key: cord-1050204-nyq2l83p
authors: Vimercati, Luigi; De Maria, Luigi; Quarato, Marco; Caputi, Antonio; Stefanizzi, Pasquale; Gesualdo, Loreto; Migliore, Giovanni; Fucilli, Fulvio Italo Maria; Cavone, Domenica; Delfino, Maria Celeste; Sponselli, Stefania; Chironna, Maria; Tafuri, Silvio
title: COVID-19 hospital outbreaks: Protecting healthcare workers to protect frail patients. An Italian observational cohort study
date: 2020-11-04
journal: Int J Infect Dis
DOI: 10.1016/j.ijid.2020.10.098
sha: 50c086727f5522eeab43bc7dfc945eea4b23e812
doc_id: 1050204
cord_uid: nyq2l83p

OBJECTIVES: to determine the prevalence of SARS-CoV-2 infection among exposed HCWs after preventive protocol implementation. METHODS: 5750 HCWs. Those in contact with COVID-19 patients were allocated into a high-risk or a low-risk group based on contact type (PPE- or non-PPE-protected); high risk workers underwent nasopharyngeal swab tests, while for low-risk workers swab tests were carried out only for symptomatic workers (active surveillance). The prevalence was determined by real-time reverse transcriptase–polymerase chain reaction on nasopharyngeal samples. RESULTS: 3570 HCWs had contact with 1065 COVID-19 patients. Among them, 3494 were subjected to active surveillance (low-risk group); 2886 (82.60%) were subjected to a swab test and 15 were positive (0.52%). 76 HCWs (2.13% of exposed) were included in the high-risk group, and a swab test was mandatory for each participant. Overall, 66 (86.84% of high-risk) were negative, and 10 positive (13.16%), resulting in a higher risk of infection than in the low-risk group [OR = 29.00; 95% CI:12.56-66.94; p < 0.0001]. CONCLUSION: To date, the SARS-CoV-2 infection prevalence is 0.70% among exposed HCWs and 0.435% among all HCWs working at the examined university hospital. The correct use of PPE and the early identification of symptomatic workers are essential factors to avoiding nosocomial clusters.

As May 20, 2020, a high severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection rate among healthcare workers (HCWs) was described: among them, 26657 were infected, corresponding to 11.7% of all Italian cases. In this regard, almost 80% worked in the hospital setting or in an extra-hospital emergency system (ISS 2020).

SARS-CoV-2 represents a major hazard for HCW safety: the biological risk is higher when medical procedures are performed on the respiratory tract, e.g., applying respiratory devices such as oxygen cannulas or noninvasive ventilation (Ferioli et al. 2020 ). To prepare for a pandemic, infection prevention measures should be focused on reducing in-hospital transmission using correct personal protective equipment (PPE) (Wong et al. 2020 ), but their utilization needs to be appropriate to avoid waste. Moreover, no drug has shown effectiveness in preventing infection among patients and HCWs (Mehra et al. 2020 ).

To date, nasopharyngeal and oropharyngeal swabs are the only useful tools to detect infections in association with reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) amplification, but the correct population of HCWs subject to screening is not clear, and an underestimation of SARS-CoV-2 prevalence is possible (Kluytmans-van den Bergh et al. 2020); antibody detection in serum is currently used as a complementary test in epidemiological studies but is not helpful for diagnosis ). Taking advantages from postponed Sars-Cov-2 outbreak in South Italy and building on Lombard experience (Fagiuoli et al. 2020) , in early March, our Operative Unit of Occupational Medicine activated a preventive protection protocol to early isolate infected workers and analysed the infection rate after a 50-day observation period at the University Hospital of Bari, one of the major COVID-19 hub centres in southern Italy.

The aim of this study was to determine the prevalence of SARS-CoV-2 infection among HCWs after preventive protocol implementation in order to verify if this strategy could be a valid alternative to massive RT-PCR screening.

