key: cord-0918093-zio4lhk6
authors: Iversen, Kasper; Kristensen, Jonas Henrik; Hasselbalch, Rasmus Bo; Pries-Heje, Mia; Nielsen, Pernille Brok; Knudsen, Andreas Dehlbæk; Fogh, Kamille; Norsk, Jakob Boesgaard; Andersen, Ove; Fischer, Thea Køhler; Juul Jensen, Claus Antonio; Torp-Pedersen, Christian; Rungby, Jørgen; Ditlev, Sisse Bolm; Hageman, Ida; Møgelvang, Rasmus; Gybel-Brask, Mikkel; Dessau, Ram B.; Sørensen, Erik; Harritshøj, Lene; Folke, Fredrik; Sten, Curt; Engel Møller, Maria Elizabeth; Benfield, Thomas; Ullum, Henrik; Jørgensen, Charlotte Sværke; Erikstrup, Christian; Ostrowski, Sisse R.; Nielsen, Susanne Dam; Bundgaard, Henning
title: Seroprevalence of SARS-CoV-2 antibodies and reduced risk of reinfection through six months: a Danish observational cohort study of 44,000 healthcare workers
date: 2021-09-17
journal: Clin Microbiol Infect
DOI: 10.1016/j.cmi.2021.09.005
sha: bcc3ef275aec00549483b637551d17c388e3f7e6
doc_id: 918093
cord_uid: zio4lhk6

OBJECTIVES: Antibodies to severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) are a key factor against COVID-19. We examined longitudinal changes in seroprevalence in healthcare workers (HCW) in Copenhagen and the protective effect of antibodies against SARS-CoV-2. METHODS: In this prospective study, screening for antibodies against SARS-CoV-2 (ELISA) was offered to HCW three times over six months. HCW-characteristics were obtained by questionnaires. ClinicalTrials.gov, NCT04346186. RESULTS: From April to October 2020 we screened 44,698 HCW of which 2,811 were seropositive at least once. The seroprevalence increased from 4.0% (1,501/37,452) to 7.4% (2,022/27,457) during the period (p<0.001) and was significantly higher than in non-HCW. Frontline HCW had a significantly increased risk of seropositivity compared to non-frontline HCW with risk ratios (RR) at the three rounds of 1.49 (95% CI 1.34-1.65, p<0.001), 1.52 (1.39-1.68, p<0.001) and 1.50 (1.38-1.64, p<0.001). The seroprevalence was 1.42- to 2.25-fold higher (p<0.001) in HCW from dedicated COVID-19 wards compared to other frontline HCW. Seropositive HCW had a RR of 0.35 (0.15-0.85, p=0.012) of reinfection during the following six months and 2,115 (95%) out of 2,248 of those who were seropositive during rounds one or two remained seropositive after four to six months. The 133 of 2,248 (5.0%) participants who seroreverted were slightly older and reported fewer symptoms than other seropositive participants. CONCLUSIONS: HCW remained at increased risk of infection with SARS-CoV-2 during the six months period. Seropositivity against SARS-CoV-2 persisted for at least six months in the vast majority of HCW and was associated with a significantly lower risk of reinfection.

The current Coronavirus disease 2019 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to more than 120 million confirmed cases and almost three million deaths worldwide(1). Vaccination combined with natural immunity against SARS-CoV-2 is expected to bring the pandemic under control, but whether immunity after vaccination or SARS-CoV-2 infection is lasting remains a key question. This is of importance to the current vaccination strategies deployed worldwide.

Currently, recommendation to vaccinate individuals with previously verified SARS-CoV-2 infection differ between countries.

Immunity against SARS-CoV-2 is established by humoral and cell-mediated immune responses, but much is still to be learned (2, 3) . In infected individuals antibodies against SARS-CoV-2 can be detected at an estimated mean of 12-15 days from symptoms onset, and virtually all SARS-CoV-2 infected, immunocompetent individuals seroconvert within 19-50 days (4) (5) (6) . Antibody development is generally thought to be one of the most important measures to prevent COVID-19 reinfection. However, reinfection has been reported in public media and case reports (7) (8) (9) (10) (11) (12) (13) (14) . In these reports re-infected individuals were often asymptomatic during the first course of infection with SARS-CoV-2. Recently, one larger cohort study found reinfection with SARS-CoV-2 to be rare and mild up to six months post primary infection but the rate of seroreversion is still inknown (3) . Seroreversion may be due to a low level of antibodies after mild infection, or immunodeficiency, but as SARS-CoV-2 is a relatively new infectious agent in humans, the percentage that serorevert over time is still being explored (15) . Also, mutations in SARS-CoV-2 may change the properties of the virus leading to reduced immunity.

