14781-koppen

42 Chapter 1 The most frequently used method of data collection was a questionnaire for children ( n =23, 62%) or parents ( n =10, 27%). Most questionnaire studies ( n =19/30, 63%) utilized the Questionnaire on Pediatric Gastrointestinal Symptoms-Rome III version (QPGS-RIII), a validated tool for diagnosing functional gastrointestinal disorders in children. Other studies utilized questionnaires or interviews which incorporated the Rome III criteria ( n =15). In four studies the Rome III criteria for FC had been modified (Table 1 and Table 2). All meta- analyses were performed using the random effect model due to significant heterogeneity. Quality assessment Adetailedoverviewof thequalityassessment of includedstudies isprovided inSupplemental file 3. In 10 studies, we scored the method of subject selection (item 1 of quality assessment tool) as ‘partial’ (1 point) because no clear statement was provided regarding the exclusion of children with known organic causes of constipation or fecal incontinence. Twenty studies (54%) used validated instruments to assess the prevalence of FC and FRNFI, these studies were scored with 2 points for using a reliable and valid survey instrument (item 4 on quality assessment tool). Two studies received maximum scores on all qualitative criteria. 25,39 Prevalence of FC The reported prevalence of FC ranged from 0.5% - 32.2% (Table 1). The pooled prevalence of FC was 9.5% (95% CI 7.5-12.1, Supplemental file 4a). Pooled prevalence and meta-regression according to the different study characteristics are depicted in Table 3. Geographical distribution A large proportion of studies (17/35, 46%) regarding the prevalence of FC were conducted in Asia. The highest prevalence of FC was reported in a Nigerian study (Supplemental file 5). 51 The pooled prevalence of FC based on geographical location is provided in Table 3. Due to the limited number of studies from the continents North America ( n =2), Central America ( n =2), Europe ( n =2) and Africa ( n =1), we decided to group results from continents with similar social and environmental circumstances together. We grouped North America and Europe together and did the same for South and Central America, these results were compared with those from Asia (Table 3). Based on this comparison, geographical location was found to be significantly associated with the prevalence of FC ( P = .018), with a high prevalence in the Americas and Europe and a low prevalence in Asia.