Minimal important change values on the Knee injury and Osteoarthritis Outome Score (KOOS) predict future Quality Adjusted Life Years (QALYS) in 3,187 patients following acl reconstruction

Abstract

Purpose: The Knee injury and Osteoarthritis Outcome Score (KOOS) is one of the most widely used self-reported outcome measures in populations with knee injury and osteoarthritis. A 10-point difference on the KOOS represents a minimal important change and a clinically meaningful difference. However, the impact of a 10-point difference on the KOOS on future health-related quality of life outcomes is unknown. The most important health-related quality of life-derived outcome for informing priority setting for resource allocation to healthcare interventions is Quality Adjusted Life Years (QALY), which are typically derived from longitudinal patient-reported outcomes and used in cost-utility (cost-effectiveness) analysis. This study aimed to investigate the extent to which KOOS scores predicted QALYs in a population following anterior cruciate ligament reconstruction (ACLR). Methods: Data from the Swedish Knee Ligament Registry, a nationwide registry comprising >90% of ACLRs performed in Sweden, were collected prospectively between 2005-2016. Main inclusion criteria consisted of primary ACLR and age >12 years. Patient-reported outcomes of KOOS and European Quality of Life-5 Dimensions (EQ5D) were administered preoperatively and at 1-, 2-, and 5-year follow-ups. The primary outcome for the current analysis was QALYs accrued between 1-5 years post-ACLR derived from EQ-5D assessments using an area under the curve approach. The KOOS4 outcome was used to combine the mean scores for four of the five KOOS subscales, covering pain, symptoms, sport/recreation difficulty, and knee-related quality of life at 1-year postoperatively (0=worst, 100=best). Generalised linear models (Gaussian family) examined associations between 1-year KOOS4 and i) EQ-5D scores at each follow-up and ii) QALYs over a 4-year time horizon (i.e. 1- to 5-years). Results: A total of 3,187 participants (mean age 29 years, 53% male) with complete 1-, 2-, and 5-year follow-up data were included, totalling 12,748 follow-up years. A 10-point greater KOOS4 score (i.e. clinically important difference) at 1-year post-ACLR, was associated with higher EQ-5D multi-attribute utility score of 0.07 (95%CI 0.0.07-0.08) at 1-year (r2=48%, p<0.001), 0.05 (0.05-0.06) at 2-years (r2=27%, p<0.001), and 0.04 (0.04-0.05) at 5-years (r2=18%, p<0.0001). A 10-point higher KOOS4 at 1-year resulted in an additional 0.21 (0.20-0.22) QALYs accrued over the following 4-years. Conclusions: These results from a large national sample suggest that a clinically important improvement in KOOS4 of 10-points within the first year after ACLR would be associated with an additional 0.21 QALYs over the following 4-years. These findings are important for informing future cost-utility (cost-effectiveness using cost-per-QALY) modelling of benefits associated with interventions that improve patient outcomes after ACLR and in other populations using the KOOS.