SARS-CoV-2 Infections in “less visible” Hospital Staff: The Roles and Safety of Environmental Services and Allied Health Professionals

Background During the early SARS-CoV-2 pandemic, all healthcare workers had specific and essential functions. However, environmental services (e.g., cleaning staff) and allied health professionals (e.g., physiotherapists) are often less recognised inpatient care. The aim of our study was to evaluate SARS-CoV-2-infection rates and describe risk factors relevant to workplace transmission and occupational safety amongst healthcare workers in COVID-19 hospitals before the introduction of SARS-CoV-2-specific vaccines. Methods This cross-sectional study (from May 2020 to March 2021, standardised WHO early-investigation protocol) is evaluating workplace or health-related data, COVID-19-patient proximity, personal protective equipment (PPE) use, and adherence to infection prevention and control (IPC) measures, anti-SARS-CoV-2-antibody status, and transmission pathways. Results Out of n=221 HCW (n=189 cleaning/service staff; n=32 allied health professionals), n=17 (7.7%) were seropositive. While even SARS-CoV-2-naïve HCW reported SARS-CoV-2-related symptoms, airway symptoms, loss of smell or taste, and appetite were the most specific for a SARS-CoV-2-infection. Adherence to IPC (98.6%) and recommended PPE use (98.2%) were high and not associated with seropositivity. In 70.6%, transmission occurred in private settings; in 23.5%, at the workplace (by interaction with SARS-CoV-2-positive colleagues [17.6%] or patient contact [5.9%]), or remained unclear (one case). Conclusions Infection rates were higher in all assessed ‘less visible’ healthcare-worker groups compared to the general population. Our data indicates that, while IPC measures and PPE may have contributed to the prevention of patient-to-healthcare-worker transmissions, infections were commonly acquired outside of work and transmitted between healthcare workers within the hospital. This finding emphasises the importance of ongoing education on transmission prevention and regular infection screenings at work.


