Survival outcomes among hospitalized patients with dementia: a propensity score matching analysis

Hospitalized patients with dementia exhibit high mortality rates, underscoring the importance of investigating variables associated with reduced survival. This study aims to determine the incidence of dementia among hospitalized patients and survival rates at 1 and 3 years post-hospitalization.

A retrospective cohort study was conducted using administrative databases from the Ministry of Health of Colombia. One- and three-year survival rates, along with adjusted hazard ratios for survival accounting for comorbidities included in the Charlson Index, were assessed using a Cox proportional hazards model. This analysis was performed for patients with dementia versus a control group without dementia. Additionally, findings were compared with those from an inverse propensity score weighting model.

6.769 (1.04%) patients were diagnosed with dementia, and 5798 (85.65%) were over 65 years of age. The unadjusted HR, the HR adjusted using the proportional hazards Cox model, and the HR obtained through propensity score matching (PSM) were 10.32 (95% CI 9.82 to 10.84), 1.69 (95% CI 1.60 to 1.78), and 1.32 (95% CI 1.02 to 1.71), respectively. The 1-year adjusted mortality rates for patients with dementia and those without were 12.5% and 1.31%, respectively, while the corresponding 3-year adjusted mortality rates were 21.25% and 2.76%. Through PSM, we determined that the mean survival time for patients with dementia, in comparison to those without, was - 0.98 months (95% CI: -0.65 to -1.94; p < 0.001).

Dementia significantly reduces survival rates of hospitalized patients, regardless of other comorbidities. Specifically, our research revealed that dementia was associated with a decrease in 3-year survival by an average of 0.98 months.