Effects of reducing contingency management values on heroin and cocaine use for buprenorphine- and desipramine-treated patients

Transcription

1 Blackwell Science, LtdOxford, UKADDAddiction Society for the Study of Addiction to Alcohol and Other Drugs Original ArticleContingency management and desipramine for cocaine Thomas Kosten et al. RESEARCH REPORT Effects of reducing contingency management values on heroin and cocaine use for buprenorphine- and desipramine-treated patients Thomas Kosten, James Poling & Alison Oliveto Department of Psychiatry, Yale University School of Medicine, VA New England MIRECC, West Haven, CT, USA Correspondence to: Thomas Kosten VA Connecticut Healthcare System Psychiatry 151D 950 Campbell Avenue, Bldg 35 West Haven, CT USA Tel: ext Fax: Submitted 28 May 2002; initial review completed 26 August 2002; final version accepted 3 December 2002 ABSTRACT Aims During 3 months where contingency management (CM) had an escalating value for each consecutive drug-free urine (escalating CM), cocaine- and heroin-abusing patients significantly increased drug-free urines. The escalating CM was eliminated during months 4 6 to assess any reduction in drug-free urines. Design Patients who completed a 3-month, randomized, double-blind, trial evaluating CM versus non-cm and desipramine (DMI) versus placebo, had an escalating CM eliminated during months 4 6. The CM and non-cm groups were compared using thrice-weekly urine samples. Setting Out-patient buprenorphine maintenance for 6 months. Participants All 75 of the 160 original study patients who completed month 3 of the clinical trial. Intervention The escalating CM was eliminated for all 3 months and during months 5 and 6 the response requirement was also increased to two and then three consecutive drug-free urines in order to obtain a voucher. Measurements Urine toxicology for opiates and cocaine. Findings After eliminating the escalating CM, the CM group showed a decline in combined opioid- and cocaine-free urines. This decline within the CM group was greater in those treated with DMI than placebo. Conclusions Buprenorphine with DMI maintained drug abstinence after eliminating the escalating CM, but not after increasing the response requirement, suggesting the need for more intensive psychosocial interventions during CM. KEYWORDS Buprenorphine, clinical trial, cocaine, contingency management, desipramine, opioid dependence. INTRODUCTION Combined opioid and cocaine dependencies are common problems with rates of 40 60% among those entering methadone treatment, but methadone or buprenorphine maintenance do not target cocaine abuse (Kosten et al. 1987; Ball et al. 1988, 1989; Zweben & Payte 1990; Ball & Ross 1991; Oliveto et al. 1999; Sorensen & Copeland 2000). Because cocaine is not addressed specifically by these pharmacotherapies, these dually dependent patients are more likely than those dependent on opioids alone to continue to use cocaine and to drop out of methadone maintenance within a month (DeMaria et al. 2000). We have conducted a series of studies examining pharmacotherapies for this dual dependence and found recently that desipramine could be combined with buprenorphine to reduce combined opioid and cocaine use significantly among these dually dependent patients (Oliveto et al. 1999). Furthermore, adding contingency management (CM) to the desipramine (DMI) substantially enhanced the efficacy of either DMI or CM alone (Kosten et al. 2003). The efficacy of this combination therapy might require sustained high levels of CM voucher values, how-

