt TESCI ER, PH.D. J. M. ERNEST, M.D. MARY Lou MOORE, R.N.C., PH.D. ROBERT MK:I-11E11TM, P11.D. PENNY SHARP, M.En.

INTRODUCTION Premature or low birthweight (LBW)births are a major causeof infant mortality and neonatal and postneonatal morbidity." Rates of LBW and very low birthweight (VLBW) births are high in southern states of the United Statesand have not shown great improvement overthe past decade.' Anexample of this lack of improvement is shown in the ratesof LBW and VLBW births in northwest North Carolina from 1980 to 1985 (see figure 9.1). Although the basic mechanism or cause of most premature births is notknown, prevention strategies may be of some use.Papiemik' and Creasy' have described their experiences Advances in the Prevention of Low Birthweight

with a system of risk assessment, patient educa- ing tocolytic drugs when appropriate. The pro- tion, intensive prenatal care, increased maternal gram was presented to private obstetricians and rest, appropriate use of tocolytic drugs, and family physicians and to county health depart- other methods which were successful in ment clinics throughout the region. Details of improving birth outcomes. Papiernik'4 reported the development and presentation of the pro- success with this program in a number of differ- gram have been described elsewhere." ent populations in metropolitan France and All health department clinics in the region Martinique; Creasy' reported success in patients chose to participate in the project. A portion of of a clinic in San Francisco. Recently, however, the private physicians chose to participate, and these encouraging reports have been tempered the remainder of the private providers chose by results from randomized prospective trials of not to participate in the program. Since patients this risk assessment-educational model. Main" of public health department clinics are low- reported a lack of effectiveness in a small group income patients known to be a high-risk group, of patients in Philadelphia, and early reports of the results from these patients were excluded the multicenter March of Dimes trial have not from the present evaluation and comparisons indicated effectiveness in the patients studied.* were made between births of patients of The purpose of this chapter is to report on providers who participated in the program and our experience with a LBW or prematurity pre- births of patients whose physicians chose not to vention project in northwest North Carolina. participate. Using birth data obtained from the North Carolina State Center for Health Statistics, birth outcomes were examined for residents of Perinatal Region 2 for 1985 and 1986, the first MEM-10M two years of full project implementation. Births This project was implemented in the 20- of less than 500 g were excluded from the county area in northwest North Caroiina consti- results, as were births to women who did not tuting Perinatal Care Region 2. This region is a receive prenatal care. Comparisons were then mixture of urban and rural counties with made between birth results of private patients approximately 21,000 births per year, approxi- enrolled in the program and private patients not mately one-fourth of North Carolina's total enrolled in the program. number of births per year. The program was based on the principles of Creasy and included the following: Risk assessment and education for all patients, routine vaginal examination at RESULTS 26 weeks' gestation to evaluate cervical change, In 1985 and 1986. 12,704 births occurred to more intensive prenatal care for patients private patients enrolled in the program and thought to he at risk, and hospital care includ- 23,757 births occurred to women whose physicians chose not to enroll patients in the pro- Morton, R. Personal communication. gram. The results are shown, categorized by

102 A Prematurity Prevention Project in Northwest North Carolina Interventions

race, in table 9.1 and figures 9.2-9.5.Births list- greater proportion of the nonprojectbirths ed as "nonwhite" are almost entirely of black were nonwhite or black,differences in birth race, as very few other nonwhitebirths occur outcomes remained when theseresults were in this region. These results are shown as the examined by race. An earlier evaluation of percentage of births less than 1500 g (VLBW), LBW births in Perinatal Care Region 2, conduct- less than 2500 g (LBW), less than 38 weeks' ed in 1985 by Buescher, used a multivariate (266 days) gestation, and less than 38 weeks' analysis. After correction for a number of risk gestation and less than 2500 g (premature low factors, including age, race, education, and birthweight EP-LBW) births). marital status, the relative odds ratio for LBW The percentage of births less than 2500 g birth to women not enrolled in the project was was significantly lower forboth white and non- 1.32 higher than for women enrolled in the white patients enrolled in the project compared project. with patients not enrolled in the project. The Although better birth outcomes occurred to percentage of births less than 1500 g waslower nonwhite or black women enrolled in the pro- in patients enrolled in the project, but thisdif- ject, the rates of premature or LBW births ference did not reach statistical significance in remained roughly twice as high for nonwhite nonwhite patients, as the absolute number of births as for white births. This disparity was births was small. Premature births (less than 38 seen both in project and nonprojectbirths. wteks' gestation) were less frequent in non- The difference in the apparent success of white patients enrolled in the project. No signif- this prevention model in this project (and in icant difference was observed for rates of the reports of Papiernik and Creasy) compared premature births in white patients orfor rates with the reports of Main and the March of of P-LBW births in white or nonwhite pdtients. Dimes trial may be related to characteristics of the populations involved. We have previously reported that LBW births in patients of a public health department clinic are frequently related DISCI .NSION to premature rupture of the fetalmembranes, Clear differences exist in birth results of impaired fetal growth, or medical problems.'" private patients enrolled in the program com- Prevention strategies are unlikely to affect the pared with those not enrolled. As all public outcome in pregnancies with theseproblems. patients in Region 2 were enrolled in the pro- In contrast, in our previous study, private gram. no similar comparison groupexists. patient LBW births were more likely to be relat- Evaluation of results in the public patients by ed to idiopathic premature labor, which may be other methods is under study. favorably influenced by prevention techniques. The results of this study may be biased by Thus, the relative success or effectiveness of differences in the characteristics of patients par- the Papiernik-Creasy prevention model may ticipating in the project compared with those depend in part on the characteristics of the not participating in the project.Although a population for whom it is employed.

Paul J. MeiS et cii. 103 102 Advances in the Prevention of Low Birthweight

Although a difference was seen in the rates 3. National Center for Health Statistics. (1984). Advance report on final natality statistics, 1982. NCHS Monthly of LBW births between women enrolled in the Vital Statistics Rcpori 33(Suppl.), 6. project and those not enrolled, no difference 4. Papiernik, E., Bouyer. J., Dreyfus, J., et al. (1985). was seen in the rates of P-LIAV births. Kessel et Prevention of preterm births: A perinatal study in al. have previously found that national trends Ilaguenall, France. Pediatrics 76, 154-158. in P-LBW birth rates may be more resistant to 5. Creasy, R. K., and Herron, NI. A. (1981). Prevention improvement than those of term LBW birth of preterm birth. Seminars in Perinatolgy 5, 295-302. rates." 6.Main, D. NI., Gabbe, S. G.. Richardson. D.. and The apparent improvement in the rates of Strong, S. (1985). Can preterm deliveries be prevent- VLBW births to patients enrolled in this project ed? American Journal of Obstetrics and Gynecology 151, 892-898. is encouraging. These births arc the most likely to have adverse outcomes," and even modest 7. Nleis. P. J, Ernest, J. NI., Moore. NI. L. Michielutte, R.. Sharp. P. C., and Buescher, P. A. (1987). Regional reductions in the rates of VLBW births can have program for prevention of premature birth in north- significant social and financial impact." western North Carolina. American journal qf In summaiy, the results of this project indi- Obstetrics and Gynecology 157, 550-556. cate that this prematurity prevention model can 8. Moore, NI. L.. Meis, P. J.. Ernest, J. NI., Michielutte, R.. be utilized in private patients in a southern Sharp, V. Grover, C., and Hill, C. (1986). A regional program to reduce the incidence of preterm birth. state. Improvement in LBW rates occurred in journal (il. Perinatology 6. 216-220. both white and black patients of private 9.Buescher, P. A., Meis, V J.. Ernest, J.NI., Moore, M. providers who utilized the program, L.. Michielutte. R.. and Sharp. P. (1988). A comparison of women in and out of a prematurity preven- ti(m project in a North Carolina perinatal care region. AMMAN/El ximENT A nwrican Journal qf Public Health 78, 264-267. 10. Meis. P. J.. Ernest. J. M.. and Moore, NI.L. (1987). The authors are indebted to Michael Causes of low birthweight births in public and pri- Patetta of the North Carolina Center for Health vate patients. American Journal of Obstetrics and Statistics for his assistance in collecting perti- Gyrwcology 156, I 165-1168. nent data from birth certificate information. 11. Kessel S SVillar, J.. I3erendes, H. W.. and Nugent, R. P. (1982). The changing pattern of low birthweight in the United States. Journal 4 the American Medical Associatiori 251. 1978-1982. REFERENCES

I.Lee, K., Paneth. N Gartner. L M., and Pearlman, NI. (1980). The very low birthweight rate:Principal predictor of neonatal mortality in industrialized populations. Journal qf Pediatrks 97, 759-764.

2. Skeoch, C.. Rosenberg, K., Turner. T., Skeoch. H., and McIlwaine. G. (1987). Very low birthweight survivors: Illness and readmission to hospital in the first 15 months of life. British Medical Journal 195, 579-580.

104 A Prematuray Prevention Project in Northut,st North Carolina Inteiventions

Statistical Findings: A Prematurity Prevention Project in Northwest North Carolina

Table 9.1 Figure 9.1 Private Births in Northwest North Carolina Rates of LBW and VLBW Births in Northwest 1985-86 North Carolina*

Project Non-Project Total Births12,704 23,715 X' Percentage Breakdown: < 1500 g 0.75 1.33 25.19 < 0.005 < 2500 g 5.94 7.32 25.06 < 0.005 < 38 wks 9.54 10.71 12.09 < 0.005 P-LBW 3.83 4.11 1.67 NS

1980 1981 1982 1983 1984 1985 White Births 11,035 19,051 Percentage rates of LBW (500-2500 g) births in north- Project Non-Project west North Carolina are shown in the upper scale of the Total Births 12,704 23,715 chart and rates of VLBW (500-1500 g) births are shown in the lower scale. Percentage Breakdown: < 1500 g 0.60 1.04 15.95 < 0.005 < 2500 g 5.35 6.05 6.34 < 0.02 Figure 9.2 < 38 wks 8.62 8.87 0.55 NS P-LBW 3.41 3.52 0.26 NS Rates of LBW Births Within the Project vs. Rates Among Unenrolled Private Patients*

Non-Whitc- Births 1,669 4,664 Project Non-Project Total Births12,704 23,715 Percentage Breakdown: < 1500 g 1.74 2.1.5 1.57 NS < 2500 g 9.83 12.09 6.21 < 0.02 1980 1981 1982 1983 1984 1985 < 38 wks 15.58 18.55 8.07 < 0.005 Non-White P-LBW 6.58 6.56 0.00 NS p < 0.005 p < att.? p < 0.02 * Percentage rates of LBW (500-2500 gt births in private patients enrolled in the project (gray bars) are compared with rates of LBW births in private patients not enrolled in the project (white bars).

Paul Meis et at. 105 104 Advances in the Prevention of Lou, Birthweight

Figure 9.3 Figure 9.5 Rates of VLBW Births Within the Project vs. Rates Rates of P-LBW Births Within the Project vs. Rates Among Unenrolled Private Patients* Among Unenrolled Private Patients*

2.5 7 Project 111 Project 6 2 fj Non-project 5 O Non-project 1.5 4

1 3 2 0.5 1 0 Total White Non-White Total White Non-White p < 0.005

* Percentage rates of VLBW (.500-1500 g) births in private * Rates of premature low birthweight (P-LBW) births in patients enrolled in the project (gray bars) are compared private patients enrolled in the project (gray bars) are with rates of VLBW births in private patients not en- compared with rates in private patients not enrolled in rolled in the project (white bars). the project (white bars).

Figure 9.4 Rates of Premature Births ( < 38 Weeks' Gestation) Within the Project vs. Rates Among Unenrolled Private Patients*

20 IIProject 15 1:1 Non-project

Total White Non-White p < 0.005

* Percentage rates of premature births ( < 38 weeks' gestation) in private patients enrolled in the project (gray bars) are compared with rates of premature births in private patients not enrolled in the project (white bars).

105

106 Statistical Findings: A Prematurity Thevention Project in Northuest North Carolina The Family Workers Project: Evaluation of a Randomized Controlled Trial of a Pregnancy Social Support Service

BRENDA SPENCER, PH.D.

INTRODIJCTION The Short Report, published in 1980 by the house of Commons Social Services Commit- tee highlighted widespread variations in perinatal health among different areas of the country. Birthweight isacknowledged to be one of the bestavailable indicators of perinatal health,' and, using this indicator, the Manchester region compared badly with the country at large. Here, 10.2 percentof all newborn infants weighed less than 2500 g, compared with a figure of 7.3 percent for England and Wales as a whole.' Low birthweight arises as a consequence of premature onset of labor or retarded fetal growth, and is more commonly found among socially disadvantaged women. Psychosocial Advances in tbe Prewntion of Low Birtbweight stress is one possible mechanism by which appropriate use of health and social services, social diyadvantage may give rise to poor preg- and community facilities; and generally acting nancy outcome.Factors often cited as con- as confidante.1' tributing to psychosocial stress are low income, The effect of intervention by the family inadequate access to and consumption of ser- workers was evaluated in the form of a ran- 1,ices, inadequate access to information, lack of domized controlled trial. The trial was designed physical effort, isolation, improper diet, poor t*i evaluate the potential impact of such a ser- living conditions, ambivalence about the preg- vice; therefore, randomization took place nancy, and lack of social support." before knowing which women would eventual- While itis not possible for health services ly accept a family worker. The "treatment" did to reverse the social disadvantage reflected in not consist of having the assistance of a family perinatal statistics, it may be possible to com- worker, but in receiving the offer of a family pensate for it by the appropriate redirection of worker. provision. This possibility prompted the institution of the South Manchester Family Worker Scheme, which aimed to provide additional social support for women at above-average risk MEITIOD of giving birth to a low :)iithweight baby. It The trial took place in two phases, from was intended that this support would reduce June 1982 to June 1983 and from June 1984 to the level of stress, thereby improving the well- September 1985. The year-long gap between being of the mothers-to-be and, ultimately, the the two phases was attributable to problems health of their babies. with funding for the family workers. An The role of the family worker is modeled overview of recruitment and participation in after that of the travailleuse familiale who had the trial is shown in figure 1. Women eligible been provided as part of the Maternal and for the trial were identified following their reg- Child Health Service in the Département of istering visit to either of the two maternity units Seine-Saint-Denis.'In Manchester, short-term within the South Manchester Health District. funding for family workers was obtained from Eligibility was determined according to criteria the Manpower Services Commission Women indicating increased risk of giving birth to a selected for the posts were hired on the basis low birthweight baby, and was mainly based of personality and general life experience; they on national birthweight statistics' and on previ- had no formal qualifications in health or social ous research.'' Any woman satisfying at least services. The service adopted a client-centered two of the entry criteria illustrated by an aster- approach, with the tasks of the worker varying isk in table 10.1 was included. according to each client's situation. In practice, In the second phase of the trial the entry these ranged from providing help with obtain- criteria were modified slightly to increase the ing state benefits, housing, shopping and other rate of recruitment. Two new criteria were domestic work, and childcare, to promoting introduced: "Interpregnancy interval 6

110 17 The Family Workers Project Interventions months" and "parity 3." Existing criteria were was to be clinically of interest, a difference in in some cases broadened; for example, what mean birthweight of 77 grams between the was formerly "previous perinatal death" was groups would need to be detected. This figure extended to "previous fetal/infant death > 12 is derived from the difference in mean birth- weeks' gestation and up to 1 year of life", and weight between social classes I and II com- "marital status single" was extended to include bined and III. IV, and V comllined); however, known cohabitees. Also, owing to the problems the chance of detecting a decrease in the pro- in using the negistrar General's Classijkation portion of low birthweight babies from 10.2 of Occupations'" to classify women's occupa- percent to 7.3 percent, significant at the 5 per- tions by social class, the decision was made to cent level, would be less than 50 percent. categorize certain occupations for women as Following recruitment. participants were belonging to social classes IV or V. although randomly allocated to either the control or they are not officially classified as such (see experimental group using random number table 10.1). tables. The women in the experimental group As the family worker was intended to pkiy were then sent information describing the a preventive rok, any women registering later scheme, as illustrated in figure 10.2. and a let- than the 20th week of pregnancy were ter indicating a date and approximat,f ime excluded. Intervention began as soon after reg- when the family worker supervisor would call. istering as could practically be arranged, Asian On her visit, the supervisor explained the women were excluded from the trial in view of scheme in more detail, ascertained whether the the lower birthweighi distribution of these eth- woman was ,nterested in having a family nic groups.:' Entry to the trial was also confined worker and, if so. which of the workers would to those living within the district and prede- be most appropriate. In the first phase, the fined adjoining areas. During the two periods scheme consisted of 10 full-time workers, plus of recruitment the notes of all women attending a supervisor and project assistant; in the sec- the registering clinics were screened. In an ond, the equivalent of 20 full-time workers average month of recruitment those eligible for were employed on a full- and pan-time basis, the trial represented approximately one-quarter plus a supervisor, assistant supervisor, and proof all women registering at the 2 clinics, giving ject assistant (although during the last 3 a final total of 1,288 women. Allowing for months the staffing was reduced to one-third). women lost to follow-up and pregnancy loss, it On average, a full-time worker would have six was estimated that each group would have clients, each of whom would receive one to approximately 600 cases; this would give a 76 two visits per week (although. as the service percent chance of detecting a difference in as flexible to the needs of each individual. mean birthweight of 77 g between the two some received more extensive help than this groups. significant at the 5 percent level (two- and others received less). Information on birth- tailed test) assuming a standard deviation of weight, together with other outcome data on 500 g. (It was estimated that, if the intervention perinatal health,as collected from the

711 Brenda Spencer 1 0S Advances in the Preamtion of Low Birthweight women's hospital records. Results have been (53.1%) of male babies than did the control analysed (1 and X' tests) using the SPSS com- group (47.7%). Therefore, in view of the fact that puter package." male babies are on average heavier than female babies, male and female birthweights are presented separately in the tables and figures. Table 10.5 presents outcome in terms of the proportion RESUCIN of babies in each group who were of low birth- The population covered by the study is weight, the proportion who were assessed as shown in table 10.1. As illustrated, the process small for gestational age, and the proportion of randomization was effective in ensuring com- wlm were born before term. Figures are also parability between the two groups. A minimum included on very preterm and very low birth- of two criteria were sufficient to qualify for entry weight births. Being small for gestational age to the study, but approximately two-thirds of was defined as those babies whose birthweight those recruited satisfied three or more criteria. fell below the 10th centile on the birthweight Of the women in the intervention group gestation charts for males and females published who were eligible to have a family worker, -11.4 by Milner and Richards." The odds ratios calcu- percent received help. As shown in table 10.2, lated on the proportions and the associated con- there were a number of reasons why the fidence intervals demonstrate the similarity of remainder of the experimental group could not, results obtained from the two groups. or did not, accept the offer of a family worker, The outcome of all pregnancies in terms of the most common being because enough sup- survival is shown in table 10.6. The proportions port was already available. of pregnancies resulting in a livebirth in the Of the 1,288 women recruited to the trial control and experimental groups were 96.5 per- (see figure 10.1), outcome data were unavail- cent and 95.-i percent, respectively (stillbirth able for 52 women (25 in the control and 27 in was defined as any baby horn dead at or after the experimental group). the most common 28 weeks gestation). Of the livebirths, three reason being relocating out of the area before babies in the control group and one baby in childbirth (see table 10.3). The characteristics of the experimental group died within the first women for whom no outcome was obtained week, the latter death being due to a lethal did not differ from those on wh(ml the final congenital malformation. The largest discrepan- analysis was conducted. In addition, twins Vc 'ere cy in the table is in the number of terminations excluded from the outcome analysis, which carried out for social reasons; however, exami- was therefore conducted on 1,227 women. nation of the relationship between the stage in Data on mean birthweight and gestation for the pregnancy at which these terminations took livebirths and stillbirths are presented in table place and the timing of the visit of the family 10.4. No statistically significant differences were worker supervisor did not suggest any connec- found between the two groups. The experimen- tion between the two. tal group contained a slightly higher proportion Details on gestational age for all pregnan-

712 The Family W`orkm Prgject Inten,entions cies, excluding those terminated by induced As with many clinical trials, the size of the abortion, are presented in table 10.7. There was study population is perhaps not sufficient for any no evidence to suggest anydifferences in distri- definite conclusions to be drawn, the final num- bution of gestational length observed in the two bers and standard deviation (550 g) providing a groups. (In order toobtain the best estimate of 69 percent power of detecting thedifference in gestational age, the woman's dates were used if binhweight felt to be clinically important. There she was certain of the date of her last menstrual was, however, no indicationin the results of an period. For those who were uncertain, gestation- outcome favoring the experimental group,and, al age was based on assessment by ultrasound indeed, the actual mean birthweight of the con- scan if this was performedbefore the 20th week trol group was greater than that of the experi- of pregnancy or, where no such reading was mental group. Moreover, the 95 percent available, was taken from the woman.s uncertain confidence interval for the difference between dates. A comparable proporti9n of women in the mean binhweights does not include adiffer- the control and experimental group were able to ence of 77 g in favorof the experimental group. report their dates with certainty.) In addition to birthweight and gestation, a num- As two maternity units were involved in ber of other outcome measures were examined the study, their entry and outcome data were with no suggestion of any clinically important studied separately to check for any possible differences between the two groups. discrepancies between the two hospitals. It was There are a number of possible ways in found that approximately equal numbersof which these results might be interpreted, each trial participants had been recruitedfrom each interpretation having different implications in center, and the profile ofthe two populations terms of policy and futureresearch. The first was equivalentboth in terms of characteristics conclusion which may be drawn is that the at entry to the trial and meanbirthweight and social support of a family worker given to a length of gestation of the baby. woman in pregnancy does notinfluence the likelihood of her giving birth to a low birthweight baby. The hypothesis that social support couki improve perinatal oukome, as measured DISCISSION by birthweight, was developed on the basisof From the results itis clear that making a substantial amountof previous research. It available the provision of a family worker ser- has variously been suggested that stressand vice to the at-risk group as defined did not sig- lack of social support give rise to adverse out- nificantly influence either the overall mean come and that intervention programs maymiti- birthweight or the proportion of low birth- gate their effect.' Our experiencein weight; nor was there any indication that either conducting the scheme strongly indicated that of the component elements of lowbirthweight, family workers increased the subjectivewell- that is, preterm birth and fetal growth retarda- being of their clients. Nonetheless, it couldbe tion, were affected. that the increase in well-being had no impact

13reilda Spencer 113 1 1 Advances in the Prevention of Low Birthuvight on the development of the fetus, or that it was service among women who were eligible insufficient to effect any change. Perhaps for because they had experience of previous those women at greatest risk the support of a adverse pregnancy outcome, or who qualified family worker was not able to counterbalance under the "social class IV or V/unemployment" the adverse factors leading to low birthweight; criterion. An analysis of variance to examine the or their help was not intensive enough; or, effect of these criteria on eventual birthweight beginning after the first trimester, it came at too outcome showed that the weight of babies born late a stage in the process. to women who entered the trial on the grounds Another line of interpretation concerns the of previous adverse pregnancy outcome was target group on whom the study was conducted. significantly lower than that of others born in It might be argued that the criterh applied result- the trial (p < .001). ed in the selection of a group which was inap- The criteria used resulted in the selection of propriate for the intervention. The group may be a population with a rate of low birthweight (live- said to he inappropriate either because just over births) of 8.6 percent The overall rate for South one-half of those offered a family worker did not Manchester during the years the trial took place or were not able to accept, or because the entry was only marginally lower than this figure.:".'" criteria did not adequately target those most like- (Earlier published figures based on the ly to benefit from additional support. Manchester area as a whole arc higher because \Xbmen eligible for the trial were not pre- the Central and North Districts have higher rates.) screened to select those who would be more Given that trial participants were identified as likely to accept a family worker. Although a cer- being at above-average risk, one might have tain level of nonacceptance was expected, since anticipated a higher rate of low birthweight rela- not all women would need additional support, tive to the local population than was obtained. lt the nonacceptance rate obtained in practice may be that the exclusion of Asian women from (SS.7%) was nonetheless higher than was felt the trial accounted in part for this discreNncy, desirable and would obviously partially account but, even taking this factor into account, it does foi the 'nonsignificant findings. Analyses were appear that the criteria did not prove sufficiently therefore conducted between those women stringent to identify those most at risk. Indeed, who did accept and those who did not accept since during the second phase they applied to combined with the control group. These (lid not approximately one-quarter of all those register- show any statistically significant differenct.s ing, they might he considered extensive. between the two groups, although. owing to Although psychosocial stress may give rise the uneven groupings, the chance of detecting a to the birth of a low birthweight infant, not all difference was unacceptably low at 52 percent, low birthweight is attributable to this cause. and, in addition, the groups were not compara- Similarly, the risk factors adopted as criteria ble because some criteria were associated with reflect both collectively and individually a com- a higher rate of acceptance than were others. bination of social and medical risk, not all of There tended to he a greater acceptance of the which could potentially be reduced by social

114 111 The Fa may Wbrkers Project Interventions support. The choice of low birthweight risk fac- under conditions similar to those which prevail tors as criteria directs the intervention to a group under nonexperimental conditions. For this rea- with a potentially higher rate of low birthweight; son, entry criteria were based on routinely however, this may nonetheless result in an inad- available data from medical records, and no equate targetting of those most likely to benefit additional prescreening took place. This from the intervention, since only a certain pro- approach is similar in concept to that of the portion of that risk could be mitigated through pragmatic trial as described by Schwartz et al.'" social support. It may therefore be argued that Given the findings of the trial, however, itis only certain groups within the overall at-risk apparent that simply providing a service and population stand to benefit from additional targetting it on one-quarter of the population social support. Particular concern is often defined as at above-average risk according to a expressed about the needs of adolescent moth- mixture of broadly based criteria, will not affect ers, for example. A subgroup analysis of women the incidence of low birthweight in that popu- less than 20 years old expecting their first child lation. On the basis of our experience with the showed a smaller proportion of low birthweight South Manchester Family Worker Scheme, we and premature babies in the experimental would thervfore advocate that future research group, but because of small numbers the results be conducted to ascertain whether social sup- were statistically inconclusive. port through family workers may be effective The adoption of lc.)w birthweight and for certain groups within the overall at-risk preterm delivery as general indicators of peri- population. This would entail closer targetting natal health is standard, but these are nonethe- of those most likely to benefit using only crite- less proxy measures and may not be entirely ria indicative of stress, social isolation, or depri- appropriate for a trial of this kind. Very low vation, and including in the trial only those birthweight is more strongly associated with women both prepared and able to accept the adverse outcome, but, since itis a much rarer help of a worker. We would also recommend event, its use as a measure would require larger that future research include outcome variables trials than itis usually feasible to mount. The which assess maternal well-being in addition to overview of research on social support during clinical indicators of the baby's health. pregnancy by Elbourne and Oakley (see pages 203-224) reports that, in common with the Manchester study, the two other trials in which REFF.RENCEs social support has been the primary interven- 1. liouse of Commons Social Services Committee. (1980. tion found a non-statistically significant fall in 19 June).Perinatal and neonatal mortality. London: His Majesty's Stationety Office, the rate of very low birthweight infants. Even Chalmers. 1.(1979). The search for indices. Lancet 2, after combining results with such small num- 1063-1065. bers, however, no conclusions can be drawn. 3. Office of Population Censuses and Surveys.(1981a). The trial was designed to be population ()PCS Monitor. infant and perinatal monality 1980. based and, broadly speaking, to be conducted Department of Health 3. 81/3.

Brenda Spencer 115

1 I_ Advances in the Prevention qf Low Birthweight

4.Office of Population Censuses and Surveys. (1981b). 18. Eedrick, J., and Adelstein. P.(1973).Influence of Birthuvight statistics 1980. Department of Ilealth 3, pregnancy spacing on outcome of pregnancy.British 81/4. Medical Journal 4, 753-756. 5. Oakley, A., MacFarlane. A., and Chalmers, I.(1982). 19. Naeye, R. L.(1979). Weight gain and the outcome of SOCial class, stress and reproduction.In:A. R. Rees, pregnancy..4 ?rierican Journal of Obstetrics and and H. Purcell (Eds.), Disease and theenvironment Gynecology135, 3-9. (pp. 11-50).Sussex. England: John Wiley and Sons. 20. Office of Population Censuses and Surveys. (1980). 6. Newton, R. W., Webster, P. A. C., Binu, P. S.. Maskrey, aamijication qf occupations1980. IAmdon: HMSO. N., and Phillips. A, B.(1979).Psychosocial stress in pregnancy and its relation to the onset of premature 21. Kurii, K. IL, and Edouard, L.(1984).Ethnic differ- British Medical Journal 2, 411-413. ence in pregnancy outcome.Public Health 98. 205-208. 7. Newton. R. W.. and Hunt, L. P.(1984).Psycluisocial stress in pregnancy and its relation to low binhweight. 22. Nie. N. C. H.. Jenkins, J, C., Steinbrenner, K.. British Medicallournal 288,1191-1194. and Bent. D 11.( I 975).Statistical package fin- the social sciences. New York: McGraw-Hill. 8. Hart. J, T.(1971).Inverse care law.Lancet 1, 405-412, 23. Milner, R. D. G., and Richards, B.(1974).An analysis of birthweight by gestational age of infants born in 9. Papiernik, E.(1984).Prediction of the preterm baby. England and Wales, 1967-1977. journal if Obstetrics anics in obstetrics and Gynaecology 11, 315-336. and 6:v1:act:Wow of Ow British Commonwealth 81, 10. Spira. N., Audras. E, Chapel, A.. Debuisson. 956-967. Jacquelin. J., Kirchhoffer. C.. Lebrun, C., and Prudent, 2.3. Sokol, R. .1 Woolf. R. B.. Rosen. M. G., and C.(1981).Surveillance A domicile des grossesses Weingarden. K.(1980).Risk, antepanum care, and pathologiques par les sages-femmes:Essai comparatif contrOle sur 996 femmes. Journal de Gynecologic, outcome impact of a maternity and infant care project. obstetrics and Gynecology 56. 150-156, Obstetrique et Biologie de la Reproduction I(). 543-548. Larson. C. P.(1980),Efficacy of prenatal and postpar- 11. Wolkind. S.. and Zajicek.E.(Eds). (1981). tum home visits (in child health and development. Pregnancy: A psychological and social study. New York: Academic Press. Pediatrks 66, 191-197. 12. Brown, G. \V.. and Harris. T.(1978). Social origins (il 2. Okis. D.. Tatdbaum, R.. Chamberlin, R.. and Roberts, depression: A study o .1' psychiatric disorder in women J.(1980, 22 Oct()ber).'The effrcts ty. a nurse home-vic- (pp. ro--293Lond()n: Tavistock. nation on impnwing the outcome of pregnancy:.4 preliminary rem.41.Paper presented at the American 13. Berkowitz. G. S., and Kasl. S. V.(1983). The role of Public licalth Association Annual Meeting, Detroit. psychomicial factors in spontaneous preterm deliver-v. Journal qf Psychosomatic Research 2'. 283-290. 27, Ri)s,, S. E. and Rutter. A. C.(1978. November). Healthiest babi.s possible: Annitreach program. 14. Stein, A.. Camplx.11, E. A., Day. A.. McPherson. K.. and Vancouver: \ancouver Perinatal I lealth Project. Coop:2r, P, J.(1987).Social adversity, low birthweight. and prderm delivery.British Medical Journal 28. Oakley, A.(1985). Sodal support in pregnancy: The 295, 291-293. "soft- way to increase birthweight.Social Science and Medicine 21.1259-1268. 15. Spencer. B.(1982). Family workers in Prance.Social Work Service (30), .4-8. )9,Office of Population Censuses and Surwvs. (1984. 31 July).Infant and perinatal monality. Department of 16. Spencer. B.. Morris, J.. and Thomas. II.(1987).The South Manchester family worker scheme.Health Health 3, 8-tr Promotion 2. 29-38. 30. Office of Population Censuses and Surveys.(1985. 13 17. Bakketeig, L. S. Hoffman. IL J.. and Harley. E. E. August).Infinu and pernuital mortality. Deparunent of' Health 3, 85. 3. (1979).The tendency to repeat gestational age and birthweight in successive births..1 ,nericanl01(rn1l rff 31. Schwartz. D Elamant, R.. and LdlGuch, J.(1980). Obstenles and Gynecology13i. 1086-1103. Clinical trials. London: Academic Press.

116 11J The Fafrnily WOrken,..- Project Interventions

Statistical Findings: The Family Workers Project: Evaluation of a Randomized Controlled Trial of a Pregnancy Social Support Seivice

Figure 10.1 Overview of Trial

Randomization

Experimental Control 655 633

271 acceptors 384 non-acceptors N. No outcome 8 No Outcome 19 No outcome 25

Twins 1 Twins I Twins 7

262 334

626 women 601 women

1227 women

Brenda Spencer 117 114 Advances in the Pmvention of Low Birth Weight

Table 10.1 Study Population

Experimental Control N =655 N= 6 3 3

Standard Standard Mean deviation Mean deviation

Age (mean, standard deviation)* 23.0 5.2 23.2 5.4 Percentage less than 20 years 45.5 43.6 Weight (kg) 57.2 10.1 (12)* 57.4 11.2 (7)* Height (m) 1.6 0.7 (20)* 1.6 0.7 (13)* Weight I I Height- 22.3 3.6 (30)a 22.6 4.0 (20)*

Number of cases for which information was unavailable

Number Percentage Number Percentage

Underweight* 247 37 222 35 Parity % '0 415 63 398 63 1-2 165 25 163 26 '3+ 75 12 72 11

Previous low birthweight (< 2500 g) 0 587 90 574 91 '1 61 9 52 8

51 8 1 8 1

Previous preterm birth (< 37 weeks) 0 605 92 578 91

1 42 6 47 7

>1 8 1 8 1

Previous spontaneous abortion 0 634 97 610 96 12-28 weeks +1 15 2 22 4

51 6 1 1 1

Previous perinatal death 0 641 98 614 97 28 weeks-lst week of life '1 14 2 19 3 51

Previous neonatal and post-natal 0 640 98 620 98 deaths 2nd week-51st week life 1 15 2 12 2

'>1 1 1

Previous terminations for 0 646 99 619 98 medical reasons 1 9 1 14 2 >1

118 115 The Family Workers Project Interventions

Table 10.1 (Continued) Study Population

Experimental Control Number Percentage Number Percentage

Previous terminations for 0 590 90 568 90 10 59 9 social reasons 1 60 >1 5 1 6 1

Inter-pregnancy interval < 6 months' 27 5 24 4 Single/widowed/divorced/separated' 398 61 376 59 Woman's social class IV or V or unemployed 511 78 477 75 Partner's social class IV or V or unemployed 216 63 219 62 (Percentage of those with partner only)

Diabetic 2 1 1 1 Smoking (number of cigarettes per day) 0 326 50 313 49 1-5 69 11 59 9 5-20 247 38 243 39 >20 11 2 14 2

t entrY criteria Women's social dass IV or V includes singleiwidoweddivorced women, nursinieaukil rit, shop'Sales asstants and harrdressersind all unemployed and unclassified women. Women unemplowd includes housewivesand students. Partner unemployed includes men in prigrn, self-employed, studentsand disabled/skk

Table 10.2 Reasons for Non-Acceptance Number Percentage Not in when visited 91 13.9 No longer/never was pregnant 15 2.3 Moving out of study area 39 6.0 Employed ful -t i me 41 6.3 Not interested 55 8.4 Well supported 143 21.8 Total non-acceptors 384 58.7 Acceptors 271 41.3 Total 655 100.0

I 19 Brenda Spencer 110 Advances in the Provntion uf Low Birth Weight

Table 10.3 Outcome Information Not Obtained

Experimental Control

Reason Outcome Information Not Obtained Number Percentage Number Percentage

Moved 11 1.7 13 2.0

Transferred to hospital outside study area 5 0.8 3 0.5

No trace 3 0.5 3 0.5

Medical record unavailable 1 0.1 0 0.0

Not pregnant 7 1.1 3 0.5

Home delivery 0 0.0 3 0.5

Total 27 4.2 25 4.0

Table 10.4 Mean Birthweight (Singleton Live and Stillbirths) And Length of Gestation

Experimental Control 95% Confidence Interval on Standard Standard Mean difference Number Meandeviation Number Mean deviation difference p between means

Birthweight: All 602 * 3179.6 549.9 581 3214.5 553.5 -34.9 0.3 -97.8 to 28.0 Female 282 3113.3 511,9 304 3146.3522.4 -33.0 0.4 -90.5 to 24.5 Male 320 .3238.0 575.8 277 3289.3 577.5 -51.3 0.3 -143.8 to 41.2

Gestational age (days) 603 279.0 18.7 581 278.4 19.1 0.6 0.6 -1.6 to 2.8

Birthweight missing Itif fflJ' stillbirth

1 1

120 'Me Family Vtr`orkenc Project InkTrentions

Table 10.5 Proportion of Low Birthweight, Small for Gestational Age, and Preterm Babies

Experimental Control Odds 95% Confidence Number Percentage Number Percentage ratio Interval on odds ratio Number of low birthweight babies (<2500 g) live births and stillbirths 54* 8.8 50 8.6 1.0 0.7 to 1.6 liyebirths only 52 8.7 49 8.4 1.0 0.7 to1.6 Number of very low birthweight babies (<1500 g) 5 0.8 6 1.0 0.8 0.3 to 2.5 Number of small for gestational age** 61 10.0 59 10.2 1.0 0.7 to 1.5 Number of preterm babies (<37 weeks)** 60 10.0 54 9.3 1.1 0.7 to 1.6 Number of very preterm babies (<33 weeks)** 9 1.5 11 1.9 0.8 0.3 to 1.9

Birthweight missing tor one stillhirth lite and stillbirths

Table 10.6 Table 10.7 Outcome of Pregnancy Categorization of Gestational Age, Excluding Terminations tor Experimental Control Both Social and Medical Reasons Outcome of pregnancy N=626 = 601

Experimental Control N = 626 = 601 Termination for Number of Weeks' medical reasons 0.5 1 0.2 Gestation 0/0 Termination for social reasons 11 1.8 1 0.2 0-24 (<174 days) 9 1.5 20* 3.3 Miscarriage: <12 weeks 3 0.5 6 1.0 25-27 (175-195 days) 2 0.3 1 0.2 >12 weeks 6 1.0 11 1.8 28-31 (196-223 days) 5 0.8 7 1.2 gestation not known 0 0.0 1 0.2 32-36 (224-258 days) 53 8.7 44 7.4 Stillbirth: antepartum 5 0.8 0 0.0 intrapadum 0 0.0 0 0.0 37-41 (259-293 days) 456 74.5 443 73.9

gestation not known 1 0.2 1 0.2 42+ (>294 days) 87 14.2 84 14.0

Tota 612 100.0 599 100.0 Live birth 597 95.4 580 96.5 Early neonatal death (1-7 days) 1* 0.2 3** 0.5 Late neonatal death (8-28 days) 0 0.0 0 0.0 Survivors at I month 596 99.8 577 99.5

to lethal c ononital malformation C;eslational age ts as cilia% ailable, tor one mi.( Idge. for this case None due to lethal ccmgenital maltorm gestation was assumed to be <1.-4 (1.14s.

Broicia Spencer 121 1.1S The Social Support and Pregnancy Outcome Study

ANN OAKLEY LINDA RAJAN, M.D.

INTRODUCTION The Social Support and Pregnancy Outcome Study (SSPO) is a randomized controlled trial of a social intervention involving home visits to women at high risk of giving birth to a low birthweight (LBW) baby. It was organized and coordinated from the Thomas Coram Research Unit (TCRU), part of the University of London Institute of Education, and funded by the British Department of Health and Social Security (DHSS) for a period of three years, from September 1985 to August 1988. The trial itself ran from January 1986 to November 1987. This chapter describes the background and methodology of the study and presents some of the initial findings. Atthe time of writ- ing, data analysis is still in progress.

1 n) Advances in the Prevention of Lou, Birthweight

BAC:1(3RM IND AND AIMS oF Tin: STUDY previous interventions in this field have provid- The SSPO study is an attempt to assess the ed social support in addition to other services effect of social support in pregnancy on a such as health education, psychosocialcoun- range of pregnancy outcomes. An important seling, or intensified clinical care, an important reason for undertaking the study was to investi- aim of the SSPO study was to attempt to sepa- gate the potential effectiveness of social sup- rate the provision of social support from other port in preventing adverse clinical outcomes, types of intervention. particularly low birthweight. The study was The broad aims of the SSPO study were: also designed, however, to evaluate the possi- 1. To conduct a randomized controlled trial ble effects of social support on outcomes (RCT) of a program of antenatal support which are conventionally described as "softer" and interviewing by midwives in a sample because they arc concerned with the psychoso- of women at high risk of delivering a LBW cial health of the mother.' infant; The general background to the study is the concept that health is a social, as well as a l'o collect information on the social cir- medical, product.2.` Health is promoted not cumstances, health, and self-perceived only by medical services but by a healthy envi- pregnancy needs of such women; and ronment which enables individuals to make 3. lb evaluate the relevance of data obtained full use of their own personal and social from (I) and (2) to the type of antenatal resources in maintaining health. These care currently provided for high-risk resources include the social relationships and women and to the future prevention of networks in which individuals are involved. LBW and other adverse pregnancy Out- The health-promoting effect of social support comes. has received increasing emphasis in epidemiological and medical sociological research over The size of the SSPO sample was initially the last 20 years. There is now a good deal of calculated in relation to the birthweight out- evidence that social support promotes health come on the basis of a comparison of mean in general.' There are also both observational birthweight between intervention and control and experimental studies demonstrating the groups rather than of the proportion of women relevanci of social support to healthy child- in the two group.i giving birth to a LBW baby. bearing and childrearing. Having reviewed The latter calculation would have demanded a these studies, it seemed important to try to considerably larger sample size. tIsing mean advance the debate concerning the effective- birthweight as an outcome required a sample ness of social support in pregnancy by under- size of 420 to show a significant increase of 150 taking an intervention study using random g in mean birthweight in the intervention group allocation to control both those factors known (power of 80%, p = 0.05). The actual achieved to influence pregnancy outcome and those fac- sample size was 509, tors whose effect is yet unknown. Since most In addition to testing the hypothesis that

1,) 124 The Social Support and Pregnancy Outcome Study Interventions a social support intervention inhigh-risk preg- appears to have beensuccessful, judging from nancy might be capable of increasing mean the obstetric and social characteristics of the hirthweight, the SSPO study was designed to final study sample, and was considerably easier test the idea that social support mightaffect to carry out than more complexrisk scoring the birthweight distribution, some birthweight systems. Women entering care very late intheir groups more than others, andparticularly pregnancy were excluded fromthe study in those women who were socially unsupported order to allow time for the intervention to be or whose obstetric historiesindicated social given. Restrictkm of the study to English speak- rather than biological reasons for previous ers was unfortunately necessary', asfunding LI3W delivery. Other initial hypotheses were was insufficient to cover the costof interpreting that the provision of additional social support and translating for non-English-speaking ethnic in pregnancy might affect maternal hospital minority mthers; however, four perceat of the admissions in pregnancy, the incidence of final sample identified their ethnicity as Afro- hypertension and other physical morbidities, Caribbean, Asian or -mixed," and five percent the length of labor, the use of analgesia, the of the partners were described as belonging to incidence of instrumental delivery and other this category. such procedures, and, in the postpartum peri- The social support intervention in the od, the incidence of maternal depression. SSPO study was provided by midwives who infant and maternal morbidity and health ser- worked as research midwives on a part-time vice use, and the women's confidence as basis for the duration of the trial, one in each mothers. of the four centers. The chief reason for using midwives to provide the intervention was that the Department of Health, which funded the study, was concerned that the interveners SIt 'I )1* DING \ employed should, for policy reasons. he repre- The study was catjed out in four centers, sentative of a group that already had an estab- two in the Midlands and two inthe south of lished role in maternity services. In the event England. Women registering at these four cen- the study demonstrated benefits of support in ters were eligible to enter the trialprovided pregnancy, the policy implicationswould then they met the following eligibility criteria: A his- be relatively easier to translate into practice tory of at least one previous lowbirthweight than if staff not linked with the maternity ser- delivery unassociated with major congenital vices had acted as sources of support. There is malformations; booking before weeks` ges- also a sound historical and professional ratio- tation with a singleton pregnancy; and reason- nale for seeing midwives as potential social able fluency in English. The first criterion was supporters, as individual social carefOr w(nnen chosen in order to select an at-risk group so has always constituted a key element in the that the chance of showing some effect of the professional ideology and practice of mid- intervention woukl be maximized. This strategy wifery.'

