Abstract
Background/Aims Although a number of international guidelines recommend antipsychotic monotherapy in patients with schizophrenia, frequently the complex pharmacological treatment for these individuals involves multiple medications. To investigate prescribing patterns for patients with schizophrenia, this study examined antipsychotic polypharmacy across multiple outpatient healthcare settings (including two HMORN sites) and their association with hospital admission.
Methods This multi-system study utilized data on patients diagnosed with schizophrenia, including 119,662 Veterans in the Department of Veterans Affairs (VA) healthcare system, 553 and 4,887 patients in two private, integrated health systems (HMORN), and outpatients (17,596,617 visits in 1-week look-back) from a nationally representative sample of U.S. residents seeking care outside federal systems (National Ambulatory Medical Care Survey, NAMCS). Antipsychotic polypharmacy was defined as use of more than one antipsychotic drug (first or second generation) during the covered period (week, year). The prevalence and trend of antipsychotic polypharmacy was assessed in each system (2002–2009 or 2005–2009) and their association with one-year hospital admission using multivariable logistic regression.
Results Annual antipsychotic treatment in the VA ranged between 74–78% each year, with the lowest rates observed in the HMORN systems (49–67% site 1, 22–41% site 2) per pharmacy fill data; NAMCS ranged between 69–84% per clinician-reported prescriptions. Polypharmacy rates depended on the defined covered period. The VA had lower polypharmacy when data were restricted to the one-week covered period used in non-federal systems (20–22% vs. 19–31% NAMCS). In each system, polypharmacy was associated with increased odds of admission (odds ratio ranging 1.4–2.4).
Conclusions The unadjusted longitudinal trends suggest tremendous system variations in antipsychotic use in patients with schizophrenia. Cross-system comparisons are inherently subject to uncertainty due to variation in the amount and type of data collected (e.g., look-back period, pharmacy records versus clinician chart review), yet represent a significant effort to understand organizational differences and context. Given current debates over healthcare access and treatment costs, electronic systems to signal polypharmacy could assist in identifying patients requiring more complex clinical and pharmacy management, individuals at substantially higher risk for adverse events. Such enhanced sentinel detection and follow-up care could ultimately lead to improved clinical practice and fiscal well-being.
