Medication-Assisted Treatment with Methadone: Assessing the Evidence Catherine Anne Fullerton, M.D., M.P.H

Home , Methadone maintenance

Assessing the Evidence Base Series Medication-Assisted Treatment With Methadone: Assessing the Evidence Catherine Anne Fullerton, M.D., M.P.H. Richard H. Dougherty, Ph.D. Meelee Kim, M.A. Allen S. Daniels, Ed.D. Cindy Parks Thomas, Ph.D. Sushmita Shoma Ghose, Ph.D. D. Russell Lyman, Ph.D. Miriam E. Delphin-Rittmon, Ph.D. Leslie B. Montejano, M.A., C.C.R.P.

Objective: Detoxification followed by abstinence has shown little success pioid dependence is a serious in reducing illicit opioid use. Methadone maintenance treatment (MMT) public health concern. In the helps individuals with an opioid use disorder abstain from or decrease O United States, approximately use of illegal or nonmedical opiates. This review examined evidence for 800,000 individuals are heroin de- MMT’s effectiveness. Methods: Authors reviewed meta-analyses, sys- pendent (1), and 1.7 million report tematic reviews, and individual studies of MMT from 1995 through 2012. a substance use disorder involving Databases searched were PubMed, PsycINFO, Applied Social Sciences prescription opioids (2). Opioid de- Index and Abstracts, Sociological Abstracts, Social Services Abstracts, pendence is associated with prema- and Published International Literature on Traumatic Stress. The authors ture mortality, criminality, violence, rated the level of evidence (high, moderate, and low) based on bench- suicide, HIV and hepatitis C infec- marks for the number of studies and quality of their methodology. They tion, and poor quality of life (3,4). also described the evidence of service effectiveness and examined ma- Detoxification followed by abstinence- ternal and fetal results of MMT for pregnant women. Results: The review oriented treatments has shown little included seven randomized controlled trials and two quasi-experimental success in curtailing illicit opioid use studies of MMT, indicating a high level of evidence for the positive im- over time (5,6). Methadone, an opioid pact of MMT on treatment retention and illicit opioid use, particularly at agonist, and buprenorphine, a partial doses greater than 60 mg. Evidence suggests positive impacts on drug- agonist, may be used in maintenance related HIV risk behaviors, mortality, and criminality. Meta-analyses treatment to improve treatment out- were difficult to perform or yielded nonsignificant results. Studies found comes. This review focused on metha- little association between MMT and sex-related HIV risk behaviors. done maintenance treatment (MMT); MMT in pregnancy was associated with improved maternal and fetal a companion review in this series ex- outcomes, and rates of neonatal abstinence syndrome were similar for amines buprenorphine (7). mothers receiving different doses. Reports of adverse events were also The Substance Abuse and Men- found. Conclusions: MMT is associated with improved outcomes for tal Health Services Administration individuals and pregnant women with opioid use disorders. MMT should (SAMHSA) describes medication- be a covered service available to all individuals. (Psychiatric Services 65: assisted treatment as a direct service 146–157, 2014; doi: 10.1176/appi.ps.201300235) that provides a person who has a substance use disorder or a mental disorder with pharmacotherapy in conjunction Dr. Fullerton and Ms. Montejano are with Truven Health Analytics, Cambridge, with behavioral therapies as treatment Massachusetts (e-mail: [email protected]). Ms. Kim and Dr. Thomas forassociatedsymptomsordisabil- are with the Heller School for Social Policy and Management, Brandeis University, ities. Treatment is individualized. Waltham, Massachusetts. Dr. Lyman and Dr. Dougherty are with DMA Health Medication-assisted treatment with Strategies, Lexington, Massachusetts. Dr. Daniels and Dr. Ghose are with Westat, methadone refers to the use of meth- Rockville, Maryland. Dr. Delphin-Rittmon is with the Office of Policy, Planning, and adone to treat individuals addicted to Innovation, Substance Abuse and Mental Health Services Administration (SAMHSA), opioids. A definition of MMT and Rockville. This article is part of a series of literature reviews that will be published in Psychiatric Services over the next several months. The reviews were commissioned by features of medication-assisted treat- SAMHSA through a contract with Truven Health Analytics and were conducted by ment are presented in Table 1. experts in each topic area, who wrote the reviews along with authors from Truven Health This article reports the results of Analytics, Westat, DMA Health Strategies, and SAMHSA. Each article in the series was a literature review that was under- peer reviewed by a special panel of Psychiatric Services reviewers. taken as part of the Assessing the

146 PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 Evidence Base Series (see box on next Table 1 page). The literature review was un- Description of medication-assisted treatment with methadone dertaken to describe MMT and its primary and secondary treatment Feature Description goals, rate the levels of evidence (methodological quality) of existing Service definition Medication-assisted treatment is a direct service that provides studies for this treatment, and de- a person with a substance use or mental disorder with pharmacotherapy in conjunction with behavioral therapies as treat- scribe the effectiveness of this ser- ment for associated symptoms or disabilities. The nature of the vice. The results provide state mental services provided is determined by the person’s current status or health directors and their staff, pur- needs. chasers of health services, state policy Methadone maintenance treatment is a medication-assisted officials, community health care admin- treatment that uses methadone to assist individuals with an opiate use disorder to abstain from or decrease the use of illegal istrators, consumers, and family mem- opiates (for example, intravenous heroin) or the use of opiates bers with an accessible summary of the in a nonprescribed manner (for example, abuse of prescription evidence for MMT and its implications pain medications). for the treatment of opioid use disor- Service goals Retention in treatment; decrease in illegal opioid use; decrease ders. To address the concerns of the in mortality; decrease in nonopioid drug use; decrease in target audiences, this review exam- criminal activity; decrease in risk behaviors related to HIV and hepatitis C ined the evidence for MMT in various populations (including pregnant Populations Adults with opioid use disorders; pregnant women with opioid use disorders women), appropriate dosing guidelines, Settings of service Methadone treatment centers and serious adverse events related to delivery methadone use. Description of MMT MMT has been available since 1964. treated at methadone clinics receive reviews and meta-analyses, major In the United States, MMT is offered inadequate doses and that many clin- clinical texts, and professional clinical through specialized methadone treat- ics place an arbitrary limit on the society reviews. We relied on system- ment programs that provide psycho- duration of treatment (9,10). This atic reviews and meta-analyses to social support as well as close patient assessment of the available research summarize relevant findings from monitoring. Typically, methadone will help inform behavioral health earlier years. To provide additional doses are dispensed daily at the policy leaders about the effects of information from recent years that methadone treatment facility to min- MMT on the lives of those with may not have been included in review imize risks of diversion. However, opioid use disorders and about its articles, we supplemented these re- individuals may become eligible for value as a treatment option and view articles with articles presenting take-home doses on the basis of ap- a covered health benefit. results from individual randomized propriate clinic attendance, absence controlled trials (RCTs) and quasi- of behavioral problems at the clinic Methods experimental observational studies. We or recent drug abuse, lack of known Search strategy considered studies that were focused criminal activity, and evidence of a We conducted a literature search of onMMTforadultswithopioiduse stable home with the ability to store major databases: PubMed (U.S. Na- disorders, including pregnant women. methadone safely. tional Library of Medicine and Nation- Specific topics, such as adverse events Because individuals remain depen- al Institutes of Health), PsycINFO and medication interactions, were also dent on methadone, MMT is not (American Psychological Association), examined. considered an abstinence treatment. Applied Social Sciences Index and The duration of methadone treatment Abstracts, Sociological Abstracts, Social Inclusion and exclusion criteria is indefinite (8). The goals of metha- Services Abstracts, and Published In- The abstracts of the identified articles done treatment are to reduce or elim- ternational Literature on Traumatic were examined to determine compli- inate illicit opioid use and, as a result, Stress. We identified meta-analyses, re- ance with inclusion and exclusion to decrease its associated negative out- search reviews, clinical guidelines, and criteria. Articles on which opinions con- comes (Table 1). For pregnant women, individual studies about MMT that were curred were accepted. The following the goals of MMT include improved published from 1995 through 2012. We inclusion criteria were used: RCTs, maternal and fetal outcomes. used combinations of the following quasi-experimental studies, systematic MMT aims to allow individuals with search terms: methadone, opioid main- review articles, meta-analyses, and opioid use disorders to minimize tenance treatment, opioid treatment, clinical guidelines; English-language many of the negative health and addiction pharmacotherapy, medication- studies conducted in the United societal outcomes associated with assisted maintenance treatment, MMT, States, including international studies opioid use. Despite the long history and pregnancy. that used U.S.-based sites and in- of methadone use, studies have sug- Additional literature was found by ternational reviews encompassing gested that a majority of individuals examining the bibliographies of major U.S.-based studies; and studies that

PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 147 and noted differences across inves- About the AEB Series tigations. We considered the quality of the research design in our con- The Assessing the Evidence Base (AEB) Series presents literature reviews clusions about the strength of the for 14 commonly used, recovery-focused mental health and substance use evidence and the effectiveness of services. Authors evaluated research articles and reviews specific to each MMT. service that were published from 1995 through 2012 or 2013. Each AEB Series article presents ratings of the strength of the evidence for the service, Results and discussion descriptions of service effectiveness, and recommendations for future Level of evidence implementation and research. The target audience includes state mental The literature search found seven health and substance use program directors and their senior staff, Medicaid – staff, other purchasers of health care services (for example, managed care RCTs (12 18) and two retrospective, organizations and commercial insurance), leaders in community health quasi-experimental studies (19,20). organizations, providers, consumers and family members, and others Summaries of these individual studies interested in the empirical evidence base for these services. The research are provided in Table 2. We also in- was sponsored by the Substance Abuse and Mental Health Services cluded 15 reviews or meta-analyses Administration to help inform decisions about which services should be that examined multiple studies (21–35). covered in public and commercially funded plans. Details about the re- Summaries of these reviews are in- search methodology and bases for the conclusions are included in the cluded in Table 3. introduction to the AEB Series (11). Because of the large number of trials included as individual studies or as part of review articles, the overall evidence rating for MMT is high. focused on MMT for individuals with ratings are based on the following Several meta-analyses, reviews, and opioid use disorders or the use of three options: two or more quasi- RCTs representing more than three MMT during pregnancy. Excluded experimental studies with adequate independent RCTs have reported on were case studies, single-subject designs, design; one quasi-experimental study the primary outcomes of MMT, which and cross-sectional studies; studies that plus one RCT with adequate design; are retention in treatment and reduc- focused on methadone use for pain or at least two RCTs with some tion of illicit opioid use (12–16,21–24). management or for detoxification methodological weaknesses or at least In addition, meta-analyses, reviews, from opioids; and reviews and meta- three quasi-experimental studies with RCTs, and quasi-experimental studies analyses that contained only articles some methodological weaknesses. Low representing more than three RCTs or that did not meet the inclusion ratings indicate that research for this two RCTs and two quasi-experimental criteria. service is not adequate to draw evidence- studies have addressed secondary out- based conclusions. Low ratings indi- comes such as other illicit drug use, Strength of the evidence cate that studies have nonexperimental HIV risk behaviors, criminal behav- The methodology used to rate the designs, there are no RCTs, or there is iors,heroincraving,andmortality strength of the evidence is described no more than one adequately designed (15–17,21,23–27). in detail in the introduction to this quasi-experimental study. series (11). The research designs of We accounted for other design Effectiveness of MMT the studies identified during the factors that could increase or decrease Research supports MMT’s positive literature search were examined. the evidence rating, such as how the impact on treatment retention and Three levels of evidence (high, mod- service, populations, and interven- suppression of heroin use, particularly erate, and low) were used to indicate tions were defined; use of statistical at higher methadone doses. Findings the overall research quality of the methods to account for baseline differ- regarding secondary outcomes are collection of studies. Ratings were ences between experimental and com- mixed, although there is general sup- based on predefined benchmarks that parison groups; identification of port that MMT has a positive impact considered the number of studies and moderating or confounding variables on criminal activity associated with their methodological quality. If ratings with appropriate statistical controls; ex- heroin use, as well as on mortality and were dissimilar, a consensus opinion amination of attrition and follow-up; risk behaviors for HIV and hepatitis C was reached. use of psychometrically sound mea- infection. High ratings indicate confidence in sures; and indications of potential MMT versus placebo or no phar- the reported outcomes and are based research bias. maceutical maintenance treatment. on three or more RCTs with adequate Most of the literature on the effective- designs or two RCTs plus two quasi- Effectiveness of the service ness of MMT versus placebo or no experimental studies with adequate We described the effectiveness of medication-assisted treatment was designs. Moderate ratings indicate MMT–that is, how well the outcomes published between the 1960s and that there is some adequate research of the studies met the goals of MMT. 1990s. In general, these and later to judge the service, although it is We compiled the findings for sepa- studies found that when MMT is pro- possible that future research could rate outcome measures and study vided at adequate dose levels, it is more influence reported results. Moderate populations, summarized the results, effective than no medication treatment

148 PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 Table 2 Individual studies of methadone maintenance treatment (MMT) included in the reviewa

Population and Study Design and objectives conditions Outcomes measured Summary of findings

