Infection rate and Immune Response to SARS-CoV-2 in Canadian Retail Workers -Cohort Description

34 Purpose: Retail workers are an understudied occupational group that may have been at 35 increased risk of contracting SARS-CoV-2 during the COVID-19 pandemic. Therefore, we 36 set up a longitudinal cohort of participants working in this sector to better document the 37 incidence of SARS-CoV-2 infection and the immune response to infection and/or 38 vaccination in this group. 39 Participants: A total of 304 participants were recruited between April 20, 2021 and 40 October 22, 2021. They were invited to attend three visits (each separated by ~12 weeks) 41 during which they provided blood samples and information on participant characteristics, 42 COVID-19 symptoms, and vaccination. An extension phase of two additional visits was 43 carried out between March 15 th, 2022 and October 3 rd, 2022 to document the impact of 44 the Omicron variant among the 198 participants who were still eligible for recruitment. 45 Participants were aged 18 to 75 and worked in grocery stores, hardware stores, bars or 46 restaurants within the Québec City metropolitan area (Canada). Findings to date: This 47 article describes participants’ demographic, socioeconomic, behavioral, clinical and 48 occupational characteristics, and their COVID-19 symptoms (where applicable). It also 49 describes SARS-CoV-2 vaccination status and any SARS-CoV-2 diagnostic test (i.e., PCR or 50 rapid antigen) performed from the beginning of the pandemic until the last visit. 51 Future plans: The incidence of SARS-CoV-2 infections will be assessed. The immune 52 response (innate and acquired) to SARS-CoV-2 infection or vaccination will be studied 53 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 4 using a variety of techniques, including reference and experimental enzyme-linked 54 immunosorbent assays, microneutralization assays with live viruses, experimental 55 pseudoneutralization with an angiotensin-converting enzyme 2-spike assay, peripheral 56 blood mononuclear cells and neutrophil stimulation, and a proliferation assay based on 57 carboxyfluorescein diacetate succinimidyl ester . 58 Registration: Not applicable. 59 Strengths and limitations 60  This cohort offers a comprehensive dataset to study the immune response to 61 SARS-CoV-2 infection or vaccination (alone), or hybrid immunity, as participants 62 provided information on a wide range of demographic, socioeconomic, 63 behavioral, clinical, and occupational variables. However, the low proportion of 64 racial minorities (i.e., 3.0%) limits the use of this cohort to study racial 65 determinants of immunity to SARS-CoV-2. 66  This cohort focuses on workers in the food and retail service sector, an 67 understudied population at high risk of occupational exposure to infectious 68 agents. 69  This study covered seven pandemic waves and thus captured a large number of 70 confirmed infections from different variants. 71  Sample collection was initiated immediately prior to COVID-19 vaccine availability 72 for this population and thus captured successive vaccination campaigns over 17 73 months. 74 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 5  In keeping with the study design, none of the participants had severe COVID-19 75 disease requiring hospitalization at baseline, and none of the COVID-19 illnesses 76 that occurred during the study required hospitalization; this cohort may not, 77 therefore, be used to study the immune response leading to severe health 78 outcomes, but is appropriate to study the immune response to mild SARS-CoV-2 79 infections. 80  Nearly 1300 blood samples were collected; furthermore, only 13 out of 304 (4.3%) 81 participants withdrew before attending all three initial visits, and 4 out of 198 82 (2.0%) participants who remained eligible in the extension phase withdrew before 83 attending the fifth visit. A series of at least 5 blood samples drawn over 48 weeks 84 is therefore available for most participants. 85 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 6

86 During the SARS-CoV-2 pandemic, workers with client-facing duties were considered 87 to be at greater risk of infection than those who worked remotely. 1-3 Although most 88 studies have focused on healthcare workers (HCW), 4 5 many non-HCWs were also 89 considered to be at risk due to their occupational exposure.6 7 90 Workers in the food and retail industry are an understudied occupational group that 91 may have been at greater risk of contracting SARS-CoV-2. These workers often have 92 below-average incomes, face precarious employment conditions and lack benefit 93 packages to cover health-related absenteeism. At the beginning of the pandemic, these 94 workers often lacked the training and the access to protective equipment used by HCWs 95 to reduce exposure. 8 However, risk is likely to vary from one sector to another. For 96 example, grocery stores were considered an essential service and therefore remained 97 open throughout the pandemic, with public health measures (e.g., mask wearing) being 98 enforced and generally well respected. In contrast, restaurants and bars were 99 intermittently opened and closed by health authorities over the same period, and public 100 health measures were more difficult to enforce due to the intrinsically social nature of 101 these businesses and their main purpose — the consumption of food and drink — that 102 precluded continuous mask wearing. 103 Compelling evidence now confirms that the risk of occupational exposure is high for 104 these workers. In a serological survey conducted in New York City prior to the approval of 105 the first COVID-19 vaccine, the seroprevalence of anti-spike antibodies was higher among 106 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 7 grocery store and restaurant workers than in most subgroups of HCWs. 1 In another 107 serosurvey conducted in Switzerland, kitchen staff and grocery store workers exhibited 108 an above-average seroprevalence compared to other essential workers. 