COVID-19 Among Doctors in Indonesia: Risk of Moderate-Severe and Behavior Adaptation in Personal Protective Equipments Usage After Infection

Doctors have a greater risk of acquiring COVID-19 due to occupational exposure. Personal protective equipment (PPE) is an essential factor in reducing COVID-19 transmission. We aimed to evaluate the risk of moderate-severe COVID-19 infection and behavior adaptation in PPE usage among doctors who survived COVID-19 in Indonesia. This was an online population-based cross-sectional survey among Indonesian doctors of COVID-19 survivors. Bivariate and multivariate analyses were performed to determine factors associated with moderate-severe COVID-19 infection. A total of 389 doctors who survived COVID-19 infection across in Indonesia was included in this study. Most participants were young doctors (20-39 years: 69.7%), general practitioners (50.4%), working in COVID-19 designated hospitals (62.5%), and worked more than 40 hours/week (57.8%). Factors associated with moderate-severe COVID-19 were IMA moderate occupational risk (aOR 4.14, 95% CI: 1.11-15.47), age 40-59 years (aOR 3.24, 95% CI: 1.99-5.29), working in COVID-19 designated hospital (aOR 1.89, 95% CI: 1.18-3.01), and higher BMI (aOR 1.88, 95% CI: 1.00-3.54). N95 respirator and other PPEs use improved after these doctors recovered from COVID-19 infection in isolation and non-isolation rooms. In conclusion, working COVID-19 designated hospital, moderate occupational risk, higher BMI, and age 40-59 were associated with moderate-severe COVID-19 among doctors in Indonesia.


Introduction
Coronavirus disease 2019 (COVID-19) has become a global public health threat and evolved into a worldwide pandemic crisis 1 . World Health Organization (WHO) declared COVID-19 as public health emergency on 2020 January 30 and called of all countries to prevent the rapid spread of COVID-19 2 .
Indonesia is a country that was hit hard by this pandemic. Con rmed cases from the beginning of the pandemic until June 2021 reached 1.531.005 with 83.700 (5.5%) active cases, 1.404.639 (91.7%) recovered cases and 42.666 (2.8%) mortality cases 3 . The Government has implemented public health protocols to control the spread of the virus, such as physical and social distancing, hand washing, and mask use, but the active cases are still increasing 4 . The public healthcare systems are not prepared to face this devastating situation.
Healthcare workers (HCWs) have a higher risk of acquiring COVID-19 due to increased occupational exposure to SARS-CoV-2 5,6 . Within three months (March-May 2020), the Centers for Disease Control and Prevention (CDC) reported that 5.9% of the hospitalized patients in United States were health workers 7 .
The lack of accurate information regarding disease management, limited availability of adequate personal protective equipment (PPE), limited diagnostic test kits, psychosocial factors such as an unsupportive work environment, and excessive workload increased the number of HCWs-infected COVID- 19, especially in the early of the pandemic 8- 10 . There was no o cial report for the number of HCWsinfected with COVID-19 in Indonesia. The rst case of an Indonesian doctor died of COVID-19 was published on March 22, 2020. In June 2021, the Indonesian Medical Association (IMA) reported that there were 374 doctor deaths due to COVID-19 11 .
Several factors that known to increase the severity of COVID-19 infection are age, pre-existing comorbidities and complications, obesity, and smoking status [12][13][14][15] . Health care workers might have additional risk factors that increase the severity of COVID-19 infection, ie excessive workload and reuse of disposable PPE 16,17 . We tried to evaluate the risk of moderate-severe among doctors in Indonesia who survived from COVID-19. In addition, behavioral adaptation in PPE usage among this population before and after being infected with COVID-19 were assessed.

