Validity of Digital Photographic Images for Dietary Assessment of Participants with Low Frequency of Home-Made Meal Intake

It is important to clarify the factors that cause errors when calculating nutrient and food group intake using dietary surveys. This study aimed to clarify the characteristics of errors in nutrient and food group intake among those who ate out and those who consumed ready-made meals when they recorded their meals using photographic and weighing methods. In October 2020, 38 second- and third-year students who were enrolled in dietitian training programs at two Japanese universities were asked to record their meals using both photographic and weighed records. Nutrient calculations were performed from the dietary records of both methods, and data from 29 subjects (all female) for whom 4-d dietary records were obtained were analyzed. The subjects were divided into two groups (high and low frequency of home-made meals) based on the median number of home-made meals, and a correlation analysis of the nutritional calculations obtained using both methods was performed. The results showed that most nutrient and food group intake levels were correlated (Spearman's correlation coefficient r=0.6-0.9) for both the high and low frequency of home-made meals. However, no correlation was found for "seasonings and spices" or their component "salt" in the low frequency of home-made meals group, indicating that there are errors in the nutritional calculations, especially for salt. It is thought that errors occurred for both the recorder and the nutrition calculator during the survey of meals eaten out; therefore, it is necessary to consider how to minimize errors before dietary surveys.