J o u r n a l P r e -p r o o f

A cohort study was carried out for all 5750 HCWs (doctors, nurses, social health assistants, technicians, administrative employers) of the University Hospital of Bari, southern Italy, starting in the early phase of the pandemic and ending the observation period after 50 days. Observation started on March 11, 2020, and ended on April 29, 2020. Informed and written consent was obtained from all participants. All subjects were informed that data from the research protocol would be treated in an anonymous and collective way, with scientific methods and for scientific purposes in accordance with the principles of the Declaration of Helsinki. Ethical approval was not necessary because all medical and instrumental examinations were performed according to Italian law concerning the protection of workers exposed to occupational risks (D.Lgs. 81/2008). Our study was compliant with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.

A preventive protocol was implemented, consulting the latest guidelines published by the World Health Organization (WHO) (2020b) and the Centers for Disease Prevention and Control (CDC) (2020c). Current definitions of suspected, probable and confirmed cases, as defined in the protocol by the European Centre for Disease Prevention and Control (ECDC), were adopted, as well as the assessment of "close contact" and "casual contact" (ECDC 2020), to identify HCWs exposed to a significant biological risk.

According to the WHO guidelines, general prevention measures were reinforced (WHO 2002), in particular infection source control to stop the transmission chain of pathogens in hospital settings (e.g., no handshakes, no coughs without surgical mask), environmental controls to prevent nosocomial infections (e.g., more than 12 air changes per hour, the disinfection of surfaces), and adequate utilization of personal protective equipment (PPE). In this regard, all HCWs were educated about correct donning and doffing procedures, which were carried out anteroom before and after making contact with suspected or confirmed cases and guided by warning signs to correctly perform the actions (CDC 2020a; Spinazzè et al. 2020).

For each participant having contact with a confirmed COVID-19 case, occupational or environmental exposure was annotated, and careful history was performed. Considering high costs of RT-PCR in a massive long-term screening and the hurry to identify infected workers in the early stage of COVID-19, we performed a risk assessment for each worker and correct PPE utilization was adopted as essential factor to define high or low infection risk: nasopharyngeal swab test in home isolation was mandatory for non-PPE guarded contacts and test was performed at least 7 days after J o u r n a l P r e -p r o o f hazardous contact in order to reduce false negative results; on the other hand, "active surveillance" was assumed as adequate for PPE-guarded contacts and included monitoring body temperature twice daily and screening everyone for evocative symptoms (anosmia, ageusia, fever, severe weakness, sore throat, rhinorrhoea, cough, diarrhoea, dyspnoea). A reporting system was activated to collect and monitor all HCW contacts with confirmed cases: in each hospital ward, a trained person in charge, a.k.a. "prevention officer", verified correct PPE handling, reported contact type with COVID-19 patients to the Operative Unit of Occupational Medicine and performed "active surveillance" for all workers in each the wards of the medical, surgical and administrative areas. Nasopharyngeal and oropharyngeal swabs were collected and stored in a sterile tube and analysed in the Hospital Virology Laboratory. Diagnostic testing for SARS-CoV-2 and specimen collection were carried out following CDC guidelines (2020b). We performed reverse transcriptase quantitative PCR (RT-qPCR) targeting SARS-CoV-2 RNA, following WHO recommendations (2020a): if the test was positive, the specimen received a second-level evaluation at the Italian College of Health Laboratories. Each positive worker was dismissed from work and retested until two consecutive negative results were received; the results of follow-up swab tests in positive patients were not included in the study. Data were collected using a Microsoft Excel Office 2019 (Microsoft Corporation™) form.

