Societal COVID-19 epidemic counter measures and activities associated with SARS-CoV-2 infection in an adult unvaccinated population – a case-control study in Denmark, June 2021

Case-control Study; Community acquired infections; COVID-19; Risk factors; SARS-CoV-2, non- pharmaceutical interventions, alcohol Word count: Main text: 3752 words. Abstract 296 words. Contains 2 figures and 6 tables. . 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 (which was not certified by peer review) The copyright holder for this preprint this version posted June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 2 1 Abstract 2 Measures to restrict physical inter-personal contact in the community have been widely implemented 3 during the COVID-19 pandemic. We studied determinants for infection with SARS-CoV-2 with the aim of 4 testing the efficiency of such measures. 5 We conducted a national matched case-control study among unvaccinated persons aged 18-49 years. 6 Cases were selected among those testing positive for SARS-CoV-2 by RT-PCR over a five-day period in 7 June 2021. Controls were selected from the national population register and were individually matched on 8 age, sex and municipality of residence and had not previously tested positive. Cases and controls were 9 interviewed via telephone about contact with other persons and exposures in the community. We included 10 500 cases and 529 controls and determined odds ratios (ORs) and 95% confidence intervals (95%CIs) by 11 conditional logistical regression with adjustment for household size and immigration status. 12 We found having had contact with another individual with a known infection as the main determinant 13 for SARS-CoV-2 infection. Reporting close contact with an infected person who either had or did not have 14 symptoms resulted in ORs of 20 (95%CI:9.8-39) and 8.5 (95%CI 4.5-16) respectively. In contrast, community 15 exposures were generally not associated with disease; several exposures were negatively associated. 16 Exceptions were: attending fitness centers, OR=1.4 (95%CI:1.0-2.0) and consumption of alcohol in 17 restaurants or cafés, OR=2.3 (95%CI:1.3–4.2). 18 For reference, we provide a timeline of non-pharmaceutical interventions in place in Denmark from 19 February 2020 to March 2022. Fitness centers and alcohol consumption were mildly associated with 20 infection, in agreement with findings of our similar study conducted six month earlier (Epidemiology & 21 Infection 2021;150:e9.). Transmission of disease through involvement in community activities appeared to 22 occur only rarely, suggesting that community restrictions in place were efficient. Instead, transmission 23 appeared to primarily take place in a confined space via contact to known persons. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 3 24 Introduction 25 During the COVID-19 pandemic, most countries have made use of widespread restrictions affecting 26 normal social life. The purpose has been to limit physical inter-personal interaction, in order to limit 27 transmission of SARS-CoV-2. A wide range of measures, sometimes referred to as non-pharmaceutical 28 interventions, have been implemented at varying time points throughout the pandemic [1]. In 2020 and 2021, 29 in many countries, a plethora of public health recommendations, and restrictions have been introduced and 30 re-adjusted on a continuous basis. In Denmark, as in many other European countries, restrictions have 31 involved public gatherings, the level of working from home, mandatory use of face masks, and regulations 32 and lockdowns of restaurant/café, bar, nightclubs, sport activities, cultural events and more [2]. However, the 33 impact these societal restrictions have had to reduce SARS-CoV-2 infection have rarely been subject to 34 study and restrictions have commonly been introduced without a defined evidence base; potentially leading 35 to mixed reactions in the populations they are applied to. This has been done for good reasons, obviously in 36 a crisis situation, implementation of measures can often not await the results of scientific studies, 37 nevertheless there is a need for more knowledge about how these regulations function in preventing SARS- 38 CoV-2 infection under real life community settings. 39 By use of a case-control design, researchers have aimed to identify determinants, private and societal, 40 for SARS-CoV-2 infection. Risk factors reported from previous studies include household overcrowding [3, 41 4], work in senior/health care [3, 5], work on-site [3, 4, 6, 7], foreign citizenship [3] and low education [3]. At 42 the societal level, only few studies have been performed. When investigating activities such as use of public 43 transport, frequenting restaurants/other dining spaces or bars, participating in indoor sports activities or 44 buying food in stores, such studies have shown conflicting results [3-6, 8, 9]. 45 Towards the end of 2020, we investigated societal activities associated with SARS-CoV-2 infection in 46 Denmark by use of a case-control study design. This was done in a period where society was partially open 47 with public gathering restrictions and mandatory face mask use indoors, the original wild type (Index) strain 48 of SARS-CoV-2 was the dominant strain in circulation and the COVID-19 vaccine rollout had not yet begun. 49 We found that having had contact, in particular close contact, to another person with a known SARS-CoV-2 50 infection was strongly associated with infection. In contrast, only few community exposures were found to be 51 associated with SARS-CoV-2 infection. They were participation in events where people sang, attending . