Aims/Design. Assessment of the association between the regular use of opiates by women during pregnancy and antepartum haemorrhage has been made difficult by the low prevalence of these conditions observed in clinical settings. As a consequence, most published studies lack statistical power due to the small number of cases. We combined all suitable published data in a meta-analysis to assess the association more accurately. Findings. Meta-analysis produced a pooled estimate of the crude odds ratio for antepartum haemorrhage in relation to maternal opiate use of 2.33 (95%CI 1.32-4.30). Conclusions. This odds ratio estimate is consistent with a moderately strong association. Studies included did not adjust for possible confounders (e.g. prenatal care, cocaine or tobacco), so confounding is likely to account for at least some of the observed effect. The relative consistency of results from studies included in the meta-analysis that were conducted in different countries and at different times adds to the strength of the evidence for an association between opiate use and antepartum haemorrhage. Well designed studies are urgently required to assess the independent effects of different types and patterns of maternal opiate use and confounders such as cigarette smoking, illicit cocaine use and antenatal care on antepartum haemorrhage.