Systematic review and meta-analysis of 33 studies (over 1.2 million people) identifying factors associated with hospital vs non‑hospital death in Parkinson's disease—hospital death was more likely in men, married people, those under 85, with suggestive higher odds in non‑white groups and lower odds…

Useful for informing end‑of‑life care planning, service provision, and targeted palliative interventions but of limited direct value for Parkinson's therapeutic discovery because it lacks mechanistic, biomarker, or intervention development insights.

IntroductionParkinson's disease is associated with increased mortality and hospitalisations are common at the end of life. However, limited evidence exists regarding the place of death and its influencing factors in people with Parkinson's disease (PwPD).ObjectivesTo identify and analyse factors associated with place of death in PwPD.MethodsWe systematically searched three electronic databases (MEDLINE, EMBASE, PsycINFO) for studies reporting on the place of death of PwPD. No restrictions on time or language were applied. Where possible, meta-analyses were conducted using random-effects meta-regression adjusted for country-level long-term care and hospital bed availability. Results are presented as odds ratios (OR) for place of death with 95% confidence intervals. Sensitivity analyses were performed to explore heterogeneity.Results33 studies were analysed, including over 1,200,000 individuals across five continents and reporting on individual, illness-level, service-level, and environmental factors. Hospital death was more likely among men (OR = 1.34; 95% CI: 1.21-1.49), married individuals (OR = 1.16; 95% CI: 1.07-1.26), and those under 85 years (OR = 1.29; 95% CI: 1.20-1.39). Lower-quality evidence suggested a higher likelihood of hospital death among non-white individuals, while receipt of palliative care was associated with reduced odds.ConclusionsThis systematic review and meta-analysis identify key factors associated with hospital death in PwPD that can inform clinical decision-making and policy planning. Our findings may support the development of targeted screening interventions and help clinicians and policymakers allocate resources effectively. Further research is needed to address gaps in evidence across different care settings.