key: cord-0297477-5d2jag0n
authors: WILSON, S.; MOUET, A.; JEANNE-LEROYER, C.; BORGEY, F.; ODINET-RAULIN, E.; HUMBERT, X.; LE HELLO, S.; THIBON, P.
title: Professional practice for COVID-19 risk reduction among health care workers : a Cross-Sectional Study with Matched Case-Control Comparison
date: 2021-09-13
journal: nan
DOI: 10.1101/2021.09.09.21263315
sha: 10d2d690199ab5f54ae6c276cdc4b5bbd130e58b
doc_id: 297477
cord_uid: 5d2jag0n

Abstract Background: Health care workers (HCWs) are particularly exposed to COVID-19 and therefore it is paramount to study preventive measures in this population. Aim: To investigate socio-demographic factors and professional practice associated with the risk of COVID-19 among HCWs in health establishments in Normandy, France Methods: A cross-sectional and 3 case-control studies were conducted in order to explore the possible risk factors that lead to SARS-CoV2 transmission within HCWs, based on an online questionnaire. The case-control studies focused on risk factors associated with care of COVID-19 patients, care of non COVID-19 patients and contacts between colleagues. Findings: Among 2,058 respondents, respectively 1,363 (66.2%) and 695 (33.8%) in medical and medico-social establishments, 301 (14.6%) reported having been infected by SARS-CoV2. When caring for COVID-19 patients, HCWs who declared wearing respirators, either for all patient care (ORa 0.39; 95% CI: 0.29-0.51) or only when exposed to aerosol-generating procedures (ORa 0.56; 95% CI: 0.43-0.70), had a lower risk of infection compared with HCWs who declared wearing mainly surgical masks. During care of non COVID-19 patients, wearing mainly a respirator was associated with a higher risk of infection (ORa 1.84; 95% CI: 1.06-3.37). An increased risk was also found for HCWs who changed uniform in workplace changing rooms (ORa 1.93; 95% CI: 1.63-2.29). Conclusion: Correct use of PPE adapted to the situation and risk level is essential in protecting HCWs against infection.

Organization on March 11th 2020. There have since been approximately 216,000,000 cases and over 4,500,000 deaths worldwide [1] . Healthcare personnel are particularly vulnerable to infection given their exposure to the virus [2] . Between March 2020 and May 2021, 85,137

HCWs have been declared infected by the SARS-CoV2 virus in France, of which there have been 19 deaths. Within the infected personnel, 69% worked in clinical areas. The professions with the highest amount of infections were nurses (24% of cases) and nursing assistants (21%) [3] .

SARS-CoV2 can be spread by respiratory droplets and fomite contact, as well as airborne transmission in specific circumstances [4, 5] . However most transmissions occur during close face-to-face contact via respiratory droplets. The virus can be transmitted by presymptomatic, asymptomatic and symptomatic carriers [6, 7] . Protection of HCWs is a key method for controlling the spread of the virus within health establishments, as vaccination does not provide complete protection against contamination. [27] . Guidelines recommend that when in contact with COVID-19 patients, and in addition to hand hygiene, HCWs protect themselves with personal protective equipment (PPE), namely surgical masks for standard care and respirators during aerosol-generating procedures, gowns and protective goggles [9, 10] . Studies have suggested that SARS-CoV2 spreads not only between patients and from patients to HCWs, but also between infected HCWs, for example during breaks [7, 11, 12] .

This study investigated sociodemographic factors, behavioral factors and professional practice associated with the risk of COVID-19 infection in healthcare workers. Secondary aims were to describe the circumstances of infection declared by the respondents, and the protective measures applied by healthcare professionals working in clinical areas, as well as during contacts with other colleagues. 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint

A cross-sectional and three case-control matched studies were performed, based on an anonymous online questionnaire.

Healthcare personnel (medical and paramedical professionals, as well as personnel from laboratories, hospital pharmacies and administration) working in health establishments (hospitals, clinics, rehabilitation and recuperation care facilities and establishments specializing in psychiatry), nursing homes and establishments for handicapped children and adults in Normandy, France, were invited to participate in the study.