2 666 Thomas Kosten et al. ever, because studies have indicated that making CM potent in an opioid maintenance setting may require substantial elevations in typical voucher values (Silverman et al. 1996, 1998, 1999, 2001; Chutuape et al. 1999; Griffith et al. 2000; Jones et al. 2001). Other studies have also suggested that increasing response requirements for vouchers in CM can lead to relapse to drug use, although notable studies have indicated that CM related gains in drug abstinence can be sustained beyond the CM period (Silverman et al. 1998; Higgins, Badger & Budney 2000a, Higgins et al. 2000b; Petry & Martin 2002). In particular, Higgins had shown that a shift from escalating vouchers for consecutive cocaine-free urines to reinforcing any cocaine-free urines without the escalation for consecutive urines could sustain the effects of CM on cocaine abstinence (Silverman et al. 1998; Higgins et al. 2000a, 2000b; Petry & Martin 2002). To examine this question of sustained efficacy of CM we used Higgins s approach by eliminating the voucher escalation, and after the first month also increased the response requirements for CM vouchers. This study was a continuation study of our previous 12-week clinical trial that showed a substantial increase in drug-free urines with DMI combined with CM (Kosten et al. 2003). During this 3-month continuation phase, we introduced monthly increases in the response requirements for vouchers while maintaining patients in four groups: CM + DMI (DC), CM + placebo (Pla) (PC), non-cm + DMI (DNC) and non-cm + Pla (PNC). We had two hypotheses. First, as the CM voucher requirements increased, the patients in CM would show either poorer treatment retention than the non-cm groups or decreases in drug-free urines compared to the non-cm groups. As a correlate of the Higgins study, we expected little change in drug-free urines for the DC group during the initial month when only the escalating voucher values for consecutive drug-free urines were eliminated (Higgins et al. 2000a, 2000b; Silverman et al. 1998; Petry & Martin 2002). Secondly, because the preceding clinical trial had found that the DC patients had substantially more drug-free urines than the other three groups, the increases in CM voucher requirements would decrease drug-free urines more in this group than in the PC group. Alternatively, the relative decline in CM efficacy might be evident only among those patients treated with DMI, as CM alone produced a much more modest increase in drug-free urines. METHODS Subjects All 75 of 160 cocaine-abusing opioid addicts who completed a 12-week randomized clinical trial of buprenorphine combined with desipramine and/or CM were entered into a 12-week CM reduction phase (weeks 13 24) of treatment (Kosten et al. 2003). The 160 subjects had been recruited from the general Greater New Haven area and gave written informed consent to participate in the full 24-week protocol including this CM reduction phase, as approved by the Yale Human Investigations Committee and the VA Connecticut Human Studies Committee. Although the 75 subjects differed from the 85 who did not complete the initial trial in both treatment retention (22 versus 6 weeks; t = 28; P < 0.001) and percentage of cocaine plus opiate-free urines during weeks 1 12 (41% versus 13%; t = 5.8; P < 0.001), the two groups did not differ in baseline characteristics before treatment. All participants fulfilled DSM-IV criteria for opioid and cocaine dependence, as determined from the SCID, and had positive urine toxicology screens for both drugs. Treatment retention and voucher reinforcement The 75 subjects among the four treatment conditions showed no significant differences on demographics (mean age = 38 years, SD = 7; 52 males; 32 African- Americans). At the time of admission to the original 12- week study, heroin use was daily for almost all subjects (28 days, SD = 5) and cocaine use every other day (14 days, SD = 10). Alcohol use was about 3 days per month (SD = 6) and sedative use was less than 1 day per month, reflecting our exclusion criterion that urine toxicologies could not contain sedatives at admission. Research design and procedures These 12 weeks of eliminating the escalating CM vouchers for consecutive drug-free urines followed a randomized double-blind clinical trial assigning 160 patients who had been stabilized on buprenorphine (range 8 20 mg) to one of four treatment conditions: desipramine (150 mg) plus contingencies (DC), desipramine without contingencies (DNC), placebo plus contingencies (PC), placebo without contingencies (PNC) (n = 40 per cell). During this CM reduction phase, desipramine dose was held constant at 150 mg daily and buprenorphine doses were maintained at 8, 16 or 20 mg daily with a median dose of 16 mg and no dosage difference across the four treatment groups. For those assigned to the CM procedure groups, each urine free from both cocaine and opioids submitted during weeks 1 12 resulted in a voucher worth a certain monetary value, which increased with every consecutive drug-free urine. This escalating CM was eliminated at week 13 and during weeks each drug-free urine resulted in a voucher worth a fixed value of $3.00, and using a fixed ratio schedule of reinforcement the number