A simple univariate and descriptive analysis (Pearson's χ2, Goodman and Kruskal's gamma) and the estimated asymptotic standard error (ASE) were carried out. Only dichotomized variables were created for risk of infection, job titles and age classes because these showed high levels of χ2 in their univariate analysis, and odds ratios were calculated. We used Stata 12 software (Stata Corp LLC, Texas, USA). A P-value < 0.05 was considered statistically significant. Among these, 3494 (97.87% of exposed workers) were PPE protected, and active surveillance was implemented (low-risk group). During observation, 608 HCWs (17.40% of HCWs belonging to the low-risk group) had no suspicious symptoms, and no further investigations were performed; however, 2886 HCWs under active surveillance (82.60% of HCWs belonging to the low-risk group) complained of at least one evocative symptom and were subjected to nasopharyngeal swab tests. In On the other hand, 76 HCWs (2.13% of exposed workers) had one close contact that was not PPE protected (high-risk group); particularly, one out the 76 was a physician who wrongly handled her safety goggles due to eye rubbing after a COVID-19 patient examination, causing a non-PPEprotected contact. Moreover, 25 out 76 were in contact with an infected colleague in a medical briefing and became symptomatic under active surveillance a few days later. Finally, 50 were cleaners who had minimal PPE equipment because they were not assigned to patients' care: prevention officers used great caution for them, identifying each as a close or not-close contact. All high-risk group workers were forced to self-quarantine in home isolation and, after 7 days, underwent a nasopharyngeal swab test: 8 were immediately positive and 68 were negative, but 2 of these latter individuals, included in active surveillance, showed evocative symptoms and were retested, resulting in positive results. The high-risk group overall had 10 positive workers (13.16% of the total high-risk group), and the risk of infection appeared statistically higher than the prevalence found in the low- 

The majority of the 25 positive-swab workers [n=15 (60.00%)] were associated with a "community exposure" outside the hospital; 10 of them were related to an "in-hospital exposure", but not during working tasks [n=10 (40.00%)].

The dichotomized variables showed a highly significant risk of positive tests only for physicians [OR=2.80 (1.26-6.22)] for job title analyses, and a statistically significant association was found only for the age group of 51-60 years [OR=1.58 (0.72-3.46)] for age class analyses.

To date, the SARS-CoV-2 infection rate is 0.70% among exposed HCWs and 0.435% among all HCWs at the examined university hospital in a 50-day observation period, and these data are lower than those in the most recent report (Kluytmans-van den Bergh et al. 2020). Overall, 3109 swab tests were performed on 2962 exposed HCWs during protocol implementation (Fig 2) , and the ratio of positive to negative tests was 0.80%.

In the same period of protocol activation, overall 1065 patients with confirmed cases of COVID-19 were admitted to the Emergency Department, and 346 were hospitalized due to their poor conditions: none of the infected HCWs needed hospital care, but all of them developed a mild viral illness. In contrast, Wang et al. (2020) found in a retrospective single-centre case series that, among approximately 138 COVID-19 patients, 29% were estimated to be HCWs working in the same hospital centre and were infected during shifts. After preventive protocol implementation, a small group of HCWs reported non-PPE-protected high-risk contacts with COVID-19 patients (76 workers): among these, 50 were cleaners working in a non-COVID-19-dedicated area, and for them, the Risk Assessment Document prescribed minimal PPE equipment. In this specific case, preventive officers used great caution, reporting that hazardous contact had occurred for each worker, broadly interpreting the ECDC definition of "close contact" (ECDC 20209). However, no cleaner results were positive, likely due to no actual close contacts. The ten high-risk positive HCWs were carefully examined to understand the infection modality: one wrongly handled her PPE, while 9 were infected in a medical briefing through non-PPE-protected close contact with one infected colleague. In the aforementioned briefing, 26 people were assembled: the infector manifested evocative symptoms during active surveillance and was rapidly tested for COVID-19, with positive results: all 25 colleagues were forced to self-quarantine and were tested while in home isolation, and 9 were positive (two after the second swab test). The hospital ward was closed for environmental sanitation. In the low-risk group, during active surveillance, 15 symptomatic HCWs were positive for SARS-CoV-2, two after the second swab test due to the occurrence of new symptoms; for them, environmental and non-occupational infection modalities were described due to close contacts with infected relatives J o u r n a l P r e -p r o o f (Table 2 ). In the latter group, the risk of infection was statistically lower than that in the high-risk group, showing that correct PPE use avoided the spread of SARS-CoV-2 infection among HCWs despite daily close contacts of these workers with COVID-19 patients; moreover, the correct use of PPE protected HCWs in the low-risk group from infected colleagues, similar to a kind of "PPEguaranteed herd protection". Indeed, no occupational infection during work tasks occurred in hospital wards characterized by higher biological risk (pulmonology, internal medicine, and intensive care units) despite the remarkable number of hospitalized COVID-19 patients, but only environmental spread due to incorrect hygienic practises (non-use of PPE, contact with infected kinsmen, handshakes). Curiously, 64% of infected workers were physicians, and they had an increased risk of infection compared to other healthcare professionals. Based on anamnesis collection, these data could reflect a low risk perception out of the hospital setting and a high frequency of wrong behaviour. hospitals (Blake et al. 2020); therefore it is possible to presume that the high prevalence of symptoms could be also linked to this mental condition. Work-related stress not only affects the physician's mental and physical well-being, but also patient care quality and the overall efficiency of the healthcare system. On this basis, establishment of preventive protocols is essential to provide safe guidance for HCWs and consequently reduce their work-related stress. On the other hand it should also be considered the role of some psychological approaches such as stigmatization and discrimination on the adherence to the preventive protocols (Baldassarre et al. 2020 ). The fear of stigmatization and the risk of being subject to quarantine measure might delay the contact tracing, contributing to the virus spread. As for the symptoms, half of the infected HCWs complained of ageusia and anosmia as early symptoms; therefore, great attention should be given to them. On the other hand, 4 out of 25 infected workers remained asymptomatic: their isolation is crucial to avoid nosocomial clustering, particularly among hospitalized frail patients (Gandhi et al. 2020) . In this regards, massive RT-PCR testing, as well as recommended by Fagiuoli et al. (2020) is helpful to identify infected worker but the test could give false negative in early stage of disease, allowing these workers to keep working and infecting other HCWs. We suggest that careful compliance to correct PPE utilization and biological risk stratification is more helpful to avoid nosocomial cluster, keeping high risk workers in home isolation as soon as possible after hazardous contacts, before RT-PCR test could detect viral genome. The very low infection rate discovered among exposed HCWs in our protocol supported this hypothesis.

