Early experiences with antibody testing in a Flemish nursing home during an acute COVID-19 outbreak: a retrospective cohort study

Abstract objectivesto assess the prevalence of COVID-19 (PCR-test) in residents and staff of a nursing home. To examine the presence of IgM and IgG antibodies in the sample and the relation between PCR and antibody test results. designcross-sectional and (retrospective) cohort study settinga nursing home for the elderly Bessemerberg in Lanaken (Belgium) with up to 130 beds. Lanaken is situated in the Belgian province with the highest COVID-19 prevalence. participantsresidents (N=108) and staff members (N=93) of the nursing home outcomesPCR, IgM and IgG resultsthe prevalence of COVID-19, based on PCR test was 34% (N=40) for residents and 13% (N=11) for staff members, respectively. Of the residents, 13% showed positive IgM results and 15% positive IgG results. In 17% of the residents, at least one of the antibodies was positive. In total 13% of the staff members had positive IgM and 16% had a positive IgG. In 20% of the staff members at least one of these antibody tests was positive. In PCR positive residents, the percentage of IgM positive, IgG positive, and at least one of both was 28%, 34%, and 41%. In PCR positive staff, we found 30%, 60%, and 60%. Additional antibody tests were performed in nine residents between day 11 and 14 after the positive PCR test. Of those, 7 (78%) tested positive on at least one antibody. When retesting three weeks later, all remaining residents also tested positive. conclusionsRecently it was reported that in Belgium antibodies are present in 3-4% of the general population. Although, the prevalence in our residents is higher, the number is largely insufficient for herd immunity. In staff members of the regional hospital the prevalence of antibodies was 6%. The higher prevalence in nursing home staff (21%) may be related to the complete absence of good quality protection in the first weeks of the outbreak. Article summaryStrengths and limitations of this study- This is the first study in Belgium examining the prevalence of COVID-19 and the presence of antibodies in residents and staff members of a nursing home- The internal procedural control was positive -with one exception- in all tests, which suggests good quality sampling and testing.- Some degree of selection bias should be assumed in residents, since some residents were absent; mostly from hospitalisation or death which can be related to the presence of COVID-related disease.- The study was set up in one nursing home and is consequently not representative for the whole of the Flemish community