Tailored Prompting to Improve Adherence to Image-Based Dietary Assessment: Mixed Methods Study

Area of Science:

• Nutrition science
• Digital health
• Behavioral science

Background:

• Accurate dietary assessment is crucial for personalized nutrition and health research.
• Current dietary assessment tools lack precision, are costly, or are burdensome.
• Image-based methods offer potential for improved dietary assessment reliability and accessibility.

Purpose of the Study:

• To quantitatively assess the impact of prompt timing on adherence to image-based dietary records.
• To qualitatively explore participant experiences with dietary assessment methods.
• To inform the design of a novel, user-centered image-based dietary assessment tool.

Main Methods:

• Randomized crossover design with 3 prompt settings: control, standard, and tailored.
• Participants completed 3-day image-based dietary records separated by washout periods.
• Qualitative data collected via semistructured interviews and questionnaires.

Main Results:

• Tailored prompting increased image capture rates by 1.78 images per day compared to control (P≤.001).
• Standard prompting showed a smaller increase of 0.83 images per day (P=.23).
• Participants preferred image-based records but noted challenges with post-meal recording.

Conclusions:

• Tailored prompting is effective in enhancing adherence to image-based dietary assessment.
• Future dietary assessment tools should integrate tailored prompting strategies.
• Combining image-based and written input options can improve overall record completeness.