key: cord-0280692-8oioivl9 authors: Rivelli, A.; Fitzpatrick, V.; Blair, C.; Richards, J.; Copeland, K. title: Incidence of COVID-19 reinfection among Midwestern healthcare employees date: 2021-09-12 journal: nan DOI: 10.1101/2021.09.07.21263100 sha: 4c86570fd3cdc46d6ff0f0b0ac611f97bcba7b40 doc_id: 280692 cord_uid: 8oioivl9 Importance: Given the overwhelming worldwide rate of infection and the disappointing pace of vaccination, addressing reinfection is critical. Understanding reinfection, including protection longevity after natural infection, will allow us to better know the prospect of herd immunity, which hinges on the assumption that natural infection generates sufficient, protective immunity. The primary aim of this paper is to provide data on SARS-CoV-2 reinfection over a 10-month period. Objective: The primary objective of this study is to establish the incidence of reinfection of COVID-19 among healthcare employees who experienced a prior COVID-19 infection. Design: This observational cohort study followed a convenience sample of 2,625 participants who experienced a COVID-19 infection for subsequent COVID-19 infection. Setting: Healthcare employees were recruited across a large Midwestern healthcare system. Positive PCR test results were administered and recorded by the system-affiliated lab serving Illinois and Wisconsin. Participants: Adult healthcare system employees who enrolled in a research study focused on SARS-CoV-2 antibodies (N = 16,357) and had at least one positive PCR test result between March 1, 2020 and January 10, 2021 were included (N = 2,625). Exposure: Positive PCR test for SARS-CoV-2 Main Outcome(s) and Measure(s): The primary outcome is incidence of COVID-19 reinfection, defined by current CDC guidelines (i.e. subsequent COVID-19 infection [≥] 90 days from prior infection). COVID-19 recurrence, defined as subsequent COVID-19 infection after prior infection irrespective of time, is also described. Results: Of 2,625 participants who experienced at least one COVID-19 infection during the 10-month study period, 156 (5.94%) experienced reinfection and 540 (20.57%) experienced recurrence after prior infection. Median days were 126.50 (105.50-171.00) to reinfection and 31.50 (10.00-72.00) to recurrence. Incidence rate of COVID-19 reinfection was 0.35 cases per 1,000 person-days, with participants working in COVID-clinical and clinical units experiencing 3.77 and 3.57 times, respectively, greater risk of reinfection relative to those working in non-clinical units. Incidence rate of COVID-19 recurrence was 1.47 cases per 1,000 person-days. Conclusions and Relevance: This study supports the consensus that COVID-19 reinfection, defined as subsequent infection [≥] 90 days after prior infection, is rare, even among a sample of healthcare workers with frequent exposure. SARS-CoV-2, the virus that causes COVID-19, has been shroud in mystery since the first confirmed case was documented in Wuhan City, China in December 2019. A year and a half later, there have been over 190 million cases, 4 million deaths, and varying degrees of successful containment and mitigation. 1 The ultimate goal is global herd immunity for COVID-19, with the two main paths to achieving herd immunity being natural infection and vaccination. 2 After six months of mass vaccination efforts against SARS-CoV-2, preliminary data suggest extremely promising vaccine immunity results. However, while some countries have vaccinated more than half of their populations, many lag behind. 3 Given the overwhelming worldwide rate of infection, especially with emerging variants, and the disappointing pace of vaccination, addressing reinfection is critical. Addressing reinfection, particularly the longevity of protection after natural infection, or natural immunity, will allow us to better understand the prospect of herd immunity, which hinges on the assumption that natural infection generates sufficient, protective immunity. 2 The primary aim of this paper is to provide longitudinal data on natural immunity after SARS-CoV-2 infection. The incidence of true COVID-19 reinfection is challenging to document, as the extensive resources necessary to confirm reinfection have not been available or practical to employ clinically. 4 Confirmation of reinfection requires multiple polymerase chain reaction (PCR) tests, viral cultures, lab testing, and collection of clinical symptoms and epidemiological risk factors. 4 This has subsequently led to probable under-reporting of reinfection in scientific journals, as evidence based on these inaccessible resources have been required for formal reporting of COVID-19 reinfection. 