With the rapid growth of the elderly population, the task of delivering proper care to all those in need has become a major concern in Republic of Korea. There are various senior- living options: living communities, assisted living, nursing homes (NHs), in-home care, temporary or day care, and palliative care. The Korean Statistical Information Service (KOSIS) reported that NHs are the most common and accessible facilities for older adults, accounting for 65.3% of senior-living options in Republic of Korea in 2019 [1]. A NH is a form of shared accommodation with the assistance of nursing home workers (NHWs). According to the Enforcement Rule of the Welfare of Senior Citizens Act in Korea, the staffing requirements for NHs include the followings: directors, facility managers, doctors, registered nurses or nurse's aides, physical therapists, social workers, nutritionists, cooks, sanitary workers, and custodians, etc. However, there is no guideline for infection control about facilities and personnel management and operation [2].

COVID-19 infection was labeled as a pandemic on the 11th of March 2020 by the World Health Organization (WHO) [3]. Initially starting in Wuhan, China, the first COVID-19 case was reported in South Korea on the 20th of January 2020 [4]. Through a number of epidemiological events, such as religious meetings in Daegu and multilevel marketing business meetings [5], the number of infected individuals increased exponentially, as did the resultant deaths. The overall fatality rate was 1.48%, but the proportion of deaths of those with 70 years and older accounted for 81.8% of all population as of the end of December 2020, according to a daily briefing that the Ministry of Health and Welfare released [6]. The rapid and efficient response to NH outbreaks led to a significant impact in lowering the morbidity and fatality rate of older adults in nursing homes.

NHs are facilities where old and vulnerable adults with geriatric diseases and syndromes such as dementia and cerebrovascular diseases cannot carry out their own daily lives; in addition, this population tends to be susceptible to the most infectious diseases. During COVID-19 pandemic, preventing infection in NHs has been essential for controlling the disease. However, complete prevention has been impossible due to transmission via asymptomatic patients and commuting NHWs [7,8]. Therefore, early detection of infected individuals and taking immediate measures have been most effective to prevent further transmission in NHs.

The government has attempted to prevent the spread of COVID-19 through strict social measures and a vaccination campaign. However, in Republic of Korea, the scale of the infection has been increasing along with several waves of COVID-19 resurgence, and infection control becomes a major challenge facing NHs despite the high vaccination rate [9]. The authors analyzed the epidemiological characteristics of the first and second NH outbreaks in Gyeonggi Province and summarized the issues surrounding infection control in NHs. We determined that more effective responses to the pandemic is possible by auditing and enhancing NHs’ response capacities, in addition to the external efforts to prevent infection.

Korea Disease Control and Prevention Agency (KDCA) prepared guidelines for collecting information and responding to confirmed COVID-19 cases and contacts based on the Middle East Respiratory Syndrome (MERS) outbreak in 2015. Electronically documented guidelines were distributed throughout the local governments, and the information obtained from the activities by the immediate infection control response team was uploaded to the National Notifiable Disease Surveillance System (NNDSS) operated under the KDCA [10]. Interviews, reviewing medical records, phone-based global positioning system (GPS), card transaction records, and closed-circuit television (CCTV) were used as methods of data collection for statistics and analysis of confirmed patients and for prevention of transmission [11]. We approached the NNDSS on July 2020 when the second NH outbreak was completed and collected data about the two NH outbreaks retrospectively.

The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Seoul National University, Bundang Hospital (No. X-2110-713-903). The informed consent was waived because all data was obtained as a result of a public health response.

Case-specific information such as age, sex, underlying disease, date of symptom onset, presenting symptoms, date of specimen collection, date of diagnosis, and activity history during the infectious period and general data about NH facilities and operations including structural arrangement, the number of certified rooms and beds, the number and location of bathrooms, the work schedule and management of NHWs, and records of visitors such as family members, etc., were included in this study. During cohort management, the facility had to fill in the health records or residents, workers, and visitors such as family members in daily report, and this information was also collected [12].

