BACKGROUND: Irritable bowel syndrome (IBS) is a common health problem, often presenting in primary care as well as in internal medicine and gastroenterology outpatient clinics. Therapeutic options are dominated by drug therapies but there is uncertainty about their effectiveness. OBJECTIVES: The primary objective of this review was to evaluate the efficacy of bulking agents, antispasmodic and antidepressant medication for the treatment of IBS. SEARCH STRATEGY: A computer assisted search of MEDLINE, EMBASE, PsychInfo and the Cochrane Library was performed for the years 1966-2001; local and national databases were searched in 10 European countries. SELECTION CRITERIA: Randomised trials comparing bulking agents, antispasmodic or antidepressant medications with a placebo, in IBS patients over 12 years of age. Only studies published as a full paper were included. No language criterion was applied. DATA COLLECTION AND ANALYSIS: The search identified 687 studies, 66 of which fulfilled all eligibility criteria. After removal of cross-over studies that did not report separately on the first phase, data from 40 studies remained for analysis. Relative risk (RR), risk difference (RD) and standardized mean difference (SMD) along with 95% confidence intervals were calculated for all subgroups. The number needed to treat (NNT) was also calculated where appropriate. MAIN RESULTS: Forty-one study reports from 40 studies, comprising 78 comparisons, were analysed. These included 11 reports on bulking agents, 6 on antidepressants, and 24 on spasmolytics.BULKING AGENTS: Three studies comprising 159 patients reported a dichotomous outcome for relief of abdominal pain. The pooled RR using a random effects model was 1.22 (95% CI 0.86 - 1.73). Three studies comprising 128 patients reported a continuous outcome for relief of abdominal pain. Using the random effects model, the SMD was 0.68 (95% CI -0.86 - 2.33). Nine studies comprising 482 patients reported a dichotomous outcome for global assessment of improvement. The pooled RR was 1.09 (95% CI 0.78 - 1.50). Five studies comprising 253 patients reported a dichotomous outcome for improvement of symptom score. The pooled RR using a random effects model was 0.93 (95% CI 0.56 - 1.54). Two studies comprising 70 patients reported a continuous outcome for improvement of symptom score; the SMD using a fixed effects model was -0.44 (95% CI -1.20 - 0.31). SPASMOLYTIC AGENTS: Eleven studies comprising 1260 patients reported a dichotomous outcome for relief of abdominal pain. The pooled RR using a random effects model was 1.34 (95% CI 1.13 - 1.59; RD=0.17, 95% CI 0.06 -0.28; NNT=6, 95% CI 4 - 15). Seven studies comprising 467 patients reported a continuous outcome for relief of abdominal pain. Using a fixed effects model the pooled SMD was -0.65 (95% CI -0.94 to -0.35). Sixteen studies comprising 1236 patients reported a dichotomous outcome for global assessment of improvement. The pooled RR using a random effects model was 1.42 (95% CI 1.17 - 1.72; RD=0.20, 95% CI 0.09 -0.30; NNT=5, 95% CI 3 - 11). One study comprising 34 patients reported a dichotomous variable for improvement of symptom score. The RR was 1.33 (95% CI 0.96 - 1.85). Three studies reported a continuous outcome for improvement of symptom score; two studies comprising 66 patients could be pooled. Using a fixed effects model, the SMD was -0.37 (95% CI -0.85 - 0.12). ANTIDEPRESSANTS: Two studies comprising 81 patients reported a dichotomous outcome for relief of abdominal pain. Using the random effects model, the pooled RR was 0.83 (95% CI 0.33 - 2.12). Two studies comprising 101 patients reported a continuous outcome for relief of abdominal pain. The SMD using a random effects model was -0.53 (95% CI -2.29 - 1.23). Four studies comprising 241 patients reported a dichotomous variable for global assessment of improvement. The pooled RR was 1.16 (95% CI 0.78 - 1.73). AUTHORS' CONCLUSIONS: The evidence for efficacy of drug therapies for IBS is weak. Although there is evidence of benefit for antispasmodic drugs for abdominal pain and global assessment of symptoms; it is unclear whether anti-spasmodic subgroups are individually effective. There is no clear evidence of benefit for antidepressants or bulking agents. The physician should be aware that global assessment is a construct containing various dimensions. For each individual, these will have a different weighting and treatment should be aimed at the most debilitating symptom. Stool problems are by definition part of the IBS symptom complex. Bulking agents may improve constipation and can be used empirically, but should be evaluated at an early stage for individual benefit. Future research should pay attention to study methodology and the use of valid outcome measures.