Globally, the number of people who inject drugs (PWID) is relatively small, rarely exceeding 1% of total population in any country. Yet, in many communities and several regions, particularly Eastern Europe, Central, and Southeast Asia, injection drug use remains a primary driver in the AIDS epidemic.1–5 Outside of sub-Saharan Africa, injection drug use accounts for about one-third of all new HIV infections and approximately 3 million PWID are estimated to be infected with HIV.1,5 PWID are estimated to have 22 times the prevalence of HIV infection relative to the general population.6 Since most of these infections are related to the injection of heroin or other opioids, alone or in combination with stimulants or benzodiazepines, effective treatments for opioid dependence have important potential for preventing new HIV infections. Although there have been no randomized controlled trials, over 20 years of observational data on the impact of methadone maintenance treatment (MMT) provide “proof of concept” on the ability of medication-assisted treatment (MAT) to reduce opioid use, injection-related risk behaviors, and HIV infections.7–16 Yet, despite evidence of the efficacy of methadone treatment as an HIV prevention strategy, access remains extremely limited.3,13,17 Only a small proportion of PWID are currently receiving MAT and most drug users at risk of HIV infection have no access to any form of evidence-based treatments for opioid use.5,18
The use of buprenorphine/naloxone (BUP/NX) has expanded rapidly since its approval as a treatment for opioid addiction in the United States in 2002. Buprenorphine is a partial opioid agonist with strong affinity to the μ-opioid receptors of the central nervous system resulting in an active half-life of up to 60 hours, making dosing intervals of up to 3 days possible.19,20 Although buprenorphine alone was first introduced for opioid dependence treatment in Europe in 1996, naloxone was included in the tablet formulation as a deterrent to diversion and injection.21 When absorbed sublingually, buprenorphine maintains adequate bioavailability, whereas naloxone is poorly absorbed and rapidly metabolized. However, if a BUP/NX tablet is dissolved for use by injection, naloxone will remain active and precipitate withdrawal. Overdose from BUP/NX alone is uncommon and withdrawal symptoms after dose reduction are less severe in intensity and duration than pure agonist medications.22–24 Similar to other opioids, side effects include nausea, vomiting, constipation, sedation, and temporary elevations in liver transaminases.
Given the critical need to expand MAT options for HIV prevention among opioid injectors, HIV Prevention Trials Network 058 (HPTN 058) was designed to compare 2 novel, 1-year, community-based opioid treatment strategies using BUP/NX and counseling to reduce incident HIV infections and mortality among opioid-dependent injectors.
HPTN 058, a 1:1 randomized open-label trial, compared two 12-month opioid treatment strategies combining medication and drug counseling in opioid-dependent PWID in China and Thailand. Those assigned to the short-term medication-assisted treatment arm (ST-MAT) received a 3-day induction to BUP/NX followed by up to 15 days of dose reduction detoxification. ST-MAT participants also received weekly drug counseling for the first 3 months of treatment and then offered monthly for the remaining 9 months of the intervention.25,26 Recognizing that opioid dependence is a chronic disease with a high likelihood of relapse, at week 26, participants in the ST-MAT arm were offered a second opportunity for detoxification. The LT-MAT arm received BUP/NX induction and dosage stabilization followed by directly observed BUP/NX administration in the clinic 3 times per week for 46 weeks. At week 47, BUP/NX dose tapering began with final dosing completed by the end of week 52. LT-MAT participants received 12 months of drug counseling on the same schedule as the ST-MAT arm. Both study arms were followed for an additional 52 weeks after treatment completion to evaluate long-term impact. The primary end point of the study was incident HIV infection or death at week 104, 1 year after the completion of the intervention. The protocol was designed to measure impact during the intervention year and to assess durability of effect after treatment completion. Interim monitoring rules established before the initiation of the study included planned analyses by the independent Data Safety and Monitoring Board (DSMB) after 25%, 50%, and 75% of the participants had completed the 104-week assessment. The protocol was reviewed and approved by ethics committees and institutional review boards at each local site and affiliated institutions.
