Two-year impact of a bundled payment program for total hip or knee replacement in the Netherlands: a difference-in-differences analysis

Abstract Background In 2018, a Dutch health insurer introduced a bundled payment program for total hip or knee replacement (THKR). Participating hospitals shared in savings and losses per episode, covering all THKR-related hospital care, including complications within one year after surgery. Hospitals were also permitted to increase THKR volume if quality targets were met. Methods We conducted difference-in-differences analyses using the insurer’s detailed claims data from 2015 to 2019, encompassing the first two years of the bundled payment program. We compared THKR episodes from 20 voluntary participating hospitals ( N = 9,960) with those from 34 control hospitals ( N = 9,147), before and after the program’s implementation. The primary outcomes were spending per THKR episode and rates of postsurgical complications. Secondary outcomes were utilization of care activities and volume of THKR episodes. The analyses were adjusted for patient characteristics, hospital and quarter fixed effects, and standard errors were clustered at the hospital level. Results After implementation, quarterly spending per THKR episode grew by €268 less, on average, in participating hospitals than in controls ( p = 0.039), reflecting a 3% reduction. This reduction was primarily driven by a decrease in the DRG price of THKR surgeries (-€240, p = 0.039). We observed no significant effects on complication rates (OR = 1.01, p = 0.951) or care activity utilization per episode. However, the program was associated with a slight increase in the volume of THKR episodes (OR = 1.09, p = 0.013). Conclusions In the first two years of the bundled payment program, spending per THKR episode modestly decreased due to lower DRG prices for THKR surgeries, without significant changes in complication rates or care activity utilization. In addition, we found a modest increase in the volume of THKR episodes. Excluding post-acute care activities from the episode likely weakened incentives to reduce spending through improved care utilization. Policymakers should evaluate whether the scope of a bundled payment program provides sufficient incentives for improving care coordination and utilization before implementation.