P1199 Early life exposures predict childhood-onset IBD in a population-based cohort
M Agrawal, A V Hansen, J F Colombel, K H Allin, T JessAbstract
Background
While many early life exposures have been associated with IBD, prediction of risk based on these variables is not yet available. Therefore, we developed and validated a prediction model for IBD based on early life exposures.
Methods
We conducted a nationwide population-based cohort study leveraging cross-linked Danish registers consisting of all liveborn singleton offspring in the Danish National Birth Cohort with available data on early life exposures based on the first four DNBC interviews (weeks 12 and 30 of pregnancy, 6 months, and 18 months of offspring age), who were alive and residents of Denmark until at least 18 months of age. Follow up was until IBD diagnoses (based on primary and secondary diagnoses codes in the National Patient Register), emigration, death or end of the study period, September 1, 2022. We used elastic net with underlying Cox regression model to predict IBD diagnosis at ≤14 and >14 years of age.
Results
A total of 50,592 offspring were included in the study for a median (IQR) of 21.8 (20.4, 22.9) years. Of these, 310 (0.6%) persons were diagnosed with IBD during follow-up. Early life risk factors were predictive of IBD risk at ≤14 years of age but not later in life [Harrell’s C statistic (SE) 0.72 (0.018) and 0.52 (0.004), respectively]. All of the 74 variable terms considered for inclusion were included in the final model, with the five most important being parental IBD, number of infection episodes age 6-18 months, household mammal pets (protective), number of infection episodes age 0-6 months, and antibiotics exposure age 6-18 months. The cumulative incidence of IBD at age 14 years of age was 0.66% (95% CI 0.49-0.84) among those in the highest 5% for predicted risk, and 0.09 % (95% CI 0.075-0.01) for those in the lowest 95% (Figure).
Conclusion
In a nationwide cohort with long-term follow up data, early life exposures were predictive of childhood-onset IBD, but not later in life. This model warrants validation and could potentially be used to determine IBD risk in children towards prevention strategies.