Introduction
At the end of 2019, a new mutation of the severe acute respiratory syndrome coronavirus, SARS-CoV-2, began its rapid global spread, causing a varying symptom complex labelled 'Coronavirus disease 2019' . Since the outbreak of the SARS-CoV-2 pandemic, healthcare workers (HCWs) around the world have been on the front lines, caring for sick and infectious patients to prevent the negative consequences of a collapsing healthcare system. However, the necessary proximity of HCWs to infectious patients resulted in higher exposure of many HCWs to the virus, leading to varying degrees of infection risk amongst certain HCW groups (1).
Here, the seroprevalence of specific IgM and IgG antibodies reflects relevant SARS-CoV-2 exposure, even in oligo-and asymptomatic infected HCWs (2,3). While it is still unclear how long seropositivity lasts after SARS-CoV-2 infection, IgG antibodies appear to be the most reliable markers of past disease or relevant SARS-CoV-2 contact in long-term follow-ups (4,5). IgG SARS-CoV-2 antibody seroprevalence is up to four times higher in HCWs than in workers in non-medical settings (6). During the early pandemic, physicians and nursing staff involved in J o u r n a l P r e -p r o o f COVID-19 patient care were commonly infected with SARS-CoV-2 and, accordingly, showed significantly higher rates of SARS-CoV-2 IgG seroprevalence than the the respective general population (6,7). These HCWs are at an increased risk of encountering SARS-CoV-2 and thus contracting and transmitting the infection, particularly without the availability or proper use of personal protection equipment (PPE). Today, after nearly two years of the continued global SARS-CoV-2 pandemic, seroprevalence varies depending on health profession and the related tasks in caring for patients with COVID-19. The highest risk is for medical personnel in intensive care units (ICUs), and infection risk is consequently decreased for professions with limited contact with COVID-19 patients (6). HCWs with low-risk exposure either did not perform direct COVID-19 patient care or wore adequate PPE and thus had significantly lower SARS-CoV-2 antibody seroprevalence than healthcare workers with direct exposure (or direct contact with patients infected with SARS-CoV-2; (6). Proper use and availability of PPE were established as key factors in reducing the risk of infections amongst HCWs (8,9).
Regarding HCWs, medical doctors and nursing staff are commonly at the centre of public attention. However, HCW groups, such as environmental services (e.g., cleaning and service staff) and allied health professionals (e.g., physical and occupational therapists), cover important tasks in hospital routine and patient care. Both these groups are essential to healthcare and played a crucial role in the SARS-COV-2 pandemic. Physical and occupational therapists are central to these patients' respiratory and functional rehabilitation, which is associated with close personal contact and sometimes mobilisation of respiratory fluids. Environmental services are mostly in close contact with the surroundings and materials of COVID-19 patients. Allied health professionals often visit staff at COVID-19 wards and thus are easily overseen in terms of nosocomial disease control (10). Similarly, cleaning staff is often neglected in terms of occupational health and nosocomial disease control (11). Further, although this HCW group is not perceived as at risk, recent findings show that their risk of contracting SARS-COV-2 in hospital settings is even higher than that of medical doctors and nurses (11)(12)(13). While physical therapists are particularly in close contact with COVID-19 patients, they are usually underrepresented (if at all considered) in HCW seroprevalence studies (14,15). A large US study supported by the Centers for Disease Control found the highest seroprevalence rates amongst nurse assistants (12.8%) and physical therapists (10.6%; (14), and a study in the UK described physiotherapists as the group with the second-largest infection risk (39% seropositive), right after ICU staff (41%; (16).
In Austria, by 23 August 2021, 672,138 confirmed COVID-19 cases and 10,564 SARS-COV-2related deaths had been registered (17). As the capital of Austria, with a comparatively high population density, Vienna suffered the most cases and deaths (18). Several large hospitals in Vienna are dedicated to treating patients with COVID-19. Our goal in this study was to evaluate the infection risk as well as the roles of environmental-services staff and allied health professionals in transmitting SARS-CoV-2 to derive recommendations for improving the safety of these healthcare-worker groups working in Viennese hospitals designated to COVID-19 patient care. We thus aimed to (a) evaluate their health statuses (e.g., chronic conditions and SARS-CoV-2 relevant symptoms); (b) assess SARS-CoV-2 seroprevalence and transmission pathways; and (c) describe risk factors relevant to occupational safety and health, such as proximity to COVID-19 patients, adequate use of PPE, and availability of infection prevention and control (IPC) education.

Ethical Issues/Statement
All participants gave informed, written consent before participating in the study. This study was approved by the Ethics Committee of the Medical University of Vienna (project number EK 1409/2020) prior to its commencing. Participation was voluntary, involved no financial or professional incentives, and was conducted in accordance with the Helsinki declaration.

Study population and Data collection
This cross-sectional study aimed to determine the seroprevalence, working conditions, and health as well as describe risk factors for infection amongst environmental-services staff and allied health professionals during the SARS-COV-2 pandemic. We recruited hospital staff at two major hospitals designated for COVID-19 patient care in Vienna, Austria. The participants were recruited between May 2020 and March 2021. The study was presented to the respective staff superiors or department heads, who allowed their staff to participate during working hours. The participants were asked to fill out a questionnaire and then provide a blood sample. If the participants were found to be SARS-CoV-2 IgG antibody positive, the examination results were discussed with the participant by a medical study member along with the confirmed or suspected circumstances of SARS-CoV-2 transmission.

Questionnaire
The participants were asked to fill out an adapted and approved German version of the standardised WHO questionnaire for HCWs that examines risk factors associated with a SARS-COV-2 infection ('Protocol for assessment of potential risk factors for 2019-novel-coronavirus-[2019-nCoV] infection amongst health care workers in a health care setting'; (19). When filling out the form, participants not fluent in German were assisted in English, Bulgarian, Croatian, or Serbian. Sociodemographic, health, and workplace-related data was collected using questions on gender, age, profession, and occupation (outpatient clinic, ward, intensive care unit, and others, including direct contact with confirmed COVID-19 cases or their materials and surroundings).
Health-related characteristics included smoking status, BMI, presence of airway (sore throat, cough, runny nose, shortness of breath), or other symptoms (chills, vomiting, nausea, diarrhea, headache, rash, conjunctivitis, muscle aches, joint ache, loss of appetite, loss of smell, nose bleed, fatigue, seizures, altered consciousness, other neurological signs, or any other symptoms since the pandemic began), followed by items concerning chronic or past somatic or mental illness. Here, we additionally asked about the presence of allergies. The questionnaire further included items pertaining to adherence to IPC measures as well as the use of PPE and other hygiene measures. In case a participant directly contacted a confirmed COVID-19 patient, a more detailed list explored the duration and quality of contact with the patient, their materials, or the surfaces around them.

