Although treatment for drug abuse has traditionally

Transcription

1 A Randomized Trial of Buprenorphine Maintenance for Heroin Dependence in a Primary Care Clinic for Substance Users versus a Methadone Clinic Patrick G. O Connor, MD, MPH, Alison H. Oliveto, PhD, Julia M. Shi, MD, Elisa G. Triffleman, MD, Kathleen M. Carroll, PhD, Thomas R. Kosten, MD, Bruce J. Rounsaville, MD, Juliana A. Pakes, Richard S. Schottenfeld, MD PURPOSE: Buprenorphine is an alternative to methadone for the maintenance treatment of heroine dependence and may be effective on a thrice weekly basis. Our objective was to evaluate the effect of thrice weekly buprenorphine maintenance for the treatment of heroin dependence in a primary care clinic on retention in treatment and illicit opioid use. SUBJECTS AND METHODS: Opioid-dependent patients were randomly assigned to receive thrice weekly buprenorphine maintenance in a primary care clinic that was affiliated with a drug treatment program (n 23) or in a traditional drug treatment program (n 23) in a 12-week clinical trial. Primary outcomes were retention in treatment and urine toxicology for opioids; secondary outcomes were opioid withdrawal symptoms and toxicology for cocaine. RESULTS: Retention during the 12-week study was higher in the primary care setting (78%, 18 of 23) than in the drug treatment setting (52%, 12 of 23; P 0.06). Patients admitted to primary care had lower rates of opioid use based on overall urine toxicology (63% versus 85%, P 0.01) and were more likely to achieve 3 or more consecutive weeks of abstinence (43% versus 13%, P 0.02). Cocaine use was similar in both settings. CONCLUSIONS: Buprenorphine maintenance is an effective treatment for heroin dependence in a primary care setting. Am J Med. 1998;105: by Excerpta Medica, Inc. Although treatment for drug abuse has traditionally been provided in specialized drug treatment programs, there are compelling reasons for developing alternative settings for this treatment. Drug abuse treatment services are provided to less than 10% of drug abusers in the United States (1). Methadone maintenance treatment slots are in particularly short supply in many US cities (2). Substance abusers reluctance to enter treatment and fear of additional stigmatization may contribute to this problem (3). Because of the risk of medical problems associated with drug abuse, many untreated drug abusers receive medical care in primary care settings (4,5), which may represent an important avenue for expanding access to drug abuse treatment (6). Primary care-based acute opioid detoxification using clonidine and naltrexone is feasible and effective (7,8), and stable long-term methadone maintenance patients From the Departments of Medicine (PGO, JMS) and Psychiatry (AEO, EGT, KMC, TRK, BJR, JAP, RSS), Yale University School of Medicine, New Haven, Connecticut. Supported by National Institute on Drug Abuse 5-R18-DA Requests for reprints should be addressed to Patrick G. O Connor, MD, MPH, Yale University School of Medicine, Primary Care Center Suite A, 333 Cedar Street, P.O. Box , New Haven, Connecticut Manuscript submitted January 16, 1998 and accepted in revised form March 20, may be successfully transferred to methadone maintenance administered in a primary care setting (9,10). However, methadone maintenance therapy is restricted to federally licensed programs (11,12). Having opioid maintenance as well as detoxification approaches available for selected patients to manage opioid dependence in primary care settings could substantially enhance access to drug treatment services (13). Buprenorphine is a new alternative to methadone for maintenance treatment of opioid dependence (14). Buprenorphine, which is currently an investigational drug for the treatment of opioid dependence, decreases opioid self-administration in dependent individuals (15,16). Clinical trials that have directly compared buprenorphine and methadone have supported its efficacy when administered in sufficient doses (eg, 12 to 16 mg/day) (17,18). A partial opioid agonist, buprenorphine may have important advantages compared with methadone as it is easier to withdraw from and less likely to cause overdose (19). Some studies have suggested that buprenorphine may also lead to decreased cocaine use among opioid-dependent individuals (20,21). In addition, buprenorphine may be effective when administered on a less than daily basis (22 24). In our pilot study, thrice weekly primary care-based buprenorphine maintenance was demonstrated to be feasible and safe: 5 of 7 patients remained in treatment for 6 months, 4 of whom had by Excerpta Medica, Inc /98/$19.00 All rights reserved. PII S (98)

