key: cord-0299636-r0ev96os
authors: Brinkmann, F.; Diebner, H. H.; Matenar, C.; Schlegtendal, A.; Eitner, L.; Timmesfeld, N.; Maier, C.; Luecke, T.
title: Seroconversion rate and socioeconomic and ethnic risk factors for SARS-CoV-2 infection in children in a population-based cohort
date: 2021-10-25
journal: nan
DOI: 10.1101/2021.10.21.21265322
sha: a9b459ba8fe3f50b62525b4bc6cbf56374428e75
doc_id: 299636
cord_uid: r0ev96os

Introduction: Socioeconomic and ethnic background have been discussed as possible risk factors for SARS-CoV-2 infections in children. Improved knowledge could lead way to tailored prevention strategies and help to improve infection control. Methods: Observational population-based cohort study in children (6mo. - 18 ys.) scheduled for legally required preventive examination and their parents in a metropolitan region in Germany. Primary endpoint was the SARS-CoV-2 seroconversion rate during study period. Risk factors assessed included age, pre-existing medical conditions, socioeconomic factors, and ethnicity. Results: 2124 children and their parents were included. Seroconversion rates among children in all age groups increased by 3-4-fold from 06/2020 to 02/2021. Only 41% of seropositive children were symptomatic. In 51% of infected children at least one parent was also SARS-CoV-2 positive. Low level of parental education (OR 3.13 (0.72-13.69)) significantly increased the risk of infection. Of the total cohort, 38.5% had a migration background. Specifically, 9% were of Turkish and 5% of Middle Eastern origin. These children had the highest risk for SARS-CoV-2 infections (OR 6.24 (1.38-28.12) and 6.44 (1.14-36.45) after adjustment for other risk factors. Discussion: Seroprevalence of SARS-CoV-2 infections in children increased by 3-4-fold within the study period. Frequently, more than one family member was infected. Children from families with lower socioeconomic status were at higher risk. The highest risk for SARS- CoV-2 infection was identified in families with Turkish or Middle Eastern background. Culture sensitive approaches are essential to improve infection control and serve as a blueprint for vaccination strategies in this population.

infection. Of the total cohort, 38.5% had a migration background. Specifically, 9% were of 23

Turkish and 5% of Middle Eastern origin. These children had the highest risk for SARS-CoV-24 2 infections (OR 6.24 (1.38-28.12 ) and 6.44 (1.14-36.45 ) after adjustment for other risk 25

Discussion: Seroprevalence of SARS-CoV-2 infections in children increased by 3-4-fold 27 within the study period. Frequently, more than one family member was infected. Children 28 from families with lower socioeconomic status were at higher risk. The highest risk for SARS-29

CoV-2 infection was identified in families with Turkish or Middle Eastern background. Culture 30 sensitive approaches are essential to improve infection control and serve as a blueprint for 31 vaccination strategies in this population. 32

Trial Registration: BMBF funding registration 01KI20173 (Corkid) 33 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in In a standardized approach from 06/2020 to 02/2021 all asymptomatic children and 53 adolescents who attended outpatient paediatric practices in three regions of Western 54

Germany for scheduled mandatory routine examinations (U-Untersuchungen) from 6 months 55 to 18 years of age and their parents were invited to participate in the study. Participants and 56 their parents were asked to fill in a tablet-based questionnaire available in 5 different 57 languages. Questions included former SARS-CoV-2 infections, medical history as well as 58 information on socioeconomic and migration background (Germany/Western Europe, Turkey, 59

Middle East, America, Eastern and Southern Europe, and Asia). Education level was defined 60 as the highest level of education of one of the parents/guardians. High level of education 61 included high school/grammar school (Fachhochschule/Abitur), medium level education 62 general secondary school (Realschule) and low-level education included education up to 63 primary school/basic secondary school (Grundschule/Hauptschule). No educational degree 64 formed a fourth group. Information on chronic diseases and medication was obtained from 65 parents and verified by matching with medical charts. 66 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted October 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 The target variable was SARS-CoV-2 infection given by seropositivity or positive PCR test, 68 respectively. We aimed at comparing demographics (age, sex), underlying medical 69 conditions, socioeconomic and migration background between children and adolescents with 70 and without evidence of SARS-CoV-2 infection. 71

Statistical Analysis: 72

Descriptive statistics are depicted in tables 1 and 2. Demographic and clinical parameters 73 (table 1) are presented along with univariate odds ratios reported with 95% confidence 74

intervals. In addition, the impact of migration background has been analysed with adjusted 75 random effect logistic regression (cf . table 3) . Thereby, we adjusted for the ratio of number of 76 available rooms to the number of persons living in the same household as well as for the 77 highest educational level within the family. The ratio has been introduced to account for the 78 correlation between number of rooms and number of persons following the rational that the 79 available space per person is the essential factor. Within the Corkid study, some families 80 participated with more than one child. Therefore, we included a random effect in our model 81

given by a family ID. Insignificance of other factors has been shown by a likelihood ratio-82 based model reduction process. For the random effect analysis, the lme4 software package 83 within statistical programming language R has been used [8]. P-values less than 0.05 were 84 considered statistically significant. 85

In this cohort, initially 2848 children and adolescents were asked to participate, 25% refused, 88 mainly because of fear of drawing blood. 