Finally, 97.87% of exposed workers, corresponding to 3494 people, had PPE-protected contacts, pointing out that there were enough PPE in hospital stores: hygienic measures to avoid overcrowding (e.g., reducing the number of HCWs in rooms hosting confirmed cases), PPE use for the maximum usage time, and nasopharyngeal swab execution by only one HCW per working shift are essential factors in our protocol to avoid waste.

This study has several limitations. First, no data about environmental contacts with COVID-19 people were available for 2180 HCWs who did not have in-hospital close contacts with COVID-19 patients; therefore, the prevalence of infection in workers due to environmental exposure is not clear in this group and the overall prevalence could be underestimated. However, no worker in this group complained to their wards' coordinators of any evocative symptoms after the observation period and any nosocomial cluster was declared to Unite of Occupational Medicine in the same period. Second, low-risk asymptomatic HCWs in active surveillance were allowed to work, and nasopharyngeal swab tests were not performed in these individuals; however, the possibility of asymptomatic patients among workers is possible (Gandhi et al. 2020) , and, therefore, the observed prevalence of infection of 0.70% in exposed HCWs could be underestimated. Third, although preventive officers were trained to recognize incorrect PPE utilization and evocative COVID-19 symptoms, overreporting or underreporting of these data could be possible (self-report bias). Overreporting of too much data would have resulted in the implementation of active surveillance and testing for SARS-CoV-2 in more individuals without infection, leading to excessive resource consumption. On the other hand, the underestimation of the aforementioned data could have led to less testing and the underestimation of the prevalence of SARS-CoV-2 infection in the overall group of HCWs. Fourth, data about the presence of antibodies in HCWs against SARS-CoV-2 were not available; however, massive immunological screening is being carried out in the same cohort to better understand the real immunization of employees.

Italy has a high number of infected HCWs, almost 80% of them working in a hospital setting. The execution of this protocol has shown good results in our hospital, recording only 25 HCWs infected by SARS-CoV-2 over 3109 swab tests, despite the high number of patients admitted to the ED and hospitalized in the period of observation. The correct use of PPE is an essential step to better assign the correct risk class to HCWs, determining whether active surveillance or home isolation is the most J o u r n a l P r e -p r o o f appropriate choice. Moreover, the careful identification of symptomatic workers is necessary to avoid missing people with mild respiratory infection and to avoid nosocomial clusters.