2 mostly opioid-free urine samples (25). We therefore conducted a 12-week randomized clinical trial to compare the effects of thrice weekly buprenorphine in a primary care setting with those in a traditional drug treatment setting on retention in treatment and rates of opioid use. METHODS Patients Patients were recruited from a methadone maintenance program waiting list and word-of-mouth referrals. Inclusion criteria were current opioid dependence according to DSM III-R criteria (26) and opioid-positive urine toxicology. Exclusion criteria were current DSM III-R diagnosis of other drug or alcohol dependence (other than tobacco), current cocaine use (defined as any use within the past month by self-report or by a positive urine toxicology for cocaine at the time of screening for admission to the study), complex medical comorbidity (eg, major cardiovascular, renal or gastrointestinal disease), complex psychiatric comorbidity (eg, psychosis, suicidality, schizophrenia, major depression), and pregnancy. Treatment Settings Eligible patients were randomly assigned to either a primary care clinic or a methadone maintenance clinic. Primary care setting. Primary care-based buprenorphine maintenance was administered at the Central Medical Unit, a licensed, free-standing primary care clinic (27) that provides care for substance users who are enrolled in drug treatment programs affiliated with the Yale University Substance Abuse Treatment Unit, as well as for patients with psychiatric disorders enrolled in an affiliated mental health program. Approximately 1,300 patients receive medical services from the Central Medical Unit, which is staffed by primary care physicians (general internists), nurse practitioners, and physician associates. We utilized a manual-guided clinical management protocol developed by Fawcett and colleagues (28) adapted for use in the treatment of drug abuse (25,29). The protocol included taking a complete history, identifying medical problems, providing advice about changing behavior, prescribing medication, and referring patients to services. The initial session (week 1), designed to correspond to a new patient visit in primary care, was up to 1 hour in length to assure that patients received full clinical and research assessment. In this session the patient met the clinician who provided care during the entire course of therapy. The goals of this session were to review the medical and substance abuse history and complications associated with the patient s substance abuse; to develop a substance abuse treatment plan including the benefits and risks of buprenorphine, and a treatment plan for medical issues; to provide a referral to group therapy; and to review program guidelines, and goals for subsequent sessions. Subsequent 20-minute sessions (weeks 2 through 12) corresponded to the follow-up visit format employed in primary care settings. These sessions reviewed substance use since the prior visit, urine toxicology results, commitment to treatment, attendance at group sessions, medication use, and medical and psychosocial issues, to determine goals for subsequent sessions. Patients met weekly in a 50-minute semistructured group session that was facilitated by a primary care nurse practitioner, and that utilized a self-help approach in which patients attempted to learn from each other concerning strategies to promote abstinence. Drug treatment setting. The drug treatment site was the Legion Avenue Methadone Maintenance Program, which has been in operation since 1968 and has a census of over 300 patients. The clinic has been the site of several studies of buprenorphine maintenance (18,20). Patients randomly assigned to buprenorphine maintenance in the drug treatment setting received the standard set of services provided to patients in the clinic including assignment to a substance abuse counselor, medication, urine toxicology, and group therapy sessions. During the initial session (week 1), patients attended an extended intake session with their counselor (2 to 3 hours in length). Thereafter they attended clinic thrice weekly to receive medication, provide urine samples, and complete a selfreport detailing drug use and opioid withdrawal symptoms. Once weekly, blood pressure and side effects were also evaluated, and patients attended a 