Strategies for Prevention of COVID-19 Transmission in Hospitals: a Nationwide Survey of Controversial Issues Related to Infection Control and Prevention in Korea

Background: Infection-control measures against the coronavirus disease 2019 (COVID-19) within a hospital are often based on expert experience and intuition due to the lack of clear guidelines. This study aimed to survey the current strategies for the prevention of the spread of COVID-19 in medical institutions. Methods: In a systematic review of national-level guidelines, 13 key topics were selected. Six hospitals were provided an open survey between August 11 and 25, 2020, to assess their responses to these topics. Using these data, an online questionnaire was developed and sent to the infection-control teams of 46 hospitals in South Korea. The survey was conducted between January 31, 2021, and February 20, 2021. Results: All 46 hospitals responded to the survey. All hospitals operated screening clinics, while 89.1% (41/46) allowed symptomatic patients without COVID-19-associated symptoms to visit the general outpatient clinics. Most hospitals (87.2%; 34/39) conducted polymerase chain reaction (PCR) tests for all hospitalised patients. Moreover, 35 (76.1%) hospitals had preemptive isolation policies for hospitalised patients, of which 97.1% (34/35) released patients from isolation after a single negative PCR test. Most hospitals (76.9%; 20/26) allowed shared-room accommodation for patients who met the national criteria for release from isolation but showed positive PCR results with cycle threshold values above a certain threshold (34.6%; 9/26) or after a certain period that satised the national criteria (26.9%; 7/26). Conclusions: Various guidelines were being applied by each medical institution, but an explicit set of national guidelines to support these guidelines was unavailable.

implemented by hospitals often rely on the experience of in-house experts and are often benchmarked against other hospitals' strategies [6].
In the light of these issues, the present study aimed to evaluate the strategies for preventing the in ow of undetected COVID-19 patients into hospitals, preventing infections among healthcare workers when treating suspected or con rmed COVID-19 patients, and preventing the spread of COVID-19 infection in medical institutions by medical personnel. Through this study, we aimed to derive the optimal strategies to prevent the spread of COVID-19 within medical institutions.

Questionnaire design
In the initial phase of questionnaire development, controversial issues related to the prevention of COVID-19 transmission within medical institutions were selected via discussion among four infectious diseases

Conducting the survey
The survey was conducted over a period of 21 days (from January 31, 2021, to February 20, 2021), targeting the six hospitals where information about real-world practices for controversial topics was gathered and 40 sample hospitals in Korea that were managing COVID-19 patients at the time of the survey. The 40 sample hospitals were selected to account for four categories of hospitals: hospitals operating state-designated isolation beds, hospitals with 500 beds or more that did not operate statedesignated isolation beds, hospitals with less than 500 beds, and hospitals without ID specialists. Ten hospitals of each category were selected while maintaining uniform distributions across regions.
A link to the online-based survey was forwarded via e-mail to the physicians or nurses who belonged to the infection-control teams of each hospital. To encourage participation, reminders were sent on the 5th, 10th, and 15th days. Only one questionnaire was administered per hospital.

Statistical analysis
To evaluate the differences in strategy according to the experience of managing COVID-19 patients, we compared hospitals that managed 100 or more cases of COVID-19 (COVID-19 ≥ 100 group) with hospitals that had treated fewer than 100 cases of COVID-19 (COVID-19 < 100 group). All statistical analyses were performed using SPSS version 24.0 for Windows (IBM Corporation, Armonk, NY, USA). Categorical variables were analysed by the Chi-squared test or Fisher's exact test. Continuous variables were analysed using the Mann-Whitney U-test or independent t-test. Variables with P values < 0.05 were considered statistically signi cant. Table 1 provides the basic information of the participating medical institutions. All 46 hospitals that were sent the questionnaire responded to the survey. Among them, 24 hospitals had treated 100 or more cases of COVID-19. The majority of hospitals in the COVID-19 < 100 group were private university-a liated hospitals, while the majority of those in the COVID-19 ≥ 100 group were public non-university-a liated hospitals. The proportion of hospitals without an ID specialist was signi cantly higher in the COVID-19 ≥ 100 group than in the COVID-19 < 100 group (P = 0.011). Screening and selective treatment policies to prevent COVID-19 patients from entering the hospital Table 2 lists the screening and selective treatment policies implemented to prevent COVID-19 patients from entering the hospital. All 46 hospitals operated screening clinics, which treated patients with respiratory symptoms, fever of unknown cause, and epidemiological association with COVID-19 patients, and accepted persons who wanted to undergo polymerase chain reaction (PCR) tests for COVID-19. Most hospitals implemented measures to prevent the in ux of COVID-19 into the hospital through caregivers and family/acquaintances. Among these hospitals, 90.5% allowed general ward access only to the patients' essential caregivers, 57.1% regularly monitored fever and respiratory symptoms of caregivers, and 64.3% performed mandatory PCR tests for caregivers.

