Development of the FiberScreen: a short questionnaire to estimate fiber intake in healthy adults

Introduction: health effects of dietary fibers continue to be the topic of many human intervention studies. Eligibility criteria for these studies often include a low fiber intake, which requires a screening step in the selection process. However, current dietary assessment methods are unnecessary extensive and burdensome for both researcher and participant. Therefore, we developed a short fiber questionnaire (FiberScreen) to identify participants with a relative low or high fiber intake. Methods: The FiberScreen was validated as part of a high-fiber personalized dietary advice study, consisting of a 6-week intervention and a 3-month follow-up. Participants were healthy adults which completed the FiberScreen (version 1) and a 247-item meal-based food frequency questionnaire (FFQ) during study screening (n=131). Furthermore, an optimized FiberScreen (version 2) was tested at 3-month follow-up (n=87). In this study, a low fiber intake was defined as <26 grams for women and <33 grams for men (based on the FFQ), which is ≥15% below the recommendation. Results: FiberScreen Version1 included 5 items assessing intake of high-fiber product categories, namely fruit, vegetables, whole grain products, pasta/rice/potatoes and legumes. Total fiber intake was moderately correlated with fiber intake from the FFQ (r=0.356), but the separate product categories had a weak correlation with similar categories from the FFQ. FiberScreen Version2 included 9 items. Total fiber was strongly correlated with the FFQ (r=0.707) and product categories were strongly or moderately correlated (rfruit=0.707, rvegetables=0.457, rwhole grain products=0.603, rpasta,rice,potatoes=0.505, rlegumes=0.731, rnuts=0.469). FiberScreen Version2 had a sensitivity of 0.66, specificity of 0.89, a positive predictive value of 0.48 and a negative predictive value of 0.92. Moreover, it correctly classified 82% of the participants as having a “truly” high or low fiber intake. Estimated FiberScreen completion time is <10 minutes, whereas completion time of the FFQ is 30-60 minutes. Discussion: FiberScreen Version2 has an adequate comparability with the FFQ and is able to rank and identify participants with a low or high fiber intake. Therefore, the FiberScreen is a suitable screening tool to identify participants with a relative low or high fiber intake.