The Development and Validation of Food Atlas for Portion Size Estimation in the Balkan Region

Assessment of portion sizes is an important factor for the accuracy of food consumption surveys. The objective of this study was to develop and validate a food atlas of commonly consumed foods in the Balkan region in order to improve the accuracy of portion size estimation for food consumption surveys. A list of 135 foods and their portion sizes was based on previously conducted food consumption surveys in this region. Food was cooked, measured and served in three or four portion sizes right before being photographed. A validation study was conducted through the visual perception method. Without receiving training on usage of the food picture book, participants were asked to evaluate two portion sizes of 20 selected foods by comparison with a photo series of each food. Portion sizes were evaluated by 18 nutrition professionals and 17 lay individuals who had no nutritional education. Mean differences and the standard deviations of the mean differences (SD) between the portions estimated by each participant and the served portion were calculated. The percentages of participants who selected the correct, adjacent or distant portion size also were calculated. The number of food items that were quantified within the predefined acceptable range (i.e., mean difference < vertical bar 0.7 vertical bar and SD < 1) was 16 (80%) among lay individuals and 17 (85%) among nutritional professionals. Among 16 photo series that were assessed as acceptable, the percentage of all participants, who selected the correct picture, was between 44.3 and 82.9%, with an average of 60.2%. Only three foods were assessed correctly by <50% participants. The percentage of participants who selected the correct or adjacent serving size was above 98% for both lay and professional evaluators. This is the first food atlas containing representative foods and recipes commonly consumed in the Balkan region. However, further adjustments of the methodology should include larger number of food items to be tested, involvement of more participants and provision of training for the users of the food atlas. This food atlas could be used in food consumption surveys in the Balkan region after further testing and validation.