Ten studies (total of 2,355 participants) were included in the review; the prevalence of IBS in included studies ranged from 21% to 78%. Five of the 10 studies were graded as level one. Assessors were blinded in all but one study .
Accuracy data were available for seven individual symptoms: lower abdominal pain; passage of mucus; feeling of incomplete evacuation; loose stools at the start of abdominal pain; more frequent stools at the start of abdominal pain; abdominal pain relieved by defecation; patient-reported abdominal distension. No individual symptom reported good discriminatory power: abdominal pain had a high sensitivity and low specificity; the other three abdominal-pain-related symptoms had higher specificity with reduced sensitivity (full results reported).
Combinations of symptoms:
The presence of three or more of the Manning criteria (see Other Publications of Related Interest, 1.) based on three studies with a total of 574 participants gave a pooled positive LR of 2.9 (95% CI: 1.3, 6.4). Those with less than three criteria present gave a pooled negative LR of 0.29 (95% CI: 0.12, 0.71). Excluding the original validation study from the analysis made no significant difference to the results. Based on one study of 602 participants the Rome I criteria had a positive LR of 4.8 (95% CI: 3.6, 6.5) and a negative LR of 0.34 (95% CI: 0.29, 0.41). Four studies with a total of 1,171 participants evaluated the Kruis model (see Other Publications of Related Interest, 2.), which includes a combination of symptoms, physical findings and laboratory tests. For a Kruis score of greater than 44 the pooled positive LR was 8.6 (95% CI: 2.9, 26.0) with a corresponding negative LR of 0.26 (95% CI: 0.17, 0.41). Again, excluding the original validation study from the analysis made no significant difference to the results.