Journal of Perinatology (2008) 28, 102–106 r 2008 Nature Publishing Group All rights reserved. 0743-8346/08 $30 www.nature.com/jp ORIGINAL ARTICLE Choice of antenatal testing significantly effects a patient’s work obligations

JM Denney1, TP Waters1, R Gorman2, M Pollock2 and AC Sciscione2 1Department of and Gynecology, Drexel University College of Medicine, Philadelphia, PA, USA and 2Christiana Health Care System, Department of Obstetrics and Gynecology, Newark, DE, USA

care in an efficient manner to fit our patients’ desires to remain Objective: We sought to compare two approaches to antenatal testing for productive, as only 27% chose to quit working during the course of their impact on the workforce. .1 Study Design: This is a prospective observational study of women who Surveillance of the fetus dates to 1818 when Mayor initiated the presented for antenatal testing. All women were given a survey upon practice of auscultation. In 1968, a large, federally funded study presentation. As per hospital protocol, nonstress testing (NST) was demonstrated a poor correlation between auscultation findings and performed semiweekly and biophysical profile (BPP) was performed neonatal outcome. Owing to the inaccuracy of auscultation, focus weekly. The choice of testing was determined by the attending physician. shifted from fetal auscultation to the use of phonocardiography w2- and Student’s t-tests were performed where appropriate. A P-value of and external fetal monitoring. In 1969, the availability of the <0.05 was considered significant. Hewlett-Packard external fetal monitoring/TM changed fetal monitoring and quickly became standard of care.2 To date, no Result: A total of 195 women were surveyed. Among them, 94 women randomized controlled trials have demonstrated the effectiveness of had an NST and 101 had a BPP. Overall, 59.2% were multiparous, 33.1% antepartum fetal surveillance in reducing fetal mortality. However, had to arrange for and 97.2% felt reassured by the testing. observational studies have correlated the (NST) with There were no differences in demographic characteristics, education, type outcomes, leading to the NST as a method of antepartum fetal of insurance or employment status between the groups. Women who had surveillance.3–9 NSTs were more likely to lose time from work than those who had BPPs Another method commonly employed to assess the fetal state is (218.4 versus 68.9 min; P<0.001). Of the women who had semiweekly the biophysical profile (BPP). The BPP combines the NST with NSTs, 80.6% would have preferred weekly testing. If the 94 women who four real-time sonographic observations: , fetal received semiweekly testing had weekly testing, a total of 534.4 h would tone, breathing motion by the fetus and amniotic fluid volume. have been available for the workforce. The use of these observations to gauge fetal well-being is supported Conclusion: Twice-weekly NST results in a significant increase in time by various observational studies.7,10–16,21–24 lost from the workforce compared with weekly BPP. Although simple and inexpensive, the weekly NST has not been Journal of Perinatology (2008) 28, 102–106; doi:10.1038/sj.jp.7211877; shown to reduce fetal mortality.17 In fact, equivalent rates of published online 29 November 2007 antenatal fetal death are demonstrated when comparing the weekly NST with no testing at all.17,18 Because performing weekly NST has Keywords: antenatal testing; biophysical profile (BPP); nonstress test (NST); work; jobs; workforce not shown any reduction in fetal mortality, NST has been utilized twice-weekly to increase likelihood of detecting fetal compromise.19 By shortening the interval of testing, the twice-weekly NST Introduction decreases the number of false-negatives in terms of identifying impending . Observational studies show that using Prior to 1978, over half of all women took a leave of absence from 19 their job upon learning they were pregnant.1 By the 1990s, 73% of the NST as a test twice a week reduces fetal mortality. F F women continued to work after a positive .1 As While both tests weekly BPP and twice-weekly NST reduce obstetricians, we face the pressure of providing adequate prenatal fetal mortality, no study has ever shown either test to be superior in reducing fetal mortality rates.3,4,19,25 When reassuring scores are Correspondence: Dr JM Denney, Department of Obstetrics and Gynecology, Drexel University obtainedF‘reactive NST’ and ‘8/8 BPP,’ both tests are reliable in College of Medicine, 2629 Brown St, APT 417, Philadelphia, PA 19130, USA. predicting fetal well-being (that is, they have low false-negative E-mail: [email protected] Received 23 April 2007; revised 10 October 2007; accepted 15 October 2007; published online rates of less than 0.5%). Likewise, the two tests correlate with 29 November 2007 one another on many levels. As per Vintzeleos et al. in 1983, antenatal testing choice effects patient’s work JM Denney et al 103