/25 Ann Oakley and Lynda Rajan 121 Advances in the Prevention of Lou, Birtbuvight

Figure 11.1 shows the consent, riindom- and low response rates. The format of the ization, and data collection procedures used in questionnaire used in the SSPO study enwur- the SSPO study. From January 1986 to May aged women to write open-ended, unstruc- 1987, the four research midwives selected tured answers if they wished to, andmany from the case notes of all thewomen register- did; assiduous follow-up of initialnonrespon- ing at the four centers those who fit the study ders produced an overall response rate of 95 eligibility criteria. The aims and design of the percent.' study were then described to eachwoman. The midwife explained that, ifa woman agreed to take part, she would not necessarily receive the social support intervention, but NATURE OF THE SOCIAL SUPPORT INTERVENTION would have a 50/50 chance of being in the The WO study was set up to evaluate control group. The importance of random allo- the potential of nonspecific social support to cation was emphasized, as was the fact that improve pregnancy outcome. As emphasized every woman would receive standard antena- above, it was not a study of a health educa- tal care, irrespective of study participationor tion, intensified clinical care or other directly allocation. The midwives also explained that service-related intervention, hut rather an every woman wno entered the study would be attempt to provide and assess the value of asked to complete a postal questionnaire six support to women in their own homes during weeks after delivery. pregnancy in whatever way seemed appropri- When a woman agreed to take part, the ate to them. The SSPO intervention consisted midwife telephoned TCRU foran allocation of a minimum package of three home visits, and then informed the woman to whichgroup ideally at 14, 20, and 28 weeks' gestation, plus she belonged. Women allocated to the inter- two telephone calls (or brief home visits for vention group received the social support women without telephones) from the research package in addition to their normal antenatal midwives. The midwives were asked to try to care; those allocated to the control group had provide this minimum package forevery standard antenatal care only. Obstetric data intervention group woman and to provide were collected after delivery on all of the more than this if asked to do so by the women participating in the study.* A postal woman (providing the midwife's caseload instrume:-.t was used to collect information allowed this). In addition, the research mid- from the mothers postpartum, both to reduce wives were provided with radiopagers, and cost and to avoid the potential problem of the intervention group women were informed midwives who had provided the social sup- that they could call the midwife assigned to port intervention collecting information they them whenever they wanted to, on a 24- might wish to see establishing the benefits of hours-a-day basis. their work. The two main disadvantages ofa During the home contacts, the midwives postal survey are loss of qualitative material Were asked to do a number of things: First of 1 .1

126 The Social Support and Pregnancy Outconw Study Interventions all, to listen to what women had to say; sec- women received less than the minimumsocial ondly, to discuss with the women their preg- support package (normally because they deliv- nancy needs and circumstances;thirdly, to ered early) and 70 percent received more than give information when required; and lastly, to the minimum package. 11,1ore than one-half of carry out appropriate referrals toother agen- the intervention women were referred by the cies, such as social workers or hospital spe- research midwives to health professionals at cialists. To provide a structure for the some stage in their pregnancies,and one- interaction between the research midwives quarter were referred to welfare agencies; one and the women they were supporting, three in five women did not receive referrals of any semistructured interview schedules were avail- kind. Information about the specific health able, and normally used. A portion of each topics of smoking, alcohol, and diet was home contact was tape-recorded to generate a requested and given to 12 percent, 8 percent, basis for comparing the midwives with one and 24 percent, respectively, of the interven- another, and also to provide some qualitative tion group;- one-third of the women were not data for the analysis of the study results. After given any lifestyle information at all. The each contact with a woman in her intervention research midwives were provided with guide- group, the midwife completed ashort data lines as to how to respond to specific topics in sheet on which she assessed the woman's order to standardize the type of information needs, her state of mind, and the type of help given. Throughout the trial regular monthly given, if any. meetings were held in London and attended Table 11.1 gives some guidance as to how by all of the midwives. These meetings served these specifications for the intervention a number of purposes,including facilitating workcd in practice. The home visits were car- the standardization of the social support being ried Out at approximately 18, 2,4, and 30 given and allowing the research midwives to weeks, somewhat later than planned, due discuss the inevitably stressful nature of the largely t-.) the women registering to hospital work involved in giving support to women and thus being recruited into the trial later with large numbers of social and obstetric than amicipated. Twenty-five percent of the problems.

* .e.:scept ji.rr tuy) teho Mot,ed away during the studyand could not be traced. tParticular thanks are due to Sarulra Stonepr this, 5 The interview ,7chedule5 were used partly as a roultof experience (il. the Perth Social Support and Preventhm of Preterm Labour Thal. in which it was iinind that sometimes contvrvationsbetween research midwives and interven- tkni ,e;roup u,omen did not easily get ol:f the ground. due tolack of an ot,ert rationale, Me PrridWities in the SSPO study were asked not to use the interview schechtleswhen they.felt it ilyis awkward or inadvisable to do so. It is important to note. botilTer; that interviewing. as used insocial Science research. IS (Ow everienced as a Suppollive eXereise by interlieu'ees."" Flirtber analysis of data penammg to these important issues willbe carried wit and published elsewhere,

Ann Oakley and Lynda Raftm I '7 I 2 ;*,' Advances in the Prevention of Lou' I3irthweight

Mil: SAMPLE had partners who were unemployed. About one-third of the women were themselves Figure 11.2 shows the SSPO sample. employed during pregnancy and similar pro- together with the pregnancy outcomes. Of the portions left school at or before 16 years of 510 women who agreed to take part in the age. These are indications that the sample was trial. 256 women were randomized to the inter- disadvantaged socially as well as being at high vention group and 254 to the control group. risk obstetrically. There are no indications that One intervention group woman proved to have the differences in pregnancy outcomes been incorrectly entered; two women, one in described below which favor the intervention each arm of the trial, moved and became inac- group might reflect the fact that this group is cessible to the study, Five women were carry- less disadvantaged socially than the control ing twins which were not diagnosed until after group; if anything, the figures suggest that the recruitment to the trial; these women were Opposite may he the case. excluded from the main analyses, as the SSPO study was mounted as a trial of social support in singleton pregnancy. Equal numbers of RE.seas women in the intervention and control groups had terminations and spontaneous abortions. 13irthweight, (kstation, Law anti Delitoy There were three stillbirths in the intervention Table 11.3 gives mean birthweight. The ini- group, three intrapartum stillbirths associated tial aim of increasing birthweight by 150 g with abruptio placentas and one antepartum proved to be overambitious, Taking singleton stillbirth due to placental insufficiency at 36 babies who survived (the most important crite- weeks (weight of 1,760 g). There were 240 live rion from the mother's point of view), the dif- births in the intervention group and 243 in the ference between the intefvention and control control group. After allowing for five neonatal groups is 50 g. For live singleton births itis deaths, 238 intervention group and 240 control about the same, but the difference is less when group babies remained at the end of the study the stillbirths are included. Itis interesting to period. note that a recent overview by David Rush" of Table 11.2 shows the comparability of the dietary interventions in pregnancy ccmcluded two groups at entry to the trial. There Was no that the overall birthweight effect of these inter- difference in mean gestation at booking, the ventions is in the range of 40-50 g. mother's age, smoking status at booking. or From table 11.4it can be seen that there parity. A slightly higher percentage of the inter- were fewer LBW babies in the intervention vention group women were married or cohabit- group. Table 11.5 shows that our intervention ing. Approximately three-quarters of both had no effect on mean gestational age. groups were categorized as working class Tables 11.6-11.9 give the findings for the according to the present or previous occupa- labor and delivery variables we identified at tion of the baby's father. One in five women the outset as possible areas of effectonset

128 Tim Social Support and Pregnancy Outcome Study Intementions of labor, mean length of labor, type ofdeliv- birth, there is also a difference betweenthe ery, andanalgesia/anaesthesia in labor. There two groups. are few differenceshere that are statistically Finally, fewet intervention group women significant in the accepted sense (the tables made use of the health services (apartfrom carry a p valuewhere this is less than 0.1). It their routine postnatal check-up), and there is clear, however, that the directionof almost was a significantdifference in the use of prima- all of these results is in favor of the interven- ry health care. tion group. Baby's Condition Maternal Physical Health and Medical Care Tables 11.13-11.15 relate to the condition Figures for some dimensionsof the of the baby. There was somedifference women's physical health in pregnancyand use observed in terms of Apgar scores. Aboutthe of medical care are given in Tables11.10-11.12. same number ofbabies in each group were There is a suggestion in Table 11.11 dill' symp- resuscitated, but there was a statisticallysignifi- toms of hypertension wereless con,mon in the cantdistinctioninthe methods used. intervention group; these women also experi- Intervention group babies were morelikely to enced fewer hospital admissions forthreatened have their airways cleared and/or simple suc- preterm labor. The lowerincidence of tiredness tion, and control group babies were morelike- and insomnia in the intervention group mayhe ly to need oxygen or more invasive methodsof linked with the lower incidence ofdepression resuscitation (see table 11.13).Approximately in pregnancy (see table 11,16).There was no the same number of babies in the two groups difference in the incidence of medical diagnoses went to the neonatal unit;however, there were of intrauterine growth retardation or usesof cer- some differences in the carereceived within vical suture and betamimetics (see table 11,11). the neonatal unit, with controlbabies using Fewer women in the intervention group more. Method offeeding at discharge from the received more than one ultrasound scan,and hospital also tended to differ (see table 11.14). fewer had antenatal cardicnocography. The most According to their mothers, intervention group striking finding in these tables is theincidence babies were somewhat healthier thancontrol of antenatal hospital admissions,which was sig- group babies. This appears tobe reflected in nificantly lower in the intervention group. the figures for babies' health service useafter Table 11.12 provides data on theperiod birth, which parallel the findings ofthe moth- after birth. There were significantdiffer.nces ers. A significantly greaterproportim of the between the two groups. with 70 percentof intervention grcnip babies were nothealth ser- the intervention women, comparedwith 60 vi,v users. About the same proportions were percent of the control women,reporting their stilln or hadbeen readmitted to the hospital, own health as good or verygood. In response but intervention babies were takenback to the to the question ofwhether or not there are hospital somewhat less than the others (see continuing physical problems as aresult of the table 11.15),

129 Ann Oakley and Lynda Raja?! 123 Advances in the Prevention of Low Birthweight

Psychosocial Outcomes talk about to a doctor. . . . I have been Information on psychosocial outcomes is looking forward to you coming. I've taken mainly from the postnatal questionnaire. been saying that woman is coming on

According to table 11.16, the most statistically Wednesday. . significant difference from this point of view Research midwife, protesting: between intervention and control group women lies in the area of "worries" about the baby after "I'm not THAT WOMAN, Fm Susan." birth Measures of perceived control over one's "I mean I can talk about it to my friends life and depression (taken in the hospital after and that, but it's just that little bit of extra delivery) also favor the intervention group,as professional help.. ." indeed do all of the outcome measures shown in this table. Table 11.17 demonstrates what The help that the research midwives in the might be regarded as a serendipitous effect of SSPO study were able to give has to be seen in the intervention, which was to increase the the context of women's ordinary experiences domestic participation of men in pregnancy and with maternity services. In both Europe and in the early weeks of parenthood. North America, there is convincing evidence that these services are frequently felt by mothers to be unsupportive and demoralizing.'" Poor doctor-patient communication was often CONCLUSION commented on in the SSPO study, as it was by This paper presents some of the initial this woman discussing a previous stillbirth: findings from an RCT of a social support inter- . . There's a lot of questions you want vention in high-risk pregnancy. On the basis of to ask...but you just can't. Everything these data, it can be said that there is no evi- is rushed and you can't think (properly) dence that this intervention did any harm to the . . . and when you do ask them, you mothers and babies who received it, and a think you're being silly because they try good deal of evidence that the effects were to put you off. . ..When we lost the last beneficial. Tables 11.18 and 11.19 givesome baby, I had to go there for the postnatal summary measures of the way in which the and we thought it was for the results of SSPO intervention was seen by the supporters the postmortem . . . and (the doctor) and by those supported. Listening was the sin- said, 'Why have you come? What have gle most valued characteristic attributed to the You come for?' So I said, 'I don't know, I research midwives. As one woman remarked: think I've come for the results of the

. . . This time I've been lucky. because postmortem,' . . . and he said, 'Don't I've been able to talk, and I mean I've you know why your baby died then?'

got a lot of fears and anxieties in rny And I said, 'No,' . . . and he said, 'Oh mind that I've been able to tell you well, at 25 weeks, what do you expect?'

about, that I wouldn't have been able to and passed it off. .. .

2 t; 130 The Social Support and Pregnancy Outcome Study Interventions

5. Berkman, L. F.(1984). Assessing the physical health Many of these women had unresolved effects of social networks and support.Review of problems to do with their past experience of Public Health 5, 413. giving birth to a LBW baby; for some, it was 6. Oakley, A.(1985).Socill support in pregnancy: the first time they had been given the oppor- The 'soft' way to increase birthweight?Social tunity to discuss their feelings fully with any- Science and Medicitw 21(11), 1259-1268. one. It would thus seemthat the therapeutic 7.Oakley, A.(1988).Is social support good for the effect of a social support intervention in the health of mothers and babies? Journal qf Infertility pregnancies of such women may have as awl Reproductity Psychology 6, 3-21. much to do with the reconstruction of the 8, Elbourne, D., and Oakley, A. An overview of trials of Social Support during pregnancy:Effects on ges- meaning of past experiences as with the medi- tational age at delivery and hirthweight.In:H. ation of present experiences. It is salutary to 13erendes (Ed.), Advances in the pretwntion qf lOW note (see table 11.19) that the importanceof birthweight. (See Chapter X, p. X) the supporters' help in these circumstances 9. Oakley, A.(1984).Alternative perinatal services: may often be undervaluedby the supporters Report of a pilot survey.In:I. M. L. Phaff (Ed.), themselves, who may feel skeptical about the Perinatal health services. London: Croom Helm. potential of the modest help they are able to 10. Oakley, A.(1981).Interviewing women: A contra- diction in terms?In:H. Roberts (Ed.), Doing femi- give in counteracting the effects of the multi- nist research. London: Routledge and Kegan Paul. ple social deprivation which affects the lives of 11. Rush, D.(1989).Effects of changes in protein and many urban families inthe 1980s. calorie intake during pregnancy on the growth of the human fetus.In:I. Chalmers, M. Encon. and MNC. Kierse (Eds.), afectity care in pregnancy and child- ACKNOWLEDGMENTS birth (Vol. I). Oxford: Oxford University Press. We are grateful to the DHSS for supporting 12. Garcia, J.(1982). \Vomen's views of antenatal care, In:M. Enkin and I. Chalmers (Eds.), Effectiveness the SSPO study. Particular thanks are due to and satisfaction in antenatal care. London: Spastics Sandra Stone for preparing the manuscript, and to International Medical Publications. Penrose Robertson for his computing expertise, 13. Rothman, 13. K.( 1982).In labor. Women and power in the birthplace. New York: W. \V. Notion.

REFERENCES

Oakley, A.(1983).Social consequences of obstetric technology: How to measure 'soft' outcomes.Birth 10, 99.

2, Dubos. R.(1960).The mimge 1-y. health,London: Allen and Unwin,

3.McKeown. T.(1979).7he mle rye medicine, Oxford: Basil Blackwell. Cohen, S., and Syme, S. L.(1985) .Social suppon and health. London: Academic Press.

Ann Oakley and Lynda Rafan 131 127 Inten'entions

Statistical Findings: The Social Support and Pregnancy Outcome Study

Figure 1 1.1 Consent, Randomizaiion, and Data CollectionProcedures

Eligible women identified by research midwives from case notes

Explanation of study and request for consent to participate

Randomization (by phone to TCRU)

Intervention Group Control Group Social Support Intervention Obstetric case note data Obstetric case note data Postnatal questionnaire Postnatal questionnaire (sent out 6 weeks after delivery) (sent out 6 weeks after delivery)

Figure 11.2 Study Sample and Pregnancy Outcome by Assignmentinto Intervention and Control

Eligible and asked 535

Agreed and Randomized 515

r- Intervention Control 258 254

Excluded Excludtd Moved Twins Singletons Singletons Twins Mo(ed 1 251 1 0 1 1 3 251 - i Mkcs Tops Tops Mists Still BirthsLive Births Live Births Still Births 243 0 6 7 -1 t, 3 240 i NNE) SWOON Surviving NM) 2 2 38 240 3

Ann Oakley and Lynda Rafan 133 1.23 Advances in the Prevention of Lou, Birthuright

Table 11.1 Table 11.2 Structure and Content of the SSPO intervention Comparability of Groups at Recruitment to Trial

Mean gestational age (weeks) at: Intervention Control 1st home contact 17.8 Mean gestation (weeks) 15.7 15.6 2nd home contact 24.0 Mother's age (years) 27.9 3rd home contact 29.7 28.1 1st telephone contact 22.3 Percentage of group: 2nd telephone contact 27.3 Under 20 years of age 4 4

Married/cohabiting 84 81 Women who received: Working class* 77 percentage number 72 Partner unemployed* 21 minimum social 20 support package 5 12 Employed during pregnancy*33 35 less than minimum Education completed at or 31 31 sodal support package 25 61 before 16 years of age. * more than minimum Smoking at time 41 40 social support package 70 170 of enrollment

Parity 1.8 1.8 No referral of any kind 22 54 Referred to: Mean number of previous LBW deliveries:

health professional(s) 52 125 1 85% 86% 11o/o 12O/0 welfare agencies 27 64 3 + 40/0 2%

No lifestyle information given 34 81 Total Numbers 255 254

Given information about: ' Information takenowl postnatal questionna fres. smoking 12 28

alcohol 8 18

diet 24 58 Table 1 1.3 Mean Birthweight

Intervention Control

Surviving singletons 2973.9 g 2923.1 g (N=238) (N=240)

Live singleton births 2956.3 g 2906.5 g (N=240) (N=243)

Singleton live and 2944.4 g 2906.5 g stillbirths (N=243) (N=243)

134 Statistical Findirigs: The Social Support and Pregnancy Outcome Study Interventions

Table 11.4 Table 11.7 Low Birthweight Mean Length of Labor

Intervention Control Intervention Control percentage number percentage number First stage (hours) 5.3 5.4 Second stage (minutes) 20.4 21.1 10.0 7.2 Live Singleton Births Third stage (minutes) < 2500 g 17 42 21 52 Total* (hours) 5.7 5.9 >2500 g 83 198 79 191 live and still- 100 240 100 243 "Based on Nz--202 (Intervention) and N.193 (Control), Total births, non-cesarean-delivered mothers, for whom information was available. Singleton live and stillbirths < 2500 g 19 45 21 52 > 2500 g 81 198 79 191 Table 11.8 Type of Delivery Total 100 243 100 243 Intervention Control percentage number percentage number Table 11.5 Gestation at Delivery Spontaneous 81 197 75 182 Forceps/vacuum 2 5 5 11 Intervention Control extraction

Mean gestational age Cesarean 17 41 21 50 at delivery (weeks) 38.2 38.2 section Percentage of deliveries: Total* 100 243 100 243 < 28 weeks 3.6 3.2 < 37 weeks 19.3 20.5 'All singleton hve births and stillbirths for wh. 'hinfomtation wasavailable. Total numbers (all pregnancies except for terminations) 249 249 Table 11.9 Analgesia/Anesthesia in Labor Table 11.6 Intervention Control Onset of Labor percentage number percentagenumber

Intervention Control None/Entonox 45 108 40 95 percentage number percentagenumber Pethidine 26 162 26 61

Spontaneous 74 180 68 163 Epidural 11 27 16 39 53 26 64 Elective 22 GA 17 41 )8 43 Emergency 1 3 cesarean 4 10 6 14 Other

100 241 100 238 Total* 100 243 100 241 Total*

available. *All singhlon live births and stillbirths for whicintormation ts'a5 available "All singleton e births and stillbirths for which information was

Ann Oakley and Aynda Rafan 13J 135 Advances in the Prevention of Low Bird:weight

Table 11.10 Table 1 1 .1 2 Maternal Physical Health in Pregnancy Maternal Physical Health After Birth

Intervention Control Intervention Control percentage number percentage number percentage number percentage number

Proteinuria Health very 70 161 60 133 on own 24 51 27 64 good/good p< ,03 with bp 8 8 11 11 Has no physical 59 69 46 52 Suspected 1UGR 21 51 22 54 problems now (N = 118)* (N =113)*

Threatened 14 9 20 No health service 40 93 31 70 miscarriage* use (except for p <.05 Admitted foi 14 34 18 43 routine postnatal) threatened pre- Still in/readmitt- 4 8 4 8 term labour ed within 6 weeks Tiredness* 78 180 83 187 of delivery insomnia* 22 50 27 62 Hospital visit 4 9 8 18 Swollen ankles* 26 59 32 73 (excluding routine postnatal)

Mother's information Visit to/from GP 29 67 39 88 p <.03

Table 1 1 .1 1 Other 6 13 7 16 Medical Care in Pregnancy Based on S4 = 2 i0 (Intervention! and= 22O (Control) surviving singleton babies whose mothers returned postnatal questionnaire and answered relevant questions. Intervention Control Number mho recorded any phyiica/ problem after the hirth. percentage number percentage number

Cervical suture 4 9 4 13 Table 11.13

Betamimetics .3 8 4 10 Baby's Condition After Biah

More than one 73 178 77 187 Intervention Control ultrasound scan percentage number percentage number

Antenatal CTG 40 97 49 118 Apgar: <7 at 1 min 12 29 15 35 p <.06 <7 at 5 min 4 4 8

Mean number 5.1 5.1 Not resuscitated 34 81 13:3 81 of hospital visits Cleared airways/ suction only 64 100 54 83 Admitted to 41 97 52 126 hospital p <.01 0: by bag and mask/ antenatally other resuscitat;on 32 50 38 59

Mean number of days 7.2 8.3 Endotracheal intubation 5 7 9 13 in hospital antenatally* (N = 157) (N = 155) p .04 for resusc itahon method. Base for calculation is number ot women admitted. Based on singleton live births for whit h rntOrmation on Apgar scores/resuscitation was available. Lii 136 Statistical Findings: The Social Support and Pregnang Outcome Study Interventions

Table 11.14 Table 11.16 Baby's Care After Birth Psychosocial Outcomes

Intervention Control Intervention Control percentage number percentage number puLentage number percentage number

To neonatal unit 15 35 15 37 Enjoyed birth 66 151 58 130 Mean number of days 9.6 17.1 p <.08

Ventilated 3 8 5 13 'Excellent' relation- 64 148 60 134 Mean number of days 5.0 6.18 ship with baby now

14 31 18 40 Supplemental 0: 3 8 5 13 Depressed in pregnancy Mean number of days 3.85 10.08 Depressed after 40 92 48 107 birth p <.08 Totally intravenously/ 5 13 5 13 tubefU Control over life 72 166 63 143 Mean number of days 6.62 18.62 p <.08 89 Breastfed at discharges 45 105 39 Worried about 16 36 28 63 baby now p< .001 Rased on N = 240 (Intervention) and N = 24(Control), all singleton live births. Means in each case based on number reiviving the prot (qua,: figures tAclude I baby who died in neonatal Lnit and 8 who are still in. itiSed on N 230 (Intervention) and N = 226 (Control), mothers of surviving singleton babies who returned postnatal questionnaire and Survivors tor whom into, lation 2.30 Or Inter-tent ion answered relevant questions. and N 226 tor Control).

Table 11.17 Table 11.15 Partner's Support and Help Baby's Health After Birth During Pregnancy, Birth, and After Birth

Intervention Control Intervention Control percentage number percentage number percentage nunther percentage number Present during labor 78 177 76 171 No problems after 74 157 66 142 'Took days off' very 30 56 27 47 p <.07 discharge often/often 'to be with you' during pregnancy No health service use 35 81 24 54 In pregnancy helped very often/often with: p <.007 shopping 85 187 77 161 p <.01 Still in/readmit ed 7 16 7 15 other children 191 89 173 to hospital 94 p <.05 Hospital visit 11 24 16 .35 After birth helped very often/often with:

Visit to/from GP 45 104 43 98 shopping 88 189 81 169 p <.02 Other 13 29 13 30 other children 97 197 92 178 p < .05 Based on N = 2 4) dnrenentiow and N = 226 (('(filtrol1 %urvn ing ton babies whose miithers returned postnatal questumnarre an(1 Rjsefi on N = 2 JO );ntervention.) and N 226 ((ontrol) n7others who answered relevant questions. returned the postal questionnaire and answered relevant questions.

137 Ann Oakkr and Iynda Ralan 1of) 42 Advances in the Prevention of Low Birthiveight

Table 11.18 Attitudes Toward the Intervention

Percentage of women who said it was important that:

She listened 80"/o She gave advice I saw her throughout pregnancy 56% She gave information 56% She was a midwife 33%

TotalN 236

Table 11.19 Value of Contact with Midwife in Pregnancy

Very/particularlyQuite Other/no helpful helpful information

Midwife considered herself* 13% 42% 450/

Woman considered midwife' 50% 44% 6%

* Mean of 3 home contact assessments.

From postnatal questionnaire.

1 3

138 Statistical Findings: The Social Support and Pregnanq Outcome Study Pretention of Preterm Deliverit BY Home Visiting System: Results of a French Randomized Controlled Trial

BFAMICE BLONDEL, M.D. josil IA LIAIX) GRAM) 1312f:ART, M.D.

Ti IE HOME VISMNG SysITM In France, the home visiting system is a part of Maternal and Child Health Services, a program created after the Second World War to reduce infant mortality and to promote the health of mothers and children under six years of age. Through clinics and home visits, Maternal and Child Health Services provide health care, screening, and immunition free of cost. The program is staffed by multidisciplinary teams consisting of doctors, midwives, nurses, nursery nurses, and social workers. During the 1960s. home visits were made by nursery nurses to children only. A home

3 -1 Advances in the &mention of Low Birdmen* visiting system for pregnant women was later also have routine visits at prenatal clinics. introduced as part of a national program which Home visits are free for every pregnant began in 1970 and which intended to reduce woman. perinatal mortality and the incidence of handi- The importance of the two basic activities capped infants.' It included allocation of finan- of domiciliary midwivessupport and care cial resources to develop health services; varies according to geographical area. In Paris, educational programs; and regulations to supervision of pregnancy complications is the improve maternity unit equipment, prenatal main activity. In 1982, 80 percent of the preg- care, and the working conditions of employed nant women received home visits because they women. The first domiciliary midwife began had a threatened preterm labor; the other indi- visiting pregnant women who were registered cations were hypertension, intrauterine growth at Antoine Beclere Hospital (Department of Dr. retardation, and multiple gestation. In Paris, Emile Papiernik) in 1975. At present, between each midwife works closely with one maternity 500 and 600 midwives make home visits in unit and visits pregnant women who attend the France. outpatient clinic of this unit. The activities of the domiciliary midwives are not defined precisely. In accordance with the regulation, home visits should be offered to two specific high-risk populations; Women who BACKGROUND AND AIM OF THE Slum' have difficult life circumstances, and women Two randomized controlled trials (RCTs) who have pregnancy complications. For the were conducted to study the effectiveness of first group, the home visiting system is consid- routine home visits during pregnancy. Olds et ered as a means of reaching women who are al. evaluated a program of prenatal and postna- reluctant to use health services, improving tal home visits among socially disadvantaged women's health habits, encouraging their rela- women in New York State. Nurses made an tives and friends to help them with housework, average of nine visits from the beginning of and establishing a link between the families pregnancy until delivery. Their activities includ- and agencies employing home helpers. For the ed parent education, enhancement of the second group of women, the task of the mid- women's informal support systems, and linkage wives is to monitor pregnancy complications. of the parents with community services. A Routine medical examination includes measure- French RCT was conducted during the same ment of blood pressure, analysis of urine glu- period (1978-1980), but in this study women cose and protein, measurement of fundal were visited at home when they had pregnancy height, monitoring of fetal heart rate and fetal complications. Seventy percent of the women movements, and assessment of cervical state. were at high risk for preterm labor.' Midwives Domiciliary midwives are not allowed to pre- made an average of six visits from the onset of scribe drugs. the complication to delivery. Both of these High-risk women who receive home visits studies failed to demonstrate any decrease in

13 5 142 Prevention of Preterm Deliveries by Home Viisitini.; System Interventions low birthweight and preterm delivery rates women's satisfaction with medical care is among the overall population, and 1 signifi- greater when they have experienced home vis- cantly higher perinatal mortality rate was its as compared with the usual care provided in observed in the intervention group in the the outpatient clinics. French trial. The publication of the French trial raised several questions concerning the effectiveness of home visits and the benefits which could he METHOD expected from this care system. The higher risk The study design consisted of a random- of perinatal mortahty associated with the home ized controlled trial. The protocol and ques- visits could be explained by the inclusion crite- tionnaires were prepared with the domiciliary ria used in the trial:It was conducted only a midwives who were working in Paris. few years after creation of the home visiting The study was restricted to women who system, health benefits may have been overesti- had threatened preterm labor because it was mated, and some high-risk women should not the main indication for prescribing home visits. have been enrolled in the trial. The eligibility criteria were defined according Moreover, against the French background. to the usual practice in Paris: Women were it may be difficult to show any health benefits recruited if they had a moderate threatened of home visits on pregnancy outcome in cases preterm labor, and wonwn who received intra- of threatened preterm labor because there is a venous betamimetics were excluded because very active management of this complication in these drugs are administered for acute threat- France and we can expect a high level of care ened preterm labor in hospital exclusively. among women visited at home and women Women were enrolled between 2( and 36 attending prenatal clinics. weeks of gestation. If the chance to show a health benefit is Two groups of women were compared. small, other effects of home visits should be Women in the intervention group were provid- considered (mainly, the cost of medical care ed one .)r two home visits per week; in addi- and the women's view). Policymakers might he tion. women had access to the domiciliary interested in this medical care arrangement if it midwives by phone calls. Considering that the reduces the overall cost of care during preg- aim of this study was to assess the existing nancy, especially the costs related to hospital- home visiting system, the intervention was con- ization. The choice of a pattern of medical care ducted by the present midwives without any should not be based on financial reasons only, extra intervention. No home visit was provided however; the women's views of prenatal care to women of the control group. Women in both arrangenlents also should be taken into groups received routine prenatal care from account. The aims of our study were to ascer- obstetricians or midwives at the outpatient clin- tain if a home visiting s}stem reduces the cost ics, Lmd they were hospitalized if necessary. of medical care during pregnancy and if the The women were allocated into the inter-

I3eatrice Blondel et al. 143 136 Advances in the Prevention of Low Birtbweight vention group and the control group by ran- views of prenatal care arrangement and domization with sealed envelopes; this random- inquired about the medical knowledge of the ization occurred in two different settings: (1) In women, bed rest, and support during the last the outpatient clinic, when a threatened trimester of the pregnancy. Data on prenatal preterm labor was diagnosed during a prenatal care, delivery. and pregnancy outcome were visit; and (2) intl .npatient ward, when the collected from medical records by one or two risk of preterm delivery of a hospitalized midwives in each hoypital. woman was decreasing and the discharge was Statistical analysis was carried out by the considered. use of X= and t tests, as appropriate. The sample size estimation was based on the number of hospital days during pregnancy, which is the main part of medical cost before delivery. The number of women required for REstrrs the study was estimated to be 90 in each A total of 158 women were randomized group. This size should enable us to detect a into either the intervention or the control statistically significant reduction of number of group. Six women were lost to follow-up after days in hospital equivalent to 50 percent of the enrollment; it was impossible to identify these standard deviatkm; this standard deviation was women and to know whether they differed determined from observational studies conduct- from the other women. The following results ed in France. The probability of observing a are based on 79 women in the intervention true difference between the two groups was 95 group and 73 women in the control group. percent. Compliance with the allocation was The study was carried out in four maternity assessed. Four women in the intervention units of the public and private sector located in group had no home visit. Two of them were Paris: Hospital of the Deacons, Notre Dame of hospitalized several days after allocation, and Bonsecours Hospital, Saint Anthony's Ilospital. they stayed in the hospital a long time. In the and Tenon Hospital. The first women were control group, eight women had home visits; enrolled in November 1985 and the last women they might have a higher risk of preterm deliv- delivered in August 1987. Data collection was ery than the other women. Six women were needed at different points in the study. A sheet hospitalized after randomization, For the analy- was completed at study entry to ascertain the sis, we retained the groups as they were origi- eligibility of each woman and to assess the risk nally allocated. of preterm delivery according to the state of the In the intervention group. the midwives cervix and the frequency of contractions. After made an average of ,6 home visits. The aver- delivery, a questionnaire was given three to age length of the home visiting service was 30 four days after birth, when the woman was still days. In general, the last visit occurred during in the maternity unit. It was a self-administered the .35th or 30th week of gestation. questionnaire. which focused on the women's At study entry, the intervention group and 137

144 Preventiwi (y. Preterin Deliveries by Home lIsititig System Intenvntims the control group were similar on the major group. the preterm delivery rate wasthree sociodemographic characteristics (see table times higher than it was in Paris in 1981.' 12.1). The distribution of age and parity was The trial was designed to assess the cost of the same; the proportiono.f women who were medical care during pregnancy mainly through marrit:d or who were cohabiting with the the cost of hospitalizations. The proportion of child's father and the proportion of women admissiens to hospital was slightly higher in with French citizenship did not differ signifi- the intervention group than in the control cantly. Furthermore, both groups had the same group. and the mean stay inhospital was a lit- social class distribution. tle longer in the intervention group; however, The distribution of risk factors fm preterm none of these differences wassignificant (see delivery was similar in the two groups. The table 12.3). The difference (intervention group proportion of women who had had a previous minus control group) of the mean stay in hos- preterm delivery was 5 percent in the inter- pital was 1.4 days (C1 =0.8. + 3.5); the c-onfi- vernion group and 8 percent in the control dence interval included a range of situations. group: this difference was notsignificant. At from a reduction of about 1 day in hospital study entry, 77 percent of the women were through the home visiting system up to an perceiving contractions in the intervention increase of 3.5 days, Thus, the chance of reduc- group versus 79 percent inthe control group ing the cost of medical care related to hospital- (see table 12.2). A vaginal examination was ization was very small. carried out for every woman before enroll- The home visiting system nevertheless ment. Dilatation of the internal osand short reduced the number of prenatal visits at the cervix were less frequent among the interven- outpatient clinic, and the difference was signif- tion group, and soft cervix and middle posi- icant: 33 percent of the women in the inter- tion of the cervix was more frequent inthis vention group had prenatal visits or mo7-e, group: nevertheless, noneof these characteris- compared to 5.4 percent of the women in the tics of the cervical state differed significantly in control group. All of the women except two both groups. were treated with tocolytic agentsafter study The proportion of women who delivered entry: the treatment wassimilar in both before 37 weeks of gestation was 18 percent in gnmps. the intervention group and 15 percent in the After delivery, we asked mothers whether control group. Two perinatal c Ieat .1S were they had been satisfied with their medical care observed in the intervention group and one in when they had had a threatened preterm the control group; these three deaths occurred labor. Four answers were proposed: Very satis- among prematurebabies whose gestational age fied, satisfied, unsatisfied, and very unsatisfied. at birth was 32 or $3 weeksand whose birth- No woman was unsatisfied or veryunsatisfied weight was between 1700 and 1800 g. These in the intervention group. and 3 women were results show that the study population had a unsatisfied or very unsatisfied in the c(mtrol high risk of preterm delivery: in the control group. The proportionof very satisfied

Beatrice Blondel et al. I5' 1 3 S Advances in the Prevention of Low Birthweight women was m. feh higher in the intervention (42%) (see table 12.6); the difference was not group (78%) than in the control group (44%), significant, but the p value was 0.07. and the difference was significant (see table In general, the women in the intervention 12.4). Another question was related to the pre- group had more help than the women in the natal care arrangement: "According to you, control group: 86 percent of the women in the what is the best prenatal care arrangement intervention group, versus 70 percent in the when a threatened preterm labor is diag- control group, said that the amount of help had nosed?" The home visiting system was pre- been higher during the episode of threatened ferred more frequently than the other preterm labor than during the first trimester of arrangements in both groups, but the propor- pregnancy. In the intervention group, 56 per- tion of women who preferred the home visit- cent of the women who had previous children ing system was 89 percent in the intervention did not participate at all in child care; this pro- group and 60 percent in the control group. portion was 30 percent in the control group. The proportion of women who would have and the difference was significant. The number preferred hospitalization was similar in both of people who took responsibility of home groups; none of the women in the interven- tasks was nevertheless not different in both tion group preferred numerous prenatal visits groups. at the outpatient clinic. but 26 percent of the women in the control group did. One of the tasks of domiciliary midwives is to provide information and support, with the CONCE SIONS objective of reducing tiring living conditions. Women's satisfaction with medical care improving home help, and encouraging women was much more important in the intervention to have rest. group than in the control group. Satisfaction Hedrest was more frequently recommend- may have 1(mg-tertn effects on such outcomes ed among the intervention group than among as postpartum depression or mother-child rela- the control group. Seventy percent of the tionship. Follow-up of the study population women in the intervention group were asked was not planned in our protocol, and we do to stay in bed the whole day, compared with not know the cc msequences of the women.s 58 percent of the women in the control group. satisfaction. but this diffeence was not significant (see table Several studies have reported that pregnant 115). The women in the intervention group women are satisfied with whatever care they were taught to identify contractions more fre- have experienced and prefer it to alternative quently than the women in the control group. possibilities. This statement was noted in rela- This difference was statistically significant. In tion to a new schedule of prenatal visits. fact. the proportion of women who staved in epidural," continuous fetal heart monitoring bed the whole day was higher (58%) in the during labor. and early discharge after intervention group than in Ow control group delivery.' It is not verified f(n. home visits, how-

1 3,

146 Prevention of Preterm Deliveries 4y Home Visiting S:ysteni Interz'entions ever; in our study, at least 60 percentof the own initiative, and only 14 percentof the hos- women in both groups preferred thehome vis- pitalizations were decided by the midwives iting system. In some cases, a new organization during a home visit. In addition, when a of medical care cannot be an improvement for woman was admitted becauseof a threatened the majority of the women; but, in general, preterm labor, the medical staff mayhave been women have difficulties consideringthe advan- inclined to keep her at the hospital for at least tages and disadvantages of a newprocedure or three to five days, with the obiective of pretreatment which they have not experienced. scribing a treatment and mmitoring the compli- Furthermore, differences between alternative cation over several days. In the trial, a short possibilities sometimes mean veiT little. On the hospital stay was not more frequent in the contrary, home visits may represent areal dif- group of women whoreceived home visits ference in care, tiredness, and relationship with than in the other group. In general, a greater midwives, and women who did not experience number of visits may induce a greater number home visits could easily imagine how this sys- of medical interventions. Women who had tem was managed. home visits had about twice as many internal This study does not show that the home examinations as the other women; therefore, visiting system reduced the number of days in the chance to detect a complication was higher hospital. In the best situation, there is a in the intervention group, Furthermore, women decrease of one day in the intervention group, in this group were more aware of the signs of but actually the cost of home visits per woman pregnancy complication manthe other women is almost equal to the cost of one day in hospi- and they might pay more attenticm to those tal. Thus, in this hypothesis, medical cost is signs. equivalent in both care systems. In the worst The results on the number of days spent in situation, the number of hospital days is much the hospital depend on the local medical prac- higher in the group of woinen visited at home tice. In a prevkms trial, Spira et al. did notfind than in the other group. any significant difference inthe mean stay in Against the background of medical practice the hospital, but this study was carried out a in Paris, it seems difficult to reduce the number short time after the creation of this care system. of days in hospital through the home visiting whereas several years are required to have an system. French obstetricianshave a very active adjustment of current medical practice to an approach to threatened preterm labor, which innovation. Therefore, it would be important to includes high rates of hospitalization, and they have results on more recent studies in other may be reluctant to decreasethis standard of areas or in other maternity units. care even if another prenatal care system This study was restricted to threatened proposed. Furthermore, the domiciliary mid- preterm labor. It does not give anyconclusion wives who were involved in the trial could not for other pregnancy complications which are interfere with admissions to hospital. The supervised by domiciliary midwives, such as majority of the women were admitted on their hypertension. It also does not give any conclu-

147 Beatrice Blondel et al. 1 4 Advances in the Prevention (f Low Birthweight sions for another type of care which is provid- Prudent, C.(1981).Surveillance a domicile des grossesses pathologiques par les sages-femmes: ed by domiciliary midwives; Out of large Essai comparatif contrOle sur 9{X) funnies. Journal towns, the midwives provide home visits to de GynC,cologie, Ohstc3trhlue et Biologie de la socially disadvantaged women; their basic Reproduction 10,5.43-518. activity is to encourage women to attend prena- (4. Rumeau-Rmiquette, C., du Mazaubrun. C.. and tal clinic and to have regular rest and appropri- Rabarison. V. (Eds.).(1984). Naltm en !Wince: Dix ans daolution. Paris: INSERM/Doin. ate diet. The benefits of such care have not yet been assessed in France. Research on the S. Porter. M.. and Maclntyre, S.(198.1).What is, must be bust: A research note on conservative or deferen- effects of home visits among underprivileged tial responses to antenatal care provision.Social women should be considered in the future. Science and Medicine 19, 1197-12(X).

Cartwright. A.(19'79).The dignity cf labor? study qf childbearing and induction. London: Tavistock. AcKNomnx;mENTs 7. Garcia. J.. Corrv, M., MacDonald, D., Elhourne, D., This study was supported by grants from and GrantA. (1985).Mother's views of continuous electronic fetal heart monitoring and intermittent aus- the Ministre des Affaires Sociales et de l'Emploi cultation in a randomized controlled trial.Birth 2, (Direction Generale de la Sante). We thank the 79-85. chiefs of the four Departments of Obstetrics 8. Blonde}, B., Garel. M., Breart, G.. and Sureau. C. and Gynecology for permission to carry out (1983).La sortie precoce des femmes apres the trial. Our thanks also go to Marieciel l'accouchement:Enquete d'opinion aupres des patientes et du personnel d'un service.journal de Allard, Dominique Decoudun, Marie-Paule Gynecologie.Obstetrique etBiologie de la Leloup, Dominique Lignac, and Frederique Rep?.oduction 12, 457-460. Teurnier, who are domiciliary midwives in Olds, D. L. Ilenderson, C. R.. Chamberlin, R.. and Paris, for their help in preparing the protocol Tatelhaum, R.(198( ).Preventing child abuse and and the questionnaires. We wish to thank neglect: A randomized trial ot nurse home visitation. Marie-Christine Nourry for preparation of the Pediatrks 78, 65-78. manuscript.

REFER1ACES

1. Rumeau-Rouquette. (.. and Blonde]. B.(198'i'). The French perinatal prognimme. In:J. Vallin and A. 1). Lopez (hIs.). Health policy, socia policy an(I mortality prolkficts (pp. 299-311), Liege: Ordina.

2. Okls, D. L.. Henderson. C. R.. Thtelhaum. R.. and Chamberlin. R.(1)8().Improving the delivery of prenatal care and Outcomes ot pregnancy:A randomized trial of nurse home visitation.Pediatrics 77, 10-28.

3. Spira, N. Audras, F., Chapel. B.. Debuisscm, E. .lacquelin. J., Kirchhoffer. C.. Lebrun. C.. and 141 148 Prevention qf Preterm Deliveries by Home Visiting System Intenvntions

Statistical Findings: Prevention of Preterm Deliveries bY Home Visiting System

Table 1 2.1 Table 12.2 Maternal Demographic and Social Contractions and Cervical State at Characteristics Study Entry

Intervention Control Intervention Control group group group group percentage percentage percentage percentage (N = 79) (N = 73) Perceived Age (years) contractions 77 79 NS < 24 25 18 25-29 34 37 NS Cervical > 30 41 45 Dilation no 5 6 Parity external os 46 40 NS 0 49 49 internal os 49 54

1 36 40 NS 2 or more 15 11 Length of the cervix (cm) Fr. citizenship 20 30 NS < 1 .35 30 2 52 54 NS Married or > 3 13 16 cohabiting with the child's father 91 93 NS Consistency firm 23 26 Social class* medium 41 46 NS 31 30 soft 36 28 II 23 15 III 17 27 NS Position IV 29 28 posterior 7 -P 59 mid 42 38 NS

anterior 1 * According to occupation of the child's father; women who were living alone were excluded. Low station 4 NS I: managers, engineers, professional workers. Expansion of IV: manual workers, unemployed. the lower uterine segment 79 79 NS

Beatrice Blundel et al. 1.4,2 /49 Advances in the Prevention Qf Lou' Birthueight

Table 1 2.3 Table 12.5 Prenatal Care After Study Entry Information about contractions and bedrest

Intervention Control group group Intervention Control group group I) Percentage percentage percentage admitted to hospital 45 36 NS Identification of contractions 77 59 < 0.05 Total number of hospital days* Hospitalized Stay in bed women 9.4 + 8.2 8.3 + 8.8 NS no recom- 9 18 mendation all women 4.3 + 7.2 2.9 + 6.6 NS some hours 21 24 NS Prenatal visits at outpatient clinic the whole day 70 58 (percentage) 0-1 17 10 2-3 40 36 NS Table 12.6 4 or more 33 54 Bedrest and Home Help * mean + s.d. Intervention Control group group percentage percentage Table 12.4 Mother's Views of Prenatal Care Bedrest during the whole day 58 42 0.07 Intervention Control group group More help percentage percentaw than during the first trimester 86 70 < 0.05 Were you satisfied with your prenatal care sinceyou had a threatened preterm labour ? Number of people who participate in house work very satisfied 78 44 0 5.3 60 quite satisfied 22 51 < 0.001 32 32 NS unsatisfied or .3 or more 15 very unsatisfied 0 5 Child care by According to you, what is the best prenatal care arrange- the father or ment when a threatened pretermlabor is diagnosed ? another person exclusively * 56 30 < 0.05 hospitalization 11 15 more visits at the outpatient chnic 0 26 < 0.001 * Multiparae only home visits by a midwife 89 60

I 4

150 Statistical Findings: Prevention qt. Preterrn Deliveries by Home Usiting System Smoking Interventions During Pregneliicy

MARY J. SEVION, M.R H.

INTRopt rnoN In this paper, three issues are addressed: (1) The importance of cigarette smoking during pregnancy as a riskfactor; (2) the causality of cigarette smoking in reduced birthweight;and (3) the effectiveness of smoking cessation assistance in achieving abstinence.The discussions below often overkill ;at least two of these issues.