9 In the 109 Netherlands, individuals working in the hospitality sector were more likely to have a 110 positive PCR test result than those working in non-close-contact occupations. 10 In Japan, 111 restaurants and bars were the second most common setting of SARS-CoV-2 outbreaks 112 after healthcare facilities.7 11 113 To date, no thorough investigation of SARS-CoV-2 exposure has been conducted 114 among Canadian workers in grocery stores, hardware stores, bars or restaurants.12 13 Such 115 an investigation could help better prepare health authorities when implementing future 116 measures, including the designation of priority groups for vaccination, and mandatory 117 lockdowns in these sectors. Accordingly, we set up a longitudinal cohort that investigated 118 the incidence of COVID-19 and the humoral and cellular immunity (innate and acquired) 119 to SARS-CoV-2 in these workers. This article describes the experimental design of the 120 project and the cohort of participants. 121 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 8 COHORT DESCRIPTION 122 Participants and setting 123 Eligibility criteria included the following: (1) providing informed consent; (2) age ≥18 124 years; (3) working either on a full-time or part-time basis in a grocery store, hardware 125 store, bar or restaurant located in the administrative regions of Capitale-Nationale and 126 Chaudière-Appalaches that include and surround the area of Québec City, Canada; (4) 127 having a public-facing role in daily work-related activities; (5) having worked ≥20 full days 128 between February 1st, 2020 and the first visit; and (6) having no history of hospitalization 129 due to COVID-19. 130 Participants were recruited using a variety of strategies: 1) an online recruitment 131 campaign conducted by a student-run communication agency; 2) email invitations to 132 members of partner union organizations - Confédération des syndicats nationaux (CSN) 133 and to sectoral organizations of hardware store workers - Association Québécoise de la 134 quincaillerie et des matériaux de construction (AQMAT); and 3) email information to all 135 students and employees at Université Laval and the Centre Hospitalier Universitaire de 136 Québec in order to publicize the study. 137 Design and procedures 138 The study was initially designed as a prospective cohort study with three sampling 139 visits, each separated by 12±2 weeks, between April 20 th, 2021 and October 3rd, 2022. In 140 response to the emergence of Omicron, an extension of two additional visits was 141 proposed to the participants who were still eligible for recruitment. An additional COVID-142 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 9 19 visit (VCoV) was also planned shortly after the occurrence of any SARS-CoV-2 infection 143 during the study period (Figure 1). 144 At the first visit (“V1”), participants signed an informed consent form and were then 145 interviewed by trained nurses to obtain information on demographic, socioeconomic, 146 behavioral, clinical and occupational variables (Table 1). The questionnaires were adapted 147 from those suggested by our funding body, the COVID Immunity Task Force (CITF).14 At or 148 after the third visit (“V3”), eligible participants received information about the extension 149 of the study and signed a new informed consent if they were interested in participating. 150 At the second (“V2”), third (“V3”), fourth (“V4”) and fifth visits (“V5”), participants 151 completed an abridged version of the V1 questionnaire that focused the COVID-19 152 vaccines that they received and SARS-CoV-2 symptoms, diagnosis, exposure and 153 associated risk factors (Table 1). Blood was drawn to study humoral immunity (i.e., at V1 154 to V5) and cellular immunity (i.e., at V1, V3, and V5) to SARS-CoV-2. Additional PCR tests 155 were carried out at V4 and V5 to detect asymptomatic carriers. 156 The VCoV visits took place at a median time of 15 days (10 to 42 days) after the onset 157 of symptoms. Blood was drawn to study humoral and cellular immunity and a 158 questionnaire focusing exclusively on SARS-CoV-2 diagnosis and symptoms was 159 completed at that visit. 160 Study exposures and follow-up 161 The two main exposures of the study were SARS-CoV-2 infection, defined as a 162 positive test result for SARS-CoV-2 (PCR or antigen detection), and participants’ 163 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 10 vaccination status. Participants were asked about possible or confirmed SARS-CoV-2 164 infection (i.e., symptoms, diagnostic test, test date and test result) and their SARS-CoV-2 165 vaccination history (i.e., number of doses, date of vaccination, type of vaccine) since the 166 beginning of the pandemic at V1, and since the last visit (at V2 to V5). Positive SARS-CoV-167 2 test results were therefore captured from the beginning of the pandemic until the 168 earliest among the last visit, withdrawal from the study, or loss of eligibility. 169 Study outcome 170 The primary outcomes were vaccine- and infection-induced immunity. We also 171 explored the humoral immunity, using different techniques and antigens, and the cellular 172 immunity (innate and acquired). 173 Confidentiality and data storage 174 This study was approved by the « Comité d'éthique de la recherche du CHU de Québec 175 – Université Laval » (registration number 2021-5744). A unique, anonymized identifier 176 was assigned to each participant and used to store the data and the samples. The samples 177 will be stored for up to 10 years, and the data for at least 15 years. 178 Patient and public involvement statement 179 No public stakeholders were involved in establishing and designing this cohort. 