Introduction:
Regional studies have reported a decline in cardiovascular (CV) hospitalizations and procedures with the onset of the coronavirus disease-2019 (COVID-19) pandemic. Factors may include patient reluctance to seek care and de-prioritization of approvals for CV admissions by hospitals. We wanted to assess these observations at a national level.
Hypothesis:
To examine national trends in CV hospitalizations for acute myocardial infarction (AMI), unstable angina (USA), and heart failure (HF), as well as left heart catheterizations (LHC), using US medical claims data.
Methods:
We interrogated IQVIA US Claims data, a verified source, from Jan 2019 to May 2020 (214 million patients; 76% private insurance claims, 19% Medicare claims, 5% Medicaid claims). Since confirmed COVID-19 cases in the US began rising in Mar 2020, this was used as reference point to identify cohorts for comparison. Trends in volumes of hospitalizations for key CV events (AMI, USA, and HF) and LHC were compared from Mar 1 to May 8, 2020 to the equivalent time period in 2019. We used a Bayesian hierarchical model to assess trends.
Results:
From Mar to May 2020, compared to 2019, there were significantly fewer hospitalizations for: key CV events (1,110,492 vs. 1,487,558; p=0.0016); AMI (277,615 vs. 412,235; p=0.0002); USA (1,007 vs. 1,688, p=0.1245); and, HF (831,870 vs. 1,073,635; p=0.0036). There were significantly fewer LHC (118,393 vs. 221,701; p=0.0002). As shown in the Figure, there was a significant decline in CV hospitalizations in 2020 compared to 2019.
Conclusions:
During the COVID-19 pandemic, CV hospitalizations have declined significantly in the US. We observed an ~25% drop in CV hospitalizations and an ~50% drop in LHC. To the best of our knowledge, this is the first national evaluation of trends in CV care during COVID-19 and validate concerns that acute CV care in the US has been delayed or deferred, potentially foreshadowing a surge of CV complications in the future.
Mortality Prediction of Patients with Cardiovascular Disease Using Medical Claims Data under Artificial Intelligence Architectures: Validation Study (Preprint)