Strain et al., Double-blind, 40-week Patients randomly assigned Primary: opioid-positive No differences in treatment retention 1999 (13) RCT to compare mod- to daily oral methadone urinalysis and treatment through week 40 (mean retention in erate versus high doses hydrochloride; patients retention high-dose group, 159 days; in moder- of methadone in treat- receiving a dose ranging ate-dose group, 157 days). The high- ment of adults with from 40–50 mg (N=97) dose group had significantly greater opioid dependence compared with those re- reduction in opioid-positive urinalysis ceiving a dose ranging compared with the medium-dose from 80–100 mg (N=95); group: 53.0% (CI=46.9%–59.2%) all received substance versus 61.9% (CI=55.9%–68.0%) abuse counseling (p=.047). Sees et al., RCT to compare out- Patients randomly assigned Primary: treatment reten- MMT resulted in greater treatment 2000 (12) comes of patients with to MMT (N=91), in- tion, heroin and cocaine retention (median retention, 438.5 opioid dependence cluding 2 hours of psy- abstinence (by self-report days versus 174.0 days for comparison treated with MMT or chosocial therapy per and monthly urinalysis), group) and lower heroin use. MMT with psychosocially week during first 6 HIV risk behaviors, and group had a lower rate of drug-related enriched, 180-day months; patients ran- functioning in 5 problem HIV risk behaviors at 12 months methadone-assisted domly assigned to de- areas (employment, fam- (mean6SD=.056.13 versus .136.19). detoxification toxification (N=88), ily, psychiatric, legal, and including 3 hours of alcohol use) psychosocial therapy per week, 14 educa- tional sessions, and 1 hour of cocaine group therapy (if needed) for 6 months McCarthy Retrospective cohort study Mothers who received Primary: rate of medication High doses of methadone were not as- et al., 2005 to compare the effects methadone (N=81) and treatment for neo- sociated with increased risks of NAS (19) of high-dose versus low- their offspring; half of natal abstinence symp- symptoms. High doses had a positive dose methadone during mothers assigned to a toms, days of infant effect on maternal drug abuse: in pregnancy on maternal high-dose group ($100 hospitalization high-dose group, 11% of infant tox- and fetal outcomes mg) and half to a low- icology screens were positive for il- dose group (,100 mg) licit drugs, compared with 27% in low-dose group (p=.05). Schwartz RCT to compare outcomes Participants (N=319) meet- Primary: rate of standard Participants who received interim meth- et al., 2006 of adults assigned to in- ing criteria for heroin MMT enrollment, self- adone treatment entered standard (14) terim methadone treat- dependence and for re- reported heroin use, MMT at a significantly higher rate ment or to a wait-list ceipt of MMT assigned opioid-positive urinaly- than those on the wait list (75.9% control group to interim methadone sis, illegal income re- versus 20.8%, p,.001). At 4 months, treatment (N=199) or ceived, and money the interim methadone treatment wait-list control group spent on drugs group reported significantly fewer (N=120) days of heroin use (p,.001), had reduced heroin-positive urine screens (p,.001), reported spending less on drugs (p,.001),andreceivedless illegal income (p,.02). Jones et al., Double-blind RCT to Pregnant women (N=175) Primary: percentage of Buprenorphine group required less 2010 (44) compare neonatal and with opioid depend- neonates treated for morphine for NAS than methadone maternal outcomes of ence; methadone group NAS, NAS peak score, group (mean dose=1.1 mg versus 10.4 opioid-dependent (N=89; 16 dropped out) duration of hospital stay, mg, p,.009), had a shorter hospital women treated with and buprenorphine morphine required to stay (10.0 days versus 17.5 days, methadone or with group (N=86; 28 drop- treat NAS, and neonatal p,.009), and had a shorter duration buprenorphine during ped out); 131 neonates head circumference; sec- of treatment for NAS (4.1 days versus pregnancy of mothers who were ondary: treatment reten- 9.9 days, p,.003); 33% of buprenor- followed to the end of tion and reduction in phine group discontinued treatment pregnancy (58 exposed opiate use before delivery, compared with 16% to buprenorphine, 73 ex- of methadone group. posed to methadone) Wilson et al., RCT to examine use of Heroin-dependent adults Primary: AIDS Risk As- For injection risk scale score, injected 2010 (17) interim methadone (N=319) randomly assessment questionnaire drugs, and sex risk score, treatment treatment on HIV risk signed to interim meth- (assesses HIV infection condition (p,.008, p,.03, and behavior among adults adone treatment without and HIV sex risk behav- p,.04, respectively) and time effects with heroin dependence counseling (N=199) or iors) at baseline and (p,.001, p,.001, p,.02) were sig- to wait-list control group follow-up nificant for injection risk, with interim (N=120) without auto- methadone group performing better matic admission after than wait-list control group. 120 days Continues on next page

PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 149 Table 2 Continued from previous page

Population and Study Design and objectives conditions Outcomes measured Summary of findings

Pizarro et al., Retrospective cohort study Pregnant methadone users Primary: rate and severity Regardless of methadone dose, rates of 2011 (20) to assess the incidence (N=174) stratified into of NAS, birth weight, NAS were similar among low-dose, of clinically significant three dose groups: low preterm birth rate, and medium-dose, and high-dose groups NAS (0–50 mg per day, neonatal morbidities (40.7%, 52.4%, and 40.8%, respec- N=59), medium (51–100 and mortality tively; p..05). No significant outcomes mg per day, N=63), and were found. high (.100 mg per day, N=52) Schwartz RCT to evaluate the im- Participants (N=244) new- Primary: treatment reten- Counseling had no significant impact on et al., 2011 pact of counseling on ly admitted to meth- tion and Addiction Se- treatment retention or rate of positive (15) the first 4 months of adone treatment programs verity Index, which urine tests for methadone group. All MMT among 3 com- from wait lists and ran- includes alcohol and groups showed reduction in self- parison groups domly assigned to drug use; medical, psy- reported days of criminal activity, emergency counseling chological, and legal money spent on drugs, and illegal only for 120 days fol- issues; family and social income compared with baseline (all lowed by standard treat- relationships; and em- p,.001). ment (N=108), standard ployment status psychosocial services (N=107), or counseling by case managers with small caseloads (N=29) Schwartz RCT to evaluate the im- Participants (N=230) from Primary: treatment reten- No significant differences were found in et al., 2012 pact of counseling on previous RCT; 3 condition and Addiction Se- treatment retention between the (16) MMT among 3 compar- tions: emergency coun- verity Index, which supervised methadone (60.6%), ison groups at 12 months seling (N=99), standard includes alcohol and standard methadone (54.8), and re- (follow-up of the counseling (N=104), or drug use; medical, psy- stored methadone (37.0%) treatment Schwartz et al. [15] counseling by case chological, and legal groups. Positive urine screens de- sample) managers with small issues; family and social clined significantly from baseline for caseloads (N=27) relationships; and em- all groups (p,.001 for heroin and ployment status p,.003 for cocaine metabolites). No significant group 3 time interactions were found for these measures. a Studies are listed in chronological order. Abbreviations: CI, 95% confidence interval; NAS, neonatal abstinence syndrome; RCT, randomized controlled trial in retaining patients in treatment and individual studies showed a greater re- had longer retention in treatment reducing illicit opioid use (21,22,28,29). duction in both outcomes among pa- compared with the detoxification group Recently, Mattick and colleagues tients receiving MMT. Three of the 11 (median of 438.5 versus 174 days). The (21) conducted a review for the studies reviewed by Mattick and col- MMT group also showed lower rates of Cochrane Collaboration of 11 RCTs leagues measured criminal activity, heroin use and lower rates of drug- (two of which were double-blinded) and four measured mortality. related HIV risk behaviors compared that assessed the effectiveness of Sees and colleagues (12) compared with the detoxification group. There MMT compared with treatments with outcomes of individuals with opioid were no differences between the two no opioid replacement therapy (that dependence who were receiving groups in sex-related HIV risk behav- is, detoxification protocols, drug-free MMT (N=91) or who were in a 180- iors or in employment, family function- rehabilitation protocols, placebo med- day psychosocially enriched detoxifi- ing, or alcohol use outcomes. ications, or wait-list control groups). cation program (N=88). One goal of Two systematic reviews and meta- The combined total of participants this study was to examine alternatives analyses have examined the impact of across 11 studies was 1,969. On the to indefinite MMT use by looking at MMT on HIV high-risk behaviors. basis of meta-analyses, the authors a six-month detoxification rather than Both reviews noted the limited num- concluded that methadone was sig- the faster detoxification programs ber of RCTs that contributed to their nificantly more effective than non- (usually one month) studied in the results. One review (N=12 studies) pharmacological treatment in retaining past. For six months the detoxification found that MMT was associated with patients in treatment and in suppress- group received psychosocial services a 54% reduction in the risk of HIV ing heroin use as measured by urine that included three hours of psycho- infection (25). The second review drug testing. No significant difference social therapy per week, 14 educa- (N=36 studies) was unable to com- was found between the two treatment tional sessions, and one hour of group bine results from the studies; the conditions (MMT and no opioid re- therapy focused on cocaine use; the authors concluded that across studies placement therapy) in their impact on group also received six months of MMT reduced drug-related risk fac- criminal activity or mortality, although aftercare. The group receiving MMT tors such as sharing of injection