BPPs of at least 8/8 are found in 95% of fetuses with reactive NSTs. their occupation, which was their stated form of employment), All hypoxic fetuses had nonreactive NSTs and fetal breathing time to complete testing (defined as time of the appointment motion was absent. Presence of fetal rate accelerations for test being performed), arrangement of child care, method of and reactive NSTs were always associated with the presence of transportation, preferred frequency of testing if given a choice fetal movement and tone. However, when the NST is nonreactive and whether they returned to work after testing as well as scores and the BPP score is not reassuring, they both have high of pre- and post-anxiety. Anxiety was quantified by the following incidences of false-positives (approximately 50% for NST and scale: 0 ¼ no anxiety, 1 ¼ little anxiety, 2 ¼ some anxiety, high false-positive rates for each component of BPP).26 3 ¼ anxiety, 4 ¼ high anxiety and 5 ¼ most ever anxiety. The Notably, the studies cited by The American College of completed forms were collected by research staff. Univariate Obstetricians and Gynecologists (ACOG) and by this paper in the analysis was performed using the Student’s t-test, Mann–Whitney use of fetal testing are all underpowered, limiting true comparison U-test, w2-test, Cochran–Mantel–Haenszel mean scores test or of the two testing formats. Nonetheless, both tests are often used the Fisher exact test where appropriate. Patients who had NSTs interchangeably, as there are no data existing to show one test were compared to those having BPPs. to be superior. At our institution, two acceptable testing protocols are used: the weekly BPPs and the twice-weekly NSTs. We employ these two tests and respective frequencies, because, in the Results limited data from underpowered studies available to base our A total of 195 women were surveyed. Among them, 94 patients practice, they have both been shown to be equivalent in reducing (48%) had NSTs and 101 patients (52%) had BPPs as their mode 19,20 fetal mortality at the frequencies we use. of testing. No intrauterine fetal demises were noted on follow-up Prior to this study, patient preference and the demands of choice in either group. Our data were assessed as exhibiting a normal of testing upon a patient have not been considered. Therefore, we distribution. Table 1 presents summary characteristics for the sought to evaluate the choice of antenatal testing for its impact on entire study cohort. Table 2 shows maternal demographic and a patient’s social and work obligations. Additionally, we evaluated employment characteristics by type of testing. There were no patient preferences regarding their method of antenatal testing. statistically significant differences in demographic characteristics, education or type of insurance. Although there were statistically significant differences between the two groups in terms of full-time Methods (P ¼ 0.01) or part-time employment (P ¼ 0.01), there was no After approval by the Christiana Hospital institutional review board and human experimentation committee, all patients referred to our Table 1 Patient sociodemographic characteristics antenatal testing unit were approached for participation. Patients Variable n ¼ 195 were enrolled in one of the hospital’s two accepted antenatal testing protocols: twice-weekly NSTs or BPPs. Method of testing was at the Age (years) (mean) 31 discretion of their physician. The primary outcome for this Nulliparous 42% investigation was the time requirements of NST versus BPP on Multiparous 56% patients presenting for routine antenatal testing. Secondary Miscarriage or 37% outcomes related to how time from work was made up (for Work 67% example, not made up, lost lunch, came to work early or stayed Full-time parent 44% late, lost vacation time, lost sick time or other) and the impact of Married 77% Returned to work after test 17% testing on maternal anxiety. A power calculation based on the Arranged for child care 27% assumption to detect a 15% difference in comparison of time Completed high school 23% commitment revealed that 90 patients in each group (NST and Advanced degree 38% BPP) were required to achieve a P-value of 0.05 and b of 0.2. Came from home to testing 74% Similarly, a power calculation based on the assumption to detect a Drove to testing 96% 15% difference in comparison of patient anxiety revealed that 90 Time to get to test (mean) 19 min per patients in each group (NST and BPP) were required to achieve a patient P-value of 0.05 and b of 0.2. Time from work (mean) 84 min per After appropriate counseling, informed consent was obtained. patient Each patient who elected to participate was given a survey to Time of child care needed (mean) 101 min per complete. This survey included the following: maternal patient demographic characteristics, type of antenatal testing, time taken Time of test (includes wait time, time for test performance, and 58 min per time spent scheduling next appointment; i.e., total time) patient from work (defined as time spent away from scheduled shifts for

Journal of Perinatology antenatal testing choice effects patient’s work JM Denney et al 104