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 4 52 fitness centers and related to consumption of alcohol in bars. Other community exposures appeared not to 53 be associated with infection e.g. supermarkets, public transport, and restaurants [10]. 54 In June 2021, the COVID-19 situation had changed. The number of infected persons was declining, 55 approximately 35% of the Danish population had received the first vaccination dose and approximately 20% 56 the second. Society was gradually reopening, with now only societal restrictions in place for those individuals 57 who were vaccinated, had recovered from infection or recently tested negative. Moreover, the Alpha SARS- 58 CoV-2 variant was now circulating. In this situation, we again sought to identify societal activities associated 59 with SARS-CoV-2 infection in Denmark. Here we present the results of a second national case-control study 60 of risk factors for infection. For context and reference, we further present an overview of the official 61 restrictions that have been in place in Denmark throughout the COVID-19 epidemic. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 5 62 Methods 63 Officially imposed societal restrictions 64 We mapped public health measures and restrictions introduced in Denmark, in the period from 65 February 2020 to March 2022. We covered measures within the following areas: public gathering restrictions 66 (indoor, outdoor and at home); schools; workplaces; public spaces: grocery shops, non-essential shops, 67 shopping malls, restaurants, bars/nightclubs, indoor cultural events, libraries, church/religious communities, 68 public transport and sport activities. We categorized them into three different levels (open without 69 restrictions, open with restrictions and fully locked down). The information was retrieved from relevant Danish 70 government ministeries and from the national COVID-19 communication partnership (coronasmitte.dk). 71 During the period of the case-control study, societal restrictions were mainly in place for those adults 72 who were unvaccinated, had not previously recovered from SARS-CoV-2 infection and who had not recently 73 tested negative for SARS-CoV-2. This status could be documented by use of a digital ‘corona passport’ 74 (accessible via a smart phone app) first introduced in May 2021, or as a printable PDF with a QR code. The 75 particular requirements for a valid corona passport within the study period was: I) Vaccination: from 14 days 76 to 42 days after the first dose, or after the second dose (mRNA vaccines), or 14 days after dose one with 77 Johnson & Johnson. After vaccination, the corona passport was valid for 8 months [11]. II) Negative official 78 RT-PCR SARS-CoV-2 test: Taken within the past 72 hours [12]. III) Recovered after SARS-CoV-2 infection: 79 Previously infected with COVID-19 documented by positive PCT test, performed at least 14 days and 80 maximum 8 months prior [11]. 81 82 Case-control study design 83 To identify societal activities associated with SARS-CoV-2 infection in Denmark, at a fixed point in time 84 during the pandemic, we conducted a national, individually matched, case-control study. The study methods 85 were largely as those previously described [10], however with minor modifications. Eligible cases were 86 unvaccinated individuals between 18–49 years old, with an address in Denmark, and an RT-PCR confirmed 87 SARS-CoV-2 infection in the period from 8 to 12 June 2021. We listed eligible cases in random order and 88 aimed to include the first 500 cases, who had not been hospitalized or travelled outside of Denmark during . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 6 89 the exposure period. Controls were matched to cases by year of birth, sex (2 levels) and municipality (98 90 levels) and were extracted from the Danish Civil Registration System [13]. Only controls unvaccinated and 91 not previously infected by 12 June 2021 were included. 92 93 Date sources 94 In Denmark, an extensive test system was built during 2020 and the first part of 2021. In addition to 95 the clinical test system, RT-PCR tests were provided for all, without indication, through widely available, free- 96 for-all public test stations. Information on SARS-CoV-2 tests was obtained at person-level format from the 97 Danish Microbiology Database [14-16]. Controls were sampled from in the Danish Civil Registration System 98 from which information on age, sex, vital status, area of residence and country of birth was also obtained 99 [17]. Information on the latter variable was obtained both for participants and their parents. Information on 100 vaccines administered against SARS-CoV-2 in Denmark are registered in the Danish Vaccination Registry 101 [18]. Through this, person level information on vaccinations given, including the date of administration, was 102 obtained. Information from other data sources were linked to by use of the unique civil registry number 103 assigned to all Danish residents [17]. 