60-minute group therapy session with their counselors, based on a 12-week manual-guided relapse prevention program designed for use in alcoholic patients (30) and modified for methadone-maintained individuals (31). On a monthly basis and as needed, patients met individually with their counselors to discuss psychosocial issues and their treatment plan. Buprenorphine administration. In both treatment settings, sublingual liquid buprenorphine administration occurred in two phases. During buprenorphine stabilization in week 1, treatment was begun at a dose of 2 mg on day 1 and gradually increased to 4 mg (day 2), 8 mg (day 3), 16 mg (day 4), and 32 mg (day 5). During buprenorphine maintenance in weeks 2 to 13, treatment was administered thrice weekly at doses of 22 mg on Monday and Wednesday and 40 mg on Friday. Clinicians dispensing buprenorphine observed the patients, who were instructed to hold it under their tongue for up to 5 minutes to assure adequate absorption. Upon completion of the study, buprenorphine was tapered during a 3-week period, and patients were referred to drug treatment programs in the community. Study outcomes. The primary outcomes of interest for this study were retention in treatment and urine toxicol- August 1998 THE AMERICAN JOURNAL OF MEDICINE Volume

3 ogy for opioids. Patients were discharged from the study if they missed medication on three successive occasions or if they failed to attend a weekly group therapy session. Discharged patients were referred to alternative drug treatment programs in the community. Urine toxicology testing was performed 3 times a week for the presence of opioids and cocaine using the Abbot Tdx system (32). Withdrawal symptoms were measured using an Opioid Withdrawal Symptom Scale that assessed self-reported experience of each of 17 opioid withdrawal symptoms on a 0 to 3 scale (none, mild, moderate, severe) (33). Statistical Analysis Baseline characteristics were compared with chi-square and t tests. Differences in retention rates were analyzed using the Kaplan-Meier product limit method and the generalized Wilcoxon test (34,35). Random regression models were chosen for the principal analysis of the effect of setting (primary care clinic or drug treatment program) on urine toxicology results and opioid withdrawal symptoms (36). This approach was designed for unbalanced repeated measures with missing data, allowing for intrasubject serial correlation and unequal variance and covariance structures across time, by incorporating data for each individual group. We estimated 95% confidence intervals (CI) to compare differences in rates or proportions (as relative risks [RR]). The BMDP statistical package was used. The primary intention to treat analyses were conducted on all 46 enrolled patients who met eligibility criteria. RESULTS Baseline Patient Characteristics Of 46 heroin-dependent individuals enrolled in this study, 23 were assigned to each treatment group. Baseline sociodemographic and clinical features of the treatment groups were similar (Table). Most patients had used heroin intravenously during the past 30 days. Both groups had similar experience with prior methadone maintenance treatment. Treatment Outcomes Overall, 65% (30 of 46) of patients completed treatment. Of the 16 patients who left treatment, 2 were transferred to other treatment programs for reasons not directly related to the treatment protocol. One patient in the primary care group was transferred to inpatient treatment after becoming suicidal, and 1 patient in the drug treatment setting was transferred to methadone maintenance for medical reasons. The remainder (14) dropped out of treatment prior to the end of the study. The proportions of patients remaining in treatment are shown in Figure 1. Rates of completion of the 12-week trial were 78% (18 of 23) in the primary care setting and 52% (12 of 23) in the drug treatment setting (RR 1.5, Table. Baseline Demographic and Clinical Characteristics of the Study Groups Primary Care (n 23) (%) Drug Treatment (n 23) (%) Age Male Caucasian Never married Education (years) Currently employed Heroin use in past 30 days Intravenous drug use in past 30 days Previous methadone maintenance Days of alcohol use in past 30 days Plus-minus values are mean SD. There were no significant differences between the two groups. 95% CI 0.96 to 2.3, P 0.06). There were significant effects of treatment setting and time on the rates of opioid-positive urine samples (Figure 2). Mean rates were lower in the primary care clinic (63%) than in the drug treatment program (85%, P 0.01). The proportion of patients who achieved 3 or more consecutive weeks of abstinence from opioids, as determined by thrice weekly urine toxicology testing, was also higher in the primary care setting (44%, 10 of 23) than in the drug treatment setting (13%, 3 of 23; RR 3.33, 95% CI 1.05 to 10.56, P 0.02). While the rate of opioid use decreased over Figure 1. Proportions of patients remaining in treatment protocol by study group. 102 August 1998 THE AMERICAN JOURNAL OF MEDICINE Volume 105