Study participants were active opioid injectors recruited from the community at 4 sites: Chiang Mai, Thailand (n = 202); Nanning (n = 161) and Heng County (n = 411) in Guangxi, China and Urumqi (n = 477), in Xinjiang, China. Sites were selected based on prior evidence of HIV incidence among PWID ranging from 2%–9% per year.27,28 All study sites had access to methadone treatment, harm reduction services, and HIV treatment. Targeted outreach and recruitment activities were guided by local epidemiologic information and knowledge of the drug-using community.
Individuals were eligible if they were at least 18 years of age, HIV seronegative, met criteria for opioid dependence (DSM-IV),29 had a positive urine drug test for opioids, and reported injecting 12 or more times in the prior 28 days. Exclusion criteria included enrollment in MMT, known allergy to BUP/NX, current alcohol or benzodiazepine dependence, pregnancy or breastfeeding, unwillingness to use contraception if female, alanine aminotransferase >3 times upper limit of normal, hemoglobin <8 g/dL (if male) or 7 g/dL (if female), bilirubin >2.5 times upper limit of normal, or platelet count <50,000/mm3.
HPTN 058 represented the first use of BUP/NX in both Thailand and China. Consequently, recruitment of participants included activities and materials devoted to community and participant education. Each site had an active community advisory board that included representatives from local advocacy groups, family members, public health, and criminal justice agencies. Before project implementation, community advisory boards and community representatives were invited to participate in formal meetings designed to solicit feedback on the purpose of the study, its procedures, the interventions, and the strategies used to protect the safety and confidentiality of study participants. The design included a safety phase and review at each site after the first 50 participants completed 1 month of treatment. No unexpected safety concerns were identified at any site and enrollment proceeded uninterrupted.30
Participants in both study arms began with a 3-day outpatient induction period. In accordance with established clinical guidelines, participants were instructed not to use opioids for 24 hours before induction, which must be initiated during withdrawal. Withdrawal symptoms were assessed using the Clinical Opioid Withdrawal Scale.31,32 Those with a score of 8 or greater were judged ready for induction and received a 4-mg dose of sublingual BUP/NX. At 1 hour postdose, if the Clinical Opioid Withdrawal Scale score was 2 or greater, a second 4-mg dose was given. Days 2 and 3 followed similar procedures with provision for up to 16 mg and 32 mg, respectively. After induction, participants assigned to ST-MAT completed detoxification with an average dose tapering of 2 mgs per day for up to 15 days. Participants assigned to LT-MAT, after establishing the maintenance dose that prevented the emergence of withdrawal symptoms, continued receiving BUP/NX for 48 weeks, attending the clinic 3 times each week for directly observed dosing. At week 48, tapering of BUP/NX began, with dose reduction of 1–2 mg per day with all medication administration completed before week 52. Dosing visits were carefully monitored and reinduction was required if more than 3 consecutive visits were missed. Women who became pregnant during the BUP/NX treatment phase were tapered from the medication and referred to MMT but continued in the counseling intervention and were followed through delivery and the completion of the study. At week 26 assessment, participants in the ST-MAT arm who had relapsed to opioid use were offered a second induction and detoxification.
All study participants received behavioral drug and risk counseling.25,26 Behavioral drug and risk counseling is rooted in cognitive behavioral theory and focused on helping participants understand their addiction as a manageable medical condition. Counseling was delivered in individual sessions of 30–45 minutes, guided by a manual, and designed to help participants develop and implement meaningful strategies for controlling drug use and avoiding injection and sexual risk. Counselors helped participants establish short-term behavioral goals that promoted completion of BUP/NX administration visits, risk reduction, avoidance of “triggers” for craving and use, and initiation of rewarding behaviors incompatible with continued drug use. Counselors developed behavioral contracts to guide goal-directed behavior between counseling sessions.
