Baylor College of Medicine
Food Intake Recording Software System: Version 4
Multiple 24-hour dietary recalls (24hdr) are the preferred method in assessing habitual dietary intake because they provide the most accurate data. This includes estimates of actual intake (as opposed to relative ranking), details on meal and snack consumption, day-of-week consumption, and detailed correlates of dietary intake (e.g. source of foods, location of eating, who was present during the eating event, whether the TV was on). Unfortunately, widespread use of 24hdr methods has been limited due to the high cost of certified dietitians performing in-person, one-on-one recall sessions, and expensive computer-based systems to assist the dietitians.
Most current 24hdr methods used with children are also limited by relying on text descriptions of foods and food portions instead of pictures. FIRSSt2 (Food Intake Recording Software System, version 2) was a low-cost, picture- based, computerized 24hdr system targeted to children for quantifying just fruit and vegetable intake. FIRSSt2's promising results eliminated the need for one-on-one dietitian interviews and more expensive dietary assessment software.
In response to FIRSSt2, the National Cancer Institute (NCI) initiated the Automated Self Administered 24hdr (ASA24) for use with adults. NCI involved the leading figures in conducting 24hdr and created a comprehensive sophisticated interactive method based on the Automated Multi-Pass Method (AMPM) and using food images in progressively larger sizes to estimate portions. To capitalize on the impressive capabilities of ASA24, we adapted the ASA24 to the needs and capabilities of children with a system called ASA24-Kids.
ASA24-Kids, like ASA24, includes 14,000+ images of foods in progressively larger portion sizes along with animated avatars to guide children through the experience. Formative research was initiated to answer several questions to optimize the effectiveness of ASA24-Kids including:
- How well children categorize foods into provided categories and whether there are ages below which food categorization schemes substantially differ,
- The screen size for food related plates, utensils and foods; optimal food search strategy, and
- The optimal approach to food portion estimation.
This grant funded three formative research projects to answer these questions. Our research revealed that children used common categories for organizing foods, but had enormous diversity in the labels used for these categories, suggesting a type-in- the-food name (search) mechanism would be more useful than a hierarchical (browse) mechanism.
Children could easily place foods in accurate categories starting at about ten years old. Placing many food pictures (in progressively larger sizes) on a screen at the same time facilitated speed of portion size estimation. Children of eight or nine years of age encountered substantial difficulties in using ASA24. The food list terms in ASA24 were simplified and questions children could not easily answer were eliminated with no diminution in accuracy of estimates.
ASA24-Kids is a computerized method for conducting 24hdrs with children that attempts to approximate the quality of 24hdrs conducted by dietitians, but at much lower cost per recall due to not needing dietitians to conduct the recalls. Increased use of inexpensive dietary assessment tools will be invaluable as part of genetic or disparities research with children when multiple days of assessment are possible.
See this project's publications and patents