4 Figure 2. Weekly opioid toxicology screens of study patients. time in the primary care clinic (P 0.001), it remained stable in the drug treatment program setting. Opioid withdrawal symptoms decreased over time in both settings (P 0.001); there was no significant effect of setting on withdrawal symptoms (P 0.6) or interaction between setting and time (P 0.77). Although no patients reported cocaine use in the month prior to admission to the study and all had negative urine toxicology at that time, a substantial proportion of urine toxicology screens during the study were positive for cocaine: 30.5% in the primary care and 38.5% in the drug treatment groups. No substantial adverse effects were noted in either group. Patients in the primary care treatment group were offered the option of continuing in that setting beyond the completion of the trial for an additional 10 weeks. All 18 patients who completed the 12-week trial accepted this option, 13 of whom remained in treatment for the full 22 weeks. Among these 18 patients, 38% of urine toxicology samples were positive for opioids during the continuation study. DISCUSSION Our results support the effectiveness of the primary care setting as an alternative to a drug treatment setting for providing thrice weekly opioid maintenance therapy to heroin-dependent patients off-the-street. Although a previous study evaluated methadone maintenance in a general medicine practice, patients in that study were required to have been on methadone maintenance at least 5 years with a high level of compliance, including clean urine toxicologies (9,37). This study also suggests that less than daily buprenorphine maintenance is effective. Most previous clinical trials have focused on daily buprenorphine, typically in doses of 8 to 16 mg sublingually (22 24). Amass et al (23) demonstrated that buprenorphine can be safely administered every 48 hours by doubling the maintenance dose. Similarly, Johnson et al (22) found that alternate-day dosing may also be effective in many patients. Because buprenorphine s effects may last up to 72 hours, we were able to design a regimen that would be practical for primary care providers routine schedules (Monday, Wednesday, Friday dosing) (24,25). Primary care-based opioid maintenance treatment may improve access to treatment. Properly trained clinicians could offer this treatment to their patients. This approach also offers the possibility of opioid maintenance treatment in communities, such as smaller towns, where methadone maintenance programs are not available. The treatment protocol used in this study was designed to be applicable to most primary care settings. We employed brief counseling techniques routinely used by primary care physicians (6,38); intervention approaches with a content similar to ours are effective in decreasing alcohol use in these settings (39 42). The visit sessions in this study were similar in time and effort to usual follow-up visits, and used an advice and clinical management approach that is well within the capabilities of most primary care providers (6,28,39,41). However, the number of visits was greater than that seen in typical brief intervention studies (41), and one might argue that weekly visits may be impractical. Other primary care patients may require a similar level of care, including selected patients with chronic conditions such as depression (43), diabetes (44), and HIV disease (45). It is possible that the frequency of visits may be diminished as patients remain in therapy for longer periods of time (25). Our initial aim in this study was to demonstrate that primary care-based buprenorphine maintenance would be at least as effective in promoting retention in treatment and decreasing opioid use as a drug treatment program approach. We were surprised to find that patients assigned to the primary care-based treatment actually did better than those assigned