104 105 Data collection 106 Cases and controls were interviewed via telephone between 15 June and 24 June, 2021, by a sub- 107 contracted private polling institute. At least two attempts were made to call each eligible case and control per 108 day. Controls were sought interviewed after their matched case had been interviewed. We aimed to include 109 one matched control per case. We sampled 10 controls per case, but sampled an additional 10 controls in 110 the instances where none of the first 10 had been reached within two days. 111 Compared to the study performed in November 2020, the period that our questions concerned was 112 shortened from a 14-day to a 6-day period. The 6-day period ran from eight to two days prior to symptom 113 onset (or test date if asymptomatic) for cases and the identical 6-day period for the matched control. We 114 refer to this as the exposure period. Further, the questions related to contact exposures were updated due to 115 changes in the national guideline set by the Danish Health Authority and therefore had a slightly different . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 7 116 wording. Contact exposures included close contact/other contact with a person with known SARS-CoV-2 117 infection, with or without symptoms. The close contact definition was: Exposure to a household member, 118 direct physical contact (for example hugging), unprotected and direct contact with secretions from an 119 infected person, having been within a distance of less than 1 meter for more than 15 minutes, or caring for 120 COVID-19 patients where the prescribed protective equipment had not been used. Other contact was 121 defined as contact with a person with known SARS-CoV-2 infection. In contrast to the first study, we did not 122 include questions on protective behavior and adherence with measures. For further and detailed information 123 on community exposures, please refer to [10]. 124 The study was performed as a national disease surveillance project, registered with the Danish Data 125 Protection Agency (reg no 21/04112) and specifically approved regarding legal, ethical and cyber-security 126 issues. According to Danish legislation, approval from an ethical committee is not needed for medical studies 127 not involving biological material. 128 129 Statistical analyses and power calculation 130 The required sample size was calculated based on an expected bar visit frequency of 10% among 131 controls [10]. With a power of 80%, an alpha-level of 0.05 and an odds ratio to detect at 2, we needed 566 132 participants following standard sample size formulae for unmatched case-control studies [19]. We assumed 133 that 30% of all cases would be infected within the household and the required sample size was then 134 calculated to be 810 (405 cases and 405 controls). We aimed to include a total of 1000 participants. 135 We compared exposures reported by cases with those of controls using conditional logistic regression 136 taking matching into account. For answers to secondary questions where matched analyses were not 137 possible, logistic regression with adjustment for the matching variables was performed. We additionally 138 adjusted for household size and migration background. For analyses concerning community exposures, we 139 excluded cases (and their matched control) who reported to be infected in their household. As a sensitivity 140 analysis, we additional excluded cases and controls (and their respective matches) who reported having 141 been close contact to an infected person during the exposure period. If such persons already during the 142 exposure period would have been aware that they were close contacts, they could have self-isolated as . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 8 143 recommended or modified their behavior and thus been less likely to participate in activities in the 144 community. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 9 145 Results 146 Official COVID-19 counter measures in Denmark 147 The first public health measures were introduced in March 2020. During the following two years, a 148 complex series of public health measures and restrictions were put in place, lifted and/or reintroduced in 149 response to the development of the epidemic. A detailed overview of these is given in Figure 1. 150 For the case-control study period in particular, the following restrictions were imposed: Restaurants, 151 cafes, bars etc. had to close at 11 pm with last servings at 10 pm. Use of face masks was mandatory for 152 those aged 12 years or older, in indoor public spaces, including shops and public transport, except when 153 seated at a table to eat or drink. Nightclubs were closed. At cultural, sport and religious events, a maximum 154 of 500 seated people could gather, facing the same direction. Further, a valid corona passport was required 155 for access to all public spaces, except pharmacies and shops selling foods. The maximum number of people 156 allowed for spontaneous or private gatherings was 50 inside and 100 outdoor [20]. 157 158 Case-control study 159 In the inclusion period 1,565 unvaccinated persons in the age from 18-49, were diagnosed with SARS- 160 CoV-2 and eligible for inclusion. A valid phone number was available for 1148 (72%) cases, 829 were 161 attempted contacted before 500 were included in the study and enrolment ended. A total of 529 matched 162 controls were included in the study (Figure 2). Eligible and included cases were similar regarding age, sex 163 and geographic region of residence. Compared to eligible cases, the included cases were less likely to have 164 migrant background (data not shown). Included cases and controls had similar household sizes, but the 165 groups differed regarding migrant background and number of contacts (Table 1). 