Prevalence and Correlates of Antipsychotic Polypharmacy From the 1970s to 2009: A Systematic Review and Metaregression

Juan A. Gallego, MD; John Bonetti, DO; and Christoph U. Correll, MD
The Zucker Hillside Hospital, Glen Oaks, New York (Drs Correll and Gallego); and New York College of Osteopathic Medicine, Old Westbury (Dr Bonetti)

This abstract presentation did not receive funding.

Background: Antipsychotic polypharmacy (APP), defined as the concurrent use of two or more antipsychotics, is common and poorly understood clinical practice. Several factors have been linked to APP, including geographical, patient, physician and illness characteristics. This study aimed to systematically review prevalence rates of APP and to examine correlates of APP. In addition to summarizing predictors of APP identified in individual studies, predictors across studies were examined using univariate and meta-regression analyses.

Methods: An electronic PubMed search without language restrictions, augmented by hand search, was conducted for articles published between 1966 and March 2009 reporting specifically on APP, using keywords, such as “antipsychotics,” “polypharmacy,” “comedications,” “coprescription,” “concomitant,” etc.

Results: One hundred fifty-four studies were identified (66.9% cross-sectional, 33.1% longitudinal, 5.8% including an intervention to alter APP rates), including 1,576,136 patients (median n=381, mean age: 40.0 years, 58.2% male, 74.9% with schizophrenia). The overall APP rate was 37.6±20.4%, with 4.4±9.0% being on >3 antipsychotics. The most common antipsychotic class combination included first-generation antipsychotics (FGAs) + second-generation antipsychotics (SGAs) (47.8%), followed by FGAs+FGAs (32.7%) and SGAs+SGAs (20.6%). Higher APP rates were significantly associated with: studies conducted in the 1970s; country of origin (Asia (43.3±20.4%) significantly higher than Europe [29.6±19.1%], U.S. [20.7±17.5%] and Oceania [15.0±9.4%]); inpatients (34.1±20.9%) versus outpatients (18.3±15.2%) and higher FGA and anticholinergic use (all p-values<0.0001); higher FGA+FGA rates (p=0.0003), longer illness duration (p=0.0005); higher rates of schizophrenia (p=0.0042); earlier study year (p=0.0042); adults+/-elderly (28.6±21.0%) versus children/adolescents (12.5±10.9%) (p=0.0068), lower FGA+SGA (p=0.018) and antidepressant (p=0.018) use; higher decanoate use (p=0.019); and lower SGA use (p=0.028). Sex, age, proportion of Whites, cross-sectional versus longitudinal assessment, and proportion of clozapine, mood stabilizer, and anxiolytic use were all not significantly associated with APP. In one metaregression model, APP was associated independently with higher anticholinergic use, and studies being conducted outside of the U.S. (r-squared: 0.24, p<0.0001). Results from additional meta-regression models will be presented.

Conclusions: APP is a complex and multifactorial phenomenon. Future studies should take the identified factors into account when designing studies to examine the relative risks and benefits of APP in severely ill patients.

References
Lerma-Carrillo I, de Pablo Bruhlmann S, del Pozo ML, et al. Antipsychotic polypharmacy in patients with schizophrenia in a brief hospitalization unit. Clin Neuropharmacol. 2008;31(6):319–332.

Rupnow MF, Greenspan A, Gharabawi GM, et al. Incidence and costs of polypharmacy: data from a randomized, double-blind, placebo-controlled study of risperidone and quetiapine in patients with schizophrenia or schizoaffective disorder. Curr Med Res Opin. 2007;23(11):2815–2822.

Valuck RJ, Morrato EH, Dodd S, et al. How expensive is antipsychotic polypharmacy? experience from five US state Medicaid programs. Curr Med Res Opin. 2007;23(10):2567–2576.

Gilmer TP, Dolder CR, Folsom DP, et al. Antipsychotic polypharmacy trends among Medicaid beneficiaries with schizophrenia in San Diego County, 1999-2004. Psychiatr Serv. 2007;58(7):1007–1010.

Back