Fig. 5
From: From first to second wave: follow-up of the prospective COVID-19 cohort (KoCo19) in Munich (Germany)
Association between potential risk factors and SARS-CoV-2 sero-positivity taking into account time between 1st and 2nd sampling, baseline result, age and sex. Age and sex was also adjusted for time between 1st and 2nd sampling, baseline result and each other (sex for age and age for sex). Unimputed (blue) and imputed (orange) GLM Models (Bayesian analysis). Main individual level risk factors were time and sex. Odds decreased by age and was lower for participants living in buildings with 3–4 apartments. Changes in the estimates by imputation were small