150 PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 Table 3 Review articles about methadone maintenance treatment (MMT) included in the reviewa

Study Focus of review Studies included Outcomes measured Summary of findings

Hall et al., Effectiveness of MMT on 6 RCTs assessing MMT, Primary: reduction in he- Although variation in outcomes be- 1998 (22) heroin use and crime and 8 additional gen- roin use and illicit opi- tween different programs was noted, eralized observational oid use, criminal activity the effectiveness of MMT in con- studies trolling heroin and illicit opioid use and crime was generally supported through the RCTs and observational studies. Fletcher Epidemiological Drug 12-month follow-up sam- Primary: treatment reten- DATOS study results for drug treat- and Abuse Treatment Out- ple based on 2,966 in- tion and various other ment outcomes were consistent Battjes, come Studies (DATOS) terviews from 76 U.S. treatment outcomes with prior evaluation findings, in- 1999 (29) conducted at multiple programs dicating that the major treatment U.S. sites modalities (including outpatient methadone treatment) are effective in reducing illicit drug use, reducing the incidence of drug-related criminal behavior, and supporting im- provement of health, mental health, and social functioning. Faggiano Efficacy and safety of var- 21 studies, including 11 Primary: retention rate, RCTs showed that high doses of MMT et al., ious dose ranges of MMT RCTs (2,279 total par- opioid use (self- were associated with better treatment 2003 (23) for opioid dependence ticipants) and 10 con- reported), opioid absti- retention (high versus low doses at trolled prospective nence (urine screen), longer follow-ups, RR=1.62, CI=.95– studies (3,715 total cocaine abstinence 2.77), opioid abstinence (high versus participants) (urine screen), and low, RR=1.59, CI=1.16–2.18; high overdose mortality versus middle, RR=1.51, CI=.63– 3.61), and cocaine abstinence (high versus low, RR=1.81, CI=1.15–2.85). At 6-year follow-up, controlled prospective studies showed lower overdose mortality at higher doses (high versus low doses, RR=.29, CI=.02– 5.34; high versus middle, RR=.38, CI=.02–9.34; and middle versus low, RR=.57, CI=.06–5.06. Center for National assessment of National assessment of Primary: methadone- Evidence suggests that an increase in Substance deaths associated with methadone-associated associated mortality methadone-attributable deaths in Abuse methadone use; recom- mortality in May 2003 1999–2004 was largely related to Treat- mendations for reduc- increased use for pain analgesia. ment, ing mortality from SAMHSA highlights the importance 2004 (32) methadone of public understanding that related mortality is essentially eliminated when methadone is prescribed, dispensed, and used appropriately. Connock Clinical and cost effective- 31 systematic reviews Primary: retention in At all doses used in the studies (MMT, et al., ness of BMT and MMT and 27 RCTs treatment and illicit 20–97 mg per day; BMT #5–18 mg 2007 (28) for the management use of opioids per day), treatment retention was of opioid-dependent better than in the placebo or no individuals therapy groups (MMT, RR=3.91, CI=1.17–13.2; BMT, RR=1.74, CI=1.06–2.87). Higher doses of MMT and BMT were almost always more effective than lower doses for treatment retention and illicit use reduction. Across comparable doses, MMT was more effective than BMT for treatment retention, except at low doses. At low doses, the two medications appeared comparable (#35 mg of MMT versus 6–16 mg of BMT, RR=1.01, CI=.66–1.54). No significant difference across studies was found in illicit opiate use between flexible-dose MMT and BMT. Continues on next page

PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 151 Table 3 Continued from previous page

