High compliance to infection control measures prevented guest-to-staff transmission in COVID-19 quarantine hotels.

Dear Editor,

We read with interest the article by Bou-Karroum et al. on the public health effects of travel-related policies on the COVID-19 pandemic, in which the authors demonstrated that early border closure and quarantine of travellers contributed positively to the control of the pandemic. In particular, the authors identified 4 observational and 2 modelling studies on quarantine of travellers, which showed that the effectiveness of quarantine increased with increasing rates of compliance with quarantine.

Hotel quarantine for incoming travellers have been implemented in many places, such as the United Kingdom, Australia, New Zealand, Canada, and mainland China. In response to the emergence of SARS-CoV-2 variants of concern (VOCs), the Hong Kong SAR government imposed mandatory quarantine at designated hotels for all persons returning from places outside mainland China since December 2020. However, quarantine hotels may serve as a hotspot for viral spread if there is lapse in infection control. There have been reports of COVID-19 transmission in quarantine hotels which involved transmission from returned travellers to staff. , A previous study has shown that SARS-CoV-2 RNA can be detected in surface swabs, pillow cover, sheet and duvet cover in guest rooms of a quarantine hotel where presymptomatic COVID-19 patients stayed, although the infectivity may be low for indirect fomite transmission.

In April 2021, two returning travellers infected with SARS-CoV-2 were diagnosed after checking out from quarantine hotels in Hong Kong, which triggered extensive contact tracing and mass testing. There were no epidemiological links to other COVID-19 cases except that they stayed on the same floors with confirmed COVID-19 cases in the two quarantine hotels. Intra-hotel transmission was suspected. Our previous investigation suggested possible airborne transmission through ingression of air from the doorway when the doors of the guest rooms were opened, and that there was a lack of fresh air supply and absence of exhaust fan in the corridors of the implicated hotels. As part of the investigation into this incident, we conducted a seroepidemiological survey of hotel staff members to assess whether silent transmission has occurred. Institutional review board approval was exempted since this is an emergency public health response.

A total of 136 individuals participated in the survey between 15th and 20th May 2021, including 90 staff members from the 2 implicated hotels (Hotels A and B), and 46 from hotel C, a third hotel under the same hotel chain which also served as a quarantine hotel but did not have any known intra-hotel transmission (Table 1 ). The questionnaire included basic demographics, COVID-19 vaccination status, work nature, exposure to quarantined guests or their belongings, personal protective equipment (PPE) usage and training on infection control.

Table 1

Demographics of hotel staff members in this study.

Characteristics	No. (%) unless otherwise specified
	Total	Hotels A + B	Hotel C	p value
	(n = 136)	(n = 90)	(n = 46)	(Hotels A + B vs. C)
Age – median years (range)	49.5 (24–70)	50 (24–66)	48 (27–70)	0.12
Sex – male (%)	76 (55.9)	52 (57.8)	24 (52.2)	0.59
Smoker (%)	45 (33.1)	28 (31.1)	17 (37.0)	0.56
Comorbidity:
No comorbidites	86 (63.2)	54 (60.0)	32 (69.6)	0.35
Hypertension (%)	32 (23.5)	24 (26.7)	8 (17.4)
Diabetes mellitus (%)	10 (7.4)	6 (6.7)	4 (8.7)
Liver disease (%)	6 (4.4)	5 (5.6)	1 (2.2)
Heart disease (%)	5 (3.7)	4 (4.4)	1 (2.2)
Lung disease (%)	4 (2.9)	4 (4.4)	0 (0)
Renal disease (%)	1 (0.7)	1 (1.1)	0 (0)
Others (%)	11 (8.1)	10 (11.1)	1 (2.2)
Work nature:
Housekeeping	42	30	12
Clerical work	30	17	13
Engineering	20	11	9
Concierge	13	8	5
Security	12	11	1
Kitchen	6	5	1
Cleaning	5	3	2
Linen room	3	2	1
Management	2	1	1
Meal delivery	1	1	0
Restaurant	1	1	0
Accounting	1	0	1
Exposure to guests under quarantine:
No exposure	63 (46.3)	36 (40.0)	27 (58.7)	0.04*
Face-to-face exposure (within 2 m)	32 (23.5)	24 (26.7)	8 (17.4)
Stayed in the same room	8 (5.9)	7 (7.8)	1 (2.2)
Contact with items which have been touched/used by quarantined guests	58 (42.6)	42 (46.7)	16 (34.8)
Vaccination status
Any COVID-19 vaccine (≥one dose)	57 (41.9)	31 (34.4)	26 (56.5)	0.02*
BNT162b2 (≥one dose)	25 (18.4)	11 (12.2)	14 (30.4)
CoronaVac (≥one dose)	32 (23.5)	20 (22.2)	12 (26.1)
Any COVID-19 vaccine (2 doses with last dose at least 14 days before joining study)	40 (29.4)	23 (25.6)	17 (37.0)
BNT162b2 (2 doses with last dose at least 14 days before joining study)	13 (9.6)	6 (6.7)	7 (15.2)
CoronaVac (2 doses with last dose at least 14 days before joining study)	27 (19.9)	17 (18.9)	10 (21.7)

