Comparison of the frailty index and frailty phenotype and their associations with postoperative delirium incidence and severity

Background Recent studies have reported an association between presurgical frailty and postoperative delirium. However, it remains unclear whether the frailty‐delirium relationship differs by measurement tool (e.g., frailty index vs. frailty phenotype) and whether frailty is associated with delirium, independent of preoperative cognition. Methods We used the successful aging after elective surgery (SAGES) study, a prospective cohort of older adults age ≥70 undergoing major non‐cardiac surgery (N = 505). Preoperative measurement of the modified mini‐mental (3MS) test, frailty index and frailty phenotype were obtained. The confusion assessment method (CAM), supplemented by chart review, identified postoperative delirium. Delirium feature severity was measured by the sum of CAM‐severity (CAM‐S) scores. Generalized linear models were used to determine the relative risk of each frailty measure with delirium incidence and severity. Subsequent models adjusted for age, sex, surgery type, Charlson comorbidity index, and 3MS. Results On average, patients were 76.7 years old (standard deviation 5.22), 58.8% of women. For the frailty index, the incidence of delirium was 14% in robust, 17% in prefrail, and 31% in frail patients (p