Evidence-Based Treatments for Substance Use Disorders

This is an exciting time for the field of addiction. Many treatments for drug and alcohol addiction have demonstrated efficacy in controlled clinical trials and in effectiveness studies conducted in real-world community-based settings (1, 2). The strongest predictors of effectiveness include longer time in treatment (retention), severity of addiction and related problems, and the comprehensiveness of services received (3–5). Despite these advances, most individuals with severe addictions will experience a fluctuating course of abstinence and uncontrolled use (6). Investigations in the neurobiology of addictions indicate that there may be lasting neurobiological changes as a result of chronic drug use.

This clinical synthesis first defines addiction and outlines a general perspective on effective addiction treatment. Behavioral therapies, psychotherapies, and pharmacotherapies with demonstrated effectiveness are briefly described. Then the latest developments in combination therapies and the need for comprehensive clinical and psychosocial services are discussed. Finally, pressing questions, controversies, and recommendations for treatment practice and for further research are summarized.

Clinical context

Although substance use disorders are initiated by voluntary acts, the persistence of powerful involuntary responses to substance-related cues and the propensity to relapse after periods of abstinence are controlled by a complex mix of environmental precipitants, genetic liabilities, and permanent pathophysiological changes in brain circuitry (6–8). These features suggest that severe addiction is a chronic illness, like diabetes, accompanied by major disabilities and requiring continuing and possibly lifetime care, frequent monitoring, and a comprehensive set of services (9–11). Substance use disorders are among the most common of psychiatric disorders (12). Moreover, the comorbidity of substance use disorders with all types of other psychiatric disorders is so widespread that all psychiatrists need to be familiar with the recognition and treatment of substance use disorders in order to effectively address the needs of their patients.

Treatment strategies and effectiveness

Behavioral and psychotherapeutic interventions

Brief interventions

Brief motivational interventions, which were developed primarily for alcohol users and cigarette smokers, are used with persons who manifest risky behavior but not dependence. They are characterized by low-intensity, short-duration counseling (three to five sessions of 5–60 minutes). In these interventions, typically a primary care clinician or an employee assistance program provider educates patients about risks and instructs, advises, motivates, and helps patients build refusal skills (13).

One form of brief intervention is motivational interviewing. In this approach a directive, client-centered counseling style characterized by warmth and empathy is used in the application of a few techniques—reflective listening and asking key questions, for example—to facilitate change in addictive behaviors (14, 15). Counselors emphasize empathy, creating discrepancy, avoiding arguments, rolling with resistance, and enhancing self-efficacy. Motivational interviewing attempts to induce a “motivational discrepancy” in the client’s mind between present behavior and desired goals. Discrepancy, an uncomfortable state, overrides resistance and motivates behavior change. A meta-analysis of studies of motivational interviewing found that most evaluate motivational interviewing either as a stand-alone intervention or as preparation for more intensive treatment, and most report favorable outcomes (16).

Cognitive behavioral therapies

A wide diversity of treatment approaches fall under the umbrella of cognitive behavioral therapies; however, all variants of cognitive behavioral therapy are based on social learning theory (17). Deficits in ability to cope with stress and substance-related cues weaken capacity to stop substance abuse, maintain abstinence, and prevent relapse. Individual and group cognitive behavioral therapy strives for self-efficacy and adaptive mastery to stressful situations through functional analyses of the antecedents and consequences of substance use. Clients are trained to use multifaceted skills to deal effectively with stressful situations and environmental cues that trigger the desire to use substances. Marlatt and Gordon’s relapse prevention approach uses many of these techniques to focus on “abstinence maintenance” (18). For alcohol, a recent meta-analysis indicates that cognitive behavioral therapy is both efficacious and effective in achieving abstinence and preventing relapse (19). Two meta-analyses of relapse prevention had similar findings (20, 21). However, cognitive behavioral therapy and relapse prevention do not appear to be more effective than other bona fide psychosocial treatments, such as 12-step facilitation (20–23).

For cocaine dependence, efficacy (24–26) and effectiveness (27) studies show that cognitive behavioral therapy and relapse prevention markedly reduce cocaine use over the course of a year, and continuing relapse prevention treatment can preserve these gains beyond a year (28). As with alcohol dependence, cognitive behavioral therapy and relapse prevention did not outperform comparison treatments of individual or group counseling (25).

Contingency management

Skinnerian operant conditioning principles provide the theoretical underpinning for contingency management. In contingency management, the consequences of continued substance abuse are altered by introducing incentives (e.g., methadone dose increases, money, and vouchers for valued items) to reduce the attractiveness of drug use and to increase the attractiveness of abstinence and prosocial behaviors (29). A meta-analysis of contingency management studies conducted in methadone maintenance populations found contingency management to be more effective than usual care in reducing illicit opiate abuse (30). The greatest improvements occurred with use of the most powerful incentives (increases in methadone dose and methadone take-home privileges), an immediate reinforcement schedule targeting a single drug (e.g., illicit opiates, cocaine, and alcohol), and frequent urine drug screening to verify abstinence (30).

