Health Care Utilization and Equity in Medicare Advantage Compared With Traditional Medicare

Importance Medicare Advantage (MA) plans use health care utilization management and other approaches that may be associated with different utilization patterns among beneficiaries from different racial and ethnic groups compared with traditional Medicare (TM). Objective To measure the MA-TM utilization gap across racial and ethnic groups. Design, Setting, and Participants This retrospective cohort study included Medicare beneficiaries 65 years or older continuously enrolled in MA and a 20% random sample of TM beneficiaries during 2021. TM beneficiaries were matched to MA beneficiaries by geographic location, age, and sex separately for Black, Hispanic, and White beneficiaries. Data analysis was conducted between August 2025 and January 2026. Exposure MA vs TM enrollment. Main Outcome and Measures Utilization rates per 1000 beneficiaries for 3 discretionary procedures (hip, knee replacement, and back surgery), 3 less discretionary cardiac procedures (cardiac catheterizations, percutaneous coronary interventions, and coronary artery bypass grafting), and 2 overall measures (emergency department visits and hospitalizations). MA utilization was measured using Healthcare Effectiveness Data and Information Set data; TM utilization measures were created following the exact same specifications. Results The final matched samples included 4 366 556 MA beneficiaries (2 442 688 [56.0%] female; 1 254 260 [28.7%] aged 71-75 years; 2 84 774 [6.5%] Black, 226 798 [5.2%] Hispanic, and 3 854 984 [88.3%] White) and 4 366 556 TM beneficiaries (2 442 688 [56.0%] female; 1 254 260 [28.7%] aged 71-75 years; 284 774 [6.5%] Black, 226 798 [5.2%] Hispanic, and 3 854 984 [88.3%] White). The MA-TM gap was largest for White beneficiaries for discretionary procedures. For example, there were 2.51 fewer (95% CI, −2.52 to −2.49) knee replacements per 1000 beneficiaries for White beneficiaries in MA vs TM, 1.26 fewer (95% CI, −1.32 to −1.21) for Black compared with White beneficiaries ( P < .001), and 0.60 fewer (95% CI, −0.66 to −0.55) for Hispanic compared with White beneficiaries ( P < .001). Among Hispanic and White beneficiaries, rates of coronary artery bypass grafting were marginally higher in MA vs TM (1.84 [95% CI, 1.67 to 2.02] vs 1.41 [95% CI, 1.25 to 1.56] per 1000 beneficiaries [ P < .001]; 2.22 [95% CI, 2.17 to 2.26] vs 2.12 [95% CI, 2.05 to 2.20] per 1000 beneficiaries [ P = .004], respectively) but not significantly different among Black beneficiaries in MA vs TM (1.37 [95% CI, 1.23 to 1.52] vs 1.45 [95% CI, 0.69 to 2.20]; P = .44). Compared with White beneficiaries, the MA-TM gap was significantly different but of similar magnitude for Black (0.09 [95% CI, 0.08 to 0.10] vs −0.07 [95% CI, −0.15 to 0.00]; P < .001) and Hispanic (0.09 [95% CI, 0.08 to 0.10] vs 0.44 [95% CI, −0.40 to 0.47]; P = .007) beneficiaries. Hospitalizations showed similar MA-TM gaps across race and ethnicity. Emergency department visit rates were lower in the MA group, with the greatest difference among Black beneficiaries. Conclusions and Relevance In this cohort study of MA and TM beneficiaries, exposure to MA was associated with a narrowing of utilization gaps for discretionary procedures with similar impacts on utilization across racial and ethnic groups for less discretionary procedures and overall utilization. Observed reductions in disparities in utilization possibly reflect reductions in discretionary use of procedures among White beneficiaries rather than increased access for Black and Hispanic beneficiaries.