Blood sampling
All participants underwent venipuncture performed by two experienced clinicians on site. The blood samples were submitted to the MedUni Wien Biobank, a central facility at the Medical University of Vienna that specialises in processing and storing human biomaterial. There, blood serum was prepared and stored at median temperatures less than −70℃ until analysis (20). The presence of Anti-SARS-CoV-2 IgG antibodies, which served as an indication of past SARS-CoV-2 infection, was determined using Elecsys® Anti-SARS-CoV-2 electrochemiluminescence immunoassays (Roche Diagnostics, Rotkreuz, Switzerland) on a fully-automated cobas e801 analyzer (21).

Statistical analysis
Mean and standard deviations were reported for continuous data, and absolute and percentage frequencies were reported for categorical data. To compare seronegative and seropositive patients as well as cleaning staff and allied health professionals, a chi-squared, or Fisher Exact, test was conducted for categorical data. For continuous data, a t-or U-test was carried out. The significance level was set to .05. Due to the exploratory nature of this study, no multiplicity adjustment was implemented, and pvalues were interpreted descriptively. All calculations were performed using R, version 4.2.2.
inclusion criteria. The final study sample included 221 participants, 17 of whom (7.7%) were seropositive for anti-SARS-CoV-2-IgG antibodies (7.9% of the environmental-services staff and 6.3% of the allied health professionals). The sociodemographic characteristics of the participants are shown in Table 1. Most of the participants identified as female (199; 90.5%), worked in inpatient wards (52; 26.1%), were between 46-50 years old (57; 25.8%), and were employed as cleaning staff (189; 85.5%). Most J o u r n a l P r e -p r o o f (188; 85.8%) reported no frequent or direct contact with COVID-19 patients, their belongings, or their environments.

Health-related characteristics
Regarding health-related characteristics, approximately half of the participants were non-smokers (106, 51.0%), and 66.1% were overweight (BMI M = 27.7, SD = 5.1). Preexisting health conditions and diagnoses are shown in Table 2a.  (Table 2b).  J o u r n a l P r e -p r o o f contracted at the workplace; three were infected during interactions with SARS-CoV-2 positive colleagues at work, and one was infected by interacting with a SARS-CoV-2 patient at the workplace. In one case (5.9%), the path of infection was unclear.

Safety and Protection
Most participants (n=131, 61.2%) reported having had more than two cumulative hours of training on IPC measures. Almost all participants (n=211, 96.3%) reported always following the recommended hygiene practices at their healthcare facility, including using an alcohol-based hand sanitiser (or soap and water) before and after touching patients (97.3% and 98.2%, respectively) or coming into contact with patients' body fluids (97.3%). They also reported always following IPC standard precautions when in contact with any patient (98.6%) and always using PPE when indicated (98.2%). Still, several participants reported not having had enough PPE at the workplace at some point during the pandemic (18, 8.2%; Table 3). occasionally 0 (0%) 0 (0%) 0 (0%) rarely 0 (0%) 0 (0%) 0 (0%) There were no notable differences in the amount of training on IPC measures, availability of PPE, and following hygienic standards between cleaning staff and allied health professionals.

Use of alcohol-based hand rub or soap and water after touching a patient's surroundings
Participants who had direct and frequent contact with confirmed SARS-COV-2-positive patients (33, 14.9%) filled out an extended version of the questionnaire. Adherence to IPC measures was high amongst seronegative (98.5%) and seropositive (100.0%) participants, and the same was found for PPE use (98.0% and 100.0%, respectively).