3 Contingency management and desipramine for cocaine 667 of consecutive urines needed for a voucher was increased to two for $6 during weeks and to three for $9 during weeks Vouchers could be exchanged for mutually agreed-upon goods and services at any time during the study. Subjects in the CM group who remained abstinent were able to earn goods and services worth a maximum of $738 during weeks 1 12, and $108 during weeks Subjects assigned to the non-cm procedure were yoked to the average value of the vouchers for the two CM groups in the previous week during weeks These vouchers were not contingent upon illicit drug abstinence and every non-cm subject received a minimum of $3 per week. During the study, subjects also participated in weekly group coping skills/relapse prevention therapy and weekly individual therapy sessions. Baseline and outcome assessments Demographics, medical and family histories were collected at intake on self-report forms modified from the NIMH Collaborative Study on Treatment of Depression. At intake, the Structured Clinical Interview for DSM-IV (SCID) interview also was completed for DSM-IV psychiatric diagnoses, including depression and substance use disorders (First et al. 1995). Substance abuse-related problems and psychosocial functioning were assessed at intake using the Addiction Severity Index (ASI), a 140- item structured clinical interview using both subjective and objective information to make severity ratings on 10- point scales in six areas: substance use, medical status, legal status, psychiatric symptoms, occupational functioning and family/social functioning (McLellan et al. 1992). Primary assessments of treatment outcome included treatment retention and illicit drug use (as measured by urine toxicologies). Urine samples were analyzed for the presence of opioids and cocaine metabolites upon submission of urine sample on a thrice-weekly basis. This frequency of urine monitoring allowed us to detect most drug use, including any regular cocaine use. A urine sample was rated positive if the quantity of drug or metabolite was >300 ng/ml for benzoylecognine and >200 ng/ml for opiates. Data analyses The four treatment groups were compared for baseline differences using c 2 for categorical characteristics and ANOVA for continuous variables such as age. Treatment retention was then compared using survival curves with Cox proportional hazards model and c 2 analyses to compare the number of completers to 24 weeks. The weekly voucher reinforcement levels were analyzed using repeated-measures ANOVA. Urine toxicologies were compared using the proportion of the three urine samples that were drug-free each week ranging between 0 and 1, as described previously (Oliveto et al. 1999). Opioid and cocaine abstinence was analyzed separately as well as abstinence from both drugs together. We examined the raw data in two ways. First, we present only those subjects remaining in the trial at each time point. Secondly, those who left before 24 weeks were considered to be using either opioids or cocaine again after they left. This attribution seemed reasonable, because patients were opioid-dependent when they left and were not transferring directly to other treatment programs. Because of the missing data, for statistical comparisons we conducted a 2 (DMI) 2 (contingencies) 24 (time) hierarchical linear modeling (HLM) analysis for these comparisons between CM and non-cm groups (Bryk & Raudenbush 1987; Gibbons et al. 1993). Because of the change in contingencies during this CM reduction phase, we compared the standard HLM to a piecewise regression analysis using HLM. In this