AGING, DISABILITY, AND END‐OF‐LIFE

Abstract

Research ObjectivePreliminary research demonstrated an increase in the reporting of schizophrenia diagnoses among nursing home (NH) residents subsequent to the Centers for Medicare & Medicaid Services (CMS) National Partnership to Improve Dementia Care. Some researchers posit that as much as 20% of the reduction in inappropriate antipsychotic use could be explained by the changes in schizophrenia reporting. Given known health disparities and higher antipsychotic use for black NH residents, we examined how race and dementia influence the rate of schizophrenia diagnoses among NH residents. To date, no study uses a formal test, concurrently adjusting for relevant covariates, to examine trends in schizophrenia reporting following the National Partnership.Study DesignUsing a quasi‐experimental study design, we examined long‐stay NH residents’ quarterly and/or yearly Minimum Data Set (MDS) 3.0 assessments between 2011 and 2015. The main outcome variable was the presence of a schizophrenia diagnosis. Using a difference‐in‐difference analysis adjusting for independent variables, we examined the relationship before and after the partnership. Our independent variables were added to our unadjusted model in a stepwise fashion: 1) age, sex, and race; 2) diabetes and heart conditions indicating poor heart health; and finally, 3) activities of daily living and changes in health, end‐stage disease, and symptoms and signs scale scores.Population StudiedLong‐stay US NH residents.Principal FindingsThere were over 4 million MDS assessments annually. Schizophrenia reporting increased by 9.1% in the dementia group as compared to 9.9% in the nondementia group (P < 0.0001). Black residents had a significantly higher likelihood of schizophrenia reporting (4%, P < 0.0001). After controlling for covariates, there was a 16.7% increase in schizophrenia reporting for blacks with dementia and 14.6% increase for non‐blacks with dementia (P < .0001). There was no racial disparity identified among the nondementia group after controlling for covariates.ConclusionsBlack NH residents were more likely to have schizophrenia documented on their MDS assessments, and following the partnership to rein in antipsychotic medication use, schizophrenia reporting rates increased faster for blacks with dementia than their non‐black peers and their peers without dementia. Further work is needed to determine whether schizophrenia diagnoses are appropriately employed in NH practice, particularly for black Americans.Implications for Policy or PracticeResearchers and policy makers should consider race and socioeconomic status when examining the impact of new policies and practices. The 2012 National Partnership may have had unintended consequences that negatively impacted black NH residents with dementia. Additionally, controlling for relevant covariates is important for properly interpreting changes in diagnoses and potentially for measuring the changes in inappropriate antipsychotics.Primary Funding SourceThe study was funded by the Agency for Healthcare Research and Quality.