All participants completed interviewer-administered assessments of injection and noninjection drug use, sexual and drug-related risk behaviors, community treatment utilization, HIV testing, and plasma storage every 26 weeks throughout the study. At baseline and weeks 12, 26, 40, and 52, specimens were collected to assess liver function (alanine aminotransferase and total bilirubin). Testing for hepatitis B and C was performed at screening and week 26, and additional testing for hepatitis C was performed at subsequent semiannual follow-up visits if previously negative. Urine drug screens and pregnancy tests were assessed monthly and semiannually during the 12-month intervention period. Participants were followed for a minimum of 104 weeks, and a maximum of 156 weeks based on their time of enrollment.
HIV testing was completed using 2 locally available rapid HIV test kits that had undergone full validation and were subject to routine proficiency testing according to HPTN protocol. Baseline HIV status and all new HIV infections were verified by the HPTN Network Laboratory using stored plasma and Food and Drug Administration (FDA) approved HIV methodologies. A commercial point of care test with a sensitivity of 300 ng/mL was used for detection of urine opioids and other drugs (Integrated E-Z Split Key Cup; ACON Laboratories, San Diego, CA).
The study had 90% power to detect a 50% reduction in cumulative HIV infection and death at 104 weeks, assuming an HIV infection risk such that 7.25% of the ST-MAT arm would be infected or have died by 104 weeks.
The study was reviewed at least annually by an independent DSMB appointed by the sponsor. At the first scheduled interim analyses, occurring October 4, 2011, when 25% of the participants had completed 104 weeks on study, the DSMB halted the study due to futility as a result of lower than anticipated HIV incidence rates. The present analysis includes data from visits up through October 4, 2011.
Cumulative incidence rates and confidence intervals (CIs) were calculated based on Poisson distribution with time to infection or death. Cox proportional hazards model was used to compute hazards ratios. Monthly urine screens and injection and related risk behaviors were compared between arms using logistic regression at each visit, adjusted for site.
A total of 2268 PWID were screened for eligibility and 1251 were enrolled and randomized between May 2007 and October 2011, 627 on the ST-MAT, and 623 on the LT-MAT (Fig. 1). At the time the study was halted, 470 ST-MAT participants and 477 in the LT-MAT completed at least 1 follow-up HIV test and are included in the primary intent to treat analysis.
As shown in Table 1, participants were predominantly (92%) male with a median age of 33 years. Over 40% were from ethnic minorities and a slight majority were married and living with a spouse or partner. Participants had been injecting for a median of 7 years with a median age at first injection of 25 years. During the month before enrollment, 90% reported injecting heroin and 11% reported injecting opium. The median number of injections in the month before enrollment was 84 [interquartile range (IQR), 60–90], or about 3 times per day. As required for enrollment, 100% of the participants tested positive for opioids. Methadone (3%), amphetamine (2%), and benzodiazepines (19%) were detected by urine drug screen.
The 3-day induction was initiated by all participants (with 1 exception) and completed by 89% (88% in the ST-MAT and 91% in the LT-MAT arm) (Table 2). In the ST-MAT arm, 80% of participants completed detoxification. At week 26, 153 of 440 (35%) ST-MAT participants initiated a second detoxification; 85% of these completed this detoxification. LT-MAT participants received observed BUP/NX doses for a median (IQR) of 170 days (IQR, 63–308).
Among the LT-MAT participants, 74% were retained in treatment through weeks 25–28 (ie, completed BUP/NX medication visits during or after the period) with 55% receiving at least 75% of expected doses in this time window. By the final month of intervention (weeks 45–48), 55% remained in treatment, with 43% receiving at least 75% of expected doses.
Among those assigned to ST-MAT, 66% completed at least 9 of 12 weekly counseling sessions, compared with 84% in the LT-MAT arm (P < 0.001). Similarly, 59% of those in the ST-MAT arm completed 6 of 9 monthly counseling sessions compared with 70% of the LT-MAT (P < 0.001).
HIV Infection and Death
Through week 104, 5 incident HIV infections and 9 deaths were identified among those in the ST-MAT arm and 2 incident HIV infections and 8 deaths among LT-MAT participants. With 642 person-years (PY) in the ST-MAT arm and 658 in the LT-MAT arm, the composite rates of HIV infection or death were 1.9 (95% CI: 1.0 to 3.3) and 1.5 (95% CI: 0.7 to 2.8) per 100 PY, for a hazard ratio of LT-MAT versus ST-MAT of 0.69 (CI: 0.31 to 1.56, P = 0.38).
