Potentially inappropriate antipsychotic use in US nursing homes: A safety concern

Background: Half of the claims for antipsychotics used in nursing homes in the United States (US) are for off-label indications, raising concerns about potential misuse of antipsychotic drugs.

Objectives: This study examines whether antipsychotic drug utilization in US nursing homes is associated with the facility financial resources, staff proficiency, and operational efficiency. Further, this study evaluates predictors of antipsychotic drug use based on the underlying demographic characteristics, cognitive impairment and physical functioning of nursing home residents.

Methods: Data were derived from Long Term Care Focus and Nursing Home Compare databases for 2008. Data include organizational characteristics and staffing measures for Medicare/Medicaid certified US nursing homes and residents’ demographics, health symptoms, clinical diagnoses and treatments, physical functioning, and psychosocial wellbeing. Multivariate regression models are performed to test for associations with antipsychotic use while controlling for the mix of residents in the facility. Analyses are conducted using STATA. Statistical tests are based on a two-sided α significance level of p<.05.

Results: Study sample contains 11,339 nursing facilities -2,857 with residents with schizophrenia/bipolar disorder (group I) and 8,482 without the diagnosis (group II). The prevalence of schizophrenia/bipolar disorder in 2008 was 16.0%; antipsychotic use was 33.6% and 21.1% in groups I and II, respectively. Antipsychotic utilization is positively associated with Medicaid as primary payer (p<.01), facility for-profit status (p<.001) and number of certified nurse assistants (p<.01); whereas, it is negatively associated with number of registered nurses in the facility (p<.001), direct care hours provided to residents (p<.001), and admissions per facility bed (p<.001). A one-percentage point increase in the number of registered nurses would decrease antipsychotic use by five percent (p<.001). To the contrary, one-percentage point increase in the number of certified nurse’s assistants would increase antipsychotic use by 0.9 percent (p<.01). Antipsychotic use is also associated with residents’ physical and cognitive impairment. A one percentage point increase in the proportion of low cognitively impaired residents is associated with a 0.1 percent decrease (p<.001) in the antipsychotic drug utilization rate. Likewise, a one percentage point increase in the residents’ activities of daily living score would increase the utilization of antipsychotic drugs by 0.2 percent (p<.001).

Conclusions: Nursing home for-profit status, payment source, and staff proficiency as well as residents’ cognitive impairment, and ability to perform activities of daily living are associated with antipsychotic use irrespective of residents’ diagnosis of schizophrenia/bipolar disorder. Antipsychotics misuse for elderly nursing home residents remains a major safety concern.