Long-Term Postoperative Recovery Trajectories Among Older Individuals Undergoing Major Operation

Older individuals undergoing major surgery may experience a prolonged recovery period or even permanent institutionalization. We sought to develop a postoperative recovery trajectory model and identify patients at risk of loss of independence after major surgery.

Patients undergoing major surgery were identified using Medicare Standard Analytic Files. Latent class group-based trajectory modeling was used to identify clusters of patients with distinct postoperative recovery patterns. Multivariable regression was performed to predict cluster membership.

Five distinct postoperative recovery trajectories were identified: routine (n = 83,603; 11.7%), slow (n = 403,715; 56.4%), protracted (n = 49,704; 6.9%), chronically dependent (n = 156,881; 21.9%), and loss of independence (n = 21,817; 3.1%). On multivariable analysis, preoperative factors associated with loss of independence included older age (relative risk ratio [RRR] 1.03, 95% CI 1.03 to 1.03), high Charlson Comorbidity Index (RRR 3.30, 95% CI 3.17 to 3.40), minority status (RRR 1.30, 95% CI 1.22 to 1.39), urgent index surgery (RRR 1.31, 95% CI 1.26 to 1.36), and frailty (RRR 3.90, 95% CI 3.66 to 4.16). Postoperative factors associated with loss of independence included major complications (RRR 1.78, 95% CI 1.72 to 1.84), ventilator support during the index admission (RRR 1.90, 95% CI 1.85 to 1.96), and ICU stay (RRR 1.09, 95% CI 1.05 to 1.13). The loss of independence risk model was made available online for broad clinical use ( https://khalil-pawlik-postoperativeindependence-calculator.streamlit.app/ ).

Patients undergoing major surgery demonstrated distinct patterns of postoperative recovery, with some experiencing a loss of independence. A risk prediction model was developed to help clinicians provide more informed guidance to patients and their families regarding future care needs.