Personal protective equipment used by obstetricians and obstetric nurses during the COVID-19 pandemic in Mansoura, Egypt

Background: This study was done to describe the pattern of personal protective equipment (PPE) use, associated factors, and adverse events among obstetricians and obstetric nurses in obstetrics & gynecology departments. Methods: A cross sectional study was conducted in Obstetrics & Gynecology departments in three hospitals (physician & nurses n=252) using an online Google form including demographic and occupational health data, type of available personal protective equipment during usual care, CS and emergency labor, infection control measures and hazards of full PPE use. Results Full PPE use was 37.7% during CS and 34.9% during emergency labor. The significant predictors of wearing full PPE during CS were daily work hours > 8 hours and receiving formal training about PPE use. During CS & emergency labor most of HCws used sterile gloves and sterile fluid resistant gowns and surgical mask.to less extent used face shields or tight fitting googles and one tenth (11.8%) only used N95. The most common health effects of full PPE use was sense of heat (79.5%) . Conclusion: During the COVID-19 pandemic more vigorous respiratory (N95 mask) and eye protection is required during aerosol-generating procedures. Formal training is an evident predictor for full PPE use.


Introduction
COVID-19 is a global pandemic affecting all populations, and subsequent worldwide aggressive measures have been taken to mitigate the spread of the infection (World Health Organization (WHO), 2020). According to a study on COVID-19 characteristics and predicting factors among healthcare providers in a developing country, approximately two-thirds of exposure occurred primarily during healthcare provision. Furthermore, one-fifth of the cases were confirmed COVID-19 cases; the majority of them had mild to moderate symptoms, with only 9.1 percent asymptomatic. Almost all became infected while on duty (97.4 percent). Infected patients (39%) were the most common source of infection, followed by colleagues (22.1%), household contacts (5.2%), and unknown sources (33.8 percent) (El-Sokkary et al., 2021).
Obstetricians have worked tirelessly for the past three decades to enhance women's outcomes through the use of evidence-based medicine. The present COVID-19 epidemic has flooded this standardized method with a deluge of contradicting material, causing uncertainty in the birthing ward about recommended practices. As the infection spreads throughout the population, obstetricians are becoming more concerned (Sichitiu & Desseauve, 2020).
Asymptomatic patients are contagious and thus are at a high risk of nosocomial infection (Rothe et al., 2020). Therefore, strict PPE usage for doctors and midwives is necessary at labor if universal screening is not performed (Umazume et al., 2020).
Regulating and adapting all aspects of infection prevention can be difficult in the unfortunate event of a maternal collapse. The delivery room is overloaded when many personnel simultaneously attempt to resuscitate the collapsed patient, perform a perimortem cesarean delivery, and resuscitate the newborn. The resuscitation team should don full PPE. The most common frequent & serious cross-infection to healthcare workers during outbreaks happened when first responders were not wearing the recommended PPE (Schwartz and Graham, 2020).
Wearing N95 respirators can reduce clinical respiratory infections by 73 per 1,000 healthcare workers (HCWs). In laboratory-confirmed bacterial colonization, N95 respirators had a protective effect. N95 respirators were found to be more effective in preventing laboratory-confirmed respiratory viral infections and influenza-like illnesses. As regards protection of HCWs, no direct high-quality evidence was found on whether N95 respirators are better than surgical masks for from SARS-CoV-2 (Iannone et al., 2020).
Despite the protective PPE value, one study reported that physicians who work in the intensive care unit (ICU) and deal with such patients are naturally anxious. They would be unable to drink, eat, or go to the toilet for approximately 6 hours after wearing full PPE as required in the ICU. Taking off PPE after duty hours necessitates training and extreme caution to avoid self-infection (Alsubaie et al., 2019).
Worldwide, reports of PPE scarcity and unavailability are emerging. HCWs and the general press use social media to report on the reusing of PPE or the use of household and self-made items in place of PPE. There is limited evidence on the effectiveness of these practices, but they have occasionally been implemented on the advice of their employers or health organizations (England, 2020;CDC, 2020). It is believed that no previous studies discuss the pattern of PPE utilization among obstetricians and obstetric nurses during the COVID-19 pandemic in our country.

Study objectives
This study aims to describe the pattern of PPE use, associated factors, and adverse events among healthcare workers in obstetrics and gynecology departments.