to a traditional methadone maintenance setting. Patients assigned to the primary care setting may have received more intensive individual attention than those treated in the methadone maintenance clinic, given the weekly visits with the primary care providers (46). The retention rates and urine toxicology results in our primary care patients were similar to those reported in other studies of patients given buprenorphine maintenance (14,17,47) as well as traditional methadone maintenance (18,47). Urine toxicology results may continue to improve after the 12-week duration of the randomized trial, as was seen in our continuation study (25,38). Finally, the willingness of our primary care patients to remain in this setting (as opposed to being transferred to opioid maintenance in a traditional setting) and their continued retention for 22 weeks suggests that patients are satisfied with this approach. Primary care-based drug treatment may have other unique advantages. Traditional primary care settings, more so than the primary care setting used in this study, may avoid some of the negative aspects of opioid main- August 1998 THE AMERICAN JOURNAL OF MEDICINE Volume

5 tenance programs, including the interactions with patients who continue to use illicit drugs and the stigma associated with drug treatment settings. Primary care settings also allow patients to receive drug treatment services and primary medical care under one roof (9,10,25,27), thus motivating patients to adhere better to both sets of services. For example, retention of our patients in primary care-based buprenorphine maintenance may have been enhanced if patients feared that they might lose medical care if they had left drug treatment at the same site. This study had several important limitations. Our clinic has special characteristics that distinguish it from other primary care settings (27). Although the primary care providers are not substance abuse treatment specialists, the clinic is closely affiliated with several drug treatment programs. Thus, providers have additional skills and resources for managing this population, such as enhanced provider-patient communication skills and easy access to substance abuse and mental health specialists. These resources may have enhanced the effectiveness of primary care based buprenorphine maintenance in our study. Although buprenorphine is as effective as methadone and has many advantages over methadone for agonist treatment of opioid dependence, it is not currently available for routine use in the United States, either in traditional methadone maintenance settings or in primary care settings. The doses of buprenorphine that we used were based on our previous experience with daily buprenorphine use (18), published literature concerning the potential use of buprenorphine as less than daily medication (23), and our pilot study of thrice weekly buprenorphine maintenance (25). There is controversy concerning the optimal daily dose of buprenorphine. In general, lower doses (4 to 8 mg/day) are thought to be less effective than standard doses of methadone (48) or higher doses (12 to 16 mg/day) of buprenorphine (49). The weekly total dose of buprenorphine (84 mg) received by our patients is equivalent to daily administration of 12 mg, a dose that is generally considered to be effective (48). Given that 16 mg/day may be more effective (47), patients may do better on a thrice weekly regimen with a greater total dose, such as 32 mg on Monday and Wednesday and 48 mg on Friday. Enrollment in this study was limited to patients without other drug dependence or severe psychiatric comorbidity. Patients with multiple substance dependencies or psychiatric disease require more services than are available in primary care settings. Despite our attempt to screen out patients who were dependent on cocaine, approximately one third of urine toxicology reports were positive for cocaine during the study. Clinicians interested in providing this treatment need to be aware of the high rate of concurrent cocaine use and to consider strategies to address this problem. More effective screening procedures may also be needed to minimize cocaine use in patients treated in primary care settings. Our results need to be interpreted in light of the small sample size and the relatively short follow-up period of 12 weeks. Although data from our pilot study demonstrated that this approach was successful for up to 6 months in most patients (25), additional studies