Rehospitalization and the Association of Postoperative Delirium With Cognitive Decline in Older Adults

Importance: Postoperative delirium is associated with long-term cognitive decline in older adults. This might be caused by the delirium itself or because delirium is more common in persons who are ill and frail, and these conditions are also associated with cognitive decline. Objective: To determine whether cognitive decline associated with postoperative delirium is mediated by illness and frailty, as measured by recurrent hospitalizations. Design, Setting, and Participants: This prospective cohort study included community-dwelling older adults (age ≥70 years), enrolled from June 2010 to August 2013 with 5 years of follow-up data in the ongoing Successful Aging after Elective Surgery longitudinal study. Data were analyzed from November 2022 to May 2026. Exposures: Incident delirium following major elective surgery, with and without rehospitalizations, combined and by type (rehospitalization alone, rehospitalization with intensive care unit stay, rehospitalization with postacute care stay). Main Outcomes and Measures: The main outcome was long-term cognitive decline, measured as change in General Cognitive Performance (GCP) score, a composite measure of 11 neuropsychological tests, between preoperative baseline and 10 repeated assessments over 5 years. Results: In the cohort of 560 older adults (mean [SD] age, 76.7 [5.2] years; 326 female [58%]), the mean (SD) GCP score at baseline was 57.6 (7.3). Each rehospitalization was associated with a decline of -0.19 (95% CI, -0.31 to -0.09) GCP units per year. Delirium was associated with more marked cognitive decline of -0.33 (95% CI, -0.67 to -0.06) GCP units per year. Rehospitalizations were more common among patients who developed delirium (adjusted incidence rate ratio, 1.42 [95% CI, 1.17 to 1.72]). However, adjustment for combined rehospitalizations and for each type of rehospitalization resulted in only a minimal percentage change that was not statistically significant (-6% to -9%) in the association of delirium with cognitive decline. Conclusions and Relevance: In this cohort study, contrary to expectations, rehospitalization did not mediate the association between delirium and long-term cognitive decline. Future work will be needed to elucidate the pathways by which delirium is associated with long-term cognitive decline.