piecewise approach, we determined whether the slope of the regression line changed at weeks compared to weeks Because the medications did not change during these 24 weeks, we focused on the contingencies and compared the fit of the standard linear and piecewise models using the difference in the likelihood ratios for the two models. This difference can be multiplied by two and treated as a c 2 value with two degrees of freedom in order to determine whether the piecewise model provides a better statistical fit to the data. Because we had found previously that during the 12-week clinical trial patients who had both DMI and contingencies had significantly better outcomes than those who received either intervention alone, we compared the groups receiving CM to those not receiving CM within the DMI- and placebo-treated groups separately using the two types of HLM models. RESULTS From the 75 patients, 63 patients completed to week 16 of 24, 54 completed to week 20 and 42 (56%) completed the full CM reduction trial. Treatment retention did not differ among the four treatment groups with the number of subjects at weeks 12, 16, 20 and 24 as follows: DC (17, 17, 12, 10), DNC (18, 14, ), PC (20, 18, 15, 10) and PNC (20, 14, 13, 11) (Wilcoxon 0.4; df = 3; NS). Reasons for premature termination from the study included the following: one subject left for medical issues, 11 left at their request, seven missed medications on three successive days and 14 failed to attend counseling and provide supervised urines leading to administrative termination. The mean proportion of counseling sessions attended was 69% for weekly groups and 83% for weekly individual

4 668 Thomas Kosten et al. sessions, with no difference across the four treatment groups. The magnitudes of voucher reinforcements attained by the four treatment groups were significantly different during the first 12 weeks and weeks (CM reduction phase). During the first 12 weeks the DC group had the greatest proportion of drug-free urines and correspondingly earned the greatest amount in vouchers $295 (SD $227) compared to $117 (SD $186) for the PC group and $125 (SD $75) for the DNC and PNC groups (F = 5.5; df = 3,74; P < 0.02). The total amount earned during weeks showed the largest drop for the DC group from $295 to $58 (SD $39), while the other three groups dropped much less: PC from $117 to $36 (SD $46), the DNC from $125 to $83 (SD $56) and the PNC from $125 to $67 (SD $56) (interaction of time with medication group: F = 7.0; df = 3, 71; P < ). Abstinence from both opioids and cocaine Because the contingencies targeted both opioid and cocaine abstinence, we examined first whether our interventions increased the rates of urines free of both illicit drugs. Figures 1 and 2 present the drug-free urines for the four groups. All treatment groups showed a substantial improvement from the first month in treatment where the rates of drug-free urines were 29% for DC and 17% for the other three treatment groups. During month 3 (weeks 9 12) at the peak CM values, the DC group clearly has the best outcome with 71% of urines drug-free compared to 25% to 29% for the other three groups (F = 17; df = 3300; P < ). As the CM values declined (months 4 6), the DC group also showed a decline in its rate of drug-free urines. Because of dropout during the trial, Fig. 2 presents the rates for months 4 6 when attributing illicit drug use to those who dropped out of treatment. In comparing Figs 1 and 2, the most apparent difference is that the DC group has a clearer decline in drug-free urines from 71 to 66% during months 3 and 4 47% at month 5 and 38% at month 6. Thus, both treatment retention and drug-free urines were maintained for the DC group during month 4 after the escalating CM was eliminated. In Fig. 2, the other three groups do not show the increases in drug-free urines that appear to occur in Fig. 1, and never rise above about 35%. Thus, the apparent increase in drug-free urines in several groups may be a result of early dropout for those who relapsed. Statistically, a simple linear time contingency interaction was significant (Z = 3.7; P < ) with an overall likelihood