This study will present a summary of the primary and unexpectedly spread cases based on the infection rate and case- fatality rate by group and location of confirmed patients. Using demographic and clinical information of confirmed patients, we analyzed mean age and sex distribution, COVID-19-related death including median time from diagnosis to death, and comorbid conditions. These variables serve as data showing the vulnerability of the elderly residents to infection. Information on structure and operation of NHs may provide insight related to the viral spread.

Considering the special circumstances involved in NHs, a COVID-19 test was carried out every three days for asymptomatic individuals or immediately when COVID-19-related symptoms developed. It was mandatory when an isolation release test performed on the 13th day from the date since contact with infected subjects occurred. Samples taken from subjects were tested at the Gyeonggi Province Institute of Health and Environment or at private laboratories according to the approved protocol by KDCA [13].

The definition of a “confirmed case” was a person who tested positive for SARS-CoV-2 infection with or without symptoms. A “contact” referred to a person who had come into contact with subjects with a confirmed COVID-19 diagnosis, and a “close-contact” was reclassified as subjects requiring self-isolation depending on the time and distance of contact with the patients. Referring to the guidelines issued by the U.S. Centers for Disease Control and Prevention (CDC) and KDCA, if there was face-to-face contact with a confirmed person within 1 m and for more than 15 min, it was classified as close contact [10,14]. The basic management policy for close contacts was self-isolation at home for caregivers and quarantine at the NH for the residents. Statistical analysis was performed using Microsoft Excel (Microsoft Corp, Redmond, WA, USA).

1. Demographic and clinical characteristics of COVID-19 patients in nursing homes A and B

In the outbreak of the two NHs, a total of 24 residents and 11 NHWs were infected with the virus. In the case of NH A, with a total of 25 confirmed patients, 18 were residents with a mean age of 86.7. Fifteen were female, and 3 were asymptomatic at the time of diagnosis. Among the 18 residents, 8 (44.4%) died of the disease with a median time of 12.5 days (range: 1-75) from diagnosis to death. All residents had comorbid conditions, including hypertension (77.8%), dementia (77.8%), diabetes mellitus (27.8%), and Parkinson’s disease (16.7%). Seven (38.9%) residents had three or more comorbid conditions (Table 1). Thirty-one NHWs were tested including eleven new caregivers, and seven NHWs (22.6%) were infected (72%).

Table 1 . Demographic and clinical characteristics of COVID-19 patients in two nursing homes.

	Nursing home A			Nursing home B		Residents (n=24)/% of residents (total n=35)
	Residents (n=18)	Workers (n=7)		Residents (n=6)	Workers (n=4)
Age (y, mean)	86.7	55.1		85.8	56.3
Age distribution
90-99	7	0		2	0	9/37.5
80-89	9	0		3	0	12/50
70-79	2	0		1	0	3/12.5
60-69	0	2		0	2	0/0
50-59	0	2		0	1	0/0
40-49	0	3		0	0	0/0
30-39	0	0		0	1	0/0
Sex
Male	3	0		3	1	6/25
Female	15	7		3	3	18/75
Conditions at the time of confirmation
Symptomatic	13	3		3	3	16/66.7
Asymptomatic	5	4		3	1	8/33.3
Died (median time from diagnosis to death, range)	8 (12.5 days, 1-75)	0		2 (23 days, 11-35)	0	10/41.7
Comorbid conditions
Hypertension	14	2		3	1	17/70.8
Dementia	14	0		4	0	18/75
Diabetes mellitus	5	1		2	0	7/29.2
Parkinson’s disease	3	0		4	0	7/29.2
Asthma	0	0		1	0	1/4.2
≥2 comorbid conditions	15	0		5	0	20/83.3
≥3 comorbid conditions	7	0		2	0	9/37.5

A total of 10 confirmed cases occurred in NH B, of which 6 (60%) were residents. The mean age of confirmed residents was 85.8 years; 3 were female and 3 were asymptomatic at the time of diagnosis. Two (33.3%) out of six died from the disease, and the median time from diagnosis to death was 23 days (range: 11-35). The comorbid conditions of confirmed residents were as follows: dementia (66.7%), Parkinson’s disease (66.7%), hypertension (50%), and diabetes mellitus (33.3%). Two residents (33.3%) had three or more comorbid conditions. Among 62 NHWs, including 6 newly deployed caregivers, four caregivers and one social welfare worker were infected. Two family contacts were confirmed to be infected but were excluded from the analysis. The above results are summarized in Table 1.

