substance abuse treatment. There is emerging consensus that lack of integration leads to poor coordination of services, interagency miscommunication, and funding conflicts, all of which contribute to attrition and poor outcomes for patients (Osher & Drake, 1996; Report to Congress, 2002). This is particularly troubling since co-occurring mental health distress is associated with substance use severity, greater psychosocial impairment, treatment resistance, and poorer long-term prognosis (Diamond, Panichelli-Mindel, Shera, Tims, & Ungemack, in press; Drake, Mueser, Clark, & Wallach, 1996; Shane et al., under review). Consequently, the most severe and chronic patients often receive the poorest care, leading to repeated visits to hospital emergency rooms and inpatient and residential facilities (Richardson, Craig, & Haughland, 1995). The end result is that comorbid patients in need of care are consuming a major portion of treatment funding (Ridgely, Goldman, & Willenbring, 1990).

The gap between substance abuse and mental health dates back to the 1930s (Rosenthal & Westreich, 1999). At that time, psychodynamic therapists, who dominated the treatment world, believed that addicts' personality structure was not amenable to the analytic method, and therefore addicts were not treatable. This attitude may persist today among practitioners in the mental health community, who tend to view addiction as inhibiting treatment of other “underlying” problems. Simultaneously, the self-help movement developed independent of the mental health community, and as the self-help philosophies and programs matured, educational and professional licensure pathways emerged that legitimized and strengthened these approaches (Rosenthal & Westreich, 1999). As often happens, these ideological differences became institutionalized and perpetuated a division that does not reflect the clinical realities of the patients.

Recognition of this schism has inspired many attempts to integrate substance abuse and mental health treatment programs for adult dual-diagnosis populations (Drake, Mchugo, & Noordsy, 1993; Miller & DelBoca, 1994; Minkoff & Drake, 1991). At least 36 studies have evaluated different versions of integrated programs at all levels of care (e.g., outpatient, day treatment, inpatient, residential, etc). Some studies added a substance abuse group to outpatient mental health services, resulting in reduced dropout, decreased hospitalization, and increased abstinence (e.g., Hellerstein, Rosenthal, & Miner, 2001; Osher & Kofoed, 1989). Studies that combined substance abuse services with inpatient, day treatment, and residential care have also shown some benefits as long as patients remained in the program. Unfortunately, attrition was often high and once patients were discharged, relapse rates were high as well (e.g., Rahav et al., 1995).

A major contribution to this area was the 1987 funding of 13 dual-diagnosis demonstration projects. These studies demonstrated that integrated programs could be implemented in a number of settings, resulting in increased engagement and services utilization and reduced drug use. Five recent studies were conducted on comprehensive integrated systems using more sophisticated treatment programs and quasiexperimental or true experimental designs (e.g., Drake, Mercer-McFadden, Muesser, McHugo, & Blond, 1998; Ridgely & Lambert, 1999; Godley, 1994; Jerrell & Ridgely, 1995). These studies showed significant reductions in substance use, program readmission, and hospitalization and improvement in other functional outcomes (Drake et al., 1998). However, there has been little or no comparable research on the effectiveness of integrated programs for adolescent substance users.

Clinical research during the past 10 years has identified a number of effective treatments for adolescent substance users. Although the field is still young, this growing body of work has yielded several important findings that support the effectiveness of carefully implemented, structured behavioral approaches for adolescent substance use (Liddle & Rowe, in press; Stanton & Shaddish, 1997; Williams et al., 2000). These can be summarized as follows:

Given the clinical importance of drug euphoria and drug craving, most pharmacological strategies for addiction target these primary reinforcers. Drug-induced reward is attenuated in animal models by a number of agents, depending on the drug in question. These medications act on dopamine, opioid, glutamate, or GABA systems. These reward-blocking medications have been tested in human substance abusers to determine whether they reduce drug euphoria under controlled settings or promote abstinence in clinical trials. Other means of reducing reward have also been tested, including vaccines that block the entry of an addictive substance into the brain, and agents like disulfiram that produce aversive symptoms when alcohol is consumed. In addition to strategies that reduce drug euphoria,

strategies that reduce craving have also been tested and prescribed. Agonist treatment (prescribing a substance that replaces the addictive drug) has been used in opioid (e.g., methadone, buprenorphine) and nicotine (e.g., nicotine gum) dependence with considerable success, providing a means of bypassing dangerous routes of administration or hazards associated with drug procurement. The reversing of clinically relevant neuroadaptations associated with chronic exposure to addictive substances has the theoretical ability to reduce craving and other aversive aspects of addiction.

Unfortunately, there has been little research directed toward the pharmacological treatment of substance dependence in adolescents. For a number of reasons, there are no controlled trials evaluating the effectiveness of substitution or replacement therapies (e.g., methadone, buprenorphine), antagonists (e.g., naltrexone), aversive therapies (e.g., disulfiram), or anticraving medications (e.g., bupropion, naltrexone) in this subpopulation. Therefore, if such medications are used in adolescents, they must be used with caution, careful monitoring, and consideration of the developmental characteristics that distinguish adult patients from adolescents (e.g., greater impulsivity and polydrug use; Solhkhah & Grenyer, 1998). More research is clearly needed in this area.

Since the most commonly abused substances by adolescents are nicotine, alcohol, and cannabis, these are the most likely drugs for which pharmacotherapy questions might arise. We will review these medications briefly, starting with those used in detoxification. Advances in our understanding of the mechanisms of drug craving and drug-induced euphoria should guide future research and shed light on more effective pharmacological treatments for addiction in adolescents.