Discussion
In our Austrian HCW groups, which included environmental services and allied health professionals, 7.7% of the participants had IgG antibodies against SARS-CoV-2, indicating relevant prior contact with the virus. In comparison, the percentage of SARS-CoV-2 seropositive HCWs rose to 14% in the south-western parts of Austria, which were most affected by the pandemic (22). The general SARS-CoV-2 seroprevalence for Austria was estimated to be 4.7% by April 2021 (Statistic Austria Report; (23,24). To our knowledge, we present the most extensive Austrian seroprevalence study and one of the earliest-data studies amongst environmental services staff in the ongoing SARS-COV-2 pandemic. Several other studies have been carried out regarding SARS-CoV-2 seroprevalence, particularly amongst hospital personnel (1,7,14,25,26). However, only a few mention seroprevalence in environmental services, or non-clinical staff members, and allied health professionals. Our findings matched the overall seroprevalence rates in European HCWs (7.7%) at this stage, whereas reported SARS-CoV-2 seroprevalence amongst healthcare workers was higher in the United States (12.4%) and lower in East Asia (4.8%; (25). The authors of these reports concluded that the differences in seroprevalence resulted from the availability of and compliance to respective precaution measures (25). Other authors reported that, with increasing community infection numbers, the seroprevalence of SARS-CoV-2 in HCWs rose similarly (22,26). In our study, we aimed to describe risk factors regarding occupational safety and health in our specific HCW groups. In addition to their general health statuses and SARS-CoV-2 transmission pathways, we evaluated the associations between J o u r n a l P r e -p r o o f seropositivity and proximity to COVID-19 patients, PPE availability and use, as well as IPC adherence and education.
In 70% of seropositive participants, SARS-CoV-2 infections reportedly occurred in private settings whereas 20% of transmissions occurred through social interactions with positive colleagues at work. In 5% of the cases, transmission occurred through patient interaction at the workplace, and in another 5%, the path was unclear. In most cases, the path was straightforward and easy to assess, mostly when close family members or work colleagues tested positive for SARS-CoV-2, and the close contact with our participants necessary for transmission was spatially and temporally comprehensible. Our findings are in line with other studies showing that work-related contact with infected individuals was less decisive than close contact with SARS-CoV-2-positive colleagues or family members (27). Still, data from different studies suggests that working in healthcare, particularly in ICUs, is associated with higher infection rates (26). In some cases, our participants struggled to explain how SARS-CoV-2 had been transmitted, since the participants had little or no social contact due to public lockdowns and strict safety measures at the workplace. We assume that this transmission may have been associated with participants being unskilled in or misinformed about the use of PPE, or that it may have occurred in public spaces, such as public transportation. That participants reported doing everything as 'recommended' does not necessarily mean they were informed or understood how to do it correctly. Since nearly all the participants stated that they always followed the hygiene guidelines, their responses may have been biased by them not wanting to admit mistakes. The rate of nosocomial infections may therefore be higher than that in our results. Social interactions with colleagues may not have been perceived as potential pathways for infection; hence, safety measures may not have been taken seriously in social settings at work (e.g., during breaks or meetings).
Patient proximity was not identified as a relevant risk factor for seropositivity amongst our healthcareworker groups (p = .432); healthcare workers who had worked in spaces dedicated exclusively to caring for COVID-19 patients had a slight (not significant) prevalence of anti-SARS-CoV-2 IgG antibodies compared to other non-COVID-19 exclusive hospital units, such as outpatient clinics. While the COVID-19 exclusive patient care subgroup reported having received the same amount of cumulative IPC training at their workplace as other units, a higher proportion of this subgroup reported 'always' following IPC standards and hospital-specific IPC measures. Further, in units exclusively dedicated to caring for COVID-19 patients, 100% of the participants stated that they had sufficient PPE, compared to other non-COVID-19-exclusive units, where PPE availability was limited, particularly in the early stages of the pandemic. Furthermore, not all environmental-services members in COVID-19-designated units directly contacted either SARS-CoV-2-positive patients, the patients' belongings, or surfaces around the patients (21.4%); most were assigned to enter patient rooms 30-45 minutes after the patients had left or had been discharged to allow potentially infectious airborne particles to settle. These precautionary measures as well as the availability and use of PPE may explain the low seroprevalence amongst this 'high-risk' subgroup. No other differences between our seropositive and the seronegative participants concerning PPE and IPC measurements were found. Nearly every participant reported strictly following all guidelines for proper workplace hygiene, and most had received more than two hours of training in IPC measures. Thus, these applied prevention measures were sufficient to withhold disease transmission from patients to healthcare workers.
Although a lack of understanding the guidelines and improper use of PPE may have contributed to misjudgments of reported nosocomial infection rates, we tentatively conclude that, if properly used, face masks, gloves, alcoholic hand rub, bodysuits, and face shields are effective at preventing infection in COVID-19 patient care.
At the beginning of the pandemic, several chronic illnesses were suspected to be relevant for a severe COVID-19 course and were accordingly evaluated amongst our study participants. We found that, due to necessity (e.g., staffing shortage) or personal reasons (e.g., fear of job loss), many participants with chronic illnesses continued working during the early pandemic. Our participants were on average overweight (BMI of 27.7), with 28.7% being obese (BMI > 30) and 49% being smokers, and 55.3% reporting chronic health conditionsparticularly, chronic airway diseases (6.9% asthma and 6.0% other chronic pulmonary diseases). A review conducted in 2022 summarised that individuals with preexisting health conditions were prone to more severe COVID-19 courses, resulting in higher morbidity and mortality. These findings were attributed to a spread of disease beyond the pulmonary system, affecting other organs, such as kidneys; organ systems, such as hepatic, cardiovascular, and central nervous; and the gastrointestinal tract. Furthermore, there is an interaction with diseases like cancer, diabetes, HIV, and gangrene (28). In our study, no significant differences in the percentages of seropositive and seronegative patients concerning their pre-existing diseases were found (Table 2a).
We further assessed the occurrence of SARS-CoV-2-related symptoms since the beginning of the pandemic as an indicator of prior SARS-CoV-2 infection. Since many COVID-19-related symptoms are unspecific, we evaluated their occurrence in association with seropositivity. We found loss of smell or taste, general airway symptoms, and loss of appetite to be the most SARS-COV-2-specific symptoms among the participants. Further pervasive symptoms included fever, chills, vomiting, nausea, myalgia, arthralgia, and altered consciousness. Both seropositive and seronegative participants reported symptoms like sore throat, dyspnea, headache, and conjunctivitis in comparable frequency; 12% of the seropositive group had no symptoms.
Another important factor significantly impacting the course of COVID-19 as a disease as well as a crucial aspect in the SARS-CoV-2 pandemic is vaccination coverage. It was possible to show that an individual's vaccination status had positive effects on the outcome of the disease. Over time, however, a new problem emerged: the virus was able to quickly mutate, creating different symptom patterns and disease severities as well as resulting in different levels of protection provided by available anti-SARS-CoV-2 vaccinations. Therefore, the adaptation of vaccines targeting new SARS-CoV-2 variants is urgently needed (29).
To quickly identify potential new target structures requires understanding the pathophysiology of SARS-CoV-2. The viral spike glycoprotein first binds to the cell's surface (S1) and then, through fusion with an angiotensin-converting enzyme-2 (S2), inserts its nucleocapsid into the host's cytoplasm. The viral RNA is then converted into 16 non-structural proteins (NSP 1-16), which promote the transcription of the viral RNA and the reproduction process. After the virus particles are assembled in the endoplasmic reticulum, they are released through the Golgi vesicle, commencing a new reproduction cycle (30). Although many checkpoints of this reproduction cycle are potential targets for a vaccine to interfere, most of the available vaccines target the binding process of the S1 and S2 spike proteins. There are more vaccine options in research, some of which are DNA-based and others of which use inactivated virus, live attenuated virus, bacterial vectors, or protein subunits (30).
While our study presents many relevant aspects of the occupational health and safety of important but sometimes neglected healthcare-worker subgroups, it also has limitations.
Firstly, a cross-sectional study design cannot establish causal connections between the reported observations. Secondly, SARS-CoV-2 IgG antibodies have their known limitations as infection markers; even though not every COVID-19 patient develops antibodies against the disease, persisting antibodies are a good marker for prior infection or for even just SARS-CoV-2 exposure. In a seroprevalence study from Austrian state Vorarlberg (22), the number of healthcare workers with positive SARS-CoV-2 IgG was 38% higher than that of healthcare workers who had a positive PCR test. These data, as well as our findings, were derived before the introduction of vaccines for these specific groups. Finally, we cannot exclude reporting bias, particularly regarding the transmission pathways. Even though the participants had clear assumptions about the origins of their infections (family member, colleague, etc.), the participants may have incorrectly reported their transmission paths. However, a low frequency of false workplace transmission reports is unlikely, as confirmed workplace transmissions would potentially lead to (financial) benefits and healthcare coverage by accident insurances. Still, a positivity bias to some extent cannot be excluded, as, for example, several participants reported not always having had enough PPE at the workplace despite PPE use being very high amongst the participants; the latter may, however, reflect the situation relative to PPE deficiency at some point during the pandemic.