Seventy three (53.7%) individuals had exposure to the guests or their belongings. Amongst them, all wore face masks during work, 93.2% (68/73) wore protective gowns, 78.1% (57/73) wore face shields, 94.5% (69/73) wore gloves, and 90.4% (66/73) wore goggles. One hundred and four (76.5%) individuals, including 90.4% (66/73) of the exposed, reported having received training on infection control and prevention, most commonly in the form of face-to-face teaching sessions (74/104, 71.2%) and self-reading materials (73/104, 70.2%), while 11.5% (12/104) individuals also attended online training class.

Fifty seven (41.9%) of the staff members had received at least one dose of COVID-19 vaccine before participating in the study. Amongst them, 43.9% (25/57) received the BNT162b2 mRNA vaccine (Pfizer-BioNTech), and 56.1% (32/57) received the CoronaVac inactivated virus vaccine (Sinovac Life Sciences). There was no statistically significant difference in the demographics and presence of underlying diseases between the vaccinated and non-vaccinated groups (Table 2 ). The vaccination rates were not significantly different between those with and without exposure to quarantined guests. However, individuals who received the BNT162b2 vaccine were significantly younger (median age 44.5 vs 52 years; p = 0.01) and were less likely to have underlying diseases than those who received the CoronaVac vaccine (proportion with underlying disease 16.0% vs 46.9%; p = 0.02).

Table 2

Characteristics of the participating staff from 3 quarantine hotels based on vaccination status.

Group	Unvaccinated (n = 79)	Vaccinated (n = 57)	BNT162b2 (n = 25)	CoronaVac (n = 32)	Unvaccinated vs. Vaccinated	BNT162b2 vs. CoronaVac
Median age (years)	49.5	49	44.5	52	p = 0.66	p = 0.01*
Sex (male%)	58.2%	52.6%	64.0%	43.8%	p = 0.60	p = 0.18
Smoking	36.7%	28.1%	40.0%	18.8%	p = 0.36	p = 0.14
Underlying disease	39.2%	33.3%	16.0%	46.9%	p = 0.59	p = 0.02*
Exposure to guests or their belongings	54.4%	52.6%	40.0%	62.5%	p = 0.86	p = 0.11
Perceived knowledge (mean, standard error)	7.56, 0.22	7.33, 0.35	7.56, 0.43	7.16, 0.53	p = 0.86	p = 0.96

We performed both anti-nucleocapsid (N) IgG test and surrogate virus neutralisation antibody test (sVNT) for all participants (See Supplementary Methods). Since our ongoing COVID-19 serosurveillance in Hong Kong showed a very low seropositive rate (Supplementary Table), a positive test in the anti-N IgG assay or sVNT would be compatible with natural infection for non-vaccinated individuals. For BNT162b2 mRNA vaccine recipients, only a positive anti-N IgG test would signify natural infection. Since CoronaVac is an inactivated whole virus vaccine, antibody test is not useful in differentiating natural infection from vaccine-induced immunity. Amongst the 104 non-vaccinated or BNT162b2 mRNA vaccine recipients, all tested negative for anti-N IgG. For the 79 non-vaccinated individuals, sVNT was positive for 1 staff and indeterminate for another. However, the sera from these two individuals tested negative by both anti-S1 IgG assay (Euroimmun) and conventional live virus microneutralisation assay. Hence, there was no serological evidence of COVID-19 infection amongst hotel staff members.

The absence of transmission from hotel guests to staff members is likely related to the adequate training and compliance of staff members to different preventive measures. Furthermore, the hotel staff in this study had higher vaccination rate than the general population in Hong Kong (41.9% amongst hotel staff in this study vs 28.2% of the Hong Kong population as of 19 June 2021), which likely contributed to the absence of transmission to the hotel staff despite possible airborne transmission inside the hotels.

There are some limitations in this study. First, some hotel staff members did not reply to the questionnaire or join the serosurveillance. Second, some vaccinated individuals have not completed the course of COVID-19 vaccination for 14 days before blood taking, which may have affected the interpretation of serology results.

In summary, we demonstrated that infection control training and strict compliance amongst hotel staff members, especially those with direct contact with quarantined persons, may have prevented guest-to-staff transmission of SARS-CoV-2, thus preventing secondary spread to other guests and in the community.

Funding

This work was supported by the Consultancy Service for Enhancing Laboratory Surveillance of Emerging Infectious Diseases and Research Capability on Antimicrobial Resistance for Department of Health of the HKSAR, and the Health and Medical Research Fund, the Food and Health Bureau, The Government of the Hong Kong Special Administrative Region (Ref. No. COVID190124).