Medical detoxification is required for alcohol, sedatives, and opiates, but not for other abused drugs. Detoxification from alcohol dependence can be effectively attained by using benzodiazepines or barbiturates, and anticonvulsants such as valproate and carbamazepine to block or reverse withdrawal symptoms (Kosten & O'Connor, 2003). These detoxification medications should be used in adolescents if withdrawal symptoms are significant, particularly because alcohol withdrawal is potentially life threatening. Detoxification from sedative hypnotic dependence can also be accomplished by prescribing descending doses of benzodiazepines and barbiturates.

For opioid dependence, the most common means of detoxification involves prescribing descending doses of methadone for a period of 2 to 4 days while carefully monitoring the patient's response. Methadone is a long-acting opioid agonist that reverses heroin withdrawal by replacing heroin at the opioid receptor. Since methadone has the potential to cause lethal opioid overdose, and opioid withdrawal is not medically dangerous, it is imperative to avoid prescribing an excessive dose of methadone to adolescents. The appropriate dose is best selected by closely monitoring the signs of opioid withdrawal, which should be given more weight than reported symptoms that might be exaggerated or feigned by drug-seeking patients.

A new treatment for detoxification and maintenance, the partial agonist buprenorphine, was made available in the United States in 2003. It may be ideally suited to adolescents and is currently in clinical trials in this population. Detoxification with this medication is very simple because overdose is almost impossible. The patient can be transferred from the opiate of abuse to buprenorphine and then the dose is gradually reduced with minimal or absent withdrawal symptoms. Yet another option is the nonopioid clonidine, an antihypertensive medication that blocks many of the opiate withdrawal symptoms (Gold, Dackis, & Washton, 1984). Most patients prefer methadone or buprenorphine because of greater comfort.

The treatment of heroin and other opioid dependence often begins with inpatient detoxification of heroin withdrawal that should also involve specialized drug rehabilitation and aftercare referral (Dackis & O'Brien, 2003b). Unfortunately, considerable availability and access problems preclude many adolescents from receiving appropriate inpatient treatment. Still,

hospitalization is the safest and most conservative treatment approach to this potentially lethal condition (Dackis & Gold, 1992). Inpatient treatment provides a controlled environment in which abstinence can be assured while a comprehensive medical and psychiatric evaluation is conducted. The high mortality rate in intravenous adolescent heroin users results not only from overdose but also from trauma, medical conditions related to the use of needles, and the concomitant use of other drugs and alcohol. Therefore, as reviewed elsewhere (Dackis & Gold, 1992), a comprehensive physical examination, medical history, and laboratory evaluation are indicated in all addicted adolescents.

Familiarity with the medical and psychiatric complications of heroin dependence enhances the clinician's ability to identify and treat these commonly occurring conditions. Infections related to intravenous heroin use include acquired immunodeficiency syndrome (AIDS), viral hepatitis, endocarditis, meningitis, tuberculosis, abscesses, infected injection sites, and pneumonia. Additionally, unprotected sex is common among addicted adolescents (Crome, Christian, & Green, 1998), leading to a preponderance of sexually transmitted diseases. Heroin often produces irregular menses in women and sexual performance problems in men, apparently by dysregulating the hypothalamic–pituitary–gonadal axis (Malik, Khan, Jabbar, & Iqbal, 1992). The most common psychiatric problem associated with heroin dependence is depression (Handelsman, Aronson, Ness, Cochrane, & Kanof, 1992), which should be expeditiously identified and appropriately treated to avoid the risk of suicide and to facilitate the recovery process.

Inpatient detoxification treatment should not be restricted merely to the medical management of heroin withdrawal. This intensive intervention provides the physician with an ideal opportunity to establish a therapeutic alliance with adolescent patients by concomitantly addressing the critical treatment issues of honesty, openness, trust, denial, and engagement. Inpatient detoxification also provides an opportunity to fully evaluate the patient, assess their readiness for change, and provide critical family therapy. Since families require education, support, and guidance throughout the process, clinicians should be familiar with psychosocial as well as medical aspects of heroin addiction. It is essential to emphasize that detoxification, in and of itself, is not sufficient treatment for heroin dependence and must therefore be followed by ongoing outpatient drug rehabilitation. The recent fad of very rapid detoxification with general anesthesia has not been shown to produce better outcomes than standard detoxification.

The nature of addiction requires that, after detoxification, complete abstinence be the treatment goal for addicted adolescents, rather than the mere reduction of drug and alcohol use. Once the cycle of addiction has become entrenched, casual use is seldom possible. Indeed, even the use of other addictive agents, such as alcohol by a cocaine-dependent adolescent, often leads to relapse to the drug of choice. Thus, total abstinence from all addicting drugs should be the goal when treating adolescents. After attaining abstinence, preventing relapse to drug dependence is the primary clinical target in adolescents, and to that end, medications for relapse prevention are likely to be useful. A few specific relapse prevention pharmacotherapies are U.S. Food and Drug Administration (FDA) approved for nicotine and alcohol, but none has been tested in adolescents. For nicotine the medications are nicotine replacement and bupropion, and for alcohol the medications are disulfiram and naltrexone. However, before medicating adolescents, it is imperative to determine that they will be cooperative, that parental consent has been obtained, and that the adolescents and parents have the same understanding of treatment goals and approaches.