Limitations
As our data was collected between May 2020 and March 2021, some reported aspects (e.g., ranking of most reported SARS-CoV-2 related symptoms) must be considered as associated with Wild-type and other early SARS-CoV-2 mutations and thus may diverge from current observations. The focus of our study, however, was the safety of 'less visible' healthcare workers at the beginning of the pandemic, before the availability of anti-SARS-CoV-2 vaccines, and before knowledge about SARS-CoV-2 was widespreadthus, partially before the establishment of clear occupational safety directives. Our group of environmental services staff included one of the largest in-hospital cleaning staff cohorts assessed during the early pandemic. We believe that this particular cohort, who are usually difficult to reach due to limitations in communication (e.g., not knowing the language [German] or having a lack of language skills or digital education options at work) and are affected by high turnover rates common amongst lower-wage workers in Austria, will face the same challenges during any other pandemic or healthcare crisis, resulting in similar health behaviour and occupational health risks as described in our study. Furthermore, while SARS-CoV-2-related symptoms and the availability of preventive measures may have changed, the associations of infection rates with work-related factors and ways of transmission, as reflected in our data, are still relevant for occupational health and safety today.
The primary aim of this study was a description of SARS CoV-2 infection rates, the health status, and risk factors. We therefore tried to collect a sample as large as possible, without a prior sample size calculation. In the results section it was only possible to interpret the p-values descriptively and not confirmatively.

