Association Between a Bundled Payment Program for Lower Extremity Joint Replacement and Patient Outcomes Among Medicare Advantage Beneficiaries
Amanda Sutherland,Emily Boudreau,Andy Bowe,Qian Huang,Joshua M Liao,Matthew Flagg,Deborah Cousins,Dana Drzayich Antol,William H Shrank,Brian W Powers,Amol S Navathe
DOI: https://doi.org/10.1001/jamahealthforum.2023.1495
2023-06-02
JAMA Health Forum
Abstract:Importance: Much of the evidence for bundled payments has been drawn from models in the traditional Medicare program. Although private insurers are increasingly offering bundled payment programs, it is not known whether they are associated with changes in episode spending and quality. Objective: To evaluate whether a voluntary bundled payment program offered by a national Medicare Advantage insurer was associated with changes in episode spending or quality of care for beneficiaries receiving lower extremity joint replacement (LEJR) surgery. Design, setting, and participants: Cross-sectional study of 23 034 LEJR surgical episodes that emulated a stepped-wedge design by using the time-varying, geographically staggered rollout of the bundled payment program from January 1, 2012, to September 30, 2019. Episode-level multivariable regression models were estimated within practice to compare changes before and after program participation, using episodes at physician practices that had not yet begun participating in the program during a given time period (but would go on to do so) as the control. Data analyses were performed from July 1, 2021, to June 30, 2022. Exposures: Physician practice participation in the bundled payment program. Main outcomes and measures: The primary outcome was episode spending (plan and beneficiary). Secondary outcomes included postacute care use (skilled nursing facility and home health care), surgical setting (inpatient vs outpatient), and quality (90-day complications [including deep vein thrombosis, wound infection, fracture, or dislocation] and readmissions). Results: The final analytic sample included 23 034 LEJR episodes (6355 bundled episodes and 16 679 control episodes) from 109 physician practices participating in the program. Of the beneficiaries, 7730 were male and 15 304 were female, 3057 were Black, 19 351 were White, 447 were of other race or ethnicity (assessed according to the Centers for Medicare & Medicaid Services beneficiary race and ethnicity code, which reflects data reported to the Social Security Administration), and 179 were of unknown race and ethnicity. The mean (SD) age was 70.9 (7.2) years. Participation in the bundled payment program was associated with a 2.7% (95% CI, 1.3%-4.1%) decrease in spending per episode (mean episodic spending, $21 964 [95% CI, $21 636-$22 296] vs $22 562 [95% CI, $22 346-$22 779]), as well as reductions in skilled nursing facility use after discharge (21.3% for bundled episodes vs 25.0% for control episodes; odds ratio [OR], 0.81 [95% CI, 0.67-0.98]) and increased use of the outpatient surgical setting (14.1% for bundled episodes vs 8.4% for control episodes; OR, 1.79 [95% CI, 1.53-2.09]). The program was not associated with changes in quality outcomes, including 90-day complications (8.8% for bundled episodes vs 8.6% for control episodes; OR, 1.02 [95% CI, 0.86-1.20]) and readmissions (4.3% for bundled episodes vs 4.6% for control episodes; OR, 0.92 [95% CI, 0.75-1.13]). Conclusions and relevance: In this study of an LEJR bundled payment program offered by a national Medicare Advantage insurer, findings suggest that physician practice participation in the program was associated with a decrease in episode spending without changes in quality. Bundled payments offered by private insurers, including Medicare Advantage plans, are an alternate payment option to fee for service that may reduce spending for LEJR episodes while maintaining quality of care.