180 Participant characteristics 181 Overall, 304 individuals were initially recruited to attend the three first visits from 182 April 20 th, 2021 to May 9 th, 2022. The cohort included 149 (49.0%) restaurant/bar 183 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 11 workers, 112 (36.8%) grocery store workers, and 43 (14.1%) hardware store workers. 184 With the emergence of Omicron, 198 participants who at the time of ethic approval were 185 still within the recruiting window, were included for two additional visits (12±4 weeks 186 apart) between March 15 th, 2022 and October 3 rd, 2022. Only 13 out of 304 (4.3%) 187 withdrew before V3, and only 4 more out of 198 (2.0%) withdrew at V5, resulting in a 188 series of at least 5 blood samples drawn over 48 weeks for most participants. 189 On average, participants were aged 41.3 years in the overall cohort (Table 2). 190 Specifically, restaurants/bar workers were on average 37.2 years old, grocery store 191 workers 44.2 and hardware store workers 48.2. Female participants represented 57.9% 192 of the cohort. Overall, 96.7% self-identified as White, 1.6% as Asian, 1.0% as Latino 193 American, and 0.7% as Black. The low proportion of racial minorities (i.e., 3.3%) is 194 consistent with the size of the visible minority population in the Québec City metropolitan 195 area (i.e., 4.9% according to census).15 Levels of education varied: 39.4% reported having 196 a high school diploma or vocational certificate, 33.2% a higher education certificate and 197 22.7% at least a university degree. In total, 76.0% of participants resided in the Capitale-198 Nationale administrative region, the remainder residing in the Chaudière-Appalaches 199 administrative region (Table 3). Most (i.e., 62.2%) lived alone or with one other person, 200 23.0% lived with children (<18 years), 15.5% with HCWs and 7.6% with teachers or 201 kindergarten workers. These distributions were similar within each occupational group. 202 According to body mass index (BMI), 41.1% of the participants had a healthy weight 203 (i.e., BMI=18.5 to 24.9 kg/m 2), 27.0% were overweight (BMI=25.0 to 29.9 kg/m 2) and 204 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 12 30.6% were obese (BMI ≥30 kg/m 2) (Table 4). Only 1.3% were considered underweight 205 (BMI <18.5 kg/m2). Cigarette (i.e., tobacco) use was reported by 17.4% of the participants 206 and e-cigarette use by 7.9%. About half (i.e., 47.7%) of participants reported having at 207 least one comorbidity. Hardware store workers had more comorbidities, probably 208 because they were slightly older (on average). Overall, 17.1% reported usually receiving 209 an annual influenza vaccine (13.8% in the year prior to the first visit). 210 Approximately half (i.e., 53.6%) of the participants reported working on average 211 more than 30 hours per week (Table 5). Most (88.8%) had attended at least one gathering 212 of 10 or more persons during the study period, and 40.8% reported attending more than 213 10 such gatherings. The predominant mode of transportation was by car (88%), followed 214 by bus (12.8%) and walking (9.2%). Traveling outside the province of Québec was reported 215 by 47.0% of the participants, with 25.7% travelling within Canada, 14.8% to the United 216 States, and 27.6% elsewhere. The distribution of participants in each occupational group 217 was similar for the workplace region, mode of transportation and travelling, but differed 218 for the weekly hours worked and the number of gatherings attended. 219 Overall, 98.7% of the participants reported wearing a mask at work, indicating 220 excellent adherence to this measure (Table 6). Other measures, such as handwashing 221 (98.4%), distancing (70.1%) and the use of Plexiglas dividers (77.3%) were also frequent. 222 The use of gloves (6.9%) and face-shields (10.5%), which were not extensively promoted 223 by the regional public health authorities, were less frequent. Outside work, all 224 participants reported wearing a mask in public (100.0%); most avoided usual salutations 225 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 13 (85.2%), practiced social distancing (84.2%) and avoided contact with vulnerable persons 226 (83.6%) and crowded places (76.6%). Most participants reported washing their hands 227 when dirty (96.7%), after using the bathroom (97.7%), when arriving at (92.1%) and 228 leaving the workplace (71.7%), before eating (87.8%) and after handling trash (79.6%). In 229 general, adherence to these measures was consistently lower among restaurant and bar 230 workers, possibly because of the nature of their work or their younger age (on average). 231 FINDINGS TO DATE 232 SARS-CoV-2 infection 233 Overall, 168 positive tests were reported in 121 participants throughout the study 234 period (Table 7). Of these tests, 117 were the first to yield a positive result as reported by 235 the participants; 40 (performed by 37 participants) were additional tests with 32 236 considered follow-up tests (90 days since previous test). The remaining 11 (including four first-ever 238 positives) were PCR tests performed as per protocol at V4 and V5 among asymptomatic 239 or pre-symptomatic individuals. In addition, 29 participants reported 31 suspected 240 COVID-19 infections based on their symptoms, although not confirmed by a PCR or 241 antigen detection test. 242 SARS-CoV-2-related symptoms 243 Among the 117 participants with a first confirmed SARS-CoV-2 infection, 94.9% 244 reported ≥1 symptom at the time of testing (Table 8). Each individual symptom was 245 experienced by ≥47.9% of participants, except for diarrhea (13.7%) and loss of smell or 246 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 14 taste (22.2%), a pattern consistent with prior studies.16-18 These distributions were similar 247 within each occupational group. 248 Vaccination for SARS-CoV-2 249 The participants were vaccinated according to local government recommendations 250 with the vaccines approved by Canadian health authorities. The COVID-19 vaccines 251 available were monovalent Comirnaty (Pfizer-BioNTech), Spikevax (Moderna) and 252 Vaxzevria (AstraZeneca), which each required two doses to complete the primary series. 