Assessing the Unintended Consequences of Antipsychotic Reduction in Nursing Homes: Medication Substitution and Changes to Diagnosis Coding

Antipsychotic medication misuse and overuse have been persistent quality of care issues in nursing homes for decades. Although potentially inappropriate antipsychotic use has declined following various quality improvement efforts, increases in other psychotropic medications, such as trazodone, and increases in the indications used to define criteria for appropriate antipsychotic use (e.g., schizophrenia) have generated concerns about medication substitution and diagnosis upcoding – two potential unintended consequences of antipsychotic reduction efforts. To understand the extent of these potential consequences, I conducted three observational studies of nearly all nursing home residents in Ontario, Canada between 2010 and 2019 using linked health administrative data. In a population-level repeated cross-sectional study, I found that sedating antidepressants and anticonvulsants continued to increase over time. Overall antipsychotic use stabilized following a modest decline in prevalence. The documentation of delusions increased substantially over time, with a clear inflection coinciding with efforts to reduce potentially inappropriate antipsychotics. Increases in sedating antidepressants and documented delusions were greatest among residents with dementia and severe aggressive behaviors.In a facility-level repeated cross-sectional study, I found modest variation in the time trends for potentially inappropriate antipsychotic and antidepressant use across facilities. More substantial variation in the documentation of delusions and associated conditions was observed and shown to be strongly correlated with a homes’ reduction in potentially inappropriate antipsychotic use over time. In a resident-level retrospective cohort study of new admissions to nursing homes, the incidence of trazodone modestly increased while the incidence of antipsychotics significantly decreased over time. Among residents receiving a potentially inappropriate antipsychotic at admission, those who were off an antipsychotic at subsequent assessments were significantly more likely to receive an incident trazodone dispensation compared to those who were on an antipsychotic. These studies provide complementary evidence consistent with antipsychotic medication substitution and changes to diagnosis coding. While these findings likely reflect an increasingly complex nursing home population, they suggest a role for improved quality of care monitoring, such as the expansion of public reporting programs to consider a broader range of quality indicators, investigations into the validity of diagnosis coding, and assessment of the safety and effectiveness of medication substitution.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.