Study design
Observational cross-sectional study with an analytic component.

REVISED Amendments from Version 1
There is no major differences in the submitted manuscript and new version; only few clarifications in logistic regression results text of and statistical analysis paragraph and type of snowball sampling technique.
Any further responses from the reviewers can be found at the end of the article

Study place and time
The study was conducted in Obstetrics & Gynecology Departments in three hospitals (one is affiliated to Mansoura University and two affiliated to the Ministry of Health & Population) from August 2020 to June 2021.

Study participants
The study included on-duty obstetric physicians and nurses during the COVID-19 pandemic. Participation was voluntary and anonymous.
The sample size was calculated by openepi.com, an online sample size calculator. It was calculated according to the primary outcome of interest which is the anticipated frequency of full PPE use (30%) based on a pilot study of 100 subjects. The current study has a target population (~954 workers), power of the study (80%), and precision degree (5%). The sample size is 242, and 10% was added to cover nonresponse from online surveys. Thus, the final sample was 266 subjects.

Data collection approach
The current study collected, first, personal communication between one of the authors and study subjects in selected hospitals to encourage participation and explain study objectives. Second, an online questionnaire was distributed (n = 266) through posting Google links to WhatsApp personal accounts or WhatsApp groups of their departments. Snowball sampling technique (exponential non discriminative) was used by asking participants to share the link with other healthcare workers in the obstetrics and gynecology departments in their hospitals. Finally, the response rate was 252 (79.2%).

Study tool
The questionnaire included demographic data; occupational history; type of available personal protective equipment during usual care, Cesarean section (CS), and emergency labor; infection control measures related to the use and discard of equipment; and hazards of full PPE during COVID-19 pandemic.
Operational case definition of full PPE use was adapted from the national baseline resources and recommendations of the International Society of Ultrasound in Obstetrics and Gynecology, Royal College of Obstetricians and Gynaecologists, Collège National des Gynécologues et Obstétriciens Français (Poon et al., 2020;Royal College of Obstetricians and Gynaecologists, 2021;Peyronnet et al., 2020). The definition of PPE use in the current study included using gown with long sleeves, gloves (sterile gloves for CS and latex gloves for usual care), respiratory protection (N95 mask or FFP or surgical mask), eye protection (face shield or goggle), and foot protection (overshoes or safety boots).

Ethical considerations
The study was approved by the Research Ethics Committee of the Faculty of Medicine of Mansoura University (code number: R/20.7.925). Informed verbal consent was obtained from the administration of hospitals affiliated with the Ministry of Health and Population. All study participants were assured of the confidentiality and anonymity of the data at the start of Google form, and participation was voluntary.

Statistical analysis
Data were analyzed using SPSS, version 23. Categorical variables are presented as numbers and percentages, and the chisquared test was used for comparison between groups. Crude odds ratios and their 95% confidence intervals (CIs) were calculated. Moreover, quantitative variables are presented as means and standard deviations. Binary stepwise logistic regression analysis was used to determine the independent predictors of full PPE use as the dichotomous outcome variable. Variable found to have statistical significance in bivariate analysis were entered into the logistic regression analysis using a Forward-Wald model. Model fitness by Hosmer and Lemeshow test was non-significant (p value > 0.05) which indicates Good fit for model. Adjusted odds ratios and their 95% CIs were calculated. A p value ≤ 0.05 was statistically significant.

Demographic characteristics
The study included 252 healthcare workers in the obstetrics and gynecology departments. Most of the studied participants were females (76.2%), has a mean age of 32.2 (9.9), and nearly half of them are from urban areas (47.2%). About two-thirds (57%) of the studied HCWs were from Mansoura University Hospitals. More than half were nurses (56.7%) and 37.7% were physicians. The median work duration in years and median daily work hours were 7 years and 8 h, respectively. Onefifth of them (20.2%) reported work in hospital isolations. Most of them reported exposure to COVID-19 patients (75.5%). However, the polymerase chain reaction-confirmed COVID-19 infection was about one-third (27.3%) of the studied HCWs. This percentage is exactly half of those were symptomatic workers (50.9%; data are not tabulated).
Factors associated with full PPE use Full PPE use was 37.7% and 34.9% during CS and emergency labor, respectively. Full PPE use during CS was significantly associated with longer work hours (≥8 h), working in hospital isolation, and receiving formal training about PPE use (p < 0.05). Moreover, it was associated with working in hospital isolations and receiving formal training during emergency labor (p < 0.05; Table 1) (Khashaba, 2022).  Table 2).