Study Focus of review Studies included Outcomes measured Summary of findings

Mattick Effectiveness of MMT 11 RCTs (1,969 total Primary: patient retention MMT was significantly more effective et al., compared with treat- participants) in treatment and her- than nonreplacement approaches in 2009 (21) ments not involving opi- oin use suppression as treatment retention and suppression of oid replacement therapy measured by urine drug heroin use (measured by self-report testing; secondary: crim- and lab analysis) (6 RCTs, RR=.66, inal activity and mortality CI=.56–.78). No significant differences were found for criminal activity (3 RCTs, RR=.39, CI=.12–1.25) or mortality (4 RCTs, RR=.48, CI= .10– 2.39). Cleary et al., Relationship between ma- 67 studies in the system- Primary: key conclusions, Meta-analysis did not demonstrate 2010 (31) ternal methadone dose atic review; 29 studies including incidence, a consistent, significant difference in in pregnancy and diag- in the meta-analysis severity, and duration NAS incidence among neonates of nosis or medical treat- of NAS outcomes in women on low versus high meth- ment of NAS relation to maternal adone doses at delivery. Nineteen methadone dose studies found a relationship between methadone dose and incidence, severity, or duration of NAS; 18 did not find a relationship; 30 did not report on the relationship. Fareed Update for clinicians about 24 studies, including 12 Primary: effect of metha- Treatment retention: 9 studies re- et al., methadone dosing, with RCTs, 10 observational done dose on retention ported that the daily dose range of 2010 (24) dose recommendations studies, and 2 meta- in treatment, illicit opi- 60–100 mg showed significant im- analyses oid use, and mortality provement for treatment retention compared with lower doses. Six studies did not find a significant difference in retention for this dose range. Illicit opioid use: 10 studies recommended a daily dose range of 60–100 mg; 2 studies suggested that doses over 100 mg are more effective for decreasing heroin use. Mortality rate: 2 long-term observational studies reported doses greater than 100 mg daily to be safe and effective in long-term MMT (the authors stated that more research is needed). Modesto- Risk factors for metha- Literature review (N of Primary: pharmacological Risk factors of respiratory depression Lowe done mortality in opi- studies not reported) of profile of methadone include advanced age, medically com- et al., oid-dependent and pain pharmacological prop- and relationship to risk promised status, liver or pulmonary 2010 (35) populations; guidelines erties and relationship factors for methadone pathology, sleep apnea, polysub- for initiating methadone to risk factors for ad- mortality stance abuse, opioid-naïve or low treatment in these pop- verse events opioid tolerance, high doses of meth- ulations to minimize risk adone, and rapid titration of metha- of death done. Risk factors for Torsades de Pointe include female sex, electrolyte imbalance, liver or cardiac pathology, unexplained syncope or seiz- ures, other drug and medication use that prolongs QT interval or inhibits CYP 3A4, prolonged QT interval, and high doses of methadone. Amato et al., Effectiveness of any psy- 35 RCTs considering 13 Primary: treatment reten- Compared with standard maintenance 2011 (30) chosocial and any agonist different psychosocial tion, opiate use during treatment, psychosocial and any main- maintenance treatment interventions (4,319 treatment, compliance tenance treatment showed no benefit compared with standard total participants) with sessions during for treatment retention (27 studies, agonist treatment for treatment, and other 3,124 participants, RR=1.03, CI=.98– opiate dependence psychological health 1.07), opiate abstinence during treat- measures ment (8 studies, 1,002 participants, RR=1.12, CI=.92–1.37), or compliance (3 studies, mean difference=.43, CI=–.05 to .92), among other findings. Comparisons of the various Continues on next page

152 PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 Table 3 Continued from previous page

Study Focus of review Studies included Outcomes measured Summary of findings

psychosocial approaches showed no significant differences in any outcomes. Fareed Effect of MMT on opiate Total of 16 studies: RCTs, Primary: effect of MMT Seven studies reported that metha- et al., craving observational studies, on subjective opiate done could reduce heroin craving, 4 2011 (27) meta-analyses, and craving and on objec- reported that MMT patients are still reviews tive measures of opiate at risk for craving, 1 study reported craving that methadone could increase heroin craving, and 4 studies reported that methadone had a neutral effect on heroin craving. Gowing Effect of oral substitution 38 studies (nearly 12,400 Primary: HIV transmis- Substitution treatment for opioid- et al., treatment for opioid- total participants). Two sion risk behaviors, dependent, injecting drug users with 2011 (26) dependent drug injectors studies were RCTs; 11 includingdruguse; methadone or buprenorphine was on behaviors associated were controlled trials, secondary: rates of consistently associated with signifi- with high risk of HIV but the intervention HIV infection cant reductions in illicit opioid use, transmission; incidence was not relevant to the injecting drug use, and sharing of of HIV infections review, and therefore, needles. It was associated with a re- these trials were used duction in the use of multiple sex as a baseline versus partners or the exchange of sex for postintervention com- money or drugs, but it was not parison; 21 were ob- associated with increased condom servational prospective use. The risk behavior reduction studies; 4 were cross- appeared to relate to reductions in sectional. cases of HIV infection, although data were not pooled because of variability and bias among studies. Martin Adverse cardiac events Expert panel examined Primary: cardiac events Results established the connection et al., associated with the peer-reviewed lit- associated with metha- between methadone and prolonga- 2011 (34) methadone erature, regulatory done, cardiac QT in- tion of QT interval and suggested actions, professional terval impact a dose-dependent effect for metha- guidance, and opioid done. Authors recommended that treatment program every opioid treatment program outcomes should have a universal cardiac risk management plan (to the extent possible) for patients with identified risk factors for adverse cardiac events. Webster Causes and risk factors for 91 documents were as- Primary: frequency, de- Risk factors for methadone-related et al., opioid-related poisoning sessed by a panel of mographic characteris- deaths were unanticipated medical 2011 (33) deaths and recommen- experts tics, and risk factors for or mental health comorbidities, payer dations to reduce death opioid-related deaths policies that encourage or mandate rates attributable to overdose methadone as first-line therapy, the in the past decade presence of additional central nervous system–depressant drugs, and sleep-disordered breathing. Cardiac irregularities in the presence of methadone remain an uncommon cause of death. MacArthur HIV risk: quantify the ef- Pooled data from 9 ob- Primary: impact of opiate Substitution treatment was associated et al., fect of opiate substitu- servational studies, in- substitution treatment with an average 54% reduction in the 2012 (25) tion treatment in relation cluding 819 incident as related to HIV in- risk of HIV infection among individ- to HIV transmission HIV infections over cidence; secondary: ef- uals who inject drugs (rate ratio=.46, among individuals who 23,608 person-years of fect of variables such as CI=.32–.67; p,.001). Heterogeneity inject drugs follow-up mode and duration of was found between studies that could treatment, geographi- not be explained by region, site of cal region, study set- recruitment, or incentives. ting, and participant characteristics a Studies are listed in chronological order. Abbreviations: BMT, buprenorphine maintenance treatment; CI, 95% confidence interval; NAS, neonatal abstinence syndrome; RCT, randomized controlled trial; RR, relative risk or risk ratio

PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 153 equipment (26). The second review re- better treatment retention and cocaine standard methadone (N=104), and ported that there were too few studies abstinence, less heroin use during restored methadone (N=27) treat- to be conclusive but stated that MMT treatment, and fewer withdrawal symp- ment. Restored methadone treatment was associated with lower rates of mul- toms. Few studies included doses refers to treatment by counselors with tiple sex partners and the exchange of above 110 mg; therefore, the data reduced caseloads, which allows them sex for drugs or money and had no were less reliable for these doses. to provide more intensive treatment. effect on the use of condoms. Only one underpowered study ex- The studies found no difference Interim methadone treatment is a amined mortality and criminal activ- between groups in treatment reten- program that allows provision of meth- ity, but a trend that did not reach tion at four months and better treat- adone under daily supervision for statistical significance suggested that ment retention for the interim and up to 120 days while the individual is individuals receiving higher doses standard methadone treatment groups awaiting placement in a standard meth- had lower mortality rates. A second at 12 months. No between-group adone program. It does not include review showed similar results; doses differences in opiate use or other counseling other than emergency above 60 mg were associated with drug use were found at the four- and counseling. One RCT examined HIV better treatment retention and fewer 12-month follow-up assessments. At risk behaviors for 319 opioid-addicted urine drug tests that were positive for 12 months, no difference was noted adults who were randomly assigned to opioids (24). between groups in arrests, criminal interim methadone treatment or a Strain and colleagues (13) con- activity, or money spent on drugs. wait list (17). Rates of drug injection ducted a 40-week, double-blinded Self-reported illegal income was andsexwhilehighondrugswere RCT comparing moderate (40–50 significantly higher in the standard significantly lower for individuals ran- mg, N=97) and high (80–100 mg, methadone treatment group. domly assigned to the interim metha- N=95) doses of methadone in the A Cochrane Collaboration system- done program. treatment of adults with opioid de- atic review by Amato and associates Another review examined the effect pendence. There were two main in 2011 (30) examined 35 studies of MMT on heroin craving and in- outcome measures: opioid-positive that evaluated whether outcomes im- cluded 16 studies (27). It found mixed urinalysis and treatment retention. proved after the addition of a specific, results; seven studies showed that The study found no difference in structured psychosocial intervention MMT reduced heroin craving, four treatment retention through week to standard agonist maintenance studies showed that patients were 40. The high-dose group had signif- treatment (either methadone or still at risk of heroin craving, one icantly greater reduction in opioid- buprenorphine) that already included study showed that methadone could positive urinalysis (53%) compared psychosocial treatment. The studies increase heroin craving, and four with the medium-dose group (62%). included 13 different psychosocial studies showed a neutral effect. In Service delivery and psychosocial interventions that were added to general, the studies that showed pos- treatments. Many methadone treatment standard treatment. Taken as a whole, itive results used higher methadone centers have wait lists, which indicate additional psychosocial treatment did doses, and those with negative or a lack of access to desired treatment. not statistically improve retention in neutral results used lower doses or Given the high social cost of opioid treatment, use of opiates during were in the setting of methadone addiction, a research group investigated treatment, session attendance during detoxification. the use of interim methadone treatment treatment, or other measures of psy- Levels of methadone doses. The as a way to improve access and decrease chological health. When the review literature has consistently shown that waiting lists. Schwartz and colleagues was limited to studies with contingency the effectiveness of MMT increases (14) conducted an RCT to compare management approaches, there still when methadone is used at doses outcomes for adults assigned to interim was no statistically significant effect above 60 mg. Two systematic reviews methadone treatment (N=199) or a of additional psychosocial services on suggested that higher doses of meth- wait-list control group (N=120). The treatment retention or decreased opi- adone were associated with improved study found that participants in the oid use. Contingency management outcomes. First, Faggiano and col- interim methadone treatment cohort describes behavioral modification pro- leagues (23) performed a systematic entered standard MMT at a significantly grams that provide rewards, such as re- review for the Cochrane Collabora- higher rate (75%) than those assigned to tail gift cards, for desired behaviors, such tion that evaluated the efficacy and the wait list (20.8%). In addition, at four as negative urinalyses. Because stan- safety of different doses of methadone months, interim methadone treatment dard treatment included psychosocial for opioid dependence. This review participants reported significantly lower treatment, Amato and colleagues could included 21 studies (11 RCTs and rates of heroin use than wait listed draw conclusions only regarding the ten controlled, prospective, quasi- participants, had fewer positive drug addition of a structured psychother- experimental studies). The authors tests for heroin, reported spending apy and not regarding the efficacy of examined outcomes for four different significantly less money on drugs, psychosocial treatment. dose ranges: low (1–39 mg), medium and received less illegal income. Pregnant women subgroup. Early (40–59 mg), high (60–109 mg), and Schwartz and colleagues (15,16) studies established the efficacy of using very high ($110 mg). Results showed compared individuals who were ad- MMT to reduce pregnancy-related that high doses were associated with mitted to interim methadone (N=99), maternal and fetal morbidity among

154 PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 opioid-addicted pregnant women (36,37). MMT during pregnancy was Evidence for the effectiveness of methadone associated with decreased illicit opioid use, increased rates of prenatal re- maintenance treatment: high tention in treatment, decreased preg- Evidence clearly shows that MMT has a positive impact on: nancy complications, and generally • Retention in treatment • improved fetal outcomes (18,38). Illicit opioid use However, MMT has been found to Evidence is less clear but suggestive that MMT has a positive impact on: put newborn infants at risk for neo- • Mortality • natal abstinence syndrome (NAS)—a Illicit drug use (nonopioid) • Drug-related HIV risk behaviors condition characterized by dysfunc- • Criminal activity tion of the autonomic nervous system, gastrointestinal tract, and respiratory Evidence suggests that MMT has little impact on: • Sex-related HIV risk behaviors system and by irritability of the central nervous system. NAS often requires detoxification treatment in the hospital with a morphine taper (19,37,39– (MOTHER) study was a large, multi- prolongation. The QT interval is 41). Reported rates of withdrawal center, double-blind RCT published a measure of time between the start symptoms among neonates born to in 2010 (44). The authors compared of the Q wave and the end of the T opioid-addicted mothers who contin- neonatal and maternal outcomes wave in the heart’s electrical cycle ued to use opiates within a week of between pregnant women treated that is measured by an electrocardio- giving birth range from 55% to 94% during their pregnancies with meth- gram. Prolongation of the QT interval (42), and rates of NAS that develop adone (dose range 20–140 mg) or canleadtoseriousheartarrhyth- among neonates as a result of treating buprenorphine (dose range 2–32 mg). mias such as Torsades de Pointes the mother with MMT during preg- Eighty-nine women were randomly (TdP) and sudden death. As a result nancy fall into this range (31). Recent assigned to receive methadone, and of this rise in mortality, the U.S. studies on the long-term impact of 86 were randomly assigned to receive Food and Drug Administration is- NAS on development are scant. Older buprenorphine. Thirty-three percent sued a physician safety alert in 2006 studies indicated no differences in cog- of women in the buprenorphine group highlighting fatalities and cardiac ar- nitive performance among four-year- discontinued treatment before delivery, rhythmias associated with meth- old children of mothers receiving MMT compared with 16% in the methadone adone (34). and children of mothers with similar group. No significant differences were Respiratory depression is most often demographic characteristics in a control found in the percentage of newborns a consequence of methadone accumu- group. However, scores of children in treated for NAS. However, infants born lation and use of concurrent illicit both groups were lower than popula- to women treated with methadone drugs or medications that also suppress tion norms (43). required higher doses of morphine to the central nervous system. Reviews To guide clinicians regarding the treat NAS, required more days of suggest that initiation into methadone necessity of tapering MMT before de- treatment for NAS, and had longer treatment is a particularly vulnerable livery, researchers have examined the hospital stays. There were no differ- time in both methadone maintenance relationship between methadone dose ences in maternal use of illicit drugs and pain therapy populations, partic- during pregnancy and the incidence and at delivery or other fetal or mater- ularly if the dose is increased rapidly severity of NAS among newborn in- nal outcomes. These results sug- (33,35). The most common drugs as- fants. Because of increased methadone gest that less severe NAS among sociated with respiratory suppression metabolism during pregnancy, preg- infants born to mothers treated are benzodiazepines and alcohol. Deaths nant women often require higher with buprenorphine may be con- from respiratory depression may also doses. Cleary and colleagues (31) per- founded by poorer treatment re- be caused by inappropriate dosing by formed a systematic review and meta- tention rates for these mothers, methadone recipients and by diver- analysis and found that methadone especially for mothers with a longer sion of methadone, which occurs dose had no consistent effect on rates history of heroin use. when individuals who have a prescrip- of NAS and other neonatal outcomes. Adverse events. Between 1999 and tion for methadone sell or give their Two of the 67 studies included in that 2004, deaths attributed to metha- methadone to others rather than us- review were RCTs, and the remaining done increased by 390%. Evidence ing it themselves. studies had quasi-experimental observa- suggests that this change was large- In 2007–2009, a panel established tional designs. Additional retrospective ly related to the increased use of by SAMHSA summarized evidence of cohort studies showed similar results; no methadone for pain analgesia rather methadone’s impact on the cardiac QT difference in NAS rate or severity was than MMT (32,33). Nonetheless, interval and derived guidelines for found on the basis of methadone dose the sharp rise of methadone-related methadone treatment programs (34). during pregnancy (19,20). deaths highlights safety issues—in The review established a connection The Maternal Opioid Treatment: particular, the risks of respiratory between methadone and prolonga- Human Experimental Research depression and cardiac QT interval tion of the QT interval and suggested

PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 155 a dose-dependent effect for metha- of MMT on secondary outcomes, de- References done. Prolongation of the QT in- velopment of a better understanding 1. Lloyd J: Drug Policy Information Clear- terval greater than 500 ms confers of the efficacy and safety tradeoffs of inghouse Fact Sheet: Heroin. Washington, significant risk with respect to ar- very high methadone doses (.100 DC, Office of National Drug Control Pol- icy, 2003 rhythmias such as TdP (34). Use of mg), confirmation of the results of additional medications that might in- interim methadone treatment as a 2. Results From the 2007 National Survey on Drug Use and Health: National Findings. crease the QT interval increases an potential avenue to improve out- NSDUH Series H-34, pub no SMA 08- individual’s risk of cardiac arrhyth- comes of MMT, and use of MMT in 4343. Rockville, Md, Substance Abuse and mias. Despite these findings, cardiac specific subpopulations, such as ra- Mental Health Services Administration, Office of Applied Studies, 2008 irregularities in the presence of meth- cial and ethnic minority groups and adone remain an uncommon cause of individuals who use prescription 3. Goldstein A, Herrera J: Heroin addicts and methadone treatment in Albuquerque: death (33). drugs compared with those who use a 22-year follow-up. Drug and Alcohol intravenous heroin. Dependence 40:139–150, 1995 Conclusions Given the poor success rates of 4. Hulse GK, English DR, Milne E, et al: The Overall, there is a high level of abstinence-based treatments for opi- quantification of mortality resulting from evidence for the effectiveness of oid use disorders, MMT is an im- the regular use of illicit opiates. Addiction 94:221–229, 1999 MMT in improving treatment reten- portant treatment option for opioid tion and decreasing illicit opioid dependence. Providers, consumers, 5. Hser YI, Hoffman V, Grella CE, et al: A 33-year follow-up of narcotics addicts. use (see box on previous page). Re- and family members should be edu- Archives of General Psychiatry 58:503–508, search findings regarding the impact of cated about the benefits of MMT in 2001 MMT on many secondary outcomes, helping individuals manage opioid 6. Termorshuizen F, Krol A, Prins M, et al: such as mortality, drug-related HIV use disorders and about appropriate Long-term outcome of chronic drug use: risk behaviors, and criminal activity, ways to avoid the significant adverse the Amsterdam Cohort Study among Drug Users. American Journal of Epidemiology are less conclusive but suggest posi- events that can occur with metha- 161:271–279, 2005 tive trends. Finally, research has not done. Providers and consumers need 7. Thomas CP, Fullerton CA, Kim M, et al: conclusively shown positive impacts on to be educated regarding appropriate Medication-assisted treatment with sex-related HIV risk behaviors, non- doses to improve efficacy and appro- buprenorphine. Psychiatric Services, 2013; opioid illicit drug or alcohol use, or priate initiation to minimize adverse doi 10.1176/appi.ps.201300256 other social consequences. Methadone events. 8. Galanter M, Kleber H (ed): The American maintenance doses above 60 mg con- Because of MMT’srelativeefficacy, Psychiatric Publishing Textbook of Sub- stance Abuse Treatment, 4th ed. Arlington, fer greater efficacy in retention and efforts should be made to increase Va, American Psychiatric Publishing, 2008 suppression of illicit opioid use; how- access to MMT for all individuals who 9. D’Aunno T, Folz-Murphy N, Lin X: ever, there is limited evidence that struggle with opioid use disorders. Changes in methadone treatment practices: doses above 100 mg provide additional Directors of state mental health and results from a panel study, 1988–1995. benefits. No evidence has emerged substance abuse agencies and commu- American Journal of Drug and Alcohol Abuse 25:681–699, 1999 to delineate the duration of MMT nity health organizations should look beyond an indefinite period. Although for methods to increase access to 10. D’Aunno T, Pollack HA: Changes in methadone treatment practices: results MMT generally is believed to reduce MMT, and purchasers of health care from a national panel study, 1988–2000. mortality risk among individuals with services should cover appropriately JAMA 288:850–856, 2002 opioid dependence, methadone is monitored MMT. 11. Dougherty RH, Lyman DR, George P, also associated with significant ad- et al: Assessing the evidence base for be- verse events, such as respiratory havioral health services: introduction to Acknowledgments and disclosures the series. Psychiatric Services, 2013; doi depression and cardiac arrhythmias, in 10.1176/appi.ps.201300214 the presence of rapid titrations or other Development of the Assessing the Evidence Base Series was supported by contracts 12. Sees KL, Delucchi KL, Masson C, et al: risk factors. There is no clear evidence HHSS283200700029I/HHSS28342002T, Methadone maintenance for opioid de- that structured psychotherapy pro- HHSS283200700006I/HHSS28342003T, and pendence. JAMA 284:694–695, 2000 HHSS2832007000171/HHSS28300001T from vided in addition to the psychosocial 13. Strain EC, Bigelow GE, Liebson IA, et al: support normally offered at methadone 2010 through 2013 from the Substance Abuse Moderate- vs high-dose methadone in the and Mental Health Services Administration treatment centers conveys additional treatment of opioid dependence: a ran- (SAMHSA). The authors acknowledge the con- domized trial. JAMA 281:1000–1005, benefit. tributions of Robert Lubran, M.S., M.P.A., Kevin 1999 MMT improves pregnancy-related Malone, B.A, and Suzanne Fields, M.S.W., ’ 14. Schwartz RP, Highfield DA, Jaffe JH, et al: outcomes by reducing illicit drug use from SAMHSA; John O Brien, M.A., from the Centers for Medicare & Medicaid Services; A randomized controlled trial of interim and increasing treatment retention. methadone maintenance. Archives of General John Easterday, Ph.D., Linda Lee, Ph.D., – However, newborn infants of mothers Rosanna Coffey, Ph.D., and Tami Mark, Ph.D., Psychiatry 63:102 109, 2006 treated with methadone during preg- from Truven Health Analytics; and Sandrine 15. Schwartz RP, Kelly SM, O’Grady KE, et al: nancy may be born with NAS irre- Pirard, M.D., Ph.D., from National In- Interim methadone treatment compared to stituteonDrugAbuse.Theviewsexpressed standard methadone treatment: 4-month spective of the methadone dose used in this article are those of the authors and findings. Journal of Substance Abuse Treat- by the mothers. do not necessarily represent the views of ment 41:21–29, 2011 SAMHSA. Potential areas for future research 16. Schwartz RP, Kelly SM, O’Grady KE, et al: include increased focus on the impact The authors report no competing interests. Randomized trial of standard methadone