We have previously reported that healthcare workers (HCW) are at higher risk of contracting SARS-CoV-2 than the general population (16) . Extended seroprevalence studies in HCW may be of importance for safety reasons, and to ensure continued staffing of the healthcare sector but may also provide knowledge on disease development and immunity before the general population reaches the same seroprevalence levels.

The aim of this study was to examine longitudinal changes in seroprevalence and seroconversion in a highly exposed population of HCW in the Capital Region of Denmark.

J o u r n a l P r e -p r o o f

In this prospective cohort study, we organized a scheme of repeated, voluntary testing for antibodies against SARS-CoV-2 among HCW in the Capital Region of Denmark (1.84 million inhabitants) from April to October 2020. Voluntary testing was supported by the administrative and political systems in the region.

All staff at somatic, psychiatric, prehospital and specialized health-care institutions as well as administrative and technical staff employed in the healthcare organization in the Capital Region of Denmark were offered screening. Further, doctors and other staff working in the primary healthcare sector as well as students working in the healthcare system were invited to participate. HCW where stratified by work assignments; at dedicated COVID-19 wards, other frontline HCW, and remaining HCW. Information on work assignment at dedicated COVID-19 wards was retrieved by questionnaires. Frontline HCW were defined as all doctors, nurses, assistant nurses, and medical and nursing students who were exposed to patients potentially infected with SARS-CoV-2 but did not work at COVID-19 dedicated wards. The rest were HCW not adhering to the first two groups.

Screening for antibodies was offered three times during 2020. The first screening was offered and April 20-22 for hospital staff and May 4-7 for staff from the primary healthcare sector. The second screening was performed June 2-4 and June 8-10 for all staff. The third and last screening was performed September 30 -October 2, and October 5-7. Participation in the second and third screening was not conditioned by prior participation.

Participants were asked at each round to fill in a survey using a smartphone or computer. The survey was accessed through a link sent to their email at the Danish, governmental, personal, password-protected, email system, e-Boks, or via a QR code at the blood sampling clinics. Participants filled in information about demographics, exposure to SARS-CoV-2, symptoms and SARS-CoV-2 polymerase chain reaction (PCR) testing (appendix p 2). Study data was collected and managed using Research Electronic Data Capture, a secure, web-based, electronic data capture tool, hosted at the Capital Region's server (17, 18) . Participation in the survey was voluntary and antibody screening was also provided to staff who did not wish to participate in the survey.

Detailed information on dissemination of information regarding the project to HCW and organization of blood sampling at clinical departments and blood sampling clinics has previously been published (16) .

For comparative purposes, SARS-CoV-2 screening in blood donors were anonymously extracted from the Danish blood bank production system from the same region and for the same time period as screening in round three was performed for the HCW (19) . This group was used as a proxy for the general working population. The age range of blood donors was 18-64 years. 

This study was presented to the scientific ethics committee of the Capital Region. They concluded that the study did not require a scientific ethical approval (J.nr-H-20026288). The study was registered with the Danish Data Protection Authorities (P-2020-361) and the protocol is registered at CinicalTrials.gov(https://clinicaltrials.gov/ct2/show/NCT04346186).

Calculations were done using R (version 6.3.1). All results were presented as mean (standard deviation ±SD) or median (interquartile range (IQR)) according to normality and tested using either chi-square test, t-test or Wilcoxon rank sum test. Possible associations between exposures and the primary outcome were explored by risk ratios (RRs) presented with 95% confidence intervals (CI), calculated using the normal approximation (Wald). Significance was examined by Fisher's exact test using the R package epitools. A pvalue of less than 0.05 (two-sided) was considered significant. A multivariable logistic regression including age, sex, asymptomatic versus symptomatic SARS-CoV-2 infection, and ever versus never smoker was used to assess risk factors of seroreversion, with each predictor presented with odds ratios (OR) and 95% CI. 