TilE IMPOMANCE OF SMOKING As A RN: FACFOR The 1985 Institute of Medicine report synthe- sized a large number of studies in the area of low birthweight. To organize the findings and discussions, risk factors were divided into the f011owing categories: Demographic riskfactors, medical risk factors that predate the current pregnancy, medical riskfactors during the current pregnancy, behavkraland environmental risk factors, and health care filo:tors. The data in

14.; Advances in the Prevention of Low Birthweight table 13.1 were abstracted from information sideration is the extent to which the observed presented in that report. For each broad catego- reduction in growth of the fetus is causal and ry of risk, the table shows the specific factor for the extent to which it can be reversed. If smok- which the literature showed the largest relative ing is causally related to birthweight, its impor- risk for babies with intrauterine growth retarda- tance is much more significant. Over the last tion (IUGR). Some estimates of relative risk three decades, hundreds, if not thousands, of were less reliable than others, and this was indi- studies have replicated Simpson's earlier obser- cated by a range of estimates. In general, how- vation that maternal smoking is related to the ever, the largest single risk factor within each of birthweight of the infant.' Despite theover- these categories showed a relative risk of about whelming consistency of the data,a hotly twofold or threefold. Smoking had about the debated issue arose almost immediately and same level of relative risk as most other factors. has continued regarding whether smoking is The data showed that smokers are at three truly causal since smokers were known to times the risk of having an ILIGR baby; the have, or more often just suspected of having. increased risk associated with smoking is, over- characteristics other than smoking that could all, as high as for other risk factors for IUGR. account for the lower birthweight of their Only a few studies have used either IUGR infants.' Arguments against smoking as a causal or preterm delivery as the dependent variable, factor could best be refuted or supported by compared with a much larger number of stud- experimental evidence. ies that have examined average birthweight or low birthweight. The reason for concentrating Results from Three RCTs on Birthweight on the IUGR relative risk, however, is that the There have been only three randomized work of Sexton and 'Jebel, as well as others, clinical trials (RCTs) reported on smoking cessa- supports the hypothesis that maternal smoking tion and birthweight. One has reported no dif- during pregnancy results primarily in a problem ference; one, a statistically significant difference; of intrauterine growth retardation. and the latest, a difference in the expected In addition to the high level of increased direction, but not a statistically significant one. risk, the importance of smoking in relation to Despite the need for such experimental evi- hirthweight from a public health point of view is dence to clarify the issue of causality, it was not further underscored by its prevalence. Over one- until 197S that Donovan reported results from fourth of all pregnancies begin with the woman his randomized clinical trial of smoking inter- smoking; of these, only a fairly modest percent- vention and birthweight conducted in England,'' age of women quit on their own and continue His trial showed no difference in the birth- their abstinence throughout the pregnancy.' weight of the babies born to the two randomly allocated groups of pregnant smokers. The THE ImPORTANC1 OF SMOMM; As A CAI:SAI, FACFOR intervention group had, on average, babies who In assessing the risk associated with stnok- weighed 12 g less than babies born to mothers ing from a clinical point of view, a major con- in the control group. Some of the details of the

4 5

154 Smoking Interventions During Pregnancy Interventions

Donovan trial will be discussed later. (baseline quitters) were included in the trial. The second of the three randomized clini- The eligibles were contacted and a baseline visit cal trials that have examined birthweight is that scheduled. At that visit, written consent to par- of Sexton and Hebei.' Itis the only reported tidpate and a questionnaire were obtained; a randomized trial that has been conducted with sample of saliva was also obtained, from which a U.S. population, although it shouldbe noted salivary thiocyanate was measured (thiocyanate that two additional U.S. trials with plans to is correlated with smoking status)." The smokers examine birth outcomes have been initiated. were randomly allocated to either a treatment The firstisa trial conducted at Kaiser or a control group. Permanente in Oregon, in which the investiga- On average, the women were approxi- tors intended to examine birthweight, but did mately 25 years of age, they had completed a not do so (probably because of the poor results little more than 12 years of schooling, and obtained in quit rates).- The second is an ongo- about 40 percent were black. About one-third ing trial in Vermont under the direction of had had no previous pregnancies, and they Secker-Walker.s Birth outcomes have been col- were, on average, at about 15 weeks gestation. lecwd; but although there are plans to examine The women smoked about a pack of them, there has not yet been a report. cigarettes at the beginning of pregnancy. but The Sexton and Hebel study was designed had reduced this to about half at tlw time of to test the causal hypothesis that a reduction in randomization. The baseline level of salivary smoking during pregnancy would increase the thiocyanate was comparable in the two groups. average hirthweight of the infant. Itis the only Both experimental groups were ccnnparable at one of the three trials that hasfound a signifi- the time of randomization. cant difference in quit rates and hirthweights. The antismoking intervention was given by Thus, the details of this trial are of special inter- project staff, outside the health care setting, est. With the cooperation of more than50 pri- with a variety of contacts, including one per- vate obstetricians in the Baltimore metropolitan sonal visit at the time of enrollment, a monthly area, 935 pregnant smokers wereenrolled. The phone call, and biweekly contact by mail (usu- pregnant women filled out a brief questionnaire ally in the form of a newsletter).'" with information related to smoking and last The smoker was provided health informa- menstmal period, and stated their willingness to tion on the risks to her own health and to that be contacted by project staff. From these of her baby. The main assistance, however, was responses, was determined: Smokers in the form of information on how to quit of 10 or more cigarettes at the beginning of through suggestions and guidance in behav- pregnancy were eligible (regardless otwhether ioral strategies. The smoking cessation coun- they were currently smoking or not) if they had SCION gave no information On caffeine, alcohol, not passed the 18th week of gestation. Ikcause nutrition, or weight gain. When a woman asked of expected high rates of recidivism, smokers about these factors, she was told to talk with who had quit prior to registration for care her physician. Discussion of these factors was

Mary" Sexton 14 155 Advances in tbe Prevention qf Low Birthuvigbt refrained from in an effort to keep the interven- MacArthur, Newton, and Knox," conducted in tion purely an antismoking one. the United Kingdom. Mothers allocated to Table 13.2 shows the results for the smok- receive assistance with their smoking had ing inform,ion obtained at the 8th month of babies weighing 34 g more, hut this difference pregnancy. Twenty percent 9f the women in failed to reach statistical significance. the control group quit smoking and 43 percent of the women in the treatment group (p< 0.01) Comparisons of the Three RCTs reported that they had quit smoking. Consistent At first glance, the results of the three trials with the quit rates are the differences in the appear to be in conflict with each other; how- distribution of smoking between the two ever, insight into the reason for the apparent groups. The biochemical measurement. salivary differences is gained by examining some of the thiocyanate, was also statistically different at a design and implementation features of the stud- significant level. Thus, the several assessments ies. Table 13.4 shows the birthweight results by showed a significant decrease in smoking for the two experimental groups for the three ran- the intervention women. domized clinical trials. As stated earlier, the first The pregnancy outcomes for the two trial, conducted by Donovan, showed no signif- experimental groups were comparable; 96.7 icant difference in birthweights between infants percent of the control pregnancies resulted in a whose mothers were in the treatment group single live birth, as did 96.6 percent of the and those whose mothers were in the control pregnancies in the intervention program. There group. The trial conducted by Sexton and was no indication of differential pregnancy loss Hebel showed a statistically significant differ- for the two groups. The birthweight hypothesis ence of 92 g. MacArthur et al. found a differ- was tested on 867 single live births. As shown enceof 34 g. The table also shows the quit in table 13.3, the babies in the control group ratesachieved in the two more recent trials. weighed 3,186 g; those in the treatment group Donovan has never reported the quit rates for weighed. on average. 3,278 g (I) < 0.05). Based his trial, but concluded that the difference on the evidence of the quit rates and the birth- between the two groups was modest at best weights, the null hypothesis of no difference in and may not have differed at all:' In the latest average birthweight was rejected. The percent- trial, MacArthur et al. reported an increase of 3 age of babies weighing less than 2500 g at birth percent in the quit rate in the treatment group is also shown; 8.9 percent of the control compared to that of the control group (9% v. babies, compared with 6.8 percent of the treat- 6%). In the Maryland study, the difference in ment babies, were low birthweight babies. The quit rates is reported to be 23 percent. If the percentages of babies weighing less than 1500 women who had already quit before enroll- g was 1.1 for control babies and 1.9 for treatment (baseline quitters) are excluded (for clos- ment babies. Neither of these last two compar- er comparability with the MacArthur et al. isons was significantly different. study), the quit rates between the two experi- The most recently reported liCT is that of mental groups in the Maryland cohort still differ

4 7

156 Snwking Interventions During Pregnancy Intenoztions substantially: 32 percent for the treatment the time of booking regardless of the amount. group versus 7 percentfor the control group. The women enrolled in their study and in that of Thus, the failure to find birthweight differences Sexton and Hebei were earlier in their gestation in the two United Kingdom randomized clinical than the enrollees in the Donovan study, giving trials results directly from their not achieving a little more time for intervention.The differ- large enough differences in the relative quit ences in the smoking criteriaused for these three rates. The reason the quit rates were soradical- studies are reflected in the averages of the ly different from the Maryland rates probably amount of smoking for thoseenrolled in the stems from the features of thestudies. study. The women in the two United Kingdom studies smoked about 18 cigarettes per day at Design Features of the Studies the time they became pregnant. The women in Some of the design features of the three the Maryland study smoked over 20 cigarettes studies are shown on table 13.5. per day at the timethey became pregnant, but were smoking only 11 cigarettes perday at the Location time of enrollment. Women in the twoUnited AS indicated, two of the studies were con- Kingdom studies had to he smoking at the time ducted in the United Kingdom, one in,c,.idon of entry into the study, so there was not as great and one in Birmingham; the Sexton and Hebei a reduction in the amountsmoked between study was conducted in Maryland. Cultural dif- onset of pregnancy and entry intothe study. If ferences alone might explain the quit rates, but the baseline quitters are excludedfrom the there seem to he other possibilities as well. Maryland cohort, the women smoked an average of 21.7 cigarettes at the beginning of pregnancy Snwkitzg Stanis and decreased to 13.5 cigarettes by the time of Amount Smoked and Gestation at Enrollment registration for prenatal care (on average at 15 The smokers included in the studies varied. weeks gestation). Thus, the Maryland study sub- Donovan included smokers of any amount atthe jects at baseline had already reducedsmoking to beginning of pregnancy if they were smoking a lower level than seenin the two United five or more cigarettes at the time they were ran- Kingdom studies. domized into the study. The average gestational age of the womenincluded in the trial WAS Assessment of Smoking almost 16 weeks (he had included women up to In the last column of table 13.5,it can be 30 weeks of gestation.) Sexton and Hebei seen that the studiesdiffered in the way in enrolled women who smoked more at the begin- which snictking status at the end of pregnancy ning of pregnancy (10 or more cigarettes),but was determined. The two1.7nited Kingdom because they had no cut-off regarding smoking studies assessed the smoking status onlyafter at the time of enrollment.they included women delivery by self-reported recall information. who had already quit at the time of enrollment. Neither reported any biochemical assessment of MacArthur et al. enrolled all pregnant smokers at smoking. In the Maryland study, both self-

157 Mat31. Sextort 1 4 :3 Advances in the Prevention of Lou, Birthweight

reported data and salivary thiocyanate were from the studies being given in the clinical set- obtained prospectively. The different assess- ting and implemented by the regular prenatal ment of smoking rates introduces some uncer- staff. In the Donovan study, the smokers were tainty about the comparability of the data. randomized on an individual allocation basis. Nevertheless, the differences in the amount In contrast, for practical reasons, the accrual of smoked by the subjects in the three trials and subjects for the MacArthur study was by four- the time the woman could be exposed to the week periods in which all smokers booking at intervention (gestational age at enrollment) do the hospital received the intervention during a not seem great enough to account for the dif- designated four-week interval followed by a fering quit rates achieved in the studies. four-week interval during which smokers received no intervention. The same clink. staff Intertvntion were responsible for patient care during the In addition to the differences in the amount intervention and nonintervention phases. of cigarettes smoked, other features of the three The brief highlighting of some of the studies varied and are the ones most likely to design differences among the three randomized explain the quit rates. These features relate to clinical trials is helpful in identifying several the intensity of the intervention sites and the possible explanations of why there was such a intervention staff. The two United Kingdom marked difference between thc reported quit studies were conducted within a small number rates in the two United Kingdom studies and of clinical settings, and the intervention was car- the quit rate in the Maryland study and the ried out by the clinic staff themselves. In the accompanying lack of difference in hirthweight. Maryland study, the smokers were recruited The United Kingdom studies had a less inten- from a much larger number of practice settings, sive intervention than the Maryland study and and the intervention was conducted outside of had less control over the implementation of the the practice setting by staff recruited and trained intervention. Furthermore, the same clinic staff by the project. In the Birmingham study, the were in contact with both treatment and con- amount of time and attention given to the inter- trol smokers, providing the opportunity for vention varied from one staff member to another contamination (although the modest quit rates and at times was not conducted as planned. for the control groups indicate that this wag not From discussions with both of the United a substantial problem). Since the United Kingdom study groups, the antismoking inter- Kingdom studies did not include an objective vention was, on average, relatively weak com- assessment of smoking, such as cotinine, itis pared to the number of contacts, intensity, and difficult to know what the precise magnitude of time given in the Maryland study. response to the intervention really was, but it was surely vei-y low. Allocation Scheme Although the two United Kingdom studies There was also some possible compromise did not produce evidence to refute the null in the integrity of the two experimental groups hypothesis, the most reasonable explanation for 14;.)

158 Smoking Intertentions During Pregnancy Intenv itions this is because the intervention was not strong trol groups for nulliparous women. These smok- enough to achieve a substantial difference in ers (see table 13.7) had quit ratesof 14 percent quit rates. The birthweight hypothesis can be and 7 percent, respectively, in the treatment tested only if there is a difference in quit rates group and control group. In contrast,the respec- that is large enough to produce a birthweight tive quit rates among multiparous women were 7 difference. Itis much more reasonable, there- percent and 6 percent for the treatment and con- fore, to conclude that the two United Kingdom trol groups. The pattern of birthweiglns was con- studies produced no evidence on whether sistent with the pattern of quit rates. The smoking causes reduction in birthweight than nulliparous women in the treatment group had to conclude that antismoking assistancehas no babies who weighed an average of 96 g more effect on birthweight. The weight of evidence than the control group babies. These findings produced by the large number of observational give further support to the causal role of studies and by the Mai}, land randomized clini- cigarette smoking: An increase in quitting causes cal trial supports a causal effect on the birth- an increase in birthweight.They also suggest, weight of the baby from maternal smoking however, that smoking cessation intervention has during pregnancy. If the mother quits smoking very little impact either on quit rates orbirth- during pregnancy, her baby will, on average. weights when the smoker has already had a have an increase in birthweight. It must be rec- pregnancy. Similar Maryland data areshown at ognized, however, that, to date, there is still not the bottom of table 7 for the same stratified a solid base of experimentalresults on which groups. Among the nulliparoussubjects. about to rest this statement. 10 percent of the control group of women quit smoking and alxmt 36 percent of the treatment group quit smoking. This represents a26 percent difference in the quit rates between the two SUBGROL ANALYSES experimental groups. Unlike the MacAnhur et al. Regardless of whether an overall effect for finding, the quit rates also differed between the the randomized clinical trial is found or not, two groups of women who had had one or more some analysis is usuallydirected toward sug- previous pregnancies. The higher quit rates in gestions of differential effects among subgroups the treatment group were consistent with the of subjects. These subgroup analyses have to finding of heavier weight babies in the treatment be viewed as unplanned anal'ses and, there- group. The multiparous women inthe treatment fore. caution must be appropriately exercised group had babies who averaged3,275 g, an in their interpretation. increase of 112 g compared with the control group babies. Whenthe quit rates in the Parity Vermont study were examined, they did not dif- The Birmingham data were analyzed and fer according to the number of previous preg- reported by parity. A larger difference in quit nancies the smoker had. The self-reported quit rates was found between the treatmentand con- rates were 14.7 percent and 1.4.1 percent, respec-

itkoyl Sexton 159 5 AdtwItCeS in the Prevention (1. Lou' Birthweight tively, for women with no previous pregnancy larger differential in birthweights. For blacks, versus those with1 or more.* While some the treatment women had ba hies w ho groups of nulliparous women, such as those in weighed, on average, 94 g more at birth, For the Birmingham study, may he more responsive whites, the treatment group babies averaged to a smoking cessation intervention, itis not a 136 g higher in weight. Nevertheless, both consistent finding. Thus, the most reasonable blacks and whites derived significant benefit conclusion at this time is that all women, regard- from the antismoking intervention. less of their parity, will benefit from antismoking In summary, the birthweights in all three intervention both by achieving higher quit rates randomized clinical trials and in the subgroup and increased birthweights of the babies. analyses seem consistently to follow clearly the pattern of quit rates, providing additional con- Race fidence in the conclusion that maternal smok- Because of the concern for the increased ing does cause a significant decrease in low birthweight rates among blacks, the birthweight. Maryland data were examined to determine for blacks the extent of response to smoking cessation intervention and the extent of Em.rnvENTss OF impact on birthweights of the babies. Very little information on quit rates, if any, has been SNIOKINO CE,SSAIION ASSISTANCE reported separately for black and white preg- int' or mation from the three trials that have nant females. On the one hand, black and reported birthweights has already been pre- white females, overall, smoke in about the sented on the differential quit rates between same proportionaround 25-30 percent. On two randomly allocated groups (see table the other hand, black females smoke fewer 13.1). Three additional randomized clinical tri- cigarettes per day than their white counter- als have reported on quit rates, hut have not parts.'' Table 13.8 shows that for the 296 reported on birthweights. These have all been blacks in the Maryland cohort who were still conducted within the prenatal care setting but smoking at time of enrollment, 39 percent of with different approaches to the interven- those in the treatment group quit smoking, tion.'"'. An indication the type of prc)gram compared with 6 percent in the control group. is shown. The Baric and Windsor studies had a Not as many whites in the treatment group fairly modest level of interventicm. consisting quit (only 28%). The quit rates for blacks and of self-help materials on smoking and a health whites ithe control group were similar. message given in the Baric study by the physi- Thus. ti:e intervention produced a larger cian or nurse and in the Windsor study by a increase in quitting among blacks compared health educator. The approach in the ongoing with whites. It did not, however, produce a Secker-Walker study is much more intensive than in the other two in terms of the frequency Seclicr-Walker. R. II. personnal communication of contacts, length. and focus of the interven-

1 r 160 Smoking Interventions During Pregnancy Interventiwts

A preliminary report On for 4. Simpson, W. J.0957). tion. The quit rate for the treatment groups cigarette smoking and the incidence of prematurity. each of the three studies is1,4 percent (in the American journal qf Ohctetrks aiul 6:rnecologi,3, Windsor study, a second treatment group had 80S-815. a 6% quit rate). The quit ratesin the control S. Yenishalmy, J.(1971).The relationship of parents' group varied from study tostudy, resulting in a cigarette smoking to outcome of pregnancy: Implication as to the pr(Olem of inferring causation differential between the two experimental from Observed associations.American journal cf groups of around 5-1D percent.It is not likely Ppidemiolojo' 93, .4.43-150, that, with the modest differentials in quit rates 6. Donovan, J. W. (197). Rand(imised controlledtrial and the sample sizes, a significant difference in of anti-smoking advice in pregnancy.Britisb birthweights would be found. These studies do ty. Pretyntire and Scicial .ttedicine 31. 6-12, show, however, that a modest level of inter- Loch. 13.K.. Waage. Cr_ and Bailey. J.(1983). Smoking intervention in pregnancy. In: W. F. Forbes, vention can increase the rate of quitting. R. C. Freeker. and Nosthakken (Eds.). Proceedings Overall, the highest quit rates were produced (!l' the Fifth Worid Conference on Smoking and He(ilth. in the Maryland study, with a 25 percentdiffer- Winnipeg. Oittada (pp. 35(.)--39i), ential in the quit rates found. Even inthat 8.Seeker-Walker. It II.. Flynn, 11,S.. Solomon. L. J. et. study, a very large proportion of smokers did al.(198(- ),Attitudes, beliefs, and other smokers: Factors affecting smoking cessation during pregnan- not quit. cy.Paper presented at the American PublicHealth From the small number of randomized tri- Association Meetings, Las Vegas. Nevada. als that have reported smoking andbirth- 9.Sexton. M.. Nowicki. P.. and Ik-bel. J.R. (1980). weight, there is room for optimism. Assistance Verification of smoking status by thiocyanate in unre- Preventive to the pregnant smokerafter she has registered frigerated. mailed sampk-s of saliva. Atedkine 15, 28--3-4. for prenatal care can be effective in raisingthe the 10. Nowicki. P.. et al.(19851.Effective smoking inter- proportion who quit and in increasing ventitm during pregnancy. Thrill 11(41, 217-224. baby's birthweight. There is still much more to 11. MacArthur, C., Newton. J. H.. and Kn)x.F. C. effective learn, however. about how to provide 1t)S7).Effect of anti-smoking health education on intervention fiw sm()king cessationduring and fetal size:A randomized clinical trial.liritisb following pregnancy. journal qi Obstetrics awl Gynaecology 9-1. I.!. Sexton. M.(1985).Smoking among women: Proceedings.thePennsylvaniaConsensus CAniference cm Tobacco and Ifealth Priorities. (pp. REi7ERENcEs -ti-791. Pennsylvania I)epartment of licalth. (1985).Preventinj; low birth. 1. Institute of Medicine. Ririe. 1,.. MacArthur. C.. and Sherwciod. M.(19-6). teet;i4bt. Washingnm. DC.: National AcadtAny Press, 13 ,A study ot health educati(In aspectsof smoking in Sext(in. M., nd !Abel, J. R.(198.4).A clinical trial pregnancy.International Journal of Health of change in maternal smoking and its effect onbirth Edtkation 19 (2. Suppl weight, journal (!f the American Medical Assocknion iL \\'indsor. H. A.. Cutter. C.. et al(1985). The effec- 251. 911-91c. tiveness of smoking cessationmethods for simikers A randomized 3. Sexton. NI.(1986).Smoking.In:Ct. (hamberlain in public health maternity clinics: and J. Lumley (Eds.). Prepregnano. caw: rnanual trial.American Journal of Public Health -S. pmct(ee. Sussex. England: John Wileyand Sons. 1389--1392.

Maly J. Scytoit r A.. Interivntions

Statistical Findings. Slizoking Interventions During Pregnancy

Table 13.1 Estimates of Rt ;atiye Risk for Principal IUGR Risk Factors

Risk Category IUGR RR Risk Category IUGR RR

Demographic: MedicalCurrent Pregnancy: Black 2.4 Preeclampsia./Toxemi 2.3-15.8

MedicalPredating Pregnancy: Behavioral and Environmental: Previous IUGR Baby 2.9-7.98 Smoking 3.04

Source: Institute of Medicine.g85). Preventing low htrth weight Health Care: Washington. D.C.: National Academy' Press. Absent/Inadequate Prenatal Care 2.0

Table 13.2 Smoking Characteristics of Control and Treatment Groups at EighthMonth of Pregnancy* Control GroupTreatment Group Thiocyanate, Control Group Treatment Group Percentage Reportiog mole/L Smoking, Cigarettes/day' 389 380 Mean + 2,452 + 1,228 2,094 + 1,209 N .395' .395 0 20.0 43.0 ' hides insIss iinitin s ho had nc)t been deh ervd hi the eIghth-nnynth ttintact. 12.7 19,1 t 1),riterern.e Iii dtstributft)ns ot number.: igairttes ;.ser ii.ii. vs,vs statistic alls 1-5 na+cant hv Nolmogorm Srria-noi test p < Ut 6-10 22.0 16.2 rnw.ing on As(i sublects. 11-20 31.4 17.8 ! I ),,qa 1711.0071g on five sublet ts potereni e in mean t,gatettes smoked via, qattstical4 NismNant fi Student st .3.9 > 20 13.9 test (f) < .017 Mean + SD1 12.8 + 11.5 6.4 + 8.7 j Vittercnce in MEW? tht0c ,Inate, isaS statrsttcalls It student t test ip c .01".

Table 13.3 Measurement of Status of Newborn* Control Group Treatment Group Mean + SD Mean + SD Primary factors Birthvveight (grams) 438 3,186 + 566 429 3,278 + 627 2.28. Percentage < 2.500 438 8.9 429 6.8 Percentage < 1,500 438 1.1 429 1.9 includes only single, live births. t p < .05 h! Ntudent'N I teNt.

Maty Sexto?1 163 5 ,3 Advances in tbe Prevention ((Low Birtbuvight

Table 13.4 Results of Three Randomized Clinical Trials

Birthweight (g) Quit Rates Randomized Clinical Trial TX Control TX Control Donovan (1972) 552 3172 3184

Sexton/Flebel (1979) 867 3278 3186 43% 20% Macarthur/Newton/Knox (1981) 982 3164 3130 9% 6% Not Reported

Table 13.5 Smoking and Gestational Characteristics by RCT's of Examining Birthweight

RC T Number at Eligibility on Amount Smoketi at Assessment ot Gestational Ra ndom - Smoking Status: Time of: Late Pregnancy Age at Entry z at ion at Pregnancy at Randomization Pregnancy Randomization Smokmg Status

Donovan 549' 17.7 15.2 ft't ft )5 pet- ti Ve 15.9 (London) self- seport 1972

Sodon, Hebef 9 15 > 10 20.8 11.2 prospec five 15.0 (Maryland) self-rep< )rt 1979 thkx yanate

Macarthur, 1156 > 1 18.2 14.1 retrospective 15.2 Newton, Knox self-reoort (Birmingham, UK) 1981

Table 13.6 Intervention and AHocation Characteristics of RCT's of Examining Birthweight

RCT Study Sites Implementation of Intervention Allocation Scheme

Donovan (London) maternity hospitals Within clinic by clinic staff Individual randomization Sexton, Nobel (Maryland) 52 private obs Outside clinic by project staft Individual randomization and university clinic

Macarthur, Newton, Knox 1 maternity hospital Within clinic by clinic staff Alternating 4-week accrual (Birmingham, UK)

1 5

164 Statistical Findings: Smoking Intert'entions During Pregnancy Interventions

I able 1 3.7 Quit Rates and Birthweights by Parity for UK andMaryland Cohorts Percentage Quit Late Pregnancy, UK* Birthweight, UK*

Parity TX (TX-C) TX (TX-C) 0 13`)/0 7% 6% 3164 3068 96 -8 > 1 7(%) ()DA) IT0 3163 3171

Percentage Quit Late Pregnancy, MD' Birthweight, MD'

Parity TX (TX-C) TX (TX-C)

0 36'N, 10% 26% 3226 3102 124 > 1 .31% 6% 2,5% 3275 3163 112 MacArthur, Newton and knot Sexton and Hebel: This excludes baseline quitters.

Table 13.8 Quit Rates and Birthweights for Blacks and Whites in Maryland Cohort Percentage Quit Late Pregnancy* Birthweight

Race TX (TX-C) TX (TX-C) Blacks (N 2961** 39% 6% 3.3t% 3080 2986 94 Whites (N439)" 28% /3% 20% 3384 3248 136 " Eighth month ** Excludes baseline quitters

Table 13.9 Additional Randomized Clinical Trials of Smoking Cessation

Investigations (1st Author, Location, Program Quit (Percentage) and Total Number) TX Control

Baric (United Kingdom); ,N = 1 0 Health provider message; materials 14% 4%

9% Secker-Walker (Vermont);N= 249 Health education; 4 individual 1,4% counseling sessions; materials

Windsor (Alabama);= 309 Health education (brief) 14% 2% N = 309 Windsor's self-help guide*

A second snuAing group using the Am('n( an 1 ung 4. .,?t1017S -Freedom iron, Sm(71..ing gunle had.1b.?;? qua rdre.

165 Mary Sexton 155 Cervical Cerclage: New Evidence from The Medical Research Council/Royal College of Obstetricians and Gynecologists

ADRIAN GRANT, M.D.

INTRODI TCTION This paper is an update of a review of cervical cerclage prepared for the symposium on preterm birth held at Evian in 1985.1 Since then, the interim results of the first 905 women participatingin the Medical Research Council/Royal College of Obstetricians and Gynecologists (MRC/RCOG) randomized controlled trial have been published.' The purpose of this update is to review the interim results of the MRC/RCOG trial in the context of the tnree smaller trials of cervical cerclage which were discussed in some detail at Evian." Full details of the MRC/RCOG trial are available elsewhere,' so only the most important features will be described here.

.1 5 G Advances in the Prevention qf Low Birthuvight

MEDICAL RESFARCI1 COUNCIL/ROYAL COLLEGE OF specified subgroups would then help to deter- OBSTETRICIANS .AND GYNECOLOGISTS TRIAL mine who (if anyone) could be expected to benefit from cerclage. The impetus behind the MRC/RCOG In practice, 70 percent of the women who Cervical Cerclage Trial was recognition of the entered the trial had had one or more second high mortality and morbidity associated with trimester miscarriages or preterm deliveries, 10 preterm delivery."4' Cervical cerclage was identi- percent had a history of possible past cervical fied as a strategy for prevention which had damage (for example. cone biopsy or cervical been poorly assessed. It was recognized that, in amputation); and, in the remaining 20 percent rare cases, the operation may have serious of the cases, cerclage was considered on the adverse effects, and that there was no sound basis of other indications, such as previous ter- evidence of its efficacy. Uncertainty about the mination of pregnancy and/or previous first place of the operation was confirmed in a sur- trimester miscarriage. The trial groups were vey conducted by the MRCIRCOG Working well balanced with respect to these various Party on Preterm Labor in 1979 (unpublished indications. The design of the trial was prag- observations), which revealed extraordinarily matic. It was recognized that the clinical situa- wide variation in the use of the procedure by tion might change as pregnancy progressed, British consultant obstetricians. This variation and the aim was to compare two policies in the was one of the reasons for choosing relatively form of recommended management at trial flexible entry criteriaobstetricians' uncertain- entry. In fact, 92 percent of those allocated to ty as to. the advisability of cerdage. There was cerclage received a suture (of those who did no possibility of reaching consensus about not, .5(zi) did not wish to have a suture and 3% where the uncertainty lay. Furthermore. it was miscarried before the stitch could be inserted), argued that cases which would meet more rigid and, of those allocated to the control group, 7 entry criteria would be entered into a trial with percent had cerclage (5(!io because the cervix more open criteria by those obstetricians was judged to be opening and 2 Q'i) because the whose uncertainty happened to coincide with patient decided after entry that she would like the rigid criteria. The advantage of this strategy a suture). was that other types of cases for which other Two hundred Obstetricians (from 12 coun- obstetricians were uncertain about the advis- tries) have contributed cases to the trial. ability of cerclage would be entered into the Although randemization witbin specialty was trial. Any beneficial effect would be expected not performed, cerclage and control cases were to be mediated by prokmgation of pregnancy evenly distributed for each participating obste- irrespective of indication. Nevertheless, the use- trician. Most cases (95) were entered before fulness of cerdage is likely to depend on the 20 completed weeks gestation: the latest gesta- indication. It was therefore hoped that a wide tion at entry was 29 weeks. Cases were entered variety of patients would be randomized and and randomized lw telephone. Most Obstetri- that secondary analysis of patients in six pre- cians used the telephone randomization service

Cerclage 168 15 -; hlterl'entions

provided by the Clinical Trial Service Unit in cent fewer Le..ver.es .)etweenI 20 and 32 Oxford, but other rand(mlization centers were weeks' gestation in the MRC/RCOG trial, which, established in Italy, Zimbabwe, and Hungary. if real, would be equivalent to the prevention of 1 preterm delivery for every 20 sutures inserted. The result is of only marginal statistical significance, however, and is compatible REVIEW OF THE OVERALL RESULTS with a wide range of possible effects of the OF THE FMR RANIX)MIZED 1RIALS operation on the occurrence of delivery before Since the MRC:RCOG trial was established, 33 weeks. The difference in the timing of deliv- the three smaller randomized controlled trials ery in the MRC/RCOG trial was alsoreflected in comparing cervical cerclage with conservative improved survival, again contrasting with the management have been reported.' These were smaller trials '.sce table 14.2). discussed at Evian' and have been reviewed elsewhere. Only Rush's study' of singleton pregnancies with an overall preterm delivery rate of 33 percent shows much similarity tothe CEIWICAL C.ERCLAGE FOR SPECIFIC INDICATIONS MRC/RCOG trial. Dor' investigated twin preg- As planned, possible differential effects of nancies only; Lazar' recruited women with sin- cervical cerclage were investigated in sec- gleton pregnancies with an overall preterm ondary analyses of the MRC/RCOG data strati- delivery rate of only six percent. fied by possible indications for cerclage, such as past obstetric history ofsecond trimester Obstetric maturgemerit miscarriage or preterm delivery. or previous The results of the MRC:RCOG trial' are surgery to the cervix, or multiple pregnancy.' consistent with those of the previously reported Three of the strata generated in this way are trials (where data are available) in suggesting broadly similar to WOMen recruited to the other increased obstetric intervention associated with three trials. cerclage as judged hy admission to hospital, the use of oral betamimetics,induction of labor. Cemical cerclage after previous second trimester and cesarean sectim (see table 14.). miscarriage and/Or preterm delivery Six hundred and thirty women (70%) in the Pregnancy outcome MRC RCOG interim analysis had singleton preg- Where the MRC. RCOG trial does differ nancies and a past history of one or more sec- from the other trials is in its suggestion of a ond trimester miscarriages or preterm deliveries beneficial effect of cerclage on length of gesta- (but no history of surgery to the cervix). These tion and vital outcome (see table 14.2). The characteristics are similar to the entry criteria of three smaller trials show a tendency toward the South African trial.' Table 14.3 provides an shorter gestation and higher mortality in the overview of the outcome in these two sub- cerclage group. In contrast, there were 5 per- gn nips with respect to delivery before 33 weeks

Adrian Grarrt 169 1 5 '3 Advances in the Prevention of Low Birthweight and fetal or neonatal death. As discussed earlier, RCOG trial. The outcome with respect to deliv- the South African trial, if anything, suggested a ery before 33 weeks and miscarriage, stillbirth, small adverse effect of cerclage, but with wide an-I neonatal death for these 74 cases is sum- confidence intervals. In contrast, the stratified marized iti table 14.4; the data are too sparse to analysis of the MRC/RCOG data suggests a large allow any conclusions. In the Dor study, 14 of beneficial effect of the operation, with estimates the 50 fetuses in the cerclage group did not of the odds ratios of 0.67 and 0.66. In fact, this survive the early neonatal period, as opposed analysis revealed that most of the differences to 11 of the 50 in the control group. The timing observed in the primary analyses of the of these hsses was generally similar in the two MRC/RCOG trial were in women with a past groups. In the sutured group, three women history of second trimester miscarriage or miscarried in the 14th, 16th, and 17th weeks, preterm delivery. Furthermore, the more early while in the nonsutured group, two women deliveries in the past, the greater the apparent miscarried in the 15th and 16th weeks. This benefit.' The MRC/RCOG trial provides 80 per- demonstrates how difficult it is to make a judg- cent of the data in table 14.3: this explains the ment in individual cases as to whether or not typical odds ratios of 0.79 and 0.76 with confi- the insertion of a cervical suture actually dence intervals between 0.52 and 1.10. To state caused a miscarriage. Three women in each these results another way, the estimate is that group subsequently delivered prior to 33 com- the insertion of about 18 cervical sutures will pleted weeks, The extra miscarriage in the prevent 1 delivery before 33 weeks, but this sutured group is largely responsible for the dif- statement cannot be made with any confidence. ference in mortality: Thirty-nine (78%) survived the neonatal period in the control group, com- Previous surgery to the cenix pared with 36 (72%) in the cerclage group. as an indication fir cervical cerclage Only 96 women in the MRC/RCOG interim Cervical cerclage jhr other reasons in women at analysis had a past history of cone biopsy or cer- moderate or low risk (?,f early delivery vical amputation, and there are no women with One hundred and fifty-five women (17%) similar histories in the other trials. The odds included in the interim analysis of the ratios for the main measures of outcome were all MRC:RCOG trial had neither a previous second near unity. but because confidence intervals are trimester miscarriage, nor preterm delivery, nor very wide, these analyses are of little practical previous surgery to the cervix, nor twin preg- use. Far larger numbers are required. nancy. Many had histories of previous first trimester miscarriages. This stratum seems most Twin pregnancy as similar to the French trial, and the outcome of an indication lbr cervical cordage these two groups is summarized in table 11.5. The 50 cases in the Israeli trial' Were all Again, the confidence intervals of the typical twin pregnancies, and there were 24 twin preg- odds ratios are wide because the numbers of nancies in the interim analyses of the MRC: events are so small.

1 70 cercleNe 1.5:; Interventions

EVIDENCE FROM THE RANDOMIZED TRIALS OF other recognized adverse effects of the opera- POSSIBLE ADVERSE EFFECN OF CEWICAL CERCJAGE tion into account, it seems sensible to limit the use of the operation to cases with a high likeli- A wide variety of possible adverse effects hood of benefit. Current evidence suggests that of cervical cerclage (in addition to increascd increasing numbers of previous second obstetric intervention) were reported in the trimester miscarriages or preterm deliveries interim analysis of the MRCIRCOG trial. It was constitute the firmest basis for making this deci- impossible to ascribe many of them to the sion. There is curremly no sound evidence to operation with any confidence. Cervical trauma support the use of cervical cerclage on the and difficulties in removing the stitches were basis of previous surgery to the cervix or multi- each reported in 6 of the 450 women treated ple pregnancy. with cerclage. Pre labor rupture of the mem- Cervical cerclage remains unsatisfactorily branes was associated with cervical cerclage in evaluated. It is unfortunate that larger numbers the South African trial,but this difference was of randomized studies have not been conducted. not observed in the Israeli trial.' Puerperal Nevertheless, the trials which have been mount- pyrexia was reported more frequently in the ed are beginning to provide good evidence cerclage group of both the MRC.,7 RCOG trial about the balance between the benefits and haz- and the South African trial. This is consistent ards. They should provide a template for future with observational studies and appears to he 3 research into the effectiveness of the procedure. real effect of the operation.

ACKNOWLEIX;MINES I am very grateful to the Department of CONCLISIONS Health and Social Security for salary and other The 1,655 cases entered into the four ran- support. The MRC/RCOG trial was supported domized controlled trials included in this by a Special Project Grant from the Medical review provide a less than adequate basis for Research Council. clinical decisions about the use of cervical cerclage. Three hundred more cases have been REFERENCES

entered into the MRC/RCOG trial; ideally, how- 1. Grant. A. (1980). Cervical cerelage: Evaluation stud- ever, far larger numbers are required, particu- ies. In: E. Papiernik. lirt:art and N. Spira (Eds.), Preventionof preterit; birth: New goals and new larly for the important analyses of subgroups. practices in prenatta care (pp. 53-)9). Paris: Unlike most medical treatments, cervical INSERM. cerclage has the paradoxical potential to both MIZCIi.COG Working Party on Cervical Cerclage. prevent and cause early delivery. The balance (1985). Interim report of' the Medical Research Councirlioyal College of Obstetricians and between these two effects is likely to depend Gynaecologists multicentre randomised trial of cervi- on the inherent risk of earlydelivery in the cal cerclage. British Journal qf Obstetrics and cases treated. Taking this considerationand the G.rnaecolug3 ,9S.

Adrian Grant 171 .16j Advances in the Prevention qf Low Birtburight

3. Dor, J.. Shalev, J., Mashiach, C.. Blankstein, .1., and Serr. D. M. (1982). Elective cervical suture of twin pregnancies diagnosed ultrasonically in the first trimester following induced ovulation. Gynecologic and Obstetric Intlestigation 13, 55-60.

4. Lazar, R. Gueguen, S.. Dreyfus. J.. Renaud. R.. Pontonnier,G.,and Papiernik.E.(1984). Multicentred controlled trial of cervical cerclage in women at moderate risk of preterm delivery. Britisb Journal qf Obstetrics andGynaecology91, 731-735.

5. Rush, R. W Isaacs, S., McPherson. K.. Jones. L., Chalmers, I.. and Grant, A. (1984). A randomised controlled trial of cervical cerclage in women at high risk of pretenn delivery. British journal of Obstetrics and Gynaecology 91. 724-730.

6. Rush, R. W., Keirse, M. J. N. C., liowat, R, Baum, J. D., Anderson, A. B. M., and Turnbull. A. C. (1976). Contribution of preterm delivery to perinatal mortality. British Medical journal 2, 965-968. 7. Sinclair,.C. (1986). Increasing costs and effectiveness of neonatal care of very low birth-weight infants. In: E. Papiernik, C. Breart, and N. Spira (Eds.), Pretvntion qfpreterm birth: New goaLs and new practices in prenatal care (pp. 31-46), Paris:

I NS ERM.

8.Verkxwe-Vanhorick. S. P.. Verwey, R. A.. Brand. R.. Bennebrock-Gravenhorst, J.. and Keirse. M. J. N. C. (1986). Neonatal mortality risk in relation to gestational age and binhweight. Lancet 1, 55-57, 9. Grant. A. (Forthcoming). Cervical cerclage. In: M. Enkin, M. J, N. C. Keirse. and I. Chalmers (Eds.), Efli,ctire care in pregnancy and childbirth. Oxford: Oxford University Press.