2. Outbreak setting, course of outbreak and cohorting

1) NH A

NH A is a facility that uses the fourth and fifth floors, in a five-story building. There are eight rooms on the fourth floor and five rooms on the fifth floor, and each room has two to four beds. All residents and NHWs use one shared bathroom per floor. Twenty-eight and seven residents on the fourth and fifth floor, respectively, reside with the help of 20 NHWs, such as social welfare workers, nurse's aides, caregivers, cooks and a physical therapist.

On 19th March, 2020, NH A reported the first confirmed case of COVID-19, an 85-year-old resident with dementia and hypertension. She went to the hospital as a result of a fall, but she had no issues related to the fall itself. Instead, she complained of cough and sputum, and the laboratory test for SARS-CoV-2 was positive. Tests for all residents and NHWs were recommended promptly due to the possibility of a cluster infection. A person presumed to be a primary case was confirmed the next day, on 20th March, 2020. She was a physical therapist for all residents, and she had a previous medical history of fever, cough, and myalgia. Symptoms started on 10th March, 2020, and the COVID-19 PCR was negative at that time, so she continued to work. All residents and workers were classified as close contacts, and NH A was managed under cohorting. On behalf of the home-isolated workers, new caregivers were deployed wearing personal protective equipment (PPE) to care for the residents. However, NH A cohorting was forcibly terminated on 8th April, 2020 as a result of a newly hired caregiver being confirmed with COVID-19; hence, there were not enough caregivers left to care for the residents. The remaining eight residents were transferred to a national quarantine hospital where they could be cared by professional medical personnel until monitoring ended.

2) NH B

NH B is also a long-term care facility consisting of two separate buildings, and the index case worked as a caregiver at one building. The four-story building houses seventy one residents and fifty six NHWs, and each floor is divided into two units that are connected by a small hallway. In the scope of space and facility, there are 10 rooms on each floor, and each room is equipped with between one and two beds and a private bathroom. However, any bathroom for the NHWs is not present on each floor, so the caregivers have to use either one in a resident's room or the one in the basement.

On 28th May, 2020, a 68-year-old caregiver was confirmed as a primary and index case. She worked every other day and was in charge of 10 residents among 6 rooms in Unit A on the second floor with another caregiver. On 26th May, 2020, she developed chills while working. Her infection was confirmed next day. Following general screening after the index confirmation, a caregiver working in the other unit on the same floor also turned out positive. Therefore, sixteen residents residing on the second floor were classified as close contacts, and cohorting was initiated as a control measure. On 3rd June, 2020, a social welfare worker serving all over the floor was confirmed, so cohort-quarantine was expanded to the entire floor in compliance with contact tracing. Cohorting ended on 20th June, 2020, when 14 days had passed since the last confirmed case.

The epidemiologic spread of COVID-19 in two NHs are illustrated in Figure 1. Table 2 summarizes the infection rate by location and the resulting case-fatality.

Table 2 . Outbreak results of COVID-19 in two nursing homes.

	No. con-firmed (n)/no. exposed (n)	Attack rate (%)	No. died (n)/case-fatality rate (%)
NH* A
Residents
Fourth floor	12/28	42.9	5/41.7
Fifth floor	6/7	85.7	3/50.0
NHWs†	7/31	22.6	-
Total	25/66	37.9	8/32.0
NH B
Residents
1st floor	0/17	-	-
2nd floor	6/20	30.0	2/33.3
3rd floor	0/18	-	-
Fourth floor	0/16	-	-
NHWs	4/62	6.5	-
Total	10/133	7.5	2/20.0

*NH, nursing home. ^†NHWs, nursing home workers..

