Dietary Assessment & Focus Groups
My Food Choices.com - A Web-based Dietary Assessment Tool
Kimberly M. Gibson, MEd, MHS, RD, LD; Marilyn H. Hughes, PhD, RD, LD; Rebecca M. Mullis, PhD, RD, LD
University of Georgia, College of Family and Consumer Sciences, Department of Foods and Nutrition
My Food Choices.com is a web-based dietary assessment program designed to assess the eating habits of children. By combining a food frequency approach with Internet technology to collect dietary data, the use of database-driven technology lessens the burden of data collection and data entry. Culturally-appropriate food photographs offer visual stimulation to children completing the dietary assessment.
Background
My Food Choices.com will be used in the 2003-2004 school year to assess the eating habits of fourth grade children participating in Creating Healthy Generations, USDA-funded dietary intervention program to increase the intake of fruits and vegetables and decrease the intake of high-fat foods of inter city African-American youth and their families. My Food Choices. Com was developed as a method of measuring dietary intake of the children before and after the intervention.
Development Phase
Development of My Food Choices.com began in the Summer of 2002. Initial discussions included the location and the functions of the program. To allow flexibility of use, the decision was made to house the tool on the Internet. Following this decision, the domain name www.myfoodchoices.com was chosen. Initially, the concept of the site included the use of an audio-based question and response option as well as animated features to entertain the intended age group of fourth grade children. Researchers with prior experience in the school system recommended that the program use only technology that would be compatible with hard drive operating systems and Internet connection speed of the school computers. It was also recommended that additional features which might encourage gaming behavior or disturb the classroom/library surroundings be avoided. In response to these recommendations, a text-based question and answer response format was chosen and small images were used to decrease download time.
In Fall 2002, dietary interviews were conducted with twenty children regarding what they had eaten during the previous day, what foods they typically chose when eating fast food, what foods they liked to have for snacks. These foods were used in the development of the Creating Healthy Generations intervention curriculum and the My Food Choices.com food items. The current version of My Food Choices.com includes 55 foods and 24 portion-size and fat-content foods.
During this time, a paper-based questionnaire was piloted with twenty children from the Atlanta Public School system. This pilot test revealed that the children did not easily understand the response categories when the answers included "5 - 6 times per week", "one time per day", and "two times per day". The decision was made to change the "one time per day" wording to "seven times per week" and remove the "per day" choices. The children also reviewed different versions of food illustrations and photographs. Observation of the children revealed that in general, they more readily recognized the food item, particularly popular brand labeling, when photographs were used.
Pilot Testing of the My Food Choices.com program
In Spring 2003, My Food Choices.com was pilot-tested with 126 children from the fourth grade classrooms of four pilot schools in the Atlanta Public School system. Approximately 30 children from each school participated. Each child completed three dietary recalls and two My Food Choices.com assessments in a one-week time period. With the exception of one dietary recall, all of the data collection efforts were conducted on Saturdays to avoid disruption of the school schedule. Each child was given a $25 Wal-Mart gift card to participate as well as free breakfast and Subway lunch.
1st Saturday School Week 2nd Saturday
Dietary recall #1
My Food Choices.com #1
Dietary recall #2
Dietary recall #3
My Food Choices.com #2
Reliability Correlations
Pearson Correlations of the test-re-test reliability method revealed correlations comparable to published correlation ranges. Due to the inherent variability in food intake on a weekly basis, the measurement of the reliability is less robust than what might otherwise be expected.
Food item Pearson correlation
Orange Juice 0.40
Toast with Jam 0.56
Bananas 0.52
Apples 0.47
Potato Chips 0.35
Popcorn 0.56
Fried Chicken 0.46
Macaroni & Cheese 0.38
Tossed Salad 0.50
Broccoli 0.51
Snack Cakes 0.55
Chocolate Candy 0.54
Validation comparisons of the instrument to the 24-hour recall data are currently underway. Food identification numbers and a ratio of the child's reported serving size to FDA serving size from the dietary recall data will be compared to the reported frequencies of the My Food Choices.com using Pearson correlations.
Future Versions
Some items will be removed from the program due to changes in the intervention curriculum, such as cottage cheese and soft drinks. Other foods will be added or modified to more appropriately meet cultural preferences, such as variety and wording of snack cake question and the addition of honey bun. Some pictures will be updated due to feedback from the children. Following these changes, the tool will be piloted again with a group of four to five children prior to the use of the tool in October for the pre-intervention data collection.
For additional information regarding the myfoodchoices.com program:
Kimberly M. Gibson, MEd, MHS, RD, LD
kimmg@uga.edu
|
