Re: The disease burden of Delta and Omicron variants of severe acute respiratory syndrome coronavirus 2 in a predominantly vaccinated and healthy cohort.

To the Editor

Since February 2021, we have been studying the sero-prevalence of anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) anti-bodies in a cohort of individuals employed by Cantonal Police Bern in Switzerland. The baseline study (February 2021), 3-month (May 2021) and 6-month (September 2021) follow-up visits correspond to the first, second and third cross-sectional analyses of the cohort, as previously reported (Fig. S1) [1,2]. Here, we presented the rates of SARS-CoV-2 infection and vaccination at the 9-month (fourth cross-sectional analysis, in December 2021) and 12-month visits (fifth cross-sectional analysis, in March and April 2022) and estimated the burden of SARS-CoV-2 within the cohort after the surges of the Delta and Omicron variants.

All participants signed written informed consent prior to enrolment in the PoliCOV-19 study. The collection of coded data and the design of the work were approved by the Cantonal Research Ethics Commission of Bern, Switzerland (ID-2020-02650). The trial was registered at ClinicalTrials.gov NCT04643444.

The cohort population included 1022 study participants. The mean age of the participants was 41 (standard deviation, 8.8) years, 72% were men and 76% had no comorbidity. Serologic methods and questionnaires were used as previously reported [1,2]. Please see supplementary material.

From late June 2021, the SARS-CoV-2 Delta variant B.1.617.2 (all sub-variants AY) was dominant in Switzerland, until its replacement by the Omicron sub-lineages BA.1 and BA.2, which became predominant from mid-to-late December 2021 to mid-February and the beginning of April 2022, respectively (Fig. S2).

The rate of SARS-CoV-2 infection per 3 months was defined as follows: (a) sero-conversion of anti-nucleocapsid (NCP) anti-bodies or (b) a self-reported PCR test from a nasopharyngeal or saliva sample. The proportion of positive agreement (PPA) between the two modalities was evaluated and adjusted when sero-conversion occurred at later follow-up visits. To estimate the burden of SARS-CoV-2 (including disease and non-disease consequences), we evaluated the proportion of study participants who reported symptoms, sought a medical doctor, required hospitalization and missed work or police academy because of coronavirus disease 2019 (COVID-19).

At the 9-month visit, the sero-prevalence of anti-NCP anti-bodies was 19.1% (n = 188/985; 95% CI, 16.7%–21.5%). Anti-NCP sero-conversion was identified in 4.8% of the samples (n = 47; 95% CI, 3.6%–6.3%). Twenty-one individuals without sero-conversion reported a positive PCR test result from a nasopharyngeal or saliva sample, of whom 16 showed sero-conversion at the 12-month visit. Hence, the infection rate within the previous 3 months was 6.9% (68/985; 95% CI, 5.5%–8.7%), and the adjusted PPA for the PCR test results and sero-conversion was 92.6% (63/68).

At the 12-month visit, the sero-prevalence of anti-NCP anti-bodies was 51.6% (n = 499/967; 95% CI, 48.5%–54.7%). Anti-NCP sero-conversion was identified in 32.7% of the samples (n = 316; 95% CI, 29.2%–35.1%). Sixty individuals without sero-conversion reported positive PCR test results. After excluding 16 individuals with a positive PCR test result at the 9-month visit and sero-conversion at the 12-month visit, the infection rate within the previous 3 months was 37.2% (360/967; 95% CI, 34.2%–40.3%). The PPA was 83.3% (300/360) and could not be adjusted for late sero-conversion because the study ended after the fifth cross-sectional analysis.

To assess these infection rates in view of the two parameters for protective immunity, we evaluated the following: (a) vaccination rates and (b) the presence of anti-spike (S) anti-bodies 2 weeks prior to the start of each cross-sectional analysis (Fig. 1 ). The increase in vaccination rate for the first (1.9%) or second (3.1%) dose between the 9- and 12-month visits was minimal. However, the proportion of individuals who received a booster dose rose from 0.6% to 60% and 63.3% 8 and 2 weeks prior to the 12-month visit. The corresponding results of anti-S anti-bodies are shown in Fig. 1.