253 Hardware store workers were the most highly vaccinated occupational group, 100% of 254 them having received two doses by February 2022 (Figure 2). By the end of the study (i.e., 255 last visit between May 10 th, 2022 and October 3 rd, 2022), nearly 70% of all participants 256 had received at least one booster dose. 257 At the time of testing positive, 17.9% of the participants had received no vaccine 258 dose, 0.9% had received a single dose of vaccine, 38.4% had received two, 39.3% had 259 received three, and 3.4% had received four (Table 9). In participants who tested positive 260 and had received at least two vaccine doses, all infections occurred after the 4 th wave, 261 when the Omicron variant was predominant. 262 Blood sample bank to study infection-induced, vaccine-induced, and hybrid immunity 263 Overall, 1299 blood samples were collected, including 304 (23.4%) at V1, 297 (22.9%) 264 at V2, 291 (22.4%) at V3, 198 (15.2%) at V4, 194 (15.0%) at V5, and 15 (1.2%) at additional 265 visits (i.e., VCoV). In total, 69.2% of the blood samples were drawn from vaccinated 266 participants with no known history of SARS-CoV-2 infection, 23.1% from vaccinated and 267 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 15 previously infected participants, 6.9% from unvaccinated participants with no known 268 history of SARS-CoV-2 infection, and 11 (0.8%) from previously infected and unvaccinated 269 participants (Table 10). 270 STRENGTHS AND LIMITATIONS 271 We set up a cohort of 304 participants to conduct a longitudinal study of COVID-19 272 immunity among food and retail workers who lived and worked within the greater 273 Québec City metropolitan region. The participants provided information on a wide range 274 of demographic, socioeconomic, behavioral, clinical and occupational variables. The study 275 covered seven waves of COVID-19 infection, including those dominated by the Alpha, 276 Delta, and Omicron variants, thus capturing a relatively large number of epidemiological 277 periods and infections. In addition, the blood samples were collected at each scheduled 278 visit regardless of participants’ infection or vaccination history, thus enabling the study of 279 infection-induced, vaccine-induced and hybrid immunity in this extensively characterized 280 cohort. Moreover, few participants withdrew from the study before the end of the initial 281 (i.e., V1-V3) and extension phases (i.e., V4-V5), resulting in complete series of at least 5 282 samples for most participants. 283 A total of 117 first (ever) COVID-19 infections were reported, and most occurred 284 between December 5 th, 2021 and October 3 rd, 2022, consistent with the emergence of 285 the highly contagious Omicron variant. In the present study, vaccine coverage was high: 286 by the time Omicron had emerged, nearly 95% of the participants had already received 287 two vaccine doses (primary series). This high rate of vaccination may be because retail 288 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 16 workers considered themselves at higher risk of SARS-CoV-2 exposure than the general 289 population and were thus more willing to get vaccinated and reduce their risk of infection. 290 Some limitations must be considered when interpreting our results. Per the study 291 design, none of the participants had previously experienced a severe COVID-19 illness 292 that required hospitalization. Therefore, the cohort may not be used to study the immune 293 response that leads to severe health outcomes, but is appropriate to study the immune 294 response to mild COVID-19 illness. Furthermore, our study was not designed to infer a 295 causal relationship between a worker’s occupational sector and the risk of SARS-CoV-2 296 infection, such that confounders probably explain part of the differences among 297 occupational groups. For example, relative to hardware store workers, grocery store 298 workers were younger (i.e., mean age: 44.2 vs. 48.2 years), included more overweight or 299 obese participants (i.e., 72.3% vs. 46.5%), and lived in more crowded households (i.e., 300 proportion with ≥3 residents: 49.1% vs. 25.6%), which may have predisposed them to 301 infection. Another limitation is that the cohort may have been subject to a sampling bias 302 as there may be less vaccine hesitancy among people willing to participate in a scientific 303 study. Hence, the study participants may not be representative of the overall population 304 of workers in these sectors. This is suggested by the 5% to 7% higher vaccination coverage 305 for the second dose as of fall 2021 compared to the general population of the province of 306 Québec. 307 The low proportion of racial minorities (i.e., 3.0%) also limits the use of this cohort to 308 study racial determinants of immunity to SARS-CoV-2. In addition, participant responses 309 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 17 may have been affected by a memory bias, particularly for those whose last infection 310 occurred months before V1. Moreover, few samples were drawn from unvaccinated and 311 previously infected participants, so that the cohort may be of limited use to study 312 immunity induced by infection alone. The high vaccination coverage also made it 313 impossible to assess the impact of vaccination on the risk of infection, since, at any given 314 time, most participants had been vaccinated. Lastly, our study may have underestimated 315 the incidence of SARS-CoV-2 infection since most infections occurred during the Omicron 316 wave, when access to PCR-based screening was limited in Québec (and only less sensitive 317 antigen detection tests were available). The serology data of these samples will shed light 318 on this question. 