Prevalence of SARS-CoV-2 Antibodies in Health Care Personnel in the New York City Area

The greater New York City (NYC) area, including the 5 boroughs and surrounding counties, has a high incidence of coronavirus disease 2019 (COVID-19),¹ and health care personnel (HCP) working there have a high exposure risk. HCP have expressed concerns about access to testing so that infection spread to patients, other HCP, and their families can be minimized.² The Northwell Health System, the largest in New York State, sought to address this concern by offering voluntary antibody testing to all HCP. We investigated the prevalence of antibodies against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) among HCP and associations with demographics, primary work location and type, and suspicion of virus exposure.

All Northwell HCP (employees) were provided with personal protective equipment from March 7, 2020, onward. SARS-CoV-2 testing by reverse transcriptase–polymerase chain reaction (PCR) began March 7, 2020, and was available for any HCP who had COVID-19–like symptoms or suspected exposure. From April 20, 2020, to June 23, 2020, all Northwell HCP were offered free, voluntary antibody testing, regardless of symptoms, at 52 sites in the greater NYC area. HCP missing all identifying data were excluded. Testing was for qualitative IgG or total immunoreactivity to SARS-CoV-2.³ Seven different assays were used (eTable in the Supplement); Northwell Health Laboratories validated all testing.

The main outcome was seroprevalence. Seroprevalence with 95% confidence interval was calculated by the exact binomial technique. HCP reported demographics, primary work location, job function, direct patient care, work on a COVID or non-COVID unit, and their level of suspicion of virus exposure: “Do you believe you were infected with COVID-19?” (range, 1-9; 1 = no; 9 = yes definitely; 7-9 = high suspicion). Associations among seroprevalence and these variables was assessed using Poisson logistic regression. All eligible persons were included in all analyses by creating a missingness subcategory for each variable. R version 4.0.1 (R Foundation for Statistical Computing) was used for analyses. P < .05 (2-sided) defined statistical significance. The Northwell Health institutional review board approved this research; all participants provided electronic informed consent.

All Northwell HCP (n = 70 812) were invited: 46 117 (65.1%) were tested as of June 23, 2020. The final consented sample of 40 329 (57.0%) (median age, 42 [interquartile range, 31.5-54.5] years) included 73.7% women, 16.0% Black, 0.8% multiracial, and 14.0% Hispanic HCP (Table 1) and 28.4% nurses and 9.3% physicians (Table 2).

Overall, 5523 of 40 329 (13.7% [95% CI, 13.4%-14.0%]) HCP were seropositive. Of 6078 with previous PCR testing results, 2186 (36.0%) were PCR positive. Of these PCR-positive HCP, 2044 (93.5%) were also seropositive, leaving 142 (6.5%) with negative antibody test results. Of the 3892 PCR-negative HCP, 3490 (89.7%) were also seronegative. Of 34 251 with no PCR testing, 3077 (9.0%) were seropositive (Table 2).

Missing data ranged from 0% to 15.4%. Working in COVID-19 units or in direct patient care were each associated with seroprevalence in bivariate analyses but not in multivariable analyses. In a fully adjusted model, several demographic variables (including increasing age and non-White race or ethnicity), a previous positive PCR test result (relative risk, 4.03 [95% CI, 3.50-4.64]; P < .001), and reported high suspicion of virus exposure (relative risk, 4.38 [95% CI, 3.32-5.77]; P < .001) were associated with seroprevalence (Table 2).

A 13.7% prevalence of SARS-CoV-2 antibodies in this large cohort study of HCP in the greater NYC area was similar to that among adults randomly tested in New York State (14.0%)⁴ but higher than among adults in Los Angeles (4.1%).⁵ HCP in a single hospital in Belgium had lower seroprevalence (6.4%), which was significantly associated only with household contact.⁶ In this study, high levels of HCP-reported suspicion of virus exposure and prior positive PCR testing results were most strongly associated with seropositivity.

Study limitations include voluntary testing, with only 56% of HCP participating; restriction to the greater NYC area; 7 different assays with variable sensitivity and specificity used; and time between PCR and antibody testing unknown and possibly too short to detect antibody response. Only HCP-reported suspicion of overall exposure was recorded, so distinguishing among community-, home-, and health care–acquired exposures was not possible.

Providing HCP with data about their SARS-CoV-2 virus exposure is important so they can protect themselves, their patients, their colleagues, and their families. High levels of HCP-reported suspicion of virus exposure may be useful as an indication for SARS-CoV-2 testing.

Corresponding Author: Karina W. Davidson, PhD, MASc, Northwell Health, 130 E 59th St, Ste 14C, New York, NY 10022 (kdavidson2@northwell.edu).

Accepted for Publication: July 23, 2020.

Published Online: August 6, 2020. doi:10.1001/jama.2020.14765

Correction: This article was corrected on December 8, 2020, for data inaccuracies in the text and tables.

Author Contributions: Drs Davidson and Chang had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Sembajwe, Jarrett, Farber, Chang, Davidson.

Drafting of the manuscript: Moscola, Sembajwe, Farber, Chang, Davidson.

Critical revision of the manuscript for important intellectual content: Sembajwe, Jarrett, McGinn, Davidson.

Statistical analysis: Sembajwe.

Obtained funding: Davidson.

Administrative, technical, or material support: Moscola, Jarrett, Farber, Chang, McGinn, Davidson.

Supervision: McGinn, Davidson.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by grants R24AG064191 from the National Institute on Aging and R01LM012836 from the National Library of Medicine.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Disclaimer: The views expressed in this article are those of the authors and do not represent the views of the National Institutes of Health, the US Department of Health and Human Services, or any other government entity. Dr Davidson is a member of the US Preventive Services Task Force (USPSTF). This article does not represent the views and policies of the USPSTF.

Additional Contributions: We thank the members of the Northwell Health COVID-19 Research Consortium who contributed to this study: Lance B. Becker, MD (North Shore University Hospital/Long Island Jewish Emergency Medical Department, Northwell Health); Dwayne A. Breining, MD, Jacqueline Moline, MD, MSc (Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, Northwell Health); Mark J. Butler, PhD, Jennifer Cookingham, MHA, Andrew J. Dominello, BA, Louise Falzon, BA, PGDipInf, Cirrus Foroughi, PhD, Jennifer C. Johnson, MS, MA, Jazmin N. Mogavero, MA, Rachel Monane, BBA, Frank Vicari, BS (Feinstein Institutes for Medical Research, Northwell Health); James M. Crawford, MD, PhD, Sharon S. Fox, BS, Stefan Juretschko, PhD, D(ABMM), Cheryl B. Schleicher, MS-HCA, BA (Northwell Health Laboratories, Northwell Health); Christopher J. Hutchins, Kristen M. McPhillips, MBA, PMP, and Abraham Saraya, MD, MSc (Northwell Health). None of these individuals received any compensation for their contributions.