Predictive Model and Web-based Calculator for the Oxford Hip Score After Total Hip Replacement

Abstract

Objective: Although previous studies have developed classical models predicting outcomes after hip replacement, no formal machine-learning based calculators have been designed to predict Oxford Hip Score based on a national sample. The aim of our study was to develop a series of machine-learning models and a web-based calculator to predict Oxford Hip Scores after total hip replacement. Methods:We made use of the National Health Service Patient Reported Outcome Measures and Hospital Episode Statistics (NHS PROMS/HES) database, evaluating pre and post-operative data from patients aged over 50 years old undergoing total hip replacement from 2010 to 2015. Predictors of Oxford Hip Score were assessed using a combination of machine-learning and tree regression models.Results: A total of 170,283 patients participated in the study. Most patients were female (60.7%), aged between 70 and 79 years, with a baseline Oxford Hip Score lower than 41. Across all machine learning models, the most significant predictors of Oxford Hip Scores were pre-operative EQ-5D index and self-perceived disability, problems while shopping, circulation diseases, and pre-operative problems while climbing stairs. The best performing models were Gradient Boosting Machines, Boosted Generalized Linear Model, and Multivariate Adaptive Regression Splines with R-Squared values of, respectively, 0.18, 0.18, and 0.18. A Web-based calculator was developed (https:// companionsite.sporedata.com/app/predicthip/). Conclusion:Highly accurate models were developed to predict the Oxford Hip Scores, which can be used in both clinical decision-making and healthcare the management of healthcare resources.

Total Hip Arthroplasty: Minimal Clinically Important Difference and Patient Acceptable Symptom State for the Forgotten Joint Score 12

“…Therefore, WOMAC and OHS scores were adopted as external anchors due to their correlation with FJS-12 being higher than 0.3 and statistically significant at the 0.05 level. Patients with WOMAC score ≥90 points, and patients with OHS score ≥40 points at 6 months follow-up were defined as responders [ 26 , 27 ]. The MCIDs calculated with those two different anchors are very close; therefore, the consistency of results across different anchors was confirmed.…”

“…High values of WOMAC and OHS scores at 6 months follow-up were used as the anchor. The cut-off value for WOMAC was 90 points, obtained from the estimate of the score at 6 months post-operative in the literature [ 26 ]; while 40 was set as the cut-off for OHS since the score from 40 to 48 indicate satisfactory joint function [ 27 ]. Therefore, patients with WOMAC values greater than 90 and OHS values greater than 40 at 6 months post-operative were considered responders.…”