Abstract
Introduction
As orthopaedics shifts towards value-based models of care, methods of evaluating the value of procedures such as a total hip arthroplasty (THA) will become crucial. Patient reported outcome measures (PROMs) can offer a meaningful way for patient-centered input to factor into the determination of value.
Despite their benefits, PROMs can be difficult to interpret as statistically significant, but not clinically relevant, differences between groups can be found. One method of correcting this issue is by using a minimal clinically important improvement (MCII), defined as the smallest improvement in a PROM determined to be important to patients.
This study aims to find demographic and surgical factors that are independently predictive of failing to achieve a MCII in pain and physical function at 1-year following THA.
Methods
A total of 976 patients were enrolled into a prospective international, multicenter study evaluating the long-term clinical performance of two acetabular shells and two polyethylene liners from a single manufacturer. All patients consented to be followed with plain radiographs and a set of PROMs preoperatively and at 1-year after surgery.
The outcomes considered in this study were achieving literature-defined MCIIs in pain and physical function at one year after THA. The MCII in pain was defined as achieving a 2-point decrease on the Numerical Rating Scale (NRS)-Pain or reporting a 1-year NRS-Pain value of 0, indicating no pain. The MCII in physical function was defined as achieving an 8.29-point increase on the SF-36 Physical Function subscore.
Univariate analyses were conducted to determine if there were statistically significant differences between patients who did achieve and did not achieve a MCII. Variables tested included: demographic and surgical factors, general and mental health state, and preoperative radiographic findings such as deformity and joint space width (JSW). Significant variables were entered into a multivariable binary logistic regression.
Receiver-operating characteristic (ROC) analysis was used to generate cutoff values for significant continuous variables. Youden's index was used to identify cutoff points that maximized both specificity and sensitivity.
Results
Of 976 enrolled patients, 630 (65%) patients had complete preoperative and 1-year PROMs and a valid preoperative radiograph. Of the final cohort, 59 (9%) patients did not achieve the MCII in pain and 208 (33%) patients did not achieve the MCII in physical function following THA.
Multivariable analysis determined that higher preoperative JSW (odds ratio (OR)=2.04; p<0.001), and lower preoperative SF-36 Mental Composite Score (MCS) (OR=0.96; p<0.001) were independently predictive of not achieving a MCII in pain. ROC analysis determined that cutoff points for preoperative JSW and MCS were 0.65mm and 47.4 points, respectively.
In a separate multivariable regression, we found higher preoperative JSW (OR=1.40; p=0.010) and higher preoperative HHS (OR=1.03; p<0.001) to be independently predictive of not achieving a MCII in physical function. Cutoff points for preoperative JSW and HHS were respectively 0.65mm and 50.5 points.
Conclusion
In the upcoming era of value-based orthopaedics, each treatment must produce a meaningful clinical improvement per dollar spent. To help achieve this goal, this study has identified that patients with less severe OA, poor mental health, and good preoperative hip function are at a higher risk for not achieving MCIIs in pain or function after THA. Surgeons can use this analysis to discuss the appropriateness of a THA with their patients, frame patient expectations, and broach the possibility of delaying surgery if the patient has risk factors for poor improvement.
