Abstract
BACKGROUND: Falls and fractures are major concerns after total hip arthroplasty (THA), although underrecognized. Their associations with preoperative physical function and balance confidence are not well understood. METHODS: This retrospective cohort study included 438 hips from 401 patients who underwent primary THA for osteoarthritis and completed a mailed survey more than 3 years postoperatively (mean follow-up, 6.9 years). Preoperative assessments included standardized measurements of hip range of motion, knee and hip strength, and gait speed. The survey assessed the number of falls in the past year, history of postoperative fractures, and the Activities-specific Balance Confidence Scale (ABCs), which measures confidence in balance. Receiver operating characteristic (ROC) analyses evaluated the ability of ABCs for identifying patients with falls and postoperative fractures. Unsupervised Gower-based clustering using fall history and ABCs was performed to classify patient subgroups. Multivariable Poisson, logistic regression, and linear models identified risk factors for falls, fractures, and low ABCs. RESULTS: Falls were reported in 29.5% of hips, fractures in 16.4%-of which 0.7% were periprosthetic-and mean ABCs was 83.4%. ABCs showed a significant negative correlation with fall frequency. ROC analysis identified a cutoff of 90.3% for falls (area under the curve [AUC] = 0.703, P < 0.001), but ABCs showed unsatisfactory discrimination for fractures (AUC = 0.599, P = 0.071). Cluster analysis revealed that patients with a fall history despite high postoperative ABCs had the highest fracture prevalence (29%). Greater preoperative knee extensor strength was predictive of fewer falls and higher ABCs, while faster preoperative gait speed and absence of recent falls were predictive of lower fracture risk. CONCLUSIONS: ABCs was useful for identifying fall risk but insufficient for predicting postoperative fractures. Cluster analysis revealed that patients with high postoperative balance confidence despite fall history were at the greatest risk of postoperative fractures. Preoperative knee extensor strength and gait speed were important predictors of postoperative falls and fractures. Evaluating physical function and promoting fall-prevention awareness-particularly among confident yet at-risk individuals-may enhance postoperative safety. LEVEL OF EVIDENCE: Level III, retrospective cohort study. See Instructions for Authors for a complete description of levels of evidence. CLINICAL RELEVANCE: Falls and fractures after THA remain underrecognized but clinically important complications. Identifying predictive preoperative factors such as knee extensor strength and gait speed, along with discrepancies between postoperative balance confidence and fall history, provides actionable insights for risk stratification and fall-prevention strategies in clinical practice.