2020. Medicare Reliance and Variation in Post-Myocardial Infarction Procedure Use
LeChauncy D Woodard, MD, MPH, Houston Center for Quality of Care and Utilization Studies, Houston VAMC and Baylor College of Medicine, T Urech, Houston Center for Quality of Care and Utilization Studies, Houston VAMC, M Thompson, Houston Center for Quality of Care and Utilization Studies, Houston VAMC, M Byrne, University of Pittsburgh, K Pietz, Houston Center for Quality of Care and Utilization Studies, Houston VAMC, LA Petersen, Houston Center for Quality of Care and Utilization Studies, Houston VAMC and Baylor College of Medicine

Objectives: Determine the effect of Medicare reliance on Network-level variation in post-acute myocardial infarction (AMI) procedure use.

Methods: We identified all VA users with AMI during fiscal year 2000 and used logistic regression to compare odds of post-AMI procedures within 90 days of admission adjusted for age, race, Diagnostic Cost Groups risk score (RS), and the interaction of age and RS. We added patient-level Medicare reliance scores to the model, defined as the proportion of total care received under Medicare financing. The reference was the Network with the median OR for the procedure.

Results: After adjustment, patients in Networks 7, 12, and 11 were more likely than those in the median Network 13 to receive coronary angiography (OR 1.79[1.35,2.39], 1.54[1.12,2.12], and 1.48[1.08,2.03], respectively), while patients in Networks 4, 15, 8, 21, and 2 were significantly less likely to receive coronary angiography (OR 0.62[0.45,0.85], 0.61[0.46,0.82], 0.60[0.45,0.78], 0.53[0.36,0.76], and 0.47[0.32,0.68], respectively). Patients in Networks 1, 6, 12, 18, and 7 were more likely than those in the median Network 22 to receive revascularization procedures (OR 2.13[1.51,3.00], 1.63[1.20,2.22], 1.59[1.12,2.26], 1.54[1.12,2.12], and 1.43[1.05,1.95], respectively), while those in Networks 8, 4, 14, 15, 5, and 2 were less likely to undergo revascularization (OR 0.72[0.53,0.98], 0.60[0.41,0.87], 0.56[0.34,0.92], 0.56[0.40,0.79], 0.53[0.33,0.84], and 0.37[0.23,0.59], respectively). Incorporating Medicare reliance into the model did not change these findings.

Conclusions: Variation in use of post-AMI procedures across VA Networks is not explained by Medicare reliance.

Impact: Clinical data are needed to assess whether procedure use variation is due to clinical indications.