Clinically important improvement thresholds for Harris Hip Score and its ability to predict revision risk after primary total hip arthroplasty

Abstract

Background: Some aspects of validity are missing for the Harris Hip Score (HHS). Our objective was to examine the clinically meaningful change thresholds, responsiveness and the predictive ability of the HHS questionnaire.

Methods: We included a cohort of patients who underwent primary total hip arthroplasty (THA) and responded to the HHS preoperatively and at 2- or 5-year post-THA (change score) to examine the clinically meaningful change thresholds (Minimal clinically important improvement, MCII; and moderate improvement), responsiveness (effect size (ES) and standardized response mean (SRM)) based on pre- to post-operative change and the predictive ability of change score or absolute postoperative score at 2- and 5-years post-THA for future revision.

Results: Two thousand six hundred sixty-seven patients with a mean age of 64 years completed baseline HHS; 1036 completed both baseline and 2-year HHS and 669 both baseline and 5-year HHS. MCII and moderate improvement thresholds ranged 15.9-18 points and 39.6-40.1 points, respectively. ES was 3.12 and 3.02 at 2- and 5-years; respective SRM was 2.73 and 2.52. There were 3195 hips with HHS scores at 2-years and 2699 hips with HHS scores at 5-years (regardless of the completion of baseline HHS; absolute postoperative scores). Compared to patients with absolute HHS scores of 81-100 (score range, 0-100), patients with scores <55 at 2- and 5-years had higher hazards (95 % confidence interval) of subsequent revision, 4.34 (2.14, 7.95; p < 0.001) and 3.08 (1.45, 5.84; p = 0.002), respectively. Compared to HHS score improvement of >50 points from preoperative to 2-years post-THA, lack of improvement/worsening or 1-20 point improvement were associated with increased hazards of revision, 18.10 (1.41, 234.83; p = 0.02); and 6.21 (0.81, 60.73; p = 0.10), respectively.

Conclusions: HHS is a valid measure of THA outcomes and is responsive to change. Both absolute HHS postoperative scores and HHS score change postoperatively are predictive of revision risk post-primary THA. We defined MCID and moderate improvement thresholds for HHS in this study.

Keywords: Clinically important improvement; Discriminant ability; Harris Hip Score; MCID; MCII; Minimal clinically important difference; Minimal clinically important improvement; Predictability; Responsiveness; Total Hip Arthroplasty.

Figures

Fig. 1

Patient selection for cohorts requiring only 2/5-year HHS scores and those requiring both preoperative and 2/5-year scores

Fig. 2

Implant survival is shown for different categories of absolute (panels a and b) or change HHS scores (panels c and d), with follow-up up to 10 years post-THA. Follow-up starts at ≥ 731 days post-surgery for 2-year and ≥ 1826 days for 5-year (Day 0 for all revision analyses), i.e., only after the administration of the 2- and 5-year HHS survey