The study was conducted in Normandy, a region located in Northwestern France comprising of 6 departments (Calvados, Eure, Manche, Orne, Seine-Maritime), populated by 3,300,000 inhabitants [13] , and with around 90,000 HCWs working in 197 health establishments, 522 establishments for handicapped adults and children, and 348 nursing homes.

[14]

The study was approved by the local ethics committee for health research of Caen university hospital (ID 2293) on March 24 th 2021. Participants received detailed information on the objectives of the study. Agreement to use the anonymous data collected via the questionnaire was obtained for all participants, and they were informed of the possibility of the withdrawal of their data at any time, according to European regulation (27th April 2016).

The online questionnaire was available from 29th March 2021 to 30th June 2021.

Healthcare personnel working in hospitals were invited to participate in the study by hospital 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint management, who relayed the online version of the questionnaire. Healthcare workers from the medico-social sector were contacted by their management following an invitation to participate in the study via an email from the Regional Health Agency. The online questionnaire covered socio-demographic characteristics of HCWs (age, sex and profession), workplace, history of COVID-19 infection (confirmed by a positive SARS-CoV2 PCR or antigenic test) with suspected exposures leading to COVID-19 infection, COVID-19 vaccination status, and personal preventative equipment and other barrier measures applied at work. For respondents who reported having been infected by SARS-CoV2, the personal preventative equipment and barrier measures were those applied at work during the ten days preceding infection symptoms (or testing in case of asymptomatic infection). For respondents with no history of COVID-19 infection, these measures were those applied at the time of filling in the questionnaire, and participants were asked if these practices had changed since September 2020.

Three case-control analyses were led, the first one describing measures applied during the care of COVID-19 patients, the second one those applied during the care of non-COVID-19 patients and the third one describing contacts with colleagues. Cases were defined as healthcare personnel who declared having had a COVID-19 infection (confirmed by a positive SARS-CoV2 PCR or antigenic test) which they reported as having been acquired in the workplace. Controls were healthcare personnel who declared no known history of COVID-19 infection over the study period and who declared no modifications of the personal preventative measures they applied since September 2020. Cases and controls were matched by sector of activity (health establishment or medico-social establishment) and by profession, with 4 controls for 1 case. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The study period for the case-control studies was defined as the period between the 1st September 2020 and the 31st January 2021, corresponding to the second wave of the COVID-19 pandemic in France, when recommendations for barrier measures had been issued [10] , and PPE was widely available. The cut-off point was chosen in order to study the effects of sociodemographic factors, behavioral factors and professional practice before wide-spread vaccination of HCWs that started in January 2021.

Qualitative variables were described with their effectives and percentages, and compared using the Chi-squared test. For each case-control study, the association between exposures and COVID-19 infection was measured by computing odds-ratios (OR) with univariate conditional logistic regression analysis, to take into account the matching of cases and controls on sector of activity and profession. As selecting a single set of controls could have led to an incorrect measurement of association due to random variations of ORs, we used a bootstrap method to perform 1,000 random samplings of controls, with replacement.

We then computed the mean ORs and their 95% confidence intervals (CI) (i.e. the 2.5% and 97.5% quantiles of the distribution of the 1,000 ORs) for each exposure. All variables significantly associated with risk of infection in the initial analysis were included in the multivariable analysis after testing for absence of collinearity between variables. Analyses were performed using multivariate conditional logistic regression analysis and the same bootstrap method, allowing to calculate adjusted ORs (ORa). P <.05 was considered significant. The analyses were performed in R version 4.1.1 (R Development Core Team).

