No evidence for stratified exercise therapy being cost-effective compared to usual exercise therapy in patients with knee osteoarthritis: Economic evaluation alongside cluster randomized controlled trial.

Background: A stratified approach to exercise therapy may yield superior clinical and economic outcomes, given the large heterogeneity of individuals with knee osteoarthritis (OA).

Objective: To evaluate the cost-effectiveness during a 12-month follow-up of a model of stratified exercise therapy compared to usual exercise therapy in patients with knee OA, from a societal and healthcare perspective.

Methods: An economic evaluation was conducted alongside a cluster-randomized controlled trial in patients with knee OA (n = 335), comparing subgroup-specific exercise therapy for a 'high muscle strength subgroup', 'low muscle strength subgroup', and 'obesity subgroup' supplemented by a dietary intervention for the 'obesity subgroup' (experimental group), with usual ('non-stratified') exercise therapy (control group). Clinical outcomes included quality-adjusted life years - QALYs (EuroQol-5D-5 L), knee pain (Numerical Rating Scale) and physical functioning (Knee Injury and Osteoarthritis Outcome Score in daily living). Costs were measured by self-reported questionnaires at 3, 6, 9 and 12-month follow-up. Missing data were imputed using multiple imputation. Data were analyzed through linear regression. Bootstrapping techniques were applied to estimate statistical uncertainty.

Results: During 12-month follow-up, there were no significant between-group differences in clinical outcomes. The total societal costs of the experimental group were on average lower compared to the control group (mean [95% confidence interval]: € 405 [-1728, 918]), albeit with a high level of uncertainty. We found a negligible difference in QALYs between groups (mean [95% confidence interval]: 0.006 [-0.011, 0.023]). The probability of stratified exercise therapy being cost-effective compared to usual exercise therapy from the societal perspective was around 73%, regardless of the willingness-to-pay threshold. However, this probability decreased substantially to 50% (willingness-to-pay threshold of €20.000/QALY) when using the healthcare perspective. Similar results were found for knee pain and physical functioning.

Conclusions: We found no clear evidence that stratified exercise therapy is likely to be cost-effective compared to usual exercise therapy in patients with knee OA. However, results should be interpreted with caution as the study power was lower than intended, due to the Coronavirus disease (COVID-19) pandemic.