Data sourcesPubMed, Web of Science, Scopus, Science Direct, EBSHost, Scientific Electronic Library online (SciELO).Study selectionPopulation-based cross-sectional and longitudinal studies assessing tooth erosion and diet, conducted in children and adolescents between eight and 19 years reporting on the permanent dentition were considered.Data extraction and synthesisStudies were selected independently by two reviewers and standard data items extracted. Study quality was assessed using the STROBE (Strengthening the Reporting of Observational studies in Epidemiology) statement and Newcastle-Ottawa Quality Assessment Scale (NOS). The pooled effect of dietary habits on tooth erosion occurrence was calculated using a fixed and a random model (OR and 95%CI).ResultsThirteen studies involving a total of 16,661 children were included. Eleven of the studies were cross-sectional and two longitudinal. Dietary habits data were mainly obtained from brief dietary assessments (69.2%) with food amount (weighed or estimated) and food frequency questionnaires used less commonly (15.4%). Most dietary assessments were self-administered (84.6%), assessed diet on a single occasion (61.5%) and required recalls of a week or more days or usual behaviours (46.2%). Meta-analyses were carried out for carbonated/soft drinks, sports drinks, milk-based drinks, yogurt, confectionery and snacks and acidic natural fruit drinks. Higher consumption of carbonated drinks or acid snacks/sweets and for acid fruit juices increased the odds for tooth erosion, while higher intake of milk and yogurt reduced the odds of erosion (see table).ConclusionsThe evidence indicated that some dietary habits (soft drinks, acidic snacks/sweets and acidic fruit juices) increased the odds for erosion occurrence, while milk or yogurt produced a protective effect. Methodological issues were shown to partly explain the heterogeneity of the data for some dietary products.