Validity and reliability of a dietary assessment method: the application of a digital camera with a mobile phone card attachment.

Abstract

This study was aimed at evaluation of the validity and reliability of an alternative dietary measurement method that assists epidemiologic studies. We validated a handheld personal digital assistant with camera and mobile phone card, called Wellnavi, in which a 1-d weighed diet record was employed as a reference method. Twenty college students majoring in food and nutrition participated in this study. They were asked to keep a diet record and to take digital photos of all these recorded food at the same time, then send them to the dietitians by the mobile phone card. In the reliability study, other twenty students from the same college were asked to take digital photos of the same meal during a day by two same instruments under the same circumstances and to send these photos to the different dietitians electronically. With respect to validity, median nutrient intakes estimated by the Wellnavi method and the diet record method are comparable. Correlation coefficients between the median nutrient intakes estimated from these two methods ranged from 0.46 for monounsaturated fatty acid to 0.93 for vitamin B12 and copper (median r = 0.77). With respect to reliability, our data show a good agreement between two Wellnavi instruments for most of the nutrients. Correlation coefficients between the nutrient intakes estimated from 2 instruments ranged from 0.55 for vitamin B1 and water-insoluble dietary fiber to 0.92 for vitamin B12 (median r = 0.78). In conclusion, the results indicate this dietary assessment instrument can usefully measure individual dietary intakes for a variety of nutrients in an epidemiologic study.