Conclusions
While SARS-COV-2 vaccines have brought some relief to the ongoing pandemic, emerging mutations have made it clear that SARS-CoV-2 will remain relevant worldwide as an occupational health issue and a nosocomial infection for years to come. Finding new target points for vaccinations as well as adapting vaccines to mutations is inevitable for controlling the pandemic (30). Even if the pandemic should subside, the lessons derived regarding the healthcare and safety of HCWs will be relevant for any pandemic to come. For this reason, it should be of paramount importance to identify risks and suboptimal working conditions and pathways of disease transmission amongst hospital HCWs to promote occupational health and safety and prevent in-hospital disease clusters that lead to temporary or even permanent loss of infected HCWs.
The SARS-CoV-2 seroprevalence in our healthcare-worker cohort was higher than in the general population in Austria. We found that, while PPE use and IPC measures successfully prevented patientto-HCW transmission, most infections were introduced into the hospital from private environments and transmitted between HCWs in social settings at work. These finding are in line with other studies that drew similar conclusions (27,31). While the prevalence of the disease amongst the general population can be effectively regulated by social distancing and lockdown measures, infections can converge in working sites of systemic importance, where employees gather and interact throughout the pandemic. Our findings emphasise the importance of regular SARS-CoV-2 screening at work and ongoing education regarding transmission prevention, especially between colleagues, in the battle against SARS-COV-2 and for any pandemics to come