14781-koppen

34 Chapter 1 METHODS Study selection and characteristics The Cochrane Library, PubMed and EMBASE databases were searched from 2006 (the year the Rome III criteria were published) until September 2017. The protocol including the full search strategy is provided in Supplemental file 1. To identify additional studies, reference lists of review articles and included studies were searched manually. No language limits were used. This systematic review was not registered. Articles were eligible for inclusion if they met the following inclusion criteria: (1) prospective studies of population-based samples (e.g., birth cohort studies and cross-sectional studies of school samples or general population samples); (2) reporting the prevalence of FC and/or FNRFI according to the pediatric Rome III or IV criteria; (3) in children 0-18 years of age; and (4) published in full manuscript form. Abstracts were screened for eligibility by two authors independently (IJNK, MHV) through Covidence (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia; available at www.covidence.org) . In case of disagreement, consensus was reached by discussion or by consulting a third author (MMT). All manuscripts that were considered eligible for inclusion were retrieved and read in full text to assess if they fulfilled the inclusion criteria. One of the Rome III/IV criteria includes findings on physical examination. However, since a physical examination is often not performed in epidemiological studies we decided not to exclude studies for omitting this Rome criterion. Moreover, studies were included if only minor changes were made to a maximum of one of the Rome criteria based on the medical history, such as the criterion regarding duration of symptoms. Any changes made to the original Rome criteria are reported in the results and were taken into account during the quality assessment. Guidelines of the Preferred Reporting Items for Systematic Reviews and MetaAnalyses (PRISMA) statement (available at www.prima-statement.org) were followed (Supplemental file 2). Quality assessment There is currently no gold-standard quality assessment tool for epidemiological studies. For this study, we used an assessment tool that was previously developed by our research group, specifically designed for quality assessment of epidemiological studies on functional abdominal pain disorders in children. 15 After adjusting this tool to address FDDs instead of functional abdominal pain disorders, we evaluated all manuscripts according to the following six questions: (1) Is the method of subject selection described and appropriate? (2) Are subject characteristics sufficiently described, i.e. do they match the target population regarding to gender and age? (3) Are FDDs diagnosed according to the Rome III or Rome