Objectives To summarise the epidemiological evidence on the relationship between second- (OC2) and thirdgeneration (OC3) oral contraceptives (OC) and the mortality associated with deep vein thrombosis (DVT) and myocardial infarction (MI), and to extrapolate and balance the evidence for these risks to the population of French OC users.

Methods All studies published on the risk of MI during OC2 and OC3 use were analysed. For DVT the Committee for Proprietary Medicinal Products public assessment report published in 2001 and more recent studies published on this topic were used. The estimates of odds ratios (OR) for risk of death from DVT or MI were extracted from the published manuscripts. ORs were used to calculate the aetiological fraction of risk for death from DVT and MI in the population; the relative impact of OC3 compared to OC2 use was expressed as an excess risk of death overall and by age group for French women.

Results Compared with OC2, the use of OC3 would prevent a maximum of 24 deaths from MI per year and induce a maximum of 16 deaths. Conversely, OC3 would induce 282–940 excess cases of DVT per year, resulting in 28–94 pulmonary embolisms and 3–19 deaths in the 4.7 million French OC users.

Conclusion Balancing the evidence, it is difficult to conclude that the overall cardiovascular risk is significantly lower for either of the two OC schemes.