Research Updates | Consumption
Louie J et al aimed to develop a systematic methodology to estimate added sugar values on the basis of analytical data and ingredients of food.
A 10-step protocol was developed, this included objective measures (steps 1-6) followed by subjective estimations (steps 7-10). This protocol was then applied to an Australian food composition database (AUSNUT2007). Inter-rater repeatability of the methodology was measured by comparing results obtained by 2 researchers.
2,977 foods (77% of all foods in AUSNUT2007) were assigned an estimated added sugar value based on objective data (steps 1-6). Based on the absence of objective data the remaining 897 foods (23%) were assigned a subjectively estimated value (step 7-10).
Both researchers agreed on choice of steps for 2,946 (76%) foods. For the remaining 24% of foods, researchers chose different steps. Overall there was no significant difference between estimated values between both researchers.
The proposed method was shown to provide an accurate estimation of added sugar content for three-quarters of the foods in an Australian database of food composition. Subjective approaches were applied to the remaining 23% of foods.
Limitations: There were a number of limitations of this protocol, including the assumptions made (for example individual types of sugars were not always available and assumptions made about added sugar content within food groups). The accuracy of added sugar estimations were limited by subjective decisions made for the foods for which no objective data was available.