Healthcare workers high COVID-19 infection rate: the source of infections and potential for respirators and surgical masks to reduce occupational infections

Objective: To analyse the work-related exposure to SARS-CoV-2 and trace the source of COVID-19 infections in tertiary hospitals healthcare workers in light of the used PPE and their ability to maintain social distances and follow governmental restrictions. Design: Cross-sectional study Setting: Tertiary hospitals in Uusimaa region, Finland Participants: Of 1072 enrolled, 866 HCWs (588 nurses, 170 doctors and 108 laboratory and medical imaging nurses) from the Helsinki University Hospital completed the questionnaire by July 15th, 2020. The average age of participants was 42.4 years and 772 (89.0%) were women. The participants answered a detailed questionnaire of their PPE usage, ability to follow safety restrictions, exposure to COVID-19, the source of potential COVID-19 infection and both mental and physical symptoms during the first wave of COVID-19 in Finland. Main outcome measures: All participants with COVID-19 symptoms were tested with either RT-PCR or antibody tests. The infections were traced and categorised based on the location and source of infection. The possibility to maintain social distance and PPE usage during exposure were analyzed. Results: Of the HCWs that participated, 41 (4.7%) tested positive for SARS-CoV-2, marking a substantially higher infection rate than that of the general population (0.3%); 22 (53.6%) of infections were confirmed or likely occupational, including 7 (31.8%) from colleagues. Additionally, 5 (26.3%) of other infections were from colleagues outside the working facilities. 14 (63.6%) of occupational infections occurred while using a surgical mask. No occupational infections were found while using an FFP2/3 respirator and aerosol precautions while treating suspected or confirmed COVID-19 patients. Conclusions: While treating suspected or confirmed COVID-19 patients, HCWs should wear an FFP2/3 respirator and recommended PPE. Maintaining safety distances in the workplace and controlling infections between HCWs should be priorities to ensure safe working conditions.


INTRODUCTION
SARS-CoV-2 spreads mainly via droplets, secretions and direct contact 1 . Lately, the possibility of airborne transmission has been discussed even in the absence of aerosol-generating procedures (AGP), especially indoors 2 3 . SARS-CoV-2 appears to be more infectious than influenza, and the reproductive number (R0) has been estimated to be as high as 2.3-5.7 in the general population in various studies 1 4-6 . The infection rates of healthcare workers (HCW) varies widely from country to country, ranging from 2.2% to 44% yet exceeding that of the general population 7 8 .

The first wave and restrictions in Finland
The COVID-19 infection reached the epidemic threshold in mid-March, and the Finnish government declared a state of emergency from March 16 th to June 16 th 2020. The primary focus was on social distancing with restrictions for travel, limitations of no more than 10 persons for public gatherings, and recommendations to avoid spending time in public places.
Additionally, visitors were banned from care institutions, healthcare units, and hospitals.
Also, the epicenter of the COVID-19 epidemic in Finland, the Uusimaa region, was isolated . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08. 17.20176842 doi: medRxiv preprint from the rest of the Finland between March 28 th and April 15 th . Helsinki University Hospital (HUS) is responsible for that region's specialised care, and its healthcare professionals are the focus of this article. By July 15 th , Finland had 7 293 confirmed cases of COVID-19, 5223 of which were in the HUS region, with a 0.3% infection rate. 9 In the HUS region, 794 (15.2%) workers in social and healthcare organizations were infected with COVID-19 by July 15 th 2020; 349 (44.0%) had occupational infection, 207 (26.1%) infections were non-occupational, and for 238 (30.0%) the source was unclear; 153 were HCWs (19.3%) working at HUS. [10][11][12] Restrictions at HUS Due to the crisis, HUS instructed personnel with several restrictions to avoid spreading the virus in the hospital facilities; non-urgent patient contact was postponed, and personnel were instructed to 1) avoid all trips abroad, 2) avoid all gatherings and favour digital meetings and remote patient contact, 3) keep at least one meter of distance from other employees, 4) maintain good hand hygiene, 5) use the required personal protective equipment (PPE), and 6) self-isolate and get tested for COVID-19 (nasopharyngeal or oropharyngeal RT-PCR) if they experience any COVID-19-related symptoms.
The debate over masks and respirators in COVID- 19 infections As the transmission routes of SARS-CoV-2 are still partly debated, there has been speculation about which PPE (especially masks and respirators) HCWs should use with various patient groups and during medical procedures. The World Health Organization (WHO) has aligned that HCWs working with COVID-19 patients should use masks throughout their shift and N95 or FFP2/FFP3 respirators during potential AGPs as well as in semi-intensive and intensive care units (ICU) 13 . During the first wave, HUS generally followed the WHO's recommendations regarding PPE instructions for the ICU, but no masks were used in the staff area. Cohorted wards had similar recommendations for PPE at first but were quickly reduced to surgical . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint mask instead of an FFP3 respirator; on March 26 th the instruction was loosened further as gowns were only required during close patient contact, and hair protection was no longer required. Cohorted wards followed droplet precautions whereas ICUs also followed aerosol precautions while doffing.
Differences between surgical masks and respirators like N95 and FFP2/3 have been studied during the recent decade concerning different respiratory viruses and bacteria but not with COVID-19. A meta-analysis and systematic review by Smith et al. shows that N95 respirators have a protective advantage over surgical masks in laboratory conditions but do not protect HCWs better against influenza in clinical work 14 . Other studies have not shown any difference in morbidity from influenza in HCWs using either an N95 respirator or a surgical mask 15 .
However, in bacterial infections, N95 respirators have been observed to prevent infection better than surgical masks 15 . These previous studies did not obserwe the possibility of nonoccupational infection and used no other restrictions to reduse the non-work or colleaguerelated exposure.
Purpose of the study The purpose of this study was to analyze the work-related exposure to SARS-CoV-2 and trace the source of COVID-19 infections in tertiary hospitals' HCWs in light of the used PPE and their ability to maintain social distances and follow governmental restrictions. The hypothesis was that usage of FFP2/3 masks prevents workplace related COVID-19 infections.

