Epidemiology, validation, and clinical characteristics of inflammatory bowel disease: the ABIS birth cohort study

Abstract

Abstract Background Birth cohort studies with linked register-based data on inflammatory bowel disease (IBD) provide opportunities to prospectively study early-life determinants of the disease. However, register-based data often lack information on clinical characteristics and rely on diagnostic algorithms. Within the All Babies in Southeast Sweden (ABIS) cohort, we examined the validity of a register-based definition of IBD, its incidence, and clinical and therapeutic characteristics at diagnosis. Methods We followed 16,223 children from birth (1997–1999) until the end of 2020 for the diagnosis of IBD as defined by a minimum of two diagnostic codes for IBD in the Swedish National Patient Register (NPR). We described the incidence and cumulative incidence of IBD. Through a medical record review of cases diagnosed by the end of 2017, we examined the positive predictive value (PPV) for IBD and described its clinical characteristics and treatment. Results By 2020, at an average age of 22.2 years, 113 participants (0.74%, 95% confidence interval [CI] = 0.61–0.89) had a register-based diagnosis of IBD, corresponding to an incidence of 31.3 per 100,000 person-years of follow-up; the incidence for Crohn’s disease (CD) was 11.1 per 100,000 person-years and 15.8 for ulcerative colitis (UC). Of 77 participants with a register-based definition of IBD by the end of 2017, medical records were identified for 61 participants, of whom 57 had true IBD (PPV = 93%; 95%CI = 0.87–1.00). While oral 5-aminosalicylic acid treatment was equally common in newly diagnosed CD and UC patients, biologics were more often used for newly diagnosed CD. The median faecal calprotectin levels were 1206 mg/kg at diagnosis and 93 mg/kg at the last follow-up (P < 0.001). Conclusions In this population-based sample of Swedish children and young adults the cumulative IBD incidence was 0.74. The validity of register-based definition of IBD was high and supports using such data to identify IBD patients in cohort studies.