key: cord-0796222-8ju469ik
authors: Jacob, Jesse T.; Baker, Julia M.; Fridkin, Scott K.; Lopman, Benjamin A.; Steinberg, James P.; Christenson, Robert H.; King, Brent; Leekha, Surbhi; O’Hara, Lyndsay M.; Rock, Peter; Schrank, Gregory M.; Hayden, Mary K.; Hota, Bala; Lin, Michael Y.; Stein, Brian D.; Caturegli, Patrizio; Milstone, Aaron M.; Rock, Clare; Voskertchian, Annie; Reddy, Sujan C.; Harris, Anthony D.
title: Risk Factors Associated With SARS-CoV-2 Seropositivity Among US Health Care Personnel
date: 2021-03-10
journal: JAMA Netw Open
DOI: 10.1001/jamanetworkopen.2021.1283
sha: 9fa2abb9beccf51f13abe30e5a257569c567a2bb
doc_id: 796222
cord_uid: 8ju469ik

IMPORTANCE: Risks for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection among health care personnel (HCP) are unclear. OBJECTIVE: To evaluate the risk factors associated with SARS-CoV-2 seropositivity among HCP with the a priori hypothesis that community exposure but not health care exposure was associated with seropositivity. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study was conducted among volunteer HCP at 4 large health care systems in 3 US states. Sites shared deidentified data sets, including previously collected serology results, questionnaire results on community and workplace exposures at the time of serology, and 3-digit residential zip code prefix of HCP. Site-specific responses were mapped to a common metadata set. Residential weekly coronavirus disease 2019 (COVID-19) cumulative incidence was calculated from state-based COVID-19 case and census data. EXPOSURES: Model variables included demographic (age, race, sex, ethnicity), community (known COVID-19 contact, COVID-19 cumulative incidence by 3-digit zip code prefix), and health care (workplace, job role, COVID-19 patient contact) factors. MAIN OUTCOME AND MEASURES: The main outcome was SARS-CoV-2 seropositivity. Risk factors for seropositivity were estimated using a mixed-effects logistic regression model with a random intercept to account for clustering by site. RESULTS: Among 24 749 HCP, most were younger than 50 years (17 233 [69.6%]), were women (19 361 [78.2%]), were White individuals (15 157 [61.2%]), and reported workplace contact with patients with COVID-19 (12 413 [50.2%]). Many HCP worked in the inpatient setting (8893 [35.9%]) and were nurses (7830 [31.6%]). Cumulative incidence of COVID-19 per 10 000 in the community up to 1 week prior to serology testing ranged from 8.2 to 275.6; 20 072 HCP (81.1%) reported no COVID-19 contact in the community. Seropositivity was 4.4% (95% CI, 4.1%-4.6%; 1080 HCP) overall. In multivariable analysis, community COVID-19 contact and community COVID-19 cumulative incidence were associated with seropositivity (community contact: adjusted odds ratio [aOR], 3.5; 95% CI, 2.9-4.1; community cumulative incidence: aOR, 1.8; 95% CI, 1.3-2.6). No assessed workplace factors were associated with seropositivity, including nurse job role (aOR, 1.1; 95% CI, 0.9-1.3), working in the emergency department (aOR, 1.0; 95% CI, 0.8-1.3), or workplace contact with patients with COVID-19 (aOR, 1.1; 95% CI, 0.9-1.3). CONCLUSIONS AND RELEVANCE: In this cross-sectional study of US HCP in 3 states, community exposures were associated with seropositivity to SARS-CoV-2, but workplace factors, including workplace role, environment, or contact with patients with known COVID-19, were not. These findings provide reassurance that current infection prevention practices in diverse health care settings are effective in preventing transmission of SARS-CoV-2 from patients to HCP.

For a given week in the figure, the mean represents the average COVID-19 cumulative incidence values for the residential zip code among the healthcare personnel who obtained serology testing that week. Similarly, the minimum and maximum COVID-19 cumulative incidence for the residential zip code were calculated among healthcare personnel who obtained serology testing that week. a Odds ratio for the relationship between the specified factor and SARS-CoV-2 seropositivity using mixed effects logistic regression controlling for all other factors in the table and adjusting for correlation within each healthcare system (via inclusion of a random intercept) b Other direct care provider = Dialysis technician, phlebotomist c Other provider = Laboratory technician, student, medical technologist, other categories unable to refine OR= odds ratio; CI = confidence interval; ref = reference group, NA = not applicable due to small sample size Chi-square p-value from a likelihood ratio test comparing model fit for the random effects and fixed effect model was <0.001 indicating that the random intercept improved model fit.

a Due to contemporary public health guidance and uncertain respirator availability, procedure masks were recommended 3/20/20 to 4/1/2020. b Available and suggested, but not required, starting 3/31/2020. c Due to contemporary public health guidance and uncertain respirator availability