(B-214) Assessing the Incidence of Major Adverse Cardiovascular Events in Medicare Data during the COVID-19 Pandemic

Background: COVID-19 significantly impacted the measurement of various conditions in claim data. Major adverse cardiovascular events (MACE) are a critical outcome, and fluctuations in its incidence during the pandemic may reflect cardiovascular care disruptions, population shifts, or changes in measurement validity. Understanding these variations informs claims-based real-world cardiovascular studies.

Objectives: To compare monthly incidence rates of 3-, 4-, and 5-point MACE in 2020 vs 2019

Methods: Using 15% national Medicare claims, we established 24 monthly cohorts in 2019 and 2020. Eligible patients had 12 months of continuous enrollment in Medicare Part A, B, and D before and including the study month. We assessed monthly incidence of 3-P MACE (stroke, myocardial infarction, cardiovascular death), 4-P MACE (+ unstable angina), 5-P MACE (+ heart failure), and each individual component. The denominator was all eligible beneficiaries per month. We calculated crude and adjusted incidence rate ratios (cIRR, aIRR) with 95% CI comparing 2020 to 2019, adjusting for demographics, healthcare utilization, comorbidities, medications, and frailty.

Results: IRRs were significantly lower from March to May 2020 across all MACE definitions, with the largest reductions in April (3-P MACE, cIRR=0.68 (0.64-0.72) aIRR=0.77 (0.73-0.82); 4-P MACE cIRR=0.66 (0.63-0.69) aIRR=0.75 (0.71-0.79); 5-P MACE cIRR=0.61 (058-0.63) aIRR=0.68 (0.65-0.72)). These reductions attenuated through later 2020, with slightly lower IRRs in August (3-P MACE cIRR=0.89 (0.85-0.94) aIRR=0.89 (0.84-0.94); 4-P MACE cIRR=0.89 (0.85-0.93) cIRR=0.88 (0.84-0.93); 5-P MACE cIRR=0.86 (0.83-0.90) aIRR =0.85 (0.82-0.89)) and November (3-P MACE cIRR=0.94 (0.89-0.98) aIRR=0.91 (0.87-0.96); 4-P MACE cIRR=0.91 (0.87-0.96) aIRR=0.89 (0.84-0.93); 5-P MACE cIRR=0.88 (0.85-0.92) aIRR=0.85 (0.82-0.89)). When comparing individual MACE components, IRRs for heart failure were consistently furthest from the null, whereas stroke’ IRRs were closest. For example, the IRRs for heart failure were cIRR=0.48 (0.44-0.52) aIRR=0.53 (0.48-0.58) and cIRR=0.75 (0.70-0.80) aIRR=0.87 (0.81-0.93) for stroke in April, comparing 2020 vs 2019.

Conclusions: We found a significant reduction in the incidence of MACE across all definitions during the initial COVID-19 surge in spring 2020, which tapered in later months. We observed significant differences across individual MACE components, potentially impacted by variation in acuteness/emergency, population shifts and healthcare-seeking behaviors. Further research is needed to understand factors contributing to these differences. This work highlights the need for careful design and analysis to account for the impact of COVID-19 on cardiovascular outcomes.