Working status of the staff and zoning at the accommodation facility
The lodging facility was a 234-room business hotel. The staff who operate the accommodation facilities are divided into four groups of prefectural staff, nurses, housekeepers, and security guards. Prefectural staff cooperate with public health centers to coordinate the admission of COVID-19 positive individuals and to manage the respective facilities. Nurses are responsible for managing the daily physical conditions of hospitalized patients. Housekeepers are in charge of patient food distribution, garbage pick-up, and cleaning, while guards are in charge of managing the movement of patients within the restricted areas. The average number of staff (range) members for day and night shifts were six (3–7) and one (1) for prefectural staff members, four (4) and two (2) for nurses, 10 (9–14) and two (2–3) for housekeepers, and three (3) and two (2) for security guards, respectively. The total number of staff members working throughout the cluster period was 99.
Each of the four occupations was based in a different area (the security guards were mainly outside), whereas individuals had to occasionally move across areas for the purposes of meetings or administrative tasks. The accommodation facility was zoned into two major areas: an area where patients temporarily stayed (area D) and “no patient access” areas (Figure 1). Furthermore, there was a barrier or wall at the boundary of the two areas, but the upper part of the barrier did not reach the ceiling, thereby making it possible for air to pass through.
Status of mutant virus detection in patients staying at the facility
The number of patients staying during the investigation period is shown in Table 1. The total number of patients was 1,419, and the mean number of patients per day was 51. Although not all patients could be tested for confirmation of delta AY.1, there were no reports regarding of delta AY.1 or delta detection from these patients or from infected individuals by these patients.
[Insert Tale 1 here]
Ventilation status
The layout and ventilation status of the ventilation systems in the area where each staff member was present, including area D, are shown in Figure 2. The areas for prefectural staff (area A), nurses (area B), and housekeepers (area C) were connected through corridors. Each ceiling fan (exhaust port) was designed to have a different air volume displacement; however, one ceiling fan in Area A was out of order. The number of individuals who could stay in each area consisted of 13 in area A, 24 in area B, and 14 in area C based on the required displacement (30 m3/h per person) in accordance with the Japanese law.
CO2 concentrations were measured in each area. The temperature and humidity at the time of measurement were 22.9°C and 61.1%, respectively. The performance of CO2 concentration meters could not be calibrated. The outdoor CO2 concentration was approximately 430 ppm, and the error range of each measuring instrument was 0–50 ppm. At the time of cluster occurrence, air conditioners in area A were not being operated, as opposed to those in areas B, C, and D, which were in operation. Therefore, conditions were similar to the conditions at the time of the occurrence for CO2 concentration measurement.
Changes in CO2 concentration at each measurement point are shown in Figure 3. Unlike in areas A and B, CO2 decreased to the baseline within approximately 30 minutes in areas C and D. Moreover, the decrease in CO2 concentration immediately after the beginning of measurement was rapid in area B, but showed a gradual and decreasing trend (Air convection due to the air conditioner was a highly likely cause, which may have exceeded the suction force at the exhaust port). In addition, differences in CO2 concentrations between the areas were not statistically significant. However, although ventilation range in area A was within legal requirements, ventilation was slightly poor compared to other areas because the exhaust port fan was broken.
Testing and diagnosis
Staff who did not exhibit symptoms suggestive of COVID-19 underwent PCR testing twice during the cluster period, and individuals who tested negative were defined as infection-free. Thirteen staff members were identified with symptoms suggestive of COVID-19. Of the 13 members, nine were required to visit a medical institution and underwent PCR testing, where they tested positive, and they were thus diagnosed with COVID-19 by the lead physician. The remaining four staff members did not visit a medical institution, since onset was on the day of the group testing. Instead, these individuals underwent PCR testing during group testing (Figure 4a). Of these four members, three tested negative and one was found to be SARS-CoV-2 positive (and was later diagnosed with COVID-19 by a physician). Since these three members also tested negative in a separate testing session that was performed two days after the initial tests, infection with SARS-CoV-2 was ruled out.
In the 10 PCR-positive staff members, except for one member with low viral load in the sample (received one vaccination dose), a genomic analysis was performed. Sanger sequencing analysis confirmed that the virus detected in the remaining nine members had K417N, L452R mutations that were characteristic of delta AY.1. In addition, whole-genome sequence analysis revealed that samples from five members with sufficient viral loads included viruses with genomes that were molecularly homologous, did not show single nucleotide polymorphisms, and thus had completely identical molecular structures. The glycosylation mutation information for SARS-CoV-2 spike proteins in these five samples and the accession IDs registered into the Global Initiative on Sharing Avian Influenza Data [GISAID] (https://www.gisaid.org/) are shown in Table 2.
[Insert Tale 2 here]
Staff interview
The health of the accommodation facility staff was routinely observed, and physical deconditioning was not observed in any of the staff members immediately prior to cluster infection. A line listing of facility staff who were SARS-CoV-2 PCR-positive is provided in Table 3 and the related epidemiological curves are provided in Figure 4b. The first episode occurred in a nurse (Case 1). There was no infection registered in nurses working together with Case 1 on night shift. The only contact between Case 1 and prefectural staff (Case 2 or Case 3) was for an approximately 5-min conversation regarding work-related tasks in Area B, during which everybody wore non-woven masks, and for approximately 5 min/meeting during staff meetings (up to 26 individuals) twice daily in Area A (where also all staff wore non-woven masks).
[Insert Tale 3 here]
Contact between Case 1 and housekeepers (Cases 4 and 6) lasted for approximately 10 min while putting on and removing PPE, and for approximately 5 min during the aforementioned general meeting and during daily communications regarding work-related tasks.
Other possible infection risks included telephone calls for approximately 30 min per day by prefectural staff in Area C where the housekeepers stayed (there was a witness stating that non-woven masks were occasionally worn incorrectly), masks being moved out of position, and conversations during light meals with individuals in the same occupation (up to six prefectural staff work in Area A and up to 10 housekeepers work in Area C), conversations during daily work while wearing non-woven masks, and smoking together in groups of 4–5 housekeepers. Hand hygiene was performed before and after putting on and removing PPE, when entering each staff area, and just before eating and drinking.
Vaccination status, symptoms at onset and oxygen saturation on admission, and prognosis of PCR-positive staff are shown in Table 4. Approximately 3 weeks before the onset of Case 4, the individual received one dose of the BNT162b2 (Pfizer-BioNTech) vaccine against COVID-19. The remaining positive individuals were not vaccinated. Of the 10 PCR individuals who tested positive, nine developed the disease (one patient was asymptomatic), and four (two males and two females) were hospitalized, but their conditions improved, and they were thus discharged from the hospital.
[Insert Tale 4 here]
Measures taken after the identification of positive individuals
Measures to prevent the spread of cluster infection included the following mandatory measures: cleaning of exhaust and air vent filters, enhancing ventilation (opening windows) from Day 7, prohibiting of having masks off or out of position during breaks, etc., and thorough hand hygiene. On Day 8, the infection countermeasure team from the Kanagawa Prefecture visited the facility to confirm the situation and to instruct all staff members to wear N95 respirators at all times. N95 respirators are respiratory protective equipment originally used to prevent infection to wearers [15]. However, since N95 respirators have a higher filtration ability and fit compared to non-woven masks and they have been suggested to be effective in preventing infection to others [16], a mandate was given to wear them.