(SLA-162) The Prevalence of Central Nervous System-Active Polypharmacy in U.S. Nursing Homes During the COVID-19 Visitation Restriction

Background: In response to the COVID-19 pandemic, US nursing homes implemented visitation restrictions. The dynamics of the pandemic, visitation restrictions, and resident stressors may contribute to changes in psychotropic drug consumption. Residents taking multiple psychotropic medications are at risk of central nervous system-active (CNS) polypharmacy, necessitating information during this period.

Objectives: To examine the prevalence, patterns of CNS polypharmacy, and associated factors in nursing homes in 2021.

Methods: A cross-sectional study was conducted using the Minimum Data Set 3.0 linked with Medicare Part D claims data. Long-stay nursing home residents aged ≥ 65 years with annual assessments in 2021 were identified (n=238,378). CNS polypharmacy was defined as the concurrent use of ≥3 medications for over 30 days of continuous exposure for: antidepressants, antiseizure medications, antipsychotics, benzodiazepines, z-drugs, opioids, and skeletal muscle relaxants. We estimated the prevalence of CNS polypharmacy and described the number of distinct medications and drug class combinations. Modified Poisson models with robust error variance and facility-level clustering were used to estimate the adjusted-prevalence ratio (aPR) of factors associated with CNS polypharmacy, compared to receiving no CNS medications.

Results: In this national study of US nursing homes, the prevalence of CNS polypharmacy was 14.3%, whereas 50.0% of residents received 1 or 2 CNS medications, and 35.7% received none. Within those exposed to CNS polypharmacy (n=33,968), 65.5% received 3 CNS medications and 24.5% received 4 concurrently. The most common drug class combination was 1 antidepressant used with 1 antiseizure medication and 1 antipsychotic (11.0%). The next frequent combination was 2 antidepressants and 1 antiseizure medication (8.1%). Gabapentin (41.3%) was the most frequent contributor to CNS polypharmacy, followed by trazodone (26.5%), sertraline (23.1%), divalproex (21.1%), and quetiapine (18.2%). Factors associated with CNS polypharmacy were comorbidities including anxiety (aPR 1.66 [95% CI:1.63-1.69]), bipolar disorder (aPR 1.36 [1.34-1.39]), depression (aPR 2.74 [2.67-2.82]), schizophrenia (aPR 1.47 [1.44-1.50]), psychosis (aPR 1.32 [1.30-1.35]), moderate/severe pain (aPR 1.51 (1.48-1.55) and recent history of falls (aPR 1.18 [1.16-1.20]), low social connectedness index (aPR 1.15 [1.13-1.18]).

Conclusions: CNS polypharmacy was common in US nursing homes, especially among residents with psychiatric comorbidity and in pain. These findings underscore the need for deprescribing strategies and medication monitoring plans to mitigate risks associated with CNS polypharmacy.