In Reply: The issues Dr. Ulus has raised are important ones—the need for caution in prescribing medications for individuals who are using substances, the risk of interactive side effects, the concern that medication may be less effective for these individuals, the importance of continuing to recommend abstinence and offering substance abuse treatment, and, finally, specific concerns about clinical, ethical, and legal risk in engaging in such practices. One major purpose of the best practice project was to provide guidelines to assist prescribers in addressing these issues in real-world clinical situations.
My first response is that all the concerns Dr. Ulus has raised are valid. Caution is warranted in the context of providing medication for a person with serious mental illness while that individual continues to use substances. Further, the best-practice principles do not say "never discontinue any medication"; nor do they say "give the patient all the benzodiazepines he or she asks for." The principle is that if the patient has a known serious mental illness and is on a regimen of a nonaddictive medication without which he or she will almost certainly decompensate, continued medication must be maintained. If the patient is engaging in specifically dangerous behavior, closer monitoring and more supervision of the medication may be required, including provision of depot injections.
Furthermore, abstinence is of course recommended, and substance abuse treatment should be offered; however, given that most patients who have serious mental illness do not routinely adhere to such recommendations, the clinician is expected to do what is necessary to maximize stabilization of the psychiatric condition while applying a variety of motivational enhancement strategies in the context of continuing treatment to engage the individual in beginning to reduce substance use.
Discontinuing medication in such instances will lead only to exacerbation of psychiatric symptoms, disruption of treatment, and an increased likelihood of a poor outcome. It is true, moreover, that medication may be less effective and compliance less consistent in the context of continuing substance use. Recent studies indicate that medication does work, and, in fact, "better medication" (for example, an atypical agent, particularly clozapine, rather than a conventional neuroleptic) produces better outcomes for both disorders (1). Further, the potential risk of interactive side effects is small compared with the definite risk of a poor outcome due to medication discontinuation.
Where is the clinical, ethical, and legal support for this recommendation? Clinically, the recommendation is based on established evidence-based best practice for individuals with the most serious mental illnesses and substance use disorders. In the integrated continuous treatment model described by Drake and others (2) as evidence-based practice in the April issue of this journal, individuals are engaged in ongoing integrated treatment for years, including continued provision of psychotropic medication and motivational enhancement for substance use disorders, while they are actively engaged in out-of-control substance use. In this model it takes, on average, four years for 50 percent of disengaged clients to establish stable abstinence.
This best-practice principle is further supported by the results of the psychopharmacologic research cited above (1). Legally, psychiatrists are on safest grounds when they follow established clinical best practices. The best practice that you have questioned was established by clinical consensus of leading psychiatrists in the area of comorbidity.
Last, but certainly not least, it is frankly unethical from a humanitarian and moral perspective to deny necessary treatment to individuals with psychiatric disorders simply because they do not adhere to all of our recommendations. We may recommend to individuals with severe medical conditions that they do not use substances, but we would not consider denying a person with diabetes access to insulin because that individual refused to listen. We need to treat individuals who have serious psychiatric disorders the same way we treat those who have "real" medical conditions, because, in fact, severe mental illnesses are real disorders, and the devastation that may occur by denying treatment is equally real.