Comparative Performance of the Risk Analysis Index and Modified Frailty Index-5 in Predicting Hip Fracture Outcomes in Steroid Users

Purpose: Frailty is a critical determinant of surgical outcomes in older adults, yet there is limited evidence directly comparing frailty assessment tools in high-risk orthopedic populations. Chronic steroid use further compounds postoperative risk but is often underrepresented in surgical risk models. This study aimed to compare the predictive performance of the Risk Analysis Index (RAI) and Modified Frailty Index-5 (mFI-5) in forecasting short-term postoperative outcomes following hip fracture surgery among adults with active steroid use. Methods: We conducted a retrospective cohort study using ACS-NSQIP data from 2015–2021, identifying 3,833 adult patients with documented steroid use who underwent hip fracture surgery. Frailty was assessed using mFI-5 and RAI scores. The primary outcome was 30-day mortality. Secondary outcomes included major and minor complications, unplanned readmission, unplanned reoperation, extended length of stay, and non-home discharge. Predictive performance was evaluated using multivariable logistic regression, AUROC analysis, and bootstrap validation. Results: Among 3,833 patients, 35.5% were classified as frail by mFI-5 (score ≥2) and 27.3% by RAI (score >30). Mortality occurred in 5.7%, and 77.0% experienced non-home discharge. RAI demonstrated superior discrimination for mortality (AUROC 0.702 vs. 0.640, p=0.003; OR 1.12 per point increase, 95% CI: 1.10–1.14) and non-home discharge (AUROC 0.673 vs. 0.557, p<0.001). mFI-5 performed slightly better for unplanned reoperation (AUROC 0.562 vs. 0.507, p=0.010). The two indices showed comparable performance for complications and readmission. Conclusion: RAI offers superior predictive accuracy for mortality and discharge outcomes in steroid-using hip fracture patients, supporting its use for outcome-specific risk stratification.