Demographic and Clinical Characteristics of Patients Who Delayed, Skipped, or Continued Care During the First Year of the COVID-19 Pandemic and the Study Design Implications

OBJECTIVES: Early in the COVID-19 pandemic there were significant disruptions in healthcare delivery in the United States. This study describes three populations of healthcare users (continuers, delayers, and skippers) and explores the implications on the design of outcomes research studies using data from 2020.

METHODS: Using the Veradigm Network EHR linked to claims data, we identified individuals with continuous claims enrollment over three time periods (Y2019 [03/01/2019–02/29/2020], Y2020 [03/01/2020–02/28/2021), and Y2021 [03/01/2021–02/28/2022]) and EHR activity during Y2019 and Y2021. These individuals were sorted into three cohorts based on their healthcare utilization in Y2020: skippers (no utilization in Y2020), delayers (delayed care by >30 days compared to Y2019), and continuers (did not delay care). We report the demographic characteristics, comorbidity burden, and healthcare utilization of these three cohorts.

RESULTS: Of the 3,322,835 individuals included in the study, 17.3% were skippers, 29.9% were delayers, and 52.8% were continuers. Skippers tended to be younger than delayers or continuers (mean [SD]: 35.5 [22.5] vs 38.1 [24.0] or 42.2 [23.3] years), non-White (42.3% of those with known Race vs 39.8%, or 38.8%) and have a lower Charlson Comorbidity Index (CCI) (mean [SD]: 0.55 [1.20] vs 0.62 [1.28] or 0.87 [1.47]). Skippers also tended to have lower baseline healthcare utilization as measured by visit count in Y2019. These differences persisted into the follow-up period, which was the 365 days following the first Y2020 visit for delayers and continuers and Y2021 for skippers. Those who delayed care had the largest mean increase in CCI between baseline and follow-up (Δ0.09 [0.98]).

CONCLUSIONS: The demographic and clinical profiles were different between patients who delayed, skipped, or continued care during the first year of the COVID-19 pandemic. These differences may be important to consider when designing studies that include a control population who indexed during this period.