Pattern of PPE use & infection control measures among study participants
During CS, most healthcare workers used sterile gloves (88%), sterile fluid-resistant gowns (75.6%), and surgical masks (79.2%). About half (52.6%) used face shields. About one-third of them used tight-fitting goggles (34.3), and one-tenth (11.8%) only used N95. Moreover, during emergency labor, most healthcare workers used sterile gloves (81.1), sterile fluid-resistant gowns (70.7), and surgical masks (83.7%). Washing gloves and reusing gowns were reported in 26.8% and 14.8% of HCWs, respectively. Most of the study subjects reported changing gloves between patients (87.7%). More than two-thirds of the subjects followed the correct sequence of donning, removing PPE, and receiving formal training on the use of PPE (67.1%, 66.4%, and 63.5%, respectively; Table 3).

Adverse health effects of full PPE use
The most common health effect of full PPE use is a sense of health (79.5%) followed by a sense of thirst and pressure areas (64.7% and 64.3%, respectively; Figure 1). More than two-thirds of affected HCWs (68.3%) had more than two symptoms. Adverse events were not associated with age, gender, job description, duration of work in years, work hours, or work in hospital isolations (p > 0.05; data are not tabulated).

Discussion
This is the first study on PPE use among healthcare workers in the obstetrics and gynecology departments during the COVID-19 pandemic in Egypt. PPE use is influenced by the workload of HCWs as a matter of hours and patients, baseline resources in these hospitals, and stockpiling of personal equipment and safety training for those HCWs (Tabah et al., 2020).
The present study showed that full PPE use was 37.7% and 34.9% during CS and emergency deliveries, respectively. Full PPE use during CS was significantly associated with longer work hours (≥8 h) and formal training on PPE use. A different description for the pattern of full PPE per facility was noticed by a nationwide survey in Japan including core facilities and affiliated hospitals of obstetrics and gynecology training programs. Authors reported that full PPE was used by doctors and midwives in 7.1% and 6.8% of facilities, respectively, taking into consideration the different definitions of full PPE use in both studies.
The present study showed that most healthcare workers used sterile gloves, sterile fluid-resistant gowns, and surgical masks during CS and emergency labor. Fewer HCWs used face shields and goggles. Only one-tenth used N95 due to the absence of baseline resources of these hospitals. Of the HCWs, less than one-third reported washing gloves (26.8%) and reusing gowns (14.8%). Surgical masks and N95 respirators are the most consistent and complete support measures used by healthcare personnel, according to a Cochrane-approved systematic review on physical interventions to reduce respiratory virus transmission undertaken in 2011. N95 respirators are non-inferior, according to the highest quality cluster-randomized controlled studies included in this systematic review (Jefferson et al., 2011).
Another web-based survey conducted by Tabah et al. (2020) and distributed worldwide in April 2020 among ICU HCWs found that more than half of the respondents (52%) reported that at least one piece of the standard PPE is not available, and 30% reported that at least a piece of single-use PPE was being reused or washed as a result of shortages. Most of the available PPEs were designed for single-use and brief duration. Hence, the authors reported that urgent design and manufacture of PPE that can be safely worn and remains effective for extended durations are needed.
Similarly, the first nationwide Japanese survey reported that N95 masks and goggles or face shields were out of stock in 6.5% and 2.7% of facilities, respectively. In addition, disposable N95 masks and goggles or face shields were reused after resterilization in 12% and 14% of facilities, respectively (Umazume et al., 2020).
Masks and respirators play a role in the protection of health workers according to high level of agreement among key agencies. However, there are current differences between these agencies in terms how and when the different products are used. Different recommendations have been made by the World Health Organization (WHO), the US Centers for Disease Control and Prevention (CDC), and other leading health organizations. For example, the WHO recommends using N95, FFP2, FFP3 standards or equivalent in care settings for COVID-19 patients where aerosol-generating procedures are used and medical masks are used in the absence of aerosol-generating procedures (WHO, 2020). In contrast, the US Centers for Disease Control and Prevention (Center for Disease Control and Prevention, 2020) recommend using respirators during both routine care of COVID-19 patients and high-risk situations.
In this study, more than two-thirds of the subjects followed the correct sequence of donning and removing PPE and receiving formal training on PPE use (67.1%, 66.4%, and 63.5%, respectively). These results came lower than those observed among ICU workers in a worldwide web-based survey who found that most of the respondents (83%) had formal training in PPE use. This included training at the start of the institution (13%) and within the last two months (60%) due to the COVID-19 pandemic (Tabah et al., 2020).
In addition, current study results revealed that receiving formal training on PPE use among obstetric physicians and nurses was a significant predictor of full PPE use together during CS and emergency labor. These results are in agreement with an Italian study on a cross-section of physicians which reported that access to adequate information on the use of PPE was associated with a better ability to perform donning and doffing procedures as an example for proper PPE use (Savoia et al., 2020).
The most common health effects of full PPE use in the current study is a sense of heat (79.5%) followed by a sense of thirst and pressure areas (64.7 and 64.3%, respectively). More than two-thirds of affected HCWs (68.3%) had more than two symptoms. These results were different from the results found in ICU workers during the pandemic where 80% had adverse events including heat (51%), thirst (47%), pressure areas (44%), headaches (28%), inability to use the bathroom (27%), and extreme exhaustion (,20%). They were all associated with longer duration of shifts wearing PPE. However, in the current study, adverse events were not associated with age, gender, job description, duration of work in years, work hours, or work in hospital isolation (Tabah et al., 2020).