Drug Use Outcomes
Monthly and semiannual urine toxicology tests are shown in Figure 2. At week 24, 23% (103/445) in the ST-MAT arm and 56% (257/461) in the LT-MAT tested negative for opioids (P < 0.001). At the initiation of dose tapering in week 48, opioid-negative toxicology results were found in 33% (78/237) of ST-MAT compared with 58% (155/268) of LT-MAT participants (P < 0.001); at completion of tapering (week 52), this remained at 32% (107/338) among ST-MAT participants but decreased to 46% (163/357) in the LT-MAT arm (P < 0.001). By 78 and 104 weeks, ST-MAT and LT-MAT arms had similar rates of opioid-negative tests: at week 78, 35% (100/283) in the ST-MAT and 32% (89/279) in the LT-MAT arm (P = 0.38); at week 104, 30% and 34% in the ST-MAT and LT-MAT arms, respectively (P = 0.43).
As shown in Figure 3A, self-reported drug use was consistent with urinalyses results; at week 26, 26% of participants in the ST-MAT arm and 58% in the LT-MAT arm reported no injection in the prior 30 days (P ≤ 0.001). At the end of the treatment phase of the study (week 52), 43% of the ST-MAT and 59% of the LT-MAT participants reported no injection during the prior month (P ≤ 0.001). No significant differences between arms were found during the second year: at 78 weeks, 48% of ST-MAT and 50% of LT-MAT participants reported no injection (P = 0.7), and at 104 weeks, 53% of ST-MAT and 49% of LT-MAT reported no injection use (P = 0.4).
Risk Behavior Outcomes
Self-reported injection-related risk behaviors showed significant reduction in both arms after baseline that was sustained through the treatment phase and the subsequent year of follow-up. Figures 3B–D show this pattern for injecting with more than 1 person, use of the same syringe/needle after another PWID, and sharing cookers, cottons, and rinse water. Significantly, fewer injections were reported by LT-MAT participants at weeks 26 and 52. Also, fewer LT-MAT participants reported injecting with more than 1 person compared with ST-MAT.
Because of the small number of new HIV infections observed in the study, we were unable to evaluate whether 1 year of BUP/NX and counseling was more effective in preventing HIV infection and death than short-term BUP/NX and counseling. Despite this, those with 1 year of access to BUP/NX in the LT-MAT arm had a significantly higher sustained reduction in opioid use and injection compared with those in the ST-MAT arm, as measured by both urinalyses and self-report, indicating a reduced risk of exposure to HIV. The differential rates of opioid use between arms disappeared by week 78, although 30% or more remained opioid free.
The study was designed to assess the durability of the LT-MAT intervention after 1 year of treatment and it is apparent that relapse to opioid use commenced as soon as the tapering began. Thrice-weekly access to BUP/NX achieved significant reductions in opioid use and injection during treatment. Since these reductions were not sustained after cessation of BUP/NX and counseling, our data clearly demonstrate the necessity of continued access to medication beyond 1 year to sustain reductions in opioid use for many PWID.
The reduction in opioid use among participants in the LT-MAT arm during the treatment phase has important implications for HIV prevention. Research conducted over the past 20 years clearly demonstrates that reduced opioid use and injection-related risk behaviors among opioid users in MAT are associated with lower risk of HIV acquisition.16 In addition, access to and retention in MAT by HIV-positive individuals has been associated with increased retention in antiretroviral treatment and viral suppression, which itself has been associated with a reduced risk of HIV transmission.33–36 Past research among PWID has demonstrated that continued drug use inhibits these objectives. Cessation of injection through effective treatments (similar to that reported here) has been shown to increase access and improve retention and adherence to antiretroviral treatment.35,37–42 Importantly, these positive impacts occur among injectors who cease injection and not merely stop sharing syringes or enter treatment.38
Injection-related risk behaviors were reduced among participants in both arms. Significant reductions in self -reports of injection, frequency of injection, needle sharing (reuse of a syringe after another injector), injection with others and sharing of cookers, cotton, and rinse water were observed between baseline and the assessment at 6 months in both arms. These behavioral changes were sustained through the intervention year and extended throughout the year after cessation of BUP/NX and counseling. The self-reported reductions in drug use and risk behavior among participants in both groups are consistent with the low rate of HIV incidence observed during the study.
