ABSTRACT
To identify patients who are hospitalized because of COVID-19 as opposed to those who were admitted for other indications, we compared the performance of different computable phenotype definitions for COVID-19 hospitalizations that use different types of data from the electronic health records (EHR), including structured EHR data elements, provider notes, or a combination of both data types. And conduct a retrospective data analysis utilizing chart review-based validation. Participants are 586 hospitalized individuals who tested positive for SARS-CoV-2 during January 2022. We used natural language processing to incorporate data from provider notes and LASSO regression and Random Forests to fit classification algorithms that incorporated structured EHR data elements, provider notes, or a combination of structured data and provider notes. Results: Based on a chart review, 38% of 586 patients were determined to be hospitalized for reasons other than COVID-19 despite having tested positive for SARS-CoV-2. A classification algorithm that used provider notes had significantly better discrimination than one that used structured EHR data elements (AUROC: 0.894 vs 0.841, p < 0.001), and performed similarly to a model that combined provider notes with structured data elements (AUROC: 0.894 vs 0.893). Assessments of hospital outcome metrics significantly differed based on whether the population included all hospitalized patients who tested positive for SARS-CoV-2 versus those who were determined to have been hospitalized due to COVID-19. This work demonstrates the utility of natural language processing approaches to derive information related to patient hospitalizations in cases where there may be multiple conditions that could serve as the primary indication for hospitalization.
Subject(s)
COVID-19ABSTRACT
Introduction. Early COVID-19 vaccine acceptance rates suggest that up to one-third of HCWs may be vaccine-hesitant. However, it is unclear whether hesitancy among HCWs has improved with time and if there are temporal changes whether these differ by healthcare worker role. Methods. In October 2020, a brief survey was sent to all participants in the Healthcare Worker Exposure Response and Outcomes (HERO) Registry with a yes/no question regarding vaccination under emergency use authorization (EUA): "If an FDA emergency use-approved vaccine to prevent coronavirus/COVID-19 was available right now at no cost, would you agree to be vaccinated?" The poll was repeated in December 2020, with the same question sent to all registry participants. Willingness was defined as a "Yes" response, and hesitancy was defined as a "No" response. Participants were stratified into clinical care roles. Baseline demographics of survey respondents at each timepoint were compared using appropriate univariate statistics (chi-squared and t-tests). Analyses were descriptive, with frequencies and percentages reported for each category. Results. Of 4882 HERO active registry participants during September 1-October 31, 2020, 2070 (42.4%) completed the October survey, and n=1541 (31.6%) completed the December survey. 70.2% and 67.7% who were in clinical care roles, respectively. In October, 54.2% of HCWs in clinical roles said they would take an EUA-approved vaccine, which increased to 76.2% in December. The largest gain in vaccine willingness was observed among physicians, 64.0% of whom said they would take a vaccine in October, compared with 90.5% in December. Nurses were the least likely to report that they would take a vaccine in both October (46.6%) and December (66.9%). We saw no statistically significant differences in age, race/ethnicity, gender, or medical role between time points. When restricting to the 998 participants who participated at both time points, 69% were vaccine-willing at both time points; 15% were hesitant at both time points, 13% who were hesitant in October were willing in December; and 2.9% who were willing in October were hesitant in December. Conclusions. In a set of cross-sectional surveys of vaccine acceptance among healthcare workers, willingness improved substantially over 2 calendar months during which the US had a presidential election and two vaccine manufacturers released top-line Phase 3 trial results. While improved willingness was observed in all role categories, nurses reported the most vaccine hesitancy at both time points.