We compared the results of randomised trials comparing taxane-containing chemotherapy regimens with regimens not containing a taxane in women with metastatic breast cancer. The specialised register of the Cochrane Breast Cancer Group was searched in March 2004. Eligibility was assessed and data extracted from eligible studies by two reviewers. Hazard ratios (HR) were derived for time-to-event outcomes, and a fixed-effect model was used for meta-analysis. Tumour response rates were analysed as dichotomous variables. Of 21 eligible trials, 16 had published some results and 12 data on overall survival. An estimated 2621 deaths among 3643 women suggest a significant difference in overall survival in favour of taxane-containing regimens (HR 0.93, 95% confidence interval (CI) 0.86-1.00, P=0.05). The treatment effect on survival was similar if only trials of first-line chemotherapy were included, although not statistically significant. There appeared to be an advantage for taxanes in time to progression (HR 0.92, 95% CI 0.85-0.99, P=0.02) and overall response (odds ratio (OR) 1.34, 95% CI 1.18-1.52, P<0.001). There was significant heterogeneity across the trials (P<0.001), partly because of the varying efficacy of the comparator regimens. Taxane-containing regimens improved overall survival in women with metastatic breast cancer. Taxane-containing regimens are more effective than some, but not all, nontaxane-containing regimens.