There were 2,058 complete responses filled in by HCWs. The majority of participants 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint establishments. A large proportion of participants (791, 38.0%) worked in non-medical areas, regardless of the type of establishment. Nurses and nursing assistants represented 31.3% of respondents (N=645), and doctors 10.3% (N=212). The percentages of nurses, nursing assistants and doctors were higher within respondents in health establishments than in medico-social establishments (p<10 -3 ). Most of the participants were women (N=1,680, 81.6%), and their age was predominantly between 30 and 49 (N=1,215, 59.0% of respondents) ( Table I) is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint was found to be associated with a higher risk of infection when compared to mainly use of surgical masks. Use of a face shield or protective goggles (ORa 3.10; 95% CI: 1.81-5.58) and use of gloves for all types of patient care (ORa 1.36; 95% CI: 1.10-1.67) were also associated with a higher risk of infection. No association with infection was found for use of gowns and plastic aprons, protective hair caps and overshoes, and for regular airing of patients/residents' rooms. Changing of outfit in the workplace changing rooms was associated with a higher risk of is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

This study covered a wide range of health establishments and professions and focused on a period during which PPE was available and before widespread vaccination of HCWs in France. We found that when caring for COVID-19 patients, HCWs who declared using respirators, either for all patient care or only when exposed to aerosol-generating procedures, had a lower risk of infection compared to HCWs who declared using mainly surgical masks.

On the contrary, when caring for non COVID-19 patients, wearing a respirator compared to a surgical mask was found to be a risk factor of infection. Numerous studies have described the transmission of SARS-CoV2, and masks and respirators are the key elements of PPE when caring for COVID-19 positive patients [5, 15] . In other situations, the discomfort of the equipment leading to HCWs touching the respirator to adjust it, therefore contaminating their hands could explain our findings. Respirators also need to be well fitting, and a badly fitting respirator could lead to a false sense of security by not providing a sufficient level of protection [16] . Violante et al. published a systematic review of scientific literature on the protective efficacy of surgical masks and respirators against airborne viral infections [17] .

Although this review was not specific to the SARS-CoV2 virus, current evidence suggests that surgical masks and respirators provide a similar level of protection, and respirators should be used selectively for greater risk situations such as aerosol-generating procedures due to the cost, discomfort and risk of badly-fitting respirators [18, 19] .

When caring for non-COVID-19 patients, we found that HCWs who declared using gloves for all types of patient care had a higher risk of infection, possibly explained by improper use of PPE increasing risk of contamination. In this situation, HCWs may have considered risk of contamination to be very low. Vigilance to correct use of PPE may 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint therefore be lowered, and risk of contamination due to misuse of PPE may increase. This has been proven for various infections, by decreasing the amount of hand hygiene HCWs perform when wearing gloves and cross contamination when HCWs do not systematically change gloves between patients [20] . Surprisingly, we found similar results for face shields and protective goggles, which HCWs however declared using very rarely in this situation. Again, improper use of these PPE or underuse of other PPE which should be associated with the face shields and goggles could be an explanation. Wearing of protective overshoes was not found to be a protective measure, in accordance with guidelines [9, 10] .

We found an increase in risk of infection in HCWs who reported changing their uniform in a workplace changing room. Guidelines recommend wearing an outfit dedicated to the workplace as a protective measure [10, 21] , however communal changing rooms could increase spread of infection due to close proximity of HCWs to each other and removal of PPE during change of clothes. Before arriving on the ward and after leaving the ward, HCWs may consider the risk of contamination to be low and protective measures may therefore seem less important than when in contact with patients. The close proximity of HCWs in changing rooms due to same arrival and leaving times likely increases risk of transmission. HCWs who declared participating in meetings were found to be less at risk of infection. Due to establishment guidelines covering meeting rooms, HCWs were likely to wear proper PPE during meetings with other colleagues, thus decreasing risk of contamination. However there was no increased risk found with breaks with other colleagues or meals at the workplace canteen, two key moments when PPE, namely masks, are not worn. As with meeting rooms, health establishments produced guidelines and rules for these communal areas, with limits on the amount of people in break rooms and staff canteens, wearing of masks whenever possible and social distancing measures. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Airing of communal areas and of patients' or residents' rooms was not found to be a protective factor. Guidelines [22, 23] recommend frequent room ventilation when possible in order to reduce potential airborne transmission [24, 25] . The lack of association found with room airing evaluated by the question: "How often do you air communal areas and patients' or residents' rooms?" may be due to ventilation systems in place in most health establishments, therefore reducing effects of opening windows to air rooms. The frequencies may have lacked precision, with respondents choosing between 5 categories of frequency (never, rarely, regularly but less than every other day, regularly and more than every other day and every day). Guidelines [22, 23] recommend airing rooms several times a day, therefore the question may have been too imprecise to provide an informative result.