Conclusions
Most obstetricians and obstetric nurses used surgical masks, gloves, gowns, foot protection, and, to a lesser extent, N95, goggles, and face shields. Work hours (>8 h) and formal training were significant predictors for full PPE use.
Based on the findings of the current study, adding N95 or FFP mask and eye protection for safety programs in obstetrics and gynecology departments during aerosol-generating procedures is recommended. Also, reducing long shift work hours that may lead to adverse events due to full PPE use is recommended. Providing simple training videos about different types, proper PPE donning and removal, disinfection of reused equipment, and hand hygiene frequently as part of CME hours of physicians and nurses are required to cope with this emerging threat.

Limitations of the study
The evaluation of PPE sufficiency during the COVID-19 pandemic was not assessed in this study. Further research is required to address PPE protective value in infected HCWs compared to non-infected controls.

Open Peer Review
The methodology section needs to be elaborated further. Since three hospitals were studied, what was the representation of the study population in the final sample itself? Moreover, were the original proportion of doctors and nurses in these 3 hospitals maintained in the final sample?
Ideally, two sampling frames are ideal, or one sampling frame containing both doctors/nurses. Why was the snowball sampling technique used in this study? It is generally used to reach hard to reach study subjects. That is not the case in this study.
Moreover, which type of snowball sampling technique was used -Linear/Exponential, Nondiscriminative/Exponential discriminative?
The output of the logistic regression was also incomplete. What about the model fitness? The variation of dependent variables explained from independent variables? What proportion can be explained from the model? Please mention as running text in the results section.
The final model was a multivariable logistic binary regression model. Please mention this.
I would recommend replacing the term 'study subjects' with 'study participants'.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes significant reservations, as outlined above.

Author Response 30 Apr 2022
Eman Khashaba, Faculty of Medicine,Mansoura University, Mansoura, Egypt Response to reviewer's comments Thanks a lot for the valuable and informative comments.
The methodology section needs to be elaborated further. Since three hospitals were studied, what was the representation of the study population in the final sample itself? Moreover, were the original proportion of doctors and nurses in these 3 hospitals maintained in the final sample?
Reply: The actual number of doctors and nurses in each hospital is not of concern as we aimed to complete the sample size calculated. Also the number on duty is not fixed due to frequent sick leaves, and many of them are not regular users of mass media to complete questionnaire in due time. In general, the number of the nurses is more than the doctors in original sample.
The proportion of each hospital was not kept in the final sample. This can be added TO STUDY LIMITATIONS. Ideally, two sampling frames are ideal, or one sampling frame containing both doctors/nurses. Reply: Yes I agree we can add "associated factors".

Reply: For our study, it is one sampling frame containing both doctors/nurses
Why was the snowball sampling technique used in this study? It is generally used to reach hard to reach study subjects. That is not the case in this study.