Defining Clinically Meaningful Thresholds for 12-Month Patient-Reported Outcomes in Total Hip Arthroplasty; Toward Improving Threshold Accuracy

Abstract

Background

Clinically meaningful thresholds for patient-reported outcomes are relevant to define and predict success of total hip arthroplasties (THAs). Defining and offering thresholds must consider preoperative symptom severity.

Methods

In this retrospective study of 40,213 primary total hip replacements registered in the Dutch Arthroplasty Register (2016-2018), receiver operating curve analysis was used to define minimal clinically important changes and patient-acceptable symptom states with the anchor transition in function. Subgroups were identified for which independent thresholds should be defined. Patient-reported outcome measures were symptoms (pain, Oxford Hip Score [OHS], Hip disability and Osteoarthritis Outcome Score) and quality of life (European Quality of Life 5 Dimensions 3L questionnaire).

Results

94.6% completed the anchor questions, of whom 80.1% reporting “much improved function” 1 year after surgery. Discriminative abilities of thresholds were not good (area under the curve < 0.8). Tercile-specific determination of thresholds improved discrimination and reliability (+10%). Minimal clinically important change values were higher for all outcomes (eg, change in OHS ≥ 24.5 vs ≥ 10.5) in patients with more severe preoperative symptoms. Patient-acceptable symptom state scores for European Quality of Life 5 Dimensions index (≥ 0.809) and OHS (≥ 40.5) showed good discrimination (area under the curve > 0.8). Patients with less symptoms required lower postoperative scores for reporting “much improved function” (postoperative OHS ≥ 38.5 vs 42.5). Tercile-specific thresholds did not improve accuracy of thresholds (Cohens kappa 42%).

Conclusions

The present study demonstrates that patients with more severe preoperative symptoms require greater change scores to achieve clinically relevant improvements than patients with less severe preoperative symptoms. This study suggests that current one-size-fits-all thresholds for success of THA should be replaced with more nuanced thresholds.

Level of evidence

Level III, Therapeutic Study.