172 161 Cen,ical Cerclage Intementiolts

Statistical Findings: Cervical Cerclage: New Evidence From theMedical Research CounciVRoyal College of Obstetricians andGynecologists

Table 14.1 Effects of Cervical Cerclage on Obstetric ManagementEvidencefrom the Randomized Trials

Cerclage No Cerdage Odds Ratio 95% Confidence Interval

A. Admission to hospital MRC/RCOG interim report (1988) 135/454 114/451 1.25 0.93-1.67

Rush et al. (1984) 30/96 19/98 1.87 0.98-3.57 Lazar et al. (1984) 85/268 41/238 2.17 1.45-3.24 typical odds ratio (95% CI) 1.55 1.24-1.93

B. Use of oral tocolytics MRC/RCOG interim report (1988) 113/454 107/451 1.07 0.79-.1.44 Rush et al. (1984) 12/96 8/98 1.59 0.63-4.01 Lazar et al. (1984) 127/268 73/238 2.01 1.41-2.87 typical odds ratio (95% CI) 1.40 1.12-1.75

C. Induction of labor MRC/RCOG interim report (1988) 85/454 69/451 1.27 0.90-1.80

Rush et al. (1984) 9/96 8/98 1.16 0.43-3.14 Lazar et al. (1984) 49/268 39/238 1.14 0.73-1.81 typical odds ratio (95% CI) 1.22 0.93-1.59

D Cesarean section MRC/RCOG interim report (1988) 68/454 56/451 1.24 0,85-1.81

Rush et al. (1984) 19196 18/98 1.10 0.54-2.24 Lazar et at. (1984) 33/268 22/238 1.37 0.78-2.40 Dor et al. (1982) 9/25 7/25 1.43 0.44-4,65 typical odds ratio (95% Cl) 1.26 0.95-1.66

Adrian Grant 173 .16:2 Advances in the Pnwntion of Low Hirthweight

Table 14.2 Effects of Cervical Cerclage on (A) Delivery Before 33 Weeks and (B) Miscarriage, Stillbirth, and Neonatal Death CombinedEvidence from the Four Randomized Trials

Cerdage No Cerclage Odds Ratio 95% Confidence Interval

A. Delivery betbre 33 weeks MRC/RCOG interim report (1988) 59/454 82/451 0.67 0.47-0.97 Rush et al. (1984) 12/96 10/98 1.26 0.52-3.04 Dor et al. (1982) 6/25 S/25 1.26 0.33-4.73 Lazar et al. (1984) 4/268 1/238 2.99 0.51-17.41 typical odds ratio (95% CI) 0.79 0.58-1.09

B. Miscarriage, stillbirth, and neonatal death MRC/RCOG interim report (1988) 37/454 54/451 0.66 0.43-1.01 Rush et al. (1984) 9/96 9/98 1.02 0.39-2.69 Dor et al. (1982) 7/25 6/25 1.23 0.35-4.28 Lazar et al. (1984) 2/268 1/238 1.74 0.18-16.84 typical odds ratio (95% CI) 0.76 0.52-1.10

Table 14.3 Effects of Prophylactic Cervical Cerclage After Previous Second Trimester Miscarriage or Preterm Delivery on (A) Delivery Before 33 Weeks and (B) Miscarriage, Stillbirth, and Neonatal Death

Cerclage No Cerclage Odds Ratio 95% Confidence Interval

A. Delivery betbre 33 weeks MRC/RCOG interim report (1988) 43/325 61/305 0.61 0.40-0.93 Rush et al. (1984) 12/96 10/98 1.26 0.52-3.04 typical odds ratio (95% CI) 0.70 0.48-1.02

B. Miscarriage, stillbirth, and neonatal death MRC/RCOG interim report (1988) 25/325 41/305 0.54 0.33-0.90 Rush et al. (1984) 9/87 9/89 1.03 0.39 -2.71 typical odds ratio (95% CI) 0.62 0.40-0.98

174 16J Statistical Findings: Cervical Cerclage Interventions

Table 14.4 Effects of Prophylactic Cervical Cerclage for Twin Pregnancy on (A) Delivery Before 33Weeks and (B) Miscarriage, Stillbirth, and Neonatal Death

Cerclage No Cerclage Odds Ratio 95% Confidence Interval

A. Delivery betbre 33 weeks MRC/RCOG interim report (1988) 1/10 4/14 0.34 0.05-2.40 Dor et al. (1982) 6/25 5/25 1.26 0.33-4.73 typical odds ratio (95% CO 0.83 0.28-2.49

B. Miscarriage, stillbirth, and neonatal death MRC/RCOG interim report (1988) 1/10 1/14 1.43 0.08-25.35 Dor et al. (1982) 7/25 6/25 1.23 0.35-4.28 typical odds ratio (95% CO 1.26 0.40-3.96

Table 14.5 Effects of Prophylactic Cervical Cerclage for Women at Moderate or Low Risk ofEarly Delivery on (A) Delivery Before 33 Weeks and (B) Miscarriage, Stillbirth, and NeonatalDeath

Cerclage No Cerclage Odds Ratio 95% Confidence Interval

A. Delivery before 33 weeks MRC/RCOG interim report (1968) W73 7/82 0.96 0.31-2.98 Lazar et al. (1984) 4/268 1/238 2.99 0.51-17.41 typical odds ratio (95% CI) 1..34 0.52-.3.47

B. Miscarriage, stillbirth, and neonatal death MRC/RCOG interim report (1988) 3/73 5/82 0.67 0.16- -2.77 Lazar et al. (1984) 2/268 1/238 1.74 0.18-16.84 typical odds ratio (95% Cl) 0.87 0.26-2.92

Adrian rant 175 1 E; Prevelltiota of Intrauterine Grou7th Retardation with Aiitiplatelet TheraAp

SERGE UZAN, M.D. M. BFAtEts, W.). M. tizAN

CTION The treatment of intrauterinegrowth retardation (with or withoutassociated hypertension) and complications ofpregnancy-induced hypertension (e.g., preeclampsia,eclampsia, and abruptio placenta) is most oftenpurely palliative. Delivery of the fetus isusually the only cure, which often results inserious prematurity and neonatal problems. Numerous studies" have attempted to analyze the phenomena responsiblefor these complications. As a result, the followingele- ments have been clarified todate: 1.Preeclampsia is often associated with adis- seminated intravascular coagulopathywith a reduction in the platelet count. a consump- Adtwnces in the Prevention of Low Birthweight

tion of factor VIII. and an augmentation of In our opinion, this ineffectiveness may have fibrin degradation products. The microscopic been due to relative late initiation of therapy. and grossly visible placental thrombotic In one epidemiological study, Crandon4 lesions which are commonly seen are a eon- reported that women who frequently used sequence of these coagulation disturbances. aspirin had a decreased incidence of gestational 2. These placental infarcts appear to be associ- hypertension or, if present, fewer additional ated with the presence of fibrin deposits in complications such as preeclampsia. the intet-villous spaces. In certain cases, suffi- By 1978, we had decided to establish a cient fibrin is accumulated to produce a true study which examined the effectiveness of the retroplacental hematoma. In addition, similar combination of aspirin and dipyrimadole in the fibrin deposits are commonly found in other prevention of pregnancy-induced hypertensive organs (e.g., the liver or the brain). These complications. There were several reasons for findings encourage us to believe that the this choice. First, aspirin effectively impedes coagulation abnormalities observed in platelet aggregation, probably through inhibi- preeclampsia are critical to the development tion of platelet cyclooxygenase. which then of both maternal and fetal complications. results in blockage of the synthesis of thromboxane A2. Second, aspirin has been extensive- In pregnancy-induced hypertension, an ly prescribed by cardiologists to reduce increase in the production of thromboxane A-) thrombotic complications in patients with valvu- (TXA)) is often found. This product (the princi- lar protheses and following coronary artery pal metabolite of arachadonic acid in platelets) surgery. Lastly, aspirin has been used to prevent is a powerful vasoconstrictor and a stimulator recurrences of arterial thrombotic problems. of platelet aggregation. The origin of the The majority of authors' '' have utilized enhanced production of TXA2 appears to be at small dosages of aspirin (less than 150 mg per the level of the placenta and the platelets them- day) in order to preset-ve the beneficial effects selves. The increase in TXA2 without a corre- of prostacydin. the hypothesis being that the sponding increase in the production of" synthesis of prostacyclin in the vascular prostacyclin leads to a relative predominance of endothelium is less sensitive to aspirin inhibi- thromboxane A-) in the uteroplacental circulation than that of thromboxane A) in platelets. tion. Prostacyclin works in opposition to throm- The second medication, dipyrimadole.'" boxane A). both in terms of vascular tone and has been associated with aspirin in a number platelet function. This imbalance 1)etween of studies; however, it has been difficult to dis- thromboxane A ) and prostacyclin)is believed tinguish its specific action from that of aspirin. to be due to pathological interactions at the Through its inhibition of phosphodiesterase. it platelet-platelet and platelet-vascular level. may potentiate the action of aspirin by retard- Previously, research teams have attempted ing the destruction of adenosine monophos- to treat preeclampsia with heparin. Results have phate. thus making platelets more sensitive to been either disappointing or. at best, transitory. prostacyclin. Dipyrimadole may also have an

178 Prevention qf IntrauterktiVmwth Retardation with Antiplatelet 'Therapy Ititenvntions autonomous action; it may stimulate prostacy- tional age at onset, number of patients includ- din synthesis in platelets. ed, and number of pregnancies observed (see These considerations logically support the table 15.2). proposal of aspirin and dipyrimadole as a pre- The expression treatment utilized means ventive therapy modality for pregnancy- uniquely antiplatelet aggregation therapy. Other induced hypertension. more traditional antihypertensive medications We will first examine a review of the thera- are not specified. peutic trials using aspirin and/or dipyrimadole. Concerning criteria for inclusion, the fol- Tables include only statistics from published tri- lowing should be noted: als; inf(wmation from brief or oral communica- 1. Study A (see table 15.1) has a heterogeneous tions and studies in progress will he included recruitment (this is probably one of its princi- in the discussion. pal faults). A certain number of patients are These studies can be analyzed according even selected under the heading,-vascular to four criteria: risk." 1.What were the criteria for inclusion 2. Study B has an original and interesting selec- intrauterine growth retardation exclusively, or tion process in that it proposes a method for complications of gestational hypertension in testing nulliparas. Their criteria for inclusion general? is an increased sensitivity to an intravenous 2. Was the study prospective? angiotensin 11 challenge. A response is considered positive if diastolic blood pressure 3. Were the patients randomized? rises by 20 mm Hg at a dose equal to or less 4. Was there a placebo-cmtrol group? than 10 mg/kg/min. At this time, to our knowledge, five con- 3. The three final studies used fetal weight trolled studies exist: The study by our group related to gestational age at delivery of either begun in 1978;s the study by Wallenburg pub- one or two preceeding pregnancies asthe lished in 1986;m the study by Wallenburg pub- sole criterkm for admission. lished in 1987;- and the two multicentric studies currently in progress in France and in Belgium (group Peigar and group Epreda ). Table 15.1 summarizes different aspects of' these studies. REst 'ut's We will next examine the meth( yds and results The principal criteria evaluated during of the three studies actually completed. these studies are shown in table 15.3. The numbers and the mode of expression (i.e., abso:ute numbers or percentages) are as pub- MATERIAL AND METIIODS lished. When two numbers are listed, the first For each study, we have indicated the cri- corresponds to the control group, and the sec- teria for inclusion, treatment regimens, gesta- ond to the treatment group.

Serge Uzan et al. 167 1 79 Advances in the Pret,ention of Lou Birthuvight

Disu SSION two arguments: (1) That the efficacy of treatment was beyond any doubt (in our Opinion, Before beginning the discussion of results, this position was not sufficiently admissible); it appears more logical to comment on the cri- and (2) that the treatment was innocuous (in teria of inclusion, certain ethical problems, the Our opinion, this argument was also uncertain). optimal dosage, and the possible adverse Because we believed that formal proof did not effects of this therapy. exist for either of these two arguments, we felt that our study was justified. criteria Jiff incluskm The most k)gical selection process (and the Optimal Dosage method being used for the two studies in In our first study, the treatment regimen of progress) is to use prior 11.1GR confirmed at 150 mg aspirin per day with 300 mg dipyri- delivery as the criteri( n for inclusion. Itis the madole per day was used; however,it only One that can be considered objective. The appeared that even smaller doses might be as two studies (I) and E on table 15.1) include effective. Dosages currently being used are two types of patients classified according to between 1 and 2 mg/kg/day for aspirin and whether there have been one or two prior 225 mg/day for dipyrimadole. In Wallenburg's episodes of It GR. This stratification attempts to study (1987), it was shown that dosages of identify future indi,.ations for therapy. Perhaps aspirin as low as 50 mg/dav and of dipyri- the benefits of therapy will be shown to out- madole as low as 22i mg/day significantly weigh the risks when the probability of recur- diminished production of platelet thromboxane rence is high (i.e., when there have been two A2, which was inferred indirectly through mea- preceding pathological pregnancies). surements of malondialdehyde. Concentrations Trial B, using the angiotensin II sensitivity of makmdialdehyde, a stable byproduct of test as its inclusion criterion, resolves only par- platelet thromboxane synthesis, were reduced tially the problem of identification in primiparas. to 5- 10 percent of baseline levels. On the other hand, levels of 6-oxoprostaglandin El, a prosta- Ethical Omsiderations cyclin Metabolite, were not significantly Participation in a study such as this posed decreased. Both of these observaticms lend fur- numerous ethical questions which were dis- ther support to the effectiveness of low dose cussed at length by our ethics committee. The aspirin and dipyrimadole.r. principal point of debate was whether or not it Each c)f the three completed studies was apprc)priate to propose to women with a unequivocally- demonstrated in its treatment history of two prior pathological pregnancies group (whether randomized or in comparimm that they attempt a third pregnancy with a pos- with the controls) the following findings: ( I) A sible placebo. Wallenburg chose not to enter reduction in the rate of intrauterine growth his patients into an internal control group retardation, and (2) a reduction in the rate of because of his implicit belief in the following secondaryilL complications.

180 Prevention of Intrauterine Groulb Retardation u'ith Antiplatelet Therapy Inten'entions

The above findings were also indirectly ing time (as measured by the IVY method, confirmed by other parameters. Placental exam- which is the most reproducible) is necessary to ination revealed a reduction in placental evaluate hemostatic changes. In patients receiv- lesions: and our study (as noted in table 15.4) ing low-dose aspirin, significant prolongation of demonstrated an improvement in plasma vol- their bleeding time may occasionally be ume, plasma uric acid,and platelet count. observed. Moreover. after stopping aspirin ther- In our study, the duration of pregnancy at apy, a delay of six toeight days is usually the time of delivery in the treatment group was required for normalization of platelet function. increased significantly, whereas in Wallenburg's Several publications have reported inci- study (1987). it was one to two weeks shorter dences of maternal hemorrhagic complications in the treatment group than in the control at the time of delivery andneonatal cutaneous group. This latter difference was notsignificant. and mueosal lesions in WOInen consuming The disparity between the two studies may per- aspirin." It must he noted. however, that these haps be explained by population differences, studies involved considerably higher dosages of with an usually high incidence of prior uterine aspirin. The same explanation can be made scarandrepeatcesareansectionin concerning the study by Daffos, in which he Wallenburg's treatment group when compared reported an umbilical cord puncture associated with his control group. with a moderately severe fetal coagulation dis- In our study (and we believe it will he turbance in a woman taking aspirin. confirmed by the larger study in progress). the Cases of premature closure of the ductus reduction in prematurity is one of the major arteriosus have been described in women on benefits of treatment. aspirin therapy (again, dosages werelarger In all three studies, no hemorrhagic compli- than those in our study). cations were n()ted in either mother or ne(mate. In addition, no fetal malformations were observed in any of the treatment groups that CONCU 'SION could be attributed to the medications. In our study, several patients receiving dipyrimadole Itis our belief that aspirin anddipyri- comp4led of headaches. Theseregressed madole will be shown to be an effective thera- rapidly with reduction in the dipyrimadole dose. py in the preventionof intrauterine growth Cessation of treatment was never required. retardatk)n in women who have had a prk)r It seems premature, however, toconclude similarly complicated pregnancy. and perhaps that the treatment regimen is innocuous. even more significantlyeffective in women When considering the potential adverse with a history of two abnormal pregnancies. effects of aspirin, itis important to note that This treatment modality appears to be both this medication does not generallymodify the effective and equally devoid of major risks. A classical coagulation parameters exceptfor controlled randomized double-blinded study those of platelet aggregation. In practice.bleed- with a placebo group. however, is felt to be

Sew Uzan et al. 181 Advances in the Pret,ention qf Lou Birthuvight

necessary before encouraging utilization of II. Masotti, G., Poggesi, L., cialanti, G., Abbate. R., and Neri Semen. G. G.(1979).Differential inhibition of these medications. In addition, the respective prostacyclin pn)duction and platelet aggregati(ni by roles of aspirin and dipyrimadole need to be aspirin. lancet 2. 1213-1216. clarified. This study is currently in progress. 12. Pedersen. A. K., and Fitzgerald. G. A.t198.1),Dose- related kinetics of aspirin: Presystemic acetylation of' platelet cyclooxygenase.Neu' Englwa Journal of Medici,w 311. 12(X1-1211. REFERENCES 13. Stuart. M. J.. Gross. S. J.. Elrad. 11., and Graeber. J. E. 1. fr.eI d..man, S. A.(1988).Preeclampsia: A review of (1982).Effects of acetylsalicylic acid ingestion on the role of prostaglandins.Obstetetrics and maternal and neonatal hemostasis.New England Gynecology 71, 122-137. Journal (tf Medicine 307, 909-912, 2. Stuart. M. J.. Sunderji. S. G.. Yambo. T. et al.(1981). Decreased prostacyclin production: A characteristic of chronic placental insufficiency syndromes, lancet 1. 1120-1128. 3. Walsh, S. W. (1985).Preeclampsia: An imbalance in placental prostacydin and thromboxane production. Anwrican Journal qf Obstetetrics and 6.'ynecology 152. 335-3-+0. 4, Crandon, A. J.. and Isherwood. D. M. (1979).Effect of aspihn on incidence of preechunpsia.lancet 1. 1356.

5. Beaufils. M.. Uzan. S.. Donsimoni. R.. and Colau. J. C.(1985).Prevention of preedampsia by early antiplatelet therapy. Lancet I. 840-842. .6. Elder, M. G.. De Swiet. M.. Robertson. A., Elder. M. and Hawkins, D. F( 1988). Low-dose aspirin in pregnancy. lancet 1. 4 10, 7. Lubbe. W. E.(1987). lAnv-dose aspirin in prevention of toxemia of pregnancy:Does it have a place? Dnigs 5I5-518. 8. Wallenburg, 1.1. C. S.. Makovitz..I. W.. Dekker. C. A.. and Rotmans. N.(198(3). Low dose aspirin prevents pregnancy-induced hypertenskm and preeclampsia in angiotensin-sensitive primigravidae. Lancet 1.1. 9. Wallenburg, 11. C. S.. and Rotmans. N.(1987). Prevention of recurrent idi()pathic fetal growth retardation by low dose aspirin and dipyridamole. American Journal of Obstetetrics and Gynecology 157, 1230-1235. 10. Moncada. S.. and Korbut. R.(1978).Dipyridamole and other phosphodiesterase inhibitors act as anti- thrombotic agents by potentiating endogenous prostacv din. Lancet 1. 1286-1289.

1 7 )

182 Prevention qf Intrauterine Growth Retardation ulth Antiplatelet Thempy Interventions

Statistical Findings: Prevention of Intrauterine Growth Retardation with AntVatelet Therapy

Table 15.1 Design Characteristics of Five ReviewedStudies

+ Prospective + + + + Randomization + + + Double-blind + 4- + + Placebo control + + 1985-' Date 1978-1984 *-1986 1983-1986 1985-*

* Start date ti)r Ntudy not available Study not completed at the time ot this wnting.

Table 15.2 Sample Sizes and Other Criteria of Five ReviewedStudies

Criteria for IUGR Acc. HTN I Pare Test IUGR > 2 RJGR: 1 IUGR: 1 IUGR > 2 Inclusion HTN ess. Terrain Angio II IUGR > 2

Treatment Asp. 150 group 1 Asp. 150 Asp. 60 Asp. 60 Dip. 300 Dip. 225 Dip. 225 Dip. 225 Asp. 150 group 2 0 0 0 Placebo Placebo group 3

16 Gestational Age 16 28 16 16 at onset

Number of Patients 93 44 48 300* :300*

300 Number of Pregnancies 93 44 57 300

Serge trzan et al. 171 783 Advances in the Prevention qf Low Birthuright

Table 1 5.3* Summary of Results of Three Completed Studies

A Treatment Control Treatment Control Treatment Control group group group group group group

Number of pregnancies 48 45 21 23 30 27

Premature births 12 3r 4 0 Mean gestational age at delivery 36.5 38.6 39 40 38 37

IUGR p 3 7 0 3 0 7 0

IUGR p 10 13 4 6 4 16 4

Hypertension 22 19 4 2 15% 6%

Intrauterine fetal development :3 0 0 1 0 1

IUFN + Neonatal death 5 0

Abruptio placenta 3 0

Preeclampsia 6 0 7 0

Edampsia 0 0 1 0 11% Cesarean sections 13 8 7 1 44% Birthweight 2625 3172 3040 3190

Normal pregnancies 12 29 Placental weight 509 599

'Whet tvro numbyri arc Iistod. the first corrcvonds to the control group.

Table 15.4 Evolution of Biomechanical Monitors

Treatment Group (N--.48) Control Group cv,45)

Baseline: Plasma uric acid (um01/1) 221 + 62 220 4-52 NS Platelet Count (x1.000/m1) 245 + 57 233 + 59 NS

Week Prior to Delivery Plasma Uric Acid 270 + 69 293 + 83 NS Platelet Count 249 + 66 209 + 57 < 0.02 Plasma Volume 56 + 7.5 49 + 8.1 < 0.02

184 Statistical Findings: Prevention (y hitrauterine Growth Retardation u'ithAntiplatelet Therapy Does Calcium Supplementation Reduce Pregnancy-Induced Hypertension and Prematurity?

JOSE VELAR, M.D. J.M. BELIZAN. M.D. J.T. REME, M.D.

INTRODLTCTION Pregnancy represents a period of very high calcium demand. There is an increased fetal requirement resulting in a total calcium accumulation at term of approximately 30 g, 30 percent of which is deposited during the third trimester. Some of the maternal adaptive mechanisms that could compensate for such high fetal demands are partially inhibited during pregnancy.) Furthermore, pregnancy has been suggested to be a period of "obligatory" high urinary calcium output,- while maternal bone calcium is protected.Finally, although intestinal absorption of calcium can increase,4 this is associated with an increased parathyroid hormone secretion during pregnancy.

17J Advances in the Prevention qf Lou' Birthweight

Itis possible, therefore, that when preg- effectof calciumsupplementation or nant women are put in a nutritional situation PIII/preeclampsia rates, Three randomized that can alter this metabolic balance, they can placebo controlled clinical trials and one be at higher risk for pathological events that uncontrolled matched study were conducted. have calcium-dependent mechanisms. This In one of our clinical trials, which was not could affect smooth muscle contractility of the designed to evaluate the incidence of PILL the vascular system and uterus. Several epidemio- placebo group had a relative risk of 2.8 of logical studies in pregnant and in normotensive developing P111 as compared with the calcium and hypertensive subjects consistently showed group (11.1% V. -4.0%): A preliminary report an inverse relationship between calcium intake from a randomized clinical trial conducted in and blood pressure.' '' Ecuador demonstrated that the group receiving Randomized controlled clinical trials of cal- 2.0 g of calcium a day (V = 46) had an inci- cium supplementation in normotensive women, dence of PHI of 6.5 percent, as compared with men, and mildly hypertensive patients have a placebo group (X = 16) which had an inci- demonstrated a significant reduction in blood dence of 28.2 percent < 0,011.° Finally, a ran- pressure. Data from two randomized clinical tri- domized controlled clinical trial was conducted als show a significant reduction in blood pres- in India in a population with a dietary calcium sure in the calcium-supplemented group at term intake of 500 mg/day. The study population (see figure 16,1). Furthermore, a dose-effect rela- was randomly assigned to a dietary intake of tionship is observed when our two studies are 375 mg/day of elemental calcium and 1200 IV combined (see figure 16.2). The effect is maxi- of vitamin D/day or to a placebo group. A non- mized with a daily dose of 2.0 g of calcium and statistically significant reduction in the inci- is less with 1.0 g of calcium supplementation. dence of preedampsia (> 140/90 mm Hg and Based on this information, we developed urinary proteins > 300 mg/21 hrs) (6%) was the hypothesis that calcium supplementation observed in the calcium-supplemented group during pregnancy can decrease the incidence CV = 188) as compared with the placebo group of pregnancy-induced hypertension (PHD/ (9%) LV = 182). It should be remembered that, preeclampsia and preterm delivery when com- to detect a difference of this magnitude with a pared with the placebo group. The effect could a = 0.05 and a power between 76 and 8,4 per- be mediated by reducing smooth muscle ten- cent, it would have been necessary to study at sion (e.g., vascular and uterine), least 8(X) patients in each group. When results from the three randomized controlled trials were pooled using the Mantel and Haenszel BACKGROI 'NI) INFORNMON: method,"women receiving calcium supple- THE RELATIONSHIP BEFWEEN CALCIV:k1 mentation had a statistically significant reduced SIPPLEMENTAPION AND PH 1/PREE0A\1pstA risk for pregnancy-induced hypertension, Table 16.1 presents summary data from pooled or typical OR = 0.41 (95, CI 0.39-0.78). four studies that provided information on the as compared with the placebo group. 171

188 Does Calcium Supplementation Reduce Pregnancy-Induced IIypertension ciul Premattolty? Intentqltions

PRELIMINARY EVIDENCE OF 11IF EFFIVI OF approved by the ohns fiopkins Hospital's Joint CALCILM St l'PLEMEN1ATION ON PREGNANCY- Committee on Clinical Investigation. INDUCED HYPERTENSION/PREECIAMPSIA Figure 16.3 summarizes the study design, criteria for patient eligibility, and follow-up We will present here preliminary results of requirements. The calcium-supplemented a double-blind, randomized,controlled clinical groups received four tabletsof calcium carbon- trial of the effect of 2.0 g of elemental calcium ate per day. Each of these tablets provides 500 (calcium carb)nate) a day or a placebo that is mg of elemental calcium, for a totalof 2.0 being conducted in two different populations. g/day. The placebo group received four tablets A white lower middle class population was of the same w,Aght, size, color, and organolep- studied at the Centro Rosarino de Estudios tic characteristics as the calcium tablets. Perinatales, Rosario, Argentina. All patients Both centers implemented a very elaborate were nulliparous and clinically healthy. Atthe mechanism of monitoring tablet intake and time of this presentation, 26 patients had been compliance as well as data quality control. enrolled and delivered their infants. The sec- Treatment compliance evaluation included ond population was obtained from the questioning patients on pill intake, pill count- Adolescent Pregnancy Clinic of the Johns ingat every visit (percentage of pills Hopkins Hospital, Baltimore, Maryland. By taken,'expected number of pills in that period). March 1988, 177 patients had been enrolled changing bottles at fixed times, and counting and delivered. All patients were less than 17 remaining tablets and urinary excretion of calci- years of age at the time of recruitment. Mostof um in a random sample of patients. the subjects were nulliparous with a known last Blood pressure, one of the main outcomes, menstrual period ( L.MP) and had singleton was obtained by trained personnelworking pregnancies. All were free of any underlying exclusively for the study; continuous monitor- medical disorders, as determined by history, ing and quality control mechanisms were physical examination, and laboratory tests. implemented. All blood pressures were Participants in bc)th centers were ranclinnly obtained using random zero sphygmomanome- assigned in a double-blind fashion before 26 ters. Table 16.2 presents reliability evaluationof weeks' gestation to one of the two treatment blood pressure measurements in the three par- groups using a randomizationschedule pre- ticipant hospitals of the Argentina center. It pared in advance for each population. Among shows high agreement values between the field white women. 230 were assigned to the calci- director and the study nurses. Table 16.3 pre- um-supplemented group and 196 to the place- sents, as an example, evidence of high repro- bo group. Among black women, 85 were ducibility of the blood pressure measure during assigned to the calcium-supplemented group pregnancy at the Ilopkins center, as well as a and 89 to the placebo group. Prenatal care was km. zero terminal digit preference for the pm- carried out according to the protocols of each ject coordinator. participant's hospital. The study prcnocol was Table 16.4 summari;.es the primary out-

Jose tillar et al. 175 IH9 Advances in the Prevention of Low Birthweight come variables and the definitions used in this observed in the calcium groups were almost study. Blood pressure values, used indepen- half the values obtained in the placebo groups. dently or with proteinuria obtained by the The rates of preeclampsia were not presented study nurses or project coordinator, and the for the black population because the indepen- diagnosis of PHI and/or preeclampsia made by dent clinical diagnoses were not available at obstetricians in charge of the clinical care of the the time of this presentation. At term, however, patients, but unrelated with the study, were the placebo group had higher mean systolic used as one of the primary outcomes. blood pressure 115 (+ 9.9) mm 11g, as com- Gestational age, obtained by using the best pared with the calcium group 109.4 (+ 11.2) obstetric estimation, which includes any ultra- mm Hg < 0.01), and higher diastolic blood sound measure, LikIP, uterine height, and the pressure 74.2 H- 6.5) mm Hg versus 68,7 (+ 9.1) time of the first fetal movements detected, was mm Hg < 0.01) than the calcium group. used to calculate the incidence of prematurity Table 16.6 presents the incidence of pre- (< 37 weeks). maturity V 37 weeks) and LBW V 2500 g) by In the Argentinean (white) population, treatment for the two studied populations. The women enrolled in the calcium-supplementel calcium-supplemented groups had a lower inci- group took 87 percent of the total expected dence of preterm deliveries, 5.4 percent for number of tablets, similar to 86 percent in the whites and -7,1, percent for blacks, than the placebo group. In the Hopkins population, placebo groups, 9.9 percent for whites and 19.1 although compliance fk ures were lower. they percent for blacks. The incidence of LBW was were very similar between the calcium-supple- also lower in the calcium group; however. mented and placebo groups (66% V. among whites, the magnitude is smaller. respectively). The calcium-supplemented and Among blacks, those supplemented had an placebo groups were very similar in most of incidence of low birthweight of 10,3 percent, the sociodemographic and baseline characteris- compared with 20.0 percent in the placebo tics for the black and white populations. There group (p= 0.07). vvre, however, differences in maternal weight at randomization, with white patients in the calcium-supplemented group and black women in the placebo group statistically significantly DISCI'SSION heavier. We have presented preliminary informa- Table 16.5 presents the effect of calcium tion on the effect of calcium supplementation supplementationontheincidenceof on gestational age, blood pressure. and PIII/preeclampsia. The calcium-supplemented PIH/preeclampsia. In agreement with previous groups had a lower incidence of Pill, as well reports (see table 16.1). there is a systematic as a lower incidence of preeclampsia in the reduction in blood pressure values and new white population and lower incidence of pro- evidence of reduction in prematurity in the cal- teinuria in the black population (NS). The rates cium-suppk.mented grtnips. There is also pre- 17,

190 Does Calcium Supplementation Reduce Pregnancy-Induced 113pertension tind Prenuaurity? Interventions liminary evidence from the white population those pregnant women receiving calcium sup- that calcium supplementation can prevent the plementation. The role of magnesium in hyper- development of preeclampsia. Nevertheless, tension has been well established.Its role as caution should he exercised in interpreting an efficacious tocolytichas also been well these results. They are obtained from two trials established.'" There is a consistency in these in progress, and some of the results couldbe observations. The interaction of calcium, modified when all of the patients are recruited parathyroid hormone, plasma renin activity, and analyzed. Itis unlikely, however, that the and magnesium results in alterations of intracel- observed direction of the effect will change lular calcium. The reduction of intracellular cal- dramatically. There remains a need to further cium will result in myocyte relaxation.There is elucidate the mechanism whereby calcium sup- no reason to believethat this effect is limited to plementation exerts its effect. Previous reports the vasculature. A similar effect should be have suggested possible effects on parathyroid expected in the myometrium, with resultant hormone,' plasma renin activity,'' and serum relaxation of the uterine smoth muscle. This magnesium.'' alteration of intracellular calcium (i.e., reduc- Increases in parathyroid hormone (PM) tion) is presumably the final commonpathway have been shown to increase intracelluhir calci- in the action of betamimetics, magnesiumsul- um in several cell types.The presumed mecha- fate, prostaglandin synthesis inhibitors, and cal- nism is an increase in membranepermeability cium channel blockers. Thus,itisnot with subsequent facilitated movement ofcalci- inconsistent to expect that calcium supplemen- um via slowchannels from the extracellular to tation could have an effect not only onblood the intracellular compartments. The question pressure but on uterine activity,and therefore remaining is what effect lcmg-term calcium sup- on gestational age, and,ultimately, birthweight. plementation will have on PTH, and whether While the results presented here are clearly chronic suppression of PTH will in fact result in preliminary, we suggest that they are provoca- lower levels of intracellular calcium. tive and encouraging. Further research atthe We have previously repurted that, in preg- clinical and basic science levels will be needed nant individuals with initial lowlevels of plas- to better characterize therole of calcium in m a renin activity (PRA),calcium supplementa- human reproduction. tion resulted in a greater reduction inblood pressure than in womenwith higher initial PRA." The uterine vascular bed, being sensitive ACKM MI.FGMENT to angiotensin II, may beaffected by calcium- Partially supported by grants from the induced changes in plasma renin activity. National Dairy Board, the National Dairy Patients with renin-dependent hyperten- Council, and the International Development sion frequently have depletion of magnesium. Research Center (IDRC), Ottawa, Ontario, Our previous report" has alsodemonstrated CANADA. increased serum magnesium levels among

Jose tillar et al. 191 17? Advances in tbe Prer'ention of Low Birthweight

REFERENCB 12. Mantel, N., and Haenszel, NV,(1959).Statistical aspects of the analysis of data from retrosp....ctive 1. Belizan, J. I.Villar, J., and Repke. J,(19881.The studies of disease.journal of the National (.-incer relationship between calcium intake and pregnancy- Institute 22. 719-748. induced hypertension:Up-to-date evidence. American Journal cy Obstetrics and Gynecology 158, 13. Altman, 1).. and Elbourne. I),Combining results 898-902, from .several clinical trials. British journal of Obstetrics and Gynaecoh)gy 95, 1-2, 2. Duggin, G. G.. Lyneham. R. C., Dale, N. E.. Evans,. A.. and Tiller, D. J.(1974), Calcium balance in preg- Repke, J. T., Villar. J.. Anderson. C,, Pareja, nancy. Lancet 2, 926. Dubin. N.. and Belizan. J. (Forthcoming). Biochemical changes assodated with calcium supple- 3. Reynokls, \X'.A., Williams, G. A.. and Pitkin. R. M. mentati( n induc.:d bl(x)d pressure reduction during (1981).Calcitropic hornume resp(msiveness during pregnancy.American journal of Obstetrics a ul d pregnancy.American Journal (1 Obstetrics and 6:311000logr, G:ynecology 139, 855, IS. Resnick.1..M.. Laragh, J.11., Sealey, J.E., and Ileaney. R. P.. and Skillman. T. G. (1971).Calcium Alderman. M. It(1983).Divalent cations in essen- metabolism in normal human pregancy. journai of tial hypertension:Relati( Ms between serum ionized Clinical Endocrinology and Metabolism 36.661. calcium, magnesium and plasma renin activity.Neu. England Journal (d.tfediciiie 309, 888. 5. Villar, J., Repke. J.. and Belizan, 3. M.( 1986), Calciutn and blood pressure.clinical Nutritm S. 16. Petrie. R. NI.(1981/.Tocolvsis using magnesium 153-160. sulfate. Seminars in l'erinatoh,gy 5, 266-27-3.

6, Parrot-(,arcia, M., and NIcCarron, 1).A. t 1984). Calcium and hypertension.Nutrition Reviews -12. 205.

7, Villar, J.. and Belizan. J. M. (198( ).Calcium during pregnancy. Clinical Nutrition 5, 55-62.

8. Villar,I., Repke. J.. and 13elizan, J.NI. ( 198'). Calcium supplementation reduces blood pressure during pregnancy:Results from a randomized controlled clinical trial.Obsteiric$ and (iynecology '0, 317-322.

9. Marya. R. K.. Rathee. S. and Manrow, M.(198). Effect of calcium and vitamin 1) st/nplementation on toxemia of pregnancy.(O'nech/gic and obstetric Itutestigation 24, 38-42. 10. Lopez-Jaramillo P.. Narvaez NI.. Weigel R.I. and Yepez R.(1989). Calcium supplementation reduces the risk of pregnancy-induced hypertension in an Andes populatkm,British Journal qf Obstetrics and Gynecology 19, 6.18-1i5S.

11, Kawasaki. N.. Matsui. K., Ito. M. et al.(1985).Effect of calcium supplementation on the vas( ular sensitivity to angiotension 11 in pregnant women. American journal c!f Obstetrics and Gynecology 153, 570-582.

17s 192 Does Calcium Supplementation Reduce Pregnancy-Induced IIJpertension and Prematurity? Interventions

Statistical Findings: Does CalciumSupplementation Reduce Pregnancy-Induced Hypertensionand Prematurity?

Table 16.1 Incidence of Pregnancy-Induced Hypertension(PIK Data From Calcium SupplementationStudies

Elemental Calcium Total Calcium Blood Pressure PIM incidence (Percentage) Supplement (mg/day) Intake (mg/day) (mm Hg) Calcium Placebo Study 21.2 p < 0.10 Kawasaki, N. et As 156 942 No effect 4.5 (22) (66) 4.0 11.1 NS Villar et al. 1500 900 4-5 (24) (25) 6.0 9.0 NS Marya R. et al. H 375 500 4-7 (182) 1,200 L11 Vit D (188) 6.5 28.2 p < 0.01 Lopez jaramillo 2,000 300

NOn-R4n(jiV771./i0---Nio Place4)0 II in pregnant tSource:Kawasaki N., Matsui K., Ito M. et al. i198S). Lttect ofcalcium supplementatitm on the vascular sensitivity to angiotension women. American journal 01Oh5fetric s and Gymvology, 153, 576-, 582. during pregnant v: Result from a random- Source: Villar 1., Rvpke I., and Belizan I. M.1 987). Cak ium supplementation reduc es blood pressure ized controlkd clinical trial.Obstetric and (ynecolergy, 70, 117-322. Eftec t ot cal( iurn and t namin D supplementation ontosemia of pregnant vGyne(ologic ItSourcv: Marva R. K., Rathee S. and Manrosv M. (1987). and CThstetric investigation, 24, 38-42. redu«.s the risk ot pregnancy-induced hyper- f Source.: Lopez-laramillo P., Narvaez M., WeigelR.14., and fvfx'l R.( 989). Calc ium supplementation tension in an Andes population. British lournalof Obstetrics and Gvnec okigy. 19, 648-655. Preeu lampsia (> 140/90 rnm I fg and urinarproteins > 100 ry1,1/..!4 hours

Table 16.2 Table 16.3 Reliability of Blood Pressure Measu(ements Reliability of Blood Pressure Measurements Rosario, Argentina (Whites) Johns Hopkins Hospital (Blacks) A. correlation coefficients betweenblood pressure values by estationaI age for project coofdinator tV Systolic Diastolic Weeks of Gestation Systolic Diastolic Hospital A 58 0.82* 0 81 28 v. 30 0.76 0.63

B 28 0.77 0.50 32 v, 34 0.61 0.51 0.81 0,71 C 54 0.74 0.81 36 v, 38 38 v. 41 0.66 0.65 13. Zero terminal digit preference (N 1,011 measures) Systolic 14.1 no 13.1 MO " Correlation cc.)ettit rents between field dire( torand stuch. nirrws Diastolic-

193 Jose t'illar et al. 17 (;) Advances in the Prevention of Low Birtinveight

Table 16.4 Table 16.6 Outcome Variables for the Randomized Clinical Effect of Calcium Supplementation on the Trial of Calcium Supplementation During Incidence of Prematurity and Low Birthweight Pregnancy Preterm LBW Blood Pressure (mm Hg) (< 37 weeks) (< 2,500 g) (study nurses and project coordinator) (Percentage) (Percentage)

Random zero sphygmomanometer Whites (diastolic B.P. 4th and 5th found) Calcium 5.4(223) 8.9(223) Seated position (Hopkins) Placebo 9.9(192) 10.4(192) Supine and lateral (Argentina) Blacks Proteinuria: 300 mg/I proteinuria two times Calcium 7.1 (85) 10.3 (86) more than 6 hours apart Placebo 19.1 (89) 20.0 (89) (Argentina) p=0.01 p=0.07 > 300 mg/1 or (+) (Johns Hopkins) Gestational age: Best obstetric estimation: LW'. Ultrasound, rundal Clinical diagnosis of PIH or Preeclampsia height. andearly tetatmovement. made by obstetricians not related to the study

Gestational age (weeks) Best OB estimation (ultrasound. LMP, Figure 16.1 uterine height, fetal movements) Differences in Adjusted Blood Pressure Mean Physical examination newborn Values at Term Between Calcium and Placebo Groups (Adjusted for Race and Initial Blood Birthweight (grams) Pressure < 26 Weeks)

-6 Table 1 6.5 -4.94" Effect of Calcium Supplementation on -s PII-1/Preeclampsia: A Randomized Double- Blood -4.09* Blind Controlled Clinical Trial pressut -4 -3.65" differenc e

t mint - Whites Blacks Preeclampsia -2 BP' > 140 BP* and > 300 > 300 mg/ mm Hg and/or mgi Proteinuria Proteinuria Proteinuria Calcium 11.7 (230) 1.7 (230) 5.9 (68) (2.0 g/day) (NS) Supine Lateral Supine Lateral Sample size Placebo 19.4 (196) 3.6 (1%) 11.4 (70) (Placebo-Cai 27-23 27-23 26-24 26-24 Systolic Diastolic Treatment started tx.tm 20 and 26tseeks" gestat 2 times > 6 hours 'Betweengrouos: p 0.(16 > 300 mg/1= (+) "Between slopes1.3 last nleasures); p0.05

,) 194 Statistical Findings: Calcium Supplementatio and Pregnancy-Induced IIlpertensUm and Prematurity Interventions

Figure 16.2 Differences in Systolic Blood Pressure at Term BetweenCalcium-Supplemented and the Corresponding Placebo Groups in Two Studies ofCalcium Supplementation During Pregnancy

-9 -8.5 mm Hg'

'Source: Author. (See chapter lb).

'Source: Marva R. K., Rathee S. and Manrow M. 11987). Effectof calcium and vitamin 0 supplernentation on toxemiaof pregnancy. Gynecologic and Obstetric Investtgation, 24, 38-42. Placebo 1 g 1.5 g 2 g (Reference Group; Calcium Croups locap 86-387 Lateral Position.

Figure 16.3 Study Design Randomized Double-Blind Controlled Clinical Trial

Rasehne Randomization Whites = 196 V Subjects eligible n: Blacks Clinically healthy Placebo Single Pregnancy Parity 0-1 Knoell t_MP Registered hetore 20 weeks NO drugs or treatments

g Flemental Calt ion a Dav Calcium Whites N.230

Chnical ex imination Blacks = 87 Blood samples 24 hour dietary, recall Ur nary sample

2t1 14 rm

fuse 1.711ar et al. 195 1 53 Magnesium Suppkmentation in Pregnancy: A Double-Blind Study

LUDWIG SPATLING, M.D.

INTRODUCTION The recent interest in the element magnesium is probably in part accident and in part due to the development of new techniques, such as atomic absorption spectrophotometry permit- ting accurate assays. These technological advances facilitate correlation of laboratory results with clinical disorders. First some important qualities of magnesium are identified. After potassium, magnesium is the second most abundant intracellular cation. On the one hand, its biological effect can be attributed tothe formation of chelation compounds. On the other hand, it is a cofactor of ATP (adenosine triphosphate), thus explain- ing its influence on al)out 300 enzymatic reactions, such as carbohydrate metabolism, lipid metabolism, and nucleic acid metabolism. It is Advances in the Prevention of Low Birthweight involved in stages of protein synthesis and is hypomagnesemia shows up in growing tissue. an important prerequisite for the proper func- The possibility that magnesium deficiency tioning of energy-releasing reactions, such as might cause early delivery was first contemplat- glycolysis and oxidative phosphorylation.2 ed following the accidental observation that the A person needs 0.16-0.21 mmol of magne- reduction of premature labor was improved by sium per kilo of bodyweight per clay, which treating calf muscle cramps with magnesium. In corresponds to 11.5-14.4 mmol per day for a a subsequent study, 10-15 mmol of magnesium bodyweight of 70 kg. During growth and preg- per day were given orally where the dosage of nancy, an additional 4 mmol per day are need- beta-adrenergic drugs could not be reduced. ed. Based on metabolic studies of pregnant Subjective and objective methods of assessment women, the World Health Organization (WHO) showed decreased incidence of uterine contrac- Food and Nutrition Boards recommend a sup- tions, and, at the same time, indicated that the ply of 18.5 mmol of magnesium per day during required beta-adrenergic dose could be pregnancy (1 mol Mg = 24.3 g Mg).' reduced.- Investigations during recent years have The publication of this pilot study caused a repeatedly shown that the magnesium content number of other studies. In a retrospective of plasma falls significantly during pregnancy, study, lower incidences of intrauterine growth particularly at the beginning.4 It is discussed as retardation, preterm rupture of the membranes, a physiologic dilution. It should be mentioned and preeclampsia by magnesium supplementa- that renal magnesium loss in pregnancy tion during pregnancy were described.8 increases by 26 percent, shown by 24 urinc' Because of the studies cited above, a dou- samples in healthy women during the whole ble-blind study was conducted to clarify pregnancy.' whether or not magnesium supplementation Chronic hypomagnesemia in the pregnant has any influence on the pregnant woman and woman carries over to the tissues. Using the fetus or the newborn, respectively. nwometriwn samples of over 100 cesarean sections, it could be documented that the magne. sium content of the smooth uterine muscles decreases significantly with increasing pregnan- Dot 'BLE-BUND 511.DY cy age. There is also a highly significant corre- A total of 568 women (normal and high- lation between the muscle magnesium content risk) who attended the outpatient clinic at the and plasma level.° Department of Obstetrics, University of Zurich, This leads to the conclusion that, although agreed to participate in the study. They were magnesium intake is not increased during the enrolled in the study as early as possible. but period of greater need, there is yet more loss not later than 16 weeks of gestation. The sam- through the kidneys due to the increased ple size was determined by the limitation of the glomerular filtration, which is not or is inade- recruitment period, which, for practical reasons, quately compensated for, and that this long-term could not exceed two years. 1SJ

198 Magnesium Supplementation in Piegnancy: A Double-Blind Study Interventions

The method of allocation was based on Median maternal weight increase was 11 kg in the subjects' even or odd date of birth. both groups, and there was no difference According to their group assignment, the sub- regarding the systolic and diastolic blood pres- jects were given either 15 mmol of magnesium- sure and different degreesof oedema. aspartate-hydrochloride (Magnesiocard, Verla Magnesium serum levels showed no differences Pharmacy, Tutzing, FRG) or 13.5 mmol of between the groups. aspartic acid per day. At each visit, patients Patients were hospitalized if they experi- were questioned onregular intake, number of enced antepartum hemorrhage, irregular tablets, and side-effects of the medication. A preterm contractions, or progressive cervix mat- subsidiary analysis was based on 437 women uration. In the magnesium group, 44 women who claimed that they had regularly ingested were hospitalized for 533days. In the placebo the magnesium or placebo. group, 65 women had to stay inhospital for The Mann-Whitney U test was used to 887 days. The average duration of each admis- compatv the central trends.Categorical vari- sion to hospital was the same in both groups. ables were analyzed using the chi-squared test. Among indications fbr admission tohospital, The results were considered significant at the 5 hemorrhage during pregnancy, incompetent percent level. Assessment alwaysused two- cervix, and preterm labor were significantly tailed tests. Values were given as medians and more frequent inplacebo-treated women. the 5th and 95th centile. For various reasons, There was also no difference in the number of such as refusal to take further tablets, delivery miscarriages. in other hospitals, or abortion, somedata were The median gestation was significantly not available for analysis. longer in women treated with magnesium, Two hundred seventy-eight women were although the difference between the medians treated with magnesium, and 290 received the was not more than oneday. Delivery before 37 placebo. Age, parity, and gravidity were the weeks occurred less in the magnesium group, same in boll groups. There was nodifference but the difference was not significant. The with regard to the birthweight of children born serum analysis showedthat low magnesium before the start of this study or the duration of levels in the first half of pregnancy are of no previous pregnancies. Based on the clinicalhis- predictable value for preterm birth. There were tory of the patients, the risk ofabnormal preg- no differences regardingthe duratkm of the nancy was comparable inthe two groups. first and second stages of labor. The same is The frequency of complaints attributed to true for the incidence of surgery orother com- the tablets svas low and comparable in the two plications during delivery. groups. In the magnesium group.1 woman There are no statistically significant differ- complained of diarrhea, .4 of nausea, 6 of vom- ences between theunselected groups with iting, and 6 of heartburn; in the placebo group, respect to placental andinfant weight; length 2 complained of diarrhea, 1 of nausea,10 of and head circumference; and birthweightbelow vomiting, 6 of heartburn, and 1 of fullness. 2500 g, below 1500 g, and below the10th per-

1S .4 Ludwig spading 199 Advances in the Preivntion of Lou Birthweight centile; however, fewer infants with a head cir- reflected a trend in the main study became sta- cumference less than 33 cm were found in the tistically significant, while other results that were magnesium group. There are also no statistical- already significant in the main study were con- ly significant differences in arterial and venous firmed at a higher level of significance. umbilical pH or frequencies of infants with an The difference between hospitalization fre- APGAR score of 7 or less. Significantly fewer quencies was significant at a higher level. The infants in the group receiving magnesium were number of infants below 2500 g and below 500 admitted to the neonatal intensive care unit. g was also significantly smaller in the group of The biggest differences were found in the mothers who regularly took magnesium. The admission rates for preterm birth and asphyxia. difference between numbers of children with a The numbers of the subgroups arc too small to head circumference below 33 cm became high- show any statistical difference. ly significant. In the subsidiary analysis, we excluded A general trend toward iiv7roved fetal out- ise women who did not fulfill the protocol of come and adaptation, already seen in the main medication as prescribed, leaving 217 women in study, was confirmed by statistically significant the magnesium group and 220 in the placebo results; birthweight, length, and head circumfer- group. As expected, the effect of magnesium ence were higher, and the number of children supplementation demonstrated itself more clear- with a 10-minute APGAR score of 7 or below ly (see table 17.1). Some of the results that only was less in the group with regular magnesium

Table 17.1 Table 17.2 Comparison Between Comparison Between Magnesium and Placebo Groups Magnesium and Placebo Groups After Exclusion of Patients Who Did Not Take Their Tablets Regularly Magnesium Placebo group group Significance Magnesium Placebo (N =278) (N =290) group group Significance Variable 50th centile50th centile (N 217) (N -= 230) Variable 50th centile50th centile Indication for Hospitlization Birthweight (g) 3340 :3300 < 0.05

Hemorrhage (N) 4 17 < 0.05 No. < 2500 g 6 18 < 0,05 Incompetent No. < 1500 g 0 6 < 0.05 cervix (N.) 8 17 < 0.05 Infant length (cm) 50 49 < 0.05 Preterm labor (N) 12 26 < 0.05 Head circum- Gestational age ference (cm) 35 34 < 0,05 at delivery No. with (weeks/days) 40/0 39/6 < 0.05 10-min Apgar Admission to score 5 7 0 5 < 0.05 neonatal intens- No, of women ive care unit (N) 20 36 <0.01 hospitalized 26 48 < 0.01 1SLJ

200 Magnesium Supplementation in Pregnancy: 4 Double-Blind Study Interventions

supplementation. Fewer mothers with mild REFERENCES edema were counted in the magnesium group. 1. Spading, L., and Spading, G.(1988).Magnesium The advantageous effect of magnesium supplementation in pregnancy:A double-blind supplementation during pregnancy is clearly study.British finirnal qf Obstetrics arid Gynaecology 95,120-25. (The results are published in parts,) documented in decreased incidences of hemor- rhages, incompetent cervix, and preterm labor, Gunther, T.(1981).Biochemistry and pathobio- chemistry of magnesium.Magnesium Bulletin 3, and in prolonged gestation. If preterm delivery 91-101 is defined as birthweight below 2500 g, in this 3, Seelig, M. S.(1980).Magnesium deficiency in the study, we could reduce the incidence from 7.7 pathop,enesis of disease.New York:Plenum to 18 percent. A statistical difference between Publishing Corporation. the frequencies of abortion could not be seen: 4, De orge, F.13.. Domingos, D., de 1.7lhoa Cintra, A. however, a positive effect of magnesium sup- B.. and Antunes. M. L.(1)05),Magnesium concen- plementation during pregnancy in respect to its tration in the blood serum of normal pregnant women. Obswtrk's and Gynecology 26, 253-255. clinical syndrome was evidenced by the low incidence of hemorrhage. Itis interesting to i.Sp;itling. L.. Kunz. P. A_ Fluch, R.. and flue)), A. (19851.Magnesium and calcium excretion during note that the two infants who were very light pregnancy. Magnesium Bulletin 91-93. for gestational length were seen in the placebo Spading. L.. Kunz. P. A., \Onderschmitt, D. J.. Iluch, group. One infant born at 31 weeks' gestation R.. and !Inch. A. (1983). Zum Magnesiumgehalt der weighed 1000 g and another one horn at 32 nerusmuskulatur im 111, Trimenon.Archives of weeks weighed 1190 g. Of course, this number Gynecology 235. .rn is still too small to establish a significant influ- SpAling. 1.. 1981/.Orate Magnesium-Zusatz- ence of magnesiumsupplementation on therapie lxi vorz.eitiger Wehent tigkeit.Geburtshiye muenbeik -4 1() 1- 102 intrauterine growth retardation.' 8. Conradt, A.. Weidinger. H., and Algayer. H.(1984). on the role of magnesium in fetal hypotrophy, pregnancy induced hypertension and preeclampsia. Magnesium Bulletin 6, 68-76. CONO.1.'SION Magnesium supplementation during pregnancy was associated with significantlyfewer maternal hospitalizations, a reduction in preterm deliveries, and less frequent referral of newborns to the neonatIA intensive care unit. The results suggest that magnesium supplementation during pregnancy has a significant influence on fetal and maternal nunbidity both bef,ire and after delivery. The results of this study strongly recommend magnesium supplementation during the whole period of pregnancy.

