1.During the years 1996-2002, the Danish National Birth Cohort (DNBC) cohort recruited about 100 000 pregnancies of 92 000 women all over Denmark. Due to the large size, general data collections, both at recruitment and during follow up, have mainly been carried out as telephone interviews and web-base questionnaires, with varying participation rates. The Danish health registers provide information about diseases in mothers and children, and here follow up i expected to be close to 100%. Please examine the data in the following table. Would you be concern with selection bias with regard to the data presented in the data? If so, why? Table 1. Selection by sociodemographic and medical factors in the Danish National Birth Cohort: impact on prevalence estimates. Initial recruitment All women (16) Prevalence source population Prevalence Participants Participation Prevalence rate ratio 95% CI 31% Only compulsory school (27) Academic level (27) 19.1% 10.8% 18% 0.57 0.55-0.59 38.0% 48.5% 40% 1.28 1.26-1.29 Health professionals (27) 10.1% 13.4% 42% 1.32 1.28-1.36 Income in lowest quartile (27) 30.1% 23.4% 24% 0.78 0.76-0.80 Living mainly on social benefits (27) 15.9% 9.9% 20% 0.62 0.60-0.65 Single mothers (27) 10.5% 7.6% 23% 0.72 0.69-0.76 Mothers 10 cig/day in pregnancy (16) 5.6% 4.0% 22% 0.71 0.66-0.76 Infertility treatment (16) 1.9% 2.3% 37% 1.20 1.11-1.30 Small-for-gestational-age infant (16) 3.2% 2.7% 26% 0.83 0.77-0.90