The report by Roll and associates that appears in this issue of the Journal describes results from a controlled trial supporting the effectiveness of contingency management for outpatient treatment of methamphetamine use disorders. This is certainly welcome news in light of the alarming increases in the prevalence of methamphetamine use disorders, the substantial morbidity, mortality, and social disruption associated with them, and the obvious need for effective treatments for them (1).

The results that Roll and colleagues report are based on a subset of data collected in a multisite trial supporting the effectiveness of contingency management for outpatient treatment of psychomotor stimulant use disorders in eight community outpatient drug abuse treatment clinics (2). The majority of the 415 patients in the parent trial met criteria for cocaine use disorders, but 113 of them (27%) from four of the eight participating clinics met diagnostic criteria for methamphetamine abuse or dependence, and these patients formed the study group on which Roll et al. report.

Contingency management is a treatment approach wherein patients receive predetermined consequences (e.g., monetary-based incentives) contingent upon achieving a therapeutic goal (e.g., abstinence from recent drug use). Contingency management is efficacious for the treatment of a wide range of different types of substance use disorders (3), and this report by Roll et al. adds methamphetamine use disorders to that list.

The efficacy of contingency management for treatment of cocaine use disorders has been well established since the mid-1990s, on the basis of results from highly controlled efficacy trials that were often, although not always, conducted in research clinics (3). The larger multisite trial from which the data of Roll et al. were obtained was conducted to examine whether similarly positive treatment outcomes would be obtained when the treatment was delivered in community drug abuse treatment clinics, where the patient population can be expected to be more heterogeneous than is typical of efficacy trials and the setting less conducive to rigorous experimental control (2). The patients were randomly assigned to receive 12 weeks of usual treatment in the participating clinics with or without abstinence-contingent reinforcement. Usual care typically was a form of cognitive behavior therapy. There is a high demand for treatment of methamphetamine use disorders in many Western and Midwestern states, and a subset of patients seeking treatment for those problems were included in the trial as well. In the parent trial, contingency management significantly increased the mean duration of continuous abstinence from stimulant and alcohol use achieved during the treatment period. We now learn that the same is true for the subset of patients with methamphetamine use disorders. No significant treatment differences in abstinence were noted at a 6-month posttreatment follow-up assessment, perhaps because of difficulties in obtaining adequate follow-up compliance. However, abstinence achieved during treatment was a positive predictor of posttreatment abstinence.

Methodologically, this strategy of separately studying a subset of patients who participated in the parent trial appears to have been well executed in terms of having adequate statistical power, comparable numbers of subjects across the two treatment conditions, and no significant differences between treatment conditions with regard to baseline sociodemographic or drug use characteristics. One potential criticism is that this analysis on patients with methamphetamine use disorders should have been handled as a subsection of the original report rather than as a separate report. However, an important advantage of reporting the analysis separately is that the focus of the report can be on methamphetamine use disorders exclusively, starting with the title, which should increase the likelihood that the information will come to the attention of policy makers and other professionals charged with addressing the alarming public health problems that methamphetamine use disorders now represent in the United States and elsewhere. Getting such information into the hands of those who have the responsibility and authority to apply it is no minor matter. Having two published reports with overlapping data sets could present problems in future reviews and meta-analyses, but Roll and colleagues were sufficiently diligent in acknowledging the overlap and citing the original report that this should be readily manageable.

Roll et al. report a positive association between during-treatment and longer-term abstinence for methamphetamine use disorders, and this relationship is known to be important in the outpatient treatment of cocaine and a number of other types of substance use disorders. With cocaine abusers, for example, the probability of posttreatment abstinence increases as an orderly function of increasing duration of continuous abstinence achieved during the treatment period (4). Another example is the strikingly robust positive association between initial and longer-term abstinence that has been well documented among cigarette smokers (5). The evidence regarding the importance of this relationship is not limited to correlational data. For example, my colleagues and I recently conducted a randomized clinical trial (6) in which the amount of continuous abstinence achieved during treatment was directly manipulated by assigning cocaine-dependent outpatients to one of two contingency management conditions involving abstinence-contingent vouchers of relatively low or high monetary value. As hypothesized, high-value vouchers not only increased the duration of continuous cocaine abstinence achieved during the 12 weeks that the incentives were available but also increased abstinence during a subsequent 21-month follow-up period. Documenting that this relationship between initial and longer-term abstinence extends to methamphetamine use disorders advances our understanding of how to effectively treat this population, and demonstrating the ability of contingency management to directly increase abstinence provides a practical tool by which the amount of abstinence achieved can be directly impacted.

Successfully completing multisite trials is a daunting task, and the authors are to be commended for doing so in this instance. Doing so in a manner that simultaneously addressed the effectiveness of contingency management for psychomotor stimulant use disorders generally and methamphetamine use disorders specifically is a particularly impressive accomplishment and a laudable contribution to the development of empirically based treatments for substance use disorders. Even more impressive still is the fact that this same general group of investigators simultaneously and successfully completed a parallel multisite trial that also supported the efficacy of contingency management for treatment of psychomotor stimulant use disorders among patients enrolled in methadone treatment (7). Whether a subset of them also used methamphetamine and will provide the grist for another report like the one in this issue by Roll et al. is not clear. What is abundantly clear from this impressive body of research, however, is the effectiveness of contingency management for treating psychomotor stimulant use disorders. Its efficacy for cocaine use disorders was first demonstrated in two controlled trials that my colleagues and I reported in the Journal more than a decade ago (8, 9). It seems fitting that the first trial extending abstinence-contingent incentives to methamphetamine use disorders would appear in the Journal as well. The consistently positive results obtained with contingency management across many rigorously controlled efficacy trials (3) and now this more recent set of multisite effectiveness trials provide overwhelming evidence that contingency management has a substantial contribution to make in treating psychomotor stimulant substance use disorders. How it eventually gets integrated into everyday community treatment services for these disorders is still unclear but will be interesting to watch unfold.

Address correspondence and reprint requests to Dr. Higgins, Human Behavioral Pharmacology Laboratory, Ira Allen School, 38 Fletcher Place, Burlington, VT 05401; Stephen.Higgins@uvm.edu (e-mail). Supported in part by research grants DA-09378, DA-14028, and DA-08076 from the National Institute on Drug Abuse. Dr. Higgins reports no competing interests.