Luduig Spat !hitt; 201 An Ovetview of Trials of Social Support During Pregnancy

DRNA ELBOURNE, PH.D. ANN OAKLEY

NTRODI ICTION The main purpose of this chapter is to update our earlier review in Evian of controlled trials of social support in pregnancy.' We begin, therefore, by summarizing the focus, methods, and conclusions of that review. We then point Out the ways in which we have altered these methods for our present work, show the updated results, and discuss their implications for current practice and future research.

1986 REVIEW We started from the presumption that a pregnant woman is more than just the carrier of her unborn child. She is a person in her own

1S7 Advances in the Prevention qf Low Birthuvight right, with a lik apart from that seen by most considered seven pregnancy outcomes: Low antenatal caregivers. This life includes a number birthweight, preterm delivery, anaesthesia, anal- of stresses and strains. Pregnancy, although gesia, prolonged labor, instrumental delivery, often a joyful experience, may also add to these and adverse psychosocial outcomes. existing pressures, and may create new stresses. Studies of the relationship between social class, Statistical nwtlxxls stress, and reproduction suggest that social sup- We followed the methods of yusuf and his port in pregnancy may be able to mitigate colleagues' to formally combine information maternal stress and improve maternal and infant from these trials in a procedure which has health.' We therefore examined the controlled become kmmn as meta-analysis or oveniew. trials of social support in pregnancy. For each trial, we calculated an odds ratio, which is the odds of getting a particular (usually adverse) outcome (e.g., low birthweight) in the experimental group, compared to or divid- Mimms ed by the odds of getting the same outcome in Selectiotz ((trials the control group. If the odds ratio isI,it In order to find these trials, we searched implies that the odds are the same in the the Oxford Database of Perinatal Trials for pub- experimental and the control group (i.e., for lished and unpublished' reports. The criteria that outcome, it makes no difference whether for entry in the computerized data base are that the intervention is experimental cw c(mtrol). If the trial participants should be human; that the the odds ratio is less than 1,it suggests that th,.. time of intervention should be during pregnan- experimental intervention offers some protec- cy or the neonatal period; and that allocation to tion against this adverse outcome. If the odds the intervention should be random or quasi- ratio is greater thanI,it suggests that the random (i.e., using alternate alk)cation or using experimental intervention may do harm. The date of birth or casenote numbers). odds ratio is .in estimate, and so the confidence For the Evian review, we selected all trials with which it can he stated varies directly with on the data base coded as -counseling, support, the size of the study; that is, the smaller the and educatim in pregnancy.- Trials included in study, the less confident one c..m be that the the analysis were either those designed explicit- estimate is a true reflection of reality, and the ly to provide social support or those utilizing wider the confidence interval is that can he what we judged to he supportive approaches to placed around the estimate. Conversely. a large antenatal care. These included such disparate sample size means greater precision and a interventions as antismoking advice; health edu- smaller confidence interval the estimate is cation; organization of maternity care; income a more precise reflection of the truth). support; and preparation for pregnancy, child- Therefore, for each study, we show an estimate birth, and parenthood. In the 29 trials of social of the odds ratio with its 95 percent confidence supportin pregnancy so identified we interval. In an overview of several trials, we use 1ES

204 An Vi 'en 'ieuf l'rials (yr Social Szeppor t During Pregnancy biterventions all of these within-trial estimates to calculate a all trials on the data base with interventions typical odds ratio, but because several trials are coded as "counseling, support, and education in included, we thereby increase the precision of pregnancy." What this disparate group had in the estimatethat is, we can narrow the confi- common was that we thought they were the dence intervals. type of' intervention that might be supportive. As a response to comments received following this 1986 review, we have limited our inclusions in the present chapter to those in which social CONGA tiIONS support was an explicit aim of' the trial (i.e., the Our conclusions in 1986 were that "In authors' intent, rather than our surmise). terms of the seven outcomes considered, the Thirdly, the overviews are presented in available experimental evidence shows these order of' their methodological quality. This effects (of the social interventions) to be either quality is assessed in terms of the likely exis- inconclusive or beneficial:1 and this is support- tence of three types of bias: Selection bias at ed by the conclusions of the observati(mal entry, selection bias after entry. and measure- studies." We ended our review by calling for ment bias. The "best" trials are put first, and the better designed, larger trials, specifically in the -worst" are put last. field of social support. The final inethodological difference between the Evian review and the present review arises because, in the interim, a number of trials primarily concerned with social support PRSENT OVERVIEW have been mounted. Thus, we have been able The aims of our present overviewto to subdivide (wr analyses into those based on review the experimental evklence of the effects trials in which social support is the primary of social supp( rt in pregnancyare largely the intervention, and those trials in which itis an same as they were three years ago. In this explicit co-intervention. We have looked for a update. IRAvever, we ad()pt thur slightly differ- dose effect, expecting that the benefits of sup- ent approaches. port would be greater in the former group. First, this chapter focuses on the two pregnancy outcomes most relevant to thetitle of the conference: Gestational age at delivery. and CI IARA(ITERItirICS OF TRIALS OF birthweight. A number of other outonnes (such EXPLICIT So XIII. SI. *PPORT IN PREGNANCY as intrapartum events, the healthof the mother and baby, health-related behavior. the attitudes We identified 35 trials of social support in and feelings of the mother, and the baby's pregnancy. The characteristics of these trials are development) will be considered and discussed described in table 18.1. The final column of this elsewhere.. Ta ble provides examples of the f()rm of words Secondly. our earlier review was based on which helped us to decide on their inclusion.

Diana Ilbourne and Ann Oakley 205 S;) Advances in the Prevention of Low Birthuvight

The results of some of the trials have nc)t five studies.''''''' Allocation is said to be ran- yet been published!' u and still must be consid- dom in the remaining 30, hut often the details ered provisional. are sketchy, or suggest that the author may be We have categorized the studies under six using the word random in a nontechnical sense. main headings. The provision of social support The sample sizes are equally variable, is the primary aim in six studies.'"'' The remain- ranging from 10 women° R) 1,763 pregnant ing 29 trials all aim to provide explicit social women.'" In two papers'''" the sample size is support, but as a co-intervention. In these trials, not given. Overall, the studies are based on at the other primary interventions are antismoking least 5.500 women allocated to a supportive advice (3);1')-" income support (1);" information intervention in pregnancy, and at least -1.600 feedback and information sharing (6);" " the women with less support serving as controls. organization of care (7);'- and preparation for As with the entry criteria, the outcomes pregnancy, childbirth, and parenthood (12)." " under investigation largely reflect the type of This last group includes such diverse interven- intervention being offered. Some include such tions as, for example, preparation for breast- traditionally clinical outcomes as birthweight feeding. nutritional advice, and enhancement of and gestation at delivery (e.g 1,40,1), whereas maternal attachment processes. The headings others empk)y a number of sociopsychok)gical are not necessarily mutually exclusive, and outcomes such as 'love scores' or postpartum some studies could be categorized under more attachment behavior." than one heading. It is worth noting at this point that the trial The entry criteria for the trials included in hy Runnerstronf9 is not only the largest (and our overview vary greatly. To a large extent. therefore the most influential in the overviews), this reflects the differing aims of the studies. For but is also the trial most likely to suffer from instance, those studies aimed at reducing the selection biases. These biases could occur both number of cigarettes smoked were targeted at at allocation (using casenote numbers) and women who WM' smokers (e.g., [191), Some after entry (by excluding complicated deliver- studies wishing to alter the organization of care ies). Therefore, where relevant, the typical odds toward more midwifery control concentrated on ratio is calculated twice, with and without the women at low risk of poor obstetric outcomes c(mtribution of Runnerstrom.s study. (e.g., [391). In others, midwifery care is compared to medical care for women at high risk (e.g., [10)). In fact, because the principal entry criterion for the study by Rehrer and Wolin' PREGNANCY O.( TCOMES was residence in a low-income neighborhood, Gestational age at deliroy the studies do not even share in common the (completed aeeks) entry criterion of being a pregnant woman! Seven of the twenty-nine trials of explicit Allocation to the intervention is clearly social support in pregnancy as a co-intervention quasi-random (e.g., using casenote numbers) in give information about the preterm (< 37 com-

206 An Overview of Thais qf Social suppert During Pregnancy Interventions pleted weeks) delivery rate. They find no evi- in the rate of very low birthweight (see figure dence to suggest that such support has any effect 18.7). In contrast, all three trials in which sup- on this outcome (see figure 18.1).The same con- port is the primary intervention suggest a non- clusion follows from the typical odds ratio based statistically significant decrease in the rate (see on four of the six trials in which support isthe figure 18.8). Although this may be indicative of primary intervention (see figure 18.2). a dose effect, the nunthers are small,and the Very few trials provide information about confidence intervals overlap. the rate of extremely preterm V 33 completed weeks) delivery, Neither the one trial in which social support is an explicit co-interventkm (see Drsu IssION figure 18.3) or the three in which it is the primary interventkm (see figure 18.4) provide Itis perhaps disappointing that this conclusive evidence of an effect of social sup- overview has provided no conclusive evidence port on this rate. about the effects of social support in pregnancy on the two selected clinical outcomesconsid- ered above. Certainly, there are no clear implications for current practice. Bllarmuo rr There are a number of possible explana- Fifteen of the thirty-five trials in which tions for this continued uncertainty about the social support in pregnancy is an explicit inter- effect of such support. vention provide some information about the It is possible. of course, that social support low birthweight V 2500 g) rate. The typical in pregnancy has such negligible effects that no odds ratio based on 10 trials in which social amount of further research will provide a clear support is an explicit co-intervention suggest answer to whether itis either beneficial or that such support may result in a statistically harmful. From these overviews, particularly significant reduction in this rate (see figure those in which support is the primary interven- 18.5). The 95 percent confidence intervals of all tion, itis possible that there are some benefits, but the last trial,however, include unity. but that these are likely to be modest. This Excluding this trial from the overview shows no implies that larger numbers are needed to avoid statistically significant effect of social support on making a "type 2- errorthat is. of not being the low birthweight rate. The same conclusion able to detect an effect, even if one exists. can be drawn from the five trials inwhich sup- Larger numbers could be obtained from port is the primary intervention (see figure 6). two main sources. Firstly, none of the outcomes Six trials give information about the very overviewed contain data from all of the 35 trials low birthweight V 1500 g) rate. All three in listed in table 18.1. It is possible that renewed which support is a co-intervention, separately attempN to elicit such information fromtheir and together, suggest that such support is asso- authors may yield some dividends. Not only ciated with a nonstatistically significant increase were many of these trials carried out quite some

Diana Mourne and Ann Oakhy 207 191 Advances in the Prevention of Low Birthweight time ago, however, but, given their heterogene- that these outcomes are of relevance but there ity (particularly in the larger group of trials (29) are additional outcomes to which we should in which support was a co-intervention), itis devote our attention. unlikely that such data were ever collected. If only because of the vvay in which 'W2. A second route to obtaining larger numbers have chosen to present our overviews, we have is by increasing (from 3;) the number of studies implicitly assumed that low birthweight and in the overview. There may be completed and preterm delivery are unwelcome outcomes. published trials which we have not been able to This assumption is certainly arguable. Both of identify from our search of the Oxford Database these outcomes are to some extent proxy mea- of Perinatal Trials. This is perhaps more likely in sures associated with adverse outcomes for the the field of social support than for other inter- child. But there are certainly also arguments ventions considered in this book, as such against considering low birthweight and research may not have been conducted by the preterm delivery as adverse in themselves, clinicians nor published in the clinical journals unless they are associated with a worse out- which have been the prime sources of entry come in the longer term. Nonetheless, in gener- onto the data base. It is also possible that trials al, the selected outcomes are associated with in which support was not the primary interven- adverse events, and so may provide useful tion were not coded sufficiently to ensure their information about the effects of support. identification as having support as a co-inter- Othei outcomes are also 4 relevance. As vention. While we continue to be alert to both indicated earlier in this chapter, we intend to these possibilities, it is unlikely that concentrat- review the effects of supportive interventions ing on them will yield great dividends. on other clinical outcomes, such as fetal and We are aware, however, of a number of tri- infant mortality and morbidity. Inevitably, some als from which results will be available in the indicators of morbidity will also be proxy mea- near future (see table 18.2). At leastfour of sures (such as Apgar scores, neonatal resuscita- these trials have already conipleted recruitment tion, and admission to special care nurseries). hut cannot yet provide data suitable for inclu- We also wish to draw on the example of the sion in the Overviews (see table 18.2A?). At least work of Olds et al." on child abuse, and by five more trials are still in progress (see table Gutelius et al."' on childhood behavior to con- l',.2B?). We are also aware of other trials which sider possible longer term effects on the moth- are at the planning stage. Asthe results from er, the child, and the interaction betweenthem. these trials accumulate, we are likely to obtain These examples also indicate that we plan to more definitive answers to our questions about examine outcomes other than those which the effects of social support in pregnancy on the might be defined as strictly clinical. These can specific outcomes considered in this chapter. be categorized broadly as health-related behav- A further explanation for the inconclusive- ior (such as cigarette smoking. bfeastfeeding. ness of the results here presented is that we are and clinic attendance) and sociopsychological considering either the -wrong" outcomes. or outcomes (e.g., postnatal depression, satisfac- ) 4 41.

208 An Ckviview of MakSocial Support During Pregnancy

c.. Interventions tion with the intervention, involvement by the process of overview in this field (as in other family members or friends in caring for many others) is ongoing and(auto)dynamic. the baby, and doing housework). The effects of social support in pregnancy are. These are of interest both in their own as yet unclear, but the wealthof information right (as outcomes of the supportive interven- which will become available over the next few tion) and also as indications of the possible years should provide someof the answers (as mechanisms for the effect of the support. For well as raise more questions). instance, does support, by reducing stress and At this stage, we can only concur with the depression, have some biochemical effect on conclusion by Bryce and his colleagues, that fetal growth? Alternatively (or additionally). 'even if the trials fail to demonstrate abenefit, does support help to encourage pregnant the cooperation between social scientists and women to attend regularly atantenatal clinics, obstetricians to undeitake this research should thereby facilitating more effective clinical benefit pregnant women in that it promulgates surveillance? a more holistic view of pregnancy:''' In addition to trying to increase the quantity and range of information insubsequent overviews of social support, we also plan to REFERENCES consider the quantity and type of interventions 1. Blondel. 13., et al.(n.d.).Prevention of preterm deliveries by home visiting midwives:Results of a being provided. To some extent, we have French randomised controlled trial.Unpublished begun this process of examining a dose effect manuscript. by separating those interventions in which sup- 2. Bryce, R. L.. Stanley, R. J.. and Enkin. M. W. (1988). port is an explicit co-intervention(low dose) The role of social support in the prevention of from those in which the support is the primary preterm birth. Birth 15. 19-23. intervention (high dose). But we also wish to 3. Campbell. S., Reading, A. E., Cox. D. N., Sledmee, take into account the length of time over which C. M.. Mooney. R., Chudleigh. P.. Beedle. 3.. and Ruddick, H.(1982).Ultrasound scanning in preg- this support is provided. The total or parthal nancy: The short-term psychologicaleffects of early absence of social support is often a long-term real-time scans. journal it/ Psycf.nymmunic Obstetrics problem. Hence, it may be unrealistic to seek rind 6:ynaecology 1. 57-61. to substantially remedy its adverseeffects with 4. Carpenter. J., Aldrich. K.. and Boverman. 11,(19M). The effectiveness of patient interviews: A controlled a short-termdose of support in pregnancy. study of emotional support during pregnancy. Indeed, it may be argued that offering and then Architvs (il- General Psychiatry 19. 110-112. quickly withdrawing such help may even wors- S. Carter-Jessop. L.(1981). Promoting maternal attach- en the situation. There may,therefore, he a ment thr(mgh prenatal intervention, Maternal-Child case for increasingthe dose by continuing sup- Naming journal 6. 107-112. port into early parenthood. Thework conduct- 6. ( halnwrs. L. Hetherington. J.. Newdick, M., Mutch. L.. Grant. A.. Enkin, M., Enkin, E.. and Dickersin. ed by Olds et al."' suggests that such a dose (1986).The Oxford Database of Perinatal Trials: effect may, indeed, exist. Developing a register of published reports of con- It is clear from the above discussionthat trolled trials,Cuntrulled Clinical Mats'. 306-324.

209 Diana Mourne and Ann Oakley 1 39 Advances in the Prevention of Low Birthweight

7.Cogan. R., and Winer, J. L.(1982).Effect of child- 18. Fischer, W. M., Huttel, F. A., Mitchell, I., and Meyer, birth educator communication skills training on post- A. F..(1972). The efficacy of the psychoprophylac- pattum reports of parents.Birth 9, 241-244. tic method of prepared childbirth.In:N. Morris 8, Dance, J.(1987). A social inteivention by linkwork- (Ed.), Psychosomatic medicine in obstetrics and ers to Pakistani women and pregnancy outcome. gynecology: Proceedings of the 3rd International Unpublished manuscript. Cmgress, London (pp, 4-44). Basel: Karger. 19. Flint. C., and Poulengeris, P.(1987). 9. Dolcetta, G.. Azzini, V., Zacche, G., Zimmermann- Know your Tansella, C., Bertagni, P., Siani, R., and Tansella, M. midwife. London: Heinemann. (1979). Traditional and respiratory autogenic training 20. Gutelius, M. F., Kilsch, A. D., MacDonald. S.. Brcioks. in psychoprophylaxis for childbirth.A controlled M. R.. and McErlean. T. M. (1977). Controlled study study on psychological and clinical effects in primi- of child health supervision:Behavioural results. parous.In:L. Zichella (Ed.), Psychosomatic Pediatrics 60, 294-30.4. medicine in obswtrics and gynecology (pp. 929-936). Proceedings of the 5th International Congress, Rome. 21. Hebei. J. R.. Nowicki. P.. a'nd Sexton. M.(1985). The effect of antismoking ntervention during preg- 10. Donovan. J. W. (1977). Randomised controlled trial nancy: An assessment of interactions with maternal of antismoking advice in pregnancy. British journal characteristics..4merican journal (..y Epideinkfty of Provntitte and Social Medicine 31, 6-12. 122.135-148. 11. Donovan, J. W., Burgers. P. L.. Hossack. C. M.. and 22. Heins. H. C., and Nance, N. W (1986). A statewide Yudkin. G. D. (1975). Routine advice against smok- randomized clinical trial to reduce the incidence of ing in pregnancy. journal qf the RQyal College (f low birthweight'very low birthweight infants in General Practice 25. 264-268. South Carolina.In:E. Papiernik. 0. Bream and N. Spira (Eds.). Prevention of preterit! birth (pp. 12. Durham, L.. and Collins. M.(198(j).The effect of music as a conditioning aid in prepared childbirth 387-410). Paris; INSERM. education. journal of Obstetric. Gynecologic, and 23. Heins. II. C., Nance, N.. and Efird, C.(n.d.). Neonatal 4Vursing15, 268-270. Multicenter randomized controlled clinical trial for the prevention of low birthweight in South Carolina. 13. Elbourne. D.(1987).Subjects' views about participation in a randomised controlled trial.Journal cf Unpublished manuscript. Repmductije and Inliint Psycholou 5, 3-8. 2.4. Hetherington, J.(1987). An international survey to 14. Elbourne. D., and Oakley, A.(Forthcoming). The indentify unpublished and ongoing perinatal trials, 6mtrolled "trials 8, 287. role of social interventions in pregnancy.In: M. Enkin, M. J.N. C. Keirse, and I. Chalmers (Eds.). 25. Hobel, C. J.. and Bemis. R. L.(1986). West area Los iffectire care in pregnancy and childbirth.Oxford: Angeles prematurity prevention demonstration pro- Oxford University Press. ject,In:E. Papiernik, 0. Bréart, and N. Spira (Eds.), Prevention (9`. preterm birth (pp. 205-222). 15. Elbourne.I)..Richardson. M.. Chalmers,I., Paris: Waterhouse, I.. and Holt, E. (1987).The Newbury INSERM. Maternity Care Study:A randomized controlled trial 20. Hol)el. C. J.. Bragonier. R.. Ross. M., Bear, M.. Bemis. to assess a policy of women holding their own R.. and Mori. B.(1987).West Los Angeles prema- obstetric records.British journal qf Obstetrics and ture prevention program:Significant impact. Gynaecology 9.4, 612-619. journal qf Perinatal Medicine 15. 112. 16,Ellis, D, J., and Ilewat. R. J.(1984).Factors related 27. Kehrer, B. 11.. and Wolin. C. M.(1979).Impact of to breastfeeding Juration.Canadian Family income maintenance on low birth weight: Evidence Physician 30, 1479-1484, from the Gary experiment. journal of Human 17. Field, T.. Sandberg. D.. Quetel. T. A., Garda. R.. and Resources1,i, .f34-462. Rosario, M.(1985).Effects of ultrasound feedback 28, King. II.. and Eiser. J. R.(1981).A strateAy for on pregnancy anxiety, fetal activity, and neonatal counselling pregnant smokers.Ikalth Education outcome. Obstetrics and Gynecology 66. 525-528. journal 40, 66-68.

210 An Overview of Ma& of Social Support During Pregnancy Intertvntions

29, Utley, J., and Forster, D. P.(1986). A randomised 40, Olds, D. L., Henderson, C. R., Tatelbaum, R., and controlled trial of individual counselling of smokers Chamberlin, R.(1988).Improving the life course in pregnancy. Public Health 100, 309-315. development of socially disadvantaged mothers: A randomized trial of home nurse visitation. 30. Little, B. C., Hayworth, J., Benson, P., Hall, F., Beard, American Journal of Public Health 78, 1436-1445. R. W., Dewhurst, H., and Priest, R. G.(1984). Treatment of hypertension in pregnancy by relax- 41. Reading, A. E., Campbell, S., Cox, D. N., and ation and biofrzedhack. Lancet I, 865.4367. Sledmere, C. M.(1982).Health beliefs and health care behavior in pregnancy. Psychologi,w1Medicine 31. Lovell, A., and Elbourne, D.(1987, 19 March). 12, 379-383, Holding the baby and your notes.Health Sewke Journal, 335. 42. Reading, A. E., and Cox, D. N.(1982). The effects of ultrasound examination on maternal anxiety lev- 32, Lovell, A., Zander, L. I., James, C. E., Foot, S Swan, els. journal of Behavioral Medicine 5, 237-247. A. V., and Reynolds, A.(1986).St. Thomas' Maternity Case Notes study: Why not give mothers 43. Reading, A. E., Cox, D. N., Sledmere, C. M., and their own case notes? London:Cicely Northcote Campbell, S.(1984).Psychological changes over Trust. the course of pregnancy: A study of attitudes towards the fetus/neonate.Health Psychology 3, 33. Lovell, A., Zander, L. I.. James, C. E., Foot, S., Swan, 211-221. A. V., and Reynolds, A.(1987).The St. Thomas' Hospital Maternity Case Notes Study: A randomised 44. Reading, A. E., and Plan, L. D.(1985).Impact of controlled trial to assess the effects of giving expec- fetal testing on maternal anxiety. Journal of tant mothers their own maternity case-notes. Reproductitv Medicine 30, 907-910. Paediatrks and Perinatal Epidemiology 1, 57-66. 45. Reid, M. E., Gutteridge, S., and Mcilwaine, G. M. 34. Oakley, A.(1985).Social support: The -soft" way (1983).A comparison of the delivery of antenatal to improve birthweight?Social Science and care between a hospital and aperipheral clinic: Medicine 21, 1259-1268. Report to Health Services Research Committee, Scottish Home and Health Department. 35. Oakley, A., Chalmers, 1., and Elbourne, (1986). The effects of social interventions in pregnancy.In: 46. Runnerstrom, L.(1969). The effectiveness of nurse- E. Papiernik, G. Brart, and N. Spira (Eds.), midwifery in a supervised hospital environment. Prevention of preterm birth:Neu, goals and new Bulletin of the American College qf Nurve-Midwifery practices in prenatal care (pp. 329-361).Paris: 14, 40-52. INSERM. 47, Sanunons, L. N.(1984).The use of music by 36. Oakley, A., Macfarlane!, A. J.. and Chalmers, I. women during childbirth.Journal of Nurse- (1982). Social class, stress and reproduction.In:A. Midwife?), 29, 266-270. R. Rees and H. Purcell (Eds.), Disease and the envi- 48. Scott, K.'.1984).Reduction of low birthweight with ronment (pp. 11-50). Chichester: John Wiley. enhanced antenatal care.Pediatric Research 18(i), 37. Oakley, A., et al. (n.d.).Social support during preg- 345A. nancy. Unpublished manuscript. 49. Sexton, M., and Hebei, J. R.(1984). A clinical trial 38. Olds, D. L., Hende-:;on, C. R., Tatelbaum, R.. and of change in maternal smoking and its effect on Chamberlin, R.(1986).Improving the delivery of birth weight. Journal of the American Medical prenatal care and outcomes of pregnancy: A ran- Association 215, 911-915, domized trial of nurse home visitation.Pediatrics 50. Shereshefsky, P. M., and Lockman, R. F.(1974), 77, 16-28. Comparison of counselled and non-counselled 39. Olds, D. L., Henderson, C. R., Chamberlin, R., and groups and within-group differences.In:P. M. Tatelbaum, R.(1986).Preventing child abuse and Shereshefsky and L. J. Yarrow (Eds.), Psychological neglect: A randomized trial of home nurse visita- aspects qf a first pregnancy and early postnatal tion. Pediatrics 78, 65-78. adaptation (pp. 151-163). New York: Raven Press.

Diana Elbourne and Ann Oakley 211 195 Advances in the Prevention of Low Birtinveight

51. Spence Cagle, C.(1984).Changed interpersonal needs of the pregnant couple: Nursing roles. JOGN Nursing 13, 56. 52. Spencer, B., and Monis, J.(1986). The family worker project:Social support in pregnancy.In:E. Papiernik, G. Bream and N. Spira (Eds.), Prevention of prekrm birth: New goals and new practices in prenatal care (pp. 363-382). Paris: INSERM. 53, Spencer, B., Thomas, H., and Morris J.(1989), The family worker project: A randomised controlled trial of the provision of a social support service during pregnancy.British JournalofObstetrics and Gynaecology 96, 281-288. 54. Spira, N., Audras, F Chapel, A.. Debuisson, J., Jacquelin, C.. Kirchhoffer, C., Lebrun, C., and Prudent. C.(1981).Domiciliary care of pathological pregnan- des by midwives:Comparative controlled study on 996 women. journal de Gynecologie Obstetrique et Biologie de la Reproduction 10, 543-548. 55. Stanley, F. J.. and Bryce, R. L.(1986). The pregnancy home visiting program.In:F. Papiernik, G. Breart, and N. Spira (Eds.), Provntion of preterm birth: Aew goals and new practices in prenatal care (pp. 309-328). Paris: INSERM. 56: Sweeney. C., Smith, H., Foster, J. C.. Place, J. C., Specht, J., Kochenour. N. K., and Prater, li.NI. (1985).Effects of a nutrition intervention program during pregnancy:Maternal data phases 1 and 2. Journal of Nurw-Afidulfety 30, 149-158. 57. Wiles, L. S.(1984). The effect of prenatal breastfeeding education on breastfeeding success and maternal perception of the infant. jOGNNursing 13, 253-2i7,

58.Yanover, NI. J., Jones. D., and Miller. M. 1).(1978). Perinatal care of low-risk mothers and infants,Neu. England Journal ofMedicine 94,702-705. 59. Yauger, R. A. (1972). Does family-centred care make a difference? Nutsing Outlook 20, 320-323. 60. Yusuf, S., P:to, R., Lewis, T., Collins. R., and Sleight. P.(1985), bol !Acvkade during and after myocardial infarction:An overview of randomised trials. Progress in Cardiozuscular Diseases 27. 336-371. 01. Zimmermann-Tansella, C.. Doketta, G., Azzini. V., Zacche. G., Bertagni. P., Skull, R., andTansella, (1979).Preparation courses for childbirth in primi- para:A comparison. journal qf Psychosomatk Research 23, 227-233, in

212 An Oven,iew ql1rials qf Social Support During Pregnancy Interventions

Statistical Findings: Prevention of Preterm Deliveries by Home Visiting System

Table 18.1: Characteristics of Trials of ExplicitSocial Support in Pregnancy

Inclusion Criteria Types of Entry Criteria Allocation to Nunthers Outcomes Study (for overview) (papers & dates) interventions Interventions

Costs; women's views, Smial support Blondel et al. Organization of care Women attending hospital Random (sealed including satisfaction (unpublished) Ill outpatient clinics or envelopes) with care; birthweight; Usual care + 1-2 hospitalised with threatened 79 gestational age. home visits a week preterm labor; 26-36 weeks' by midwives, gestation,

Usual care for 71 women with threatened preterni labor. 4 Nervousness and use of "Provision ot emotional Carpenter et al. Lmottunal s,uppw-t Women registered in a lime of prenatal visits medication before, support" ip.109), 1968 141 hospital prenatal during, and after labor; interviews at intervals clinic. 52 length of labor. throughout pregnancy, labor, and the puerperium with 1st year medical students,

Usual care SO ,t- "Maternal attachment Postpartum attac hrnent Carter-lessop Preparation for Women between 32 and 37 Random paxess , enhancedby behavior. 1981 151 parenthomi weeks pregnant, from specific prenatal private Obstetricians' intervention" (p,1093. Enhancement of practices. Primiparae, maternal svhite, married, expecting 'Mothers encouraged to attat hment proc ess. nom omplicated increase awareness of pregnancies, attention( e fetal activity and notice Usual antenatal care, at childbirth classes, how they cart affect their activity" ip,111i. _ Parents feelings and 'Nonjudgmental wpport Cagan and Winer Preparation tor Childbirth educators. Random perceptions of events in for expectant parents" childbirth 1982 171 pregnant v, labor and 43,241). puerperium. Communication 22 workshop to give instruction and practit e in listening and responding to people's behavioral needs. Usual training. 21 - - Birthwerght, maternal Antenatal supptin Immigrant Pakistani , Alternate Dance 1987 Socia/ support linkstorkers win-nen from rural physic al ps hostx tat (unpublished) 181 health, intrapartum care, Usual care + "link- villages in Azad Kashmir 25 ; neonatal health. know workers" (minimum 3 and Rawlpindi, with ledge of health. home visits and 2 history at 1 or more telephone low birthweight babies, riot ass(t wed with 25 Usual care. elettise cesarean, gross malformation, or multiple birth.

213 Diana Elbourne and Ann Oakley 197 Advances in the Prevention of Low Birthweight

Table 18.1: Characteristics of Trials of Explicit Social Supportin Pregnancy (Continued)

Study Types of Entry Criteria Allocation to NumbersOutcomes Inclusion Criteria (papers & dates) Interventions Interventions (for overview)

Donovan 1977 1101 Antismoidng advice Women attending ante- Table of random Birthweight; gestational *Support needed and Donovan et al. natal clinics at 3 numbers age; other neonatal contact ... maintained" 1975 L111 Doctor at each ante- hospitals; smoking > 280 measures; smoking in (Donovan et al. 1975, natal visit giving cigarettes daily; < 35 pregnancy; smoking p.266). intensive individual years; < 30 weeks' advice; instrumental/ antismoking advice. gestation; no previous operative delivery rate. perinatal deaths, Usual antenatal care. 308

Durham and Preparahon (or PrimagravIda couples Table of random frequency of intrapartum -Relaxation decreases Collins 1986 1121 childbirth receiving private child- numbers medication. anxiety ... music birth education therapy as an aid to Childbirth education instruction. 15 psychological and following textbook by physical control of pain" Hasid and tape- (p.268). recorded music during conditioning exetcises 'Music.. provided and relaxation/ common ground for- breathing techniques. couples to relate with each other" (p.270). Childbirth education 15 following textbook by Hasid.

Elbourne et at. Informal on feedbacWWomen < 34 weeks' Random Women's feelings of Hypothesized that lwomen 1987 1151 sharing gestation registered for (consecutively satisfaction, being holding their own Mourne 1987 1131 antenatal care with 1 numbered, sealed informed, anxious, obstetric notesl would Women held sole obstetrician in a opaque envelopes) 147 confident, in control; feel less anxious, more obstetric record from peripheral antenatal father's involvement; confident, more in registration until 10 clinic. depression; health- control .. and would days postpartum. related behavior; find it easier to analgesia/anesthesia; communicate with staff" Women held usual 143 mode of delivery; length (Mourne et al., p.613). cooperation cards. of labor; birthweight; gestational age; administrative effects. --t-- Ellis and Hewat Preparation for Pregnant women planning Random Breastfeeding at 1 4nd "Erxoaraged to telephone 1984 1161 parenthood to breastfeed; able to three months; use ot nurse clinician for communicate in English; semi-solids at 3 and support" (p.1481) Nurses educated about cohabiting: resident in 62 six months, fireastfeeding. Mothers urban toll-free zone. given in-hospital assistance by nurse clinician and prenatal breastfeeding classes, and encouragement to telephone for support, and postnatal help.

Nurses educated about 60 breastfeeding. Mothers given in-hospital assistance by nurse clinician. t Field et al, 1985 Information feedback/Women referred tor Random Anxiety; fetal activity; "Attempt . , . to 1171 sharing ultrasound assessments maternal sleep; obstetric alleviate pregnancy ot gestation, and postnatal complica- anxiety" tp.525). Women viewed ultra- 20 tions; neonatal behavior. sound monito r. and given running descriptions.

Women told about fetus' 20 well-being and gestation.

214 Statistical Findings; An Overview of Trials of Social Support Dr;ring Pregnancy Interventions

Table 18.1: Characteristics of Trials of Explicit Social Supportin Pregnancy (Continued)

NumbersOutcomes Inclusiun Criteria Study TYPes of Entry Criteria 11/ Allocation to (for overview) (papers & dates) Interventions Interventions

-Emphasized that martyrdom Fischer et al. Preparatkut for Printiparae. By EDD Length of labor: analgesia in labor. was not the point" ip.381. 1972 1181 chikfbirth

"Stress husband Course on psycho- 31 participation" ip.40). prophylactk method ot prepared childbirth Two sessions with 6 couples at a time

Usual care. 41 4-- 'Mothers will esperrence Flint and Organisation ot care1-Womenof low obstetric risk Random (sealed Clinical details greater emotional Poulengeris 1987 enrolling for full envelopest inc luding birthweight: 503 gestational age: length satisfaction." 1191 'know VOW midwife' consultant antenatal and c are. Team of 4 mid- detiverv care. of labor; use of wives thrmighout analgesia and pregnancy, latxu, and anesthesia; mode ot puerperium. delberY; mothers' VieWS.

Normal hospital care. 498

; Intant and maternal "Counseling nd Gutelius et al. Preiuration tor Unmarried printigravidae tiv stern at random beh.w(or. antic ipatory guidame" 1977 1201 parenthood ages 15-18 before 7 ! numbers months gestation (p.294i. 47 Proied nurse from 7th "frequent contact tp,29h). month of preAnancv. Routine health care and crxinseling on preparation for infant iand «mtinuity of care 1 for 3 years)

Usual prenatal care 48 One visit neonatally,

Birthweight; gestation "Social support and stress Heins et al. 1986 Svc support < weeks' gestation; Random R omputer. at delivery: maternal reduction" . 1221 previous low birthweight gerwrated random weight gain Heins et al. intensive antenatal tare baby; > 10 t m Crvasv risk- numbers in sealed 684 (unpublished)1231 from nurseimidssite in 5 s( oring system; free of opaque envelopes at reponal low hirthweight medic al or obstetric coordinating center, c inics; sot ial support complications (e.g., opened in response and stress reduction; hypertension, renal to tetephone call from increased number of disease, diabetes, or 1 of S tertiary visas; assessment of multiple pregnant centers) cervis: education re signs of preterm labor and on health- relatt-si behavior: emphasis on weight gain; continuity of care.

Usual antenatal care in h87

! high-risk clinic. -t income matntenance Rarxiom twang Watts- ; Birthyseight Kehrer and Wolin ! Income supix)rt Residents ot low-income program to encourage 1979 1271 ne;ghllorhoods.131at k Conlisk ally( ation increased utilization of Negative income tas head of household; at ! model) ! 102f3' prenatal care" tp.440). tami I les received fv%e least 1 tkpendent 241, monthly payments when < 18 years. "May have broad influence previous month's on social behavior" ip 457i earnings below /111 'break even" level) 771 Families No negative income tas 148: " Single live births

Diana Mourne and Ann Oakky 215 19) Advances in the Prevention of Low Birthuvight

Table 18.1: Characteristics of Trials of Explicit Social Support inPregnancy (Continued)

Study Types of Entry Criteria Alkscation to Numbers Outcomes Inclusion Criteria (papers & datesi Interventions Interventions dor overview)

Liffey and Fonter Anti-smoking advke Smokers (> 1 cigarette Simple random in Smoking in pregnanty. *Individual counseling" 1986 1291 per day1 at first ante- hltxksuf 8 ip.3011. Including advice and natal visit; < 28 weeks' 77 booklets at first ante- gestation *Reinforcing advice . . natal visit plus additional encouragement

reinforcing letter after . . home visit" ip.310i. 2 weeks and home visit further 2 weeks later.

Usual antenatal care. 74

1- Little et al. 1984 Informatton feedback/Women with HP 135/ ; Sequentially Antenatal hospital "Extend relaxation 1301 sharing 85 mm Hg at 2 admission; length of achieved in biofeedback successme antenatal antenatal stay; blood and relaxation sessions to A. Relayation tra n ng.visits 18 pressure in fate stresses experienced in pregnancy: pregnantv everyday life" (p.865). O. Relayation training 18 outcomes, and biofeedback.

C. Usual practice. 24

Lovell et al. 1986 Information feedback'Women attending hospital Random iwnsecutivelv Women's feelings of 'Communication .. women's 132) sharing antenatal c knit. ot 1 numbered sealed satistaction; sense of choices" ip.t Lovell et al. 19E37 obstetrician. env-elopes) control; communications 1331 Women held sok. 115 with staff; lather's "Encourage information Mourne 1987 011 obstetric record from involvement: health- sharing and increased registraticm to delivery. related behavior: participation in decision clinic al outcomes such making.- Women held usual co- 1 20 as birthweight, gestation, operation cards intrapartum anesthesia, length of labor, mode of delivery; administrative effects. -7. - -.

Oakley et al Sochi] support Previous lens birthweight Random by central Hirthweight; gestational Soc ial support to (unpublished) I71 baby iwithout maior telephone alkx.ation age at delivery; maternal 'influent e mother's Sot ial support Inc furl. formation, multiple 251 maternal and infant physical and mental health mg home visits by pregnancy, or elective morbidity. maternal iantenatally and post- midwives. and delivery). Fluent psychological condition natally), affec t process telephone calls to English.speakers, postpartum, and of labor and delivery . . . women and 24-hour satisfaction with care and inc Tease women's self- 'hotline to midwives. ontidenc e about mothering.'

Usual antenatalare. 254

Olds et al. 1986a Soc ial support Prim yams vvrimen either Stratified by marital Obstetrical labor and 'Apple( iation fur the full 1481 < 19 years or single status, rat e, geograph- neonatal details: use ot set of stressful family Olds et al. 19861) A C ontrol no services parent or lossk4X W- ical regions. 9(1 health services: bealth and community influences 139) pre wided through e(0410(111(status: < Assigned at random habits inc luding dret on wornen's health habits Okls et 31 On rt-se,eri h projec sseeks gestation. Women dress assign and smoking: gestational and behaviors" ip.16). press) 1401 mem dec k ot age at delivery; birth. Free transportation ards. Decks recon- 94 weight. 'Home visitation by nurses and regular prenatal stituted periodically should be an ettedise and well,child care tel in-errepresent means cM . responding treatment with flexibly to the oressful (As (81lus ourse- smaller number of 100 Irte c ire umstances with home visitor during sulsiec ts ias E.fron's Shk.h sex tally disadvantaged pregnant s biased win design). women must «Intend' ip.17). 'Parent education. enharu D. As ar plus nurse 16 home-v isitor until ment ot the women's informal support sYstems to c hrld rear bee. 2 emphasize the strengths of years of age the ssonwn and their families' tp.171 2,..

216 Statistical Findings: An Overview of Thals cf Social Support During Pregnancy Interventions

Table 18.1 : Characti_ristics of Trials of Explicit Social Support inPregnancy (Continued)

Induskm Criteria Study Types of Entry Criteria Allocation to Numbers Outcomes (for overview) (papers & dates) Interventions Interventions

informative Reading et a). Infonnation feedback' Primievae having real- RAMum Attitudes to scan, about ^Scanning pregnancy, about fetus, and emotionally rewarding 1984 1431 sharing time ultrasound jr Reading et al. 10-14 weeks. Caucasian, (especially anxiety): when specific and detailed (eedback made 19821411 High feedbackshown in stable relationship; 67 health related. Reading and Cox screen and given feed- ages 18-32; without available to mother" (p.59). 1982(42) back about size, shape, history of miscarriage, Campbell et al. and movement. infedility, Of risk of "Enhances awareness oi fetus" (p.60). 1982131 congenital malformations. 62 Low feedbacknot -Shtirt-term effects on shown screen, global ansiety levels" tp.239t. evaluation of progress. -t Attitudes to scan, "Psychologic impact of Reading and Platt Information feedtxu is` High.risk population. Random about pregnanc y. about technology. .. concern 1985 1441 t.haring fetus (especially stressful e(fects- High feedback 11 anxiety). (p.907). ultrasoundsaw screen and nurse indicated *Anxiety levels' (p 910). features visualized. Low feedback 8 ultrasoundnot shown screen, global evaluation. Fetal monitoringoon- stress or contractions stress test using external feral heart monitor. Video controlshosyn ultrasound record that they were aware was not their own.