319 COLLABORATION 320 More detailed, participant-level information is publicly available on an online platform 321 developed by Maelstrom Research. 19 Researchers with other enquiries or collaboration 322 proposals may contact Sylvie Trottier — the principal investigator in charge of setting up 323 the cohort — at [email protected]. Data on participants’ immune 324 responses to SARS-CoV-2 infection and vaccination will be shared through peer-reviewed 325 publications. 326 Funding: This project is being supported by funding from the Public Health Agency of 327 Canada, through the Vaccine Surveillance Reference group and the COVID-19 Immunity 328 Task Force (grant number: 2021-HQ-000134). 329 Competing interests statement: Nothing to declare. 330 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 18 Data sharing statement: All participant-level information is publicly available on an online 331 platform developed by Maelstrom Research. 332 Author contributions: The co-principal investigators of the study were DB and ST, who 333 conceived the study and led the proposal. Protocol design and development team: J.N.P., 334 C.G., J.-F.M., M.B., D.B., and S.T. K.S. constructed the data base, coordinated the study 335 and contributed to the data analysis. M.T. did the maintenance and the development of 336 the data base, the cleaning of the data and contributed to data analysis. N.B. oversaw 337 the data analysis. S.R. participated to data analysis and wrote the first draft of the 338 manuscript. All authors critically reviewed and approved the final manuscript. 339

The authors thank the participants and all the staff involved in 340 planning and preparation of this study. David Simonyan helped with statistical analysis. A 341 special thank you to our partners: CSN Federation of Commerce and AQMAT. 342 ORCID iDs 343 Sylvie Trottier https://orcid.org/0000-0002-3986-5146 344 Denis Boudreau https://orcid.org/ 0000-0001-5152-2464 345 Caroline Gilbert https://orcid.org/0000-0003-2722-1180 346 Jean-Francois Masson https://orcid.org/0000-0002-0101-0468 347 Mariana Baz https://orcid.org/0000-0002-1230-0735 348 Joelle Pelletier https://orcid.org/0000-0002-2934-6940 349 Mathieu Thériault https://orcid.org/0000-0002-7293-7623 350 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 19

351 1. Pathela P, Crawley A, Weiss D, et al. Seroprevalence of Severe Acute Respiratory 352 Syndrome Coronavirus 2 Following the Largest Initial Epidemic Wave in the 353 United States: Findings From New York City, 13 May to 21 July 2020. J Infect Dis 354 2021;224(2):196-206. doi: 10.1093/infdis/jiab200 355 2. Feehan AK, Velasco C, Fort D, et al. Racial and Workplace Disparities in 356 Seroprevalence of SARS-CoV-2, Baton Rouge, Louisiana, USA. Emerg Infect Dis 357 2021;27(1) doi: 10.3201/eid2701.203808 [published Online First: 20201110] 358 3. Sim MR. The COVID-19 pandemic: major risks to healthcare and other workers on the 359 front line. Occup Environ Med 2020;77(5):281-82. doi: 10.1136/oemed-2020-360 106567 [published Online First: 20200401] 361 4. Gomez-Ochoa SA, Franco OH, Rojas LZ, et al. COVID-19 in Health-Care Workers: A 362 Living Systematic Review and Meta-Analysis of Prevalence, Risk Factors, Clinical 363 Characteristics, and Outcomes. Am J Epidemiol 2021;190(1):161-75. doi: 364 10.1093/aje/kwaa191 365 5. Gholami M, Fawad I, Shadan S, et al. COVID-19 and healthcare workers: A systematic 366 review and meta-analysis. Int J Infect Dis 2021;104:335-46. doi: 367 10.1016/j.ijid.2021.01.013 [published Online First: 20210111] 368 6. Boucher E, Cao C, D'Mello S, et al. Occupation and SARS-CoV-2 seroprevalence 369 studies: a systematic review. BMJ Open 2023;13(2):e063771. doi: 370 10.1136/bmjopen-2022-063771 [published Online First: 20230228] 371 7. Vachon MS, Demmer RT, Yendell S, et al. SARS-CoV-2 Seroprevalence Survey in 372 Grocery Store Workers-Minnesota, 2020-2021. Int J Environ Res Public Health 373 2022;19(6) doi: 10.3390/ijerph19063501 [published Online First: 20220316] 374 8. Parks CA, Nugent NB, Fleischhacker SE, et al. Food System Workers are the 375 Unexpected but Under Protected COVID Heroes. J Nutr 2020;150(8):2006-08. doi: 376 10.1093/jn/nxaa173 377 9. Stringhini S, Zaballa ME, Pullen N, et al. Large variation in anti-SARS-CoV-2 antibody 378 prevalence among essential workers in Geneva, Switzerland. Nat Commun 379 2021;12(1):3455. doi: 10.1038/s41467-021-23796-4 [published Online First: 380 20210608] 381 10. de Gier B, de Oliveira Bressane Lima P, van Gaalen RD, et al. Occupation- and age-382 associated risk of SARS-CoV-2 test positivity, the Netherlands, June to October 383 2020. Euro Surveill 2020;25(50) doi: 10.2807/1560-7917.ES.2020.25.50.2001884 384 11. Furuse Y, Sando E, Tsuchiya N, et al. Clusters of Coronavirus Disease in Communities, 385 Japan, January-April 2020. Emerg Infect Dis 2020;26(9) doi: 386 10.3201/eid2609.202272 [published Online First: 20200610] 387 12. Decarie Y, Michaud PC. Counting the Dead: COVID-19 and Mortality in Quebec and 388 British Columbia During the First Wave. Can Stud Popul 2021;48(2-3):139-64. 389 doi: 10.1007/s42650-021-00053-z [published Online First: 20210917] 390 13. Shim E. Regional Variability in COVID-19 Case Fatality Rate in Canada, February-391 December 2020. Int J Environ Res Public Health 2021;18(4) doi: 392 10.3390/ijerph18041839 [published Online First: 20210214] 393 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 20 14. COVID-19 Immunity Task Force standardized core survey data elements [Available 394 from: https://www.covid19immunitytaskforce.ca/covid-19-immunity-task-force-395 releases-standardized-core-survey-data-elements/ accessed February 15th, 2023 396 2023. 397 15. Census profile, 2016 census, Québec City [Available from: 398 https://www12.statcan.gc.ca/census-recensement/2016/dp-399 pd/prof/details/page.cfm?Lang=E&Geo1=CSD&Code1=2423027&Geo2=CD&C400 ode2=2423&Data=Count&SearchText=quebec&SearchType=Begins&SearchPR401 =01&B1=All&TABID=1 accessed February 16th, 2023. 