ratio of 594. However, the piecewise model for Rate both cocaine- and opiate-free urines Month Figure 1 Monthly proportion of opioid- and cocaine-free urines during the six successive monthlong time periods corresponding to the first 3 months of escalating CM voucher values followed by 3 months of monthly declines in CM voucher values. The comparison is among the four treatment groups: desipramine hydrochloride (0 or 150 mg/day) plus contingency management (DC) or a non-contingent voucher control (DNC), placebo plus contingency management (PC) or a non-contingent voucher control (PNC), in 75 opioid-dependent cocaine abusers maintained on buprenorphine 16 mg daily (median dose)., DC;, DNC;, PC;, PNC Rate opioid- and cocaine-free urines Month Figure 2 Monthly proportion of opioid- and cocainefree urines during the six successive month-long time periods corresponding to the first 3 months of escalating CM voucher values followed by 3 months of monthly declines in CM voucher values. In contrast to Fig. 1, drug-positive urines were attributed to subjects leaving before the end of the trial (week 24). The comparison is among the four treatment groups: desipramine hydrochloride (0 or 150 mg/day) plus contingency management (DC) or a non-contingent voucher control (DNC), placebo plus contingency management (PC) or a non-contingent voucher control (PNC), in 75 opioid-dependent cocaine abusers maintained on buprenorphine 16 mg daily (median dose)., DC;, DNC;, PC;, PNC

5 Contingency management and desipramine for cocaine 669 these same data showed a positive slope in the rate of urines becoming drug-free during weeks 1 12, but a negative slope for the CM groups during weeks (Z = 5.4; P < ). Moreover, the likelihood ratio for this piecewise model was 579, which yielded a c 2 of 30 (df = 2; P < ) in comparison to the simple linear model. Thus, the piecewise model fitted the data significantly better than a simple linear model. To examine whether the combined DC treatment group was the major contributor to the reduced efficacy of CM in the piecewise HLM model, we conducted HLM model fitting for the DMI and placebo groups separately. For the DMI group alone, the simple linear HLM model comparing the CM to non-cm groups gave a log likelihood of 299 compared to 271 for the piecewise model (c 2 = 58; df = 2; P < ). The rate of urines becoming drug-free changed significantly between weeks 1 12 and (Z = 5.4; P < ) going from positive to negative, as both parts of Figs 1 and 2 illustrate. For the placebo groups, the simple linear and piecewise models showed no differences (log likelihood of 247 for both models) and the time contingency interaction was not significant (Z = 1.8). Abstinence from opioids or cocaine alone The rates of opioid-free or cocaine-free urines showed a similar effect of reducing CM values. The DC group s opioid-free urines declined from 82% to 42% after the CM values were reduced and their cocaine-free urines dropped from 72% to 46%, while the other three groups showed no change from months 4 6. Statistically for opioids alone, the simple linear time contingency interaction was significant (Z = 2.97; P < 0.003) with a log likelihood ratio of 802, but the piecewise model had an improved fit to the data with a c 2 of 34 (df = 2; P < 0.001). Statistically for cocaine alone, the simple linear time contingency interaction was significant (Z = 2.5; P < 0.003) with a log likelihood ratio of 627, but the piecewise model had an improved fit to the data with a c 2 of 10 ( df = 2; P < 0.01). When we considered the DMI group alone, the piecewise model was considerably better at describing the comparison of CM to non-cm for opioids alone (320 versus 351 likelihood ratios) (c 2 =62; df = 2; P < ) and for cocaine alone (244 versus 259 likelihood ratios) (c 2 =30; df = 2; P < 0.001). However, for the placebo groups the piecewise model showed no improvement over a simple linear HLM model with either opioids or cocaine alone. DISCUSSION During the 3-month placebo-controlled clinical trial preceding this continuation study, the combination of CM and DMI (DC) had substantially increased opioid- and cocaine-free