MATERIAL AND METHODS
Out of 17 740 nurses, midwives, and doctors working at Helsinki University Hospital, 1072 (6.0%) enrolled to the study, and 866 (4.9%) answered the questionnaire completely between June 12 th and July 15 th , 2020. The minimum study size of 366 was calculated by is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020.  16 17 . Additionally, some HUS workers have been tested with antibodies as part of the employer's COVID-19 strategy, and all who tested positive with the first antibody test were tested for neutralising antibodies to confirm the diagnostics 18 . Potential AGPs were listed according to the THL as intubation, extubation, resuscitation, direct laryngoscopy, bronchoscopy, upper gastrointestinal endoscopy, non-invasive ventilation, use of nebulizer, high-flow nasal oxygen, open suction of the mucus from airways, and oral and ENT surgery 19 .
The number of COVID-19 infections in HCWs was calculated and compared to the general population in the same area. The infection rate was calculated for all HCWs and each occupation separately. PPE usage was compared between infected and non-infected HCWs.
All infected participants were contacted, and their answers were confirmed regarding the tracing of infection, the usage of PPE at the time of the assumed transmission, and the ability to maintain social distance at that time. The original tracing was done after the initial . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. were traced successfully, so they were marked "unclear".

Statistical analysis
The statistical analysis was performed using statistical software (IBM SPSS Statistics 25, Chicago, USA). Differences between nominal variables were tested using a chi-square test or Fischer's exact test with a P<0.05 significance level. Odds ratios (ORs) and a 95% confidence interval (CI) were calculated using logistic regression with variance calculation. Some of the participants didn't answer to all questions, leading to varying sample size from question to question. The number of answers per question is presented accordingly.

Patient and public involvement
Due to the exceptional situation during the COVID-19 pandemic, no patients or public were directly involved in the development, implementation, or interpretation of this study. Our study design was established on questionnaires, RT-PCR, and antibody testing, that were aimed at HUS' HCWs and required strong identification for enrolment. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint

Characteristics, infection rates, and exposure among studied HUS personnel
Of the 866 HCWs who participated in the study, 820 (94.7%) said that they had followed all the restrictions imposed by the Finnish government. Characteristics of the participants are presented in Table 1.
Lung disease that is not clinically stabilised Diabetes that involves organ damage Diseases that weaken the immune system Medication that significantly . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020.   is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint  Further information about the workplace of participants during the first wave of the COVID-19 epidemic, ability to maintain 1-meter distance and ORs is seen in Table 3. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. .  41 (4.7) n/a n/a n/a n/a n/a n/a n/a * One missing value. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint Only 13 participants (1.8%) told that they used a minimum of FFP2/3 respirators with noninfection patients (Figure 1), and that figure increased to 169 (28.5%) with suspected COVID-19 patients, but even with confirmed COVID-19 patients, fewer than half (210, 41.7%) used at least an FFP2/3 respirator.  Table 4. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint

HCWs' COVID-19 infection rate vs the general population
Overall, 41 or 4.7% of participants had tested positive for COVID-19, which was more than the overall HUS HCWs (0.9%) and much higher than that of general population in HUS-area (0.3%). Healthcare worker´s higher risk for COVID-19 infection was also been reported earlier, and our findings are in line with earlier studies 8 13 21 . The confirmed or likely occupational infections represented 53.7% of participants infection, somewhat higher than that of all HUS-area HCWs (44.0%) 10 . It is notable that the participants also had non-occupational contacts with other HCWs, and in our study, 5 (33%) of confirmed non-occupational infections were still from a colleague.

