Assessing Gastrointestinal Tolerance in Healthy Adults: Reliability and Validity of a Weekly Questionnaire

Abstract

When examining gastrointestinal tolerance to nondigestible carbohydrates, a weekly vs. daily symptoms questionnaire may lessen participant burden. This secondary analysis examined the reliability and validity of the Gastrointestinal Symptom Rating Scale (GSRS) in healthy adults. The internal consistency reliability of the GSRS syndromes and a daily questionnaire (DQ) comparator were determined. The GSRS syndromes prediction of slow transit stool form was assessed by ROC analysis. The DQ (α = 0.76) and GSRS syndromes of constipation (α = 0.73; ω = 0.74), and diarrhea (α = 0.76; ω = 0.77) exhibited acceptable reliability, as did the GSRS overall (α = 0.76; ω = 0.87) but not the syndromes of abdominal pain (α = 0.54; ω = 0.54), reflux (α = 0.69; ω = 0.67), or indigestion (α = 0.64; ω = 0.67). The GSRS syndromes predicted slow transit stools (AUC = 0.855), and the GSRS items of stomach pain, nausea, flatus, constipation, and diarrhea were moderately correlated (ρ = 0.55-0.64; P < 0.001) with the corresponding DQ items. The GSRS may be useful to assess gastrointestinal tolerance and efficacy of nondigestible carbohydrates given its performance at predicting slow transit stools, suggestive of constipation.

Keywords: gastrointestinal symptoms; gastrointestinal tolerance; nondigestible carbohydrates; stool form; stool frequency.

FIGURE 1

Receiver-operating characteristic (ROC) curve of the optimal prediction model of the Gastrointestinal Symptom Rating Scale syndromes using a boosted regression tree estimation to predict slow transit stools (any reporting of BSF 1 or 2) in healthy adults. Area under the curve (AUC) = 0.855. Blue and red lines indicate “slow transit” and “no slow transit,” respectively, with equal prediction.