Validity and reliability of the Bristol Stool Form Scale in healthy adults and patients with diarrhoea‐predominant irritable bowel syndrome

Abstract

The Bristol Stool Form Scale (BSFS) is a 7-point scale used extensively in clinical practice and research for stool form measurement, which has undergone limited validity and reliability testing. To determine the validity and reliability of the BSFS in measuring stool form in healthy adults and patients with diarrhoea-predominant irritable bowel syndrome (IBS-D). One hundred and sixty-nine healthy volunteers provided a stool sample and used the BSFS to classify stool form, which was compared with measured stool water content and with values from 19 patients with IBS-D. Eighty-six volunteers used the BSFS to classify 26 stool models to determine accuracy and reliability. Volunteers' classifications of stool type correlated with stool water (Spearman's rho = 0.491, P < 0.001), which increased in hard (Types 1-2), normal (Types 3-5) and loose stools (Types 6-7) (P < 0.001). The BSFS detected differences in stool form between healthy volunteers (mean 3.7, s.d. 1.5) and IBS-D patients (mean 5.0, s.d. 1.2) (P < 0.001). Overall, 977/1204 (81%) stool models were correctly classified (substantial accuracy, κ = 0.78), although <80% of Types 2, 3, 5 and 6 were classified correctly. On 852/1118 (76%) occasions, volunteers classified covert duplicate models to the same stool type (substantial reliability, κ = 0.72), but with only moderate reliability for Types 2 (63%, κ = 0.57) and 3 (62%, κ = 0.55). The BSFS demonstrated substantial validity and reliability, although difficulties arose around clinical decision points (Types 2, 3, 5, 6) that warrant investigation in larger clinical populations. Potential for improving validity and reliability through modifications to the BSFS or training in its use should be explored.