Frontline HCW had a significantly increased risk of being seropositive throughout the three rounds as compared with non-frontline HCW with a RR of 1. Gradual increases in seropositivity from round one to three were seen in the three main categories of HCW and were most pronounced in HCW in COVID-19 wards ( Figure 2 ). Correspondingly, the incidence of seropositivity was highest in HCW at COVID-19 wards (6.35% (190/2,992), 7.66% (257/3,354), and 3.14%

(78/2,486) in rounds one to three respectively) (Supplementary figure 7) . During the study period the incidence of seropositivity mostly decreased from round one to three for HCW working frontline (4.5% J o u r n a l P r e -p r o o f Seroprevalence stratified according to job categories, age and specialty Figure 3 shows the seroprevalence among doctors, nurses, and assisting nurses stratified by specialty and round (incidence for each round shown in supplementary figure 8) . Overall, the differences between the specialties were maintained during the three rounds.

The seroprevalence and -incidence for different job categories are shown in supplementary figures 9-10.

The highest seroprevalence was observed among medical students where 25.23% (222/880), 28.02%

(167/596), and 34.25% (186/543) were seropositive in rounds one to three, respectively. The lowest seroprevalence was observed among midwifes, where 2.3% (13/555), 1.7% (7/404), and 3.7% (13/356) were seropositive in rounds one to three, respectively. Overall, the differences between the job categories were maintained during the three rounds.

In a subgroup analysis of HCW compared to blood donors stratified by under or over 30 years of age at round three, the difference between blood donors and HCW persisted in both the young and old (Supplementary figure 11) . In a subgroup analysis looking only at participants >30 years of age the incidence of seropositivity was similar to the one seen for all HCW (supplementary figure 12 ). S/CO-ratio for antibodies against SARS-CoV-2 Figure 4 shows the S/CO-ratio for seropositivity for 817 participants who were seropositive in round one and participated in all rounds as well as the fluctuation of S/CO-ratios for seroreverters and HCW with possible reinfections through rounds one to three. The S/CO-ratio of seropositive participants that participated in all rounds increased significantly through rounds one to three (median 11.92, IQR 6.49-17.98; median 16.97, IQR 8.11-23.68; median 19.31, IQR 14.28-23.68 for rounds one to three respectively, p<0.001 for all). Looking at the S/CO-ratio for these groups in round one, the participants who were seropositive in all three rounds had a significantly higher S/CO-ratio (median 11.92, IQR 6.49-17.98) than J o u r n a l P r e -p r o o f both participants who seroreverted in round two and did not subsequently seroconvert (median 1.64, IQR 1.38-2.52, p<0.001), the participants who seroreverted between rounds two and three (median 6.22, IQR 1.70-14.91, p=0.01), and the participants with possible reinfections (median 2.76, IQR 1.70-4.76, p<0.001).

There was no clear difference in S/CO-ratio among the seropositive participants working on the frontline or at dedicated COVID-19 wards (Supplementary figure 13) .

A total of 2,811 participants were found to be seropositive at least one time during the study period of which 2,248 were seropositive in one of the first two rounds. Of these 2,248 participants, 113 (5.0%) had seroreverted at a subsequent round. During the study period of six months 948 (94.5%) of 1,003 seropositive participants that participated in both round one and round three stayed seropositive. Table 2 shows a comparison of basic characteristics for seroreverters to participants that stayed seropositive. The participants who seroreverted were significantly older than the other seropositive participants, reported milder illness and were less likely to think they had been ill because of SARS-CoV-2. In a multivariable logistic regression model including age, sex, asymptomatic versus symptomatic SARS-CoV-2 infection, and ever versus never smoker, only age (OR 1.03, 95% CI 1.01-1.04 p<0.001) and asymptomatic SARS-CoV-2 infection (OR 7.46, 4.8-11.86, p<0.001) were significantly associated with an increased risk of seroreversion.

Supplementary table 6 shows basic characteristics of seroreverters compared to all other participants.

We found a gradual and significant increase in seroprevalence in HCW from April to October 2020. HCW with antibodies to SARS-CoV-2 had a 65% reduction in risk of reinfection during the following six months and approximately 95% of those who were seropositive during the first round remained seropositive after six months. Seroreverters were slightly older and had a milder course of disease.

