Abstract

Introduction: Although depression-one of the most studied neuropsychiatric symptoms (NPS)-has been associated with poorer survival, the prognostic significance of nondepressive NPS has not been well established. We aimed to examine the association between nondepressive NPS and all-cause mortality in older adults with dementia.

Methods: This meta-analysis was registered in PROSPERO (CRD42024621462), conducted in accordance with the Cochrane Handbook, and reported following PRISMA 2020 guidelines. Studies published between January 1990 and December 2024 were identified from MEDLINE, Embase, PubMed, and PsycINFO. Eligible studies included individuals aged ≥65 years with dementia and NPS, examining all-cause mortality as the primary outcome. Ten nondepressive NPS (hallucinations, delusions, agitation, anxiety, apathy, disinhibition, irritability, aberrant motor behavior, sleep disturbances, and eating disorders) were analyzed individually and cumulatively.Adjusted hazard ratios (aHRs) were extracted from multivariable Cox regression models and meta-analysed in random effects models.

Results: We included 10 studies with 55,281 participants across multiple countries. Affective symptoms (excluding depression) were significantly associated with increased all-cause mortality (pooled aHR = 1.24, 95% CI: 1.09-1.41, p = 0.0007, I² = 100%, 10 studies), with apathy showing the strongest association (aHR = 1.47, 95% CI: 1.06-2.02, p = 0.007, I² = 99%). The cumulative burden of NPS was also associated with higher all-cause mortality (pooled aHR = 1.16, 95% CI: 1.08-1.25, p <0.0001, I² = 100%, 10 studies). In contrast, no significant association was found for the remaining individual NPS.

Conclusions: Both cumulative NPS burden and affective symptoms-especially apathy-are associated with higher mortality. These findings highlight the need for further research and standardized assessments in dementia care.