This study presents some limitations. The questionnaire was filled in retrospectively by participants, and for respondents with history of COVID-19, the questions covered a short 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint performed. The questionnaire was sent to health establishments' management and then relayed to HCWs. There was no feedback as to how many HCWs had access to the questionnaire from then on. Despite these biases, the fact that our results are consistent with the data in the published literature and correspond to current recommendations allows us to believe that they are reliable.

Since the beginning of the pandemic, studies have demonstrated the higher risk of infection for HCWs, particularly exposed to the virus [11, 26] . Until widespread vaccination, hand hygiene and correct use of PPE were the main barriers against the spread of COVID-19

infection [9, 10, 15, 16, 27] . With the possibility of vaccines being less effective against variants of the virus, studying correct use of PPE is paramount [8] . 

We thank all of the participants in the study, as well as management and infection control teams who relayed the information about the study to health care workers in Normandy. All of the authors thank Josiane Lebeltel for her help in conveying the study to health establishments.

Mouet A, Jeanne-Leroyer C, Borgey F, Humbert X, Le Hello S and Thibon P designed the study. Thibon P applied for the ethics. Odinet-Raulin E, Humbert X, Le Hello S and Thibon P contributed to data acquisition. Wilson S and Thibon P analyzed data and wrote the manuscript.

All authors revised the manuscript. All authors contributed to data interpretation and reviewed 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint and approved the final version. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Thibon P, Wilson S and Le Hello S had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

The data set analyzed in this paper is available from the corresponding author on reasonable request.

We declare no conflicts of interest.

None.

. CC-BY-NC-ND 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 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint 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 September 13, 2021. ; https://doi.org/10.1101/2021.09.09.21263315 doi: medRxiv preprint

COVID-19 Dashboard by the Center for Systems Science and Engineering

COVID-19 in health care workers -A systematic review and meta-analysis

Recensement national des cas de COVID-19 chez les professionnels en établissements de santé

Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1

Dismantling myths on the airborne transmission of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)

Pathophysiology, Transmission, Diagnosis, and Treatment of Coronavirus Disease 2019 (COVID-19): A Review

Healthcare associated coronavirus disease 2019 among health care workers in Normandy, France: a multi-center study. Infection Prevention in Practice

New variants of SARS-CoV-2

Rational use of personal protective equipment for coronavirus disease (COVID-19) and considerations during severe shortages: Interim guidance

Avis relatif aux mesures de prévention et contrôle de l'infection dans le contexte de la pandémie COVID-19 dans tous les secteurs de soins

Specific risk factors for SARS-CoV-2 transmission among health care workers in a university hospital

Assessing coronavirus disease 2019 (COVID-19) transmission to healthcare personnel: The global ACT-HCP case-control study

Bilan démographique 2019 : la population normande poursuite sa baisse

Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis

Personal protective equipment during the coronavirus disease (COVID) 2019 pandemic -a narrative review

Surgical masks vs respirators for the protection against coronavirus infection: state of the art

Avis relatif aux mesures d'hygiène pour la prise en charge d'un patient ou résident considéré comme cas possible, probable ou confirmé d'infection à SARS-Cov-2, 17 janvier 2021

COVID-19 and the Risk to Health Care Workers: A Case Report

Glove utilization in the prevention of cross transmission: a systematic review

Société Française d'Hygiène Hospitalière

Recommandations de la SF2H relatives à l'organisation du parcours des patients, à la protection des patients et des personnels à l'heure du déconfinement et de la reprise de la l'activité médico-chirurgicale non COVID-19 en milieu de soins

Roadmap to improve and ensure good indoor ventilation in the context of COVID-19

Aerodynamic analysis of SARS-CoV-2 in two Wuhan hospitals

Airborne Precautions

Epidemiology of and Risk Factors for Coronavirus Infection in Health Care Workers: A Living Rapid Review

Effect of Vaccination on Household Transmission of SARS-CoV-2 in England