Abstract
Objectives:
Anterior cruciate ligament reconstruction is a commonly performed orthopaedic procedure. PROMIS (Patient-Reported Outcome Measurement Information System) was developed by the National Institutes of Health in an effort to advance patient-reported outcome (PRO) instruments by developing question banks for major health domains. Our goal was to compare the responsiveness and construct validity of the PROMIS physical function (PF) computer adaptive test (CAT) with current PRO instruments utilized in patients who undergo anterior cruciate ligament reconstruction.
Methods:
A total of 174 patients ages 14-53 scheduled to undergo anterior cruciate ligament reconstruction were asked to complete PROMIS PF-CAT, Short Form-36 Health Survey (SF36-PF and -GH), Marx activity rating scale (Marx), Knee Injury and Osteoarthritis Score (KOOS-ADL, -Sport, -QOL), and the EuroQol five dimensions questionnaire (EQ5D) at their preoperative visit. These surveys were repeated at six weeks and six months after surgery. Correlations between PRO instruments was defined as excellent (>0.7), excellent-good (0.61-0.7), good (0.4-0.6), and poor (0.2-0.3) using Spearman Correlation Coefficients. The effect size (Cohen d) and standardized response mean (SRM) were used to describe the responsiveness of each PRO at the 6 week and 6 month follow-up visits and were defined as small (0.2), medium (0.5) and large (0.8). Ceiling and floor effects were defined as present if ≥15% of participants scored the highest or lowest score on a PRO, respectively. Subgroup analyses were performed comparing change in PRO scores at follow-up between participants with and without additional arthroscopic procedures (meniscal debridement and/or repair, microfracture, or OATS vs ACL reconstruction only) using linear mixed models.
Results:
There were excellent and excellent-good correlations between the PROMIS PF-CAT and physical function PROs including the SF36-PF (r=0.75-0.80, p<.01), KOOS-ADL (r=0.62-0.70, p<.01) and KOOS-Sport (r=0.32-0.69, p<0.01) at most time points, respectively. There was also excellent-good correlation with the EQ5D (r=0.60-0.71, p<.01) and good correlation with the KOOS-QOL (r=0.52-0.58, p<0.01). As expected, there was no (p>0.05) to poor correlation with Marx activity (r=0.24, p<0.01) and the SF36-GH (r=0.32-0.34, p<0.01). No ceiling or floor effects were noted for the PROMIS PF-CAT; there was a ceiling effect noted for KOOS-ADL at the 6 month visit (38.1%). Effect size estimates for physical function PROs increased from small to large changes from the 6 week to 6 month visits and were largest in the PROMIS PF-CAT (1.34) followed by KOOS-ADL (1.19) and SF36-PF (1.06). Patients answered on average 4 questions utilizing the PROMIS. Baseline subgroup analyses showed no significant differences in physical function PRO scores, and at follow-up, both groups showed significant improvement in physical function PRO scores that was not statistically different.
Conclusion:
Our results support the construct validity of the PROMIS PF-CAT in patients who undergo ACL reconstruction. Responsiveness to change was highest by 6 months postop and greater for the PROMIS PF-CAT than for other measures of physical function with no ceiling or floor effects and a low time-burden. Taken together, these findings suggest that the PROMIS PF-CAT is a beneficial alternative to measuring and tracking changes in physical function in adults undergoing ACL reconstruction.