Abstract
IMPORTANCE: There is growing concern that Medicare Advantage (MA) plans are shifting the costs of care to the Veterans Health Administration (VHA) for veterans dually enrolled in both systems, particularly in high-veteran MA plans that disproportionately enroll veterans. However, empirical evidence evaluating the sources of payment for veterans' surgical care is lacking. OBJECTIVE: To evaluate differences in payment sources for surgical care between high-veteran MA plans and other MA plans. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study used 2021 US national MA and VHA data from veterans dually enrolled in MA and VHA care for inpatient surgical episodes at VHA facilities (VHA-paid direct care), non-VHA community hospitals paid by VHA (VHA-paid community care), and community hospitals paid by MA (MA-paid community care) among veterans dually enrolled in MA and VHA care. Data were analyzed from April 1, 2024, to November 30, 2024. EXPOSURE: Enrollment in high-veteran MA plans. MAIN OUTCOMES AND MEASURES: Likelihood of utilizing VHA-direct care, VHA-paid community care, and MA-paid community care. High-veteran MA plans were defined as plans with 20% or more veteran enrollees; others were categorized as other MA plans. Multinomial logistic regression was used to evaluate the association of veteran enrollment in high-veteran MA plans with the likelihood of surgical care paid by each payment source, adjusting for veteran and surgery characteristics, and state fixed effects. Stratified analyses were conducted based on surgical complexity and source of admission. RESULTS: A total of 54 754 inpatient surgical episodes were analyzed, including 53 036 male (96.9%); 3133 Hispanic (5.7%), 47344 non-Hispanic Black (13.4%), 2933 non-Hispanic White (78.4%), and 1354 other or unknown race and ethnicity (2.5%); 601 (1.1%) were younger than 55 years, 3301(6.0%) aged 55 to 64 years, 22 381 (40.9%) aged 65 to 74 years, and 28471 (52%) aged 75 or older. Among these episodes, 52.1% were through MA-paid community care, 18.8% through VHA-direct care, and 29.1% through VHA-paid community care. Veteran enrollees in high-veteran MA plans were significantly less likely to have MA-paid surgeries (adjusted difference, -25.7 percentage points; 95% CI, -26.7 to 24.6 percentage points) and more likely to have surgeries paid through VHA-direct care (adjusted difference, 11.0 percentage points; 95% CI, 10.0-12.0 percentage points) and VHA-paid community care (adjusted difference, 14.7 percentage points; 95% CI, 13.6-15.8 percentage points) compared with veterans in other MA plans. As surgical complexity increased, differences in the use of VHA-paid direct care narrowed between high-veteran MA and other MA plans. Payment source differences were also less pronounced for nonelective surgeries admitted through emergency departments. CONCLUSIONS AND RELEVANCE: The findings of this cross-sectional study suggest substantial cost shifting in veterans' surgical care from MA to VHA among high-veteran MA plans, underscoring the urgent need for policy reforms to improve the efficiency of veterans' care.