FFP2/3 respirator vs Surgical mask in occupational infections
This is the first study that analyses the source of infection and compares the impact of using of either a surgical mask or an FFP2/3 respirators on HCWs' COVID-19 infections, which alings with an earlier meta-analys indicating that respirators might have a stronger protective effect than surgical masks toward SARS-CoV-2 21 .
In our study, none of the ICU HCWs got sick while using the recommended PPE ( is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint work, and the infection rate for HCWs is even higher 8 22 . With other viral infections, contamination while D&D has been observed 23 , marking the importance of assisted D&D and training 24 . In our data, only under 40% of participants had trained D&D, and prevention of possible contamination during D&D needs more attention.
In an earlier study, surgical masks were shown to reduce SARS-CoV-2-virus transmission 21 , but in our study, especially in the wards with high exposure, the surgical mask is obviously less protective against COVID-19.
Out of the 413 participants who perform AGPs with infection patients, only 180 (43.6%) use FFP2/3 respirators. Furthermore, 42 (10.7%) participants said that they are not using FFP2/3 respirators even while performing AGPs with COVID-19-positive patients. The WHO and other international and THL recommendations suggest always using FFP2/3 respirators in AGPs 13 19 25 . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint (e.g., influenza) to avoid any possibility of cross-transmission. 13 At HUS, the guideline has been to use surgical masks only during patient contact, avoid any personal gatherings, and keep at least 1 meter social distance. A recent systematic review 21 shows a clear association in the reduction of infections by having at least 1 meter of physical distance, and this study also supports the importance of social distances also for HCWs. Despite the restrictions, 83.3% of those who got the infection from a colleague at the workplace and 60.1% of all participants were not able to main the instructed 1-meter radius from other people. Of all participants, 69 were guaranteed due to COVID-19 exposure from a colleague, and 46 of those (66.6%) were not able to maintain the instructed safety distance. This highlights the difficulty of following this instruction in a crowded hospital setting and urges hospitals to find additional ways of reducing transmission between colleagues. Accordingly, it can be presumed that pregnant women should not engage in direct contact with COVID-19 patients 27 31 .
Strengths and weaknesses of this study: This is the first study to analyse the impact of using surgical masks vs FFP2/3 respirators in . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08. 17.20176842 doi: medRxiv preprint HCWs during the COVID-19 pandemic. The masks and respirators were tested by the employer before usage, and the availability of masks was good during the whole study. The positive COVID-19 infections were proven by COVID-19 RT-PCR and antibody tests, which are the gold standard in COVID-19 diagnostics 32 . The study was conducted prospectively during the COVID-19 pandemic, and all of the nurses and doctors working at HUS were informed about the study. Of those that opted to participate, 95% stated that they have followed the state restrictions. These factors, combined with the low density of 176/km² in Uusimaa, greatly reduce the likelihood of non-occupational infections of HCWs and hence increase the reliability of analysis on workplace-related infections.
It could be argued that the high infection rate of HCWs is due to their more extensive testing.
However, the guidelines in Finland have been to test everyone with matching symptoms through universal healthcare. Although the testing HCWs has been prioritised, the testing capacity has been sufficient during most of the first wave, leaving only a limited number of potential symptomatic COVID-19-infected non-HCWs without diagnosis; thus, testing cannot explain the major difference in infection rates.
The characteristics of the participants reflect the overall personnel, although females are slightly overrepresented in this study, and as usual, the people related to the topic participate more frequently, as evident by the number of COVID-19 infected participants. The high number of infected participants gives this study a higher reliability in analysis of infection sources and PPE usage of those infected.
The participants reported that the availability of PPE, masks and respirators has been good compared to the use recommendations, and this does not limit the reliability of our results unlike in many other countries 33 . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint CONCLUSIONS HCWs have a higher risk of COVID-19 infection compared to the general population. Despite the high exposure to SARS-CoV-2, none of the ICU workers got clear occupational COVID-19 infection, whereas working in a cohorted COVID-19 ward or a normal ward with COVID-19 patients seems to have a high-risk association for occupational COVID-19 infection. The main difference is that the ICU workforce has used FFP2/3 respirators and aerosol precautions, whereas in other departments, the PPE quality has been significantly lower or completely absent. Notably, 29.3% of the infections were from colleagues, thus also requiring special attention for social distances and infection control measures between co-workers.
We recommend that PPE similar to what is used in the ICU be used in all COVID-19-related treatments. Per the Finnish Employee Protection Law, this should be done regardless of additional PPE expenses, especially as the global supply of respirators has increased substantially. The possible role of aerosols in the transmission of SARS-CoV-2 needs to be further studied. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 18, 2020. . https://doi.org/10.1101/2020.08.17.20176842 doi: medRxiv preprint