Reductions in injection-related risk behaviors must be considered in light of the counseling intervention that was available to participants in both arms. The approach and content were the same and attendance had potential to be clinically meaningful. Those assigned to the LT-MAT arm completed significantly more counseling sessions, suggesting that the regular contact required for medication administration also facilitated counseling attendance. Since participants in both arms received the same counseling, it is not possible to directly measure its impact on drug use and risk behaviors in this study. However, both arms displayed similar patterns of risk reduction. The data reported here support the position that counseling can contribute to risk reduction even among those who continue to use drugs.
Study sites were chosen based on earlier studies in these same communities that found HIV incidence rates as high as 8% per year.27,28 During the intervening time, HIV testing was expanded, access to antiretroviral treatment was scaled up, and methadone treatment was introduced in each of the 3 communities in China. It is likely that these public health programs contributed to the low HIV incidence observed in HPTN 058.
The findings reported here should be considered in light of several limitations. First, participants in this study were recruited from the community and volunteered to receive treatment for their addiction. Consequently, data from this trial may not generalize to all opioid-dependent injectors. Despite efforts to include women, 92% of the participants were men, and we cannot be certain that the treatment outcomes apply to female opioid-dependent injectors. Data on frequency of injection, injections with others, and sharing of syringes and other injection equipment can only be measured by self-report. However, it is important to note that we observed a high correlation between self-reported opioid use and urine toxicology results. Additionally, the decline in opioid use and injection in both arms of the study may reflect a regression to the mean bias since all participants had to have a positive opioid test to be enrolled. Finally, the study was discontinued prematurely due to lower than expected event rates in both arms of the study, and we were unable to address the primary outcome of the trial.
MAT approaches are based on a body of evidence showing that drug dependence is most effectively treated when viewed as a chronic medical condition with both biological and behavioral components. As in the treatment of other chronic medical conditions, disease management is more likely to be successful when appropriate medications are used in combination with behavioral interventions. In the United States, data on the efficacy of BUP/NX treatment coupled with its safety profile led to FDA approval for administration through office-based practice in which physicians with brief training can provide prescriptions for pharmacy distribution and self-administration. For some opioid-dependent individuals, however, the clinic-based directly observed medication administration used in HPTN 058 may be a more effective delivery strategy and lead to increased adherence and participation in counseling.
The participation we achieved in both medication and counseling is comparable or superior with that seen in other drug treatment modalities and provides evidence of the feasibility and acceptability of this strategy.33 Thrice-weekly directly observed dosing, combined with counseling, has a much lower burden on both patients and staff than daily MMT and has not previously been assessed in a large-scale community-based trial. This intervention was safe and effective for reducing both opioid dependence and injection-related risk. Access to MATs for opioid-dependent individuals who are at risk for HIV and other blood-borne infections is extremely limited, particularly in areas where HIV transmission is most commonly linked to injection drug use.5,17 It is hoped that the success of the BUP/NX strategy tested in HPTN 058 will lead to expanded coverage of effective MAT as HIV prevention.
1. Beyrer C, Malinowska-Sempruch K, Kamarulzaman A, et al.. HIV in people who use drugs 7 Time to act: a call for comprehensive responses to HIV in people who use drugs. Lancet. 2010;376:551–563.
2. De Cock KM, Jaffe HW, Curran JW. The evolving epidemiology of HIV/AIDS. AIDS. 2012;26:1205–1213.
3. Vlahov D, Robertson AM, Strathdee SA. Prevention of HIV infection among injection drug users in Resource-limited Settings. Clin Infect Dis. 2010;50:S114–S121.
