Validation of food portion size photographs among women in Nairobi, Kenya: A cross-sectional study

Abstract Background Validated food portion size photographs can increase accuracy of estimatingportion sizes during dietary surveys. Our objective was to assess the validity of food portion size photographs in estimating portion sizes to be used in 24-hour dietary recall food consumption study. Methods We recruited two hundred and six women of reproductive age (13-45 years) residing in Roysambu sub-county in the northern parts of Nairobi City, Kenya. Eleven foods from the Photographic Food Atlas for Kenyan Adolescents (9-14 years) were examined. Participants were served pre-weighed portions. After eating, each participant was asked to estimate the amount of food they consumed, using the Food Atlas. Validity was assessed by calculating percentage of estimates within and outside of ±10% of consumed portion size, the mean percent difference between estimated and consumed portions, Spearman’s correlation coefficients, and Pearson’s chi-square test. Results The proportion of participants with estimates within ± 10% of the consumed portion size ranged between 15-65%. Mean differences between the consumed and estimated portion sizes varied -45% for stewed beans to +60% for watermelon. Generally, small portions were overestimated while large portions were underestimated. Correlation coefficients ranged from 0.12 to 0.77 and all the coefficients were statistically significant except for watermelon (p=0.22). Accuracy of estimations was not associated with participants’ age or educational level. Conclusions The validity of the tested food proportion size photographs was adequate for quantifications of most food items. However, the study’s findings also indicated that further improvement is needed before wider use in Kenya.