Accuracy and cost-effectiveness of technology-assisted dietary assessment in healthy adults

Design: The study design will use a controlled feeding study cross-over design to compare the accuracy of three technology-assisted methods of assessing dietary intake: Arm 1: Automated Self-Administered 24-hour dietary recall (ASA24); Arm 2: Self-administered online 24-hour recall (Intake24); and Arm 3: Image-assisted mobile food record (mFR24). The study is a 3-period cross-over study, in which the sequence of the three groups, will be randomised for each participant. This study will obtain three (3) non-consecutive 24-hour recalls (24HR) from each of participants attending a face-to-face session. Each participant will therefore be exposed to each of the three techniques at different time periods. The crossover design, a repeated measurements design, will allow both between group and within group method comparisons. This is also the design of choice in dietary assessments, as both the errors between groups and errors within groups are important to examine. The wash-out period between each feeding day will be 7 days. 24 Hour Dietary Recall: The 24-hour Dietary Recall (24HR) is the current standard method used in population surveillance studies to assess usual intakes (kilojoules, nutrients and food and beverages) both in Australia and internationally. The 24HR, conducted as a structured interview, is designed to capture detailed information of foods and beverages consumed in the previous 24 hours. The 5-step multiple-pass method provides a structured interview format with specific probes and involves five structured sets of probing. ASA24 and Intake24 are automated self-administered web versions of the 24HR recall. Participants will undertake these self-administered methods on the day following the controlled feeding study through a web link emailed to their device. The ASA24 and Intake24 will be completed using the dedicated software on a computer without assistance. ASA24 and Intake24 take between 30-45 minutes to complete. The participant is guided through questions on food and beverages consumed at each meal, and includes a meal-based quick list, detail pass, final review and forgotten foods review. The mFR24 method will use images of meals and snacks captured by the participant in the previous 24-hour period. The images and related metadata are automatically uploaded to a secure cloud based server, thus viewing of the images and the metadata by the participant can be controlled so that the images are only available for viewing after midnight. On the day after the face-to-face feeding day, participants will review their eating occasion images on their mobile device using the “review” module of the mFR app and label the food and beverage items. This software allows the user to zoom in and show the time and date stamp that may assist with remembering the items consumed. A trained interviewer will conduct a 30 minute video interview where the participant will use a food model booklet that contains drawings and photographs to assist people with portion size estimation. All participants will be followed up by video/telephone to ensure they have completed the 24HR recall and to conduct a short survey on the method acceptability. Procedures: Prior to attending a face-to-face session, eligible participants will be randomised using a random number generator. The order the participant completes the recall will be allocated to ensure there is no order effect. Participants will be invited to attend a University study centre where they will have their height, weight and waist circumference measured, complete a demographic questionnaire and additional questionnaires to assess response, memory and cognitive biases. They will consume three meals (breakfast, lunch and dinner) of true intake (composition and amounts). The mFR24 group will be instructed on how to use the mFR App to take a before and after image of their meals and any food or beverage consumed outside of the meal times. Participants will select from a menu and will be blinded to the weight of the foods selected. The menu items will consist of commonly eaten food and beverage items taken from the Australian Health Survey data. All food items will be weighed pre- and post to determine the plated weight of each food and beverage item selected. Once the participant has finished eating, the tray is collected so that plate waste can be recorded by the research staff. The true weight of each food and beverage item will be recorded. Research staff will take a before and after image of participant’s food tray to be used for improving automated image analysis methods. Participants will return in weeks 2 and 3 to undertake the second and third feeding days and undertake their allocated 24HR the following day. At the end of the third feeding day, participants will complete an additional online questionnaire to evaluate method preference and their perceptions on each dietary assessment method using closed and open-ended questions. Focus groups – A qualitative methodological approach using focus groups will explore and describe perceptions and beliefs regarding the dietary assessment methods. Approximately 10% of the recruited sample (approximately 20-30 participants stratified by age group and gender) will be invited to attend a 1 hour focus group (5-8 participants) or a 30 minute in-depth telephone interview conducted by a trained researcher assistant. The focus group script with prompts will be developed by the research team and will cover questions on what would motivate them to take part in studies assessing diet, barriers and enablers to participation and retention and incentives most favoured. Discussions will be audio-recorded and transcribed verbatim for efficacy and accuracy by a professional transcription service. Analysis of the transcripts will commence immediately at the end of the first year, so that findings of earlier focus groups can be used to inform the discussions of subsequent focus group during Year 2. Transcripts will be managed using NVivo software. Sampling and analysis will continue until saturation. Mode of delivery: Participants will be required to attend face-to-face sessions for breakfast, lunch and dinner on the same day. A follow-up interview the day after the feeding study will be conducted either in-person or via a video interview. Location: The face-to-face sessions will take place in a University study centre and food laboratory.