166 In total, 80% of cases reported knowing where they had been infected. This was primarily reported to 167 have happened in the household (20%), at the workplace (16%), or among friends or family members (other 168 than the household,16%). Cases further reported education facilities (5.4%), leisure activity (5.2%), other 169 events (2.5%) or ‘other place/exposure’ (13%) as places of likely infection (Table 2). In total, 87% of the 170 cases reported to have experienced symptoms of COVID-19. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 10 171 Overall, 47% of the cases and 8% of the controls had been in contact with a known infected person in 172 the exposure period. Most reported ‘close contact’. ‘Other contact’ with an infected person with symptoms was 173 reported by 2.9% of the cases and 1.1% of the controls, resulting in a matched odds ratio (mOR) estimate of 174 3.3 (95% CI: 1.2-9.2). Close contact with an infected person without symptoms was reported by 15% of cases 175 and 3.1% of controls had, resulting in a mOR of 8.5 (95% CI: 4.5-16). Close contact with an infected person 176 with symptoms was reported by 27% of cases and 2.3% of controls, mOR: 20 (95% CI: 9.8-40, Table 3). 177 178 Community determinants of SARS-CoV-2 infection 179 Controls were more likely to report having been to restaurants than cases, mOR: 0.66 (95% CI: 0.49- 180 0.90). However, cases were more likely than controls to report, that they or others in their company had 181 consumed alcohol during the restaurant visit, adjusted odds ratio (aOR): 2.3 (95% CI: 1.3-4.2). The same 182 trend was seen for bar and indoor cultural events, where more controls than cases reported to have visited a 183 bar and participated in indoor cultural events, but more cases than controls reported alcohol consumption 184 (Table 4). 185 In total 24% of cases and 19% of controls had been in a fitness centre at least once during the 6-day 186 exposure period, mOR: 1.40 (95% CI: 0.98-2.0). Apart from this, controls were more likely to participate in 187 indoor sport activities, mOR: 0.57 (95% CI: 0.32-1.0), and outdoor sport activities, mOR: 0.71 (95% CI: 0.49- 188 1.0). For private social events, a higher proportion of controls than cases reported to have participated in 189 small or medium-sized private social events, while there was no difference among the proportion 190 participating in large private social events (Table 4). Finally, a higher proportion of controls than cases had 191 visited shops/supermarkets, used public transport, or participated in religious events in the period. For 192 participation in events, which involved singing, no difference was observed (Table 4). 193 In the sensitivity analysis, excluding cases and controls (and their matched case or control) who reported 194 to have been close contacts to an infected person during the exposure period, we obtained largely similar 195 results, showing the same trends (Table 5). . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 11 196 Discussion 197 In this case-control study performed in June 2021, we found that known contact to an infected person 198 was the most important risk factor for infection. For transmission in the community, outside the household, 199 we identified only weak associations. Infections were associated with fitness centers and alcohol intake, 200 while a large series of other possible societal exposures were not found to be associated with SARS-CoV-2 201 infection; some were even found to be negatively associated with infection. 202 We prepared and present a detailed account of the officially imposed restrictions to free movement 203 that were in place during the epidemic period from February 2020 to March 2022. Because of the possibility 204 to perform population register–based studies and because of the particularly high number of SARS-CoV-2 205 tests that were performed in the country, Danish data has become a focus of interest in COVID-19 research 206 [14, 16, 18, 21-27]. Besides its relevance for the current study, we hope with this account of the official 207 restrictions imposed to be able to provide context to studies of the epidemiology of COVID-19 being based 208 on Danish SARS-CoV-2 data. For comparison of restrictive measures that were implemented in individual 209 countries in the European Union, including Denmark, the European Centre for Disease Prevention and 210 Control has published reports hereof [2]. 211 This study is the second in a series of two. Six month prior to the current study, we performed a first 212 case-control study using similar methodology [10]. The main difference in set-up between the two studies 213 related to the study size, the current was based on inclusion of 1000 cases and controls, the first study on 214 600 only, and the fact that the exposure period inquired about was shortened from two weeks to six days, 215 with the aim of increasing the specificity. Apart from that, the main changes concerned external factors: the 216 differences in restrictions in place, society this time being far more open, the Alpha rather than the original 217 viral strain being in circulation and, importantly, the older population segments and other risk groups having 218 been vaccinated and therefore excluded from the study population. The pattern of risk factors seen in the 219 current study was remarkably similar to what we found in our previous study, where also, apart from contact 220 to persons with known infections, fitness centers and alcohol consumption in bars and in addition 221 participation in events which involved singing, were identified as being associated with SARS-CoV-2 222 infection. This second study may therefore be seen as corroborating the findings of the first and it would 223 appear that besides direct contact with infected individuals, under the restrictive measures in place in . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 12 224 Denmark, use of fitness centers and social activity involving alcohol constituted actual risk factors. Whereas 225 many other investigated activities, notably use of public transportation, supermarkets and cultural and sports 226 gatherings were as common in cases than amongst controls. 227 Several other case-control studies of community determinants have been published. Early in the 228 pandemic (May to June 2020), a case-control study conducted in Ohio and Florida, found no association 229 between infection with SARS-CoV-2 and attending private or public gatherings or use of public transport [6]. 230 Another study from July 2020 in the USA among outpatients also did not indicate an association between 231 SARS-CoV-2 infection and use of public transport, shopping or visiting friends and family. However, cases 232 were more likely to have been dining at restaurants and visiting bars/coffee shops than controls [9]. In 233 Portugal, in September to October 2020, no association was found with use of public transport, restaurant 234 visits, mall/supermarket visits, attending gym or sports activities and being infected with SARS-CoV-2 [3]. A 235 large case-control study from France (October and November 2020), in a period with broad-reaching public 236 health and social measures, found an increased risk of infection associated with bar and restaurant visits, 237 but no association with attending cultural gatherings [4]. Later in the pandemic, in a period with Delta 238 circulation in France, another case-control study was performed (May to August 2021). Here people under 239 40 years of age attending bars, nightclubs or private parties were found to be at increased risk of infection. 240 For public transport, cases were more likely to have used the subway, but not buses, trams or trains. For 241 private gatherings, there was an association with ceremonies, but no association with other private 242 gatherings, cultural events nor shopping (except from convenience stores). No association was seen for 243 outdoor sports activities, but for indoor sport activities [8]. A Danish (as yet unpublished) case-control study, 244 performed in October to December 2020 by a different group from ours, found associations similar to those 245 seen in our studies. The most important risk factor identified was contact to a person with known infection. 246 Moreover, this study also found an association with fitness centers but not with shopping, use of public 247 transport and participating in outdoor sport activities. However, contrary to our findings, it identified 248 participating in indoor sport activities, larger events, and restaurant and bar visits as risk factors [5]. 249 Taken together, the available literature has not been able to show an association between SARS-CoV-2 250 infection or hospital admission and potential risk factors such as: supermarkets, outdoor sport activities or 251 use of public transport in situations where basic preventive measures – mask use, keeping a distance – have 252 been in place. Certain other activities have been found to be associated with infection, such as indoor sport . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 13 253 activities or restaurant and bar visits. However, this appeared to vary, depending on the setting of the 254 particular study. 255 Methodological strengths and limitation outlined in our previous 2020-study also apply for the current 256 study. Among the limitations of the first study was the small sample size, therefore we went from 600 to 1000 257 participants to strengthen the power of the present study. Compared to our first study, we also shortened the 258 exposure period inquired about, aiming to provide more specific estimates of associations. The use of the 259 Danish Vaccination Registry enabled us to swiftly and objectively exclude those who had been vaccinated by 260 the time of the study. 261 A potential bias would arise from systematic differences in behavior between cases and controls. Some 262 persons who recently had been in close contact with a person with known infection, would likely have been 263 in isolation and therefore not exposed in the community. Because we frequently found controls to be more 264 exposed than cases (resulting in OR estimates below 1), we were suspicious of such a bias being at play. To 265 explore this further, we performed a sensitivity analysis, in which we excluded all participants who reported to 266 have been close contacts to infected persons. However, this did not change the results. Another potential 267 concern relates to the selection of controls. We used matched controls sampled from the general population, 268 which was made possible because of our access to the Danish Civil Registration System. A different 269 possibility, which we did not opt for, would have been control selection with recruitment from the pool of 270 persons testing negative in PCR test in the same period as the pool of cases tested positive. This option has 271 been used by others, and we cannot say how it would have influenced on our results [5, 9]. We note 272 however, that controls were probably unlikely to have been positive without knowing so. In the study period, 273 participation in societal activities generally required testing as part of the corona passport in our 274 unvaccinated, not previously infected study population, and almost 4,000,000 public tests were performed 275 every week. 276 In conclusion, we show results of a study of risk factors for SARS-CoV-2 infection and compare with a 277 similar study done six month earlier. Under the constraints of the methodology of case-control studies, we 278 conclude that societal restrictions in use in the spring of 2021 were in fact efficient; people did not appear to 279 become infected while conduction public activities, including presence at cultural events, in restaurants, 280 shops and public transportation; exceptions being fitness centers and alcohol consumption which did . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 14 281 constitute risks. Instead, transmission primarily took place via contact to other persons, often colleagues or 282 family members, known to be infected. Finally, we provide a timeline of non-pharmaceutical interventions 283 that were implemented in Denmark from February 2020 to March 2022. 284 285 Acknowledgments 286 We thank the participants of this study for taking their time to answer questions. We thank Caroline Eves for 287 critical reading of the manuscript. 288 289 Author Contributions 290 Conceptualization and design: all authors; Formal analysis: Christian Holm Hansen, Pernille Kold Munch; 291 Funding acquisition: Tyra Grove Krause, Steen Ethelberg; Project administration: Pernille Kold Munch, Laura 292 Espenhain; Supervision: Steen Ethelberg, Tyra Grove Krause; Validation: Laura Espenhain; Writing – 293 original draft: Pernille Kold Munch, Steen; Writing – review & editing: all authors. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 15

1. European Centre for Disease Prevention and Control. Guidelines for the implementation of nonpharmaceutical interventions against COVID-19. Stockholm: 2020. 2. European Centre for Disease Prevention and Control. COVID-19 country overviews https://covid19- country-overviews.ecdc.europa.eu/#11_Denmark2021. 3. Leite A, Leão T, Soares P, Severo M, Moniz M, Lucas R, et al. A Case-Control Study of Contextual Factors for SARS-CoV-2 Transmission. Front Public Health. 2021;9:772782. Epub 2021/11/23. doi: 10.3389/fpubh.2021.772782. PubMed PMID: 34805081; PubMed Central PMCID: PMCPMC8599439. 4. Galmiche S, Charmet T, Schaeffer L, Paireau J, Grant R, Chény O, et al. Exposures associated with SARS-CoV-2 infection in France: A nationwide online case-control study. The Lancet Regional Health Europe. 2021;7:100148. Epub 2021/06/15. doi: 10.1016/j.lanepe.2021.100148. PubMed PMID: 34124709; PubMed Central PMCID: PMCPMC8183123 of France. All other authors have nothing to declare. 5. Nordsjællands Hospital. Projekt COVID-19 Mapping: Afdækning af smittespredningsmønsteret i Region Hovedstaden. https://www.nordsjaellandshospital.dk/nyheder/Documents/Projekt%20COVID-19- Mapping%20Data%20rapport_kort_09032021.pdf: 2021. 6. Speaker SL, Doherty CM, Pfoh E, Dunn A, Hair B, Daboul L, et al. Social Behaviors Associated With a Positive COVID-19 Test Result. Cureus. 2021;13(2):e13064. Epub 2021/03/09. doi: 10.7759/cureus.13064. PubMed PMID: 33680606; PubMed Central PMCID: PMCPMC7929545. 7. Fisher KA, Olson SM, Tenforde MW, Feldstein LR, Lindsell CJ, Shapiro NI, et al. Telework Before Illness Onset Among Symptomatic Adults Aged ≥18 Years With and Without COVID-19 in 11 Outpatient Health Care Facilities - United States, July 2020. MMWR Morb Mortal Wkly Rep. 2020;69(44):1648-53. Epub 2020/11/06. doi: 10.15585/mmwr.mm6944a4. PubMed PMID: 33151918; PubMed Central PMCID: PMCPMC7643895. 8. Grant R, Charmet T, Schaeffer L, Galmiche S, Madec Y, Von Platen C, et al. Impact of SARS-CoV-2 Delta variant on incubation, transmission settings and vaccine effectiveness: Results from a nationwide case-control study in France. Lancet Reg Health Eur. 2022;13:100278. Epub 2021/12/02. doi: 10.1016/j.lanepe.2021.100278. PubMed PMID: 34849500; PubMed Central PMCID: PMCPMC8616730. 9. Fisher KA, Tenforde MW, Feldstein LR, Lindsell CJ, Shapiro NI, Files DC, et al. Community and Close Contact Exposures Associated with COVID-19 Among Symptomatic Adults ≥18 Years in 11 Outpatient Health Care Facilities - United States, July 2020. Center for Disease Control and Prevention - Morbidity and Mortality Weekly Report. 2020;69(36):1258-64. Epub 2020/09/12. doi: 10.15585/mmwr.mm6936a5. PubMed PMID: 32915165; PubMed Central PMCID: PMCPMC7499837. 10. Munch PK, Espenhain L, Hansen CH, Müller L, Krause TG, Ethelberg S. Societal activities associated with SARS-CoV-2 infection: a case-control study in Denmark, November 2020. Epidemiol Infect. 2021;150:e9. Epub 2021/11/18. doi: 10.1017/s0950268821002478. PubMed PMID: 34784997; PubMed Central PMCID: PMCPMC8755547. 11. The Danish Ministry of Health. Orientering om ændringer i regler vedr. coronapasset pr. 21. maj 2021. https://sum.dk/Media/637572575767187044/Orientering%20om%20%C3%A6ndringer%20i%20regler%20ve dr.%20coronapas%20pr.%20215%20NY%20VERSION.pdf2021. 12. The Danish Ministry of Health Ministry of Industry Business and Financial Affairs and The Ministry of Employment. Retningslinjer om ansvarlig indretning af liberale serviceerhverv og køreskoler i lyset af udbruddet af COVID-19. https://em.dk/media/14167/retningslinjer-for-ansvarlig-indretning-af-liberale- serviceerhverv.pdf2021. 13. Pedersen CB. The Danish Civil Registration System. Scandinavian Journal of Public Health. 2011;39:22-5. Epub 20 July 2011. doi: 10.1177/1403494810387965. PubMed PMID: 21775345. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 16 14. Hansen CH, Michlmayr D, Gubbels SM, Mølbak K, Ethelberg S. Assessment of protection against reinfection with SARS-CoV-2 among 4 million PCR-tested individuals in Denmark in 2020: a population-level observational study. Lancet. 