Figure 1. Confirmed coronavirus disease (COVID-19) cases in two NHs in Gyeonggi Province, by date of symptom onset and relationship. In NH A, the primary case, a physical therapist worked during the symptomatic phase with residents on all floors, confirmed patients occurred in 7 out of 8 rooms on the fourth floor, and in 4 out of 5 rooms on the fifth floor (A). The primary case of NH B, a caregiver who also worked while having symptoms and infected patients, occurred in 4 out of 6 rooms on the second floor which she was in charge of in a four-story building (B). Red bold squares indicate NHWs who cared for close contacts while wearing PPE during cohorting. NH: nursing home. C: caregiver. NC: new caregiver. N’s: nurse’s aid. Ph: physical therapist. Ck: cook. S: social welfare worker. PPE: personal protective equipment.

3. Analysis of unexpected spread to NHWs wearing PPE

The last and unexpected case in NH A was a 59-year-old caregiver who was working on behalf of a home-isolated caregiver. The new caregiver started the job on the 23rd of March and cared for close-contact residents solely on the fourth floor. On 8th April, 16 days following her new employment and the 8th day after the last confirmed case occurred at her workplace, she was confirmed to be an asymptomatic patient. In order to discover the cause of the unexpected spread, her statement, work schedule, and CCTV footage were analyzed. Although there was no evidence of direct contact between her and the confirmed residents, she frequently touched goggles and PPE with her gloved hands and was witnessed drinking water and talking to other caregivers with the PPE half-removed while on duty, according to the CCTV footage.

In the case of NH B, a social welfare worker was the unexpected, confirmed case during the outbreak. When the first confirmed case occurred, he was not a close contact and continued working as an essential staff member. His symptoms of fever and ache occurred on 3rd June, and he was immediately tested. Upon the confirmation of COVID-19 next day, an investigation was conducted to reveal the route of transmission through interviews, analysis of shift schedules and CCTV. The social welfare worker had come into contact with confirmed cases before the onset of symptoms, and he monitored them in an independent space while wearing PPE until they were transferred to COVID-19-designated hospitals on 29th May. As a result of reviewing his activities on the CCTV footage, he was witnessed to have frequently touched goggles with gloved hands that had touched the confirmed residents. After transporting them, he had moved around the undisinfected space after removing the goggles and part of PPE. In addition, according to his statement, his PPE was torn while caring for the confirmed residents, but he considered it insignificant at that time. Except for this event, he had no direct contact with confirmed cases and his symptom developed five days later. Meanwhile, he worked as an essential person wearing PPE to deliver daily necessities to each floor of the building and to help residents bathe and move. We summarized the characteristics of primary case and unexpected spread infection in two nursing homes in Table 3.

Table 3 . The characteristics of primary case and unexpected spread in two nursing homes.

	Nursing home A	Nursing home B
Primary case
Age/sex	62/F	68/F
Role/position	A physical therapist/fourth and fifth floors	A caregiver/2nd floor, unit A
Time interval symptom onset to confirmation (working days)	10 days (3 days)	2 days (1 day)
Case of unexpected spread
Age/sex	59/F	34/M
Role/position	A new caregiver/fourth floor	A social worker/all floors
Symptoms at the confirmation	No	Yes
Level of PPE* on duty	Double gloves, coveralls, face shield or goggles, N95 mask, shoe covers	Double gloves, coveralls, face shield or goggles, N95 mask, shoe covers

*PPE, personal protective equipment..

To prevent the spread of COVID-19, the government issued restrictions such as social distancing while emphasizing personal hygiene such as handwashing and wearing a mask. COVID-19 vaccination distribution began approximately 13 months after the first case was confirmed in Republic of Korea with the aim of achieving herd immunity in a safe way. To reduce mortality and morbidity, elderly residents and workers in long-term care facilities were classified as a high-priority group for vaccination, and the percentage for those who received a first dose, and second dose was 85.9% and 66.6%, respectively, as of 30th June, 2021 [15]. However, the outbreak of COVID-19 infection in NHs has been a constant concern and there have been few epidemiological reports in Republic of Korea with no systematic reviews, especially in Gyeonggi Province, the most populous province in Republic of Korea [16,17]. Social measures over a long period of time affect the occurrence of confirmed cases according to strength [18], and it is reported that the vaccine distribution and immunization effect decreases over time [19]. The authors recognized that external factors such as social measures or vaccination cannot completely prevent cluster infections of NHs. Therefore, we monitored the issues that occurred during the response to cluster infections in NHs and provided solutions.