with larger numbers of patients and longer term follow-up are needed before this approach can be generalized. Properly trained primary care physicians, including general internists, are well suited to care for patients with opioid dependence (13,50). Prior research suggests that the pharmacologic and nonpharmacologic treatments employed in this study can be provided by appropriately trained primary care providers (7,8,10,25,41,42,46,51). New approaches to managing substance use in primary care settings are central to ongoing efforts to link substance abusers to primary medical care (53). An additional barrier that would need to be addressed is the method of reimbursement for these services in primary care settings. Capitated, full-risk managed care plans may provide an impetus for this type of comprehensive treatment. In summary, this study demonstrates that thrice weekly primary care-based buprenorphine maintenance is an effective approach to promoting retention in treatment and deceased opioid use in managing heroin-dependent patients. Future research should more precisely define the dose schedules of buprenorphine, examine this approach in other primary care settings, and identify the subpopulation of drug users for whom primary carebased treatment may be suitable. REFERENCES 1. Primm BJ. Future outlook: treatment improvement. In: Lowinson JH, Ruiz P, Millman RB, et al, eds. Substance Abuse: A Comprehensive Textbook. 2nd ed. Baltimore: Williams and Wilkins; 1992: Dole VP. On federal regulation of methadone treatment. JAMA. 1995;274: O Connor PG, Selwyn PA, Schottenfeld RS. Medical progress: medical care for injection drug users with human immunodeficiency virus infection. NEJM. 1994;331: Stein MD. Medical complications of intravenous drug use. J Gen Intern Med. 1990;5: Cherubin CE, Sapira JD. The medical complications of drug addiction and the medical assessment of the intravenous drug user: 25 years later. Ann Intern Med. 1993;119: Fleming M, Rivo M, Kahn R. Tobacco, alcohol, and drug use in primary care: prevalence, treatment and costs. In: Sirca C, ed. Training About Alcohol and Substance Abuse for All Primary Care Physicians. New York: Josiah Macy Jr. Foundation; O Connor PG, Waugh ME, Carroll KM, et al. Primary care-based ambulatory opioid detoxification: the results of a clinical trial. J Gen Intern Med. 1995;10: O Connor PG, Carroll KM, Shi JM, et al. Three methods of opioid 104 August 1998 THE AMERICAN JOURNAL OF MEDICINE Volume 105

6 detoxification in a primary care setting. A randomized trial. Ann Intern Med. 1997;127: Novick DM, Pascarelli EF, Joseph H, et al. Methadone maintenance patients in general medical practice. A preliminary report. JAMA. 1988;259: Novick DM, Richman BL, Friedman JM, et al. The medical status of methadone maintenance patients in treatment for years. Drug Alcohol Depend. 1993;33: National Institute on Drug Abuse. Methadone in maintenance and detoxification: joint revisions of conditions for use. Fed Regist. 1989;54(40): Rettig RA, Yarmolinsky A. Federal Regulation of Methadone Treatment. Washington, DC: Institute of Medicine, National Academy Press; Cooper JR. Including narcotic addiction treatment in an officebased practice. JAMA. 1995;273: Strain EC, Stitzer ML, Liebson IA, Bigelow GE. Comparison of buprenorphine and methadone in the treatment of opioid dependence. Am J Psychiatry. 1994;151: Leander JD. Opioid agonist and antagonist behavioral effects of buprenorphine. Br J Pharmacol. 1983;78: Mello NK, Bree MP, Mendelson JH. Reinforcing properties of buprenorphine: a behavioral analysis. NIDA Res Monogr. 1981;34: Johnson RE, Jaffe JH, Fudala PJ. A controlled trial of buprenorphine treatment for opioid dependence. JAMA. 1992;267: Kosten TR, Schottenfeld R, Ziedonis D, Falcioni J. Buprenorphine versus methadone maintenance for opioid dependence. J Nerv Ment Dis. 1993;181: Kajiwara M, Aoki K, Ishii K, et al. Agonist and antagonist actions of buprenorphine on three types of opioid receptor in isolated preparations. Jpn J Pharmacol. 1986;40: Kosten TR, Kleber HD, Morgan C. Treatment of cocaine abuse with buprenorphine. Biol Psychiatry. 1989;26: Carroll ME, Carmona GN, May SA, et al. Buprenorphine s effects on self-administration of smoked cocaine base and orally delivered phencyclidine, ethanol and saccharin in rhesus monkeys. J Pharmacol Exp Ther. 1992;261: Johnson RE, Eissenberg T, Stitzer ML, et al. Buprenorphine treatment of opioid dependence: clinical trial of daily versus alternateday dosing. Drug Alcohol Depend. 