Seropositivity and risk factors for SARS-CoV-2 infection in staff working in care homes during the COVID-19 pandemic.

To the Editor,

We recently reported in the Journal of Infection very high rates of SARS-CoV-2 infection in four London care homes experiencing prolonged outbreaks of COVID-19 during the peak of the pandemic in England. These outbreaks were associated with two deaths and two additional hospitalisations among staff members before the onset of this investigation.

In order to control the outbreak in the four care homes, weekly nasal swabbing was performed for all residents and staff for four weeks starting 3 May 2020 to rapidly identify and isolate infected individuals. The results of weekly swabbing and antibody testing in residents have already been reported. The staff took their own nasal swab and completed a detailed questionnaire on employment, work patterns, household composition and travel pattern (Supplement Figure S1). In total, 443/596 staff were on shift when swabs were taken, and provided at least one swab and completed a questionnaire. None of the 1305 nasal swabs from 443 staff tested positive for SARS-CoV-2 RNA by RT-PCR. After the four-week period, a blood sample was taken from consenting staff (N = 327) for SARS-CoV-2 anti-N IgG antibody testing, performed according to manufacturer instructions, and 154 (47.1%) were seropositive (Supplement Figure S2). We explored risk factors for antibody positivity in staff including age, gender, co-morbidities, ethnicity, employment, household composition and travel to work using Pearson's Chi-squared test and included factors with p<0.25 in a multivariate logistic regression model (Table 1 ). All analyses were performed using the tidyverse (version 1.3.0) and arsenal (version 3.5.0) packages in R (version 4.0.2).

Table 1

Summary statistics and Chi-squared test p-value for potential risk factors for SARS-CoV-2 infection.

Risk factor	N	N Seropositive	Chi-squared test p-value
DEMOGRAPHIC FACTORS
Sex			0.357
Female	254/325 (78.2%)	123 (48.4%)
Male	71/325 (21.8%)	30 (42.3%)
Age			0.014
0–39	142/327 (43.4%)	55 (38.7%)
40–59	149/327 (45.6%)	83 (55.7%)
60+	36/327 (11.0%)	16 (44.4%)
Co-morbidity			0.742
No comorbidity	231/308 (75.0%)	110 (47.6%)
At least 1 comorbidity	77/308 (25.0%)	35 (45.5%)
Ethnic group			0.185
Asian	70/282 (24.8%)	38 (54.3%)
Black	87/282 (30.9%)	49 (56.3%)
White	111/282 (39.4%)	46 (41.4%)
Mixed	6/282 (2.1%)	2 (33.3%)
Other	8/282 (2.8%)	3 (37.5%)
EMPLOYMENT FACTORS
Care Home			0.100
Home A	135/327 (41.3%)	56 (41.5%)
Home B	63/327 (19.3%)	31 (49.2%)
Home C	77/327 (23.5%)	35 (45.5%)
Home D	52/327 (15.9%)	32 (61.5%)
Contract type			0.637
Permanent	261/311 (83.9%)	121 (46.4%)
Temporary	50/311 (16.1%)	25 (50.0%)
Work in another care home			0.390
FALSE	219/229 (95.6%)	101 (46.1%)
TRUE	10/229 (4.4%)	6 (60.0%)
TRAVEL FACTORS
Car	80/327 (24.5%)	32 (40.0%)	0.144
Bus	145/327 (44.3%)	78 (53.8%)	0.030
Train	64/327 (19.6%)	33 (51.6%)	0.425
Underground	40/327 (12.2%)	17 (42.5%)	0.534
Bicycle	12/327 (3.7%)	5 (41.7%)	0.701
Walk	34/327 (10.4%)	16 (47.1%)	0.996
Other	44/327 (13.5%)	17 (38.6%)	0.227
HOUSEHOLD FACTORS
Resides at place of work			0.075
FALSE	311/327 (95.1%)	143 (46.0%)
TRUE	16/327 (4.9%)	11 (68.8%)
Household member works in a care home			0.044
FALSE	265/310 (85.5%)	116 (43.8%)
TRUE	45/310 (14.5%)	27 (60.0%)
Household member works in hospital/GP surgery			0.144
FALSE	273/305 (89.5%)	131 (48.0%)
TRUE	32/305 (10.5%)	11 (34.4%)

Chi-squared tests showed that age group (p = 0.014), ethnic group (p = 0.185), care home of employment (p = 0.100), travelling to work by car (p = 0.144), bus (p = 0.030) or other (0.227) method, residing at their place of work (p = 0.075), having a household member who works at a care home (p = 0.044) and having a household member who works at a hospital/GP surgery (p = 0.144) were predictors of positive antibody status at p<0.25 (Table 1). When these were reanalysed in a multivariate analysis, we found statistically significant associations with having a household member who works in a care home (OR 2.97, 95%CI 1.33–7.04), Asian ethnicity (OR 2.03, 95%CI 1.04–4.05) or having a household member who works in a hospital/GP surgery (OR 0.36, 95%CI 0.13–0.92), and weak evidence for 40–59 age group (OR 1.78, 95%CI 1.00 – 3.22), Black ethnicity (OR 1.79, 95%CI 0.94–3.43) or working at care home D (OR 2.17, 95%CI 0.98–4.88) (Table 2 ). The characteristics of staff members that had an antibody test result (N = 327) did not differ from staff who completed the risk factor questionnaire (N = 443) but did not provide a blood sample.