Table 2 Maternal characteristics by type of testing Table 4 How time missed from work was made up by patients regardless of testing protocol Variable NST (n ¼ 94) BPP (n ¼ 101) P-valuea How was time from work made up? n ¼ 89 Age (years) (mean) 30.7 31.9 0.23 Nulliparous 41 (43.6%) 41 (40.6%) 0.19 No time made up 33% Multiparous 55 (58.5%) 63 (62.4%) 0.31 Lunch time 7% History of miscarriage or abortion 39 (41.5%) 33 (32.7%) 0.10 Came in early or stayed late 18% Work outside home 60 (66%) 75 (75%) 0.52 Vacation time 2% Full time (of those who work) 43 (46%) 58 (58%) 0.01 Sick time 6% Part-time (of those who work) 5 (5%) 17 (17%) 0.01 Other 34% Home business (of those who work) 9 (10%) 8 (8%) 0.67

Abbreviations: BPP, biophysical profile; NST, nonstress test. Table 5 Important testing qualities to patients, recorded by numbers of patients aBy w2-test. Factor Very important Not important

Table 3 Effects on work and patient opinion/preference by type of testing Accuracy 181 5 Time to complete test 14 49 Variable NST (n ¼ 94) BPP (n ¼ 101) Reasurance 87 2 Leave work for test 21 (22.3%) 22 (21.8%) Frequency 24 23 Lost wages 5 (5.3%) 4 (4%) Cost of test 6 74 Arrange child care 24 (25.5%) 31 (31%) Cost to patient 9 124 Prefer weekly testing* 37 (39.4%)* 83 (82.1%)* None of the above values varied by testing protocol. The only values recorded from the Prefer more than weekly* 43 (45.7%)* 11 (11%)* surveys and included in the analysis were qualities patients felt were ‘very important’ or Is test time important (yes) 33 (35.1%) 30 (30%) ‘not important.’ If you have two or more NST per week, do 60 (83%) N/A you prefer weekly? Reassured by test 84 (89%) 92 (91%) of testing on time lost from work, we evaluated the amount of time Time to get to test 18 min 20 min that would be gained if a patient who was evaluated by a twice- Time from work 82 min 87 min weekly NST was instead evaluated by weekly BPP. In our analysis, Time of child care needed 118 87 if the 96 women who received semiweekly testing had weekly Time of test 57 58 testing, a total of 534.4 h would have been available for the Hours from work versus no. test 83 min* 166 min* workforce. In short, those enrolled in twice-weekly NSTs had both per week* more time in minutes lost from work and more episodes of Total time missed* 218 min* 103 min* absences from work. Weekly testing 16 97 For patients who had to leave their job to attend testing, we Twice-weekly testing 60 6 asked how their time away from work was covered. Table 4 presents Three or more tests per week 19 1 a summary of these results. While the majority of patients did not Abbreviations: BPP, biophysical profile; NST, nonstress test. have to make up their time, a number of patients had to make *P<0.05. arrangements to attend their testing, which included losing vacation time, lunch or sick time. We also asked patients if their statistically significant difference in terms of those whose work was testing had a negative impact on their lives. Nine patients (5%) outside the home (P ¼ 0.52). There was also no significant reported that they lost wages to attend their testing, and four difference in women working in a home business (P ¼ 0.68). patients stated their testing compromised their work and threatened Overall, 67% of patients worked and 17% of patients returned to their job. Twelve patients (6%) stated that they had to skip a work once their testing was complete. A total of 44% stated they meal to have their testing. were a full-time parent and 27% of patients had to arrange for One hundred and eighty one (90.5%) and eighty-seven (43.5%) child care to attend their testing. patients found accuracy and reassurance, respectively, very Table 3 shows the impact of testing on the workplace. Overall, important. Only 3% (n ¼ 6) and 4.5% (n ¼ 9) found cost of women who had twice-weekly NST lost more time from work than the test and cost of the test to the patient to be very important, those who had a BPP (218.4 versus 68.9 min; P<0.001). Notably, while 37% (n ¼ 74) and 62% (n ¼ 124) found the same qualities time listed is a mean time in minutes lost from work rather than not important. Similarly, more found duration of testing not to time in minutes to actually perform the test. Mean values were be important rather than very important (that is, n ¼ 49 versus used due to a normal distribution. To ascertain the impact of mode n ¼ 14, respectively), as per Table 5. When questioned about

Journal of Perinatology antenatal testing choice effects patient’s work JM Denney et al 105