2021;397(10280):1204-12. Epub 2021/03/21. doi: 10.1016/s0140- 6736(21)00575-4. PubMed PMID: 33743221; PubMed Central PMCID: PMCPMC7969130. 15. Schønning K, Dessau RB, Jensen TG, Thorsen NM, Wiuff C, Nielsen L, et al. Electronic reporting of diagnostic laboratory test results from all healthcare sectors is a cornerstone of national preparedness and control of COVID-19 in Denmark. APMIS - Journal of Pathology, Microbiology and Immunology. 2021. Epub 2021/05/06. doi: 10.1111/apm.13140. PubMed PMID: 33949007. 16. Espenhain L, Tribler S, Sværke Jørgensen C, Holm Hansen C, Wolff Sönksen U, Ethelberg S. Prevalence of SARS-CoV-2 antibodies in Denmark: nationwide, population-based seroepidemiological study. Eur J Epidemiol. 2021;36(7):715-25. Epub 2021/08/23. doi: 10.1007/s10654-021-00796-8. PubMed PMID: 34420152; PubMed Central PMCID: PMCPMC8380416. 17. Schmidt M, Pedersen L, Sørensen HT. The Danish Civil Registration System as a tool in epidemiology. Eur J Epidemiol. 2014;29(8):541-9. Epub 2014/06/27. doi: 10.1007/s10654-014-9930-3. PubMed PMID: 24965263. 18. Gram MA, Nielsen J, Schelde AB, Nielsen KF, Moustsen-Helms IR, Sørensen AKB, et al. Vaccine effectiveness against SARS-CoV-2 infection, hospitalization, and death when combining a first dose ChAdOx1 vaccine with a subsequent mRNA vaccine in Denmark: A nationwide population-based cohort study. PLoS Med. 2021;18(12):e1003874. Epub 2021/12/18. doi: 10.1371/journal.pmed.1003874. PubMed PMID: 34919548; PubMed Central PMCID: PMCPMC8726493. 19. Kirkwood BR, Sterne JAC. Essential Medical Statistics. 2nd edition ed. Massachusetts, USA: Blackwell Science Ltd; 2003. p. 413-28. 20. Nationalt Kommunikations Partnerskab COVID-19. Genåbning af Danmark https://coronasmitte.dk/Media/637572003274448907/Gena%CC%8Abning%20af%20Danmark%202105202 1.pdf2021. 21. Fonager J, Bennedbæk M, Bager P, Wohlfahrt J, Ellegaard KM, Ingham AC, et al. Molecular epidemiology of the SARS-CoV-2 variant Omicron BA.2 sub-lineage in Denmark, 29 November 2021 to 2 January 2022. Euro Surveill. 2022;27(10). Epub 2022/03/12. doi: 10.2807/1560- 7917.Es.2022.27.10.2200181. PubMed PMID: 35272746; PubMed Central PMCID: PMCPMC8915403. 22. Husby A, Hansen JV, Fosbøl E, Thiesson EM, Madsen M, Thomsen RW, et al. SARS-CoV-2 vaccination and myocarditis or myopericarditis: population based cohort study. BMJ. 2021;375:e068665. Epub 2021/12/18. doi: 10.1136/bmj-2021-068665. PubMed PMID: 34916207; PubMed Central PMCID: PMCPMC8683843. 23. Hviid A, Hansen JV, Thiesson EM, Wohlfahrt J. Association of AZD1222 and BNT162b2 COVID-19 Vaccination With Thromboembolic and Thrombocytopenic Events in Frontline Personnel : A Retrospective Cohort Study. Ann Intern Med. 2022. Epub 2022/02/02. doi: 10.7326/m21-2452. PubMed PMID: 35103482; PubMed Central PMCID: PMCPMC8805866 www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M21-2452. 24. Pottegård A, Lund LC, Karlstad Ø, Dahl J, Andersen M, Hallas J, et al. Arterial events, venous thromboembolism, thrombocytopenia, and bleeding after vaccination with Oxford-AstraZeneca ChAdOx1-S in Denmark and Norway: population based cohort study. Bmj. 2021;373:n1114. Epub 2021/05/07. doi: 10.1136/bmj.n1114. PubMed PMID: 33952445; PubMed Central PMCID: PMCPMC8097496. 25. Lyngse FP, Mølbak K, Skov RL, Christiansen LE, Mortensen LH, Albertsen M, et al. Increased transmissibility of SARS-CoV-2 lineage B.1.1.7 by age and viral load. Nat Commun. 2021;12(1):7251. Epub 2021/12/15. doi: 10.1038/s41467-021-27202-x. PubMed PMID: 34903718; PubMed Central PMCID: PMCPMC8669007. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 17 26. Bager P, Wohlfahrt J, Fonager J, Rasmussen M, Albertsen M, Michaelsen TY, et al. Risk of hospitalisation associated with infection with SARS-CoV-2 lineage B.1.1.7 in Denmark: an observational cohort study. Lancet Infect Dis. 2021;21(11):1507-17. Epub 2021/06/26. doi: 10.1016/s1473-3099(21)00290- 5. PubMed PMID: 34171231; PubMed Central PMCID: PMCPMC8219488. 27. Espenhain L, Funk T, Overvad M, Edslev SM, Fonager J, Ingham AC, et al. Epidemiological characterisation of the first 785 SARS-CoV-2 Omicron variant cases in Denmark, December 2021. Euro Surveill. 2021;26(50). Epub 2021/12/18. doi: 10.2807/1560-7917.Es.2021.26.50.2101146. PubMed PMID: 34915977; PubMed Central PMCID: PMCPMC8728489. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 18 Table 1. Number and proportion of included cases and controls by demographic characteristics and p-value of test for deviations, Denmark, June 2021. Demographic characteristics Included cases (n=500), n (%) Matched controls (n=529), n (%) P value1 Age group na 18-24 years 199 (40) 212 (40) 25-34 years 159 (32) 170 (32) 35-44 years 85 (17) 88 (17) 45-49 years 57 (11) 59 (11) Sex na Male 263 (53) 283 (54) Female 237 (47) 246 (47) Region na Capital Region of Denmark 195 (39) 200 (38) Region Zealand 56 (11) 62 (12) Region of Southern Denmark 61 (12) 66 (12) Central Denmark Region 117 (23) 128 (24) North Denmark Region 71 (14) 73 (14) Migration background2 <0.05 Denmark 403 (81) 462 (87) Western country 26 (5.2) 23 (4.3) Non-western country 71 (14) 44 (8.3) Household size3 0.64 1 78 (16) 91 (17) 2 141 (28) 165 (31) 3 109 (22) 100 (19) 4 105 (21) 108 (20) ≥5 67 (13) 65 (12) Number of contacts <0.05 0-5 290 (58) 208 (39) 6-10 88 (18) 156 (29) 11-15 47 (9.4) 51 (9.6) 16-20 27 (5.4) 30 (5.7) 21-49 30 (6.0) 52 (9.8) Over 50 18 (3.6) 32 (6.0) Employment 0.34 Employed 307 (61) 339 (64) Under education 150 (30) 138 (26) Other 43 (8.6) 52 (9.8) 1 P-value for t-test for deviation between included cases and matched controls. 