Food habits and nutrition education--computer aided analysis of data.

Nutrition education messages should take into account the food habits of those who are to be educated. These can be revealed by computer analysis of weighed intake data, which has been collected for calculation of nutrient intakes. Seventy-six students and staff at Robert Gordon's Institute of Technology weighed their food for 1 week and the records were used to determine the frequency of consumption of foods and portion sizes, as well as nutrient intakes. There were only very minor relationships between the number of different foods chosen and nutritional variables. Nutrition students had successfully changed the frequency of consumption of certain foods relative to others and as a result consumed diets containing a lower proportion of energy from fat. Messages to non-nutrition students might profitably incorporate those beneficial changes that nutrition students had easily accomplished. This study revealed that certain (otherwise common and nutritionally unsound) food choices were not a major part of the subjects' habits, and could be given low priority in educational messages. It was suggested that foods exhibiting high variability of portion weight might be under greater individual control and hence more amenable to change. A study of the distribution of portion weights reveals information about number of slices, biscuits, etc, taken in each portion. This varied for different kinds of biscuit. It was concluded that messages should target specific foods rather than stress variation in the diet. It is suggested that educators should consider whether messages would be more effective in terms of frequency of consumption or size of portion for particular groups. The meal distribution pattern also shows which foods are most commonly consumed at home or in the canteen, hence whether education might be best directed to the individual or the caterer, respectively.