Table 6 Patient anxiety (both arms of the testing protocols as a collective preference toward an equivalent weekly test. As evidenced by this group) study, scheduling different testing protocols have differing effects Level of anxiety from 0 to 5 012345 and demands on the lives of patients involved. In our group, where 0 ¼ none, 1 ¼ little compliance was not an issue. However, if patients were to possess anxiety, 2 ¼ some anxiety, knowledge that two testing protocols are equivalent in terms 3 ¼ anxiety, 4 ¼ severe anxiety of identifying a distressed fetus but differ drastically in time and 5 ¼ most anxiety ever commitments, it appears clear that they would choose the testing protocol that required the least amount of time commitment Pretest anxiety (n 196) 48 49 28 48 19 4 ¼ on their end. Post-test anxiety (n ¼ 187) 94 56 21 11 5 0 The generalizability of this study is limited because many testing centers perform the ‘modified BPP’Fconsisting of an NST pre- and post-test anxiety, patients reported lower anxiety levels with the amniotic fluid index. The modified BPP is utilized for after testing (P<0.0001), as reported in Table 6. Finally, we asked several advantages, including time, accuracy and the ability to patients if they could choose their method of testing or frequency approximate the BPPs. If the modified BPP is abnormal, a backup of testing, what method would they prefer. Overall, 120 patients test is performed, typically the remainder of the BPP. Additionally, (61%) stated they would prefer weekly testing over twice-weekly the study is limited by the lack of adequately powered, randomized evaluations. For patients who were having twice-weekly testing, controlled trials comparing NSTs versus BPPs. The American 80% reported they would prefer only weekly testing. College of Obstetricians and Gynecologists does not take a on the frequency of either test NST or BPP. Furthermore, in the most recent practice bulletin, the American College of Obstetricians Discussion and Gynecologists provides no guidelines regarding frequency The results of this study demonstrate that many obstetrical patients of antenatal testing, and offers no comments regarding existing receiving antenatal testing experience consequences outside of the literature on antenatal testing protocols and effects on fetal pregnancy due to required testing. We demonstrate the need to mortality. The American College of Obstetricians and Gynecologists arrange transportation and child care, loss of time at work, wages, states that either testFNST or BPPFis an acceptable test to vacation time, sick time, lunch breaks and additional hours implement. Owing to the lack of adequately powered trials, we required either before or after work to meet the obligations of would like to call for a large, multicenter, prospective study that their antenatal appointments. While both forms of testing placed could be done by the Maternal Fetal Medicine Units, for example, demands upon our patients, in our evaluation, weekly BPP was to address optimal antenatal testing. far less taxing on a patient’s time commitment. This study also Currently, the individual practitioner is left to decide what demonstrates that patients who had twice-weekly NSTs would conditions to institute fetal testing protocols, when to start testing have preferred weekly BPPs. as well as the frequency of testing.3 Testing protocols may vary Our patients face the pressure of balancing work as well as widely among individual practitioners, institutions and different taking steps to reassure themselves and their physicians that their regions of the country, depending on training, medical–legal unborn child is not compromised and in need of delivery. When environments and individual/institutional perspectives on existing the NST is done twice a week, a reduction of fetal mortality is literature, as many studies contradict one another.4 In light of accomplished at a rate similar to that when weekly BPP is used the effects and stress on our patients, we might consider utilizing (that is, there is no statistically significant difference in fetal the weekly BPPs rather than twice-weekly NSTs when medical mortality from weekly BPP when NST is done twice-weekly). Given and/or obstetrical issues indicate a need for fetal surveillance, that 83% of the patients having twice-weekly NST had preferred an until further clinical evidence demonstrates a difference between equivalent weekly test, consideration should be given to patient these two protocols. desires when we send expectant mothers to testing. Using weekly BPP would require less time taken from the workforce. From the patient’s perspective, the most important testing qualities are References accuracy and reassurance, while time and cost are not important. 1 Armour S. Pregnant workers report growing discrimination. USA Today, 2/16/05 Patient anxiety levels decreased following antenatal testing. www.usatoday.com/money/workplace/2005-02-16.htm. Two hundred and eighteen minutes in the twice-weekly NST 2 McCartney PR, Schmidt JV. Intrapartum fetal monitoring: a historical perspective. group versus 103 min in the weekly BPP group in regard to total Fetal Heart Monitoring Principles and Practice, 3rd edn. 2003; 3–22. 3 American College of Obstetrics and Gynecology. ACOG practice bulletin: antepartum missed time per week shows that use of twice-weekly NST results fetal surveillance, no. 9, 1999; Compendium 2007; II: pp 503–513. in more than doubling the amount of time spent away from work. 4 Devoe LD, Jones CR. Nonstress test: evidence-based use in high-risk pregnancy. Of those enduring the twice-weekly testing, 83% reported a Clin Obstet Gynecol 2002; 45(4): 986–992.

