Reprise: Medication adherence and attitudes in patients with bipolar disorder and current versus past substance use disorder

Abstract

Original citation: Teter CJ, Falone AE, Bakaian AM, Tu C, Ongür D, Weiss RD. Medication adherence and attitudes in patients with bipolar disorder and current versus past substance use disorder. Psychiatry Research 2011 [Epub ahead of print].Substance use disorders are a well-documented risk factor for medication nonadherence among patients with psychiatric illnesses. Numerous explanations have been put forth to explain why this is the case (e.g., substance users living a disorganized lifestyle or being more impulsive, or possible neurocognitive impairment from substance use). However, no definitive etiologic explanations have been identified. In order to further tease apart this phenomenon, we examined the impact of substance use disorder (SUD) history among patients with bipolar I disorder (BD) in regards to medication-taking behaviors and attitudes. We compared patients with BD and no history of SUD (BD-NH), BD and past history of SUD (BD-PH), and BD and current SUD (BD-C). It was our thinking (based on prior research) that these sub-groups possess unique characteristics that must be considered separately.Interviews were conducted with inpatients hospitalized for BD, which included diagnostic instruments (e.g., Structured Clinical Interview for DSM-IV; SCID-I) and measures of attitudes and beliefs concerning psychiatric medications. The primary outcome variable was a standardized medication adherence ratio (SMAR) of [medication taken]/[medication prescribed]. The secondary outcome consisted of the 10-item Drug Attitude Inventory (DAI-10). Additional variables were collected, as described in the manuscript.Fifty-four patients with a BD diagnosis participated, which included BD-NH (n=26), BD-PH (n=19), and BD-C (n=9). The SMAR was significantly different among the three groups as shown in Table 1 and this finding remained significant after controlling for numerous patient characteristics. Attitudes regarding medications, measured by the DAI-10, were positive among a significantly higher percentage of BD-PH and BD-NH patients (please see Table 1).Patients with BD-C demonstrated poor medication adherence and attitudes concerning medication management for their BD. In stark contrast, patients with a past history of SUD (i.e., at least one year of not meeting DSM-IV criteria for any SUD prior to the interview) were taking their medications as prescribed and demonstrated very positive attitudes and beliefs towards their BD pharmacotherapy. Perhaps helping patients with BD achieve remission from SUD may lead to a more successful course of medication management. In fact, the current study is part of a growing body of literature which has shown positive outcomes among patients ‘achieving' a past history of SUD (also referred to as recovery or remission).In the published manuscript, we provide a full item analysis which allows the reader to see each individual item contained in the DAI-10. This instrument appears to be effective as a brief screener to identify patients with BD at risk for medication nonadherence. We also provide a medication-specific analysis, although small sample sizes in sub-groups limited our ability to generalize the findings. During our medication-level analysis, it became very apparent that patients receiving lithium (usually in combination with another medication) demonstrated the highest rates of medication adherence. While we were initially surprised, we identified (and briefly described in the manuscript) at least one other recent study describing similar findings. Although the study methods differed between the two studies, they suggest patients with BD take lithium appropriately, despite all of the required monitoring (or perhaps because of the monitoring).A concerning finding was that approximately half of the patients in the entire study endorsed “did not need” BD medication as a reason for lifetime nonadherence. This finding is consistent with research dating back to at least the mid-1990s. It appears that we as clinicians need to focus more on this particular aspect of medication nonadherence. In our paper, we mention a model proposed by Clatworthy and colleagues to help address this particular treatment issue. The authors of this theoretical model suggest that medication nonadherence continues to be associated with doubts about the need for treatment and that these concerns can be placed into the context of the Necessity-Concerns Framework. Perhaps clinicians can use such a model to incorporate patient beliefs regarding their treatment and thus develop targeted, individualized interventions.In conclusion, our findings strongly suggest that a targeted approach of providing interventions on the benefits of pharmacotherapy to patients with BD in specific patient subgroups (e.g., current substance users, patients with negative attitudes towards their BD medications) could be a helpful component of treatment. Further, there appears to be something unique about patients with BD who have remitted from substance use that is associated with improved medication-taking behaviors and better attitudes. This is an important area for future research.There appear to be meaningful differences between patients with a current versus past SUD, and paying attention to these differences is essential for the effective management of patients in the BD population.