To deal w ith "failure of Reid et al. 1981 ONanizaPon ot care Women referred by CP Random with Obstetric morbidity and from Easterhouse area stratification mortality; birthweight; communic ation nd (ack of 1451 continuity of care' Community antenatal for ddivery at Glasgow 100 intrapartum anesthesia clinic. Royal Maternity Hospital. and analgesia; (P.3). instrumenteoperatise Routine hospital sersice.:; 100 delivery rate; pattern of usage of ditties; financial effects; communication; women's perceptions of pregnancy and childbirth and reported management of first months ot infant lite.

prenatal visits; "f eel full-time nurse- Runnerstrom 1969 ()rganization oft are Women coded tasenote nunlbers analgesia and midwifery service with ' "uncomplicated" attend ng 1461 commitment to patient and lbstetric care given by large training hospital. anesthesia; length of family c an provide more (1) nurse midwives 758 labor; birthweight; mode of delivery: integrateci and ccmsecutive (21 obstetric residents. 1().05 comphcations; conditioncare, comhaning of baby. fragmentation" (p.41).

. _ Use ot music during "Uplifting effects ot Sammons 1984 Preparation tor Women attending Lamaze Random designation labor; perceived effect music arise (rom music's (471 childbirth childbirth classes of«msecutive I class series of musk. ability to OforitOre inter- cwrsonal relationships, to Exposure to taped 24 facilitate ac hievement of music during active cell-esteem and to energize rehearsal of breathing and bring order through patterns. rhythm" (p.2bbi. Usual antenatal classes. 30

Diana Mourne and Ann Oakkv 20 1 217 Advances in the Prevention of Low Birth:might

Table 18.1: Characteristics of Trials of Explicit SocialSupport in Pregnancy (Continued)

Study Types of Entry Criteria AlIncatIon to Numbers Outcomes Inclusion Criteria (papers & dates! Interventions interventions (for overviessi

SexMn and Hebe) Anti-smoking advke Smokers of at least 10 Random Smoking habits; preg- "Assistance with smoking 1984 1211 cigarettes per day at nancy aukardes: birth- cessation . . support" Hebei et al. 1985 Staff assistance with beginning of pregnancy: 441.i weight; gestational age; tp.9191. 1491 smoking cessation. At < 18 weeks' gestation. intrapartunt anesthesia least 1 personal visit and analgesia; mode ot supplemented by delivery; length ot labor. frequent telephrme and mail contacts. Encouragement through information, support, practical guidance, behavioral strategies.

Usual antenatal care. 472

Shereshefsky and Preparation for "Normal" women and Random E motional disturbance ounsel ng wry ice" LOckman 1973 1501pregnancy/childbirth/ their husbands during in pregnancy; physical 41.1511 wrenthood tipit pregnancV complications in pregnancy and diddhirth; "Dem ted to emotional Soc 1.31 work; relationship between needs ot pregnant women counseling service. husband and wite in the arid their tamdies- pregnanc y. ip.1529. Control.

Spence Cagle 1984 Preparation lin. Coupk., parte loafing Random Caring relaticm- ',stinted at couple inter- (511 pregnark in Lamaze classes. inventory: fundamental personal needs tor interpersonal relation.. control, affection, and In-( lass homework orientation behavior. inclusion' cp.Rn. aimed at couple. Interpersonal needs tor control, affection, and inclusion during pregnancy.

Usual Lamaz classc

Spencer and Morris So(' stipport Women at increased risk Random l3inihsse'ight; outcome ot "Social support service." 1986 1521 ite having loss birth- pregnancy; gestational Spencer ('t al. Otter of family worker, weight baby regktering ti55 age; intrapartum anal: lin press) 1531 Client-centered at 2 maternity units in gesia and anesthesia; approach. South Man( hester. length of tailor: mode tI dellyery. Usual care. . . _ - Splra et al.1981 Organization of care ; Pregnant women at high Random Preterm delis VrIV.: "( ant inuit t c' aro at 1541 risk of poor auk cures. hi rttuvelght, death. home, net by strangers in Dom it iliary r are hy (-pathologic at 4'18 antenatal admission, hospital.' midwives. pregnant les"). Operative/ Instrument,i1 Usual care deloery rate 425 . . Sweeney et al, Nutritional advice Able to commune ate in Random if fron'. Maternal protein and "Providing indisidual c are 1985 15W English: tree ot pre- biased coin design) alorie intake and according to the spec Mc Individual nutritional existing medic al within pre-gravid 22 maternal weight gain in needs ot eat h unique ads it e based on tfiggins' disorders: not private weight, and weight late pregnant y. Ante- mother-intant unit" formota from Montreal patients or teenagers; gain in test 20 natal, intranartum, and Diet Dispensary singletons. weeks. strata postpartum maternal west optic in and ornplic meets: infant counseling. birthweight and gestation at delivers . Usual care. 21

2 r.'

218 Statistical Findings; An Overvitiw of Trials qt. Social Support DuringPregnancy Interventions

Table 18.1: Characteristics of Trials of Explicit Social Support in Pregnancy(Continued)

Inclusion Criteria Study Types of Entry Cnteria Allocation to Numbers Outcomes (papers & dates) Interventions Interventions dor overview)

Wiles 1984 1571 Preparation for breast- Primigravtdae; > 32 Randomly assigned Breastfeeding; maternal 'Anticipatory guidance . . feedirw and parenthood weeks' gestation; taccording to the perception of infant. for assisting parents* planning to breastfeed; class in which they (p,253). Breastieeding education attending chlkibirth were enrolled) 20 class. education classes; vaginal delivery; no Normal childbirth medical complications; 20 preparation class, gestation at delivery 36-43 weeks; 5 minute Apgar:. 7; normally formed.

Yanover et at Organization of care Parity 0, 1; 19-15 years Random Length of labor; 'Endeavor to respond to wishes of numerous 1976 1581 old; low risk medically; analgesia; anesthesia: Family-centered care 12th grade education: 44 mode of delivery; length families to enhance family mainly postpartum but parents 1Mng together; of hospital stay; participation . including continuity of father prepared to attend complications in mother collaborative perinatal care from prenatal antertatal classes , Or baby, continuity ut care.- classeS. Traditional care, 44 -

'Family-centered care . . Vauger et at 1972 Organization ot care Residents of specified Random Health knowledge; identification of problem 1591 areas; 3-8 months health behavior: amily-centered care: pregnant; at least 30 health status, and needs of family and nursing service, child < S years. provision of appropriate identifying pwbtems I service for even, member" and needs of family ip.3207. and providing appropriate care. Minimum of 4 home

Usual care. 4----

Anxiety; p.Un; ciulidition -autogenous training . . . Zimmermann- Preparation tin- Married primiparae ages Random effects of deep relaxation Tansella et al. childbirth 20-15; rto physical ot baby. including diametrically opposed to 1979 161, abnormalities. hirthweight; analgesia: Doketta et al Prenatal classes 14 length ot labor; mode ansiety ..related to feelings ut calmness' 1979 191 including autogenous ot delivery. trainingdeep relaxation.

Prenatal c lasses in( kid- ing specific gymnastic exert. ises.

Diana Mourne and Ann Oakley 203 219 Advances in the Premntion of Low Birthweight

Table 18.2: Uncompleted Trials of Explicit Social Support in Pregnancy

A. Recruitment Complete Awaiting Data Analysis and Report(s)

Principal investigator Nature of the intervention Sample Allocation Outconws of intervention and place

S. Elliott Social support-education First- and secondarme Week of clinic enrollment Incidence of psychiatric disorder, espet ratiy T. level-ton groups led by psychologkt mothers ts =2001 postnatal d vression ; psychological well- London, England or health visitor being; use of support networks; hirthweight

I. king Supportive antismoking Smokers attending first ante- Random Number of t igarettes smoked; mother's I.R. Eiser advice reinforced by natal visit (s w 300) cooperation Bristol, England 1281 counseling and bkxhemical testing at home _t C. Larson Prenatal home visits by w 1,548) Random Health and social status of mother and t had Montreal, Canada community health nurses to 3 years postpartum, childrearing attitudes and behavior of mothin - K. Scott Enhanced maternal care Women with history Of loi Stratified random Patient compliance and behavior change; Nova SCotia, Canada 1481 from physical and lifestsk. birthweight delivery antenatal referrals; obstetric interventions; assessment by community =, 10,574t birthweight; stillbirths; NICU admissions; health nurses infant mortality, morbidity

8. In Progress

Principal insestigator Nature tit the intervention Sample Allot ation Outt ames ot intervention and place -+ S. Ng Intense crenatal tare including iistory of preterm dells eft, Stratified randomization so Preterm delivery; stillbirths, spontaneous New York, U.S.A. weekly visit, ptblyic stillbirth. of spontaneous that equal numbers abortions in 211d or Ird trimester: hirthweight examination; edut aticm on abortion > 2052: age .4 1O, assigned to each treatment self-palpation; ho ITW Uter ine > 33; registered for prenatal at any point in time ontraction monitor; %O.( ial care More 1b/52 is 41t1) servke referral

F. Stanley Social support via home Anv tI the tollinsing: Computer-generated random Preterm delivery; birihweight: C ost: ansietv; R. Bryce s tsning bY midwives ores mus preterm birth; numbers in blot ks of 4 sot ial support; focus ofontrol; ondut t of

Perth, Australia ISSI perinatal death; spontaneous . labor mid-trimester abortion; > brst trimester abortions; APH, low birthwerght (s ..:72000)

J. lar !tome visits by trained Women attending Random stratified by center IUGR, LBW1, preterm debverv; maternal E. Kest ler women (including nurses, specified antenatal clinic weight gam: obstetric interventions, labor Latin Ament a sot sal KV rkers, and lay before 20152, without t ()wilt awns, breasneeding. neonatal (Argentina, Cuba w(nker%) to provide personal important t fink al problems, morbidity; use of health and soc al services: Guatemala, Mesir and ramify support health but haying I or more risk satisfaction; support; portpartum depression education, and use I), stal fat tors: previous tow and health services birthweight birth; age 17 malnutrition: low income, maternal educ anon, lack ot support is = 2000)

. . Citobel Psvc hosot (al supptotsotial High-risk women in S Random assignment or Gestational age at deliver% hirthweight R. Bemis work to reduce stress espenmental c linKS c fink s to ever i talent.) Los Angeles, O.S.A, 125.21i1 or ontrof status, then random assignment of high. risk women within esperimental clinics to 1 of interventions--1 cif whic h is support

't4

220 Statistical Findings: An Overview of Thals of Social Support During Pregnancy Interventions

Figure 18.1: Effect of Social Support inPregnancy (as An Explicit Co-Intervention) on Preterm Delivery Rate (< 37 Weeks)

Log Odds Rat(o Graph of Log Odds Ratios Study Experiment Control & Confidence Interval (%) (%) (95% Confidence Interval)

,01 .1 .5 1 2 10

Blonde), 8. (unpublished) 14779 07.72) 11/73 (15.07) 1,21 (0.51-2.85) 0.8fr Sexton, M.el al. (1984) 34/429 (7.931 4(/438 (9.13) 83.5.1-1.38)

Mourne, D. R. et al. 0987) 10/140 t7,1 4) 9/127 (7 09) 1 01 (0.40-2.56) 0.98 Flint, C. et al. t1987) 31/488 (6,35) 11/479 (6.47) t0.59-1.64)

Lovell, A. et al. (198b) 5N5 (5.26i 4/102 0.92) 1 .3b (0.3o-5.16) 1.43 Spira, N. et al. (1981) 45/458 (9.83) 30/425 (7.06i (0.89-2.29)

181279 it) 45) 0.8; Donovan, I. 11977) 1 252 t5,511) (0.42-1.75)

1 DO Typical Odds Ratio (0 al-1 34) OS"), Confidence Interval)

Figure 18.2: Effect of Social Support inPregnancy (as a Primary Intervention) on Preterm Delivery Rate (< 37 Weeks)

Log Ocids Ratio 0,10to Log Odds Ratios Study Eperimvnt Control & Confident e interval ("el (93% Confident e Interval)

,01 .1 .5 1 2 10 1,04 Oakley. A. tunpubtichedt 34/243 0 3,991 33/243 0 3.581 (0.b2-1.75) 1.08 Spencer, B. et al. (198b1 60/602 (9.971 54/581 i9.29i 10.71-1.59) ----- 0,77 Olds, D. L. et a). (1986) 11/1nn (h.63) 12/142 18,45) al. 33- 1 80i 1 30 Oame, I. iunpublishetil 3/25 )20,00) 4!2 I tr.001 10.31-5,441 1.03 Typical Odds Ratio 10.78-1. 17, i95°.;, Confidence Interval!

Figure 18.3: Effect of Social Support inPregnancy (as an Explicit Co-Intervention) on Extremely Preterm Delivery Rate (< 33Weeks)

Control Log Odds Rano Graph or Log Odds Ratios Study Lsperiment ("o) (93"., Contitienc e Interval; is. Confident e Intenal

.01 1 5 1 2 10 0.91 Mourne, (), R. et al. (1987 t 3/140 2.14) I/ 27 (2.36r t0.18-4.56) 0.91 Typical Odch Ratio (0,18-4.56t (95% Confidence interval,

Diana Mourne and Ann Oakky 205 221 Advances in the Prevention of Low Birthweight

Figure 18.4: Effect of Social Support in Pregnancy (asa Primary Intervention) on Extremely Preterm Delivery Rate (< 33 Weeks)

Study Experiment Control Log Odds Ratio Graph of Log Odds Ratios (%) (%) 195% Confidence interval) & Confidence interval

.01 ,5 1 2 10 Oakley, A. (unpublished) 13/243 (5.35) 13/243 (5.35) 1 00 (0.45-2.20) Spencer, EL et al. (1986) 9/603 (1.49) 11/581 (1.89) 0.79 (0.32-1.90) Dance, J, (unpublished) 2/25 (8.00) 1/25 (4.00) 2.00 (0.20-20.20) Typical Odds Ratio 0.91 (95% Confidence interva)) (0.1(4-4.56)

Figure 18.5: Effect of Social Support in Pregnancy (asan Explicit Co-Intervention) on Low Birthweight Rate (< 2500 g)

Study Experiment Control Log Odds Ratio Graph of Log Odds Ratios (%) (%) (95% Confidence lruerval) & Confidence interval

.01 .1 .5 1 2 10 Blonde), B. (unpubhshed) 9/79 (11.39) 6/73 4822) 1,41 (0.49-4.13) __. Sexton, M.1. et al. (19134) 29/429 (6.76) 39/435 (8.97) 0,74 (0.45-1.21) Elbourne, D. R, et a) (1987) 11/143 17.691 12/130 (9.23) 0 82

_ (0,35-1.92) Flint, C. et al, (1987) 31/478 (6.49) 18/471 (8.07) 0.79 (0.48-1.29) Lovell, A. et al. (1986) 5/95 (5.26) 11/102 (10.78) 0.48 10.17-1 131 Reid, M. E. et al. (19831 6/76 (7.89) 10/79 (12.661 0.60 (0 .21- 1.68) """ Spira, N. et A. (1 981) 51/458 (11.14) 40/425 19.4n 1,20 (0.78-1.86) A _ ___ . Donovan, .1, W. 0 9771 27/263 (10.27) 261289 0.00) 1.16 (0.66 2,04) 1,1444. Zimmerman-Tansella, C. et al. 1/14 (7.14) 1/20 (5.(0) 1.46 APA,P11.0.1 DATE? (0.08- 25.42) Runnerstrom, L. (1969) 61/768 (7,94) 121/1005 (12.04) 0.64 0.47-0.87) Typical Odds Ratio 0.81 (95% Confidence interval) 10.68-0.97) Typical Odds Ratio minus Runnerstrom 0.91 (95% Confidence interval) t 0.74 1.141

222 Statistical Findings: An Overview of Trials qf Social Support DuringPregnancy Interventions.

Figure 18.6: Effect of Social Support in Pregnancy (as a Primary Intervention) on Low Birthweight Rate (< 2500 g)

Study EAperiment Control Log Odds Ratio Graph of tog Odds Ratios i'!.) 195% Confidence Interval) tig Confidence Interval

01 .1 .5 1 2 10 52f243 (21.40) 0.84 Oakley, A. (unpubl(shed) 45/243 118.52) .."641 (0.54-1.30)

Heins, a C. et al. (unpublished) 118/633 (18.64) 137/642 (21.34) 0.84 (0.64-1.11)

sPencer. a. et al. 0986) 54/602 18.971 50/581 (8.61) 1 05 (0.70-1.561 ads, a L. et al. (1986) 101166 16.02) 4/142 12.821 2.09 10.71-6.11) 0.72 Dance, ). tunpubfiYhed) 3/25 0 2.00) 4125 (16.00) : /....4.4,...4, ,i.,,... (0.15-3.51)

Typical Odck Ratio 0.91 10.75-1.111 1

Figure 18.7: Effect of Social Support in Pregnancy (as an Explicit Co-Intervention) on Very Low Birthweight Rate (< 1500 g)

Study EAperiment Control Log Odds Ratio Craph of Log (kids Ratios

A (93% Confidence interval) ey Confidence interval

01 .5 1 2 10

Reid, M. E et .31.1198 it 2/76 (2.63) 1/79 (1.27) 2.04 (0.21-19.95) Sexton, M. I. et al. (1987) 8/429 11.8bi 5/436 i1 15) --.1 62 (0.54-4.85)

Elbourne, D. R. et al. i19871 .3/143 t2.1W 2/110 1.16 (0.23-7 991

Typical Odds Ratio .61 Contidence interval( (0.68 1 811

Figure 18.8: Effect of Social Support in Pregnancy (as a PrimaryIntervention) on Very Low Birthweight Rate (< 1500 g)

Graph ot I og Mds Ratios Study k %penmen! Control Log Odds Ratio i"o (9c", Confidence Interval( & C orthdenc v Inter

01 1 .5 1 2 10

Hein, Ff. C. et .0. (unpublohedi26,3J 1.3.4b1 i0/1142 (4.b7; 0.74 (0.42- 1.28(

Oakley, A. tunpublished) 4/243 11 .63, 9/243 , 1.70( 0.45 4 (0.15-1. 0.80 Spencer, 8. et al. 11986) 5/603 i0.81) 6'581 (1.01, 41 (0.24- 2.631

_ . . TyPi1:11 Odds Ratio 0 69 1 0.43 1 O8) 4

223 Diana Elbourne and Ann Oakky 207 Inhibition of preterrn Labor: Is It Worthwhile?

MARc J.N.C. KEIRSE, M.D.

INTRODUCTION For a normally formedinfant, there is no greater risk than tobe born too early in pregnancy. The transitionfrom fetus to newborn is a hurdle tosurvival that is more than 100 times higher for the preterm infant than it isfor the infant born at term." Yet, there is alack of understanding of the mechanisms that initiate labor too early in pregnancy. andthere are large deficiencies in our ability toforetell and timely recognize them. Failure tounderstand and prevent preterm labor has madeclinicians turn their attention to meansand methods that could stop it in time to prevent its main consequence, pretermdelivery. At first glance, they appear to havebeen extremely successful in doing so. Most of the approachesintroduced in the past few decades haveshown such high success rates inthe initial reports documenting

20S Advances in the Prevention of Low Birthweight their use that one would expect the problem of of the mortality associated with preterm preterm labor to have been resolved many None of that mortality would he affected by years ago (see table 19.1). abolishing preterm delivery, and none of these This, however, has not happened. Nor preterm labors are worth inhibiting. have all of these treatments disappeared from What applies to mortality also applies to the obstetric armentarium On the contrary, morbidity. This is emphasized by the fact that, some of them have been supplemented with currently, about one-third of preterm deliveries other treatments in the hope that two unvali- in singleton pregnancies do not result from dated treatments will be better than one. Thus, spontaneous preterm labor. They result from the term tocolysis, introduced by Mosler in deliberate obstetric intervention to end pregnan- Germany in 1964 and now generally accepted cy because the obstetrician believes that the as synonymous for inhibition of labor, gave rise risl.s of a continuing pregnancy to mother or to the term Zusatztokolysis to refer to the vari- fetus have become unacceptably high." The fact ety of treatments that are added to either aug- that approximately half of all preterm births are ment the effects or counteract the side effects associated either with twin pregnancies (10%), of tocolysis. In addition, new drug treatments with fetuses who died before labor or have continue to capture the imagination of clini- lethal malformations (10-15%), or with elective cians desiring to solve the issue of preterm obstetric intervention (25-30%) is rarely empha- labor once and for all. The introduction of sized in the literature.° This may, in part, explain these new treatments also continues to follow the failure of many preterm prevention pro- the traditional pattern in which enthusiasm for grams to lower the (crude) incidence of pre- the treatment seems to be at least as com- term birth. It certainly emphasizes that, for about pelling as evidence that it works. 40 percent of preterm deliveries, inhibition of labor is not only superfluous, but harmful. Also readily forgotten is the fact that spontaneous preterm labor and delivery are often WHAT MAY OR MAY NOT BE WORTH INHIBITING? associated with clinically significant maternal Preterm delivery and preterrn labor and fetal pathology. In comparison to birth at There is a widely held misconception that term, preterm birth is associated with a much resolving the problem of preterm labor will higher incidence of inadequate fetal growth,' resolve all of the problems associated with prelabor rupture of the membranes,8 placenta preterm birth. This is not so. Studies which praevia,9 placental abruption,4 fetal congenital have examined causes of perinatal mortalityLs malformations,' and severe maternal disease."" have shown that more than 50 percent of that Studies which have specifically addressed the mortality is due to fetal death before the onset incidence of maternal and fetal pathology in of labor, or to lethal congenital malformations, spontaneous preterm birth' have indicated that or both. These infants account for 12-15 per- about half of all singleton preterm births that cent of all preterm births, but for 60-65 percent result from spontaneous preterm labor are asso-

226 Inhibition of Preterm Labor Interventions dated with such pathology. Often that patholo- and without specific treatment. Several authors gy is the very same pathologywhich, in the have attempted to improve upon the classical absence of preterm labor, leads the obstetrician diagnostic criteria by measurement of hormone to artificially end pregnancy, either byinduction levels," assessment of fetal breathing move- of labor or by elective cesarean section.' ments,".36 measurement of prostaglandin All of this suggests that, in many instances, metabolites,'' or assessment of thromboxane nothing can be gained, but a great deal can be excretion.'m Some have attempted to forestall lost by inhibition of preterm labor and delivery. the problem by serial cervical examinations in pregnancy,° by antenatal monitoring of uter- Preterm labor and preterm contractions ine contractions,2'-24 or by screening for the risk Judging whether labor has started or not is of preterm delivery." Some of these approaches by no means easy. Predicting whether it will or have resulted in the administration of labor- will not lead to delivery can be extremely diffi- inhibiting drugs to more than 40 percent of cult, as witnessed by the high "success" rates of low-risk pregnant women,26 demonstrating that placebo "treatments" in stopping preterm the best results of treatment are obtained when labor." Lack of precision in the diagnosis of there is nothing to be treated. labor would create no problems if one could It is fair to conclude that, despite all of await the signs and symptoms that invariably these attempts, the diagnosis of preterm labor clarify the diagnosis. Signs of steady progress in has remained as problematic as ever. This has descent and dilatation may be useful to prepare immediate consequences for the inhibition of for delivery. They are not very helpful, howev- preterm labor and for assessing the effectsof er, if delivery needs to beaverted and if labor tocolytic treatments. In many instances, appar- itself is undesirable and needs to be stopped. ently progressive preterm labor will stop irre- The more advanced labor is, the more difficult spective of whether or not any treatment is it is to stop. On the other hand, the less instituted. The finding that uterine contractility advanced it is, the more likely it is to stop of its is suppressed does not necessarily mean that own accord without any tocolytic treatment. treatment has been effective, nor does it neces- Successful inhibition of preterm labor, sarily mean that delivery will be postponed to therefore, depends on early and accurate diag- an extent that is clinicallyuseful. nosis. Early diagnosis, however, particularly Numerous criteria have been applied to diagnosis based on uterine contractions, is describe "successes" of one treatment or anoth- notoriously inaccurate for predicting whether er (see table 19.1). Someof the most common- preterm delivery will occur or not. Some2have ly used treatments relate to temporary arrest of even suggested that this diagnosis maybe erro- uterine contractions, number of hours or days neous about 80 percent of the time. We recent- gained before delivery, number of deliveries ly found that 31 percent of women who caine delayed until 36 or 37 weeks' gestation, num- to the delivery unit with pretermuterine con- ber of infants weighing more than 2500 g at tractions could safely return home undelivered birth, and increases in mean gestational age or

Marc Keirse 227 210 Advances in the Prevention of Low Birthuvight

mean birthweight at delivery. None of these erogeneous mixture. This is exemplified in table outcomes are of great relevance to mother and 19.2, which summarizes some of the data on all baby unless they are accompanied by an infants with a gestational age of less than 32 increase in the number of survivors or by an weeks who were born alive in the Netherlands increase in the quality of life for the infants and in 1983.4' Even in this category (0.6% of all their mothers. births in the country in 1983), which would seem to be the one most likely to benefit from Early and late preterm gestations inhibition of preterm labor, 15 percent of births Preterm is internationally defined as less (excluding stillbirths) resulted from deliberate than 37 completed weeks (259 days) of gesta- obstetric intervention to end pregnancy and tioni-29 but inhibition of labor is not equally another 44 percent followed prelabor rupture of useful at all gestational ages that are character- the membranes (see table 19.2). ized as being preterm. Cut-off points drawn for Thus, the following statements appear to demographic and public health reasons, particu- be true. First, the majority of preterm deliveries larly when they require international consensus, occur at gestational ages that are advanced have a tendency to follow the reality of clinical enough to offer little potential for improving practice at a safe distance rather than to light infant outcome by prolonging the pregnancy. the way. This has been emphasized in the past Second, even at gestational ages at which the by the expression prematurity, originally infant might theoretically benefit from prolon- defined as a birthweight of 2500 g or less30 and gation of pregnancy, many other factors need later defined as a gestational age of less than 37 to be taken into account if that theoretical ben- weeks,m which was abolished and replaced by efit is to be matched by a real benefit. If deliv- preterm-"A2 long after it had become obvious ery can be postponed or pregnancy prolonged that "born too early" does not mean the same as for a duration which is thought to be clinically "born too small" and that two different defini- significant, it does not necessarily follow that tions of the same word are not very helpful in the outcome for the infant will be improved. In distinguishing these two situations. addition, drug treatments that are powerful The incidence of preterm delivery increas- enough to suppress preterm labor effectively es with increasing gestational age up to the are bound to have other effects on the mother cut-off point at 37 weeks. From the limited or the baby, some of them undesirable, that data that are available on geographically must be taken into account. defined populations, less than one-quarter of The conclusion of all of this should be all preterm deliveries occur below 32 weeks' clear. Emphasizing the uterine relaxant effects gestation.'"" of tocolytic treatments, without considering Itis particularly these deliveries, generally what is gained in terms of quantity and quality referred to as being very preterm, that present of survival and without considering the possi- the greatest challenge. It should be realized, ble hazards to mother and baby, is needlessly however, that these births also form a very het- naive.

228 Inhibition qf Preterm Labor 211 Interventions

Assessment of the effects of tocolyticdrugs Betamimetic drugs A large number of tocolytic agentshave Betamimetic agents are used more exten- been employed for the inhibition of preterm sively than any of the other approaches that are labor. Not all of these are stillrelevant. Table employed to achieve tocolysis in preterm 19.3 lists those that were reported tobe in use labor.""7 Since the first reports in 1961;4°9 a in the literature of the past 10 years.Some of great variety of betamimeticdrugs have been these, such as antibiotic treatmene9and oxy- applied to the inhibition of preterm labor. They tocin analogues,4° have been recentintroduc- have included, among others, isoxsuprine,44." tions that will require validationbefore being orciprenaline,"° mesuprine,"ritodrine ,62 introduced into clinical practice."' Others,such fenoterol," salbutamol,"4 buphenine, hexopre- as ethanol, havebeen used for many years," naline,"" and terbutaline."' but are now mainly of historical interest aswit- There is a wealth of literature on thephar- nessed by the results of questionnaire surveys macological effects of betamimetic drugs.They conducted among obstetricians in Europe"''' are all chemicallyand pharmacologically relat- and Australia."' Still others, such as magnesium ed tothe catecholamines, epinephrine sulfate24.4'.4" and diazoxide,49 have beenreported (adrenaline) and norepinephrine(nora- to be widely used in somecenters in the drenaline), and all stimulate the 13- receptors United States, although there are nocontrolled that are present in the uterus and inother trials which have shown them tobe more organs throughoutthe body."8 Many efforts effective than placebo for inhibiting preterm have been directed at developing agentsthat labor. Admittedly, magnesium sulfatehas been would selectively stimulate the S-2 receptorsin compared with other drug treatments infour the uterus. These efforts have onlypartially controlled trials."--" Neither of thesehas, on been successful, and all betamimetic agents balance,'' invalidated the opinion expressed available also stimulate to some extentS-1 even by proponentsof this tocolytic treatment receptors.Beta-2 selectivity currently only that, "once labor has broughtabout cervical means that interactionwith 1-2 receptors dilatation, the drug is reasonablyineffective."4- occurs at lower agonistconcentrations than In the light of present dayevidence, only interaction with 1-1 receptors. Stimulationof 13- two categories of drugsmerit consideration in 1 receptors is responsible for actionssuch as an terms of their potentialbenefits for inhibiting increase in heart rate and strokevolume, relax- preterm labor. These arebetamimetic agents and ation of intestinal smooth muscle,and lipolysis. inhibitors of prostaglandin synthesis.Calcium Beta-2 stimulation mediates glycogenolysisand antagonists, although comprising awide catego- relaxation of smooth muscle in thearterioles, ry of differentchemical compounds with inter- the bronchi, and the uterus. esting properties, have been sopoorly evaluated for the inhibition of preterm labor493."'that they Placebo controlled trials cannot, as yet, beconsidered to have any clini- Only three of the many betamimetic agents cal value for this indication. available (isoxsuprine, ritodrine,and terbu-

229 Marc Keirse 212 Advances in the Prevention of Low Bird:weight

Wine) have ever been compared witha control women treated with placebo or some standard group, who received either no active treatment treatment. All but one" of these trials involved or placebo, for inhibition of preterm labor. the use of oral betamimetic maintenance thera- Some of the drugs (such as salbutamolor py after acute tocolysis had been achieved. The fenoterol) that were reported to be used by large majority of the trials (12 of the 16) dealt one-third of the obstetricians in Belgium," the with ritodrine, 3 tested terbutaline, and 1, the Netherlands," and the United Kingdom" have oldest trial, dealt with isoxsuprine. Allocation to never been tested against placebo or no treat- the treatment or control groupwas stated to ment in preterm labor. Yet, they entered obstet- have been blind in 11 of the 16 trials. In the rical practice at about the same time"as other others, the method of allocationwas either not drugs, such as ritodrinetu or terbutaliner'" describeduor was such that the investigators which have been subjected to several placebo may have known in advance to which group controlled trials. the woman would be allocated.""" For 12 of the Seventeen studies have been published in 16 trials, data were available on allwomen which one or another betamimetic agent was entered. That information had often not been reported to have been compared againstno published in the original reports, but King et labor-inhibiting drugs in preterm labor.1"2°""'"" al." succeeded in obtaining additional data Five of these 17 studies'2"4.".' were either con- from the authors in order to minimize the intro- ducted or reported in a manner that precludes duction of bias after trial entry.'" unbiased evaluation of the treatment given. King and his colleagues" conducted a for- Penney and Daniell"' reported on the use of a mal meta-analysis of these 16 trials, using meth- prolongation index for their trial, but provided od.s described by Peto et al;"' and Yusuf et al." no data on outcome by treatment allocation. This approach, which is extensively discussed by The reasons why the four other trial reports Chalmers et al.,'" is based on calculating a ratio were considered to introduce too much bias in (odds ratio) for each trial between the odds ofa evaluating the effects of betamimetic treatment particular outcome, among women allocated to have been elaborated by King and his col- betamimetic drug treatment, and the odds of the leagues." Data on four unpublished ritodrine same outcome, among women allocated to the trials,conducted by Barden,'" Hobel," control group. The sum of the differences Mariona,* and Scommegna and Bieniarzr and between active treatment and control derived included in the report of Merkatz et al.- could from the independent trials, used in combination be obtained by courtesy of the investigators with the sum of the variances of these differ- and the company (Duphar) that manufactures ences, is then used to calculate a typical odds the drug." The characteristics of the resulting ratio. The latter provides a more sensitive esti- total of 16 (published and unpublished) trials mate of the effect of treatment than can be ob- with adequate data have been described by tained from the individual, and often small trials. King et al." These 16 trials involved a total of This is illustrated in figure 19.1, which 484 betamimetic-treated women and 406 shows the effect of betamimetic drug treatment

230 2 I.a Inhibition of Preterm Labor Interventions

on the incidence of deliverywithin 24 hours of Overall, the data from the controlled com- trial entry in the 14 studies that provided infor- parisons of betamimetic drug treatment in mation on this outcome. For each of thetrials, preterm labor thus convincingly demonstrate the figure shows the ratio between the odds in that the treatment results in a significant delay the betamimetic treated group and in the con- of delivery and in a lower incidence of preterm trol group, with its 95 confidence interval. An birth than observed without such treatment odds ratio of 1 indicates that the tiutcome (see figure 19.2). As mentioned earlier, in all occurs with the samefrequency in the treat- hut one of the trials,-' acute tocolysis was fol- ment as in the control group. Anodds ratio of lowed by oral maintenance treatment, and it is less than 1 indicates that fewer women in the possible that this component of the betamimet- treatment group delivered within 24hours than ic drug administration may havecontributed to in the control group, while an odds ratioabove the overall gain in gestational age. The three 1 indicates the reverse. A 95 percentconfi- placebo controlled trials that have been report- dence interval that crosses the vertical line, ed on oral maintenance treatment, after an drawn at an odds ratio of 1, indicates that the acute episode of preterm labor hadbeen over- difference between the treatment and control come,'''-"4 showed that oral maintenance treat- groups is not statisticallysignificant. A confi- ment will to some extent prevent a recurrence dence interval that does not cross the vertical of preterm labor.4' line, drawn at an odds ratio of 1, indicates a The data on the incidence of low birth- statistical significance at the 5 percent level or weight also suggest a beneficial effect of less. The figure illustrates that there werefewer betamimetic drug treatment on this outcome, deliveries within the first 24 hours in the group although this did not reach conventional levels allocated to betamimetic treatment than in the of statistical significance, with a 95 percent confi- control group in 12 of the 14 studies, butthat dence interval from 0.55 to 1.02 (see figure 19.2). this reached statistical significance in only6 of The effects on delay of delivery, gestation- them. The typical odds ratio acrosstrials, al duration, and birthweight did not, however, however, was statistically highly significant, as result in a detectable decrease in the incidence illustrated in figure 19.2. of more serious infant outcomes. The incidence Figure 19.2 shows the typical odds ratios of perinatal death not attributable to lethal con- (across trials) with their 95 percentconfidence genital malformations and that of severe respi- intervals for the incidences of deliverywithin ratory disorders, includingrespiratory distress 24 hours of trial entry, deliverywithin 48 hours syndrome, was similar in the treatment and of trial entry, delivery before 37 weeks of gesta- control groups. The typical odds ratio for respi- tion, delivery of a low birthweightinfant, respiratory distress syndrome and severerespiratory ratory distress and/or respiratorydisease, and disorders in the newborn across the 12 trials, perinatal death not due to lethalcongenital which provided information on that outcome, malformations. Further details on all of these was 1.07, with a confidenceinterval from 0.71 outcomes have beendescribed by King et al." to 1.61 (see figure 19.2). Datafor the individual

Marc Keirse 231 214 Advances in the Prevention of Low Birthweight trials are illustrated in figure 19.3. The typical teroids had been given before delivery, odds ratio across the 15 trials, which provided although this treatment is known to reduce data on perinatal death, was 0.95, with a confi- perinatal mortality and morbidity."' dence interval from 0.55 to 1.67 (see figure 19.2). These results are compatible with an Controlled comparisons with other drugs impressive reduction in perinatal mortality of Eight reports on controlled comparisons about 40 percent, but they are equally compati- between betamimetic agents and ethanol for ble with an increase in perinatal mortality of the inhibition of preterm labor have appeared about the same magnitude. The lack of statisti- in the literature.")197-h"2 Castren et al.' reported cal power in the individual trials is amply illus- a study in which every other patient was allo- trated in figure 19.4. cated to buphenine and the others to ethanol, The apparent lack of effect of betamimetic but this alternate allocation resulted in 43 drug treatment on the serious adverse out- buphenine- and 50 ethanol-treated women. comes of mortality and respiratory morbidity This study was excluded from meta-analysis, as may be due to a number of factors. The trials it was too likely to introduce bias into the com- may have included too large a proportion of parison. The report of Merkatz et al." contained women in whom postponement of delivery and no usable data on the 153 women who had prolongation of pregnancy were unlikely to been randomized to ethanol or ritodrine, but confer any further benefit to the baby. A treat- 150 of these women had been included in a ment that is effective in stopping preterm labor previous report by Lauersen et al.,"" and some at 36 weeks may well reduce the likelihood of of these had also been reported on by Fuchs:* delivery within 24 hours, of delivery before 37 The characteristics of the trials that were weeks, and of delivering a low hirthweight less unlikely to have introduced bias into the infant; but at that gestational age it will have lit- comparison and the betamimetic agent used in tle potential for reducing perinatal mortality or these trials are summarized in table 4. Although serious morbidity. The lack of effect on infant the three largest trials all excluded some mortality and respiratory morbidity may also be women after randomization, no attempts were due to adverse effects of the drug treatment. made to obtain unpublished data from the These need not be direct adverse effects, but authors. Widely different cut-off points have may also result from prolongation of pregnancy been used in these trials for describing delay of when this is contrary to the best interest of the delivery, but four reports"-'"2 provided infor- baby, as may be the case with clinically unrec- mation on the number of deliveries within two ognized placental abruption, severe pregnancy- or three days after treatment allocation, In 2 induced hypertension or intrauterine growth reports this related to delivery within 48 retardat;on. It is also possible that too little use hours;'"'" in 1it related to delivery within 3 was made of the time gained by postponing days;'" and in the other, to an insufficient delay delivery. Only four of the trial reports, for of delivery for a full 36 hours course of cord- instance, indicated whether or not corticos- costeroids to be given,9- These data have been

232 215 Inhibition of Preterm Labor Interventions combined in figure 19.5. The typical odds ratio Unwanted effects of 0.38, with a confidence interval from 0.23 to A number of subjective maternal symp- 0.64, convincingly shows that betamimetic toms may occur as a result of betamimetic- agents are superior toethanol for delaying induced changes in many bodily functions. The delivery. most frequently observed arepalpitations, Two controlled comparisons have been tremors, nausea, and vomiting. Headache, reported between magnesium sulphate and a vague uneasiness, thirst, nervousness,and rest- betamimetic agent, either ritodrine" or terhu- lessness may also occur. Chest discomfort and taline." One of these only reported on the inci- shortness of breath should alert to the possibili- dence of hypothermia with both treatments." In ty of pulmonary congestion. the other report, the method of randomization From the placebo controlled trials dis- was not mentioned andoral ritodrine was used cussed above, there is little evidence that for maintenance treatment in both groups." betamimetic drug treatment frequently poses Several controlled comparisons have been great hazards to either mother orbaby. This is conducted between different betamimetic not necessarily convincing in viewof the fact drugs.'"3-"" The interpretation of such studies is that all 16 trials together contained less than difficult, however, as the authors of the studies 500 betamimetic-treated women. This is proba- themselves often recognize. None of the stud- bly less than one percent of the number of ies has been large enough to have had a women who annually receive treatmentwith chance of detecting or excluding important one of these agents becausethey are at risk of differences in the outcomes that really matter, delivering preterm. The likelihood that rare but such as infant mortality and morbidity. Nor did serious adverse effects of betamimetic drugs, if the trials show that any one of the drugs test- they exist, would have been uncovered by any ed was remarkably free of maternal side one of these trials must beinfinitely small. effects. When differences in the incidence of Other data in the literature, however, indicate maternal side effects or in average gain in ges- that these drugs are not entirely harmless. tational age were detected, itis never entirely In the late 1970s in West Germany, inci- clear whether the drugs were used in equipo- dents of pulmonary edema, congestive cardiac tent doses. The widedifferences among failure, and even death started to be noticed betamimetic drugs in their ratio between drug in young women who had received the weight and drug effect make this nearly betamimetic drug, fenoterol, in combination impossible to assess. Without a dealer delin- with corticosteroids for preterm labor.""-"2 eation of both benefit and harm, such small Within a year, a case of severe pulmonary differences are difficult to interpret, irrespec- edema following administration of terbutaline tive of whether or not they reach statistical sig- and dexarnethasone was reported from the nificance. None of the studies thus far has United States,-4' and by 1980, a large number investigated the differential in the costs of of such reports had appeared in the medical treatment between one drug andanother. literature." 4. 1"

Marc keirse 216 233 Advances in the Prevention of Low Birthweight

Initially, much of the blame for this dread- pressures. Nevertheless, liberal administration of ful complication was directed at conicosteroid crystalloid fluids became standard practice, with administration, which had been the more as much as 400 mrm' to 1,000 of intravenous recently introducedof the two drug treat- fluids often being administered routinely for 30 ments, or at a cumulative and potentiating minutes even before starting infusion with effect of corticosteroids and betamimetic drugs. betamimetics. Apparently, this practice has As more of these reports became available, pul- caused more harm than it prevented. monary edema was recognized to be a compli- The frequency with which pulmonary cation not only of the combination of edema develops during betamimetic drug betamimetics with corticosteroids, but also of administration is difficult to estimate. Katz et betamimetics used alone. observed clinical signs and symptoms of Many of the cases described have been pulmonary edema in 5 percent of 160 women secondary to fluid overload, and no instance of treated with terbutaline for preterm labor, but pulmonary edema has been reported in women half of the women with this complication had receiving betamimetics orally. Fluid overload twin pregnancies. Multiple pregnancy, as well during betamimetic treatment can occur by two as underlying heart disease and the use of cor- mechanisms: (I) By too vigorous administration ticosteroids or multiple drugs in addition to the of intravenous fluids; and (2) by decreased betamimetic agents, is known to increase the renal excretion of sodium, potassium, and risk of pulmonary edema. If 5 percent were a water as a direct result of high doses of reasonable approximation of the frequency of betamimetics. The antidiuretic effect of this complication, it would have been observed betamimetics is most pronounced in the first 48 and described much earlier than in the late hours of tocolytic treatment, and that is when 1970s, and it would have been noted in several most cases of pulmonary edema are observed. of the women who participated in the placebo In experimental animals, the development of controlled trials. Itis possible that this compli- intravascular hypervolaemia during betamimetic cation occurs more frequently with terbutaline administration has been directly correlated with than with other betamimetic drugs, as suggest- increasing doses of the betamimetics and with ed by Robertson et al."' Whether or not this is increasing rates of crystalloid infusion.'2' Both true and whether or not it relates specifically to of these concur when betamimetic drugs are this agent or to the way in which it is adminis- administered in dilute solutions; the higher the tered is not clear:" 12- dose, the larger the amount of fluid. Myocardial ischaemia has been described Excessive hydration has long been prac- as the other main, but rare complication of ticed, apparently to combat the risk of hypoten- betamimetic drug treatment.''' )44 This compli- sion,us which is rarely a problem since the cation is a separate entity from that of pul- betamimetic influence on blood pressure mainly monary edema. Diffuse micronecrosis in the consists of a widening pulse pressure, due to an myocardium has been known since 1959, when increase in systolic and a decrease in diastolic it was induced in the rat by administration of

234 21; Inhibition of Preterm Labor Intetventions isiproterenol.L"Similar lesions have been unwanted effects of the betamimetic drugs. observed in the myocardium after betamimetic- Some have involved the use of different infu- related death. Unlike genuine myocardial sion fluids,"" but it is clear that the amount of infarcts, micronecrosis is not a direct result of fluid is far more important than the type of hypoxia. It relates to the increasing energy and fluid. Others have centered on the use of calci- oxygen requirements of thebeta-stimulated um antagonists or 1-1 receptorblockers to myocardial cell; if demand exceeds supply, reduce the systemic, and especially the cardio- ischaemia develops. vascular, effects of betamimetic agents. To be Betamimetic drug administration results in of value, such combination treatments should a marked increase in cardiac output in preg- be demonstrated to achieve at least one of the nancy, which is roughly of the sameorder as following three aims. First, infant outcome in that observed during moderate exercise.9s terms of mortality and serious morbidityshould This increase is attributed to the combination of be better with/ qe combined treatment than an increase in heart rate and adecrease in with the singl treatment. Second, the addition peripheral vascular resistance due to relaxation of these drugs should decrease the incidence of of vascular smooth muscle. In late pregnancy, seriousmaternal complications during cardiac output is already 40 percent above betamimetic drug treatment. Third, the com- prepregnancy values, and the increase is even bined treatment should significantly decrease larger in twin pregnancies." The additional the incidence of less serious, but troublesome, work imposed on the myocardium by maternal side effects. Thus far, there are no betamimetic drug treatment may thus become indications that any one of these goals has too much for women with preexisting, overt, or been adequately met. hidden cardiac disease.s4) These women should not be given betamimetics, as thehazards for inhibitors ofprostaglandin synthesis them are likely to be greater than any benefit Thereissubstantial evidencethat that could be derived for their infants. For the prostaglandins are of critical importance in the same reason, itis wise to insist on a normal initiation and maintenance of human labor."' electrocardiogram before betamimetics are Suppression of endogenous prostaglandin syn- administered. The likelihood of finding an thesis is therefore a logical approach to the abnormal electrocardiogram in a normal preg- inhibition of preterm labor. Several agents with nant woman without symptoms or suggestive widely different chemical structures and phar- history must he small, however. Nor can it be macokinetic properties'42 inhibit prostaglandin implied that a normal electrocardiogram before synthesis. They are sometimes referred to as treatment will protect against subsequentdevel- prostaglandin synthetase inhibitors. Since there opment of pulmonary edeina.'2'' is no enzyme of this name, they arebetter A few trials have addressed the question as referred to as inhibitors of prostaglandin syn- to whether the combinationof betamimetics thesis. The inhibitors that have been used to with other treatments could reduce some of the treat preterm labor include naproxen,'"flufe-

