Determinants of atypical antipsychotic use among antipsychotic users in community-dwelling elderly, 1996-2004.

OBJECTIVES

This study examined the determinants of atypical antipsychotic use among antipsychotic users in community-dwelling elderly in the United States.

METHODS

The study involved analysis of household and prescription files of the Medical Expenditure Panel Survey (MEPS) data from 1996 to 2004. The analysis focused on the use of six atypical antipsychotic agents namely, clozapine, olanzapine, risperidone, quetiapine, ziprasidone, and aripiprazole among the elderly of 60 years or older. Multiple logistic regression analysis within the conceptual framework of Andersen's Behavioral Model was used to examine the determinants of atypical antipsychotic use among antipsychotic users in community-dwelling elderly.

RESULTS

An average of 0.62 million elderly received antipsychotic agents annually during the study period. A majority of the elderly using antipsychotic agents were female (70%), white (86%), non-Hispanic (95%), and living in metropolitan statistical areas (79%). Frequently reported diagnoses among the elderly taking antipsychotic agents were dementia (26.12%), anxiety (20.42%), and schizophrenia (6.62%). Of the elderly receiving antipsychotic agents, 50.39% received atypical agents and 51.88% received typical agents during the study period. The most frequently used atypical agents were risperidone, olanzapine, and quetiapine. Multivariate logistic regression analysis revealed that need (perceived mental health, p < 0.01) and enabling (time, p < 0.01) factors were significantly associated with atypical antipsychotic use after controlling for predisposing factors. The study found that elderly patients with relatively poor perception of mental health (need) and utilization of antipsychotic agents after 1998 (enabling) were more likely to involve the use of atypical agents.

LIMITATIONS

This study was limited to the use of antipsychotic agents in community settings and cannot be extrapolated to other settings. Correlates examined in this study were limited to variables available from the data source and those used by previous researchers.

CONCLUSIONS

Need and enabling factors play a vital role in the use of atypical agents in the elderly. The findings have important implications in understanding the use and outcomes of atypical agents in the elderly. Future pharmacoepidemiological research can use these variables to control for confounding and selection bias when evaluating health care outcomes in observational studies.