urines (Kosten et al. 2003). However, this continuation study from months 4 6 found that the combined intervention showed a significant reduction in efficacy as the response requirement for the CM voucher increased from one to three consecutive drug-free urines in months 5 and 6. When the dropouts were considered to be non-abstinent after leaving treatment, then the drug-free urine rate for the DC group declined to the rate for the other three treatment groups during months 5 and 6. As no one dropped out from the DC group before month 5 and the decline in drug-free urines was minimal between weeks 13 and 16, the initial loss of the escalating CM (for consecutive drug-free urines) appears to be relatively unimportant for sustaining abstinence. However, the increase in response requirement for CM vouchers may have been critical in the reduction in drug-free urines. The apparent increase in drug-free urines for the non-cm as well as the PC groups shown in Figs 1 and 2 probably reflected some selection factors, because treatment retention dropped by almost 50% during this 3- month increase in response requirements for vouchers. While the PC group showed no significant reduction in drug-free urines when the CM values were reduced, this probably reflects little reduction in their drug use during the escalating CM voucher phase of this trial. Finally, because the contingencies targeted abstinence from both opioids and cocaine, this outcome was the most relevant, but we also found that either opioid or cocaine use alone followed this pattern of reduced opioid-free or cocainefree urines for the DC group. Other trials of CM have indicated that CM s efficacy can be sustained for about a month after discontinuation of any vouchers, and our DC patients continued to sustain high rates of drug-free urines for the first month after a marked reduction in their CM earnings from $54 for week 12 to $6 per week for weeks (Silverman et al. 1998; Higgins et al. 2000a, 2000b; Petry & Martin 2002). However, during months 5 and 6 the DC group had substantially fewer drug-free urines as the response requirement rose for the CM vouchers. This inability to sustain drug-free urines beyond about 2 months after stopping CM has been found by others. Silverman et al. (1998) described a decline from 47 to 59% cocaine abstinence to 29% at a 2-month follow-up, which was not significantly greater than the 16% rate in the control group. Petry reported that the day to first use of opioids or cocaine was substantially longer for the CM than the non-cm groups on methadone maintenance (50 versus 6 days), but by about 2 months the CM and non-cm groups did not differ, with about 70% versus 80% having relapsed to drug use. Higgins et al. had shown that eliminating the escalating vouchers for consecutive cocainefree urines and simply reinforcing any cocaine-free urines

6 670 Thomas Kosten et al. could sustain the effects of CM on cocaine abstinence for well beyond 2 months (Higgins et al. 2000a, 2000b; Silverman et al. 1998; Petry & Martin 2002). This finding suggested that our increase in the response requirement during months 5 and 6 was critical to the loss of a sustained response, but primary cocaine abusers as in the Higgins study may also have a better prognosis for sustained abstinence after CM than opioid-maintained cocaine abusers (Griffith et al. 2000). As our novel approach to increasing the response requirement of CM appears to have little advantage over discontinuing the contingencies, reducing the cost of CM by eliminating the escalating CM vouchers for consecutive drug-free urines merits further testing alone as a way to increase CM cost efficiency. Because the antidepressant did not appear to sustain the drug abstinence attained by the addition of CM, DMI alone holds little promise as a relapse prevention agent in this population. These antidepressants alone have also not shown efficacy for abstinence initiation (Gawin et al. 1989; Weddington et al. 1991; Arndt et al. 1992; Kosten et al. 1992; Galloway et al. 1994; Covi et al. 1995; Grabowski et al. 1995; Margolin et al. 1995; Nunes et al. 1995; Batki et al. 1996), although