Determinants of 90-day case fatality among older patients admitted to intensive care units: A retrospective cohort study

Abstract

BackgroundA recent systematic review identified highly variable case-fatality rates among studies of older patients admitted to intensive care units (ICUs). However, structural and process determinants including patient resident status, tertiary ICU status, and treatment limitations were unavailable. ObjectiveThe objective of this study was to evaluate the role of determinants such as resident status, tertiary ICU, and treatment limitations on 90-day case fatality among older ICU patients. MethodsA retrospective cohort of all Queensland residents aged 75 years and older admitted to four ICUs within the Metro North Hospital and Health Service was included. The impact of Metro North Hospital and Health Service resident status, tertiary ICU, treatment limitations, and other known determinants on 90-day all-cause case fatality (case-fatality) was assessed. ResultsOf the 2144 eligible first admissions included, 1365 were residents, and 893 were nonelective admissions. The case-fatality rates were higher in residents (21% vs 12%, p < 0.001), nonelective admissions (32% vs 7%, p < 0.001), and non–tertiary ICU admissions (27% vs 16%, p < 0.001). The case fatality increased progressively with age, being highest (29.6%) in the >90 years age-group. Multivariable mixedeffects logistic regression modelling demonstrated that presence of treatment limitations was strongly associated with case fatality, but neither resident status nor the tertiary ICU was associated. ConclusionThe presence of treatment limitations should be considered when evaluating variations in case fatality among cohorts of older ICU patients, in addition to variables with well-established association with case fatality such as comorbidities and illness severity.