402 16. Zhu J, Ji P, Pang J, et al. Clinical characteristics of 3062 COVID-19 patients: A meta-403 analysis. J Med Virol 2020;92(10):1902-14. doi: 10.1002/jmv.25884 [published 404 Online First: 20200624] 405 17. Burke RM, Killerby ME, Newton S, et al. Symptom Profiles of a Convenience Sample 406 of Patients with COVID-19 - United States, January-April 2020. MMWR Morb 407 Mortal Wkly Rep 2020;69(28):904-08. doi: 10.15585/mmwr.mm6928a2 [published 408 Online First: 20200717] 409 18. Government C. COVID-19 signs, symptoms and severity of disease: A clinician guide. 410 [Available from: https://www.canada.ca/en/public-health/services/diseases/2019-411 novel-coronavirus-infection/guidance-documents/signs-symptoms-severity.html. 412 19. Boudreau DT, Sylvie; Masson, Jean-François; Baz Etchebarne, Mariana; Gilbert, 413 Caroline; Pelletier, Joëlle;. CISACOV investigators. Cellular Immunity and 414 Seroprevalence of Antibodies against SARS-CoV-2: Characterisation of Three 415 Populations of Food Workers. , 2021-2022. 416 20. Canada Go. COVID-19 vaccination in Canada: Government of Canada; 2023 [updated 417 2023-03-03. Available from: https://health-infobase.canada.ca/covid-418 19/vaccination-coverage/. 419 420 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 21 TABLES AND FIGURES 421 Table 1. Participant visits 422 COVID-19 Visit 1 Visit 2 Visit Visit 4 Visit 5 visit (VCoV) N=304 N=297 N=291 N=198 N=194 N=15 Retention 100% 98% 96% 100% 98% - Eligibility assessment X X Consent form X X Participant characteristics Demographic data X Household data X Clinical data X Occupational data X X X X X X Retrospective questionnaire form COVID-19 positive tests & symptoms X X X X X X Vaccine status X X X X X X Cross-section interventions Humoral immunity blood samples X X X X X X Cellular immunity blood samples X X X X COVID-19 PCR test at visit X X 423 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 22 Table 2. Detailed demographics of study participants at the first visit 424 Overall study Restaurant/bar Grocery store Hardware store population workers workers workers Total COVID-191 Total COVID-191 Total COVID-191 Total COVID-191 (N=304) (N=117) (N=149) (N=62) (N=112) (N=42) (N=43) (N=13) Age (years), Mean±SD 41.3 ± 15.9 39.6 ± 14.6 37.2 ± 14.8 36.2 ± 14.3 44.2 ± 15.3 42.4 ± 14.1 48.2 ± 17.3 46.5 ± 14.6 Age groups, N (%) 18-59 257 (84.5%) 106 (90.6%) 135 (90.6%) 57 (91.9%) 93 (83.0%) 37 (88.1%) 29 (67.4%) 12 (92.3%) 60-75 47 (15.5%) 11 (9.4%) 14 (9.4%) 5 (8.1%) 19 (17.0%) 5 (11.9%) 14 (32.6%) 1 (7.7%) Sex, N (%) Female 176 (57.9%) 72 (61.5%) 95 (63.8%) 39 (62.9%) 57 (50.9%) 25 (59.5%) 24 (55.8%) 8 (61.5%) Male 128 (42.1%) 45 (38.5%) 54 (36.2%) 23 (37.1%) 55 (49.1%) 17 (40.5%) 19 (44.2%) 5 (38.5%) Race/ethnicity,2 N (%) White 294 (96.7%) 114 (97.4%) 142 (95.3%) 61 (98.4%) 109 (97.3%) 40 (95.2%) 43 (100.0%) 13 (100.0%) Asian 5 (1.6%) 0 (0.0%) 4 (2.7%) 0 (0.0%) 1 (0.9%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Black 2 (0.7%) 2 (1.7%) 0 (0.0%) 0 (0.0%) 2 (1.8%) 2 (4.8%) 0 (0.0%) 0 (0.0%) Latino American 3 (1.0%) 1 (0.9%) 3 (2.0%) 1 (1.6%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Educational attainment, N (%) Less than high school 14 (4.6%) 3 (2.6%) 4 (2.7%) 2 (3.2%) 9 (8.0%) 0 (0.0%) 1 (2.3%) 1 (7.7%) High school 79 (26.0%) 27 (23.1%) 42 (28.2%) 14 (22.6%) 23 (20.5%) 7 (16.7%) 14 (32.6%) 6 (46.2%) Professional certificate 41 (13.5%) 13 (11.1%) 19 (12.8%) 6 (9.7%) 16 (14.3%) 6 (14.3%) 6 (14.0%) 1 (7.7%) CEGEP and college certificate 101 (33.2%) 45 (38.5%) 51 (34.2%) 26 (41.9%) 35 (31.3%) 14 (33.3%) 15 (34.9%) 5 (38.5%) University baccalaureate 54 (17.8%) 22 (18.8%) 28 (18.8%) 12 (19.4%) 21 (18.8%) 10 (23.8%) 5 (11.6%) 0 (0.0%) Graduate studies 15 (4.9%) 7 (6.0%) 5 (3.4%) 2 (3.2%) 8 (7.1%) 5 (11.9%) 2 (4.7%) 0 (0.0%) Abbreviations: CEGEP = General and professional teaching college 425 Notes: 426 1. Subset of participants who contracted COVID-19 at least once during the study period. 427 2. Self-reported by study participants 428 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 23 Table 3. Household characteristics of the participants at the first visit 429 Overall study Restaurant/bar Grocery store Hardware store population workers workers workers Total COVID-19 1 Total COVID-19 1 Total COVID-19 1 Total COVID-19 1 (N=304) (N=117) (N=149) (N=62) (N=112) (N=42) (N=43) (N=13) Region, N (%) 112 (75,2%) 45 (72,6%) 86 (76,8%) 35 (83,3%) 33 (76,7%) 11 (84,6%) 37 (24,8%) 17 (27,4%) 26 (23,2%) 7 (16,7%) 10 (23,3%) 2 (15,4%) 100 (67,1%) 48 (77,4%) 57 (50,9%) 19 (45,2%) 32 (74,4%) 9 (69,2%) 36 (24,2%) 10 (16,1%) 48 (42,9%) 19 (45,2%) 9 (20,9%) 3 (23,1%) 13 (8,7%) 4 (6,5%) 7 (6,3%) 4 (9,5%) 2 (4,7%) 1 (7,7%) 35 (23,5%) 12 (19,4%) 29 (25,9%) 14 (33,3%) 6 (14,0%) 3 (23,1%) 23 (15,4%) 9 (14,5%) 18 (16,1%) 5 (11,9%) 6 (14,0%) 3 (23,1%) 5 (3,4%) 1 (1,6%) 5 (4,5%) 2 (4,8%) 0 (0,0%) 0 (0,0%) 9 (6,0%) 4 (6,5%) 4 (3,6%) 3 (7,1%) 0 (0,0%) 0 (0,0%) 90 (60,4%) 41 (66,1%) 49 (43,8%) 19 (45,2%) 19 (44,2%) 6 (46,2%) 59 (39,6%) 21 (33,9%) 63 (56,3%) 23 (54,8%) 24 (55,8%) 7 (53,8%) Capitale-Nationale 231 (76,0%) 91 (77,8%) Chaudière-Appalaches 73 (24,0%) 26 (22,2%) Household size, N (%) 1-2 residents 189 (62,2%) 76 (65,0%) 3-4 residents 93 (30,6%) 32 (27,4%) ≥5 residents 22 (7,2%) 9 (7,7%) Co-residents, N (%) Underage (<18) 70 (23,0%) 29 (24,8%) Healthcare worker 47 (15,5%) 17 (14,5%) Kindergarden worker 10 (3,3%) 3 (2,6%) Teacher 13 (4,3%) 7 (6,0%) Household bedroom, N (%) 0-2 158 (52,0%) 66 (56,4%) ≥3 146 (48,0%) 51 (43,6%) Notes: 430 1. Subset of participants who contracted COVID-19 had least once during the study period. 