Acceptance of Different Self-sampling Methods for Semiweekly SARS-CoV-2 Testing in Asymptomatic Children and Childcare Workers at German Day Care Centers A Nonrandomized Controlled Trial

IMPORTANCE Closure of day care centers (DCCs) to contain the COVID-19 pandemic has been associated with negative effects on children's health and well-being. OBJECTIVE To investigate the acceptance of self-sampling methods for continuous SARS-CoV-2 surveillance among asymptomatic children and childcare workers (CCWs) in DCCs. DESIGN, SETTING, AND PARTICIPANTS This nonrandomized pilot study included children and CCWs at 9 DCCs in Wuerzburg, Germany, from May to July 2021. INTERVENTIONS Twice weekly testing for SARS-CoV-2 was conducted by self-sampled mouthrinsing fluid (saliva sampling [SAL], with subsequent pooled polymerase chain reaction test) plus nasal rapid antigen self-test (RAgT) (group 1), SAL only (group 2), or RAgT only (group 3) in children and CCWs. MAIN OUTCOMES AND MEASURES Main outcomes were rates for initial acceptance and successful (>= 60% of scheduled samples) long-term participation. The probability of SARS-CoV-2 introduction into DCCs was modeled as a function of age-adjusted background incidence and DCC size. RESULTS Of 836 eligible children, 452 (54.1%; 95% CI, 50.7%-57.4%) participated (median [IQR] age: 4 [3-5] years; 213 [47.1%] girls), including 215 (47.6%) in group 1, 172 (38.1%) in group 2. and 65 (14.4%) in group 3. Of 190 CCWs. 139 (73.2%; 95% Cl. 66.4%-79.0%) participated (median [IQR] age: 30 [25-46] years; 128 [92.1%] women), including 96 (69.1%) in group 1, 29 (20.9%) in group 2. and 14 (10.1%) in group 3. Overall, SARS-CoV-2 PCR tests on 5306 SAL samples and 2896 RAgTs were performed in children, with 1 asymptomatic child detected by PCR from SAL. Successful long-term participation was highest in group 2 (SAL only; children: 111 of 172 [64.5%]; CCWs: 18 of 29 [62.1%]). Weekly participation rates in children ranged from 54.0% to 83.8% for SAL and from 44.6% to 61.4% for RAgT. Participation rates decreased during the study course (P < .001). The probability of SARSCoV-2 introduction into a DCC with 50 children was estimated to reach at most 5% for an age-adjusted SARS-CoV-2 incidence below 143. CONCLUSIONS AND RELEVANCE Self-sampling for continuous SARS-CoV-2 testing was well accepted, with SAL being the preferred method. Given the high number of negative tests, thresholds for initiating continuous testing should be established based on age-adjusted SARS-CoV-2 incidence rates.