Testing strategies for the control of COVID-19 in nursing homes: Universal or targeted screening?

Dear Editor,

We read with interest the outbreak investigation carried out in late March/early April 2020 in four nursing homes affected by COVID-19 outbreaks in central London. During this study, 27 days after the first death and 21 days after the first resident tested positive, 126 (40%) of nursing home residents were found SARS-CoV-2 positive. A striking finding of this investigation was that 60% of SARS-CoV-2 positive residents were either asymptomatic or only had atypical symptoms for COVID-19 during the two weeks prior to testing. Asymptomatic nursing home staff (4%) were identified as potential source of viral transmission to the residents. Additionally, genomic analysis identified one cluster involving one staff member and two residents in the same home. This work highlights the need for evaluations of the accuracy of testing strategies to identify the reservoir of asymptomatic COVID-19 cases in nursing home (NH) and adjust the control measures. In light of our local experience, and in the context of a post-lockdown period with a low incidence of COVID-19 in the general population, we analysed the results of strategies adopted by 50 NH to assess the accuracy of testing professionals, residents or both for investigating the spread of COVID-19 around a positive case.

In France, a general lockdown was implemented from 03/17/2020 to 05/11/2020. Since 07/04/2020, the French national screening recommendations in NH rely on testing all professionals when a COVID-19 positive case is identified (resident or professional). The screening of residents is recommended up to three suspected cases. Since 27/05/2020, the health authorities in the Vendée department, located in western France (Fig. 1 ), recommended to systematically screen all residents and professionals after the identification of a COVID-19 positive case in NH. Moreover, a universal RT-PCR screening was organized in voluntary NH to estimate the invisible reservoir of COVID-19 at the end of the lockdown.

Fig. 1

Geographical distribution of the 50 participating nursing home in the Vendée department, France.

In the Vendée department, a total of 71 COVID-19 positive residents and 49 professionals were identified from the 04/01/2020 to the 06/30/2020 in 23 (17%) of the 136 NH, with 8 clusters of three cases or more. From the 17/04/2020 to 26/06/2020, 50 NH situated in the Vendée department tested a median number of 42 (IQR: 15–76) residents and 54 (35–73) professionals by nasopharyngeal sample and RT-PCR, totalizing 2003 residents and 2822 professionals. (Figure) Overall, 25 (1.25%) residents and 25 (0.88%) professionals were positive for COVID-19, among whom 19 (76%), and 22 (88%) asymptomatic, respectively. Among the 14 NH which tested both residents and professionals to investigate around a COVID-19 positive resident, 6 (42%) NH did not found any positive case, 4 (29%) found one to two cases, and 4 (29%) identified ≥3 cases. (Table 1 ) Among the 4 (8%) NH which tested residents and professionals to investigate the spread of COVID-19 around a positive professional, one (25%) NH did not found any positive case, two (50%) found one to two cases, and one (25%) NH identified ≥3 cases. All positive cases were asymptomatic. Among the 32 NH which performed a universal screening in absence of known COVID-19 positive case on the day of testing, two (6%) identified one positive resident.

Table 1

Description of testing results performed to investigate the spread around a COVID-19 positive resident, professional, and in nursing homes free of case.

	Total	Residents	Professionals
Number of NH with screening results	50	35	48
Screening performed to investigate the spread around a COVID-19 positive resident
Number of NH with screening results	14 (28%)	14 (40%)	14 (29%)
Number of NH without positive case identified	6 (42%)	7 (50%)	7 (50%)
Number of NH with 1 to 2 positive cases identified	4 (29%)	4 (29%)	4 (29%)
Number of NH with >3 positive cases identified	4 (29%)	3 (21%)	3 (21%)
Total number of participants tested	1538	754	784
Total number of positive cases identified	42 (2.7%)	22 (2.9%)	20 (2.5%)
Screening performed to investigate the spread around a COVID-19 positive professional
Number of NH with screening results	4 (8%)	4 (11%)	4 (8.5%)
Number of NH without positive case identified	1 (25%)	3 (75%)	1 (25%)
Number of NH with 1–2 positive cases identified	2 (50%)	1 (25%)	2 (50%)
Number of NH with >3 positive cases identified	1 (25%)	0	1 (25%)
Total number of participants tested	410	238	172
Total number of positive cases identified	6 (1.4%)	5 (2.1%)	1 (0.6%)
Universal screening in NH free of COVID-19 confirmed case on the day of testing
Number of NH with screening results	32 (62%)	17 (49%)	30 (62.5%)
Number of NH without positive case identified	30 (94%)	17 (88%)	30 (100%)
Number of NH with one positive case identified	2 (6%)	2 (12%)	0
Total number of participants tested	2897	1011	1886
Total number of positive cases identified	2 (0.07%)	2 (0.2%)	0

These results suggest that 7/14 (50%) NH would have missed asymptomatic residents by testing professionals only to investigate around a positive resident. Half of NH would also miss cases by testing residents only. A quarter of NH would have missed one asymptomatic resident by testing all professionals only to investigate the spread around a positive professional. Three facilities would have missed from one to three asymptomatic professionals by testing residents only. Finally, 2/32 (6%) NH which performed a blinded universal testing without any known case would missed one case. In the literature, 40.7 to 57% of residents/staff tested positive for SARS-CoV-2 in high prevalence NH were asymptomatic on the day of testing.4, 5, 6 Our results suggest a higher rate of asymptomatic persons (41/50, 82%) in a low incidence context. Symptom-based screening of NH residents might fail to identify all SARS-CoV-2 infections. Asymptomatic NH residents and professionals might contribute to SARS-CoV-2 transmission. In a post-lockdown context with an estimated mean daily incidence rate of 0.59 COVID-19 cases per 100.000 population, the universal screening in NH free of known positive case seems inefficient. In such context, supplies should be saved, and adverse effects (pain, complication of nasopharyngeal sampling, logistics, and costs) can be avoided. However, once a facility has confirmed a COVID-19 case, extensive testing should be performed for all professionals and residents, and all residents should be cared for using personal protective equipment.

Declarations

Funding: The research was funded by the National Institute for Health Research Health Protection Research Unit (NIHR HPRU) in Healthcare Associated Infection and Antimicrobial Resistance at Imperial College London in partnership with Public Health England (PHE). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health or Public Health England. GB has received an Early Career Research Fellowship from the Antimicrobial Research Collaborative at Imperial College London, and acknowledges the support of the Welcome trust. RA is supported by a NIHR Fellowship in knowledge mobilisation. The support of ESRC as part of the Antimicrobial Cross Council initiative supported by the seven UK research councils, and also the support of the Global Challenges Research Fund, is gratefully acknowledged.

Declaration of Competing Interest

The authors declare that they have no competing interests.