431 432 433 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 24 Table 4. Clinical characteristics of study participants at the first visit 434 Overall study Restaurant/bar Grocery store Hardware store population workers workers workers Total COVID-191 Total COVID-191 Total COVID-191 Total COVID-191 (N=304) (N=117) (N=149) (N=62) (N=112) (N=42) (N=43) (N=13) BMI scores, Mean±SD 27,3 ± 6,1 27,5 ± 6,4 27,0 ± 6,9 26,83 ± 6,7 28,1 ± 5,3 27,94 ± 5,2 26,3 ± 5,3 27,51 ± 6,6 BMI categories, N (%) <18.5 (underweight) 4 (1,3%) 1 (0,9%) 2 (1,3%) 1 (1,6%) 2 (1,8%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 18.5-24.9 (healthy weight) 125 (41,1%) 49 (41,9%) 73 (49,0%) 30 (48,4%) 29 (25,9%) 13 (31,0%) 23 (53,5%) 6 (46,2%) 25.0-29.9 (overweight) 82 (27,0%) 23 (19,7%) 31 (20,8%) 10 (16,1%) 39 (34,8%) 9 (21,4%) 12 (27,9%) 4 (30,8%) ≥30 (obesity) 93 (30,6%) 44 (37,6%) 43 (28,9%) 21 (33,9%) 42 (37,5%) 20 (47,6%) 8 (18,6%) 3 (23,1%) Smoking, N (%) Cigarette user 53 (17,4%) 22 (18,8%) 33 (22,1%) 14 (22,6%) 16 (14,3%) 6 (14,3%) 4 (9,3%) 2 (15,4%) E-cigarette user 24 (7,9%) 8 (6,8%) 20 (13,4%) 6 (9,7%) 3 (2,7%) 2 (4,8%) 1 (2,3%) 0 (0,0%) Comorbidities,2 N (%) Hypertension 39 (12,8%) 11 (9,4%) 13 (8,7%) 4 (6,5%) 18 (16,1%) 6 (14,3%) 8 (18,6%) 1 (7,7%) Chronic pulmonary disease 33 (10,9%) 17 (14,5%) 13 (8,7%) 6 (9,7%) 14 (12,5%) 10 (23,8%) 6 (14,0%) 1 (7,7%) Diabetes mellitus 18 (5,9%) 5 (4,3%) 4 (2,7%) 1 (1,6%) 11 (9,8%) 3 (7,1%) 3 (7,0%) 1 (7,7%) Hypothyroidism 16 (5,3%) 4 (3,4%) 6 (4,0%) 2 (3,2%) 6 (5,4%) 1 (2,4%) 5 (11,6%) 1 (7,7%) Cancer 10 (3,3%) 3 (2,6%) 4 (2,7%) 1 (1,6%) 5 (4,5%) 2 (4,8%) 1 (2,3%) 0 (0,0%) Cardiovascular disease 8 (2,6%) 2 (1,7%) 2 (1,3%) 0 (0,0%) 3 (2,7%) 1 (2,4%) 3 (7,0%) 1 (7,7%) Immune deficiency 7 (2,3%) 5 (4,3%) 3 (2,0%) 3 (4,8%) 1 (0,9%) 0 (0,0%) 3 (7,0%) 2 (15,4%) Chronic neurological disorder 6 (2,0%) 1 (0,9%) 2 (1,3%) 1 (1,6%) 3 (2,7%) 0 (0,0%) 1 (2,3%) 0 (0,0%) Liver disease 2 (0,7%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 1 (0,9%) 0 (0,0%) 1 (2,3%) 0 (0,0%) Blood disorder 1 (0,3%) 1 (0,9%) 1 (0,7%) 1 (1,6%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 0 (0,0%) Obesity3 1 (0%) 0 (0%) 1 (1%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Kidney disease 0 (0,0%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 0 (0,0%) 0 (0,0%) Influenza vaccination, N (%) Usually received 52 (17,1%) 21 (17,9%) 23 (15,4%) 12 (19,4%) 18 (16,1%) 7 (16,7%) 11 (25,6%) 2 (15,4%) Received in the last year 42 (13,8%) 12 (10,3%) 17 (11,4%) 5 (8,1%) 15 (13,4%) 5 (11,9%) 10 (23,3%) 2 (15,4%) Notes: 435 1. Subset of participants who contracted COVID-19 had least once during the study period. 436 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 25 2. Comorbidities related to an increased risk of hospitalisation at the first visit. 437 3. 30.6% of participants had a BMI in the range of obesity, but only one reported to be obese. 438 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 26 Table 5. Occupational and behavioral characteristic of study participants 439 Overall study Restaurant/bar Grocery store Hardware store population workers workers workers Total COVID-19 1 Total COVID-19 1 Total COVID-19 1 Total COVID-19 1 (N=304) (N=117) (N=149) (N=62) (N=112) (N=42) (N=43) (N=13) Workplace region,2 N (%) Capitale-Nationale 240 (78,9%) 96 (82,1%) 123 (82,6%) 49 (79,0%) 85 (75,9%) 36 (85,7%) 32 (74,4%) 11 (84,6%) Chaudière-Appalaches 64 (21,1%) 19 (16,2%) 26 (17,4%) 12 (19,4%) 27 (24,1%) 6 (14,3%) 11 (25,6%) 1 (7,7%) Weekly hours worked,3 Mean±SD 27,9 ± 11,5 28,9 ± 11,4 24,7 ± 11,0 24,5 ± 11,0 32,6 ± 10,4 34,2 ± 9,8 26,6 ± 11,9 32,3 ± 10,4 Participants working Full time (≥30) 141 (46,4%) 64 (54,7%) 49 (32,9%) 23 (37,1%) 73 (65,2%) 32 (76,2%) 19 (44,2%) 9 (69,2%) Part time (<30) 163 (53,6%) 53 (45,3%) 100 (67,1%) 39 (62,9%) 39 (34,8%) 10 (23,8%) 24 (55,8%) 4 (30,8%) Gathering of 10+ persons,3 Mean±SD 17,1 ± 27,5 22,5 ± 31,4 23,2 ± 33,9 31,5 ± 38,9 10,4 ± 15,8 13,1 ± 15,9 13,7 ± 22,1 9,9 ± 10,4 Per group, N (%) None 34 (11,2%) 6 (5,1%) 16 (10,7%) 2 (3,2%) 15 (13,4%) 4 (9,5%) 3 (7,0%) 0 (0,0%) 1 to 10 gatherings 146 (48,0%) 47 (40,2%) 57 (38,3%) 16 (25,8%) 62 (55,4%) 22 (52,4%) 27 (62,8%) 9 (69,2%) 11 to 50 gatherings 101 (33,2%) 51 (43,6%) 58 (38,9%) 33 (53,2%) 32 (28,6%) 14 (33,3%) 11 (25,6%) 4 (30,8%) >50 gatherings 23 (7,6%) 13 (11,1%) 18 (12,1%) 11 (17,7%) 3 (2,7%) 2 (4,8%) 2 (4,7%) 0 (0,0%) Transportation,2 N (%) Car 266 (87,5%) 103 (88,0%) 127 (85,2%) 54 (87,1%) 98 (87,5%) 37 (88,1%) 41 (95,3%) 12 (92,3%) Carpooling 2 (0,7%) 2 (1,7%) 1 (0,7%) 1 (1,6%) 1 (0,9%) 1 (2,4%) 0 (0,0%) 0 (0,0%) Bus 39 (12,8%) 16 (13,7%) 23 (15,4%) 9 (14,5%) 10 (8,9%) 5 (11,9%) 6 (14,0%) 2 (15,4%) Bicycle 13 (4,3%) 6 (5,1%) 7 (4,7%) 3 (4,8%) 4 (3,6%) 2 (4,8%) 2 (4,7%) 1 (7,7%) Walking 28 (9,2%) 11 (9,4%) 13 (8,7%) 6 (9,7%) 14 (12,5%) 5 (11,9%) 1 (2,3%) 0 (0,0%) Travel,3 N (%) Any destination 143 (47,0%) 70 (59,8%) 79 (53,0%) 40 (64,5%) 48 (42,9%) 25 (59,5%) 16 (37,2%) 5 (38,5%) In Canada 78 (25,7%) 39 (33,3%) 45 (30,2%) 24 (38,7%) 23 (20,5%) 13 (31,0%) 10 (23,3%) 2 (15,4%) To USA 45 (14,8%) 20 (17,1%) 25 (16,8%) 11 (17,7%) 16 (14,3%) 8 (19,0%) 4 (9,3%) 1 (7,7%) Other destination4 84 (27,6%) 45 (38,5%) 47 (31,5%) 25 (40,3%) 28 (25,0%) 16 (38,1%) 9 (20,9%) 4 (30,8%) Notes: 440 1. Subset of participants who contracted COVID-19 had least once during the study period. 441 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 27 2. At the time of the first visit (i.e., week 0). 442 3. During the entire study period. 443 4. Includes travel to Cuba, Ireland, Great-Britain, Luxembourg, Dominican Republic, South Africa, Bahamas, Morocco, Guadeloupe, 444 Panama, Costa Rica, Greece. 