Funding source: This research did not receive any specific grant from funding agencies in the public, commercial or nofor-profit sectors.

The flowchart extensively describes the preventive protocol initiated in the cohort of 5750 workers. Between March 11 and April 29, 3570 HCWs (62.09% of the total) had contact with 1065 COVID-19 patients. Based on preventive officers' reports, 3494 HCWs, having PPE-protected close contacts (97.87%), were subjected to active surveillance (Low-risk Group): among them, 2886 (82.60%) manifested at least one evocative symptom (described in table 2) and were subjected to a swab test; 2871 (99.48% of low-risk symptomatic HCWs) were negative, and only 15 HCWs (0.53%) were positive, of whom 2 were positive after the second test. On the other hand, 76 HCWs were included in the high-risk group due to non PPE-protected contacts, and a swab test was mandatory for each participant. Overall, 66 HCWs (86.84% of the highrisk group) were negative, and ten were positive (13.16%), of whom 2 had a statistically higher risk of infection after the second test than the low-risk group [OR= 29.00; CI95%:12.56-66.94; p<0.0001]. In yellow boxes, checkpoints are performed by prevention officers. 

Stigma and Discrimination (SAD) at the Time of the SARS-CoV-2 Pandemic

Mitigating the Psychological Impact of COVID-19 on Healthcare Workers: A Digital Learning Package

Work-related stress levels and musculoskeletal disorders among municipal solid waste collectors in Ankara

Hand Hygiene Guideline. Page reviewed online on

Interim guidelines for collecting, handling, and testing clinical specimens from persons for coronavirus disease 2019 (COVID-19)

S. guidance for risk assessment and public health management of healthcare personnel with potential exposure in a healthcare setting to patients with coronavirus disease (COVID-19)

Work-related stress risk and preventive measures of mental disorders in the medical environment: an umbrella review

Perceived stress, disturbed sleep, and cognitive impairments in patients with work-related stress complaints: a longitudinal study

European Centre for Disease Prevention and Control. Case Definition and European Surveillance for Covid-19, as of 2

Covid-19 Bergamo Hospital Crisis Unit. Adaptations and Lessons in the Province of Bergamo

Protecting healthcare workers from SARS-CoV-2 infection: practical indications

Asymptomatic Transmission, the Achilles' Heel of Current Strategies to Control Covid-19

Italian College of Health (Istituto Superiore di Sanità -ISS)

Prevalence and Clinical Presentation of Health Care Workers With Symptoms of Coronavirus Disease 2019 in 2 Dutch Hospitals During an Early Phase of the Pandemic

The association of work stress with somatic symptoms in Chinese working women: a large cross-sectional survey

Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis

COVID-19 outbreak in Italy: protecting worker health and the response of the Italian Industrial Hygienists Association. Ann Work Expo Health

Temporal profiles of viral load in posterior oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort study

Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

Preparing for a COVID-19 pandemic: a review of operating room outbreak response measures in a large tertiary hospital in Singapore

Prevention of hospital-acquired infections: A practical guide

2019-nCoV) in suspected human cases

World Health Organization. 2020b. Novel Coronavirus (2019-nCoV)

All authors participated in the drafting and revision of the work and gave the final approval of the version to be published. All authors agreed to be responsible for all aspects of the job in ensuring that issues relating to the accuracy or integrity of any part of the job are properly investigated and resolved.Patient Consent: Informed and written consent was obtained from the participants. The patients were informed that data from the research protocol would be treated in an anonymous way, with scientific methods and for scientific purposes in accordance with the principles of the Helsinki Declaration. Ethical Approval is not necessary because all the medical and instrumental examinations were performed according to the Italian laws concerning the protection of workers exposed to occupational risks (D. Lgs. 81/2008).

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.J o u r n a l P r e -p r o o f

3 (30%) 3 (20%) Cough, n (%) 6 (60%) 8 (53.33%) Conjunctivitis, n (%) 1 (10%) 5 (33.33%) Diarrhoea, n (%) 6 (60%) 9 (60%) Headache, n (%) 1 (10%) 1 (6.67%) Abdominal pain, n (%) 0 (0%) 1 (6.67%)