Association between delirium superimposed on dementia and mortality in hospitalized older adults: A prospective cohort study

BACKGROUND Hospitalized older adults with preexisting dementia have increased risk of having delirium, but little is known regarding the effect of delirium superimposed on dementia (DSD) on the outcomes of these patients. Our aim was to investigate the association between DSD and hospital mortality and 12-mo mortality in hospitalized older adults. METHODS AND FINDINGS This was a prospective cohort study completed in the geriatric ward of a university hospital in São Paulo, Brazil. We included 1,409 hospitalizations of acutely ill patients aged 60 y and over from January 2009 to June 2015. Main variables and measures included dementia and dementia severity (Informant Questionnaire on Cognitive Decline in the Elderly, Clinical Dementia Rating) and delirium (Confusion Assessment Method). Primary outcomes were time to death in the hospital and time to death in 12 mo (for the discharged sample). Comprehensive geriatric assessment was performed at admission, and additional clinical data were documented upon death or discharge. Cases were categorized into four groups (no delirium or dementia, dementia alone, delirium alone, and DSD). The no delirium/dementia group was defined as the referent category for comparisons, and multivariate analyses were performed using Cox proportional hazards models adjusted for possible confounders (sociodemographic information, medical history and physical examination data, functional and nutritional status, polypharmacy, and laboratory covariates). Overall, 61% were women and 39% had dementia, with a mean age of 80 y. Dementia alone was observed in 13% of the cases, with delirium alone in 21% and DSD in 26% of the cases. In-hospital mortality was 8% for patients without delirium or dementia, 12% for patients with dementia alone, 29% for patients with delirium alone, and 32% for DSD patients (Pearson Chi-square = 112, p < 0.001). DSD and delirium alone were independently associated with in-hospital mortality, with respective hazard ratios (HRs) of 2.14 (95% CI = 1.33-3.45, p = 0.002) and 2.72 (95% CI = 1.77-4.18, p < 0.001). Dementia alone did not have a significant statistical association with in-hospital mortality (HR = 1.69, 95% CI = 0.72-2.30, p = 0.385). Finally, while 24% of the patients died after discharge, 12-mo mortality was not associated with dementia or delirium in any of the diagnostic groups (DSD: HR = 1.15, 95% CI = 0.79-1.68, p = 0.463; delirium alone: HR = 1.05, 95% CI = 0.71-1.54, p = 0.810; dementia alone: HR = 1.19, 95% CI = 0.79-1.78, p = 0.399). Limitations to this study include not exploring the effects of the duration and severity of delirium on the outcomes. CONCLUSIONS DSD and delirium alone were independently associated with a worse prognosis in hospitalized older adults. Health care professionals should recognize the importance of delirium as a predictor of hospital mortality regardless of the coexistence with dementia.