OBJECTIVES: Rapid recovery (RR), i.e., mobilization within 6h after a hip or knee replacement, was first introduced to free capacity and to save provider costs. Literature has confirmed cost savings for hospitals, and suggests that also patients benefit from RR compared with conventional care (CC). However, no study has so far investigated the payer’s perspective regarding the cost-utility and cost-effectiveness of RR.

METHODS: Patient-level observational data from nine German hospitals in 2020/21 were used. Through a causal forest estimating double-robust propensity scores of individual treatment effects, the average treatment effects (ATEs) on costs (payer's perspective) and utility (QALYs with utility values from the EQ-5D-5L) were calculated. Based on those, incremental net monetary benefits (INMBs) were obtained to investigate the cost-utility of RR. A second analysis will use joint-specific patient reported outcomes to investigate cost-effectiveness of RR. A set of sensitivity analyses will be implemented to confirm the robustness of the results.

RESULTS: For hip replacements, RR significantly reduced costs but did not significantly increase utility in terms of health-related quality of life. For knee replacements, no significant effect was found. Assuming a willingness to pay per QALY of EUR 50,000, the INMB for RR was EUR 1,458.58 (95%CI: -1,577.63; 2,887.24) for hip replacements and EUR 654.81 (95%CI: -781.06; 3,698.22) for knee replacements, indicating no cost-utility. Potential cost-effectiveness is expected to be found in the second part of the analysis.

CONCLUSIONS: An innovative approach to determine the cost-utility and cost-effectiveness of RR for hip and knee replacements from the payer’s perspective was implemented, by exploiting a machine learning methodology for the analysis of retrospective observational data. RR shows no cost-utility from the provider’s perspective, but it is likely to be cost-effective as joint functionality improvement is the main aim of RR for joint replacement.