445 446 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 28 Table 6. Protective measures taken at work and elsewhere by study participants at first visit 447 Overall study Restaurant/bar Grocery store Hardware store population workers workers workers Total COVID-19 1 Total COVID-19 1 Total COVID-19 1 Total COVID-19 1 (N=304) (N=117) (N=149) (N=62) (N=112) (N=42) (N=43) (N=13) Protection measures at work,2 N (%) Mask 300 (98,7%) 116 (99,1%) 147 (98,7%) 61 (98,4%) 111 (99,1%) 42 (100,0%) 41 (95,3%) 13 (100,0%) Handwashing 299 (98,4%) 114 (97,4%) 145 (97,3%) 59 (95,2%) 111 (99,1%) 42 (100,0%) 43 (100,0%) 13 (100,0%) Plexiglas 235 (77,3%) 88 (75,2%) 96 (64,4%) 39 (62,9%) 98 (87,5%) 37 (88,1%) 41 (95,3%) 12 (92,3%) Social distancing 213 (70,1%) 86 (73,5%) 111 (74,5%) 48 (77,4%) 74 (66,1%) 29 (69,0%) 28 (65,1%) 9 (69,2%) Protective glasses 77 (25,3%) 12 (10,3%) 30 (20,1%) 4 (6,5%) 35 (31,3%) 5 (11,9%) 12 (27,9%) 3 (23,1%) Faceshield 32 (10,5%) 4 (3,4%) 16 (10,7%) 1 (1,6%) 13 (11,6%) 2 (4,8%) 3 (7,0%) 1 (7,7%) Gloves 21 (6,9%) 3 (2,6%) 10 (6,7%) 0 (0,0%) 9 (8,0%) 3 (7,1%) 2 (4,7%) 0 (0,0%) Face Cover 7 (2,3%) 1 (0,9%) 3 (2,0%) 0 (0,0%) 4 (3,6%) 1 (2,4%) 0 - 0 (0,0%) Other3 135 (44,4%) 59 (50,4%) 85 (57,0%) 41 (66,1%) 43 (38,4%) 16 (38,1%) 7 (16,3%) 2 (15,4%) Behavioral protection measures,2 N (%) Mask wearing in public spaces 304 (100,0%) 117 (100,0%) 149 (100,0%) 62 (100,0%) 112 (100,0%) 42 (100,0%) 43 (100,0%) 12 (92,3%) Avoid usual salutations 259 (85,2%) 99 (84,6%) 119 (79,9%) 48 (77,4%) 103 (92,0%) 39 (92,9%) 37 (86,0%) 12 (92,3%) Social distancing 256 (84,2%) 93 (79,5%) 115 (77,2%) 41 (66,1%) 102 (91,1%) 40 (95,2%) 39 (90,7%) 12 (92,3%) Avoid contacts with vulnerable persons 254 (83,6%) 93 (79,5%) 114 (76,5%) 42 (67,7%) 100 (89,3%) 38 (90,5%) 40 (93,0%) 13 (100,0%) Avoid crowded places 233 (76,6%) 81 (69,2%) 102 (68,5%) 35 (56,5%) 94 (83,9%) 34 (81,0%) 37 (86,0%) 12 (92,3%) Quarantine if exposed to COVID-19 126 (41,4%) 70 (59,8%) 65 (43,6%) 36 (58,1%) 46 (41,1%) 25 (59,5%) 15 (34,9%) 9 (69,2%) Pre-emptive isolation 36 (11,8%) 23 (19,7%) 21 (14,1%) 13 (21,0%) 12 (10,7%) 9 (21,4%) 3 (7,0%) 1 (7,7%) Handwashing habits,2 N (%) After using the toilet 297 (97,7%) 115 (98,3%) 147 (98,7%) 60 (96,8%) 109 (97,3%) 42 (100,0%) 41 (95,3%) 13 (100,0%) When dirty 294 (96,7%) 113 (96,6%) 140 (94,0%) 58 (93,5%) 112 (100,0%) 42 (100,0%) 42 (97,7%) 13 (100,0%) When entering workspace 280 (92,1%) 106 (90,6%) 136 (91,3%) 55 (88,7%) 107 (95,5%) 37 (88,1%) 37 (86,0%) 13 (100,0%) Before eating 267 (87,8%) 104 (88,9%) 124 (83,2%) 51 (82,3%) 105 (93,8%) 41 (97,6%) 38 (88,4%) 12 (92,3%) Before & after handling food 246 (80,9%) 96 (82,1%) 127 (85,2%) 55 (88,7%) 93 (83,0%) 32 (76,2%) 26 (60,5%) 9 (69,2%) After handling trash 242 (79,6%) 97 (82,9%) 119 (79,9%) 51 (82,3%) 95 (84,8%) 38 (90,5%) 28 (65,1%) 8 (61,5%) When exiting workspace 218 (71,7%) 81 (69,2%) 104 (69,8%) 41 (66,1%) 87 (77,7%) 32 (76,2%) 27 (62,8%) 8 (61,5%) Other4 50 (16,4%) 11 (9,4%) 29 (19,5%) 4 (6,5%) 12 (10,7%) 2 (4,8%) 9 (20,9%) 5 (38,5%) Notes: 448 1. Subset of participants who contracted COVID-19 had least once during the study period 449 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 29 2. Includes customer registry, QR code, customer limit in store, thorough cleaning of workplace, worker temperature surveillance. 450 3. Includes after touching the cash register, handling money, in between clients. 451 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 30 Table 7. Number of COVID-19 positive tests 452 Overall Bar/ Grocery Hardware cohort Restaurant store store Reported positive COVID-19 tests First occurrence 117 62 42 13 PCR 40 24 13 3 Antigen detection 77 38 29 10 Additional test1 8 5 2 1 PCR 1 1 0 0 Antigen detection 7 4 2 1 Positive PCR test for COVID-19 at V4 or V5 11 6 5 0 First occurrence 4 1 3 0 Additional test <90 days from previous test 4 3 1 0 ≥90 days from previous test 3 2 1 0 Reported symptoms (no positive test) 31 17 12 2 Notes: 453 1. All 90 days or more since a previously positive test. 454 455 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 31 Table 8. COVID-19 symptoms at the first COVID-19 positive test reported by the participant 456 Overall Bar/ Grocery Hardware cohort Restaurant store store (N=117) (N=62) (N=42) (N=13) Asymptomatic infection, N (%) 6 (5,1%) 2 (3,2%) 4 (9,5%) 0 (0,0%) Symptomatic infection, N (%) 111 (94,9%) 60 (96,8%) 38 (90,5%) 13 (100,0%) Runny nose or nasal congestion 76 (65,0%) 39 (62,9%) 26 (61,9%) 11 (84,6%) Cough 74 (63,2%) 41 (66,1%) 25 (59,5%) 8 (61,5%) Headache 71 (60,7%) 36 (58,1%) 26 (61,9%) 9 (69,2%) Sore throat 70 (59,8%) 41 (66,1%) 23 (54,8%) 6 (46,2%) Fever 68 (58,1%) 39 (62,9%) 19 (45,2%) 9 (69,2%) Muscle pain 65 (55,6%) 35 (56,5%) 22 (52,4%) 8 (61,5%) Shortness of breath 56 (47,9%) 27 (43,5%) 21 (50,0%) 8 (61,5%) Loss of sense of smell or taste 26 (22,2%) 11 (17,7%) 10 (23,8%) 5 (38,5%) Diarrhea 16 (13,7%) 7 (11,3%) 8 (19,0%) 1 (7,7%) 457 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 32 Table 9. COVID-19 vaccination status at the first reported COVID-19 positive test 458 Overall Bar/ Grocery Hardware cohort Restaurant store store (N=117) (N=62) (N=42) (N=13) Vaccination status, N (%) No vaccine 21 (17.9%) 11 (17.7%) 10 (23.8%) 0 (0.0%) One dose 1 (0.9%) 0 (0.0%) 1 (2.4%) 0 (0.0%) Two doses 45 (38.5%) 24 (38.7%) 15 (35.7%) 6 (46.2%) Three doses 46 (39.3%) 26 (41.9%) 14 (33.3%) 6 (46.2%) Four doses 4 (3.4%) 1 (1.6%) 2 (4.8%) 1 (7.7%) 459 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 33 Table 10. Blood samples collected by vaccination status 460 Overall cohort Blood samples (N = 1299) Infected Not infected No Vaccine, N (%) 11 (0.8%) 89 (6.9%) Vaccinated, N (%) 300 (23.1%) 899 (69.2%) 1 dose 14 (1.1%) 61 (4.7%) 2 doses 145 (11.2%) 584 (45.0%) 3 doses 122 (9.4%) 231 (17.8%) 4 doses 20 (1.5%) 22 (1.7%) 461 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 34 Figure 1. Timeline of the study illustrating the visits (colored dash lines) and the first occurrences of COVID- 19-positive tests (grey bars) . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint 35 Figure 2. Evolution of the vaccination coverage (Comirnaty [Pfizer-BioNTech], SpikeVax [Moderna] or Vaxzevria [AstraZeneca] vaccine) during the study period (A) in the overall cohort stratified by the number of doses received, and (B) stratified by occupational group. The study data are compared to those of the Public Health Agency of Canada (Qc population) for the whole province.20 . . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.18.23294172doi: medRxiv preprint