adding these antidepressants to CM appears to improve initiation of abstinence (Schmitz et al. 1998). Future studies will need to sustain the effects of CM in this population by focusing, perhaps, on both broader psychosocial changes and reduction in drug use such as the community reinforcement approach, rather than simply focus on urine toxicologies or psychosocial changes alone (Hunt & Azrin 1973; Azrin 1976; Iguchi et al. 1997). ACKNOWLEDGEMENTS This study was supported by the National Institute on Drug Abuse grants R01-DA05626 (TRK), K05-DA0454 (TRK), P50-DA and the Veterans Administration Mental Illness Research, Education and Clinical Center (MIRECC). REFERENCES Arndt, I. O., Dorozynsky, L., Woody, G. E., McLellan, A. T. & O Brien, C. P. (1992) Desipramine treatment of cocaine dependence in methadone-maintained patients. Archives of General Psychiatry, 49, Azrin, N. H. (1976) Improvements in the community-reinforcement approach to alcoholism. Behavior Research and Therapy, 14, Ball, J., Corty, E., Bond, H., Myers, C. & Tommasello, A. (1988) The reduction of intravenous heroin use, non-opiate abuse and crime during methadone maintenance treatment: further findings. NIDA Research Monograph, 81, Ball, J. C. & Ross, A. (1991) The Effectiveness of Methadone Maintenance Treatment. New York: Springer-Verlag. Ball, J. C., Ross, A. & Jaffe, J. H. (1989) Cocaine and heroin use by methadone maintenance patients. NIDA Research Monograph, 95, 328. Batki, S. L., Washburn, A. M., Delucchi, K. & Jones, R. T. (1996) A controlled trial of fluoxetine in crack cocaine dependence. Drug and Alcohol Dependence, 41, Bryk, A. S. & Raudenbush, S. W. (1987) Application of hierarchical linear models to assessing change. Psychological Bulletin, 101, Chutuape, M. A., Silverman, K. & Stitzer, M. (1999) Contingent reinforcement sustains post-detoxification abstinence from multiple drugs: a preliminary study with methadone patients. Drug and Alcohol Dependence, 54, Covi, L., Hess, J. M., Kreiter, N. A. & Haertzen, C. A. (1995) Effects of combined fluoxetine and counseling in the outpatient treatment of cocaine abusers. American Journal of Drug and Alcohol Abuse, 21, DeMaria, P. A. Jr, Sterling, R. & Weinstein, S. P. (2000) The effect of stimulant and sedative use on treatment outcome of patients admitted to methadone maintenance treatment. American Journal on Addictions, 9, First, M. B., Spitzer, R. L., Gibbon, M. & Williams, J. B. W. (1995) Structured Clinical Interview for DSM-IV. Patient Edition. Washington, DC: American Psychiatric Press, Inc. Galloway, G. P., Newmeyer, J., Knapp, T., Stalcup, S. A. & Smith, D. (1994) Imipramine for the treatment of cocaine and methamphetamine dependence. Journal of Addictive Diseases, 13, Gawin, F. H., Kleber, H. D., Byck, R., Rounsaville, B. J., Kosten, T. R., Jatlow, P. I. & Morgan, C. (1989) Desipramine facilitation of initial cocaine abstinence. Archives of General Psychiatry, 46, Gibbons, R. D., Hedeker, D., Elkin, I., Waternaux, C., Kraemer, H. C., Greenhouse, J. B., Shea, M. T., Imber, S. D., Sotsky, S. M. & Watkins, J. T. (1993) Some conceptual and statistical issues in analysis of longitudinal psychiatric data. Application to the NIMH treatment of Depression Collaborative Research Program dataset. Archives of General Psychiatry, 50, Grabowski, J., Rhoades, H., Elk, R., Schmitz, J., Davis, C., Creson, D. & Kirby, K. (1995) Fluoxetine is ineffective for treatment of cocaine dependence or concurrent opiate and cocaine dependence: two placebo-controlled double-blind trials. Journal of Clinical Psychopharmacology, 15, Griffith, J. D., Rowan-Szal, G. A., Roark, R. R. & Simpson, D. D. (2000) Contingency management in outpatient methadone treatment: a meta-analysis. Drug and Alcohol Dependence, 58, Higgins, S. T., Badger, G. J. & Budney, A. J. (2000a) Initial abstinence and success in achieving longer term cocaine abstinence. Experimental and Clinical Psychopharmacology, 8, Higgins, S. T., Wong, C. J., Badger, G. J., Ogden, D. E. & Dantona, R. L. (2000b) Contingent reinforcement increases cocaine abstinence during outpatient treatment and 1 year of followup. Journal of Consulting and Clinical Psychology, 68, Hunt, G. M. & Azrin, N. H. (1973) A community-reinforcement approach to alcoholism. Behavior Research and Therapy, 11, Iguchi, M. Y., Belding, M. A., Morral, A. R., Lamb, R. J. & Husband, S. D. (1997) Reinforcing operants other than absti-