4. Kilmarx PH. Global epidemiology of HIV. Curr Opin HIV AIDS. 2009;4:240–246.
5. Mathers BM, Degenhardt L, Ali H, et al.. HIV prevention, treatment, and care services for people who inject drugs: a systematic review of global, regional, and national coverage. Lancet. 2010;375:1014–1028.
6. Bergenstrom AM, Abdul-Quader AS. Injection drug use, HIV and the current response in selected low-income and middle-income countries. AIDS. 2010;24:S20–S29.
7. Gowing L, Farrell MF, Bornemann R, et al.. Oral substitution treatment of injecting opioid users for prevention of HIV infection. Cochrane Database Syst Rev. 2011;8:CD004145.
8. Metzger DS, Woody GE, McLellan AT, et al.. Human-Immunodeficiency-Virus seroconversion among intravenous-drug users in-of-treatment and out-of-treatment—an 18-month prospective follow-up. J Acqui Immune Defic Syndr. 1993;6:1049–1056.
9. Gowing LR, Farrell M, Bornemann R, et al.. Methadone treatment of injecting opioid users for prevention of HIV infection. J Gen Intern Med. 2006;21:193–195.
10. Qian HZ, Hao C, Ruan YH, et al.. Impact of methadone on drug use and risky sex in China. J Subst Abuse Treat. 2008;34:391–397.
11. Thiede H, Hagan H, Murrill CS. Methadone treatment and HIV and hepatitis B and C risk reduction among injectors in the Seattle area. J Urban Health. 2000;77:331–345.
12. Metzger DS, Woody GE, O'Brien CP. Drug treatment as HIV prevention: a research update. J Acquir Immune Defic Syndr. 2010;55:S32–S36.
13. Sullivan LE, Metzger DS, Fudala PJ, et al.. Decreasing international HIV transmission: the role of expanding access to opioid agonist therapies for injection drug users. Addiction. 2005;100:150–158.
14. Sorensen JL, Copeland AL. Drug abuse treatment as an HIV prevention strategy: a review. Drug Alcohol Depend. 2000;59:17–31.
15. Farrell M, Gowing L, Marsden J, et al.. Effectiveness of drug dependence treatment in HIV prevention. Int J Drug Policy. 2005;16:S67–S75.
16. MacArthur GJ, Minozzi S, Martin N, et al.. Opiate substitution treatment and HIV transmission in people who inject drugs: systematic review and meta-analysis. BMJ. 2012;345:e5945.
17. Mathers BM, Degenhardt L, Adam P, et al.. Estimating the Level of HIV prevention coverage, knowledge and protective behavior among injecting drug users: what does the 2008 UNGASS reporting round tell us? J Acquir Immune Defic Syndr. 2009;52:S132–S142.
18. Degenhardt L, Mathers B, Vickerman P, et al.. HIV in people who use drugs 2 Prevention of HIV infection for people who inject drugs: why individual, structural, and combination approaches are needed. Lancet. 2010;376:285–301.
19. Amass L, Kamien JB, Mikulich SK. Thrice-weekly supervised dosing with the combination buprenorphine-naloxone tablet is preferred to daily supervised dosing by opioid-dependent humans. Drug Alcohol Depend. 2001;61:173–181.
20. Johnson RE, Jones HE, Fischer G. Use of buprenorphine in pregnancy: patient management and effects on the neonate. Drug Alcohol Depend. 2003;70(2 suppl):S87–S101.
21. Emmanuelli J, Desenclos JC. Harm reduction interventions, behaviours and associated health outcomes in France, 1996-2003. Addiction. 2005;100:1690–1700.
22. Johnson RE, Strain EC, Amass L. Buprenorphine: how to use it right. Drug Alcohol Depend. 2003;70(2 suppl):S59–S77.
23. Bridge TP, Fudala PJ, Herbert S, et al.. Safety and health policy considerations related to the use of buprenorphine/naloxone as an office-based treatment for opiate dependence. Drug Alcohol Depend. 2003;70(2 suppl):S79–S85.