1995;40: Amass L, Bickel WK, Higgins ST, Badger GJ. Alternate-day dosing during buprenorphine treatment of opioid dependence. Life Sci. 1994;54: Walsh SL, Preston KL, Stitzer ML, et al. Clinical pharmacology of buprenorphine: ceiling effects at high doses. Clin Pharmacol Ther. 1994;55: O Connor PG, Oliveto AH, Shi JM, et al. A pilot study of primarycare-based buprenorphine maintenance for heroin dependence. Am J Drug Alcohol Abuse. 1996;22: American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed, revised. Washington, DC: American Psychiatric Association; O Connor PG, Molde S, Henry S, et al. Human immunodeficiency virus infection in intravenous drug users: a model for primary care. Am J Med. 1992;93: Fawcett J, Epstein P, Fiester SJ, et al. Clinical management imipramine/placebo administration manual. NIMH Treatment of Depression Collaborative Research Program. Psychopharmacol Bull. 1978;23: Carroll KM, Rounsaville BJ, Gordon LT, et al. Psychotherapy and pharmacotherapy for ambulatory cocaine abusers. Arch Gen Psychiatry. 1994;51: Monti PM, Abrams DB, Kadden RM, Cooney NL. Treating Alcohol Dependence. New York: Guilford Press; Avants SK. Coping Skills Training for Methadone-maintained Patients (unpublished manual) Poklis A. Evaluation of TDx cocaine metabolite assay. J Anal Toxicol. 1987;11: Kosten TR, Rounsaville BJ, Kleber HD. Comparison of clinician ratings to self reports of withdrawal during clonidine detoxification of opiate addicts. Am J Drug Alcohol Abuse. 1985;11: Dawson-Sanders B, Trapp RG. Basic and Clinical Biostatistics. Norwalk, Conn: Appleton & Lang; Selvin S. Statistical Analysis of Epidemiologic Data. New York: Oxford University Press; Gibbons RD, Hedeker D, Elkin I, et al. Some conceptual and statistical issues in analysis of longitudinal psychiatric data. Application to the NIMH treatment of Depression Collaborative Research Program dataset. Arch Gen Psychiatry. 1993;50: Novick DM, Joseph H, Salsitz EA, et al. Outcomes of treatment of socially rehabilitated methadone maintenance patients in physicians offices (medical maintenance): follow-up at three and a half to nine and a fourth years. J Gen Intern Med. 1994;9: McCrady BS, Langenbucher JW. Alcohol treatment and health care system reform. Arch Gen Psychiatry. 1996;53: Babor TF, Grant M, Acuda W, et al. A randomized clinical trial of brief interventions in primary care: summary of a WHO project. Addiction. 1994;89: Bien TH, Miller WR, Tonigan JS. Brief interventions for alcohol problems: a review. Addiction. 1993;88: Wilk AI, Jensen NM, Havighurst TC. Meta-analysis of randomized control trials addressing brief interventions in heavy alcohol drinkers. J Gen Intern Med. 1997;12: Fleming MF, Barry KL, Manwell LB, et al. Brief physician advice for problem alcohol drinkers. A randomized controlled trial in community-based primary care practices. JAMA. 1997;277: Callahan CM, Hui SL, Nienaber NA, et al. Longitudinal study of depression and health services use among elderly primary care patients. J Am Geriatr Soc. 1994;42: Fasching P, Derfler K, Maca T, et al. Feasibility and efficacy of intensive insulin therapy in type 1 diabetes mellitus in primary care. Diabet Med. 1994;11: Selwyn PA, Budner NS, Wasserman WC, Arno PS. Utilization of on-site primary care services by HIV-seropositive and seronegative drug users in a methadone maintenance program. Pub Health Rep. 1993;108: McLellan AT, Arndt IO, Metzger DS, et al. The effects of psychosocial services in substance abuse treatment. JAMA. 1993;269: Schottenfeld RS, Pakes JR, Olivetto A, et al. Buprenorphine vs methadone maintenance treatment for concurrent opioid dependence and cocaine abuse. Arch Gen Psychiatry. 1997;54: Ling W, Wesson DR, Charuvastra C, Klett CJ. A controlled trial comparing buprenorphine and methadone maintenance in opioid dependence. Arch Gen Psychiatry. 1996;53: Schottenfeld RS, Pakes J, Ziedonis D, Kosten TR. Buprenorphine: dose-related effects on cocaine and opioid use in cocaine-abusing opioid-dependent humans. Biol Psychiatry. 1993;34(1 2): Lewis DC. The role of internal medicine in addiction medicine. J Addict Dis. 1996;15: Babor TF, Ritson EB, Hodgson RJ. Alcohol-related problems in the primary health care setting: a review of early intervention strategies. Br J Addict. 1986;81: Samet JH, Saitz R, Larson MJ. A case for enhanced linkage of substance abusers to primary medical care. Substance Abuse. 1996;17: August 1998 THE AMERICAN JOURNAL OF MEDICINE Volume