Journal of Perinatology antenatal testing choice effects patient’s work JM Denney et al 106

5 Lee CY, DiLoreto PC, O’Lane JM. A study of fetal heart rate acceleration patterns. 16 Manning FA, Platt LD, Sipos L. Antepartum fetal evaluation: development of a fetal Obstet Gynecol 1975; 45: 142–146. biophysical profile. Am J Obstet Gynecol 1980; 136: 787–795. 6 Powell OH, Melville A, MacKenna JL. Fetal heart rate acceleration in labor: excellent 17 Lumly J, Lester A, Anderson I, Renou P, Wood C. A randomized trial of weekly prognostic indicator. Am J Obstet Gynecol 1979; 134: 36. cardiotocography in high-risk obstetrical patients. BJOG 1983; 90: 1018–1026. 7 Baser I, Johnson TR, Paine LL. Coupling of fetal movement and fetal heart rate 18 Kidd LC, Patel N, Smith R. Non-stress cardiotocographyFa prospective randomized accelerations as an indicator of fetal health. Obstet Gynecol 1992; 80:1. clinical trial. BJOG 1985; 92: 1156–1159. 8 Rochard F, Schifrin BS, Goupil F, Legrand H, Blottiere J, Sureau C. Nonstressed 19 Boehm FH. Improved outcome of twice weekly nonstress testing. Obstet Gynecol 1986; fetal heart rate monitoring in the antepartum period. Am J Obstet Gynecol 1976; 67(4): 566–568. 126(6): 699–706. 20 Platt LD, Walla CA, Paul RH, Trujillo ME, Loesser CV, Jacobs ND et al. A prospective 9 Murata Y, Martin CB, Ikenoue T, Hashimoto T, Taira S, Sagawa T et al. Fetal heart trial of the fetal biophysical profile versus the nonstress test in the management of rate accelerations and late decelerations during the course of intrauterine death in high-risk . Am J Obstet Gynecol 1985; 153: 624–633. chronically catheterized rhesus monkeys. Am J Obstet Gynecol 1982; 144: 218. 21 Manning FA, Snijders RL, Harman CR, Nicolaides K, Menticoglou S, Morrison I. Fetal 10 Dawes GS, Fox HE, Leduc BM, Liggins GC, Richards RT. Respiratory movements and biophysical profile score. VI: correlation with antepartum umbilical venous pH. Am J paradoxical sleep in foetal lambs. J Physiol 1970; 210: 77. Obstet Gynecol 1993; 169: 755–763. 11 Boddy K, Dawes GS, Fisher R, Pinter S, Robinson JS. Foetal respiratory movements, 22 Bocking AD, Gagnon R, Milne KM, White SE. Behavioral activity during prolonged electrocortical and cardiovascular responses to hypoxemia and hypercapnia in sheep. hypoxemia in fetal sheep. J App Physiol 1988; 65: 2420–2426. J Physiol 1974; 243: 599–618. 23 Manning FA, Baskett TF, Morrison I, Lange I. Fetal biophysical profile scoring: a 12 Devoe LD, Abduljabbar H, Carmichael L, Probert C, Patrick J. The effects of maternal prospective study in 1184 high-risk patients. Am J Obstet Gynecol 1981; 140: 525–530. hyperoxia on fetal breathing movements in third-trimester pregnancies. Am J Obstet 24 Manning FA, Morrison I, Lange IR, Harman CR, Chamberlain PF. Fetal assessment Gynecol 1984; 148: 790. based on fetal biophysical profile scoring: experience in 12,620 referred high-risk 13 Natale R, Clewlow F, Dawes GS. Measurement of fetal forelimb movements in lambs pregnancies. I. Perinatal mortality by frequency and etiology. Am J Obstet Gynecol in utero. Am J Obstet Gynecol 1981; 158: 545. 1985; 151: 343–350. 14 Manning FA, Platt LD, Sipos L. Fetal movements in human pregnancy in the third 25 Manning FA, Morrison I, Lange IR, Harman CR, Chamberlain PFC. Fetal biophysical trimester. Obstet Gynecol 1979; 54: 699. profile: selective use of the nonstress test. Am J Obstet Gynecol 1987; 56: 709–712. 15 Cohn HE, Sacks EJ, Heyman MA, Rudolph AM. Cardiovascular responses to hypoxemia 26 Vintzileos AM, Campbell WA, Ingardia CJ, Nochimson DJ. The fetal biophysical profile and acidemia in fetal lambs. Am J Obstet Gynecol 1974; 120: 817–824. and its predictive value. Obstet Gynecol 1983; 62: 271–278.

Journal of Perinatology