At all three rounds, HCW working at COVID-19 wards had the highest seroprevalence followed by other frontline personnel. The lowest rate was seen in the remaining HCW. At the end of the study period this group did, however, still have a significantly higher seroprevalence than blood donors, which served as a proxy for the general working population. Seroprevalence was lower among females and decreased with age. The highest seroprevalence at all rounds was seen in medical students. HCW with the medical specialties respiratory medicine followed by infectious diseases and hematology had the highest seroprevalence. The differences remained stable during the study period.

Following seroconversion HCW reported lower rates of positive PCR tests as compared to seronegative HCW, indicating protection against reinfection in good agreement with findings in the general population of registry studies (20, 21) . This is reassuring, as it indicates that not only seropositive individuals in the general population, but also seropositive HCW seem to be similarly protected despite their increased SARS-CoV-2 exposure.

We have previously reported a higher seroprevalence in HCW than in blood donors and that the risk of SARS-CoV-2 infection was related to exposure to infected patients (16) . Despite this knowledge, the observed increase in seroprevalence during the study period was also highest in those exposed to patients with COVID-19. HCW at dedicated COVID-19 wards should use adequate protective measures, especially during the second and third screening round, since they were directly exposed to patients with COVID-19.

Frontline HCW could, on the other hand, be expected to have had a higher incidence, since compliance with protection measures in treatment of patients without a confirmed diagnosis of COVID-19 may be lower. The reason for our finding of an even greater seropositivity among HSW at COVID-19 wards is unclear. Provision of healthcare during the COVID-19 pandemic is of utmost importance, and safety for HCW is crucial. The present findings strongly support the need for protection of HCW against transmission and justifies the prioritized vaccination of HCW in most countries.

In a study of British HCW from the Oxford area(3), a seroprevalence of 9.4% was found in samples collected from March to June 2020. As in our study, the seroprevalence was higher in HCW than in blood donors in the same area which was approximately 4% in both March and June 2020 in the UK(22). No stratification J o u r n a l P r e -p r o o f according to specialty or information on whether personnel were working as front-line personnel or in COVID-19 wards was reported in the British study (3) .

The observed antibody-test characteristics may, in part, explain seroreversion between test rounds.

However, the importance of the potential, although limited, seroreversion is apparent, but it is reassuring that our data indicate that the antibodies developed in response to infection with SARS-CoV-2 persist in the vast majority for at least six months. Similar findings have been reported in a recent study by Wajnberg et al, where stability of IgG antibody titers was found over 148 days in individuals with mild to moderate COVID-19 (23) . While it is unknown to what degree previous infection with SARS-CoV-2 protects individuals from reinfection, a recent study as well as our data suggest that seropositivity is associated with protection against reinfection with SARS-CoV-2(24).

The HCW who seroreverted during the follow-up reported no or mild symptoms of COVID-19. This is in line with previous studies reporting a more rapid decline in antibody levels amongst infected individuals with no or mild symptoms(25). In a large study of three million patients who were tested for antibodies against SARS-CoV-2, seroreversion was seen in 18.4% of seropositive individuals during a median follow-up of 54 days (24) . However, these data were based on results from several laboratories. Antibody-test characteristics and differences in handling of samples may in part also explain the observed, high rate of seroreversion as compared to our findings.

The high seroprevalence among Copenhagen medical students was also observed in a seroprevalence study of medical students from autumn, 2020 (26) . This may be caused by outbreaks at social events for medical students. Danish students have also been working at SARS-CoV-2 testing facilities during the pandemic and behavioral patterns may be in play. Also, seroprevalence among medical students contribute to the observed decrease in seroprevalence with increasing age as medical students are younger than the other participants.

Our study has several limitations. Only 42% of participants participated in all study rounds. This may be explained by the introduction of widespread screening options by nasal swab for PCR testing for HCW during the second and third round of our study. In comparison, PCR testing was only available for symptomatic individuals in March and April, while an antibody test made it possible to tell if an individual was likely to have had SARS-CoV-2. It seems plausible that individuals who did not find it likely that they J o u r n a l P r e -p r o o f would have seroconverted since the last screening were less motivated to participate in the consecutive follow-up rounds.

While participation in the questionnaire was possible from home, severely ill persons would not necessarily have been able to participate in follow-up rounds which may have resulted in a small drop-out among persons who were infected during the study period. Unfortunately, information on ventilator-free days, length of hospital admission, and mortality were not available in the current questionnaire-based study.

