Factors affecting healthcare consultation in children with chronic constipation: a school based survey

Abstract

BACKGROUND: Constipation is a common paediatric disorder. To date, there are no community-based studies to detect healthcare consultation among affected children, and the factors determining it. OBJECTIVE: To identify factors determining healthcare consultation in Sri Lankan children with chronic constipation. METHOD: A cross sectional survey was conducted among 10-16-year old children. Five schools were randomly selected from 3 randomly selected provinces of Sri Lanka. From each school, children were randomly selected from academic years (grades) 5-11. Previously validated questionnaire based on Rome Ml criteria was used in the survey and questions were included regarding healthcare seeking for constipation. Questionnaire was administered in an examination setting to ensure confidentiality and privacy. Constipation was diagnosed using Rome III criteria. A child who had consulted a doctor during the previous 12 months for symptoms of constipation was considered as a healthcare consulter. RESULTS: Of the 2770 questionnaires distributed, 2694 (97.3%) properly filled questionnaires were included in the analysis. Four hundred and sixteen (15.4%) children had chronic constipation. Of them, only 16 (3.8%) had sought medical advice during the previous 12 months. Healthcare consultation was significantly associated with age less than 12 years and presence of family history of constipation (p<0.05, chi-square test). There was no such association with sex, social class and exposure to stressful life events (p>0.05). Healthcare consultation in children with stool frequency <3 per week, hard stools, painful defaecation, large volume stool, faecal incontinence, withholding posture and blood stained stools are respectively 18.8%, 31.3%, 31.3%, 50%, 31.3%, 68.8% and 37.5%. CONCLUSIONS: Healthcare consultation was significantly associated with age less than 12 years and presence of family history of constipation (p<0.05). There was no such association with sex, social class and exposure to stressful life events (p>0.05).