156 PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 treatment compared to initiating metha- infection. Cochrane Database of Systematic 35. Modesto-Lowe V, Brooks D, Petry N: done without counseling: 12-month find- Reviews 8:CD004145, 2011 Methadone deaths: risk factors in pain and ings. Addiction 107:943–952, 2012 addicted populations. Journal of General 27. Fareed A, Vayalapalli S, Stout S, et al: Internal Medicine 25:305–309, 2010 17. Wilson ME, Schwartz RP, O’Grady KE, Effect of methadone maintenance treat- et al: Impact of interim methadone main- ment on heroin craving, a literature re- 36. Jones HE, O’Grady KE, Malfi D, et al: tenance on HIV risk behaviors. Journal of view. Journal of Addictive Diseases 30: Methadone maintenance vs. methadone Urban Health 87:586–591, 2010 27–38, 2011 taper during pregnancy: maternal and neonatal outcomes. American Journal on 18. Kandall SR, Doberczak TM, Jantunen M, 28. Connock M, Juarez-Garcia A, Jowett S, Addictions 17:372–386, 2008 et al: The methadone-maintained pregnancy. et al: Methadone and buprenorphine for Clinics in Perinatology 26:173–183, 1999 the management of opioid dependence: 37. Kaltenbach K, Berghella V, Finnegan L: a systematic review and economic evaluation. Opioid dependence during pregnancy. 19. McCarthy JJ, Leamon MH, Parr MS, et al: Health Technology Assessment 11:1–171, iii– Effects and management. Obstetrics and High-dose methadone maintenance in iv, 2007 Gynecology Clinics of North America 25: pregnancy: maternal and neonatal out- 139–151, 1998 comes. American Journal of Obstetrics and 29. Fletcher BW, Battjes RJ: Introduction to Gynecology 193:606–610, 2005 the special issue: treatment process in 38. Hulse GK, Milne E, English DR, et al: The DATOS. Drug and Alcohol Dependence relationship between maternal use of her- 20. Pizarro D, Habli M, Grier M, et al: Higher 57:81–87, 1999 oin and methadone and infant birth maternal doses of methadone does not in- weight. Addiction 92:1571–1579, 1997 crease neonatal abstinence syndrome. Jour- 30. Amato L, Minozzi S, Davoli M, et al: nal of Substance Abuse Treatment 40: Psychosocial combined with agonist main- 39. Finnegan L, Kaltenbach K: Neonatal ab- 295–298, 2011 tenance treatments versus agonist main- stinence syndrome; in Primary Pediatric tenance treatments alone for treatment of Care. Edited by Hoekelman RA, Friedman 21. Mattick RP, Breen C, Kimber J, et al: opioid dependence. Cochrane Database SB, Nelson NM, et al. St Louis, Mo, Methadone maintenance therapy versus no of Systematic Reviews 10:CD004147, Mosby-Year Book, 1992 opioid replacement therapy for opioid de- 2011 pendence. Cochrane Database of System- 40. Ebner N, Rohrmeister K, Winklbaur B, atic Reviews 3:CD002209, 2009 31. Cleary BJ, DonnellyJ,StrawbridgeJ, et al: Management of neonatal abstinence et al: Methadone dose and neonatal ab- syndrome in neonates born to opioid main- 22. Hall W, Ward J, Mattick R: The Effective- stinence syndrome-systematic review and tained women. Drug and Alcohol De- ness of Methadone Maintenance Treatment meta-analysis. Addiction 105:2071–2084, pendence 87:131–138, 2007 1: Heroin Use and Crime. Amsterdam, 2010 Harwood Academic Publishers, 1998 41. Dashe JS, Sheffield JS, Olscher DA, et al: 32. Methadone-Associated Mortality: Report Relationship between maternal metha- 23. Faggiano F, Vigna-Taglianti F, Versino E, of a National Assessment, May 8–9, 2003. done dosage and neonatal withdrawal. et al: Methadone maintenance at dif- SAMHSA pub no 04-3904. Rockville, Md, Obstetrics and Gynecology 100:1244–1249, ferent dosages for opioid dependence. Substance Abuse and Mental Health 2002 Cochrane Database of Systematic Reviews Services Administration, Center for Sub- 3:CD002208, 2003 stance Abuse Treatment, 2004 42. American Academy of Pediatrics Commit- tee on Drugs: Neonatal drug withdrawal. 24. Fareed A, Casarella J, Amar R, et al: Metha- – done maintenance dosing guideline for opioid 33. Webster LR, Cochella S, Dasgupta N, Pediatrics 101:1079 1088, 1998 dependence, a literature review. Journal et al: An analysis of the root causes for 43. Kaltenbach K, Finnegan LP: Developmental of Addictive Diseases 29:1–14, 2010 opioid-related overdose deaths in the United States. Pain Medicine 12(Suppl 2): outcome of children born to methadone 25. MacArthur GJ, Minozzi S, Martin N, et al: S26–S35, 2011 maintained women: a review of longitudinal Opiate substitution treatment and HIV studies. Neurobehavioral Toxicology and transmission in people who inject drugs: 34. Martin JA, Campbell A, Killip T, et al; Teratology 6:271–275, 1984 systematic review and meta-analysis. Brit- Substance Abuse and Mental Health Ser- ish Medical Journal 345:e5945, 2012 vices Administration: QT interval screen- 44. Jones HE, Kaltenbach K, Heil SH, et al: ing in methadone maintenance treatment: Neonatal abstinence syndrome after 26. Gowing L, Farrell MF, Bornemann R, report of a SAMHSA expert panel. Jour- methadone or buprenorphine exposure. et al: Oral substitution treatment of inject- nal of Addictive Diseases 30:283–306, New England Journal of Medicine 363: ing opioid users for prevention of HIV 2011 2320–2331, 2010

PSYCHIATRIC SERVICES ' ps.psychiatryonline.org ' February 2014 Vol. 65 No. 2 157