Remote assessment of instrumental activities of daily living and cognitive impairment in older surgical patients

Background: We aimed to investigate the prevalence of instrumental activities of daily living (IADL) dependence in older surgical patients, its association with cognitive impairment, and risk factors associated with postoperative IADL dependence. Methods: In this multicenter prospective study, 307 older adults aged ≥ 65 years undergoing non-cardiac surgery were evaluated for IADL dependence pre- and postoperatively at 30, 90 and 180 days. Changes in IADL scores over time were analyzed using a linear mixed-effects model stratified by preoperative IADL status. The association between IADL and cognition was examined by stratification of cognitive status. Preoperative risk factors of IADL dependence at 30 and 180 days were explored using multivariable logistic regression models. Results: Preoperatively, 38% of patients were IADL dependent. Among preoperatively IADL dependent patients, we observed a decline in IADL at 30 days, with subsequent recovery to baseline by 90 days. Those who were IADL independent showed no change in their IADL scores. Regardless of baseline cognitive status, postoperative IADL trajectories were similar, with cognitively impaired and non-cognitively impaired patients exhibiting a temporary decline at 30 days, with recovery by 90 days. Orthopedic surgery and preoperative cognitive impairment were associated with higher adjusted odds of 30-day IADL dependence (aOR 2.81, 95% CI [1.33-6.39]; aOR 5.31, 95% CI [1.56-25.0], respectively), whereas older age and preoperative frailty were associated with long-term IADL dependence at 180 days (aOR 1.57, 95% CI [1.14, 2.18]; aOR 2.83, 95% CI [1.01-8.05], respectively). Conclusions: Older adults with preoperative IADL dependence experienced transient postoperative decline with recovery by 90 days, whereas independent patients maintained function. Preoperative IADL assessment may help identify patients at risk and guide postoperative care planning.