A study of the use of a photographic food atlas to estimate served and self-served portion sizes

Background: Although it is generally accepted that a weighed intake is the most valid way of assessing nutrient intake, this is not always the most appropriate method. Photographs have been suggested as a useful aid in assessing portion size when other methods of recording dietary intake are used. Method: Male and female subjects (n=100) were recruited to assess portion sizes of served amounts of mashed potato or cornflakes using food photographs. A second group of subjects (n=40) were recruited to assess portion sizes of self-served mashed potato or cornflakes using the same photographs. Results: Estimates varied from -70.6% underestimation to + 198.7% overestimation for mashed potato, with less variation for the cornflakes where portions were served. Where the food was self-served, results ranged from -38% and -64% underestimation to +61% and +88% overestimation for cornflakes and mashed potato, respectively. Correlation coefficients between estimated and actual weights of food showed them to be of statistical significance (P<0.05). Conclusion: Food photographs are a useful and convenient aid in the estimation of food portion sizes.