5 Additionally, most individuals around the world who became infected during the first COVID-19 pandemic wave did not access a PCR or antibody test and/or were not treated in the hospital, delaying efforts to recognize and track overall COVID-19 reinfection early on in the pandemic. [5] [6] [7] [8] [9] While the consensus is that reinfection is rare, more longitudinal studies focused on reinfection incidence in a variety of populations and time between confirmed infections will help corroborate this. 3, [10] [11] [12] The most up-to-date research suggests that infection provides natural immunity for at least three months 13 and immunity remains stable up to 6-8 months after the initial infection. 12, 14 Furthermore, the maximum duration of SARS-CoV-2 ribonucleic acid (RNA) shedding in the upper respiratory tract, indicating recurrence, has been reported to be between 83 and 104 days, 15 4 Use of these definitions in research would promote more clarity and unity in results reporting. This study aims to contributes longitudinal data on epidemiological reinfection in a large cohort of healthcare workers in the United States (US) with documented cases of COVID-19, as defined by positive PCR test results. This study is an extension of two previous studies among the same cohort that addressed factors related to seroprevalence of SARS-CoV-2 Immunoglobulin G (IgG) 21 and 3-month incidence of COVID-19 recurrence by SARS-CoV-2 IgG status. 22 In this prior publication, recurrence was used as an umbrella term that comprised numerous scenarios, including persistent illness, prolonged viral RNA shedding, increased virus replication, a different symptomatic viral infection in the presence of remnant SARS-CoV-2 RNA, and/or true reinfection with disease. 4, 7, 9, 19 This current study defines 'recurrence' the same, as all instances of subsequent reinfection after initial infection during the study period. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. This study will address 10-month cumulative incidence of COVID-19 reinfection using the CDC's current guidelines. To provide context around reinfection, this study will also describe 10-month cumulative incidence of recurrence. We will also describe time to reinfection and recurrence, overall and stratified by clinical role in order to shine a light on the role of exposure frequency to SARS-CoV-2 in incidence and time to recurrence and reinfection. This prospective cohort study recruited healthcare employees across a large Midwestern healthcare system, which consists of 26-hospitals and over 500 sites of care in Illinois and Wisconsin. SARS-CoV-2 IgG was measured in serum specimens obtained from all participants using the SARS-CoV-2 IgG Abbott Architect assay. Performance characteristics of the SARS-CoV-2 IgG assay were validated at ACL Laboratories, determining a sensitivity of 98.7% and specificity of 99.2%. [23] [24] To detect SARS-CoV-2, this study used the Aptima Panther SARS-CoV-2 Assay, which uses qualitative detection of RNA from SARS-CoV-2 isolated and purified nasopharyngeal, oropharyngeal and nasal swab specimens obtained from individuals who meet COVID-19 clinical and/or epidemiological criteria. 25 Both the SARS-CoV-2 Antibody Assay and the Aptima Panther TMA SARS-CoV-2 Assay were approved for use under Emergency Use Authorization in US laboratories certified under the Clinical Laboratory Improvement Amendments of 1988. 26 Prior to recruitment, this study obtained approval by the Institutional Review Board (#20-168E). This study includes English-and Spanish-speaking adults ages ≥ 18 employed by the healthcare system as of June 8, 2020 (study initiation) who had at least one positive SARS-CoV-2 PCR test results in the system's Electronic Medical Record (EMR) system between March 1, 2020 and January 10, 2021. This sample of participants was drawn from the overarching study, which enrolled a convenience sample of 16,357 participants meeting the same inclusion criteria to test for SARS-CoV-2 IgG assay results between June 8, 2020 and July 10, 2020. 21 After enrollment, all participants' positive SARS-CoV-2 PCR test results documented in the system's All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.07.21263100 doi: medRxiv preprint EMR between March 1, 2020 and January 10, 2021 were collected. It is implicit that team members were tested at a system-affiliated lab, if tested at all, due to no cost, convenience and employment implications. On June 6, 2020, a detailed recruitment email was sent to all team members' work email addresses. The email provided instructions for participation in the study, including an alteration of consent and a study-specific passcode required for study registration. Interested team members were instructed to register in their active online health portal. Team members who met study inclusion