24. Fudala PJ, Woody GW. Recent advances in the treatment of opiate addiction. Curr Psychiatry Rep. 2004;6:339–346.
25. Chawarski MC, Mazlan M, Schottenfeld RS. Behavioral drug and HIV risk reduction counseling (BDRC) with abstinence-contingent take-home buprenorphine: a pilot randomized clinical trial. Drug Alcohol Depend. 2008;94:281–284.
26. Chawarski MC, Zhou W, Schottenfeld RS. Behavioral drug and HIV risk reduction counseling (BDRC) in MMT programs in Wuhan, China: a pilot randomized clinical trial. Drug Alcohol Depend. 2011;115:237–239.
27. Wei L, Chen J, Rodolph M, et al.. HIV incidence, retention, and changes of high-risk behaviors among rural injection drug users in Guangxi, China. Subst Abus. 2006;27:53–61.
28. Zhang Y, Shan H, Trizzino J, et al.. HIV incidence, retention rate, and baseline predictors of HIV incidence and retention in a prospective cohort study of injection drug users in Xinjiang, China. Int J Infect Dis. 2007;11:318–323.
29. Woody G, Schuckit M, Weinrieb R, et al.. A review of the substance use disorders section of the DSM-IV. Psychiatr Clin North Am. 1993;16:21–32.
30. Lucas GM, Beauchamp G, Aramrattana A, et al.. Short-term safety of buprenorphine/naloxone in HIV-seronegative opioid-dependent Chinese and Thai drug injectors enrolled in HIV Prevention Trials Network 058. Int J Drug Policy. 2012;23:162–165.
31. Wesson DR, Ling W. The clinical opiate withdrawal scale (COWS). J Psychoactive Drugs. 2003;35:253–259.
32. Tompkins DA, Bigelow GE, Harrison JA, et al.. Concurrent validation of the Clinical Opiate Withdrawal Scale (COWS) and single-item indices against the Clinical Institute Narcotic Assessment (CINA) opioid withdrawal instrument. Drug Alcohol Depend. 2009;105:154–159.
33. Fiellin DA, Weiss L, Botsko M, et al.. Drug treatment outcomes among HIV-infected opioid-dependent patients receiving buprenorphine/naloxone. J Acquir Immune Defic Syndr. 2011;56(suppl 1):S33–S38.
34. Uhlmann S, Milloy MJ, Kerr T, et al.. Methadone maintenance therapy promotes initiation of antiretroviral therapy among injection drug users. Addiction. 2010;105:907–913.
35. Wood E, Hogg RS, Kerr T, et al.. Impact of accessing methadone on the time to initiating HIV treatment among antiretroviral-naive HIV-infected injection drug users. AIDS. 2005;19:837–839.
36. Cohen MS, Chen YQ, McCauley M, et al.. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505.
37. Roux P, Carrieri MP, Villes V, et al.. The impact of methadone or buprenorphine treatment and ongoing injection on highly active antiretroviral therapy (HAART) adherence: evidence from the MANIF2000 cohort study. Addiction. 2008;103:1828–1836.
38. Roux P, Carrieri MP, Cohen J, et al.. Retention in opioid substitution treatment: a major predictor of long-term virological success for HIV-infected injection drug users receiving antiretroviral treatment. Clin Infect Dis. 2009;49:1433–1440.
39. Wood E, Milloy MJ, Montaner JSG. HIV treatment as prevention among injection drug users. Curr Opin HIV AIDS. 2012;7:151–156.
40. Wood E, Kerr T, Tyndal MW, et al.. A review of barriers and facilitators of HIV treatment among injection drug users. AIDS. 2008;22:1247–1256.
41. Volkow ND, Baler RD, Normand JL. The unrealized potential of addiction science in curbing the HIV epidemic. Curr HIV Res. 2011;9:393–395.
42. Spire B, Lucas GM, Carrieri MP. Adherence to HIV treatment among IDUs and the role of opioid substitution treatment (OST). Int J Drug Policy. 2007;18:262–270.