A substantial number of parents perceive that their children have adverse reactions to food, but it is well documented that objective assessments agree with only one-quarter to one-half of parentally reported reactions. In order to prevent wrong diagnoses and curtail unnecessary or inadequate diets, primary health care providers need to deal with the parental perception of adverse reactions to food. A description of the prevalence and pattern of parentally perceived adverse reactions to food in children is needed to meet this challenge. The aim of the present study was to estimate the prevalence, incidence and cumulative incidences of parentally perceived adverse reactions to food in children younger than 2 years of age, and to study the duration of the reactions. A population-based cohort of 3623 children born in Norway was followed from birth until the age of two. At 6-month intervals, the parents completed questionnaires regarding the occurrence and type of any reaction to food. Information was available on the outcome measure at all age points for 77.4% of the families and these were used in the analyses; 3.8% of the cohort were entirely lost to follow-up. The cumulative incidence of adverse reactions to food was 35% by age two. Fruits, milk and vegetables accounted for nearly two-thirds of all reported reactions. Milk was the single food item most commonly incriminated, the cumulative incidence being 11.6%. The cumulative incidences of reported reactions to fruits and vegetables were 20.4% and 7.3%, respectively, with citrus fruits, strawberry and tomatoes as the most common food items in these groups. The cumulative incidences were less for food reactions associated with eggs (4.4%), fish (3%), nuts (2.1%) and cereals (1.4%). The duration of the reactions was short – approximately two-thirds of the reactions were not reported again 6 months later. However, the probability of remission depended on the food item concerned, the age at onset of reactions, and whether the reaction had been reported previously or not. Adverse reactions to food are reported by the parents of one-third of children in Norway before the age of two. The most striking feature of this study is the short duration of the food reactions, as approximately two-thirds of the reactions are not reported again 6 months later. Nevertheless, the high frequency of reactions attributable to milk is of concern. Milk is an important part of the Norwegian diet for children, and if removed from the diet its nutritional value is not easily replaced. Further studies are needed to assess the degree to which parents alter the diet of their children based upon perceived reactions to food.