OP19 Perinatal factors do not affect paediatric inflammatory bowel disease risk: A Scottish Nationwide Cohort study using administrative health data 1981–2017

C. Burgess1,2, C. Schnier3, I. Chalmers4, R.K. Russell5, R. Hansen5, P. Henderson1,2, R. Wood6,7, D.C. Wilson1,2

1Department of Child Life and Health, University of Edinburgh, Edinburgh, UK, 2Department of Paediatric Gastroenterology and Nutrition, Royal Hospital for Sick Children, Edinburgh, UK, 3Centre for Medical Informatics, Usher Institue, Edinburgh, UK, 4Department of Paediatric Gastroenterology and Nutrition, Royal Aberdeen Children’s Hospital, Aberdeen, UK, 5Department of Paediatric Gastroenterology and Nutrition, Royal Hospital for Children, Glasgow, UK, 6Information Services Division, NHS National Services Scotland, Edinburgh, UK, 7Salvesen Mindroom Research Centre, University of Edinburgh, Edinburgh, UK

Background

The incidence of paediatric-onset inflammatory bowel disease (PIBD) continues to rise, with genetic-environmental interactions expected to play a vital role. Given gestation and infancy are sensitive periods for environmental change, we aimed to determine whether a mode of delivery, gestational age or type of infant feeding contribute to the development of PIBD in a nationwide cohort of high-risk Scottish children using administrative health data research methodology.

Methods

All children born in Scotland 1981–2017 were identified within the Scottish Morbidity Record maternity inpatient dataset (SMR02). Data linkage was applied to determine mode of delivery, gestational age and type of infant feeding, with PIBD cases defined as any inpatient International Classification of Diseases (ICD) coding for Crohn’s disease (CD; 555/K50), ulcerative colitis (UC; 556/K51) or inflammatory bowel disease unclassified (IBDU; 558.9/K52.3) until age 16. Validation was performed within an entire Scottish health board (16% of the total population) via individual case-note verification. Hazard ratios (HRs) were calculated for each exposure using Cox proportional hazards models with the following potential confounding factors included as covariates: sex, maternal age, year of birth, postcode area and deprivation quintile. Dependent interactions were also explored between the three main exposures.

Results

A study population of 2 013 851 children were identified including 1721 PIBD cases. Individual validation was performed on all 298 (17%) PIBD-coded patients within NHS Lothian; 242/261 CD and UC cases were validated as true positive (positive predictive value 93%); all 37 IBDU cases were false positive with this ICD code excluded from all further analysis. Children delivered vaginally did not have an altered risk of developing PIBD compared with Caesarean section; adjusted HR 0.95 [95% CI 0.84–1.08] (p = 0.424). Children born prematurely did not have an altered risk of developing PIBD; 24–31 weeks gestation HR 0.97 [95% CI 0.56–1.68] (p = 0.922); 32–36 weeks gestation HR 0.96 (95% CI 0.76–1.20] (p = 0.699). Children not exclusively breast feeding at 6 weeks of age did not have an altered risk of developing PIBD; formula feeding HR 0.97 [95% CI 0.81–1.15] (p = 0.705); other/unknown feeding type HR 0.90 [95% CI 0.74–1.08] (p = 0.259). No two-way interactions were significant and none of the three exposures was significantly related to CD or UC in PIBD subtype analysis.

Conclusion

This robust, validated population-based study of over 2 million Scottish births utilising administrative health data confidently demonstrates no association between mode of delivery, gestational age, type of feeding and the development of PIBD.