All rights reserved. No reuse allowed without permission.

perpetuity.

preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in This large population-based study identifies seroconversion rates and risk factors for SARS-123

CoV-2 infections in asymptomatic children and adolescents, attended legally required 124 preventive examinations. Our key findings are: 125

• less than half of seropositive participants of all age groups did recall symptoms of 126 infection suggesting that more than half of SARS-CoV-2 infections go unnoticed [9] . 127

• the incidence of SARS-CoV-2 infections is significantly higher in families of Turkish or 128

Middle Eastern decent independent of other risk factors i.e. after adjustment for other 129 socioeconomic confounders. 130

A public health antibody screening indicates a marked increase of SARS-CoV-2 exposure 184 rate in children during the second wave

SARS-CoV-2 Infections Among Children and Adolescents With Acute Infections in the Ruhr 188

Racial and/or Ethnic and Socioeconomic Disparities of SARS-CoV-2 Infection Among 192

Emergency Department Visits for COVID-19 by Race and Ethnicity -13 States

MMWR Morb Mortal Wkly Rep

Demographic 199 predictors of hospitalization and mortality in US children with COVID-19

Assessment of 135 794 Pediatric Patients Tested for Severe Acute Respiratory Syndrome 204

Coronavirus 2 Across the United States

Ethnic differences in SARS-CoV-2 infection and COVID-19-208 related hospitalisation, intensive care unit admission, and death in 17 million

England: an observational cohort study using the OpenSAFELY platform

Fitting Linear Mixed-Effects Models Using 213

SARS-CoV-2 Infection in Children and Their Parents in Southwest Germany

SARS-CoV-2 218 transmissions in students and teachers: seroprevalence follow-up study in a German 219 secondary school in

Clustering and longitudinal 223 change in SARS-CoV-2 seroprevalence in school children in the canton of Zurich

Switzerland: prospective cohort study of 55 schools

Household Transmission Evaluation Dataset: preliminary findings from a novel passive 231 14

Lower household transmission rates of SARS-CoV-2 from children compared to adults

Race/Ethnicity Among Children With 237 COVID-19-Associated Multisystem Inflammatory Syndrome

Between Race and COVID-19 Outcomes Among 2.6 Million Children in England

Epub ahead of print

Lancet Reg Health Eur

Epidemiological and clinical characteristics of the COVID-19 epidemic in Brazil

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted October 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 Our cohort had increasing seroprevalence rates from 0.5% in Mid-2020 to almost 6% in the 131 beginning of 2021 like other seroprevalence data from German or Swiss school cohorts 132 [10, 11] . The dynamics in increase of seroprevalence of SARS-CoV-2 infections matches well 133 with national data of acute SARS-CoV-2 infection in the same geographic region [2, 12] . As 134shown before [2] infection rates have been comparable across all age groups in keeping with 135 the findings of other authors [1, 6] . 136It is not surprising, that the known exposure to SARS-CoV-2 positive individuals doubled the 137 risk for seroconversion. In most families, at least one parent also has evidence of SARS-138CoV-2 infection, which is slightly more than described in other cohorts [13, 14] . However, 139 50% of these SARS-CoV-2 exposed children did not develop evidence of infection. 140

Recent study from the US state that socioeconomically disadvantaged children, especially 142 those from ethnic minorities, are at higher risk of infection [3, 5, 6] . Adult data from the UK and 143 the US also show increased morbidity and mortality from SARS-CoV-2 in adults from ethnic 144 minorities and with poor socioeconomic status. [4, 7] . Limited access to healthcare systems 145 and migration background might also play a role in spreading the disease [15, 16] . 146The most relevant risk factor for SARS-CoV-2 infection in our study population is a Turkish 147 or Middle Eastern migration background. Although socioeconomic factors confound the 148 factor of migration history to some extent, our findings result from an adjusted regression 149 model and are therefore independent of educational standard and housing conditions. 150Reasons for this observation could include different family and social structures favouring 151 closer contacts and therefore increasing the risk of transmission as described in a tight knit 152Jewish Orthodox community in the US by Gaskell et al. [17] . Apart from that, language as 153 well as cultural barriers might play a role [7] . Visiting friends and relatives in countries with 154 higher incidence of infection might also further increase the risk. Scepticism regarding politics 155 and health authorities are an additional issue, especially in prevention and vaccination 156 programs [1] . To approach these families, tailored, culture sensitive strategies need to be 157 developed and can serve as a blueprint for vaccination programs. 158Limitations: Paediatric patients were recruited only from three regions (counties) in Western 159Germany and not all paediatric practices in the area participated. In addition, initially low 160 seroprevalence rates could have introduced a bias. Genetic risk factors could predispose 161 certain populations to infections with SARS-CoV-2, which was not considered in this study. 162The proportion of children from Turkish families, in contrast to those of other origins, was 163 lower than expected at 9% (see table 1), whereas ca. 20% of the families in the regions 164 studied are of Turkish origin. However, this rather indicates an underestimation of the true 165 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted October 25, 2021. ; https://doi.org/10.1101/2021.10.21.21265322 doi: medRxiv preprint All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted October 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021