Background/aim of study Constipation is a frequent cause of referral to specialist clinics. Prediction of those who need investigation would be useful. Previous literature has looked at both autism and allergy as potential aetiologic factors in constipation. We studied factors predicting transit study results.

Methods Retrospective review of 90 consecutive colon nuclear transit studies in children. Clinical data as well as transit study results were reviewed.

Main results A significant minority had a history of delayed passage of meconium. These children tended to be smaller (Z score average-0.15) than both the general population and the overall group (NS). Moreover children with a history of delayed passage of meconium were more likely to have an abnormal study than those without a history of delayed passage of meconium. All children had a rectal biopsy, none had Hirschsprung disease.

The constipated group generally were more obese than the national average.

Overall 40.7% had diffuse slow transit; 38% had recto sigmoid retention and 22.3% were normal.

An allergic history was present in onethird, but did not predict the study result.

38% had an underlying neuropsychiatric problem, and these children were slightly less likely to have a normal study than neurotypicals: 13% vs. 30% (p = 0.0634 Fisher exact test).

Conclusion Early onset of symptoms (delayed passage of meconium) and a neuropsychiatric disorder both seem to predict abnormal transit study results. Food allergy is clinically relevant, but our data does not show an association with the outcome of the transit study.