Of the 25 participants, who were seropositive, then seronegative and again seropositive in round one, two and three, it is unknown if the seroconversion represents actual reinfection. None of them reported to have had a positive PCR test after round one. It is possible that the antibody test results from these individuals are either false-positive or false-negative.

In conclusion, this study found HCW working frontline and HCW working at dedicated COVID-19 wards to remain at a significantly increased risk of infection with SARS-CoV-2 during the six months period of the pandemic. Seropositivity against SARS-CoV-2 was found to last for at least six months after infection in nearly all HCW and importantly, reduced the risk of contracting SARS-CoV-2 by 65%. Longitudinal seroprevalence studies with longer follow-up are needed to assess if immunity is lasting and protects against future SARS-CoV-2 infection.

Lundbeck Foundation[R349-2020-731]. The benefactor had no role in study design, data collection, analysis, interpretation, or writing of the article.

J o u r n a l P r e -p r o o f 

Professor Iversen received grants from Lundbeck Foundation to his institution

Folke has received research grant from the Novo Nordisk Foundation and an unrestricted research grant from the Laerdal Foundation to the Copenhagen EMS. Dr. Nielsen has an unrestricted research grant from the Novo Nordic Foundation. Dr. Knudsen has received a grant from the Danish Heart Foundation unrelated to this work. Dr. Dessau reports personal fees from Roche Diagnostics, outside the submitted work. Professor Benfield received grants for his institution from the Novo Nordisk Foundation, Simonsen Foundation, Lundbeck Foundation, Kai Foundation, and Erik and Susanna Olesen's Charitable Fund. Also, he received an unrestricted grant for his institution and advisory board from GSK

Boehringer Ingelheim as principal investigator for a clinical trial; from Gilead Sciences an unrestricted grant for his institution, as principal investigator for a clinical trial and advisory board; from MSD an unrestricted grant for his institution, as principal investigator and advisory board

Professor Benfield received payment or honoraria from GSK, Pfizer, Gilead Sciences, Boehringer Ingelheim, and Abbvie for lectures

Professor Benfield received a donation of trial medication from Eli Lilly

Original Draft: JHK, RBH, MPH, LH, HB References 1. WHO Coronavirus (COVID-19) Dashboard | WHO Coronavirus Disease (COVID-19) Dashboard

Immunological considerations for COVID-19 vaccine strategies

Antibody Status and Incidence of SARS-CoV-2 Infection in Health Care Workers

Antibody responses to SARS-CoV-2 in patients with COVID-19

Humoral response and PCR positivity in patients with COVID-19 in the New York City region, USA: an observational study. The Lancet Microbe

Antibody response to SARS-CoV-2 infection in humans: A systematic review

Genomic evidence for reinfection with SARS-CoV-2: a case study

Symptomatic SARS-CoV-2 reinfection by a phylogenetically distinct strain

Re-Infection by a Phylogenetically Distinct SARS-CoV-2 Variant, First Confirmed Event in South America

Coronavirus Disease 2019 (COVID-19) Re-infection by a Phylogenetically Distinct Severe Acute Respiratory Syndrome Coronavirus 2 Strain Confirmed by Whole Genome Sequencing

Asymptomatic Reinfection in 2 Healthcare Workers From India With Genetically Distinct Severe Acute Respiratory Syndrome Coronavirus 2

Clinical immunity in discharged medical patients with COVID-19

Clinical recurrences of COVID-19 symptoms after recovery: Viral relapse, reinfection or inflammatory rebound?

Reinfection with SARS-CoV-2: considerations for public health response

Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections

Risk of COVID-19 in health-care workers in Denmark: an observational cohort study

The REDCap consortium: Building an international community of software platform partners

Research electronic data capture (REDCap)-A metadata-driven methodology and workflow process for providing translational research informatics support

Estimation of SARS-CoV-2 infection fatality rate by real-time antibody screening of blood donors

Assessment of protection against reinfection with SARS-CoV-2 among 4 million PCR-tested individuals in Denmark in 2020: a population-level observational study

Seropositive Antibody Test With Risk of Future Infection

Robust neutralizing antibodies to SARS-CoV-2 infection persist for months. Science (80-)

Real-world data suggest antibody positivity to SARS-CoV-2 is associated with a decreased risk of future infection. medRxiv Prepr Serv Heal Sci

Seropositivity Among Medical Students in Copenhagen. Open Forum Infect Dis