Methods
Participants and procedure. This was an online population-based cross-sectional study. The survey was conducted from October-December 2020 among Indonesian doctors of COVID-19 survivors. The questionnaire was designed for Google survey tool (Google Forms) and the generated link was shared to public on social media and also shared personally to the contact list of investigators. The inclusion criteria of this survey were Indonesian doctors who registered in the Indonesian Medical Association, run a medical practice, and have a history of con rmed COVID-19 infection. The diagnosis of COVID-19 was con rmed by real-time polymerase chain reaction (RT-PCR) for SARS-CoV-2 done on nasopharyngealoropharyngeal swab specimens. Participants who do not ll out incomplete survey data were excluded from the analysis.
Measures. Socio-demographic information collected were age, gender, marital status, profession, working in COVID-19 designated hospital, working hours per week, occupational risk strati cation, body mass index (BMI), smoking status, suspected sources and location of transmission. Occupational risk strati cation based on Indonesian Medical Association was categorized into four groups: low risk (do not have contact with suspect/probable/con rmed COVID-19), moderate risk (have contact with many people who unknown their COVID-19 status), high risk (have close contact with suspect/probable/con rmed , and very high risk (have close contact and doing aerosol medical procedures in COVID-19 patients) 18 . Description of COVID-19 infection history included date of rst PCR-SARS-CoV-2 con rmed, clinical symptoms, comorbidities, disease severity, hospitalization, and long-haulers COVID-19 symptoms.
We evaluated the use of several PPEs before and after COVID-19 infection in several working locations. Special isolation ward with aerosol generating procedure was considered as isolation room, while emergency department, non-isolation ward, critical care unit, operating room, other procedural room, and outpatient unit were considered as non-isolation room. We asked the participants about the use standardized ltering face-piece respirator N95 and other type of mask, including non-standardized KN95, surgical mask, and fabric mask. N95 was considered an adequate standard for the COVID-19 pandemic. For those who used N95, we further evaluated the reusing practice of N95 respirator. Other personal protective equipment interests were surgical cap, eye protection (goggles), face protection, coverall, long sleeve gowns, and gloves. The use of shoe covers (boots) and coverall was asked only for those working in isolation room. Each PPE item use had a possible response of "never", "sometimes", and "always".
These possible responses were also applied in the reusing practice of N95 respirator. Statistical analysis. The data analysis was performed using SPSS version 25.0 (IBM® SPSS® Statistics 25). Bivariate and multivariate analyses were performed using chi-square test and binary logistic regression to determine factors associated with moderate-severe COVID-19 infection.
Ethical considerations. This study was approved by the Health Research Ethics Committee, Faculty of Medicine, Universitas Indonesia-Cipto Mangunkusumo Hospital (KET-1114/UN2.F1/ETIK/PPM.00.02/2020). All participants who agreed to participate in this study provided online informed consent form on the rst page of the survey before continue responding to the online self-report survey. All procedures in this study were conducted in accordance with the Declaration of Helsinki, as revised in 2013.

Results
Socio-demographics characteristic. A total of 389 doctors who survived COVID-19 infection were included in the nal analyses. These participants were spread across 25 of 34 provinces in Indonesia. The highest number of participants were Jakarta, West Java, and Banten (46.1%, 11.0%, and 6.4% of total participants, respectively) as seen in Figure 1.
Most of the participants were young doctors (20-39 years: 69.7%), general practitioners (50.4%), working in COVID-19 designated hospitals (62.5%), and had worked more than 40 hours per week (57.8%). Almost half of suspected transmission source was from the patients (47.8%), followed by their medical colleagues (15.4%), and other health care workers (7.5%). Eighty percents HCWs thought that they got infection in their workplaces (80.5%). Fifty ve percents of participants were not hospitalized during infection as seen in Table 1.   (Figure 2A).
Outside the isolation room, before contracting with COVID-19, N95 respirators were routinely used by rank in operating room (56.0%), critical care unit (46.7%), emergency department (45.1 %), medical procedure room (42.3%), non-isolation ward (38.9%), and outpatient clinic (34.4%). Some doctors were using surgical mask only: in outpatient clinic (always 17.2%, sometimes 9.7%), in medical procedure room (always 14.1%, sometimes 0.7%), in non-isolation ward (always 12.2%, sometimes 6.8%), in critical care unit (always 8.2%, sometimes 4.9%), in emergency department (always 8.0%, sometimes 7.4%), and less in operating room (always 6.0%). Although, we demonstrated better trends in N95 respirator used after these doctors recovered from COVID-19, there were still some doctors used surgical mask only in all nonisolation room as seen in gure 2B-G. We identi ed there were some doctors who even used fabric masks only when working in non-isolation room: 3 doctors in medical procedure room, 2 doctors in emergency department, 2 doctors in outpatient clinic, and 1 doctor in non-isolation ward. After recovering from COVID-19, one doctor still used fabric mask only when working in emergency department.
Overall practices of using other PPE were improved after doctors survived from COVID-19 infection as seen in gure 3. Of the 114 doctors working in isolation rooms, 83.3% doctors routinely used headcap, 83.3% routinely used face protectors, 62.3% routinely used eye protectors, 81.6% routinely used hazmat suit, routinely used gown 53.5%, 71.1% boots, and 86.8% routinely used gloves before getting COVID-19 infection. The trends were increasing in all types of PPEs after they survived from COVID-19 infection as seen in gure 3A.
In non-isolation room, the routinely use of headcap, face protectors, gown, and gloves were also increased once these doctors doing medical practice after recovering from COVID-19 as seen in Figure   3B-G. Better patterns were seen in operating room even before getting COVID-19 infection, except the use of face protector, eye protector, and boots ( Figure 3G).