Marc Keirse 235 21S Advances in the Prevention of Low Birthweight

namic acid,'" and acetyl salicylate,14"46 but the Controlled comparisons most widely used has been indomethacin.'" Only two studies have been reported that All of these drugs act by inhibiting the purported to compare a prostaglandin synthesis activity of prostaglandin endoperoxide syn- inhibitor with placebo in preterm lahor.'"1" thase, an enzyme also known as cyclooxyge- Both used indomethacin as the active treat- zase. This enzyme converts fatty acids, ment, and both were stated to have been con- arachidonic acid in particular, into prosta- ducted in a double-blind manner. Neither of glandin endoperoxides. Itis present in high them was entirely placebo controlled, however, concentrations in the myometrium of pregnant since a number of women in whom treatment women,"8 but is found throughout the body was considered to have failed received other both in and outside pregnancy. Inhibition of tocolytic drugs. In addition, three controlled that enzyme doesnot only suppress studies have been conducted in which a prostaglandin synthesis. It also suppresses the prostaglandin synthesis inhibitor was either formation of prostacyclin and thromboxane A2, added or not added to treatment with other both of which may have a number of impor- labor-inhibiting drugs. All of these trials also tant, though largely unknown functions in used indomethacin, while the other drug treat- pregnancy.'"" Inhibition of the enzyme is not ment used ethanol in one trial" and the always achieved in the same way. Aspirin, for betamimetic agent, ritodrine, in the other example, causes an irreversible inhibition of two.'41" Eight reports were available on this the enzyme, whereas indomethacin results in a total of five trials.t22 " competitive and reversible inhibition. This is There are reservations al)out the heteroge- because aspirin acetylates the enzyme, and neous nature of these trials and about potential therebyincapacitates it permanently. bias in many of them. On the whole, these tri- Indomethacin, on the other hand, competes als are of inferior quality compared with those with arachidonic acid for utilization by the of the placebo controlled trials of betamimetic enzyme; it leaves the enzyme itself intact, and, agents. Nevertheless, the data that could reli- when indomethacin levels decrease, the ably be extracted from all of the3e reports have enzyme can resume activity. been combined in a formal meta-analysis, the All prostaglandin synthesis inhibitors are results of which are shown in figure 6. Some effective inhibitors of myometrial contractility. reservations should be expressed, however, both in and outside pregnancy. There is also with regard to the interpretation of these data. no doubt that they are more effective in this The results show that indomethacin, either respect than any of the betamimetic drugs. No alone or in combination, was statistically signif- case has been reported in which a betamitnetic icantly more effective in delaying delivery for at drug resulted in suppression of uterine contrac- least 48 hours, for at least 7 to 10 days, and tility after inhibition of prostaglandin synthesis beyond the preterm period than the control had failed, while the reverse has repeatedly treatments, which consisted of placebo, been observed." ethanol, and betamimetic agents. The typical

236 Inhibition of Preterm Labor Interventions odds ratios for these outcomes, however, are interchangeable."' They may roughly fulfill the each based on only 2 of the 5 trials. There also same function, but thisdoes not mean that they was a statisticallysignificant reduction in the will all have the same effects and ill-effects. incidenceof low birthweightinthe The most serious potential side effects are indomethacin-treated group across the 4 trials peptic ulceration, gastrointestinal and other that reported on this outcome (see figure 19.6). bleeding, thrombocytopenia, and allergic reac- All reports provided data on fetal and tions. Gastrointestinal irritation is commonwith neonatal death. Overall, the data show no the use of prostaglandin synthesis inhibitors, reduction in the incidence of fetal and neonatal and it can occur irrespective of the route of death with the use of indomethacin in preterm administration. With indomethacin it is less fre- labor. The typical odds ratio of 0.61 had a con- quent with rectal than with oraladministration, fidence interval from 0.33 to 1.11. The inci- and, as the bioavailability of the drug is identi- dence of respiratory distress syndromealso cal with both routes of administration,'''' the showed no real difference between the rectal route offers some advantage. Nausea, indomethacin and the control groups across tri- vomiting, dyspepsia, diarrhea, and allergic rash- als, with a typical odds ratio of 0.62 and a wide es have all beenobserved in women treated, confidence interval from 0.25 to 1.58. even briefly, withprostaglandin synthesis None of the trials indicated that the useof inhibitors in preterm labor. Headache and inhibitors of prostaglandin synthesis was associ- dizziness may occur at the very start of treat- ated with an increased incidence of major ment. Gamissans and his associates's'reported problems for either mother or baby. On the systematically on the incidence of headache, other hand, none of these trials were truly maternal tachycardia above 120 heats per placebo controlled, and the number of women minute, vomiting, epigastric pain, andrectal included in these trials has not been large intoleranceintheirtrialcomparing enough to stand a chance of uncovering rare indomethacin with placebo in association with adverse effects. ritodrine treatment. Only two of these side effects were observed more frequently in the Unwanted effixts indomethacin-treated group. Epigastric pain Inhibitors of prostaglandin synthesis are was observed in 6 (4%)and rectal intolerance not innocuous. Three pointsneed to be consid- in 7 (5%) of 148 indomethacin treated women; ered. First, there are numerous potentialside these symptoms occurred in only 2 of 149 effects because of the ubiquitous nature of the women in the control group. prostaglandins. Second. the drugs and doses Prostaglandin synthesis inhibitors cross that are used for inhibition of pretermlabor from the mother to the fetus1' and may influ- also suppress prostacyclin andthromboxane ence several fetalfunctions. A great deal of synthesis. Third, the drugs are both chemically information has been gathered on the fetal and pharmacologically so different from each effects of these drugs in experimental animals. other that they should not be considered as The results are not always easy to interpret,

Marc Keine 237 220 Advances in the Prevention of Low Birthweight

however, because of the variety of species monary hypertension have been reported in studied, and differences in the type of drug the controlled trials that we reviewed. Both of used and in the dose, route, and duration of these, one in the placebo group and one in the administration. Apart from a prolonged bleed- indomethacin-treated group, occurred in ing time, which is a constant feature in infants Gamissans' trial in women with ruptured mem- born with detectable levels of such drugs, branes.'" effects in human fetuses and neonatesare Wiqvisr6 compiled reports from controlled mostly based on anecdotal reports. The most and uncontrolled clinical studies in whichcare- consistent observations relate to the cardiopul- ful pediatric examination of the newborn had monary circulation and to renal and hemostatic been carried out. For a total of 730 mothers functions.'" included in these studies, he found 17 infants The major worries about the use of such with persistent pulmonary hypertension (2.3%); drugs for the inhibition of preterm labor have 14 of these infants recovered within a few days resulted from their influence on the ductus and 3 died. A similar approach was followed arteriosus. Closure of the ductus after birth con- by Gamissans and Balasch,t44 who found 19 sists of an initial functional closure by muscular cases (1,5%) among a total of 1,235 women contraction followed by definitive anatomical who received prostaglandin synthesis inhibitors closure, which is a much slower process that is in preterm labor; 16 of the infants survived and rarely accomplished within the first week of 3 died. Whether or not this incidence is higher life. Prostaglandin synthesis inhibitors cause than it would have been without inhibition of constriction of the ductus in the neonate, an prosuglandin synthesis is impossible to deter- effect that has been conclusively demonstrated mine from such data. in placebo controlled trials of neonatal Data both from experimental animals and indomethacin administration.'"3."4 Autopsy and human neonates suggest that the responsive- cardiac catheterization data from infants who ness of the ductus to indomethacin is lower at presented with congestive heart failure at or lower gestational ages. If such a difference in after birth, have suggested that severe constric- responsiveness exists in utero, it would imply tion of the ductus may also occur before birth that the risk of ductus constriction and its in association with inhibition of prostaglandin potential sequelae would be smallest when synthesis.119.'m most gain is to be made from arresting preterm Constriction of the ductus during fetal life labor, and largest at gestational ages which probably has little effect on fetal oxygenation hardly provide an indication for inhibition of in the short term, as effective shunting can be labor. Itis also probable, although this is not maintained through the foramen ovate. borne out by the available controlled and ,Prolonged prenatal constriction of the ductus uncontrolled data in preterm labor, that the arteriosus can lead to pulmonary hypertension duration of treatment is of influence. The and possibly to tricuspid insufficiency in the longer prostaglandin synthesis inhibition is con- newborn.119 Only two cases of persistent pul- tinued, the greater the risk is likely to be.

238 221 Inhibition of Preterm Labor Intenfentions

Indomethacin treatment may alter both enzyme, but that also the platelets themselves fetal and neonatal renal function. Renal dys- are rendered permanently nonfunctional. They function and reduced urinary output has cannot restore normal hemostasis; for this to repeatedly been noted in infants treated with occur, they must be replaced by new platelets. indomethacin to close a patent ductus arteriosus.'""-"'9 The effect is apparently dose related and transient. Renal function usually returns toward pretreatment values within 24 hours CONCLUSIONS after stopping the treatment."'s Several reports Hopes that inhibition of uterine contrac- have indicated impaired renal function in fetus- tions can resolve the entire issue of preterm es and in the neonates at birthfollowing birth and its associated mortality and morbidity administration of prostaglandin synthesis are unrealistic and naive, at best. On the inhibitors to the mother.''2 Long-term mater- whole, the proportion of preterm births that nal treatment may influence fetal urine output can be and are worth being averted by tocolyt- enough to alter amniotic fluid volume, although ic treatment is not larger than the proportion of other mechanisms may also be involved in the preterm births that is actually provoked with reduction of amniotic fluid volume occasionally the same hopes of avoiding Mortality and mor- seen during indomethacin treatment.'' There is bidity. This is not to say that there are no situa- no evidence from either maternal or neonatal tions in which inhibition of preterm labor is indomethacin treatment that the use of this worthwhile. Rather, it emphasizes that stopping drug in preterm labor would lead to permanent uterine contractions does not necessarily mean impairment of renal function in the infant." improving outcome either for the mother or for Inhibitors of the cyclooxygenase enzyme the baby. Clinicians cannot escape their corn- all inhibit platelet aggregation and prolong mitmments and they must provide care for bleeding time. They do so in the mother, in the preterm labor within the constraints of the fetus, and in the neonate at birth.'N Since imperfect knowledge that is available. That will neonates, and particularly preterm neonates, include tocolytic treatment, and the main issue eliminate these drugs far less efficiently than is how to maximize potential benefit and mini- their mothers,"2.r4 these effects will be of longer mize potential harm. duration in the baby than in the mother. There In that context, only two categories of are major differences in this respect between drugs presently merit consideration for the inhi- different inhibitors of prostaglandin synthesis. bition of preterm labor: Betamimetic agents and Salicylates are particularly troublesome. As inhibitors of prostaglandin synthesis. All others mentioned earlier, they acetylate the cyclooxy- are either obsolete or in an experimenul stage. genase enzyme and permanently incapacitate There is no longer a place for ethanol, relaxin, it. Unlike most cells in the body, blood or progesterone in the treatment of preterm platelets cannot manufacture new enzyme. This labor. Oxytocin analogues and calcium antago- implies that not only the cyclooxygenase nists have been insufficiently studied to asses6

Marc Keirse 239 Advances in the Prevention of Low Birthweight whether they have any beneficial effect. The that they are only useful when the time that is use of other drugs, such as magnesium sul- gained before delivety is used to implement phate or diazoxide, should only be permitted effective measures. Such measures could within the context of adequately controlled tri- include transfer of the mother to a center with als to determine whether or not their acclaimed adequate facilities for intensive perinatal and benefits exist and outweigh their known neonatal care, the administration of corticos- adverse effects. teroids to reduce perinatal mortality and mor- This does not imply that the evidence in bidity, or the judicious use of "expectant favor of either the betamimetic agents or the management" in the period of gestation in inhibitors of prostaglandin synthesis is beyond which the infants' chances of intact survival are reproach. On the contrary, there are many very poor. flaws in the evidence available, particularly Powerful drugs are dangerous when used with regard to prostaglandin synthesis inappropriately. Administration of these drugs inhibitors. Moreover, these two categories of in preterm labor requires a valid indication and drugs contain many compounds, not all of careful control of maternal and fetal condition. which can be assumed to have the same Betamimetic agents are currently the drugs effects. It is also worth remembering that there of choice. For women with cardiac disease, are, in these wide classes of agents, drugs that hyperthyroidism, and diabetes mellitus, howev- have never been evaluated against "placebo er, the risks of betamimetic drug treatment will treatment" in preterm labor; have never been nearly always outweigh its potential benefits. shown to be superior to other, more validated, Maternal side effects are inevitable with drug treatments; and, yet. have caused serious betamimetic drug treatment, hut serious com- complications, including maternal death. plications are largely avoidable. There is no Specific agents that have never been tested evidence that concurrent administration of cal- against placebo or no treatment, or have not cium antagonists or S-1 receptor blockers pro- been shown conclusively to be superior to oth- tects the mother against complications of ers, should probably be dropped from clinical betarnimetic drug treatment. Nor is there any practice. There is something to be said for firm- evidence that such combinations are of benefit ly convincilg those pharmaceutical industries to the baby. There is enough evidence, albeit which propagate such agents that preterm observational, that vigorous hydration causes labor is too serious a problem to be subjected more harm to the mother than it prevents. to drug treatments that have not been evaluat- On the whole, prostaglandin synthesis ed by randomized controlled trials. inhibitors are more powerful inhibitors of uter- Betamimetics and prostaglandin synthesis ine contractions than the betamimetic agents. inhibitors are effective in postponing delivery There are too few data from controlled compar- and in prolonging pregnancy. There is no evi- isons, and their quality is too poor, to recom- dence that the use of these drugs reduces mend prostaglandin synthesis inhibitors as a infant mortality or morbidity. This would imply first link: approach in the inhibition of preterm

240 Inhibition of Preterm Labor ') Intoventions labor. They would appear to be the logical 7. Tamura, R. K., Sabbagha. R. E, Depp, R., Vaisrub, N,, Dooley, S. L., and Socol, M. L.(1984).Diminished choice, however, if labor needs to be inhibited growth in fetuses born preterm after spontaneous in women with cardiac disease, hyperthy- labor or prelabor rupture of the membranes. roidism, or diabetes or if betainimetic treatment A merka rt journ(4l of obstetrics and Gynecology 148, 1105-1110. fails at very young gestational ages. Their 8. Keirse, M. J. N. C.(1989).Preterm delivery.In 1. potential hazards, weighed against potential Chalmers,M. J. N. C. Keirse, and M. W. Encon benefits, do not justify the use of such drugs, in (Eds.), Effective care in pregnancy and childbirth the doses that are necessary to inhibit uterine (Vol. 2, pp.1ro-i292). Oxford: Oxford University contractions, for any longer than is necessary Press. (two or three days). Nor does the available evi- 9. Kaltreider, 1). E. and Kohl, S.(1980), Epidemiology of preterm delivery.Clinical Obstetrics and dence justify the use of aspirin and other salicy- Gynecology 23. 17-31. lates (in the large doses that are required) for 10, Kanhai. 11. 11. 11.(1981). Achtergnmden en konsek- inhibition of preterm labor. wenties van vroeggelx)orte. Unpublished MD thesis. Leiden Ilniversity, Leiden, The Netherlands. 11. King. J. E. Grant, A., Keirse. M. J. N. C., and REFERENCES Chalmers, I.(1988).lietamitnetics in preterm labor: An overview of the randomized controlled trials. 1. Keirse. M. J. N. C., and Kanhai. 11. II. IL (1981). An British Journal of Obstetrics and Gynaecology 95, obstetrical viewpoint on preterm birth with particular 211-222. reference to perinatal morbidity and mortality.In: 11. J. Iluisjes (Ed.). Aspects qf perina(al morbidity 12. O'Driscoll. NI.(19 77).Discussion. In: A. (pp 1-35). Groningen. The Netherlands: Anderson. R. Beard. J. M. Brudenell, and P.NI. Universitaire Boekhandel Nederland. Dunn (Eds.). Pre-term labor Proceedings of the Fifth study (iroup of. the Royal Colhwe of N. C.. Howat. P.. Baum, .1. 2' Rush, R. Keirse. Obstetricians and Gynaecologists (pp. 309-370). D.. Anderson. A. li. M., and Turnbull, A. C.(1976). London:Royal College of Obstetricians and Contribution of preterm delivery to perinatal mortali- Gynaecol()gists, ty.British Medicallouriud 2.965-9M. 13. Cousins, I.. M., llobel, C. J.. Chang. R. J.. Okada. D. 3. Christensn. K. K Ingemarsson. 1., Leideman. T., NI., and Marshall, J. R.(1977).Serum pnigestenme Solon), 11 and Svenningsen, N.(1980),Effect of and estradio1-17-beta levels in premature and term ritodrine on labor after premature rupture of the labor. American journal of Obstetrics and membranes. Obstetrics tind Gynecology 55, 187-190. Gynecology 127. 012-615. Rush, R. W., Davey, 1). A.. and Segall. M. L.(1978). 14. Suzanne. E. Fresne. J. J.. Portal, 15.. and Baudon, J. The effect of preterm delivery on perinatal mortality. (1980).Essai therapeutique de l'indometacine dans British ournal (y. Obstetrics aiul Gynaec(?/ogy 85, les menaces craccouchement premature.71.Yrapie 800-811. 35, 75I-760.

S. Van Kamp. 1.. Kanhai. II., and Keirse, M. N. C. 15. Agustsson. P., and Patel, N. B.(1987). The predic- (Forthetnning).The changing pattern of preterm tive value of fetal breathing movements in the diag- birth. nosis of preterm labor.British Journal cf Obstetrics aml Gynaecolc)gy 94, 860-8b3, 6. Keirse, N1.). N. C. (19'9), Epidemiology of pre-term labor.In: M. J. N. C. Keirse. A. 15. M. Anderson, and V). Castle. 15. NI., and Turnbull. A. C.(1983). The pres- J.Bennebroek Gravenhorst (Eds,), Human ence or absence of fetal breathing movements pre- Parturition: New Concepts and Developments (pp. dicts the outcome of preterm labor.Lancet 2. 219-23.4). The Ilague: Leiden University Press. 471-473.

Marc Kelm, 241 224 Advances in the Prevention of Lout Binhuteight

17. Fuchs, A. R., Husskin, P, Sumulong, L.. Mieha, J. P., 26. BrC!art. G., Gouiard. J.. Blondel, B., Maillard, F., Yusoff Dawood, M., and Fuchs, F.(1982),Plastna Chavigny, C., Sureau. C., and Runwau-Rouquette, C. levels of oxytoein and 13,14-dihydro-15-keto- (1981).A comparison of two policies of antenatal prostaglandM F2a in preterm labor and the effect of supervision for the prevention of prematurity. ethanol and ritodrine.American Journal of International Jouriud of Epidemiology 10,241-244. Obstetrics and Gynecology 144, 753-759. 27. FIGO News. (1976).Lists of gynecologic and obstetri- 18. Noort, W A., De Zwart, F. A., Kragt, H., and Keirse, cal terms and definitions.International Journal of M. J. N. C.(Forthcoming). Can urinary thrombox- GYnaecology and Obstetrics 14, 570-576. ane excretion presage preterm delivety? Eunpean 28. World Health Organization.(1977).Recommended Journal of Obstetrics. Gynecology. and Reproductii definitions, terminology and format for statistical tables Biokk) related to the perinatal pericx1 and use of a new certifi- 19. Papiernik, E.. Bouyer, J., Collin, 11, Winisdoerffer. cate for cause of perinatal deaths.Acta Obstetricia G., and Dreyfus, .1.(1986).Precocious cervical Gynecol(frgica kandinavica 56, 247-253. ripening and preterm labor.Obstetrks and 29. World Health Organization.(1969).Prevention of Gynecology 67, 238-242. perinatal morbidity and mortality.Public health 20. Stubblefield, P. G. (1978). Pulmonary edema occur- papers 42 Geneva: World Health Organization, ring after therapy with dexamethasone and terbu- 30. World Health Organization. 0950).Expert group on taline for premature labor:A case report. prematurity final report.kchnical report series 27 American Journal (f Obstetrics and Gynecokgy 132, Geneva: World Health Organization. 341-342. 31. World Health Organization (1961).Public health 21. Anderson, A. B. M., and Turnbull, A. C.(1969). aspects of low birthweight.Technical report series Relationship between length of gestation and cervi- 21 7 Geneva: World Health Organization. cal dilatation, uterine contractility and other factors during pregnancy.American Journal of Obstetrics 32. \Thrld Health Organization.0979-1.Definitions and and 4ynecoh)g3' 105, 1207-121-4. recommendations. International statistical classifka- lion of diseases (9th rev.). Vol.1. 763-768.Geneva: 22. Bell, R.(1983). The prediction of preterm labor by World ficalth Organization. recording spontaneous antenatal uterine activity. British Journal of Obstetrics and Gynaecology 90, 33. Chamberlain. R.(1975).Birthweight and length of 884-887. gestation.In:British births /97(/tiI.I. The first uwek (00, (pp, 50-88). lAindon: lieinemann. 23. Katz, M.. Newman, R. B.. and Gill. P. J.(198(). Assessment of uterine activity in ambulatory patients Ileinonen. K..I lakulinen, A., and Jokela, V.(1988). at high risk of preterm labor and delivery. Survival of the smallest: Time trends and determinants American journal (-,f Obstetrics arid Gyneccilogy 154. of mortality in a very preterm popubtion during the 44-47. 1980s. Lancet 2. 201-20-7. 24. Morales, W. J., Diebel, N. D., Lazar. A. .1.. and 3i. flollman. H. J.. and Bakketeig. L. S.(198.4).Risk fac- Zadrozny. D.(1986),The effect of antenatal dex tors associated with the c)ccurrence of preterm birth. anwthasone on the prevention of respiratory distress clinkal obstetrics and Gynecology 27. 539-552. syndrome in preterm gestations with premature rup- 36. National Center for Health Statistics.(1985).Monthly ture of membranes. Amerkan Journal of Obstetrics vital statistics report, 34 (Suppl.). 6. and G:ynecology 154, 591-595. 37, Verloove-Vanhorick. S. P.. Verwev. R. A.. Brand. R.. 25. Alexander, S.. and Keirse. M. J. N. C. (Forthcoming). Bennebroek Gmvenhorst. J.. Keirse. M. .1. N. C.. and Formalized risk assessment.In:M. W. Enkin. Ruvs, J.If.(1986), Neonatal numality risk in relatk in N. C. Keirse. and I. Chalmers (Eds.). Iffective care in to gestati(mal age and birthweight:Results of a pregnancy and childbirth.Oxford:Oxford national survey of preterm and %.ery-I( iw-birthweight University Press. infants in the Netherlands. lancet 1. 55-57,

242 Inhibition (y- Preterm Labor 225 `Z:541.

intementions

38. Verkmve-Vanhorick, S. P.. and Verwey. R. A.(1987). tocolytic agent to prevent premature delivery. Project on preterm and small for gestational age American journal of Obstetrics and Gyiwcology 142, infants in the Netherlands 1983: A collaborative sur- 840-8.45. vey.Unpublished MD thesis. Leiden University, 49, Adamsons, K.. and Wallach. It C. (1984). Dia/oxide Leiden. The Netherlands. and calcium antagonists in pretenn labor.In: F. 39. McGregor. J. A., French, .1.I., Re Her, L. 13 Todd. J. Fuchs and P. C. Stubblefield (Eds.), Preterm birth: K.. and Makowski. E.L. (198().Adjunctive ery- Causes. prevention and management. (pp. 249-263). thromycin treatment for klic)pathic prewrm labor: New York: Macmillan. Results of a randomized, double-blinded, placebo 50. Eskes, T. K. A. 13., Kornman, J. J. C. NI.. Buts, It S. G. American journal q,1 Obstetrics and cc)ntRilkd trial. M., Hein. P. R., Gimbnl.re. J. S. F., and Nimk. J. T. C. Gynecology 15.1. 98-103. (1980).Maternal morbidity due to beta-adrenergic 40, Akerlund. M.. Straml)erg. P. Hauksson. A., Andersen, therapy:Ike-existing cardiomyopathy aggravated by

L. F.. Lyndrup, J.. Trojnar, J.. and Melin. P.(1987). fe not e ro I. ropea Pi jou rual of Obstetrics. Inhibition of uterine contractions of premature labor 6:ynecologv. and Reproductity Biology 10. 41-44. with an oxytoein analogue:Results from a pilot 51. Parsons. NI. T.. Owens. C. A., and Spellacy. W. N. study. British journal of Obstetrics awl Gyiuwcology (1987 ).Thermic effects of tocolytic agents: 94, 1040-1044. Decreased temperature with magnesium sulfate. 8.Keirse, M. J. N. C.(1989).Preterm delivery.In 1, Obstetrics and qynecology 69, 88-90. Chalmers.NI. J. N. C. Keirse. and NI. W. Encon (Eds.), Effective care in pregnancy and childbirth 52, Steer. C. NI.. and Petrie. R. 11.(1977). A comparison (Vol. 2, pp. 1270-1292). Oxford: Oxford University of magnesium sulfate and alcohol for the prevention of -preMittlre labor.American journal of Obstetrics Press. and Gynecohmgy 129 42. Turnbull, A. C.(1987). Commentary: An oxvtocin inhibitor for suppressing preterm labor.British 53. Tchilinguirian, N. C., Natem. R.. Sullivan. C. 13.. and journal of Obswtrics and Gynaecology 9i, Craparo, F. 3.(1984). The use of ritodrine and magnesium sulfate in the arrest of premature labor. 1009-1013. International louriud of 6:ynaecology and Obstetrics 43. Fuchs. F., and Fuchs. A-R.(1984).Ethanol for the 22,117-123. prevention of preterm birth.In:F. Fuchs and P. C. Stubblefk.ld (Eds.), Preterm birth:Otuses. prawn- 51. ,Andersscin, K. F. (1977). Inhibition of uterine activi- tion and management (pp. 207-222).New York: ty by the calcium antagonist nifedipine.In: A. nderson, R. Heard. J. M. lirudenell. and P. NI. Dunn Macmillan. (Eds.), Pre-term labor Proceedings (y. the Fifth Stiuly 44. Keirse, NI. J. N. C.(1984).A survey of tocolvtic Group of the Royal (ollege of Obstetricians and drug treatment in pretenn labor.British Journal of Gynaecologists (pp. 101-11.0.London:Royal Obstetrics and Gynaeccdov 91. 421-130, College of Obstetricians and (...lynaecologists. 45. Lewis. P. J.. de Swiet, M Boylan, P., and Bulpitt, C. 55. Read, NI. I) and \XCliby, D. E.(1986). The use of a (1980). How obstetricians in the Vnited Kingdom calcium antaw mist (nifedipine) to suppress preterm manage preterm labor.British journal of Ohqvirics labor.British journal mf Obstetrics mind Gynaecokmgy Wid Gynaecology 87,574-577. 93,9,23-937. Trudinger, B. J.. and Hoshell, L.(1987). A survey of Ilmsten, Andersson. K. E., and Wingerup. L. the management of premature labor by Australian (1980). Treatment of premature labcir with the calci- obstetricians.Australian and Neu. Zealand Journal mn antagonist nifedipine.Archires qf Gynecology of obstetrics wul Gynaecology 27, 188-195. 2"). 1-c. .17. Petrie, R. 11. (1981). I( wolysis using magnesium sul- Keirse, M. 1. N, C.(1984).lietamimetic drugs in the fate. Seminws in Pehnatology 5, 2(6-23, prophylaxis cif preterm labor:Extent and rationale Spisso, K. R.,I larbert. G. M.. Jr. and Thiagarajah. S. of their use.l3ritish Journal of Obstetrics amid (1982). The use of magnesium sulfate as a primary (J:vnaecofiNy 91. t3 1-437.

Marc Keirse 2,43 226 Advances in the Prevention of Low Birthweight

58. Bishop, E.11,. and Woutersz, T. B.(1961). 69. lngemarsson. 1,, Azulkumaran, S.. and Kottegoda, S. Ismcsuprine, a myometrial relaxant:A preliminary R.(1985). Complications of betamimetic therapy in report. Obstetrics and Gynecology 17, 442-446. pretenn lalxir. Australian and Arm' Zealand Journal 59. Hendricks, C. H., Cibils, L. A., Pose, S. V.. and (fObstetrics and Gynaecology 25,182-189. Eskes. T. K. A. B.(1961). The pharmacologic con- 70. Adam, G. S.(1966).Isoxuprine and premature trol of excessive uterine activity with isoxsuprine. labor.Australian and New Zealand Journal of American journal Obstetrics and Gynecology 82. Obstetrics and Gynaecology 6, 294-298. 1064-1078. 71. CaiderA. A., and Patel, N. B.(1985),Are 60. Baillie, P., Meehan, F.P., and Tyack, A. J.(1970). bvtamimetics worthwhile in preterm In: R. Treatment of premature labor with oreiprenaline. W, Beard and F. Sharp (Ed.s.), Pre-term labor and its British Medical Journal 4, 154-155. consequences: Proceedings qf the Thirteenth Study Group of the Royal C.'cillege of Obstetricians and 61. Barden. T. P.(1971).Inhibition of human prema- Lute labor by inesuprine hydrochloride.Obstetrics Gynaecologists (pp. 209-218).London;Royal College of Obstetricians and Gynaecologists. and Gynecology 37. 98-105. 62. Wesselius-De Casparis, A., Thiery, M., Yu Le Sian, Castren, 0,, Gummerus, NI., and S ..rikoski, S. A., Baumgarten, K.. Brosens. I,, Gamissans, 0., (1975).Treatment of imminent premature labor: A Stolk, J. G., and Vivier, W. (1971).Results of dou- ccnnparison between the effects of nylidrin chloride multicentre study with ritodrine in prema- and isoxuprine chloride as well as of ethanol. Acta Obstetrica et Gynecologica Scandinarica 5.i, 95-100. ture labor.British Medkal Journal 3. 144-147. 73. Cotton, D. 13,, Strassner, H. T., Hill. L. M., Schifrin, B. 63. Edelstein, H., and Bane. P.(1972).The use of fenoterol (Berotec) as compared with orciprenaline S., and Paul, R. H.(1984). Comparison of magnesium sulfate. terbutaline and a placebo for inhibitkm (Alupent) in the treatment of premature labor:A companaive study. Med Prot* 18, 92-96. of preterm labor: A randomized study. ournal of Repmductity Medicine 29, 92-97. 64. Liggins, G, C., and Vaughan, G. S.(1973). 74. Csapo, A. I.. and Herczeg, J. Intravenous infusion of salbutamol in the manage- (1977).Arrest of premature labor by isoxsuprine. ment of premature labor. Journal qt. Obstetrics and American Journal qf Gynaecology of the Britith Commonwealth 80, Obstetrics and Gynecology129, 482-491. 29-33. 75. Das. R. (1969). Isoxsuprine in premature labor. journal of Obstetrics and Gynaecology ofIndia. 65. Gummerus. M.(1975).fiemnning der drohenden 566-570. FrOhgeburt mit Nylidrin und Verapamil.Zeitschrill Geburtshilfe und Perinatologie 179, 261-266. 70. Howard, T. E., Killam. A. P., Penney, L.L., and 66. Lipshitz. Bane. P., and Davey. D. A.(1976), A Daniell. W. C.(1982).A double-blind randomized comparison of the uterine 14-2-adrenoreceptor selec- study of terbutaline in premature labor.Military tivity of fenoterol, hexoprenaline. ritodrine and Medicine 147. 305-307. salbutamol,South African Medical journal SO, 77. Larsen. .1.F.. Eldon. K.. Lange, A. P. Leegaard. 1969-1972, Osler. M.. Sederberg Olsen. J.. and Permin, M. 67. Ingemarsscm, I.(1976),Effect of terbutaline on (1986).Ritodrine in the treatment of preterm labor: premature labor:A double-blind placebo con- Second Danish multicenter study.Obstetrics and trolled study,American Journal qf Obstetrics and Gynecology67,607-613. Gynecology 125, 520-524. 78, Larsen, J.F.. Hansen, M. K.. Hesseldahl, H., 68. Eskes, T. K. A. 13.. and Essed. G. G. NI.(1979). Kristoffersen. K., Larsen. P. K.. Osier. M., Weber, J., Inhibition of uterine contractility with Is-mimetic Eldon. K., and Lange. A.(1980).Ritodrine in the drugs.In:M. J. N. C. Keirse. A.13. M. Anderson. treatment of preterm labor:A clinical trial to com- and J. Bennebroek Gravenhorst (Eds.), Human par- pare a standard treatment with three regimens turition (pp. 165-187).The Hague:Leiden involving the use of ritodrine.British journal of University Press. Obstetrics and Gynaecology87, 949-957.

244 Inhibition qt. Preterm Labor 2 :? -7 Interivntions

79. Levin, D.L. (1980).Effects of inhibition of 91Yusuf, S., Peto, R., Lewis, J., Collins, R., and Sleight. prostaglandin synthesis on fetal development, oxy- P. (1985).8-bk)ckade during and after myocardial genation, and the fetal circulation.Seminars in infection:An overview of the randomized trials. Perinatology 4, 35-44. Progress ift Cardiovascular Di.seases 5, 335-371. 80. Merkatz, I.R., Peter. J. B., and Barden. T. P.(1980). 92. Brown. S. Nt, and Teiani. N. A.(1981). Terbutaline Ritodrine lwdrochloride: A beutnimetic agent fm use sulfate in the prevention of* recurrence of premature in preterm labor.11.Evidence of efficacy.Obstetrics labor.Obstetrics and Gynecology 57, 22-25. and Gynecoky 56, 7-12. 93. Creasy, K. K Go!bus, M. S., Laros, R. K., Parer, J. T., 81. Penney, L. L and Daniell, W. C.(1980).Estimation and Roberts, J. M.(1980).Oral ritodrine mainte- of success in treatment of premature labor: nance in the treatment of preterm labor.American Applicability of prolongation index in a double-blind, Journal of Obstetrics and Gynecology 137, 212-219. controlled, randomiz.ed trial.American Journal of 9.1. Smit. D. A.(1983).Efficacy of orally administered Obstetrics and Gynecokw. 138, 345-346, ritodrineafterinitialintravenoustherapy. 82. Sivasamboo. R.(1972).Premature labor.In: K. Unpublished MD thesis, University of Limburg, The Baumgarten and A. Wesselius-De Casparis (Eds.), Netherlands. Procivdings of the International Symposium on the 95. Crimley, P.(Forthcoming).PlInnoting pulmonary Treatment o.f Fetal Risks, Baden. Austria (pg. 1(-20). maturity.In:NI. W. Enkin, M. J. N. C. Keirse, and I. Vienna: University of Vienna Medical School. Chalmers (Eds.). Effective care in pregnancy and 83. Spellacy, W. N., Cruz, A. C., Birk. S. A., and Buhi, W. childbirth. Oxford: Oxford University Press. C.(1979).Treatment of premature labor with filo- 96. Liggins, G. C.. and Howie, R. N.(1972).A con- drine:A randomized controlled study.obstetrics trolled trial of antepartum &corticoid treatment for and Gynecology 54, 2Z0-223. prevention of the respiratory distress syndrome in 84. Barden. T. P.Unpublished data included in the premature infants. Pedknrics 50. 515-525. reports of Merkatz et al. (1980) and King et al. (1988). 97. Caritis, S. N., Carson, D., Greebon, D., McCormick, 85. Hobel, C. J. Unpublished data included in the reports M., Edelstone. D. L. and Mueller-Heubach, E. (1982). A comparison of terbutaline and ethanol in of Merkatz et al. (1980) and King et al. (1988). the treatment of preterm labor. American Journal of 86. Mariona, F. G.Unpublished data included in the Obstetrics and Gynecology 142, 183-190. reports of Merkatz et al. (1980) and King et al. (1988). 98. Finley, J Katz, NI., Roias-Perez. NI.. Roberts. J. NI., 87. Scommegna, A., and Bieniarz,.Unpublished data Creasy, R.K.. and Schiller,N.B. (1984). included in the reports of Merkatz et al. (1980) and Cardiovascular consequences of placebo controlled King et al. (1988). agonist tocolysis:An echocardiographic study. Obstetrics cind Gynecology 64, 787-791. 88, Leveno, K. Klein, V. R., Guzick, D. S., Young, D. C.. Hankins. G. D. V., and Williams, M. L.(1986). 99. Lauersen, N. IL, Merkatz. R.. Tejani, N., Wilson, K. Single-centre randomised trial of ritmirine hydrochlo- II_ Roberson, A., Mann, L. I., and Fuchs. F.(1977). ride for preterm labor. Lancet 1. 1293-1296. Inhibition of premature labor: A multicenter comparison of rit()drine and ethanol..4mericanjournal 89.lialmers, I., Hetherington, J., Enkin, M., and Keirse. of Obstetrics and Gynecolopi, 127, 837-845. M. J. N. C. (Forthcoming). Meth(xis used for synthe- sizing evidence.In:NI. W. Enkin, NI. J, N. C. Keirse, 100.Reynolds, J. W. (1978). A comparison of salbutamol and I. Chalmers (Eds.), afective care in prcNnancy and ethanol in the treatment of premature labor. and childbirth. Oxford: Oxford University Press. Australian and New Zealand Journal of Obstetrics and Gynaecology 18, 107-109. 90. Peto, R Pike, M. C., and Armitage. P.(1976 ). Design and analysis of randomised clinical trials 101.Sims, C. D.. Chamberlain, G. V.P., Boyd. I.E., and requiring prolonged observation for each patient: Lewis. P. J.(1978). A comparison of salbutamol and Introduction and design.British journal (?/. Cancer ethanol in the treatment of preterm labor.British 34, 585-612. Journal of Olytetrics ami G:ynaecc 'logy 85, 761-766.

Marc Kora, 245 22S Advances in the Prevention of Low Birthweight

102.Spearing, G.(1979),Akohol, indomethacin and 112.Kubli, F.(1977),Discussion.In:A. Anderson, R. salbutamol: A comparative trial of their use in Beard. J. M. Brudenell, and P. M. Dunn (Eds.). Pre- pretetm ObsMrics and Gyrwcology term labor Proceedings qf the Ili% Study Groupef 103,Beall. M. IL. E-ar. B. W.. Paul. R. IL. and Smith- the RQvat oitlege of Obstetricians and Gmaeccdogists Wallace, T.(1985). A comparison of ritodrine. terbu- (pp. 218-220). London: Royal College of taline. and magnesium sulfate for the suppression Of Obstetricians and Gynaecologists. preterm labor.Anwikan journal cf Obstetrics and 113.Abr.untwici, It. Lewin. A., Lissak. A.. and Palant, Gynecology 153, 85,t-859, (1980).Nlaternal pulmonary edema occuring after 104.Caritis, S.N.. Toig. G., lieddinger. L.A., and therapy with ritodrine for premature uterine contrac- Ashmead, G. (1984).A double-blind study compar- tions. A etaObstet riciaetGyilecologica ing ritodrine and terbutaline in the treatment of scandinaviea 59, 555 preterm labor, American journal (,/' Obsretrks and 114,Babenerd. J.. and Flehr. I.(1979).MOtterliche G3.neco1ogy 150, 7-1.1. Zwischentalle umer (ier Tokolyse mit Fonoterol. 105.Essed, G. G. M.. Eskes. T. K. A. B.. and ,fongsma. 11. Medizinische Uclt 30. 537-5-II. W.(1978).A randomized trial of two beta-mimetic drugs for the treatment of threatening early labor: 115.Barden. T. P. Peter, J. B., and Nkrkatz. I. R.(1980). Ritodrine hydrochloride: Clinical results in a prospective comparative study A betamimetic agent for use in preterm labor.I.Pharmact logy. clinical his- with ritodrine and fenoterol.Fungrean journal of tory, admiffistration, side effects, and safety. Obstetrics, Gynecofirgy. and Reproductivt, Biology 8, Obstetrics and Gynecology 56, 1-6. 341-348. 106.Gummerus, M.(1983).Tocolysis with hexoprenalin I1O.Daubert. J. C., Gosse. P.. Rio, NI.. Gran. J.Y.. and salbutamol in clinical comparison.Geburtsbilfi, Bourdonnec. C., Pony. J. C.. and (iouffault. und Frauenheilkunde 43. 151-155. (1978).Myocardiopathies en cours de grossesse: Role possible des bOtamimetiques.Archives des 107,Karlsson. K.. Krantz_ M.. and Ilamberger. L (1980). Maladies du Coeur et des litiwaux 71. 1283-1290. Comparison of various betamimetics on preterni Libor survival and development of the child.journal (y. 117.Elliott. It R., bdulla. L. jj,f iLlyeti, J.(1978). Perbtatal Medicine 8. 19-26. Pulmonary oedema asstrciated with riux.frine infusion and betamethasone administration in premature 108.Kosasa. T. S.. Nakayama. R. T.. liale. R. W.. Rinzler. G. Ialmr.British Medkal journal 2.-99-800, 5,. and Ereitas. C. A.(1985).Ritodrine and terbu- !aline eimipared for the treatment of preterm labor. 118.1 acobs. NI.N1.. Knight. A.13., and Arias. F.(1980). Ada Ohstetricia el Gynecohwica Scandinarica 04. Materna! pulmcmary edemaITS LI III ii g from 421-426. lvtamimetic and glucocorticoid therapy,Obstetrics and Gynecology 5(,. 56-59. 109.Ryden. G.(1977).The effect of salbutamol and terbutaline in the management of premature lalx)r. I19.Milliez. J.. Blot. P.. and Sureau. C.(1980).A case Acta ohNtetrieia et Gynecologic(' Seandirtatiea report of maternal death associated with betamimet- 293-296. ics mid betamethasone administration in premature laben. Film/wan Journal of Obstetrics. Gyilecol(gy. 110.Bender. If. G.. Goeckenian. Meger. C.. and and Reproductive 1üolo,e0'11,95-104). NIUntefering. IL (197). Zuni mutterlichen Risiko der medikamentOsen Thkolvse mit Fenoterol (Partusisten). 120.Neihuhr-Jorgensen, 1'.(1980).Pulmonary oedema GeburtshilW und Fraeuribeilku rule 37. 065-674. folkiwing treatment with rit(kIrinc and betamethasone in premature labor.Danish MedicalBulletin 111.jonatha, Goessens. 1... Traub. E.. and Dick. W. 2'.99-100, (197').Eulmonale Komplikaticmen wahrend der Tokolyse.In:IL Jung and E. Friedrich (Eds.). 121.Rogge, :' Yming. S.. and Goodlin. R.(1979). Fenoterol (Partusisten ) bei (ler Bebandlungin der Post-partum pulmonary oedema ass(sciated with pre- Geburtsbilie unit Peril/Wu/ogle (pp.6S- I (8 ). ventive therapy for premature labor.lancet 1. Stuttgart: Georg Thieme, 102-1027.

