Frailty and decisional regret after elective noncardiac surgery: a multicentre prospective cohort study

Background: Frailty is associated with morbidity and mortality after surgery. The association of frailty with decisional regret is poorly defined. Our objective was to estimate the association of preoperative frailty with decisional regret status in the year after surgery. Methods: We conducted a secondary analysis of a prospective, multicentre cohort study of patients aged >= 65 years who underwent elective noncardiac surgery. Decisional regret about having undergone surgery was ascertained at 30, 90, and 365 (primary time point) days after surgery using a 3-point ordinal scale. Bayesian ordinal logistic regression was used to estimate the association of frailty with decisional regret, adjusted for surgery type, age, sex, and mental health conditions. Subgroup and sensitivity analyses were conducted. Results: We identified 669 patients; 293 (43.8%) lived with frailty. At 365 days after surgery, the unadjusted odds ratio (OR) associating frailty with greater decisional regret was 2.21 (95% credible interval [CrI] 0.98-5.09; P(OR>1)=0.97), > 1) = 0.97), which was attenuated after confounder adjustment (adjusted OR 1.68, 95% CrI 0.84-3.36; P(OR>1)=0.93). > 1) = 0.93). Similar results were estimated at 30 and 90 days. Additional adjustment for baseline comorbidities and disability score substantially altered the OR at 365 days (0.89, 95% CrI 0.37-2.12; P(OR>1)=0.39). > 1) = 0.39). There was a high probability that surgery type was an effect modifier (non- orthopaedic: OR 1.90, 95% CrI 1.00-3.59; P(OR>1)=0.98); > 1) = 0.98); orthopaedic: OR 0.87, 95% CrI 0.41-1.91; P(OR>1)=0.36). > 1) = 0.36). Conclusions: Among older surgical patients, there appears to be a complex association with frailty and decisional regret, with substantial heterogeneity based on assumed causal pathways and surgery type. Future studies are required to untangle the complex interplay between these factors.