Introduction Current diagnostic practice for paediatric CD in the UK includes increased screening of family members and at- risk groups for example, diabetes mellitus.

Aims To identify all incident cases of CD in SE Scotland over the 20-year period of 1990–2009 to assess trends in incidence, symptomatology, age at diagnosis and the impact of active screening of at-risk groups. Utility of routine laboratory tests at diagnosis was also evaluated (2005–2009 diagnoses only).

Methods A retrospective review of case notes, pathology databases, endoscopy and patient records was performed for all children diagnosed with CD <16 years from 1990 to 2009 on duodenal biopsy in SE Scotland (at-risk population group of 233 000 aged <16 years). Data were age-sex standardised and analysed in 5-year epochs, with Poisson regression models used to calculate changes in incidence over time.

Results 266 biopsy positive children were diagnosed from 1990 to 2009 with an increase in incidence from 1.8 (95% CI 1.1 to 2.7) to 11.7 (95% CI 9.8 to 13.9) per 100 000 children aged <16 years during the 1990–1994 and 2005–2009 epochs respectively (p=0.001). The median age (IQR) at diagnosis also increased significantly from 29 (16–53) months to 90 (53–132) between these epochs (p<0.0001), and non-classical presentation (children with a mono-symptomatic presentation and those with extra-intestinal symptoms for example, fatigue, pallor, irritability) increased significantly from 5% to 21% respectively (p=0.008). Additionally, 7% of children were diagnosed through targeted screening in 1990–1994 compared to 23% in 2005–2009 (p=0.002). When cases identified via active screening or with non-classical symptoms were removed, a significant rise from 1.51 (95% CI 0.91 to 2.38) in 1990–1994 to 6.59 (95% CI 5.17 to 8.28) in 2005–2009 (p=0.006) remained. Routine blood investigations demonstrated that over 25% were anaemic and over 50% were iron deficient at presentation in 2005–2009.

Conclusion The incidence of paediatric CD has increased 6.5-fold over the last 20 years in SE Scotland. Children are older at diagnosis, presenting with fewer classical and more varied symptoms, and with iron deficiency anaemia the commonest laboratory abnormality. We show that both increased screening of at-risk groups and a lower threshold for serological screening (as a result of greater clinician awareness of the condition's heterogeneic nature) do not fully explain the overall rise in incidence.

Competing interests None declared.