2 Non-Danish migration background was defined as first or second generation immigrants from either Western countries (primarily neighboring European countries) or from non-Western countries (the five most frequent being Turkey, Iraq, Lebanon, Pakistan, and Somalia). 3 Number of registered persons on the same address. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 19 Table 2. Number and proportion of likely place of infection as indicated by cases (n=516). Likely place of infection n (%)1 Household 105 (20) Friends/other family than household 80 (16) Workplace 84 (16) Education 28 (5.4) Leisure activities 27 (5.2) Other events 13 (2.5) Public transport 8 (1.6) Other place/exposure 68 (13) Don’t know 103 (20) 1It was possible to indicate more than one likely place of infection. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 20 Table 3. Number, proportion and matched odds ratios (mOR) related to type of contact with a person with known SARS- CoV-2 infection (with/without symptoms), Denmark, June 2021. Type of contact with infected person with/without symptoms Cases (n=487), n (%) Controls (n=525), n (%) mOR (95% CI) No contact with infected person 258 (53) 485 (92) Ref. Other contact, without symptoms 10 (2.1) 6 (1.1) 3.04 (0.93-10) Other contact, with symptoms 14 (2.9) 6 (1.1) 3.25 (1.2-9.2) Close contact, without symptoms 73 (15) 16 (3.1) 8.53 (4.5-16) Close contact, with symptoms 132 (27) 12 (2.3) 20 (9.8-39) Note mOR estimates were adjusted for type of contact with/without symptoms, migration background and household size. All ‘do not know’ responds were excluded in the analysis. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 21 Table 4. Number, proportion and odds ratios related to community exposures without household transmission, Denmark, June 2021. Community exposures1 Cases (n=395), n (%) Controls (n=421),n (%) OR (95% CI) Restaurant or café 126 (32) 176 (42) 0.66 (0.49-0.90) Alcohol vs. no alcohol 35 (28) 26 (15) 2.33 (1.29-4.21) Bar 78 (20) 102 (24) 0.76 (0.53-1.09) Alcohol vs. no alcohol 65 (83) 80 (78) 1.31 (0.57-3.01) Indoor cultural events 23 (5.8) 44 (10) 0.58 (0.34-0.98) Alcohol vs. no alcohol 5 (22) 7 (16) 1.65 (0.34-7.87) Spectator at sport events 21 (5.3) 33 (7.8) 0.69 (0.39-1.22) Alcohol vs. no alcohol 6 (29) 10 (30) 0.59 (0.14-2.45) Indoor fitness center 95 (24) 82 (19) 1.40 (0.98-2.01) Indoor sport activities 24 (6.1) 41 (9.7) 0.57 (0.32-1.01) Outdoor sport activities 68 (17) 94 (22) 0.71 (0.49-1.03) Shopping (grocery) 299 (76) 379 (90) 0.36 (0.24-0.54) Shopping (other) 119 (30) 205 (49) 0.45 (0.33-0.61) Private social events <10 persons 125 (32) 212 (50) 0.46 (0.34-0.63) Alcohol vs. no alcohol 57 (46) 93 (44) 1.06 (0.65-1.72) Private social events 10-20 persons 48 (12) 80 (19) 0.61 (0.41-0.90) Alcohol vs. no alcohol 30 (63) 49 (61) 1.05 (0.45-2.46) Private social events >20 persons 34 (8.6) 36 (8.5) 1.05 (0.62-1.78) Alcohol vs. no alcohol 23 (68) 25 (69) 1.13 (0.32-4.03) Public transport 123 (31) 179 (43) 0.56 (0.41-0.78) During rush hour 43 (35) 83 (46) 0.59 (0.36-0.96) Religious events 5 (1.3) 18 (4.3) 0.29 (0.11-0.79) Events with singing 46 (12) 51 (12) 1.04 (0.65-1.65) For analyses on exposure mOR adjusted for migration background and household size are shown (italic). For sub-analyses on details within an exposure unmatched OR adjusted for sex, age, region, migration background and household size are shown (non italic). 1Never versus at least once in the period. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint 22 Table 5. Number, proportion and odds ratios related to community exposures without close contact cases and controls, Denmark, June 2021 Community exposures (n cases/n controls)1 Cases (n=267), n (%) Controls (n=277), n (%) OR (95% CI) Restaurant or café 78 (29) 108 (39) 0.59 (0.39-0.89) Alcohol vs. no alcohol 17 (22) 14 (13) 1.72 (0.75-3.94) Bar 55 (21) 72 (26) 0.78 (0.50-1.22) Alcohol vs. no alcohol 44 (80) 56 (78) 1.00 (0.38-2.68) Indoor cultural events 12 (4.5) 28 (10) 0.48 (0.24-0.96) Alcohol vs. no alcohol 2 (17) 8 (29) 1.25 (0.04-41) Spectator at sport events 14 (5.2) 23 (8.3) 0.55 (0.27-1.10) Alcohol vs. no alcohol 4 (29) 6 (26) 0.36 (0.03-4.50) Indoor fitness center 61 (23) 51 (18) 1.58 (0.92-2.37) Indoor sport activities 19 (7.1) 26 (9.4) 0.84 (0.42-1.66) Outdoor sport activities 44 (16) 59 (21) 0.74 (0.46-1.20) Shopping (grocery) 202 (76) 253 (91) 0.36 (0.22-0.60) Shopping (other) 80 (30) 142 (51) 0.42 (0.29-0.63) Private social events <10 persons 85 (32) 148 (53) 0.40 (0.27-0.59) Alcohol vs. no alcohol 34 (40) 74 (50) 0.74 (0.40-1.35) Private social events 10-20 persons 30 (11) 54 (19) 0.53 (0.32-0.87) Alcohol vs. no alcohol 15 (50) 33 (61) 0.40 (0.12-1.30) Private social events >20 persons 15 (5.6) 23 (8.3) 0.64 (0.31-1.33) Alcohol vs. no alcohol 10 (67) 16 (70) 0.65 (0.07-5.64) Public transport 84 (31) 118 (43) 0.57 (0.38-0.86) During rush hour 29 (35) 58 (49) 0.48 (0.26-0.90) Religious events 1 (0.37) 8 (2.9) 0.13 (0.02-1.07) Events with singing 25 (9.4) 35 (13) 0.70 (0.38-1.30) For analyses on exposure mOR adjusted for migration background and household size are shown (italic). For sub-analyses on details within an exposure unmatched OR adjusted for sex, age, region, migration background and household size are shown (non italic). 1Never versus at least once in the period. . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint . 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 June 2, 2022. ; https://doi.org/10.1101/2022.05.31.22274922doi: medRxiv preprint