7 Contingency management and desipramine for cocaine 671 nence in drug abuse treatment: an effective alternative for reducing drug use. Journal of Consulting and Clinical Psychology, 65, Jones, H. E., Haug, N., Silverman, K., Stitzer, M. & Svikis, D. (2001) The effectiveness of incentives in enhancing treatment attendance and drug abstinence in methadonemaintained pregnant women. Drug and Alcohol Dependence, 61, Kosten, T. R., Morgan, C. M., Falcione, J. & Schottenfeld, R. S. (1992) Pharmacotherapy for cocaine-abusing methadonemaintained patients using amantadine or desipramine. Archives of General Psychiatry, 49, Kosten, T. R., Oliveto, A., Feingold, A., Poling, J., Sevarino, K., McLance-Katz, E., Stine, S., Gonzalez, G. & Gonsai, K. (2003) Desipramine and contingency management for cocaine and opiate dependence in buprenorphine maintained patients. Drug and Alcohol Dependence, in press. Kosten, T. R., Rounsaville, B. J. & Kleber, H. D. (1987) A 2.5-year follow-up of cocaine use among treated opioid addicts. Have our treatments helped? Archives of General Psychiatry, 44, Margolin, A., Kosten, T. R., Avants, S. K., Wilkins, J., Ling, W., Beckson, M., Arndt, I. O., Cornish, J., Ascher, J. A. & Li, S. H. (1995) A multicenter trial of bupropion for cocaine dependence in methadone-maintained patients. Drug and Alcohol Dependence, 40, McLellan, A. T., Cacciola, J., Kushner, H., Peters, R., Smith, I. & Pettinati, H. (1992) The fifth edition of the Addiction Severity Index: cautions, additions, and normative data. Journal of Substance Abuse, 9, Nunes, E. V., McGrath, P. J., Quitkin, F. M., Ocepek-Welikson, K., Stewart, J. W., Koenig, T., Wager, S. & Klein, D. F. (1995) Imipramine treatment of cocaine abuse: possible boundaries of efficacy. Drug and Alcohol Dependence, 39, Oliveto, A. H., Feingold, A., Schottenfeld, R., Jatlow, P. & Kosten, T. R. (1999) Desipramine in opioid-dependent cocaine abusers maintained on buprenorphine vs methadone. Archives of General Psychiatry, 56, Petry, N. & Martin, B. (2002) Low-cost contingency management for treating cocaine- and opioid-abusing methadone patients. Journal of Consulting and Clinical Psychology, 70, Schmitz, J. M., Rhoades, H. M., Elk, R., Creson, D., Hussein, I. & Grabowski, J. (1998) Medication take-home doses and contingency management. Experimental and Clinical Psychopharmacology, 6, Silverman, K., Chutuape, M. A., Bigelow, G. E. & Stitzer, M. L. (1999) Voucher-based reinforcement of cocaine abstinence in treatment-resistant methadone patients: effects of reinforcement magnitude. Psychopharmacology, 146, Silverman, K., Higgins, S. T., Brooner, R. K., Montoya, I. D., Cone, E. J., Schuster, C. R. & Preston, K. L. (1996) Sustained cocaine abstinence in methadone maintenance patients through voucher-based reinforcement therapy. Archives of General Psychiatry, 53, Silverman, K., Svikis, D., Robles, E., Stitzer, M. L. & Bigelow, G. E. (2001) A reinforcement-based therapeutic workplace for the treatment of drug abuse: six-month abstinence outcomes. Experimental and Clinical Psychopharmacology, 9, Silverman, K., Wong, C. J., Umbricht-Schneiter, A., Montoya, I. D., Schuster, C. R. & Preston, K. L. (1998) Broad beneficial effects of cocaine abstinence reinforcement among methadone patients. Journal of Consulting and Clinical Psychology, 66, Sorensen, J. L. & Copeland, A. L. (2000) Drug abuse treatment as an HIV prevention strategy: a review. Drug and Alcohol Dependence, 59, Weddington, W. W. Jr, Brown, B. S., Haertzen, C. A., Hess, J. M., Mahaffey, J. R., Kolar, A. F. & Jaffe, J. H. (1991) Comparison of amantadine and desipramine combined with psychotherapy for treatment of cocaine dependence. American Journal of Drug and Alcohol Abuse, 17, Zweben, J. E. & Payte, J. T. (1990) Methadone maintenance in the treatment of opioid dependence. A current perspective. Western Journal of Medicine, 152,