From the original SARS-CoV-2 strain to the Omicron variant: Predictors of COVID-19 in ambulatory symptomatic children

International audienceObjectives: To determine the predictors of a positive SARS-CoV-2 test in a pediatric ambulatory setting.Patients and methods: We performed a cross-sectional prospective study (November 2020-February 2022) of 93 ambulatory settings in France. We included symptomatic children < 15 years old tested for SARS-CoV-2. For each period corresponding to the spread of the original strain and its variants (period 1: original strain; period 2: Alpha, period 3: Delta; period 4: Omicron), we used a multivariate analysis to estimate adjusted odds ratios (aORs) associated with COVID-19 among age, signs, symptoms or contact, and 95 % confidence intervals (95CIs).Results: Of 5,336 children, 13.9 % (95CI 13.0-14.8) had a positive test. During the first three periods, the positivity rate ranged from 5.6 % (95CI 4.6-6.7) to 12.6 % (95CI 10.8-14.6). The main factors associated with a positive test were contact with an infected adult at home or outside the home (aOR 11.5 [95CI 4.9-26.9] to 38.9 [95CI 19.3-78.7]) or an infected household child (aOR 15.0 [95CI 4.8-47.1] to 28.4 [95CI 8.7-92.6]). By contrast, during period 4, aORs for these predictors were substantially lower (2.3 [95CI 1.1-4.5] to 5.5 [95CI 3.2-7.7]), but the positivity rate was 45.7 % (95CI 42.3-49.2).Conclusions: In pediatric ambulatory settings, before the Omicron period, the main predictor of a positive test was contact with an infected person. During the Omicron period, the odds of these predictors were substantially lower while the positivity rate was higher. An accurate diagnostic strategy should only rely on testing and not on age, signs, symptoms or contact