Patient-specific meaningful improvement thresholds for EQ-5D index scores after total hip or knee replacement: a retrospective observational cohort study

Abstract Purpose: The aim is to calculate EQ-5D-5L index score thresholds for patients after total hip and knee replacement (THR and TKR), considering patient-specific characteristics to enable a realistic evaluation of surgery outcome. Methods: In this retrospective observational cohort study, average and patient-characteristics specific meaningful improvements of the EQ-5D-5L index score were calculated via anchor-based receiver operator characteristics curves. Estimations are based on patient-level data from 5,217 primary total hip and knee replacement from two German hospitals. Results: Post-surgery improvements in health-related quality of life, measured by the EQ-5D-5L index score, were dependent on patient characteristics. The choice of anchor, mobility improvement or pain reduction, determined the size of the unstratified 12-months MI thresholds. For THR, the MI was 0.264 for the anchor mobility improvement and 0.204 for the anchor pain reduction, 0.211 and 0.203 for TKR, respectively. Accuracy of calculated thresholds for both anchors and treatment groups increased for characteristics-specific thresholds. Conclusions: MI thresholds for post-surgery EQ-5D-5L index scores should consider patient characteristics, are dependent on the anchor and are specific to time windows. Level of Evidence: Level III, Prognostic study