Fig. 1

Vaccination rates and the presence of anti-spike (S) anti-bodies in the police cohort during the 1-year study period. The blue bars (anti-S anti-body titres) were inserted at dates 2 weeks prior to the start of each cross-sectional analysis. The time points for anti-S anti-bodies were biased by the per-protocol predefined serum sampling time points. Prior to the 9-month visit (i.e. after the third cross-sectional analysis), 87.0% of the individuals displayed anti-S anti-bodies with a titre of ≥2 U/mL and 81% with a titre of >250 U/mL. Prior to the fifth cross-sectional analysis (i.e. after the fourth cross-sectional analysis), 93.0% of the individuals displayed anti-S anti-bodies with a titre of ≥2 U/mL and 88% with a titre of >250 U/mL. The red, green and purple curves illustrate the cumulative vaccination rates over time. The self-reported vaccination dates were collected using questionnaires. The vaccination rates are the values 2 weeks prior to the start of each cross-sectional analysis (except 65.5% for the 3rd vaccination; this value is the vaccination rate at the end of the study).

At the 12-month visit, we found no statistically significant difference between the infection rates among individuals who received a booster vaccine dose (36.6%; 242/661) and those who did not (38.6%, 118/306; difference, 2.0%; 95% CI, 4.6%–8.5%, p 0.559).

At the 9-month visit, 0.5% of the study participants reported having had no symptoms, and at the 12-month visit, 2.0% did so. At the 9-month-visit, 6.1% reported that they sought a doctor appointment, 16.3% reported that they missed working days (or school days at the police academy) and 1.2% reported that they were hospitalized because of COVID-19. After the 12-month visit, the proportions were 7.3%, 33.4% and 1.8%, respectively. The median numbers of missed working days were 7 (interquartile range, 2–10) days during the surge of the Delta variant and 4 (interquartile range, 2–5) days during the surge of the Omicron variant.

This study evaluated the burden of the surges of the Delta and Omicron variants in a healthy population cohort with a relatively high vaccination rate. During the surge of the Omicron variant, an infection rate of 37.2% (95% CI, 34.2%–40.3%) was noted in the cohort within 3 months despite the fact that more than 88% had received two doses of an mRNA vaccine and more than 63% had received an additional booster dose. The sero-prevalence of anti-NCP anti-bodies rose from 15% in September 2021 to 19% in December 2021 to 52% in April 2022 (Fig. S1). The latter number is likely under-estimated considering that 60 (6.2%) study participants tested positive for SARS-CoV-2 using a nasopharyngeal or saliva sample and did not (yet) reveal sero-conversion at the time of serum sampling.

The high infection rate observed from January to April 2022 is likely due to the transmissibility and immunologic escape properties of the Omicron variant [3]. The disease and non-disease consequences of the Delta and Omicron variants in a defined cohort have not been previously quantified. From a disease perspective, these data reinforce previous findings and support information statements by health authorities that COVID-19 vaccine remains protective against hospitalization and severe infection [4]. From a non-disease perspective, considerable absence from work within 3 months was observed, illustrating the disruption potential of the Omicron variants in emergency service providers.

Because we found no difference in the infection rate between the non-boosted and boosted individuals within the 3-month window at which the Omicron variants were predominant and because 95% among the boosted group received their third vaccine dose 8 weeks prior to the cross-sectional analyses, the results suggest that protection from non-severe Omicron infection is not durable. However, we were unable to confirm this conclusion because we did not inquire about the precise infection dates in the questionnaires.

The data provided in this study are highly relevant for the preparation and management of large-scale absence in workplaces for both public and private sectors [5]. They indicate that the burden of SARS-CoV-2 may shift from hospitalization because of severe disease to work shortage because of the Omicron variants.

Transparency declaration

The authors declare that they have no conflicts of interest. There was no funding via a grant or an external institution. The manufacturer of the electrochemiluminescence immune-assay tests (Roche) provided no funding for this study because the test was commercially purchased by the investigators. The study was funded in part by the Cantonal Police of Bern, Bern, Switzerland. The Institute for Infectious Diseases of the University of Bern and the Interregional Blood Transfusion, Swiss Red Cross, Bern, Switzerland, supported the study by providing working hours of their employees specifically for this study and providing material and consumables at cost or for free.