Factors associated with COVID-19 infection and infection severity in Dutch health care workers during the COVID-19 epidemic in the Netherlands
MetadataShow full item record
Background: The BCG-Corona trial began in March 2020 and followed 1511 healthcare workers in the Netherlands for a year: 753 participants received BCG vaccine and 758 placebo vaccine. The trial found no impact of BCG vaccination on SARS-CoV-2 infection incidence or infection severity. As a secondary analysis, we examined factors associated with cumulative SARS-CoV-2 infection and infection severity. Methods: Participants completed a baseline and endline questionnaire. Additionally, they completed daily/weekly diaries within a mobile phone application. Blood sample collection for SARS-CoV-2 serology testing occurred twice during follow up; serology results could distinguish natural infections from COVID-19 vaccination responses. Infections were classified as proven (by participant-reported positive test with or without seroconversion), possible (seroconverted but no participant-reported positive test), unlikely (due to inconsistent data), or unknown (due to insufficient data). The 55 participants in the latter two categories were excluded from the analyses. Furthermore, only first infection episodes were included. Infection severity was classified as: 1) no infection; 2) asymptomatic infection; 3) mild infection; and 4) moderate/severe infection. We used logistic regression to model cumulative infection and ordinal logistic regression to model infection severity. Each covariate of interest was first modelled in univariate models, followed by stepwise forward multivariable models using a cut-off of p=0.05 for retention in the model. Results: During follow-up, 277 infections occurred in 273 participants of 16.6% were asymptomatic, 61.0% mild, and 22.4% moderate/severe; only four participants acquired a second infection. In univariate models, taking hypertension medication, having a higher number of patient contact hours per week, having had COVID ward duty, working in internal medicine departments (compared to urgent or intensive/medium care or other hospital departments) and being female were associated with higher risks of infection as well as severe infection. In the multivariable models, the increased risk associations with hypertension medication and the work-related variables persisted. Conclusions: During the first year of the COVID-19 epidemic in the Netherlands, healthcare workers were exposed to SARS-CoV-2 at work and their levels of risk for both infection acquisition as well as infection severity were associated with work-related conditions.