A comparison of children with functional abdominal pain and irritable bowel syndrome in primary care and secondary care

Context: Most children with functional abdominal pain (FAP) and irritable bowel syndrome (IBS) could be managed in primary care (PC). Nevertheless, many children are referred to pediatric care. It is not clear which children are referred and what the differences are between PC and hospital care populations. Insight in patient characteristics in both settings might invite to reflect upon referral strategies of general practitioners (GPs) and interpretation of study results from hospital care.

Objective: To compare demographic, clinical and psychosocial characteristics of children with FAP or IBS in PC with those seen in hospital care by a pediatrician.

Study Design and Analysis: This study compared baseline data from two randomized controlled trials (RCTs) studying effectiveness of hypnotherapy. Comparison was based on an estimation of potential clinical relevance.

Setting: One RCT was performed in PC and the other RCT was performed in a hospital setting including secondary and tertiary care patients.

Population studied: Children in the PC study were 7-17 years old, had FAP or IBS according to their GP, and were not being treated by a pediatrician. Children in hospital care were 8-18 years old and had a diagnosis of FAP or IBS based on Rome III criteria. Exclusion criteria for all children were organic gastrointestinal disease, intellectual disability, and previous hypnotherapy.

Outcomes: Age, gender, duration of abdominal pain symptoms, diagnosis based on Rome III (hospital study) or Rome IV (PC study) criteria, pain intensity and frequency scores, somatization scores, school absenteeism, pain beliefs, anxiety and depression scores and health-related quality of life.

Results: We included baseline data from 412 children, of whom 152 children were included in the PC study and 260 children in the hospital study. Sixteen children in PC did not have a Rome diagnosis of FAP or IBS (10.5%). Compared to PC, children in the hospital were older, had higher pain intensity and frequency scores, higher somatization scores, and more school absenteeism. Other characteristics were comparable between groups.

Conclusions: This study shows that GPs seem to refer children with higher age and school absenteeism, and more pain and somatization symptoms. GPs do not use Rome criteria, which might imply that they offer inappropriate treatment or withhold adequate treatment. Treatment effects found in hospital settings might not translate to PC, and RCTs in PC are thus needed.