Evaluation of web-based, self-administered, graphical food frequency questionnaire

Area of Science:

• Nutrition Science
• Dietary Assessment Methods
• Public Health

Background:

• Traditional paper-based food frequency questionnaires (FFQs) have limitations in data collection.
• Computer-administered FFQs offer enhanced features like complex skip patterns and visual portion size estimation.
• The Graphical Food Frequency System (GraFFS) is a web-based FFQ designed to overcome these limitations.

Purpose of the Study:

• To evaluate the performance of the web-based Graphical Food Frequency System (GraFFS).
• To assess the intermethod reliability (vs. 24-hour recalls) and test-retest reliability of the GraFFS.
• To compare the measurement characteristics of the GraFFS with existing paper-based FFQs.

Main Methods:

• Participants (n=74) completed two administrations of the GraFFS, separated by time.
• Dietary intake was also assessed using six telephone-administered 24-hour dietary recalls over 12 weeks.
• Intakes of energy, macronutrients, and 17 micronutrients/food components were analyzed.

Main Results:

• Bias between GraFFS and recalls was small for energy (-9%) and macronutrient energy percentages (-5% to +4%).
• De-attenuated correlations (intermethod reliability) between GraFFS and recalls were high, with a median of 0.61 for micronutrients (range 0.40-0.92).
• Test-retest reliability correlations for GraFFS were also strong, with a median of 0.67 for micronutrients (range 0.49-0.82).

Conclusions:

• The Graphical Food Frequency System (GraFFS) demonstrates measurement characteristics comparable to or better than most paper FFQs.
• High intermethod reliability suggests the GraFFS is a valid tool for assessing dietary intake.
• Further development of computer-administered FFQs, like GraFFS, is supported by these findings for nutritional research.