Contingency management strategies, alone or in combination with other therapy, have been demonstrated to be more effective than standard treatment for cocaine dependence (31–36). In two controlled efficacy trials, Carroll and colleagues demonstrated that contingency management increased compliance with naltrexone treatment and decreased opioid and cocaine use in a group of opioid-dependent individuals after detoxification (37, 38). Other recent studies have demonstrated that contingency management is effective in alcohol dependence (39), drug use in pregnant and early postpartum women (40, 41), and community settings (39, 42).

Community reinforcement approach

Drawing from operant (contingency management) and social learning approaches (cognitive behavioral therapy and relapse prevention), the community reinforcement approach’s core procedures include functional behavioral assessments of substance use situations; skills training in refusal of alcohol or drugs, in communication, and in problem solving; and identification and use of natural reinforcers in community settings to decrease substance use. Limited research on the community reinforcement approach has shown promising results for alcohol (43), cocaine (35), and opiates (44). The community reinforcement approach has been expanded into a strategy for training a significant other to motivate treatment-refusing patients to enter treatment. This community reinforcement approach and family training intervention brought two-thirds of treatment refusers into treatment and also served as an effective therapy for significant others (45–48).

Multidimensional family therapy

A recent meta-analysis of controlled efficacy trials concluded that family approaches to adolescent drug abuse were superior to individual or peer-group therapy or usual treatment (49). Of the family therapies, those emphasizing multilevel interventions with family systems and other key people were more effective than multifamily or adolescent group therapy in reducing substance abuse and improving family relations and prosocial behaviors (49). One such approach is multidimensional family therapy, a stand-alone, family-based, developmental-ecological, multiple-systems approach (50). In a comparison of multidimensional family therapy with multifamily psychoeducation groups and group cognitive behavioral therapy conducted by Liddle and colleagues, multidimensional family therapy achieved greater reductions in substance abuse and greater improvements in school performance and family functioning (51).

Brief strategic family therapy

Brief strategic family therapy, designed for Hispanic families, intervenes at the family and social level to achieve behavior change in adolescents (52). For severe adolescent substance abuse and disorganized family systems, therapists increase intervention intensity by conducting treatment in the homes of families and other community settings (53).

Behavioral couples therapy

Originally developed for heterosexual couples—men with alcohol dependence cohabiting with a women without substance use problems—behavioral couples therapy employs cognitive behavioral principles to reduce alcohol use and improve relationships. Behavioral couples therapy prescribes a three-phase, intensive, time-limited, 24-week therapy. After a 4-week individual phase emphasizing skills training, the therapist pursues a 12-week conjoint phase that requires negotiation of a “sobriety” contract, daily support of abstinence with positive reinforcement, increased quality and frequency of communication, and shared recreational activities. A meta-analysis indicated that behavioral couples therapy is more effective compared with individual and group therapies that involve only the person with addiction problems in decreasing use and improving relationship functioning (54). Recent studies have reported on the use of behavioral couples therapy in treatment of drug use disorders (55), in reducing partner violence (56), and in dyads in which the female partner is substance dependent (57).

Matching client characteristics to treatment

Treatment matching assumes that subpopulations of persons with addictions differ in ways that may predict differential response to treatment. Project MATCH, a large multisite efficacy trial, tested a wide variety of interactions, or “matches,” of client characteristics with cognitive behavioral therapy, motivational enhancement therapy, and 12-step facilitation therapies delivered in individual format (23, 58). Surprisingly, none of the matches proved to be significant. In another large multicenter effectiveness study, the Department of Veterans Affairs (VA) examined 12-step, cognitive behavioral therapy, and combined treatments, each delivered in group format (59), and replicated Project MATCH’s findings of equal efficacy across psychotherapies (60).

Pharmacotherapies with demonstrated efficacy

With the recent explosion of information about the neurobiology of addictions, interest in pharmacotherapeutic interventions has dramatically increased. Although a number of studies have investigated treatments for cocaine, marijuana, opioid, and alcohol dependence, efficacy has been demonstrated only for pharmacotherapeutic treatments of opioid and alcohol dependence. However, there are promising pharmacological treatments on the horizon, including development of a cocaine vaccine (61, 62) and agonists and antagonists of the cannabinoid receptor system (63).

Opioid dependence

In recent years the number of persons addicted to illicit opioids has increased. At the same time, the complexity of their disorders has increased along with the purity of heroin and with medical problems associated with injection drug use, particularly HIV and hepatitis C infections (64–66). These trends make the need to optimize the treatment of opioid dependence more urgent.

