Prescriptions for insulin and insulin analogues in children with and without major congenital anomalies: a data linkage cohort study across six European regions

Are children with major congenital anomalies more likely to develop diabetes requiring insulin therapy, as indicated by prescriptions for insulin, than children without congenital anomalies? The aim of this study is to evaluate prescription rates of insulin/insulin analogues in children aged 0–9 years with and without major congenital anomalies. A EUROlinkCAT data linkage cohort study, involving six population-based congenital anomaly registries in five countries. Data on children with major congenital anomalies (60,662) and children without congenital anomalies (1,722,912), the reference group, were linked to prescription records. Birth cohort and gestational age were examined. The mean follow-up for all children was 6.2 years. In children with congenital anomalies aged 0–3 years, 0.04 per 100 child-years (95% CIs 0.01–0.07) had > 1 prescription for insulin/insulin analogues compared with 0.03 (95% CIs 0.01–0.06) in reference children, increasing ten-fold by age 8–9 years. The risk of > 1 prescription for insulin/insulin analogues aged 0–9 years in children with non-chromosomal anomalies (RR 0.92, 95% CI 0.84–1.00) was similar to that of reference children. However, children with chromosomal anomalies (RR 2.37, 95% CI 1.91–2.96), and specifically children with Down syndrome (RR 3.44, 95% CIs 2.70–4.37), Down syndrome with congenital heart defects (RR 3.86, 95% CIs 2.88–5.16) and Down syndrome without congenital heart defects (RR 2.78, 95% CIs 1.82–4.27), had a significantly increased risk of > 1 prescription for insulin/insulin analogues aged 0–9 years compared to reference children. Female children had a reduced risk of > 1 prescription aged 0–9 years compared with male children (RR 0.76, 95% CI 0.64–0.90 for children with congenital anomalies and RR 0.90, 95% CI 0.87–0.93 for reference children). Children without congenital anomalies born preterm (< 37 weeks) were more likely to have > 1 insulin/insulin analogue prescription compared to term births (RR 1.28, 95% CIs 1.20–1.36). Conclusion: This is the first population-based study using a standardised methodology across multiple countries. Males, children without congenital anomalies born preterm and those with chromosomal anomalies had an increased risk of being prescribed insulin/insulin analogues. These results will help clinicians to identify which congenital anomalies are associated with an increased risk of developing diabetes requiring insulin therapy and allow them to reassure families of children who have non-chromosomal anomalies that their risk is similar to that of the general population.