Association between frailty, delirium, and mortality in older critically ill patients: a binational registry study

Frailty and delirium are prevalent among older adults admitted to the intensive care unit (ICU) and associated with adverse outcomes; however, their relationships have not been extensively explored. This study examined the association between frailty and mortality and length of hospital stay (LOS) in ICU patients, and whether the associations are mediated or modified by an episode of delirium.

Retrospective analysis of data from the Australian New Zealand Intensive Care Society Adult Patient Database. A total of 149,320 patients aged 65 years or older admitted to 203 participating ICUs between 1 January 2017 and 31 December 2020 who had data for frailty and delirium were included in the analysis.

A total of 41,719 (27.9%) older ICU patients were frail on admission, and 9,179 patients (6.1%) developed delirium during ICU admission. Frail patients had significantly higher odds of in-hospital mortality (OR: 2.15, 95% CI 2.05-2.25), episodes of delirium (OR: 1.86, 95% CI 1.77-1.95), and longer LOS (log-transformed mean difference (MD): 0.24, 95% CI 0.23-0.25). Acute delirium was associated with 32% increased odds of in-hospital mortality (OR: 1.32, 95% CI 1.23-1.43) and longer LOS (MD: 0.54, 95% CI 0.50-0.54). The odds ratios (95% CI) for in-hospital mortality were 1.37 (1.23-1.52), 2.14 (2.04-2.24) and 2.77 (2.51-3.05) for non-frail who developed delirium, frail without delirium, and frail and developed delirium during ICU admission, respectively. There was very small but statistically significant effect of frailty on in-hospital mortality (b for indirect effect: 0.00037, P < 0.001) and LOS (b for indirect effect: 0.019, P < 0.001) mediated through delirium.

Both frailty and delirium independently increase the risk of in-hospital mortality and LOS. Acute delirium is more common in frail patients; however, it does not mediate or modify a clinically meaningful amount of the association between frailty and in-hospital mortality and LOS.