Basic information of medical institutions
All medical institutions provided education concerning the use of masks for patients and caregivers, while 35 hospitals made regular public address announcements guiding patients and caregivers to wear masks; these strategies were used more frequently in the COVID-19 < 100 group (P = 0.024 Preemptive isolation policies for patients with suspected COVID-19 Table 3 presents the data for preemptive isolation policies for patients with suspected COVID-19. Thirtyve hospitals implemented preemptive isolation for inpatients with suspected COVID-19, of which 34 released patients from preemptive isolation only after the rst negative PCR test after hospitalisation, and two released patients from isolation after 10-14 days of symptom onset or exposure to COVID-19, regardless of the PCR test results. Regarding the criteria for preemptive isolation before con rmation with PCR results, 23/35 hospitals applied preemptive isolation to all patients admitted into the hospital, while 12 hospitals applied preemptive isolation for patients with suspected symptoms of COVID-19, those with imaging results suspected to be pneumonia, and those with an epidemiological association with a COVID-19 patient. Additional isolation policies for patients with suspected or con rmed COVID-19 are provided in the Supplementary Table 1. Policy for COVID-19 patients who showed consistently positive PCR results but whose symptoms had improved and were released from isolation Table 4 summarises the policies for COVID-19 patients who showed consistently positive PCR results but also showed improvement in symptoms and were released from isolation. Of the hospitals that responded to the survey, 27 treated COVID-19 patients whose PCR results were consistently positive but were released from isolation.  hospitals allowed shared-room accommodation when the cycle threshold (Ct) value in the PCR test was above a certain threshold, and 7/26 hospitals allowed shared-room accommodation after a certain interval from the time of meeting the national isolation release criteria.
The strategy of procedures or operations for patients with suspected or con rmed COVID-19 Table 5 shows the strategy of procedures or operations for patients with suspected or con rmed COVID-19. Thirty-ve responding institutions performed emergency procedures or operations for suspected COVID-19 patients; emergency procedures or operations were performed in a higher proportion of hospitals in the COVID-19 < 100 group (P = 0.024).  3 11 hospitals that did not have a preemptive isolation policy for patients suspected of COVID-19 and 1 hospital that did not perform elective procedures or operations were excluded 4 7 hospitals that did not have an isolation policy for patients con rmed with COVID-19 and 6 hospitals that did not perform elective procedures or operations were excluded 5 19 hospitals that did not have a policy for patients with COVID-19 whose PCR results are consistently positive but whose symptoms have improved and are released from isolation and 2 hospitals that did not perform elective procedures or operations were excluded As for elective procedures or operations, 29/34 of the hospitals postponed all procedures or operations for suspected COVID-19 patients until the patient was released from preemptive isolation, and 26/33 postponed all procedures or operations for con rmed patients until the isolation was removed. Of the hospitals that treated COVID-19 patients who showed consistently positive PCR results with improvement in symptoms and were released from isolation, 10/25 performed procedures or operations when the Ct value of the PCR test met certain criteria, and 6/25 performed the procedures after a certain interval from the time of release from isolation regardless of the PCR test results.
Hospital work-restriction policy for healthcare workers Table 6 presents the data for hospital work-restriction policies for healthcare workers. For healthcare workers who had visited high-risk areas of COVID-19 without fever or respiratory symptoms, 19/44 medical institutions restricted the workers from working and performed PCR tests if they became symptomatic. However, all hospitals restricted healthcare workers with a fever or respiratory symptoms from work and conducted PCR tests. Over 60% of the participating institutions restricted healthcare workers from performing certain activities outside the hospital. As for the conditions for returning to work among COVID-19 infected employees who met the national isolation release criteria, 14/46 hospitals required COVID-19-infected employees to show negative PCR results before returning to work, and the proportion of such hospitals was higher in the COVID-19 < 100 group. Only 12/46 medical institutions allowed healthcare workers to return to work immediately after meeting the national isolation release criteria. Additional data for decision-making system for COVID-19 related issues is provided in the Supplementary Table 3. All hospitals operated screening clinics, and nearly 90% allowed patients with symptoms not considered to be associated with COVID-19 to enter their general outpatient clinics. Because screening clinics