Although the primary case of NH A was confirmed to be negative at the time of symptom onset by PCR test, the false- negative rate was reported to be 3.2% & 4.3% in some studies [20,21], and in particular, it reached 38% on the day of symptom onset by Kucirka et al. [22]. Therefore, repeated testing would be important to prevent further transmission when infection was suspected. Our study showed that the confirmed residents tended to be older and female and had more comorbidities, similar to other studies, and these facts had a serious subsequent effect, such as a high fatality rate [23-25]. Therefore, NHs face a problem of blocking the additional transmission after the influx of the COVID-19 infection was unveiled.

As COVID-19 is transmitted by droplet and direct contact [10,26], there was a high possibility of transmission from NHWs to residents in particular when a physical therapist was in charge of all residents as in NH A. The transmission between residents was likely as they used a communal restroom and located at the same table to eat. In addition, since physical therapy was provided every day according to the needs of the residents without documentation, there were limitations in identifying the exact contacts by the time and distance. On the other hand, the primary case of NH B cared for a limited number of residents in a designated space, the possibility of additional transmission appeared low. However, as long as there was only one staff lounge and no staff restroom on each floor, allowing the caregivers to share it with residents, the capability of additional spread will remain. These distinctions can be reflected in the difference in infection rate, where 51.43% of the residents and 38.88% of the NHWs at NH A and 8.45% of the residents and 6.45% of the NHWs at NH B were infected (Table 2). According to WHO and KDCA, SARS-CoV-2 was transmitted by aerosol, which has increased risk of spread under special circumstances with the production of respiratory droplets and poorly ventilated spaces [10,27]. Since confirmed cases were recognized, air-borne routes were also considered as one of the methods of virus propagation. Two NHs do not have spaces that generates a lot of respiratory droplets while singing or worshiping. Both NHs are located on the ground level, and there are enough windows in every room. The immediate response team provided guidance on sufficient ventilation following the KDCA guidelines [12,19]; therefore, the possibility of airborne transmission appeared to be low.

The appropriate choice of PPE was based on the subject's nature of interactions; therefore, medical staff who treated COVID-19 patients were recommended to wear an N95 mask or an equivalent respirator, eye protection, and gowns or coveralls with foot covers and gloves. The same level of PPE was also be applied to healthcare workers while caring for close contacts [28]. In our experience, NHWs with excessive fear of the virus were equipped with comprehensive PPE including coverall, and there were no alternatives to the KDCA guidelines [10,27]. However, in the process of on-site management, questions were constantly raised as to whether it had been appropriate for caregivers to wear comprehensive PPE due to poor handling and inconsistent reliance on them. Moreover, we found that PPE could act as another source of infection for less skilled NHWs. Referring to the guidelines of other countries, coveralls were not mandatory and easy-to-maintain protective clothing such as gowns or aprons were also allowed [29,30]. Therefore, it is warrantly required that guidelines for these various alternatives are reinforced. Cohorting by badly trained NHWs and in inefficient facilities for infection-control methods also resulted in unexpected outcomes such as enforced termination and expansion of cohorting, so cohorting by itself has limited effectiveness in suppressing the spread of infection.

We have described experiences of two NHs in response to an emerging infectious disease. There were a small number of cases, and we have not been able to measure the objective level of infection control knowledge or the relevant capabilities of the NHWs. However, these two NHs were operated with facilities and personnel that met the legal standards, and they followed standardized guidelines during cohorting in the process and response to COVID-19 outbreaks. We carefully considered that the issues raised by these examples could be representative.

We observed two COVID-19 outbreaks in NHs in which infection initiated by NHWs transmitted elderly residents and affected cohorting measures. Transmission in facilities could not be suppressed by cohorting implement alone due to dependence on NHWs. The purpose of this study was to understand the characteristics of group infections in NHs, analyze the transmission, and inform that it is important to improve the infection control ability of NH and NHWs.

This research received no external funding.

The authors declare no conflict of interest.