The COVID-19 Pandemic and Associated Inequities in Acute Myocardial Infarction Treatment and Outcomes

The COVID-19 pandemic placed substantial strain on the US health care system and disrupted timely care for urgent and emergent conditions, including AMI. Prior work documented excess non–COVID-19 mortality, including cardiovascular deaths, during the pandemic, with disproportionate impacts on racially and ethnically minoritized populations. Black and Hispanic individuals experienced markedly higher COVID-19 death rates and excess deaths than White individuals. Potential mechanisms include delayed presentation, reduced availability of usual treatments due to staff/bed shortages, and broader systemic stress on hospitals. This study asks: (1) Were AMI treatments and outcomes (revascularization, 30-day mortality, readmissions, nonhome discharge) different during weeks when hospitals had high COVID-19 burdens compared with low burdens? (2) If differences existed, were they more harmful for Black and Hispanic patients compared with White patients?

Prior studies reported higher mortality after emergency surgery in hospitals with high COVID-19 case volumes and increases in AMI mortality during the pandemic. However, evidence on whether these effects disproportionately affected racial and ethnic minority groups was limited or mixed. Work has also shown reduced hospital admissions for acute coronary syndromes during the pandemic and declines in urgent/emergent procedures, raising concerns about delayed care and access. Longstanding literature documents racial inequities in cardiovascular procedures among Medicare beneficiaries and persistent inequities in access to advanced heart failure therapies, potentially driven by structural and interpersonal racism and differences in access to high-quality care.

Design: Cross-sectional analysis using national Medicare data for beneficiaries aged 65 years and older hospitalized with AMI (NSTEMI or STEMI) from January 1, 2016, to December 31, 2020. December 2020 admissions were excluded to allow 30-day outcome assessment through November 2020. Data sources: Medicare claims provided demographics (age, sex, self-reported race/ethnicity using RTI race code), diagnoses/procedures (ICD-10), admission details, discharge disposition, death date, and hospital identifier. These were merged with CMS Impact Files for hospital characteristics (region, rurality, beds, mean daily census, disproportionate share percentage, resident-to-bed ratio). Study population: Initially identified 1,512,924 admissions among Hispanic, non-Hispanic Black, and non-Hispanic White beneficiaries with NSTEMI or STEMI. Exclusions: December 2020 (n=19,152), elective admissions (n=70,447), 30-day readmissions (n=102,022), transfers from hospice (n=76), and hospitals not in CMS Impact Files (n=1,954). Final sample: 1,319,273 admissions (1,022,439 NSTEMI; 296,834 STEMI) across 3,078 hospitals. Exposure: Weekly hospital COVID-19 burden defined as the proportion of Medicare inpatients testing positive for COVID-19, categorized as 0–2.0%, 2.1–10.0%, 10.1–20.0%, 20.1–30.0%, and >30.0% (COVID period months: March–November 2020). Outcomes: (1) Revascularization during index admission (PCI or CABG); (2) 30-day all-cause mortality; (3) 30-day all-cause readmission; (4) nonhome discharge (death or discharge to skilled nursing facility/nursing home, inpatient rehabilitation facility, long-term care hospital, or transfer). Statistical analysis: Interrupted time-series logistic regression models with a weekly linear time trend and monthly indicators for seasonality. Pandemic period deviations captured via monthly indicators (March–November 2020). Main exposure was hospital weekly COVID-19 burden; associations estimated overall and separately for NSTEMI and STEMI. Baseline model included age and race/ethnicity; expanded models added patient risk factors (age, sex, admission status and source, dual eligibility, MI location, heart failure subtypes, AMI complications, prior cardiac procedures, dialysis, COVID-19, Elixhauser comorbidities) and hospital characteristics (rurality, resident-to-bed ratio, DSH percentage, proportion of Black/Hispanic patients, AMI volume). Interactions between COVID-19 burden and race/ethnicity assessed differential changes by race/ethnicity. Cluster-robust variance estimators accounted for clustering within hospitals. Adjusted rates estimated via average marginal effects. Two-sided P<.05; no adjustment for multiple comparisons. Software: Stata/MP 17.0. Post hoc analyses: Negative binomial regression to estimate weekly AMI admissions by period (prepandemic: Jan 2016–Feb 2020 vs pandemic: Mar–Nov 2020) and AMI type (NSTEMI vs STEMI). Additional mortality models included interaction between weekly hospital COVID-19 burden (linear) and early (Mar–Jul 2020) vs later (Aug–Nov 2020) pandemic periods.