Studies have repeatedly demonstrated the finding that 80–100 mg/day of methadone is more efficacious than the commonly prescribed lower doses (30–60 mg/day) in treating withdrawal symptoms (67, 68), increasing retention in treatment, reducing needle sharing, and reducing morbidity and mortality rates (69). In a 3-year study of 245 patients in methadone maintenance treatment, Maxwell and colleagues clearly demonstrated the safety of methadone dosing above 100 mg/day (70).

Levomethadyl acetate, or levo-alpha acetyl methadol (LAAM), is an opioid agonist with a longer elimination half-life than methadone, permitting a more convenient in-clinic dosing schedule of three times a week. In 2001, the Food and Drug Administration (FDA) approved unsupervised, take-home use of LAAM, further increasing its convenience (71, 72). However, an association of LAAM use with prolongation of the QT interval and cardiac arrhythmias led the FDA to issue a “black box” warning and a recommendation to obtain pretreatment and follow-up ECGs to ensure safety (73).

Buprenorphine, a partial opioid mu-receptor agonist and weak kappa antagonist, offers greater safety than and equal efficacy to methadone and LAAM. Buprenorphine has a ceiling effect, whereby at high doses its clinical effect reaches a plateau and it begins to act more like an antagonist, which reduces the risk of overdose, respiratory depression, and diversion to other opioid users (74). Optimal dosing is 8–12 mg/day. The sublingual formulation achieves nearly equal distribution as the parenteral formulation, and the inclusion of naloxone in both formulations blocks euphoria if the medication is injected, thus increasing its safety (75).

Johnson and colleagues directly compared opioid agonist therapies and found the longest continuous abstinence with LAAM but superior retention in treatment with methadone (72). This may be a result of LAAM’s longer dose induction period (76) and the greater reinforcement with methadone, a full agonist, compared with buprenorphine (69). Studies using opioid agonist therapy as “medical maintenance” in primary care settings show equivalent outcomes compared with treatment delivered in specialty clinics (77–79). The FDA recently approved buprenorphine sublingual tablets (with a schedule III designation) for routine medical maintenance treatment of opioid-dependent patients in office-based practice.

Alcohol dependence

Naltrexone, an opioid antagonist, was approved by the FDA in 1996 for the treatment of alcohol dependence. It is hypothesized that naltrexone blocks the rewarding effects of alcohol through its effect on endogenous opioid systems. Reviews of naltrexone studies (80, 81) found a modest reduction in relapse rates and percent drinking days, a modest decline in relapse to alcohol dependence, and a decrease in alcohol craving. A recent VA multicenter trial of naltrexone for alcohol dependence produced null findings (82). Further investigation is needed to better define which patients are most likely to benefit from treatment with naltrexone. Naltrexone compliance may be enhanced by combined use of a sustained-release formulation (83) and behavioral incentives (38).

Acamprosate (calcium acetylhomotaurinate) is a novel compound believed to exert its therapeutic effects through excitatory amino acid systems involved in alcohol withdrawal. Thus it may decrease the craving related to conditioned alcohol withdrawal. In a meta-analysis of 11 studies comparing acamprosate to placebo, acamprosate was found to be superior across several measures of alcohol consumption. However, evidence for prevention of relapse to heavy drinking remains sparse. In comparing meta-analytic findings for naltrexone compared with acamprosate, Krantzler tentatively concluded that both agents demonstrate modest efficacy in maintenance of abstinence (81).

A number of other promising pharmacological treatments are available for alcohol use disorders. In a placebo-controlled study of the efficacy of ondansetron, a 5-HT₃ antagonist, Johnson and colleagues demonstrated that it decreased alcohol consumption more in persons who developed alcohol dependence at an early age compared with persons who developed dependence later in life (84). In a placebo-controlled study of the efficacy of sertraline for treatment of alcohol dependence, Pettinati and colleagues found that it decreased alcohol consumption markedly more for persons without a history of depression compared with persons with lifetime depression (85). These studies suggest that the serotonergic system is involved in the pathophysiology of alcohol dependence, and they support the hypothesis that there is a spectrum of biologically based subtypes of alcohol dependence, each of which may respond uniquely to specific pharmacological treatments.

Questions and controversies

In spite of the wealth of new information about treatments for substance use disorders, obvious questions remain. The interventions described above target substance abuse, but they often do not address functioning in other key life domains. Most persons with severe addictions report chronic medical problems, other disabilities, unemployment, and poverty. Few addiction therapies or treatment programs address these problems directly with appropriate health and human services (86).

Matching severe problem needs with services

In a trial matching specific needs with type and intensity of targeted services, McLellan and colleagues found that patients receiving services for specific problems had greater rates of retention in treatment and of improvement in psychiatric, family, and employment performance and decreased relapse rates (87). A 5-year prospective naturalistic study replicated these findings for a nationwide sample of addiction treatment programs (88, 89).

