Crowdsourcing Helps Rate Diet Quality Accurately

This article originally appeared here.
Crowdsourcing Helps Rate Diet Quality Accurately
Crowdsourcing Helps Rate Diet Quality Accurately

(HealthDay News) — Crowdsourcing can provide basic feedback on overall diet quality, according to a study published online August 4 in the Journal of the American Medical Informatics Association.

Gabrielle M. Turner-McGrievy, PhD, RD, from the University of South Carolina in Columbia, and colleagues examined the potential of crowdsourcing dietary ratings of food in photographs for assisting with dietary self-monitoring. They examined how closely crowdsourced ratings of foods and beverages in 450 pictures (rated by 5,006 peers; mean 18.4 ratings per photo), rated using a simple "healthiness scale," were related to the ratings of the same pictures by trained observers (raters). Raters provided a healthiness score based on criteria from the 2010 U.S. Dietary Guidelines.

The researchers found that for all photos there was a high correlation between the average of all three raters' scores with the peer healthiness score (r = 0.88; P<0.001). Peer ratings were in accordance with the hypothesized direction for foods/beverages to increase and to limit. Higher healthiness peer scores were seen for photos with fruit; vegetables; whole grains; and legumes, nuts, and seeds. Lower peer healthiness scores were seen for processed foods, food from fast food restaurants, refined grains, red meat, cheese, savory snacks, sweets/desserts, and sugar-sweetened beverages.

"The findings suggest that crowdsourcing holds potential to provide basic feedback on overall diet quality to users utilizing a low burden approach," the authors write.

Abstract
Full Text

Loading links....