Mason et al. (1996) randomly assigned 71 patients with alcohol(Drug information on alcohol) dependence and depression (sober for a median of eight days) to a six-month course of desipramine (Norpramin) or placebo plus encouragement to attend counseling and AA. Desipramine patients had less depression and a longer time to relapse than those on placebo, although most patients in each group relapsed by six months. In another study, McGrath et al. (1996) randomized 69 actively alcoholic outpatients to 12 weeks of imipramine(Drug information on imipramine) or placebo plus weekly counseling. All patients had onset of depression prior to the alcohol dependence or during periods of remission. Imipramine was associated with more improvement in depression than placebo with no clear effect on alcohol use, although imipramine patients whose mood improved showed more decrease in alcohol consumption. Cornelius et al. (1997) randomized 51 severely depressed alcoholics to a 12-week course of fluoxetine(Drug information on fluoxetine) (Prozac) or placebo. There was a strong treatment effect, with fluoxetine patients having significantly more improvement in depression and less alcohol use than placebo patients. In one of the few pharmacotherapy studies of anxious alcoholics, Kranzler et al. (1994) randomized 61 detoxified anxious alcoholics to weekly relapse prevention plus buspirone(Drug information on buspirone) (BuSpar) or placebo. At 12 weeks, buspirone patients had less anxiety and a slower return to heavy alcohol use than placebo patients. No serious adverse events attributable to medication were reported in these studies, although there were more dropouts due to medication side effects in the McGrath et al. study (1996).
Regarding bipolar disorders (BDs) and comorbid substance-use disorders, Brady et al. (1995) treated nine acutely manic patients with substance-use disorders using divalproex (Depakote) and found good improvement in mania, no adverse effects and a decrease in substance use. In one of the few random assignment studies with patients with BD, Geller et al. (1998) randomly assigned 46 adolescents with BD and substance-use disorders (mainly marijuana and alcohol dependence) to psychosocial treatment plus lithium(Drug information on lithium) (Eskalith, Lithobid) or lithium placebo. Lithium patients had more improvement in BD and substance use than those who received placebo.
Studies of schizophrenia have routinely used medication and focused mainly on treatment delivery models. These studies found that delivering treatment in the same location and using the same staff for both the schizophrenia and the substance-use disorder provided the best results (Hellerstein et al., 2001).
For attention-deficit/hyperactivity disorder (ADHD), a pilot study of bupropion (Wellbutrin) plus psychosocial treatment in 13 adolescents with ADHD, substance-abuse disorder and conduct disorder suggested that bupropion is safe and effective (Riggs et al., 1998). A single-blind study of bupropion for 11 adults with cocaine dependence showed that both ADHD and cocaine use decreased at 12-week follow-up (Levin et al., 2002). These findings were similar to their single-blind study of sustained-release methylphenidate(Drug information on methylphenidate) (Concerta) (Levin et al., 1998).
As in the case of medication studies in psychiatric patients who are not abusing substances, not all results have been positive. However, findings from the studies summarized here are consistent with the idea that medication and psychotherapy or counseling can be combined for patients with substance-use and other psychiatric disorders with additional benefits and few adverse events. These studies suggest that patients who have one or more independent psychiatric disorders in addition to their substance-abuse disorder(s) are those most likely to benefit from combined therapy and that the most likely result is a reduction in psychiatric symptoms, not in substance use. However, reductions in both are seen in some studies. These findings are consistent with clinical experience and common sense--patients with substance-use and additional psychiatric disorders do best if both disorders are treated. Furthermore, substance-use and other psychiatric disorders are on different tracks, with each having a negative influence on the other. All the studies discussed here delivered parallel, integrated treatment. This model is also consistent with common sense and other research findings that are not reviewed here, but that have become increasingly difficult to implement in the age of carveout reimbursement models.