Fruit & Vegetable Screeners in the Eating at America's Table Study (EATS): Overview
In the early 1990s, NCI staff worked with individual grantees to the National Cancer Institute's 5 A Day for Better Health Program to develop a short 7-item assessment tool to measure intake of fruits and vegetables (Thompson, FE and Byers, T. Dietary Assessment Resource Manual. J Nutr 1994;124(11S):2305S). This tool has been used widely to track changes in fruit and vegetable intake in specific population groups.Because fruits and vegetables seemed to be underreported using this tool, staff developed a new short assessment form in 1996. This form asked about consumption of fruits and vegetables other than salad and potatoes by time of day (morning, midday/afternoon, evening), while continuing to ask about the usual daily consumption of fruit juice, salad, and potatoes overall without reference to time of day. The performance of the new screener was compared to the performance of the 5 A Day screener. Both screeners produced large underestimates of total fruit and vegetable intake, as assessed by the 24-hour recalls, analyzed within a measurement error model. The new by-meal screener performed somewhat better, however. (See Thompson FE et al. Evaluation of 2 brief instruments and a food-frequency questionnaire to estimate daily number of servings of fruit and vegetables. Am J Clin Nutr 2000;71:1503-10.)
In 1998, cognitive interviews of about 30 men and women were conducted to refine the wording of specific food categories, portion size questions, and time frame issues (all day vs. by meal). These findings were incorporated into two versions of new screeners. Both of these screeners feature:
- more precise wording of foods asked;
- an additional question about dried beans;
- three additional questions about vegetable mixtures; and
- portion size questions for every food item.
Both new versions contain the same food item description and portion size ranges; they differ in that one asks about usual intakes of all items (termed the All-Day version), and the other asks about usual intakes of fruits and other vegetables by time of day (termed the By-Meal version). Both are machine scanable. In one timed test, the By-Meal screener took an average of 14 minutes to complete.
The performance of each new screener was evaluated in the Eating at America's Table Study (EATS). A sample of 462 adult men and women, living throughout the US, completed one of the two screeners and four 24-hour recalls, 1 per season, taken over a year's time. Fruit and vegetable intake estimated from each screener were compared to the 24-hour recalls using a measurement error model (see Thompson FE et al. Fruit and vegetable assessment: performance of 2 new short instruments and a food frequency questionnaire. J Am Diet Assoc 2002;102(12):1764-72).