A major challenge in nutritional epidemiology is assessing dietary intake accurately enough to be able to reveal diet health associations. Read our Q&A with Professor Carol Boushey, University of Hawaii, USA.


How did you originally become involved in creating innovative dietary measurement approaches?
The real answer so dates me. I was working with Dr. Jean Hankin at the University of Hawaii on a project collecting dietary data among adolescents in Alaska. One of the goals was to examine iron bioavailability. An individual with an entrepreneurial spirit developed a computer program to analyse iron bioavailability which provided my first opportunity to analyse dietary data with a stand-alone computer running a pre-DOS operating system. Wow! I realized immediately that the process was a lot easier than sending things to the mainframe computers. I became sold on technology and started to work toward tapping into the positive attributes of technology to make gathering dietary measures faster and easier.

How does new technology help to accurately measure dietary intake?
As noted above, technology is faster and easier. Removing burden from researchers and practitioners, as well as patients and study participants helps accuracy. Ease helps cooperation which reduces non-reporting.

We often hear that specific nutrients are linked with specific diseases or indeed, disease prevention. How important is new technology in enabling researchers to make these associations?
The studies driving these results have primarily come from the food frequency questionnaire (FFQ). I think it important that dietitians and scientists know that the FFQ was an innovative breakthrough. The optical reader in a scanner could process thousands of completed questionnaires from cohort participants. These diet assessment results from millions of people provided the statistical power to find these links.

In your experience, what is the most successful method for accurately measuring intake? 
I am a true believer in maintaining and cultivating multiple methods of collecting dietary information. Whether the method is accurate is driven by the research question and study design not the tool itself. Thus, most methods have earned their time in the limelight.

Does using mobile technology allow for more accurate analysis of wider ethnic groups as participant burden is reduced?
Mobile technology likely allows for more accurate data collection due to the reduced burden and the wide distribution of mobile device ownership by ethnic groups and age (e.g., self-report with images from 5 year olds without help from mom or dad). The actual analysis is mixed as the automated methods and the trained analyst methods are still evolving. Nonetheless, the amount of information collected is so much greater than other methods that the errors with analysis are not readily apparent.

What could be the next development for an even more accurate measurement of dietary intake? 
If we could embed a working miniature mass spectrometer in an individual's molar (back chewing tooth) to measure compounds entering the body, along with a mini-mini camera over the ear; we could develop some pretty accurate assessments of dietary intake. Of course, the dental work and camera may reduce cooperation. Thus, we have to keep in mind maintaining the balance between accuracy and cooperation. I am also a fan of one day having a biomarker for every food using samples collected from blood, urine, sweat, and stools.