Case management

Little research has been conducted on case management in addictions treatment. Most programs use a brokered model of case management, in which a provider coordinates services within and across agencies to meet needs comprehensively. One controlled study of a brokered model for alcoholic homeless persons found modest improvements in safe housing, income supports, and alcohol use (90). In another study, case-managed patients received more medical and psychosocial services, which resulted in markedly reduced alcohol and drug use, improved medical and mental health, and better family functioning compared with a control group (5, 91). Other studies demonstrate that case management strategies improve psychosocial functioning, increase retention in treatment, and decrease relapse (92, 93). In a naturalistic study of 300 African-American HIV-positive injection drug users, case management markedly improved access to and use of optimal outpatient medical care (94).

Linkage of addiction treatment with medical services

When addiction treatment programs co-locate addictions and medical services rather than referring patients out for medical services, access and utilization of both services dramatically increases (95), and substance use and general health problems show marked improvements. In a study of injection drug users with serious medical problems treated by an inner-city methadone maintenance program offering on-site medical services, illicit opiate use and health gains were markedly better than those of patients referred out for medical care (96). This finding has been replicated in studies of men with severe alcoholism and severe alcohol-related illnesses served in a VA primary care clinic (97) and for persons receiving integrated substance abuse and primary care in a California health maintenance organization (98). When referral out for services becomes necessary, if program providers personally transport clients to services, then utilization by clients is virtually guaranteed. However, if a program simply issues vouchers to clients for covering transportation costs, they rarely use the vouchers to access needed services (99).

Community support groups

In the United States, persons with addictions rely more on community support groups (e.g., Alcoholics Anonymous) than on the formal addictions treatment system (100, 101). Use of community support groups has increased markedly over the past two decades, while the availability and comprehensiveness of formal addictions treatment services have declined (86), possibly reflecting a service substitution effect. Three meta-analyses and one box-score review report that the overall quality of research evaluating effectiveness of community support groups is so poor that no firm conclusions can be drawn about the impact of community support groups on addiction outcomes (19, 102–104). Such studies fuel the heated, decades-long debate among formal treatment providers and community support group members about the groups’ effectiveness and about whether each should recommend use of the other for management of addiction.

However, recent well-designed controlled and observational studies of community support groups provide evidence that they do help mitigate severity of addictive illness. The most crucial aspect of community support groups in promoting abstinence from alcohol is long-term participation (i.e., years) beyond conclusion of formal treatment episodes (i.e., weeks to months) (105–107). This finding has been replicated in recent studies of cocaine and opiate dependence (108, 109). All studies report an association of a threshold effect of at least weekly participation in a community support group in order for this approach to be effective. Clients are more likely to participate in community support groups over the long term if their formal treatment program (e.g., 12-step programs and cognitive behavioral therapy) encourages such participation. The 12-step-oriented programs seem to motivate more participation (110–113). The mechanisms of behavior change promoted by participation in community support groups remain unknown, but recent penetrating evaluations of such groups suggest that, as with formal treatment approaches, community support group activities drive change with therapeutic processes common to both (107, 114). The long-term (even lifelong) orientation of community support groups fits well with views of addictions as chronic illnesses in need of continuing care (10).

Recommendations for treatment and further research

A wide variety of evidence-based treatments are currently available in the addictions field. Many of these treatments demonstrate effectiveness in typical practice settings. Research has demonstrated the superiority of combinations of therapies, such as using contingency management to enhance compliance with medication regimens. While treatment matching by client characteristics has not shown success in psychotherapy research, pharmacotherapy studies show promise in subtyping individuals on the basis of individual client characteristics (e.g., age at onset of substance dependence) to provide a rationale for choosing the most appropriate medication.

Given the neurobiological diversity of individuals with addictive disorders and the availability of a broad menu of potential treatments, further research in identifying optimal matches between specific pharmacotherapies and well-defined subpopulations of persons addicted to drugs and alcohol would help in maximizing the use of limited treatment resources. Further research on the most appropriate treatment for individuals with psychiatric comorbidity and substance use disorders is also needed, given lack of response to standard treatments, high rates of disability, and overuse of expensive emergency services and hospital inpatient services. Thus, at present, the choice of therapy (or therapies) should be based on the needs of the individual client, within the constraints of service availability and capacity to pay. For persons with chronic severe addictions and complex medical and psychosocial problems, a comprehensive approach with a continuing treatment format and participation in community support groups is essential. Research should be directed toward maximizing therapeutic impact by addressing patients’ multifaceted needs with specific combinations of therapies (behavioral, psychological, and pharmacological), psychosocial services, and community support activities.