criteria and completed a lab blood draw to test for SARS-CoV-2 IgG were participants in this study. Data gathered for this study included demographics and all system EMR-documented positive SARS- preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.07.21263100 doi: medRxiv preprint and their fifth documented infection was 92 days after their initial infection, so their initial and fifth infections and the days between were used in the reinfection analysis. This explains why there are more reinfection cases (156) than recurrence cases occurring at 90+ days (115). It should be noted that, if all first and last infections were included in reinfection analyses, there would be an additional 1162 person-days added to the overall person-time, reducing the incidence rate per 1,000 person-days a negligible amount. The secondary outcome, incidence of COVID-19 recurrence, represents the second documented SARS-CoV-2 positive PCR result after the initial documented SARS-CoV-2 positive PCR result, irrespective of time between positive results. Data management and analysis were performed by the study research team and conducted using SAS statistical software (Version 9.4; SAS Institute, Cary, NC). Descriptive statistics are reported as counts (%) or means (standard deviation) and median (interquartile range), as appropriate, particularly days to outcome. Demographic and baseline variables are also reported across primary and secondary outcome statuses. Corresponding measures of association include mean difference in age between those who did not experience reinfection or recurrence from those who did experience recurrence or reinfection and, for the remaining categorical variables, the odds ratio (OR), or the relative odds of participants of a given variable category experiencing COVID-19 recurrence or reinfection relative to the reference category of that variable. Variable reference levels were chosen based on lowest presumed risk. Corresponding p-values were generated from Student's T-tests for continuous variables and logistic regression Wald tests to represent differences in recurrence or reinfection. Cumulative incidence of COVID-19 recurrence was calculated as number of participants who experienced a subsequent infection at/after 90 days of prior infection (reinfection) or who experienced a subsequent infection at all (recurrence) by total number of participants at risk of a subsequent infection between earliest positive PCR test result (March 1, 2020) and study end (January 10, 2021). Incidence rate (IR) was calculated as the number of participants at risk who experienced each outcome by person-days contributed to follow-up before the outcome was experienced or participant was censored at study end. The entire study period All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.07.21263100 doi: medRxiv preprint was counted as 315 days (the number of days between earliest positive PCR test result and study end). Incidence measures were calculated overall and by clinical role category. Incidence rate ratio (IRR) represents the relative IR between clinical role categories. Finally, categories of days to reinfection and recurrence are described as counts and percentages. This study was funded internally. The healthcare system had no influence over the study design, conduct, results, or dissemination of findings. The authors received no direct financial support for the research, authorship, and/or publication of this article. (2.56%) had non-clinical roles within the healthcare system. Cumulative incidence of reinfection within 10 months was 5.94% overall, 6.70% among COVID-clinical participants, 6.23% among clinical participants, and 1.73% among non-clinical participants. IRRs indicated 3.77 times and 3.57 times increased risk of COVID-19 reinfection among COVID-clinical and clinical participants, respectively, relative to non-clinical participants. system, which put individuals in clinical roles at more than 3.5 times increased risk of COVID-19 reinfection as compared with individuals working remotely or in non-clinical roles. This study enrolled and followed a large cohort of healthcare employees to determine risk of reinfection, as defined by the CDC, in a population likely to be re-exposed to COVID-19. This study provides much needed data to contribute to existing research on reinfection. PCR tests for COVID-19 infection were performed within system-affiliated labs, resulting in test performance and reporting consistency. All data was stored in EMR system and extracted by the healthcare system's analytics team, resulting in data collection consistency. There are several limitations to this study. Most important, there was no viral testing done to participants' blood samples, eliminating the ability to conclusively determine whether two SARS-CoV-2 test results in the same individual were due to true reinfection or recurrence. Second, abstracted data for this study did not include symptomatology; therefore, we cannot determine 1) reasons participants tested multiple times, 2) sickness severity of participants with positive SARS-CoV-2 test results, or 3) commonalities among individuals with positive results. This information could have contributed to the body of literature that correlates All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.07.21263100 doi: medRxiv preprint viral load with the ability to transmit the virus. 