(P12-008-24) Development and Validation of Photographic Food Atlas: Guidance for Portion Size Estimation, Carbohydrate Counting, and Energy Intake Estimation

Objectives: This study developed and validated a photographic food atlas, utilizing domestic measures and weighed food portions to provide specific carbohydrate and energy information. Specific objectives included identifying commonly consumed foods by adults with diabetes, determining the proximate composition of the foods, and developing and validating the food atlas with portion sizes.

Methods: A cross-sectional survey sampled adults with diabetes and domestic measure sellers in the market to identify commonly consumed foods and common domestic food measures used in households. Proximate composition analysis of 23 selected foods was conducted using standard methods. Energy content was calculated using Atwater factors for the energy-giving food nutrients. Standardized photographs captured portion sizes supplying specific amounts of carbohydrates and energy for carbohydrate counting. Additionally, photographs of different sizes of domestic measures were respectively captured with their weights, carbohydrate serving, and energy content. Validation of the photographic food atlas involved adults with diabetes in a hospital. Statistical analysis used descriptive tools (mean, standard deviation, percentages) and inferential tools (Chi-square, student t-test), with significance set at p< 0.05.

Results: Rice dishes (rice + stew and jollof) were the most frequently consumed. Among banana varieties, red decca had the highest carbohydrate content (19.5g/100g), while the green mutant had the lowest (12.3g/100g). Energy varied, with the highest in the green mutant banana (125kcal/100g) and the least in the gross Michel banana variety (95kcal/100g). Carbohydrates in fruits and local dishes displayed diverse compositions. The photographic food atlas comprised 232 photographs from 23 food samples. Validation results revealed that more than half (50.1%) of participants scored 50% and above in identifying the correct photograph with the scooped portion, with 33.3% scoring < 40%. Validator's gender (p=0.035) and occupation (p=0.015) significantly influenced scores.

Conclusions: The photographic food atlas provided validated guidance for portion size estimation, carbohydrate counting, and energy intake estimation. It is applicable in surveys and clinical settings.

Funding Sources: None