Antipsychotic treatment adherence and associated mental health care use among individuals with bipolar disorder.

BACKGROUND: Up to 48% of patients with bipolar disorder are either nonadherent or partially adherent to antipsychotic drug treatment. Medication adherence may differ by bipolar disorder subtype.
OBJECTIVE: This study evaluated the association between antipsychotic treatment adherence and mental health care use among individuals with bipolar disorder with predominantly manic/mixed symptoms or predominantly depressive symptoms.
METHODS: Individuals with bipolar or manic disorder who had at least 1 medical claim with International Classification of Diseases, Ninth Revision, Clinical Modification codes 296.4-296.8 (bipolar disorder) or 296.0 or 296.1 (manic disorder) were identified from medical and pharmacy claims in the PharMetrics database for the period from January 1999 through December 2004. Adherence was measured by intensity (medication possession ratio [MPR]) and treatment duration. The association between adherence and health care use during and after antipsychotic treatment was evaluated using multiple regression analysis. The traditional P < 0.05 threshold was used for statistical significance; however, results that approached significance at P < 0.10 were also noted.
RESULTS: Claims data were examined for 13,941 antipsychotic treatment episodes occurring in 12,952 individuals with bipolar or manic disorder. Of these, 6153 treatment episodes occurred in 5711 individuals with predominantly manic/mixed symptoms, and 2617 occurred in 2381 individuals with predominantly depressive symptoms. The remaining 5171 treatment episodes occurred in 4860 individuals with unspecified bipolar disorder and were not included in the analysis. In individuals with manic/mixed symptoms, a higher MPR was associated with reduced total and outpatient mental health expenditures over subsequent stages of treatment (reduction in total expenditure per 1-point increment in MPR: $123-$439; P < 0.001). In individuals with predominantly depressive symptoms, the association between MPR and subsequent mental health expenditure reached statistical significance only in months 10-12, the 3rd of the 4 treatment segments examined (total mental health expenditure: -$714 [P < 0.001]; outpatient mental health expenditure: -$468 [P < 0.001]). A higher MPR was also associated with a lower likelihood of acute mental health care (inpatient hospitalization or an emergency department visit) in subsequent months in individuals with manic/mixed symptoms or depressive symptoms (odds ratio = 0.545 [95% CI, 0.30- 1.00] and 0.395 [95% CI, 0.14-1.12], respectively; both NS at the P < 0.05 threshold), and was not associated with mental health inpatient days. In both subgroups, a longer duration of treatment was associated with lower total and outpatient mental health expenditures during the 4 months after the termination of treatment (both, P < 0.01).
CONCLUSIONS: In these individuals with bipolar or manic disorder, improved adherence to antipsychotic treatment was associated with lower subsequent total and outpatient mental health care expenditures. This association was less pronounced in individuals with predominantly depressive symptoms than in those with predominantly manic/mixed symptoms.