Learn More about Technology in Dietary Assesssment
The measurement of dietary intake has, by necessity, traditionally relied on self-report instruments (see 24-hour Dietary Recall Profile, Food Record Profile, Food Frequency Questionnaire Profile, and Screeners Profile). A variety of such instruments exist, but they have high respondent burden, require costly processing, and the resulting data are prone to varying degrees of [glossary term:] measurement error (see Key Concepts about Measurement Error). To overcome some of these limitations, several research teams are developing instruments designed to improve [glossary term:] accuracy, reduce respondent and researcher burden, and automate the processing of data.
These instruments are at varying stages of development and are part of a growing technological toolbox that includes automated versions of 24-hour dietary recalls (24HR), food records, and food frequency questionnaires (FFQ). Although these newer tools have some expanded abilities regarding data capture and processing, the recommendations elsewhere in this Primer about data analysis and caveats concerning limitations of the data do still pertain (see Choosing an Approach for Dietary Assessment).
Examples by Type of Tool
24-hour Dietary Recall
The National Cancer Institute's (NCI) Automated Self-Administered 24-Hour Dietary Assessment Tool (ASA24) is an automated, self-administered web-based tool for collecting 24HRs.
- This freely available tool provides a complete system for probing, coding, and calculating dietary intake, including foods and supplements.
- It is available in English and Spanish, and versions tailored for children (ASA24-Kids) and Canada (ASA24-Canada) have been developed. Efforts are also underway to develop a version for Australia.
- Evaluation and validation efforts are completed and ASA24 has been shown to perform well relative to true intake and to interviewer-administered recalls.
Several automated mobile device food records (mdFR) have been developed that integrate camera, video, voice, text, and automated image processing options in different ways. Evaluation and validation efforts are currently underway for these tools.
- The Technology Assisted Dietary Assessment (TADA) project includes a mobile device food record (mdFR) that integrates camera, text, and sophisticated image processing based on images of foods taken before and after eating to automatically identify foods and portion sizes.
- Viocare's Food Intake and Voice Recognizer (FIVR) is an mdFR that combines video and voice to record foods before and after eating, and uses computer vision techniques and speech recognition software to automatically identify foods and portion sizes.
- The Diet Data Recorder System (DDRS) is a handheld device that records optical, acoustic, and direct-entry of data of food intake. It is not fully automated as coding of foods and portion sizes is done by trained coders.
- The eButton is a wearable button-like sensor system that includes video, global positioning system (GPS) sensor, accelerometer, ultraviolet sensor, and digital compass to assess both diet and physical activity. All coding is done by trained coders.
Food Frequency Questionnaire
Automated web-based versions of FFQs are available online and through private companies. These tools are a logical extension of their original machine-readable paper and pencil versions.
- NCI's Diet History Questionnaire.
- Harvard School of Public Health and the Brigham and Women's semi-quantitative food frequency questionnaires.
- Dr. Gladys Block's Block Food Frequency Questionnaire.
- Viocare's VioFFQ, a collaborative effort between Viocare, Inc., Dr. Alan Kristal of the Fred Hutchinson Cancer Research Center, and Dr. Phyllis Stumbo of the University of Iowa.
For More Information
Kirkpatrick SI, Subar AF, Douglass D, Zimmerman TP, Thompson FE, Kahle LL, George SM, Dodd KW, Potischman N. Performance of the Automated Self-Administered 24-hour Recall relative to a measure of true intakes and to an interviewer-administered 24-h recall. Am J Clin Nutr 2014 Apr 30;100(1):233-240. [Epub ahead of print] [View Abstract]