are equipped with minimal facilities and a workforce that can only provide a minimal examination [7], most hospitals manage patients with fever and respiratory symptoms who are unlikely to have COVID-19 at the general outpatient clinic, where careful evaluation and management can be provided. Unfortunately, there are no clear criteria for the entry of symptomatic patients in general outpatient clinics, which can lead to confusion among frontline medical professionals [5]. As observed in this study, many hospitals implemented e cient strategies for patient selection that can be benchmarked by other hospitals, including (i) negative COVID-19 test results within two to three days, (ii) previously treated in general outpatient clinics for a disease related to the current symptoms, and (iii) minimal probability of COVID-19 as determined by experts.  [11]. A study conducted in long-term care facilities in the United States showed that the prevalence of COVID-19 in facilities that performed broad preemptive PCR tests on inpatients was 0.5%, while it was 28.0% in facilities that did not perform them [12]. Nevertheless, approximately 15% of the hospitals in this study did not perform PCR tests on patients without fever or respiratory symptoms. Although this strategy seems to be effective for detecting asymptomatic COVID-19 patients, its cost-effectiveness remains unproven. Thus, considering the local COVID-19 prevalence, it is necessary to determine whether to proceed with PCR tests when asymptomatic patients are hospitalised [13]. According to the Infectious Diseases Society of America (IDSA), screening asymptomatic patients is expected to be effective in regions with more than 2% prevalence, considering the results of missing a COVID-19 diagnosis and the sensitivity of PCR tests [14]. However, based on the strategy used by many of the hospitals in this study, screening PCR tests for selected patients such as those requiring general anaesthesia, those admitted to the intensive care unit, and those transferred from other medical institutions and nursing homes can be considered in communities with a low prevalence of COVID-19.
Although the probability of infectious SARS-CoV-2 is very low after 10 days from the onset of COVID-19 symptoms in most cases, severely ill or immunocompromised patients can transmit infectious virus particles even after 10 days [18, [20][21][22][23], and the continuing possibility of COVID-19 transmission from these hospitalised patients, especially those who need aerosol-generating procedures, has been a topic of concern [22,[24][25][26]. For these reasons, only 19% of the hospitals treated these patients in a shared room in a general ward regardless of the PCR test results or the time of release from isolation in this survey.
Since 50% of patients show positive results in real-time polymerase chain reaction (RT-PCR) tests of nasopharyngeal swabs at around 18 days after the onset of symptoms, and some patients show a positive result even eight weeks later, the inclusion of negative PCR test results in the criteria for release from isolation is likely to lead to unnecessary isolation and excessive use of PPE [26,27]. Since the cutoff values for quantitative RT-PCR and quantitative immunoassays tend to be correlated with the infectivity of COVID-19, some researchers have suggested that certain cut-off values could be used as surrogate markers for the decision to release hospitalised patients from isolation [22,28,29].
Although the current guidelines include many recommendations on PPE for healthcare workers, clear guidelines for the management of healthcare workers were not available. As a result, work-restriction policies for healthcare workers differed across hospitals. Since medical personnel can also spread COVID-19 to other people in medical institutions, many hospitals applied stricter return-to-work criteria than the national guidelines for healthcare workers infected with or exposed to COVID-19. However, strict criteria for returning to work could result in a lack of a su cient workforce [30]. Thus, the implementation of suitable criteria considering the speci city of the hospitals' situations is essential.
This study had some potential limitations. First, the survey was conducted in February 2021, and the situation may have changed by the time the results are presented. Second, this survey was performed only in South Korea, and since the prevalence of and social response to COVID-19 differs among countries, additional investigations might be required in other regions or countries. Nevertheless, the ndings remain signi cant since they highlight the importance of appropriate guidelines and indicate key topics relevant to real hospital settings for further research based on the results of this study.

Conclusions
In conclusion, because of the lack of clear guidance on infection-control strategies for preventing COVID-19 transmission within hospitals, individual hospitals are currently relying on experience to frame relevant guidelines. Thus, systematic research on these areas and guidance at the national level is needed for greater consistency and standardisation in hospitals' measures to prevent COVID-19 outbreaks.