Summary

Meaningful recalibration response shift was clearly identified in IKDC and KOOS scores one and two years after ACLR, indicating traditional change scores are not adequately detecting patient perceived change in outcomes and suggesting treatment effects may frequently be underestimated.

Abstract

Background

Patient reported outcome measures (PROMs) are used to understand recovery following anterior cruciate ligament reconstruction (ACLR). However, PROM scores are largely dependent on respondent perspectives and individual frame of reference. Individuals’ priorities, standards, and conceptualizations of key constructs are likely to change over time, particularly following a health-related event. This phenomenon is labelled and quantified as “response shift” and may limit our ability to accurately interpret true change in PROMs.

Purpose

To identify whether recalibration response shift effects are present in PROM scores following ACL reconstruction in a prospective cohort study.

Methods

Patients having primary ACLR were recruited to complete the Knee Injury and Osteoarthritis Outcome Score (KOOS) and International Knee Documentation Committee Subjective Knee Form (IKDC) pre-, one year, and two years post-operatively. At the post-operative timepoints patients were asked to complete a second KOOS and IKDC measure according to their recollection of their pre-operative status. In addition they completed the brief-COPE and life event questionnaires. Unadjusted change (traditionally measured change), adjusted change (accounting for patient perspective), and magnitude of response shift (difference in change scores) were calculated for all outcomes according to then-test methodology. Patients were grouped into one of three categories for each PROM: positive recalibration (change was underestimated), negative recalibration (change was overestimated), or no meaningful recalibration, based on the magnitude and direction of response shift reported.

Results

A total of 171 participants were included in the study (mean age 24 years, 56% male). All participants have reached one-year follow-up (92% PROM completion) and 53% of the cohort have reached two-year follow-up (94% PROM completion). Adjusted change was significantly greater than unadjusted change (p < 0.05) for all PROMs except KOOS Sport (1yr: p = 0.17, 2yr: p = 0.08), with a difference in change scores ranging from 3 to 15 points. At one- and two-years post-operative, 75% and 73% of participants’ scores indicated meaningful recalibration response shift in at least one outcome, meaning pre- to post-operative change in PROM scores was over- or under-estimated by at least 10 points using traditional measures. The KOOS Symptoms score had the greatest proportion of positive recalibration (12m: 62%, 24m: 60%) while the KOOS Sport score had the greatest proportion of negative recalibration (12m: 27%, 24m: 20%).

Conclusions

Meaningful recalibration response shift was clearly identified in IKDC and KOOS scores one and two years after ACLR. Positive recalibration was more common than negative recalibration, however both processes were evident. The presence of response shift suggests traditional change scores are not adequately detecting true change in participants’ perceived outcomes following ACLR and suggests treatment effects may frequently be underestimated.