BACKGROUND: In 2024, 6.9 million Americans lived with Alzheimer's disease and related dementias (ADRD), with nursing homes serving as a major site of care. Antidepressants are the most prescribed psychotropic medications among nursing home residents with ADRD, yet detailed information on prescribing patterns, potential indications, and associated resident and facility characteristics remain limited. METHODS: Using 2018 minimum data set 3.0 assessments linked to Medicare claims, we conducted a cross-sectional study of long-stay nursing home residents aged ≥ 65 years with ADRD who were continuously enrolled in Medicare fee-for-service for 120 days before their annual assessment. We described usage by drug class and type and compared characteristics of users versus non-users overall and by potential indications: depression, anxiety, pain, and insomnia. RESULTS: Among 232,543 residents with ADRD, 51.6% used antidepressants. Fewer than 5% had moderate or severe depressive symptoms (PHQ-9 ≥ 10). Use was highest among residents with depression or anxiety (69.5%), pain (61.9%), and insomnia (60.0%). Among those without these conditions, 14.1% were prescribed antidepressants. SSRIs were the most prescribed class (60.7%) overall. Citalopram, mirtazapine, sertraline, and trazodone were the most common medications. Residents with any current level of depression severity were more likely to use antidepressants compared to those without symptoms, whereas all levels of cognitive impairment were associated with lower use compared with cognitively intact residents with ADRD. Polypharmacy was strongly associated with increased use, while diabetes, heart failure, and stroke were associated with reduced use. CONCLUSIONS: Antidepressants were frequently prescribed to residents with ADRD despite limited documentation of active depressive symptoms. Limitations in accurately capturing depressive symptoms in nursing home records, including underreporting by residents due to cognitive impairment and reliance on staff observation rather than self-report, may contribute to apparent discordance between symptoms and prescribing. Further research should evaluate treatment appropriateness, deprescribing opportunities, and risk-benefit balance of chronic treatment in this population.
Liang et al. (Wed,) studied this question.