Patterns and predictors of antipsychotic medication use among the U.S. population: Findings from the Medical Expenditure Panel Survey

Abstract

Given the importance of pharmacological treatment in mental disorders, it is important to have a thorough understanding of predictors and variations in antipsychotic use. To provide a description of patient characteristics associated with antipsychotic use and to examine predictors of atypical antipsychotic use among antipsychotic users. Data were obtained from the 2004 and 2005 Medical Expenditure Panel Survey. Dependent variables were annual, self-reported, atypical and typical antipsychotic use. Independent variables included predisposing, enabling, and need characteristics according to Andersen's Behavioral Model. In addition to descriptive statistics, logistic regression analyses were performed to examine the determinants of antipsychotic use. Patients aged 65 and older were 0.63 times as likely to use antipsychotics as patients aged 26-45. Poor and near-poor patients were 1.55 and 1.37 times as likely to use antipsychotics as middle- to high-income patients, respectively. The odds of antipsychotic use were 2.95 and 1.99 times for patients with public and prescription insurance coverage, respectively. Patients with a usual source of health care were 1.51 times as likely to use antipsychotics as those without. Compared with typical antipsychotic use, patients aged 25 and younger were 3.88 times as likely to use atypical antipsychotics as patients aged between 26 and 45. Urban residents were 1.87 times as likely as rural residents to use atypical antipsychotics. The odds of antipsychotic and atypical antipsychotic use for the poor mental health population were 8.73 and 3.87 times as patients with good to excellent mental health status. Predisposing and need factors play important roles in determining the use of antipsychotics. However, among antipsychotic users, the use of atypical versus typical antipsychotics appears to have been influenced primarily by need. These findings should be useful to clinicians and policy makers in directing antipsychotic treatments to patients in need.