Table 2

Multivariate analysis of significant risk factors for SARS-CoV-2 in univariate analysis (Table 1). Controlling for ethnicity and having a household member who works in a care home, travel to work by bus is not a significant predictor of seropositivity.

Risk factor	OR	95% CI – lower OR	95% CI – higher OR	p-value	Concordance
(Intercept)	0.364	0.163	0.792	0.012	0.694
Age category
40–59	1.784	0.998	3.215	0.052
60+	1.105	0.435	2.741	0.831
Ethnic group
Asian	2.034	1.038	4.049	0.040
Black	1.788	0.941	3.425	0.077
Mixed	0.807	0.101	4.781	0.820
Other	0.917	0.150	5.211	0.921
Care home of employment
Home B	1.164	0.513	2.640	0.716
Home C	1.186	0.581	2.428	0.639
Home C	2.166	0.984	4.875	0.057
Travel by car	0.797	0.375	1.679	0.552
Travel by bus	1.154	0.615	2.163	0.654
Travel by ‘other’	0.890	0.376	2.089	0.790
Resides at place of work	2.802	0.848	10.336	0.100
Household member works in care home	2.973	1.326	7.041	0.010
Household member works in hospital/GP surgery	0.356	0.125	0.915	0.039

In the four care homes under investigation, nearly half the staff members had antibodies against SARS-CoV-2. Since no staff had a positive RT-PCR test result for SARS-CoV-2 during the 4 weeks prior to the antibody test, it is likely that the vast majority of infections occurred early in the course of the pandemic in London (March-April), at a time when there was widespread community infection and limited personal protective equipment (PPE) and SARS-CoV-2 testing available in care homes. Both PPE and testing for SARS-CoV-2 are now available for care homes across the UK.

Our findings are consistent with other care home investigations reporting very high seropositivity rates among staff irrespective of their PCR-positivity or symptom development. Seropositivity rates among care home staff are several fold higher than reported in any other occupational setting, including frontline hospital healthcare staff, highlighting the degree of virus exposure experienced by care home staff and residents.

Antibody testing after four weeks of negative nasal swabs demonstrates its usefulness is assessing past exposure to the virus. It is increasingly clear that all nearly all adults who are exposed to SARS-CoV-2 have an detectable antibody response within 28 days of infection, and the majority of people who have detectable antibodies also have neutralizing antibodies, , as reported in staff and residents in other London care homes.. High rates of seropositivity may provide a degree of protective immunity in these settings, although the duration of such protection is at present unclear. Close monitoring of further cases and outbreaks will be important to help contribute to our understanding of protective immunity.

The negative weekly nasal swabs following the high infection rates prior to testing (as evidenced by the staff seropositivity rate) and small sample sizes meant that we were not able to identify additional risk factors for SARS-CoV-2 infection per se, such as working across different care homes as reported in other investigations. The independent association of SARS-CoV-2 antibody positivity with Asian ethnicity, however, is consistent with other healthcare settings and the community. , This is compounded by having other members of the household – usually a spouse – also working in a care home, for which there was strong evidence of an increased risk in our cohort. The reduced risk associated with having a household member who works in a hospital/GP surgery likely reflects less exposure to potentially infected individuals (other staff or patients) when compared to the close contact with residents in care homes. Non-occupational risk factors such as travel to place of work, which have been identified as important risk factors for SARS-CoV-2 infection in other cohorts , , were not significantly associated with seropositivity in this cohort.

In summary, we found SARS-CoV-2 seropositivity rates among staff working in care homes affected by COVID-19 outbreaks to be several times higher than community seroprevalence in London. The high seropositivity rates precluded more detailed assessment of risk factors for SARS-CoV-2 infection, especially in relation to non-occupational risk factors. SARS-CoV-2 antibody positivity was, however, significantly, independently and positively associated with Asian ethnicity and having a household member who also worked in care homes. Further studies are needed to assess the level and duration of protection offered by the antibodies against SARS-CoV-2 re-infection and onward transmission in institutional settings.

Ethics: PHE has legal permission, provided by Regulation 3 of The Health Service (Control of Patient Information) Regulations 2002, to process patient confidential information for national surveillance of communicable diseases and as such, individual patient consent is not required.

Declaration of Competing Interest

None.