Purpose To verify the risk of pediatric eating disorders in preschoolers who were born prematurely and relate it to neonatal aspects, breastfeeding and oral habits. Methods Descriptive and longitudinal study. The sample was made up of preschoolers who were born prematurely and who were followed up until 24 months of corrected age. Data collection consisted of analysis of medical records, interviews and application of the Brazilian Infant Feeding Scale (EBAI). Results 19 preschoolers participated, with a mean age of 5.11±0.51, born prematurely (gestational age of 32.05 ± 3.26 weeks), healthy and with normal oral motor function at 2 years of corrected age. The application of the EBAI made it possible to identify the risk of pediatric eating disorders in 31.57% of the sample. There was no difference between the groups with and without risk of pediatric eating disorders regarding neonatal variables, breastfeeding and oral habits. Only the variable feeding time during the neonatal hospitalization period showed a difference between the groups, being higher in children who were not at risk for pediatric eating disorders. Conclusion The risk of pediatric eating disorders occurred in almost a third of pre-school-age premature infants. The dietary transition time during the neonatal hospitalization period was the only variable that showed a difference between the groups with and without risk of pediatric eating disorders. Thus, one may consider that, the moment of introduction of oral feeding and the way in which the diet progresses until the full oral route, during neonatal hospitalization, may be related to eating behavior in other phases of child development.