27 Finally, because there is no universally accepted definition of reinfection, the study team used CDC retesting guidelines and some recently published guidance on proposed operational definitions of the terms to define reinfection and considered all subsequent positive test results to be recurrence. Overall, this study indicates that reinfection is possible but unlikely, and both reinfection and recurrence are more likely among high-exposure groups like clinical healthcare workers. Individuals in high-exposure groups should continue to abide by previous public health precautions, irrespective of policy easement. Widespread vaccination may be a solution to easing up on public health recommendations, but more long-term data is needed on vaccine efficacy, transmission and duration of protection in high exposure-risk populations. Vaccination rates will need to increase, as well, if we are ever to reach herd immunity since individuals will always be in higher-exposure groups. The current study end timeline was before the healthcare system began vaccinating front-line workers, which would have likely confounded the incidence of recurrence and reinfection. A future follow-up study using the same cohort will explore reinfection pre-and post-vaccination. Herd immunity: Understanding COVID-19. Immunity Six months of COVID vaccines: What 1.7 billion doses have taught scientists Definitions for coronavirus disease 2019 reinfection, relapse and PCR re-positivity Immunity to SARS-CoV-2 variants of concern The possibility and cause of relapse after previously recovering from COVID-19: a systematic review COVID-19 and post-infection immunity: limited evidence, many remaining questions Scientific consensus on the COVID-19 pandemic: we need to act now. The Lancet Clinical recurrences of COVID-19 symptoms after recovery: Viral relapse, reinfection or inflammatory rebound? COVID-19 reinfections are rare -but without better data, we don't know how rare All rights reserved. No reuse allowed without permission preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Study finds COVID-19 reinfections are rare, more common for those above age 65. The Lancet SARS-CoV-2 infection rates of antibody-positive compared with antibody negative health-care workers in England: A large, multicentre, prospective cohort study (SIREN). The Lancet Evaluation of potential COVID-19 recurrence in patients with late repeat positive SARS-CoV-2 testing Immunological memory to SARS-CoV-2 assessed for up to 8 months after infection SARS-CoV-2, SARS-CoV-1 and MERS-CoV viral load dynamics, duration of viral shedding and infectiousness -a living systematic review and metaanalysis. The Lancet Microbe Prolonged detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA in an obstetric patient with antibody seroconversion Prolonged virus shedding even after seroconversion in a patient with COVID-19 Profile of RT-PCR for SARS-CoV-2: A Preliminary Study From 56 COVID-19 Patients COVID-19 reinfection: Prolonged shedding or true reinfection? New Microbe and New Infect Interim Guidance on Ending Isolation and Precautions for Adults with COVID-19. cdc Disparities in seroprevalence of SARS-CoV-2 immunoglobulin antibodies in a large midwestern health care system Incidence of COVID-19 recurrence among large cohort of healthcare employees Comparison of Commercially Available and Laboratory Developed Assays for in vitro Detection of SARS CoV-2 in Clinical Laboratories SARS-CoV-2 IgG for use with Architect. U.S Food & Drug Administration COVID-19) Emergency Use Authorizations for Medical Devices. U.S. Food and Drug Administration COVID-19 Where are we on vaccines and variants The authors would like to thank the many people at Advocate Aurora Health who supported this study, specifically the executive team who financially supported staff testing, Public Affairs and Marketing, AndyMarek and Chris Blumberg in Analytics, ACL leadership and staff, the Health Informatics Technology (HIT) team, Advocate Aurora Research Institute (AARI), IRB, and Maureen Shields for early work on this project.