246 2') InhiNtion of Preterit, Ittbor interventions

122.Tinga. 1). J.. and Aarnoudse. J. G. (1979).Post-par- 132.Michalak. D., Klein. V., and Marquette, G. P. tutn pulmonary (iedema ass(wiated with preventive (1983),Myocardial ischemia:A comphcation of therapy for premature labor, lancet 1. 1026. ritodrine tocolysis.American journal (f Obstetrks and Ciptecolgy 146, 861-862. 123.Wolff. F.,Mc ier,U., and Hoke. A. (1979). Untersuchungen zum Pathoinechanismus schwerer 133.Ries. G. H.( 1979). Kasuistische Mitteilungither das kardicipulmonaler Komplikationen tinter tokolytisch- Auftrcten einer Myokardischilmie unter medikamen- er Behandlung mit 1A-adrenergenSubstanzen und loser Tokolyse mit Ritodrin (Pre-Par).Geburtthilfe Betanwthason.Zeitschrill ftir Geburtisbilfe rind and Entuenbeilkande 39, 33--37. Perinatologie 183. 343-347. 134.Ving, V. K., and Tejani. N. A.(1982).Angina pec- 124.Grospietsch. G.. Fenske. M., Dionch, B., and Ensink. toris as a complication of ritodrine hydrochloride Obstetrics and E. B. M (1982).Effect of the tocolytic agent therapy in premature. labor. fenoterol on body weight. urine exeretkin. blood (l)wecologv 60,385-388. hematocrit. hemoglobin. serum pr()tein. and ek.c- 135.Rona, G., Chappel, C. I., Balazs, T and (haudry. R. trolyte levels in non-pregnant rabbits.American (1959), An infarct-like myocardial lesion and other journal cy'Obstetrics and Gynecolt v. 143. (th7-672. toxic manifestations produced by is(iproterenol in thc r;at. rchii les ofPat/Nikki.67. 443-455. I25.Caritis. S. N.. Edelstone, D. 1., and Mueller-lIeubach. E, (1979).Pharmacologic inhibition of preterm 136.Bieniarz. J.. Ivankovich. A., and Scommegna. A. labor, American journal of Obstetrics and t 19-74). Cardiac output during rimdrine treatment in Gynecology 133, 557-578, premature labor.Amerkan journal of Obstetrics aiitl Gyttecology 118. 910-919. 126.Katz. M., Robertson, P. A.. and Creasy. R. K.(1981). Cardiovascular complications associated with terbu- 137.1)e Swiet, M.(1980).The cardiovascular system. taline treatment for preterm labor.American In:F. E. Hytten and ( . Chamberlain (Eds.), lout-rad of011stetrics and 6)necology 139, 605-608, Clinicalphysi(//ogy in pregnancy (PP. 3-42). Oxliird: Blackwell Scientific Publications. 127.Robenson, P. A., Herron. M.. Katz, M.. and Creasy. R. E.. Hensleigh. P. A.. and Kredenster, K.(1981).Maternal morbidity associated with isox- 138.Eerguson, 3. suprine and terinitaline tocolysis. Eumpean journal I).(1984),Adjunctive use of magnesium sulfate qi obstetric:, Gynecology, and Reproductin, Biologr with ritodrine for preterm labor tocolysis. 11, 371-378. American Journal of Obstetrics and Gynecology 148, 166-171. 128.Benedetti. T. .1.0983).Maternal complications of parenteral 14-sympathomimetic therapy for premature. 139.1.enz. S Detk.fsen. G.. Rygaard. C.. and Veicrslev. L. Isotonic dextrose and isotonic saline. as sol- labor. American Journal of Obstetrics and (1985). vents for intravenous treatment of premature labor Gynecology with ritodrine,journal of Obstetrics and 129.1)hainaut.I..Boutonnet. G., Weber. S.. and G:ynaecology 5,151-154. ResponsabiW des lu-Aa-2- Degeorges. M.(1978). 140.Philipsen. T., Eriksen, P. S.. and Lynggard. F. minu'liques Cliins la genese chine cardiomyopathie (19811. Pulmonary edema following ritc)drinc..-saline du postpartum Nouvelle Nesse .1tedicale-, 058. infusion in premature labor.Obstetricsand 130.Hendrklis. S. K.. Keroes. .1., and Katz, M.(1986). 0:31teculo,L1.58.30-4-308. Electrocardiographic changes associated with rito- 14I,Keirse, M, I. N. C., Noon. W.. and Erwich..1. J. H. NI. tachycardia and drine-induced maternal (1987). Therole.of prost agl a nd i n sand American Journal (,1* obstetrics and hypfikalemia. prcistaglandin synthesis in the pregnant uterus.In: 921-923, Gynecoltigy 1 M. J. N. C. Keirse and II. J. de Koning Gans (Eds.), 131.Kubli. F.(1980). Commentary on maternal morbidi- Printing and induction (!f* labor hy pr()staglandins: ty due to beta-adrenergic therapy.Eu ropea n A state. of the art (pp. 1-25).Leiden. The journal of obstetnCs. Gmecology. and Reproductitv Xe'! he'rla nds jc ,c'rl, aa e Comm it t eefor Iiiologvlii.1.4-4-i. Postacademic Nk.dical Education.

,Ilarc Kelm, 247 23f.) Advances in the Prevention of Low Birthweight

142.Keirse. M. J. N. C.(1981).Potential hazards of 153.Zuckennan. IL. Shaky. E.. Clilad, G., and Katzuni. E. prostaglandin synthetase inhibitors for management of (1984).Further study of the inhibition of premature pre-term labor. Journal (y. Drug Research 6. 915-919, 1:111(ir lw indomethacin.Part II.Double-blind study. WAVicivist. N.(19797.The use of inhibitors of journal ql Perinatal Medicine 12, 25-29. prostaglandin synthesis in ohstetrics.in:M. J. N. C. 15.I.Gamissans.0..aniBalasch,J. (198,1). Keirse. A. B. M. Anderson. and .1. Bennebroek Prostaglandin synthetase inhibitors in the treatment Gravenhorst (Eds.). Human parturithm (pp. 189-200). of preterm lab(ir.In:F. Fuchs and P. G. Stubblefield The Ilague: Leiden Iniversity Press, (Eds.). Preterit) birth:Causes, preivition and man- 1+LSchwartz. A.. Brook, L. Inskr. V., Kohen. E., /or. U., agonent (pp. 223-248). New York: Macmillan. and Lindner. IL R.(1978).Effect of tlufenamic acid 155.Katz. Z,. 1.ancet. NI.. Yemini. NI., Mogilner. B.NI.. on uterine contractions and plasma kvels of 15-keto- Feigl. A.. and Ikn-Hur. H. (1983). Treatment of pre- 13.14-dihydro-prostaglandin F2a in preterm labor. mature labor contractions with combined ritodrine Gyneco/ogic and obgetric buestimuicin 9, 139-1 +9, and indomethacine./n/ernational jou nut/ uf 145.Dornhöfer.W.. and Nlosler.K. If. 19-57. (iputecology and ()Work's 21. 337-3.42, Prosuglandine und 1;-Stimulatoren.In:11. ung and E. 156,1ilake, D. A., Nielwl.3.R., White, R. D., Rumor. K. K. Kliick (Eds.), Th Partusistett 1 bet der NI,. Dubin. N. If,. Robinson, J. C.. and Egner, E. G. Behandlung in der Gehurtshilfe und Perinatologie (pp. (19801. Treatment of premature labor with 194-202/. Stuttgart: Georg Thienw. indomethacin.Advances inProstaglandin and 1.+0.GyOry, G., Kiss. C.. Benyo. T. BAgdany. S., Szalay, J., Throniboxane Research8, 1.466-1,467. Inhibition of Ialuir by Kurcz. Nt. and Virag. S.(1974). 157.Gainissans, Canas. E., Cararach. V.Ribas, prostaglandin antagonist in impending alxirtion and Puerto. B.. and Edo. A.(1978).A study of preterm and term lalxir.Lancet 2, 293, ind(miethacin combined with ritoklrine in threatened 1.17.Witter. E.R.. and Niebyl. J. R.(198(0.Inhibition of preterm labor.Eurolwan Journal of Obstetrics, arachidonic acid metabolism in the perinatal perit)d; o:inecolow. and Reproiluctne Biology S. 123-128. Pharmacology. clinical application. and potential 1.58,Gamiss.ns 0.. Cararach. V.. and Serra. J. adverse effects.SeMillilIN rClilltlioi(.k).10. 310-333. (1982). The role of prostaglandin-inhibitors. beta-adrenergic 1-18.M(xmen. P.. Klok. G.. and Keirse, NI. ,1. N. C.(1984). drugs and glucoeorticciids in the management of Increase in c()ncentrations of prosuglandin endoper- threatened preterm labor.In:It Jung and G, Oxide synthase and prcistacyclin synthase in human Limberti (Eds.). Betamimetic drugs in obstetrics and nlymnetrium in late pregnancy.P).ostag/(Indins 28. perina1 Hhigri3p."1-8.47. Stuttgart: (;org 309-322. 159.Alvan. C.. Orme. NI., Bertillson. I_ strand, R. E.K., 149.Nlitchell. M. 1),(1(...*407,)'athways of arachidonic acid and Pa Ini&r,L. (19-5).Pharmacokinetics of metab()Iism with spedfic application to the fetus and indomethacin. Clinical Pharmacology and mother.Seminars in Perinatologr 10. 2.12-2-1.i. Therapeutics IS, 0i-373, 150.11alle. II.. Ilengst. P.(198I.ZusatztoWysc Lim-6 ImAVallusch. NV. NV.. Novak, II.. Leopold, G.. and Netter. Prostaglandinsvthetasehemmung Indoinctacin. K.41. (19-8t.Comparative bitiavaiLtbility:Influence Zeitschrif Gelnirlshile rind Perinatologu,182. ()ivariousdiets ( )11 thehioavailabilityof 307-370. indomethacin. hiternatumn/ journal of Clinical 151.Van Kets. I I.. Thiery. NI.. Deront. R.. Van Egmond. rmactdog v arid lihpilarrnacy 10. ,( and Back.. G.(1979).Perinatal hazards of chronic antchAd tocolYsis with ind(mlethain. Ph,staglandins 101.Traeger. NOschel, IL. and ZatimseiLl.(19-3), 18. 893-90'. /Air Pharmakokinctik vc)n Indomethazin icrhte.1.1i Scwangeren.h Kreissenden und 15.2.Niebyl. J.R.. Blake. 1).,A.. White. R. D,. Kumor. K. Neugeborenen. Zentralblatt far Gyna4.,ologie 95. NI., Dubin. N. IL. Robinson, .1.C.. and Egner, P. G. 035-0-11. 119801.The inhibition of premature labor with indomethacin.ArnericanJournalof ohstetrks and IO2AVilkinson. A. K. 1980). Naproxen levels in preterm Gyneoclow 136, 1014-1019, infants atter materna! treltment. Lancet 2, 591-59

248 Inhibition ofPretertnLabor 231 Interventions

163.6ersony. W. M,, Peckham, G. J., Ellison. R. C.. 174.Eriedman, Z., Whitman. V,. Maisels, M. J.. Berman, Miettinen. 0. S., and Nadas, A. S.(1983).Effects of W., Jr., Marks, K. H., and Vesell, E. S.(1978). indomethacin in premature infants with patent ductus Indomethaein disposition and indomethacin-induced arteriosus:Results of a national wilaborative study. platelet dysfunction in premature infants. Journal (y. Journal of Pedia(ric.. 102, 895-905. Pinomacology 18, 272-279. Mahony. L.. Carnero, V.. Brett, C., Heymann, M. A., 175.Abnunson, D., and Reid. D. E. (1955).Ilse of relax- and Clyman. It I.0982), Prophylactic indomethacin in in treatment of threatened premature labor. therapy for patent ductus arteriosus in very-low-birth- journal qf Endocrinolcw, and Meutbolism weight infants. Neu. England Journal (1*.tredicine 300, 15. 206-209. 506-510. 176.Chng, P. K. (1981). An analysis of preterm singleton 16i.Arcilla. It A.. Thilenius, 0. G., and Ranniger. K. deliveries and assoeiated perinatal deaths in a total (1969), Congestive hean failure from suspected ductal popukttion.British Journal q' Obstetrics and closure in uter(). _Journal (y* Pi,diatrics Gynaecohlgy 88. 814-818. 166.Wiqvist. N. (1981). Preterm labor: Other drug possi- r7.1)avies, A. E.. and Robertson. NI. J. S.( 1980). bilities including drugs not to use.In:M. G. Ekler Pultmmary oedema after the administratkm of intra- and C. II. Hendricks (Eds.). Preterrn labor (pp. venous saihutamol and erpnnetrine:Case report. 148-175). London: Butterworths. Britisb puma! (y. Obstetrics and Gynaecokkgy 87. 539-541. 167.11etkerur. M. V., Yeh, T. E., Miller. K., (ilasser. It J., and Pikles, It S.(1981).Indomethacin and its effect on 178.Erny, R., Pigne. A.. Prouvost. C., Ciamerre. NI., Malet, renal function and urinary kallikrein excretion in pre- C,. Serment. H., and Murat, J.11986). The elkcts of mature infants with patentductus arteriosus. oral administration of progesterone for premature Mhatrics 68, 99-102. labor. .4 merican Journal of Obstetrics and Gynecology 154,525-529, 168.Gleason. C. A.(1987).Prostaglandins and the developing kidney. Seininarc in Perinatologv 11. 12-21. 179.Fuchs. F.(1976).Prevention of prematurity. American Journal qf Obstetrics an(1 6:rnecolcv126, 169.11eymann. M. A., Rudolph, A. M.. and Silverman. N. II. 809-4420. (19761.Closure of the ductus arterk)sus in premature infants by inhibition of prosuglandin synthesis. New 180.Fuchs. F., l'uchsA-R.. Poblete. V. C., and Risk. A. England ,l(nrrnal (..?1:11tylicipte 29c. 53(}-533. 11967).Effect of alc.thol on threatened premature American Journal of Obstetrics and 170.Cantor. B.. Tyler. T., Nelson, R.NI.. and Stein. (3. H. labor. ( 1980),Oligohydramnios and transient neimatal viwco/ogr anuria:A pcis,sible association with the maternal use 181.Keirse.NI.J.N. C. (197) ). Endogermus of prostaglandin synthetase inhibitors. ournal ...aglandins in human parturition.In:NI. J. N. C. Reproductive Medicine 2i. 2211-223, Keirse (Ed.). Human parturition: New concepts and Leiden. The 171.Itskovnz, f. Abram(>vice. IIand Brandes. I. I. developments (pp.10 I -1421. Netherlands: Leiden t *niversity Press. 11980),Oligohydramnion. meconiom and perinatal cleath concurrent with indomethacin treatment in 182,Morrison, I. C.. Martin. J. N.. MarVin. H. W.. human pregnancy. Journal Qt. Reproductite Medicine Gookin. K.S.. and Wiser, W. L. ( 1987). 2 I. 1$7-1.10. Prevention of preterm birth by ambulatory assess- 172Neerseina. D.. de .Iong, P. A.. and Van \\lick. ,1. ment of uterine activity:A randomized study. (1983).inchimethacin and the fetal renal nonfunetion American journal qf Obstetrics and Gynecology syndrome. European Jou nut/ ff Obstetrics. 156. 536-5-0. Grneothigr, and RepnidUCti1,0 Bio/ogy 1(. 113-121. 183.Stubbs. T. NI.. Van Dorsten. J.P.. and Miller. NI. C. 173.1)e Wit. W,. Van Mourik. 1.. and Wiesenhaan. l(1988). t 19861.The preterin cervix and preterm labor: Relative risks, predictive values, and change over Prokinged maternal indomethacin therapy associated and vith oligohydramnios: Case reports.British journal .1.1 time. American jt,turnal (4. obstetrics Obstetrics and Gynaec o/ogy 95.303-305. Gvnecolozrlii, 829-834.

Marc Keirse 249 4, Advances in the Prevention of Low Birthweight

184.Tamby Raja, It L., Anderson. A. B. M., and Turnbull, A. C.(1974).Endocrine changes in premature labor. Brita Medical Journal 4, 67-71. 185.Tsuie, 3. J., Lai, Y. E, and Sharma, S. D. (1977). The influence of acupun.lure stimulation during pregnancy. Obswtrics and G:rnecologr 50.479-488. 186.Tumer. G.. and Collins, E.(197-if.Fetal effects of regular salicylate ingestion in pregnancy.lancet 2, 338-339. 187.Zuckerman, II., Reiss. II.. and Rubinstein. I.(1974). Inhibitionofhuman prematurelaborby indomethacin.Obstetrics and Gynecoloky 44. 787-792.

250 23 ' (y. Preterm Labor Interventions

Statistical Findings: Inhibition of Preterm Labor: Is It Worthwhile?

Table 19.1 Summary of Approaches Used to Inhibit PretermLabor in the Past 35 Years and TheirPurported Success Rates in the First English LanguagePublication Documenting Their Use

Number of Criterion of Percent success StICCeSS Agent Year Authors women delivery after 36 weeks 100 Relaxin 1955 Abramson and Reid 121 5 contractions delayed 24 hours 82 lsoxsuprine 1961 Bishop and Woutersz 1231 120 deliverydelayed 72 hours 67 Ethanol 1967 Fuchs et al, 1621 52 delivery after 36 weeks 70 Orciprenaline 1970 Baillie et al. 113) 30 delivery delayed 24 hours 53 Mesuprine 1971 Barden 1141 17 not delivered during treatment 80 Ritodrine 1971 Wesselius-DeCasparis vial. 117 43 28 ddivery delayed 1 week 71 Fenoterol 1972 Edelstein and Baillic 150) 88 delivery delayed 24 hours 81 Salbutamol 1973 Liggins and Vaughan 11131 arrest of contractions 80 lndomethacin 1974 Zuckerman et al. 1186) 50 diminished uterine activity 100 Sodium Folicylate 1974 Gyory et al. 1721 50 birthweight2,500 g 86 Buphenine 1975 Castrtin et al.1.331 41 not delivered during treatment 80 Terbutaline 1976 Ingemarsson 1801 15 delivery delayed 3 days or more 100 Nifedipine 1977 Andersson 1101 10 31 contractionsstopped24 hours 77 Magnesium sulphate 1977 Steer and Petrie 115.31 delivery after 36 weeks 92 Acupuncture 1977 1suie et al. 11631 12 delivery delayed 24 hours 83 Elufenamic acid 1978 Schwartz et al. 11451 18 complete Cessation of contractions 94 Diazoxide 1984 Adamsons and Wallach 141 118 decrease in contraction freciuency 76 Oral progesterone 1986 Erny et al. 1521 57 inhibition of contractions 100 Oxytocin analogue 1987 Akerlund et al. 161 13 Table 19.2 Characteristics of Birth and In-Hospital Mortalityand Morbidity in Infants Born Alive Before 32 Weeks (224 Days) of Gestation in theNetherlands in 1983* Number Percentage Characteristics Number Pertentage Characteristks newborn (iwracferistics of birth Charat teristits and morbidity (confirmed diagnoses only) ot 96 9.5 After pwlabor rupture of membranes 453 44.8 Congenital malformations 171 16.9 More than 24 h atter rupture of membranes 226 22,4 Weight below 10th tentile for gestation 417 41.2 After use of tocolytic drugs(anytimet 533 52.8 Respiratory distress syndrome 251 24.8 After corticosteroidadministration 173 17.1 Intracranial hemorrhage 66 6.5 Part of a multiple pregnancy 261 26.0 Convulsions 113 13.2 Wee( h presentation 293 29.0 Septicaemia 155 15..3 Elective del iveryt tn-hospaal mortality of lweborn infants Cesarean section 321 .31.8 Neonatal deaths (28 days) 285 28.2 In-hospital deaths 110 30.7 DatafromVerloove-Vanhorick andVenvey148711701. t Defined as anv delivery following arty obstetricalintervention aimed at !ringing pregnantly to an end beforethe onset ot spontaneous Total 1,010 100.0 labor andor spontaneous rupture of the membranes.

Marc Keince 23 4 Advances in the Prevention of Low Birthweight

Table 19.3 Agents Reportedto be Used for the inhibition of Preterm Labor in the Literature Since 1980 Prostaglandin synthesis Oxytocin analogues Diazoxide Calcium antagonists inhibitors (receptor Antimicrobial agents Magnesium sulfate Betamimetic agents blocking agents) Ethanol Oral progesterone

Table 19.4 Characteristizs of the Trials Compar:ng Betamimetics With Ethanol for Inhibition of Preterm Labor

No. randomized:reported on

Authors (date) Betamimetic Method of Allocated to Allocated treatment allocation betamimetic to ethanol Other ddails

Caritis et al. (1982) Terbutaline random by 92:85 corticosteroids were iv oral sealed \ generally given; maintenance envelopes ?:45 ?:40 magnesium sulphate for 5 days if given if assigned membranes intact treatment failed; ruptured and intact membranes reported separately.

Spearing (1979) Salbutamol iv alternate 22:20 22:22 salbutamol was given -4. oral for 48 hours if ethanol failed after contractions and vice versa

Reynolds (1978) Salbutamol iv alternate 42:42 42:42 all ethanol treated + 200 mg sodium women received phenobarbitone 500 mg methyl prednisolone iv

Sims et al. (1978) Salbutamol iv random by 100:88 all women random- no maintenance open list / \ ized to betametha- ?:42 ?:46 sone vs. placebo; 5 women ethanol betamimetic; 2 betamimetic -4. ethanol

Lauersen et al. (1977) Ritodrine iv -0. oral random 150:135 maintenance for by sealed / \ 4 weeks or envelopes ?:68 ?:o7 until term

Fuchs (1976) all women are also included in the report oft auersen et al. (1977)

252 Statistical Findings: inhibition of Preterm Labor 235 Inteiventions

Figure 19.1 Effect of Betamimetic Drug Treatment in Preterm Labor on theIncidence of Delivery Within 24 Hours in the 14 Trials Which Provided Data on This Outcome*

Odds Ratio

0.01 0.03 0.1 0.3 1 4 10

1 4 1 4 1 Christensen 1980 Spellacy 1979 Barden

Hobe( 1 Cotton 1984 Howard 1982 Ingemarsson 1976 Larsen 1986 Calder 1985

Scommegna 1 Mar lona Wessel lus 1971 4 Leveno 1986 1 Larsen 1980 4

1 1 1 4 4 1 4

See text for explanation on how to interpret thi.; and subsegtwnt figures.

Figure 19.2 Effects of Betamimetic Drug Treatment in Preterm Labor:'Typical' Odds Ratios, With Their 95 Percent Confidence Intervals, Across Trials for theVarious Outcomes Studied Odds Ratio

0.2 0.4 0.6 0.8 1 1.5 2

--1--- 1 1 I 4- Delivery < 24 hours Delivery < 48 hours 4 Preterm birth Low birthweight Respiratory disease Perinatal death

4 4 1 1 1 4

Marc Keirce 253 Advances in the Prevention of Low Birthavight

Figure 19.3 Effect of Betamimetic Drug Treatment in Preterm Labor on the Incidence of Severe Respiratory Disorders Including Respiratory Distress Syndrome

Odds Ratio

0.01 0.03 0.1 0.3 1 4 10

1 Christensen 1980 Spellacy 1979 Barden Hobel

Cotton 1984 4 Howard 1982 Ingemarsson 1976 Larsen 1986 Calder 1985 Scommegna Leveno 1986 Larsen 1980

1 1 1 1 1 1

Figure 19.4 Effect of Betamimetic Drug Treatment in Preterm Labor on the Incidence of Perinatal Mortality Not Attributable to Lethal Congenital Malformations Odds Ratio

0.01 0.03 0.1 0.3 1 4 10

4 4- 4 4 Christensen 1980 I Spellacy 1979 Barden

Hobe! 1

Cotton 1984 4. Howard 1982 Ingernarsson 1976 Larsen 1986 4--

Calder 1985 -4 Scommegna Marlona .-

Wessellus 1971 1 .- Leveno 1986 Larsen 1980 4- Adam 1986 4

254 Statistical Findings: Inhibition qf Preterm Labor Is it Worthwhile? 1 .3 liztoventions

Figure 19.5 Comparison Between Betamimetic Drugs and Ethanol for Treatment of Preterm Labor: Effects on the Incidence of Delivery Within 48 to 72 Hours After Entry into the Trial

Odds Ratio

0.1 0.2 0.4 0.6 0.8 1 1.5 2

I I 4 1 1 I Caritas et al. 1982 Spearing 1979 Sims et al. 1978 Lauersen et al. 1977

Typical Odds Ratio 1

1 4 1 4 -4-- I

Figure 19.6 Effects of Indornethacin in Preterm Labor: 'Typical' Odds Ratios, With Their 95 Percent Confidence Intervals, Across Trials for the Various Outcomes Studied Odds Ratio

0.1 0.2 0.4 0.6 0.8 1 1.5 2

4 I 4- -4- 1 Delivery < 48 hours -4- Delivery < 7-10 days 4- Preterm birth 4- Low birthweight Respiratory distress 4- Perinatal death -4-

1 -4-

Mary Keim, 255 23 S 239 Appendices

Appendix 1".. SymposiumProgram

Sunday, May 8,1988

6:00 p.m. REGISTItkTION

Monday, Mgy 9,1988

7:30 a.m. BREAKFAST

8:30-8:45 a.m. ,SE;SION I: INTRODIVIIONS Presiders: Heinz Berendes:Woodie Kessel; SumnerYaffe Rapporteur PhyllisZucker

WELCat Ir Development Duane Alexander,National Institute ofChild Heahh and Human Development Vince Hutchins, Bureauof Maternal and ChildHealth and Resources

8:45-10:00 a.m. SEsSION II: TRENDS 13rian McCarthy:David Rush Rapporteli?.: HowardI-loffman

BuMi WEIGIRN IN MP17N177:71 SEIZES 8:45 a.m. TREV/T)s LV ktThs op Low Speaker: Marie McCormick,Boston, Massachusetts

Low Brnin WEIG FRAAIX 9:05 a.m. 7kEvps PREIERE DEMERY .'iND .Speaker Gerard Brea rt. Paris,France

9:25 a.m. Disci ;s:C/0,V

10:00 a.m. Ni7RT770N AAP FITNEsIsBREIK

10:30 a.m.-- SESsIoN DETERNI1NANIN Charlotte Catz 1230p m. Presiderc: Ann Koontz: Rapporteur Karla Damus

AMit "HMV GROW A.V) GES11170.\A/ 10:30 a.m. EPIDEVIOIDGICAL DE771411LVINIS or Dt i4ThX, 0 RRENTSEARS .4.7) GUN LV al?KNOW .LEI K :rE Spmker: Michael Kramer,Montreal. Canada

24 0 Advances in the Prevention of Low Birthweight

10:50 a.m. BIOLOGICAL DE7P:RMINANIS OF R774.11.17ildNE GROWTH21417) GmAnaw DminoN, Ct1RRENT STA7IN AAD GUNiv OIIR KNOWIRXiE Speaker: Miles Novy, Portland, Oregon

11:10 a,m. BEIL4170RAL DE7ERMIAAN7S OF A7k417ER/AEGROW 771itm) GEszvioNAL DtRAnas CURRINT STAIT5' AND aiRS' LV OIR Kvotvwxd:' Speaker:Jeanne Brooks-Gunn, Princeton,New Jersey

11:30 a.m. Disamtav

12:30-3:30 p.m. LUNCH

2:00-3:30 p.m. SEsSION IV: INTERVENTIONS Presidem Henry Foster; Jose Belizan; GianDi Renzo Rapporteurs: Patricia Shiono; Donald MeNellis

2;00 p.m. ME M4RC7 1 OF DBMS PR1:7ERII BIRM CLINICALna,11, Spaiker: Robert Creasy, Houston, Texas

2:30 p.m. Rh:VI:IS OF A MREE-YhAR P1?OSIV:17VE CUV1ROLLEDKINDOMIZ.ED DUAL ON PRE17:181 BIIM1 IIREVEMON AT 771E B71ERCITYOF PTI7SIII:R671 Speaker: Eberhard Mueller-Heubach, Pittsburgh,Pennsylvania 3:00 p.m. DiSt.EssioN

3:30 p.m. NUIRrnON AND RINESS BREAK

4:00-5:30 p.m. SESSION V: INTERVENTIONS Presiders: Marshall Klaus; Steven Ng Rapponeur Margaret Freda

4:00 p.m. DEaLV AND RISE OF PREIERM &KW MIESFRO31 /1 1 5 DAR .517 111 ,Speaker Emile Papiernik, Paris, France

4:30 p.m. D'alIE ON 771E WEST LOS ANCIELES PRE11117RE PREIEVIRAV PRQIECT Speaker Cal Ilobel, Los Angeles. California

5:00 p.m. DISCI WON

5:30 p.m. AI )JOI 'RN

260 2 4 I Alpendix Appendices

7:00 p.m. DINNER SEIIPOVIIM P4R77CIIMNIN AM) US. PUBLICHEAL77I SERIICE EVERT PANEL ON 77IE COVENT OF PRENA7A1. C4RE hroSt: Ruby Heam, Robert WoodJohnson Foundaticm

Tuesday, Mqy 10,1988

7:30 a.m. BREAKFAST

8:30-10;00 a.m. SESSION VI: INTERVENTIONS Presiders: Sara Depersio; Michael Ross;Dyanne Affanso Rappodeur:Jean Konte

8;30 a.m. PIE Sat M C4ROLIN4 RANIXWIZEDOINICAL TRIAL USING NURSE MIDIVIM 7t) REIN 7CE Low' &MI Wiic;iris Veaker Henry Heins, Charleston, SouthCarolina

RF.111tE ruE 9:00 a.m. REst EIS FROM A R&M AAL PR(X,RAMBASED ON RISK ASSESMIENT kim oF PRE7ERM DWIERY Veaker Paul Meis, Winstcm-Salem, NorthCarolina

9:30 a.m. Asa xsioN

10:00 a.m. NimmoN AND FrrNEss BREAK

10:30 a.m. SINSION VII: I N1ERVENTIONS 12:30 p.m. Presiders: John Kennel; David Olds; Diedre13I:mk Rapporteur: Ruth Merkatz

CoNIROLLED 10:30 a.m. DIE FAMILY WORKERS PROIELT FINALEIALI ;AMA OF A kiNDOMED DeLAL OF 77 IE PRol1sION OF A SOC;IAL Si PA NTSERI ICE DI VAG PREaVAA0 Veaker Brenda Spencer. Manchester,United Kingdom

11:00 a.m. SK.nl. SI l'PORT Di "RING PREGNANo` Veaker Lynda Rajan, London, UnitedKingdom

26/ Symp. Wrim Prognun 2 4 ') Advances in the Prevention of Low Hirtbuteight

11:30 a.m. PRET EV770N OFIIRE7Eat Danl.:MKS' BF HOME V1S177AGAlownks: Rains OF A FKENtli R1NDOMIZFD Coymormn speaker: Beatrice Blondel, Paris, France

12:00 noon DtsciMION

12:30-2:00 p.m. Lxcii

2:00-3:30 p.m. SESSION VIII: INTERVENTIONS Presiders: Steven Caritis: Leslie Cooper; Lynda Rajan Rapportenr: Richard David

2:00 p.m. LV7717dEvnavs. RAG PREUVAM1' Speaker Mary Sexton, Baltintore, Maryland

2:30 p.m. REct ns OF.-1 MIlnamER RIVM/MIR) CLINIC-1L ML-11.OF 01117C.11. CER0AGE ,Speaker Adrian Grant, Oxford, United kingdom

3:00 p.m. DISCI wax

3:30 p.m. NITRMON AND FUNESS BREAK

4:00-5:30 Rm. SESSION IX: INTERVENTIONS Presiders: I)enise Main; Robert Romero; Milton Lee Rapporteltrc: Ann 1-1(),ckett: Nancy Nance

8:30 a.m. (NE OF ANIIPLVELET DI/RUT MR 7711: PRI:ITN/70N CE,IVM`CEI) PRIM:RM DEL117111ES iw PRE1EVI7ON OF PREFC.7.'IMINIA Veaker Serge t zan, Paris, France

4:20 p.m. CALCIllt S( P1'LEMLN7A170N 7t) REIN a PIHAND PN77:11.11 DELI1TRY Speaker: j()se Villar, Bethesda, Nlaryland

4:40 p.m. 1,1,161vEsit 31 S IPLEMIEAT177M A PREGNAVO A Doi .BLE-BLAD Sfil/1 Speaker Ludwig Spat:ding. Herne. West Germany

5:00 p.m. Dm :vim

5:30 p.m. AQIOURN

262 2 J Appei 1 Appendices

8:00 p.m. DINNER SKI/PM/11! P4h71C/MAIN AM) U.S. Pt WICMUM SIUMCE EVERT PIXEL ON nx CONMV1* OF PRENVAL CU& Host: Richard Morton, M.D., Marchof Dimes Birth Defects Foundation

Wednesday May 11 1988

7:30 a.m. BREAKFAST 8:30-10:00 a.m. SESSION X: INnIAENTIONS Presiders: George Rhoades; John Hauth: TarigSiddigi Rappwiews: Rose Bemis; DonaldMeNellis

8:30 aan. A kLYDGMIZED CLI.VICAL ALAI, OF 711EMIAMI/NT OF URIAPLASIL4 C0IDNIZTI1ON AM) PRP:7E1M DELIVERY Speaker Davkl Eschenhach, Seattle, Washington

8:50 a.m. DISO

9:05 a.m. S(X1,11. St PIVRT ARM; PRE6MAC);AV 014:R1 IEW O Cumout TRL-tis Speaker: Diana Elbourne, Oxfcgd, United Kingd(nri

9:25 a.m. IIBMOV OF PROW 114Bol?IS IT 1170R1 1 MIME? Veaker Marc Keirse, Leiden, Netherlands

9:15 a.m, D s (.1 Way

10:00 a.m. NI-MUM AND FITNESS BMA':

10:30 a.m. SESSION Xl: FUTURE DiRamoNs 12:30 p Presiders: Woodie Kessel: Sumner Yaffe: Heinz13erendes Rapporteu rs: Godfrey Oakley: RobertKliegman

10:30 a.m. D/REcnms toR Ft IER RESE-IROl Panel Moderator: Richard Behrman, Cleveland,Ohio Pane/ Members: Adrian Grant Michael Kramer Jeanne Brooks-Gunn Miles Novy Emile Papiernik Irwin Merkatz Mort Rosen Peter Nathanielsz

Symposia nz Progra m 263 244 Appendix 11: List of Panictxints

titt.'VCN. (:aritis. MA), Diana Elboume, Ph,D. Dyannv Alfonso. R.N., Ph.D. National Perinatal Epidemitik)gy ['nit Pniessor. of Nursing Assiwiate Professor Deparunent of Obstetrics and Radclitk Infirmary Department of Muni ly Nursing Oxford 0X2 01E. United Kingdom University of California San Francisco Gynecology Niagee-Women's Ilospital Mail Stop N-111V David A. Eschenbach, M.D. S;in Frandsco, CA 9-4143 Forbes and Ilalket Street Pinstnirgh, PA 15213 Professor of Obstetrics and Gynecolcigy Departnwrit of Obstetrics and Duane Alexander. MD Charkine Cat/. M.D. Gynecology DIM: ICH% Nations! Institute of Child Iniversity of Washington Health and Human Devehipment Chif. Pregnancy and Perinatology National Institutes of Health Branch Seattle, WA 98195 Center for Research for Mothers and Building 31. It )(nit 2A03 limy Foster, M.1), Bethesda, MD 20892 Children Professor and Chairman Natkinal Institutes of Ilealth Department of Obstetrk-s and Executive Plaza North. Room 0,13 Richard liehrtnan. Gyneci)logy Dean, Schinil of Medicine Bethesda, MD 20892 cast. kx!estern Re,eree I'niversity Meharry Medical College 1005 Todd Boukvard 2119 Ahington Road Leslie Cooper. RN.. M.P.H. Nashville, TN.r208-3iv9 Clevelimd. 011 .14100 Epidemioloa Branch Prevention Research Program Ntargaret Freda. Ph.D. 105e M. Relii.an, M.D. Nati(mal Institute cif (hild Ilealth and Assistant Pnitesstir Centro Ro.arino de Estudios Perinatales 1)evehipment Department of 01)stetrics and Br. Orono 5000 Executive Plaza North, Room Oatt tlyttecculogy 2t.KRI Rosario. Argentina Bethesda, MD 20892 Anvil Einstein College of Medicine Dirccuir of Ediicituin and Oimmunity Rose Bemis. R.N.. M.S. Robert K. Creasy. M.P. Affairs tor PROPP Division i)f Maternal Fetal Nledicine Chairman, Department of Obstetrics and South Nurses Residence 1300 Nforris t'CLA \Wylie:it Center. Building I. clynecology Psrk Avenue Bronx, NY 10401 1124 Wem Carson Stwet Reproductive lk.alth Torrance. CA 90504 l'niversity of Texas Medk.al School Adrian (rant. N1.1). ot31 Fannin Street. Risint 32-0 National Perinatal Epidemiology Heinz W. Berem.k.s. M.D.. M.11.S. louston, 'IN 77030 Radcliffe Infirmary Direct(ir. Pre% ention Research Migrant Oxford ON2l IF. rnited Kingdom NatknLd Institute (if Child liealth and Karla Dainus. Ph.D. !Ionian Devel(wwnt Assistant Profe.ssor. Department of )(mini !Will, M.D. National Institutes of I lealth DB .GYIN Chairman, .kmerican College Executive Plaza Ninth, 11(sn'n 610 Albert Einstein College (it Ntedicine )bstetrics and Gynecology BCtittlida. MD 20892 1)iteckir (if Reseatch Epidemiokigy Maternal 'Fetal Medicine t'ainuitittee Bureau of ',Maternity Services andFamily Department of Obstetrk.s and Diedre [flank. U.N. , Ph.D. Disease Prevention Planning Gynt'Culogy Ikmith Pumnition 'niversitv of Alabama National t'xnter fin. Nursing Research New l'ink City Department (if fie:dill PROPP 'niyersity SLIM m National Institutes of I lealth Birmingham. Al. 3529i Bldg 38A. Room B2E17 South Nurses Residence Bethesda MD. 20892 130) Morris Park Avenue Bronx. NY 10461 Ruby Hearn, Phi). Vice president Beatrice Blondel. NED. Robert \\A).1)(1.1nlinson Finindation o Dr. G. Breart Richard C. David Founder. Iiikos Nledical Corpiiratain Colkge Road INSERM Princeton. NI 08i.i0 123, Ixf de P(irt royal 182.1 East Dyer Road. Suite 21K) Santa Ana. CA 9270i 7501-4 l'aris, France Henry Ileins. Nara I hpvlsif Profess( ir (erard Brean, M.1). 1)epartment of Obstetrics and (-(ynecoli)gy de Recherches epidemi- Medical Director for MCI1 Syr% ices )1dah(nna Dept of Ilealth Medical I'Mversity of South Carolina olergiques stir la Ntere ci 12Entant Charleston, SC 29425 INSERM 1000 N.E. loth Street on 703 123. Kt de Port Royal Oklahoma City. OK, 7.4152 Calvin J.I lohel, NED. 7501-i Paris, France Gian Carlos Di Rento, M.D. Professor. Division of Maternal and Fetal Cattedra di Purricultura Prenataic Medicine Jeanne Bro4iks-Gunn. Ph.D. Ilarbor General licispital Educational Testing Service Institut() di Ginecologia ed )steincia Monteluce 1000 West Carson Street Iiiisedale Drive Torrance. CA 9)509 Princeton. NI °Mil 1-06100 Perugia. Italy Advances in the Prevention of Low Birthweight

Anne Hocken, M.S. Ann Koontz. CNA_ 1)r.P,11, Irwin Merkatz, M.D. Infant Mortality Commimion Chief. Maternal and Infant Health Branch Professor and Chairman Switzer Building. Room 2000 Division of Maternal. Child. and Infant Department of Obstetrics and 330 C Street. S.W. Ilealth Gynecok)gy 1X7ashington, DC 20201 Maternal and Child Health Bureau Albert Einstein College Of Medicine Parklawn Building, Room 6-19 1300 Morris Park Avenue Howard Hoffinan. M.A. 50X) Fishers Line Bronx, NI' 10461 Chief. Biometry Branch Rockvilk., MD 208i7 Prevention Research Program Richard F. Morton. M.D. National Institute of Chikl Ikalth and Michael Kramer. M.D. Vice President for Medical Services Iluman Development Associate Professor of Pediatrics March of Dimes Birth Defects Executive Plaza North, Room 030 McGill University Faculty of Medicine Foundation Bethesda, MD 20892 1020 Pine Avenue West 1275 Mamanmeck Avenue Montreal. Quebvc. Canada 113A1A2 White Plains, NY 10(115 Vince L. Hutchins. M.D., MAUI. Associate Director Nlilton Lee. M.D. Eberhard Muelkr-lieubach, MD, Maternal and Child Health Bureau chief of Obstetrics Ass(s.late Pnifessor of Obstetrics and Parklawn Buikling. Roont 5(100 Department of ()tweaks and C,yneeology Fishers Lane Rockville. MD 20857 Gynecokigy 'niversity of Pittsburgh Drew Medical Sdlool MaCiee Wimiensliospital Mare J.N.C. Keirse, M.D, 12021 S. Wilminoin Ave, Forbes and llalket Streets Depannwnt of Obstetrics Los Angeles. CA 90059 Leiden t 7niversity Pittsburgh. PA 15213 Rijnsburgerweg 10 Brian McCarthy, M.D. Nancy Nance, 2333 AA Leiden Pregnancy Epidemiology Branch The Netherlands Division of Reproductive Ilealth Department of Obstetrics & Gynecoliigv Centeri tOr Disease Control Medical I 'niversity of soutliCandina Charleston, SC 29425 John H. Kennel, M.D. Building 1. Room +119 WO Professor of Pediatrics 1(1(X) Clifton Road, N.E. Peter Nathanielsz. Chief. Child Development Divisiim Atlanta. GA 30333 Case Western Reserw 1.raversity Chief of Repoiductive Sciences Sdlool of Medicine Marie McCormick, M.D. Cornell I 'niversitr Rainbow Babies and Childrens Iluspital Joint Pnigram and Neonatology 2074 Abington Road Brigham & \\Omen's !hospital Steven Ng. M.D. Cleveland, 011+1106 -7S Francis Street Assistant Po)lessor Boston. MA 0211i C,ennide Sergievsky Center Wcxxlie Kessel, M.D., M.P.H. 03o 1O8th Street Director. Division of MCH Pnigram Donald McNellis. M.D. New. York, NY 10032 Coordination and Systems Center for Research k r Mt )thers and Dcreli,pinent Children Miles Novy, M.D. Maternal and Chikl !health Bureau Ithaca. NY 31(r Southwest Nottingham Drive Parklawn Building. Room (1-17 5600 National Institutes of Ilealth New York. NY I(X.)32 Fishers Lane Rockville. MD 20857 Executive Plaza North. Room o.+3 llethesda. MD 20892 Ann Oakley. X.Yz. Marshall Klaus, M.D. "Uinta. Cc wam Research I 'nit Children's Ilospital, Oakland Denise Main, M.D. 41 Brunswick Square 747 52nd Street Department of Ohstetrics and London WC1N 1AZ Oakland, CA 94609 Gynec()logy 'nited Kingdont l'niversity of California. San Francisco Robert Khcgman, P.O. Box (1132 (;odl-rey Oakley, 51.1). Vice Chairman of Education San Francisco, CA 9i1,13 Direckir. Division of Birth D...fects and Rainbow Babies Si Children's flospital Deyelipmental 2101 Adelbert Road Rcx.nn 60 Paul Meis, M.D. Center for Envinninent.d I health and Cleveland. OH 44100 Departnwnt of Obstetrics and Injury Control Gynee(slogy Centers for Disease Contn)I Jean Konte, R.N., M.S. Bowman Gray Sclusil of Medicine Kiiger Center. Rismi 2031 Coordinator ot North Cciast Perinatal 3(X) South Ilawthorne Road 1000 Clifton Road Pn)tect U'inston-Sakm, NC 2-103 Atl.tnta, (.1.N 30333 Department of Obstetrics and Gynecology and Reproductive Sc.-lent es Ruth Merkatz, R.s.M.s. David Olds, Ph.D. 'niversity of California. SF Allwrt Einstein College Ot Medicine 1)epartment of Pvdiatrics P.O. Box 0982 13c)0 Morris Park Avenue t 'niversity of Michester San Franciseo, CA 941-13 Bnmx. NY 10101 Riklicster, NY 110.12

266 2.1i; ApperldixII Appendices

Emile Papiernik. Tariq Siddiqi M.D. Professor of Obstetrics and Gynecok )gy Depannent of Obstetrics Ifunun and Repnx.luctive Research College of Medicine. Ilospiul Antoine Bedew I.'niyersity of Cincinnati 157, Rue de la Porte de Trivaux 231 Bethesda Avenue 921,40 Clamart, France Cincinnati 011 45207-05.+1

Lynda Rajan. M.D. Ludwig Spaetling, Thomas Comm Research I 'nit 'fflyersitats-Frauenklinik Bochum University of London Institute of Ma rienhe )spital Educatkin I it Akeskampring -40 41 Brunswick Square D .4090 Iferne I London WC1N1AZ, I'Mted Kingdom Federal Republic of Germany

George Rhoads. M.D. Brenda Spencer. Ph.D. Chief, Epidoniohigy Branch Department of Epklemiolt)gy and Social Prevention Research Prt)gram Researdi Nationa1 Institute of Child Ilealth at;d 'niversity Ilospital of South Manchester Ifuman Development Kinnaird Road Executive Plaza Ncirth, Rot-nu (40 N1anchester N1209y1.. I nited Kingdom Bethesda, MI) 20892 Serge l'zan. M.1). Roberto Ronwro M.D. f lead of the Obstetnc staff ale Prenatal Testing Center Maternity Guy Le Loner 03 York Street hipital Tentin New Ilaven, CI. 00510 .4 rue de la Chine 75020 Paris. France Monimer Rtrien, NM, Professor and Chairman Jose Villar. Department of Obstetrics and Visiting Scientist Gynecoltigy Prevention Research Pn)gr,:-,, Columbia Presbyterian Medical Center Nalitnul Institute of Child I lealth and 022 West 108th Street Itiman Development New York, NY 10032 National Institutes of I lealth Executive Plaza North, lit min 040 Nfichael Ross. M.D. liethesda. MD 20s92 Department of 011stetrics and Gynecokigy Sumner Yaffe. M.D. Harbin' General Hospital Director, Center for Research UCIA Medical School Ntc idlers and Chiklren 1000 \Xest Cas( in Street National Institute of Child I lealth and Torrance, CA 90509 I finnan Development National Instill ne fI fealth David Rush, \El), Executive Pla/a North. Room Department of Pediatrics and Gynecology Ikthesda, 20892 Albert Einstein College of Medicine LINE Rose. Kennedy Center Phyllis Zucker, MS 1410 Pelham Parkway Director, Divisitm t)t Policy Analvs s Bronx, NY 10401 Oftic e of I fealth Plannnig and Evaluation Mary Sexton, M.P.H. Office of the Assistallt Secretary lOr Department of Epidemiol()gy fealth 1 lumphrey Huikling. Rm AK; 'niversity N1ary land School ot !filbert If. Preventive Medicine 2a 11 ndependent (' Avenue. SW 655 \X'. Baltimore Street Washington. IX: 20201 Baltimore, MD 21201

Patricia shit nio. Ph.D. Epidemiok igy Branch Prevention Research Pnigrain Nati( mai Insiitutv of Child Ilealth and !Ionian Develripment Executive Plaia North. Room Ow Bethesda. MD 20592

267 List qf Pa rtk 247