Highlight
Agitation is prevalent among dementia patients in nursing homes, often managed with pharmacologic agents. This comparative study found that initiation of selective serotonin reuptake inhibitors (SSRIs) resulted in modest but statistically significant improvements in agitation symptoms compared to second-generation antipsychotics (SGAs). The benefits of SSRIs were most notable in residents with mild-to-moderate agitation and cognitive impairment. These findings provide important real-world evidence favoring SSRIs as safer pharmacologic alternatives to SGAs, which carry FDA black box warnings due to associated safety risks.
Background
Agitation occurs in up to 80% of individuals with dementia, particularly affecting nursing home residents. It manifests as restlessness, aggression, or disruptive behaviors that significantly reduce quality of life and challenge caregiving environments. Pharmacologic management often involves second-generation antipsychotics despite recognized serious safety concerns, including increased risks for cerebrovascular events and mortality in elderly dementia patients. Consequently, there is an unmet need for safer, effective therapeutic options. SSRIs have been proposed due to their favorable safety profile and plausible mechanisms in mood and behavior regulation, yet comparative real-world data remain sparse.
Design
This study employed a national, active comparator, new user cohort design utilizing US Medicare claims data linked with Minimum Data Sets (MDS) 3.0 from 2011 to 2018. The population included long-stay nursing home residents aged 65 years or older with diagnosed dementia and documented agitation who newly initiated either SSRIs or SGAs. The primary outcomes measured were changes in agitation severity using the Agitated and Reactive Behavior Scale (ARBS), a validated tool ranging from 0 to 12, with higher scores indicating worse agitation, assessed over 60 to 120 days post-initiation. Additionally, researchers evaluated the proportion achieving a minimal clinically important improvement, defined as at least a 1-point decline in ARBS score. Analytical methods included weighted linear models with overlap propensity scoring and inverse probability of censoring weights within generalized estimating equations to estimate adjusted treatment effects under both intention-to-treat and as-treated scenarios.
Key Findings
The study cohort comprised 19,508 new SSRI users and 21,541 new SGA users, with baseline characteristics well balanced after propensity weighting, minimizing confounding bias. In as-treated analyses, SSRI initiators experienced a greater mean reduction in ARBS scores compared with SGA initiators, with a difference of -0.14 points (95% confidence interval [CI]: -0.19 to -0.09), indicating modest but statistically significant improvement. Furthermore, SSRIs were associated with a 3.6% higher probability (95% CI: 2.7% to 4.6%) of achieving minimal clinically important improvement relative to SGAs. Subgroup analyses revealed that these benefits were most pronounced among residents with mild-to-moderate agitation and mild-to-moderate cognitive impairment, suggesting a differential treatment effect based on baseline severity. Adverse safety outcomes were not detailed in this study but remain a consideration given existing knowledge of SGA risks.
Expert Commentary
This large, well-conducted observational study contributes valuable comparative effectiveness data for two common pharmacologic classes used to manage agitation in dementia within nursing homes. The modest superiority of SSRIs versus SGAs aligns with clinical imperatives to mitigate the risks associated with antipsychotic use highlighted by FDA black box warnings. While effect sizes were small, they are clinically relevant in a vulnerable population where treatment options are limited and nonpharmacologic interventions may not suffice. Limitations include potential residual confounding inherent to observational designs, lack of detailed safety outcome reporting, and generalizability constrained to US nursing home residents aged 65 and older. Prior randomized controlled trials have yielded mixed results, but real-world evidence such as this enhances external validity.
Mechanistically, SSRIs’ modulation of serotonin pathways may alleviate behavioral dysregulation and agitation, while SGAs’ dopamine antagonism, though efficacious for psychosis, may impose greater adverse effects in this demographic. The findings underscore the need for individualized treatment decisions weighing symptom severity, cognition level, and safety profiles, along with prioritizing nonpharmacologic strategies when feasible.
Conclusion
This comprehensive national cohort study provides evidence supporting SSRIs as modestly more effective and potentially safer pharmacologic options than second-generation antipsychotics for agitation management among nursing home residents with dementia. Clinicians should consider SSRIs preferentially, especially in patients with milder agitation and cognitive impairment, while vigilantly monitoring for treatment response and adverse events. Further research is warranted to explore long-term outcomes, safety profiles, and integration with behavioral interventions to optimize dementia care. Given the significant burden of agitation and limitations of current therapies, these findings have important implications for clinical practice guidelines and policy makers aiming to improve quality of life in this vulnerable population.
Funding and ClinicalTrials.gov
The original study did not specify funding sources or clinical trial registration. It utilized administrative and clinical data from Medicare and the Minimum Data Sets (MDS), representing observational research.
References
- Liang S, Lapane KL, Ott BR, et al. Comparative Effectiveness of Antidepressants and Second-Generation Antipsychotics for Agitation Among Nursing Home Residents with Dementia. J Gen Intern Med. 2026 Jul 20. PMID: 42477265.
- Reus VI, Fochtmann LJ, Eyler AE, et al. The American Psychiatric Association Practice Guideline on the Use of Antipsychotics to Treat Agitation or Psychosis in Patients with Dementia. Am J Psychiatry. 2016;173(5):543-546.
- Ballard C, Waite J. The effectiveness of antipsychotic drugs in treating the behavioural and psychological symptoms of dementia. CNS Drugs. 2006;20(5):405-415.
- Kales HC, Gitlin LN, Lyketsos CG. Assessment and management of behavioral and psychological symptoms of dementia. BMJ. 2015;350:h369.
