[Seroprevalence, risk factors and clinical presentation after the first wave of COVID-19 in nursing homes of the UNIVI group: The SEROCOVID study].

INTRODUCTION

The pandemic caused by SARS-COV-2 infection spread rapidly during the "first wave" through France between March and May 2020. It was responsible for high mortality in subjects with comorbidities and the elderly who lived in nursing homes. In May 2020, 75% of the deaths occurred in people over 75 years old in nursing homes. It is difficult to estimate accurately the prevalence of COVID-19 infection during this period because only 50% of the diagnoses in nursing homes were made by RT-PCR. During this period, the diagnosis was mainly based on the clinical symptoms.

POPULATION AND METHODS

We carried out a prospective study among residents of the 27 EHPADs in the UNIVI group (SEROCOVID study) between August 31 and October 16, 2020 using rapid ELISA serotests carried out by pricking the fingertip. We looked for the seroprevalence by the use of rapid serotests as well as the overall prevalence by cumulating the positive results of the RT-PCR when done and of the rapid serotest. The secondary objectives were the study of risk factors for infection by multivariate analysis as well as the description of the symptoms that led to the diagnosis.

RESULTS

In total, 1145 residents were included aged on average 89±7.5 years old (female 78.7%). The time between the COVID-19 disease and the rapid inclusion serotest was on average 5±1.7 months. The prevalence estimated by the three diagnostic evaluation methods (medical diagnosis, RT-PCR or by rapid serotest ELISA) is about 14%, underestimated compared to the overall prevalence at 22.7%. The study of risk factors in multivariate analysis shows that the most dependent residents, living in a protected unit due to behavioral disorders or whose close contact with a person with COVID-19 had significantly higher rates of infection. Finally, the symptoms most frequently observed in residents differed from those in younger subjects with geriatric characteristics, such as the higher frequency of digestive symptoms and geriatric syndromes. Fever has only been observed in one third of cases in the elderly. Smell and taste disorders were seldom described.

CONCLUSION

Our study provides an estimate of the overall prevalence as well as the mean seroprevalence of COVID-19 in EHPAD residents five months after the diagnosis of COVID-19 disease. The difference between the two estimates is probably explained by the frailty and decreased immunity of the nursing home residents. Therefore, it would need to be reactivated by vaccination of all residents, even those already infected with SARS-COV-2. These elements corroborate the governmental strategy of vaccination deployed in all residents of EHPAD regardless of their previous contact with the virus.