Discussion
In this nationwide cross-sectional survey, there were more asymptomatic or mild cases than moderate or severe cases. Our participants were doctors, so they tend to have a higher awareness of health problems and initial symptoms of COVID-19. However, we are aware that some doctors who experienced severe or critical COVID-19 infection are not survived and accounted for in this survey. At the end of this survey period, 374 doctors in Indonesia died of COVID-19, as reported by IMA. 11 This study showed that the most suspected transmission source was from the patient (47.9%), and the most suspected transmission location was in health care facilities or their workplaces (80.5%). This result was similar to Zabarsky, et al., who reported that exposure at the workplace to the infected patient was the most transmission source in HCWs with COVID-19 infection in Cleveland, USA, especially in the early pandemic. Exposures to infected patients occurred when recognition of COVID-19 was delayed due to atypical clinical manifestation 19,20 . Di culties in distinguishing patients with COVID-19 and non-COVID-19 due to limited diagnostic tests might impact the prolonged exposure of SARS-CoV-2 in health care workers 8 . Therefore, some of the COVID-19 patients might be seen in non-isolation rooms, especially in emergency department and non-isolation wards. Until late 2020, SARS-CoV-2 antibody was used routinely as screening to differentiate the COVID-19 and non-COVID-19 cases before the availability of PCR SARS-CoV-2. Serological tests might be cheaper and easier to implement at the point of care during pandemic. Because of the higher risk of bias and lower accuracy of serological test, the doctors should be careful when interpreting the result 21 .
Exposure to other doctors was thought to be responsible for 15.4% of the infection, while 7.5% was from other HCWs such as nurses, midwives, and nutritionists. These transmissions probably occurred when these doctors or other HCWs were asymptomatic and still worked despite having symptoms of COVID-19 infection. This condition also associated with occupational risk strati cation and severe COVID-19 19 . Doctors who had close contact with patients who were unknown their COVID-19 status (IMA moderate occupational risk) had a greater risk of severe COVID-19. Early study in Wuhan Hospital in China identi ed 34.9% HCWs who infected COVID-19 were asymptomatic 22 .
Gibson, et al., showed that older age was correlated with severe COVID-19 illness in HCWs who worked directly with the patients. Doctor age ≥65 years old had 9.2 fold higher (95% CI: 6.3-13.3) to get severe COVID-19 infection 23 . Older age has a role in the e ciency of the binding of S Protein SARS-CoV-2 with ACE-2 24,25 . Compared to doctors age 20-39 years, the proportion of moderate-severe COVID-19 cases was higher in doctors age 40-59 years, but not in doctors age above 60 years. In Indonesia, older doctors age ≥60 years were encouraged not to practice or work directly with the patients during the pandemic. We only had 12 doctors age above 60 years old responded to the study invitation. This group was under representative in this survey, probably because the high mortality due to COVID-19 infection. A large epidemiological study in Jakarta showed that the highest mortality rate was among patient above 70 years (34%), followed by 60-69 years (22%) 26 .
In our study, we found that there was not association of comorbidities and moderate-severe COVID-19 infection. Our study is contrast with the latest meta-analysis by Li 31,33 . In those two meta-analyses, the pooled prevalence of fatigue as long COVID-19 symptom were 63.9% (95% CI: 54-73%) and 58% (95% CI: 42-73%), respectively 31,33 . This symptom will extend on several occasions beyond seven months after the onset of illness and causing signi cant disability 34 .
Moreover, meta-regression analysis showed that fatigue was associated with poor quality of life 33 . Though our study showed a lower proportion of long COVID-19 symptoms, because doctors played an essential role in managing during the pandemic, good rehabilitation and clinical management strategies to overcome these long COVID-19 symptoms are warranted.
Although fabric mask should not be worn by HCWs as the level of security is not acceptable in medical facilities, some of doctors used this type of mask before getting COVID-19, even in isolation room. Several studies shown that fabric masks have only marginal protection in preventing droplet infection compared to N95 mask 35,36 . The di culties to access the appropriate mask due to shortage was probably the strongest reason for this unsafe practice. Not all health care facilities provided appropriate type and number of masks for their staffs, especially in the early pandemic 17 . Therefore, most of the doctors need to provide their own masks and other PPEs as needed.

Limitation
This study should be seen in light with some limitations. The study was an online-based survey, thereby underreporting was very likely. The invitation was using social media platforms. It depends on the willingness of participants and might be restricted to only those with good internet access. As a consequent, our study was unlikely to represent of the whole Indonesian doctors survived from COVID-19 infection, especially from east Indonesia. Participants might also give wrong information in this data collection method that can not be con rmed. Recall bias may affect the accuracy of self-reported symptoms and disease classi cation, not to mention the PPE practice before being infected with COVID-19. However, to the best of our knowledge, this is the rst report of COVID-19 infected doctors during the pandemic in Indonesia. Another limitation was occured as a result of a cross-sectional study design.
Therefore, the de nitive cause and effect associations can not be determined.

Conclusion
In this cross-sectional survey, several factors were associated with moderate-severe COVID-19 infection among doctors in Indonesia, including IMA moderate occupational risk, middle-aged doctors, worked in COVID-19 designated hospitals, and higher BMI. Adequately PPEs were a crucial factor that could be preventing doctors from getting COVID-19 infection. Although, getting adequate PPEs was still challenging, these doctors showed improvement in PPEs used after recovered from COVID-19 infection both in isolation and non-isolation rooms.

Declarations
Data availability The data could be obtained by request to author's email (evy.yunihastuti@gmail.com.) Authors contributions  Other PPEs use in isolation (A) and non-isolation rooms (B-G) before and after COVID-19 infection.