Send to:

Choose Destination
See comment in PubMed Commons below
J Clin Gastroenterol. 2011 Oct;45(9):790-3. doi: 10.1097/MCG.0b013e3182053cea.

Importance of diarrhea in evaluating constipation in irritable bowel syndrome clinical studies.

Author information

  • 1GI Motility Program, Cedars-Sinai Medical Center, Los Angeles, CA 90048, USA.



Irritable bowel syndrome (IBS) is a chronic functional gastrointestinal disorder characterized by alterations in bowel function, abdominal discomfort, and bloating. The symptoms of IBS vary widely between constipation and diarrhea predominance. In this study, we examine "constipation minus diarrhea" score (C-D) to account for the contribution of diarrhea and to create a reliable method for predicting constipation predominance in IBS.


A nested study was conducted in participants with IBS by Rome I criteria. After consent, they completed a daily stool diary for 1 week. This included recording frequency of bowel movements and describing consistency of stool using the Bristol stool scale. After submitting their diaries, participants rated their bowel symptoms using a visual analog score (VAS) scale from 0 to 100 mm, with 100 mm representing maximum severity. VAS scores for C, D, and C-D were subsequently compared with the true stool events of the previous week by stool diary to validate the potential usefulness of the simple C-D method.


Eighty-four participants with IBS completed stool diaries and VAS surveys for bowel symptoms, from which C-D scores were compiled. By correcting for the predominance of constipation compared with diarrhea, C-D scores predicted constipation based on the frequency of bowel movements and Bristol stool score. Furthermore, a second method was used to validate the C-D technique by comparing the mean C-D score in participants with constipation (≤3 d with bowel movement in previous week). In this case, the mean C-D score was 74.1 ± 3.8 compared with -3.0 ± 5.9 for patients with >3 days with bowel movement in that week (P<0.001). Although the constipation severity alone also distinguished these 2 groups, the difference was less impressive (P<0.01). In participants with mean Bristol stool scores of ≤2 in the preceding week, the C-D mean score was 25.4 ± 6.1 compared with participants having mean Bristol stool score of >2 where the C-D was -11.1 ± 5.9 (P<0.01). Using constipation severity by VAS alone, there was no significant difference between these same 2 Bristol stool categories (P=0.29).


IBS patients' self-assessment of constipation and diarrhea severity using the C-D score is an effective predictor of C-IBS and constipation severity. The C-D score may be a useful tool in future IBS constipation trials, as it seems to predict constipation and it correlates well with stool diaries.

[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Lippincott Williams & Wilkins
    Loading ...
    Write to the Help Desk