- Sample and setting: 1,319,273 AMI admissions (1,022,439 NSTEMI; 296,834 STEMI) at 3,078 hospitals. - NSTEMI revascularization: Odds decreased by 9% at hospital COVID-19 burden 20.1–30.0% (aOR 0.91; 95% CI 0.83–1.00; P=.049) and by 27% at >30.0% burden (aOR 0.73; 95% CI 0.64–0.83; P<.001) vs pre-pandemic baseline. - NSTEMI outcomes vs pre-pandemic: 30-day mortality increased with higher hospital COVID burden: aOR 1.10 (95% CI 1.01–1.21; P=.04) at 10.1–20.0%, aOR 1.21 (95% CI 1.05–1.38; P=.007) at 20.1–30.0%, and aOR 1.51 (95% CI 1.29–1.76; P<.001) at >30.0%. Readmissions were not significantly higher at >30% burden (aOR 1.13; 95% CI 1.00–1.28; P=.06). Nonhome discharge increased by 32% at >30% burden (aOR 1.32; 95% CI 1.15–1.52; P<.001). - STEMI outcomes vs pre-pandemic: Readmissions and nonhome discharge were not significantly higher during high COVID burden weeks. Mortality at >30% burden was borderline and not statistically significant (aOR 1.28; 95% CI 1.00–1.64; P=.05). - Baseline racial/ethnic disparities (overall, irrespective of COVID burden): Revascularization was lower for Black and Hispanic patients compared with White patients. NSTEMI: Black OR 0.56 (95% CI 0.54–0.59; P<.001), Hispanic OR 0.64 (95% CI 0.59–0.69; P<.001). STEMI: Black OR 0.55 (95% CI 0.53–0.58; P<.001), Hispanic OR 0.65 (95% CI 0.59–0.71; P<.001). Among NSTEMI, Black patients were more likely to be readmitted (OR 1.15; 95% CI 1.11–1.19) and have nonhome discharge (OR 1.13; 95% CI 1.10–1.16). Hispanic NSTEMI patients were more likely to die within 30 days (OR 1.07; 95% CI 1.03–1.11) and be readmitted (OR 1.13; 95% CI 1.07–1.19), and less likely to have nonhome discharge (OR 0.82; 95% CI 0.76–0.87). - Differential changes by race/ethnicity during high COVID burden: No significant differential changes in revascularization during high-burden weeks for Black (aOR 0.98; 95% CI 0.94–1.03; P=.48) or Hispanic (aOR 1.03; 95% CI 0.98–1.09; P=.25) NSTEMI patients. For NSTEMI mortality, Black patients experienced a modestly greater increase per 10% rise in hospital COVID burden (aOR 1.07; 95% CI 1.00–1.15; P=.04), corresponding to adjusted mortality increasing from 12.8% to 14.8% for Black patients versus 13.1% to 14.4% for White patients when burden rose from 30% to 40% (difference ≈0.7 percentage points). For STEMI, Black and Hispanic patients did not experience greater increases in mortality, readmissions, or nonhome discharge during high-burden weeks relative to White patients. - Post hoc analyses: Weekly AMI admissions decreased by 5.2% during the pandemic vs prepandemic (IRR 0.95; 95% CI 0.90–0.997; P=.04). The NSTEMI-to-STEMI ratio decreased by 12.1% (IRR 0.88; 95% CI 0.83–0.93; P<.001); overall STEMI-to-NSTEMI IRR 0.29 (95% CI 0.28–0.29; P<.001). No significant difference in 30-day mortality between early (Mar–Jul 2020) and later (Aug–Nov 2020) pandemic periods for STEMI (aOR 0.95; 95% CI 0.86–1.04; P=.26) or NSTEMI (aOR 1.05; 95% CI 0.98–1.13; P=.16).

Hospital strain from high COVID-19 inpatient burden was associated with reduced revascularization and worse outcomes among NSTEMI admissions compared with pre-pandemic patterns, aligning with broader evidence that the pandemic compromised urgent cardiovascular care. STEMI care appeared more resilient, potentially reflecting the less discretionary nature of emergent PCI. Despite no marked worsening in racial/ethnic disparities during high-burden weeks, substantial baseline inequities persisted: Black and Hispanic patients had markedly lower revascularization rates for both NSTEMI and STEMI and higher risks of adverse outcomes in several domains. The modest additional rise in NSTEMI mortality for Black patients with increasing hospital COVID burden suggests some unequal spillover effects, though absolute differences were small. These findings underscore the need to ensure continuity of evidence-based cardiovascular care during system stress and to address longstanding structural and interpersonal drivers of inequitable access to invasive treatments.

In over 1.3 million Medicare AMI hospitalizations, weeks with high hospital COVID-19 burden were associated with lower revascularization and higher 30-day mortality, readmission, and nonhome discharge, particularly for NSTEMI. Racial/ethnic inequities did not significantly widen during the pandemic, yet large preexisting gaps in revascularization persisted for Black and Hispanic patients. Policies and clinical strategies should bolster hospital capacity to deliver equitable, evidence-based AMI care during periods of strain and target the root causes of persistent procedural inequities. Future research should evaluate long-term cardiovascular sequelae of reduced revascularization and develop interventions to mitigate structural and interpersonal contributors to inequity.

- Primary presentation of baseline disparities emphasized unadjusted analyses; adjusting for baseline health and hospital differences could attenuate observed disparities, but may also adjust away prehospital effects of structural racism. - Study population limited to Medicare beneficiaries aged 65 years and older; findings may not generalize to younger patients, the uninsured, or all Medicare Advantage enrollees. - Hospital COVID-19 burden measured using Medicare patients only, not all adult inpatients. - Analyses conditional on hospital admission; do not capture disparities in AMI incidence or prehospital mortality. - Observational, nonrandomized design with potential unmeasured confounding; causal inferences are limited.