Debate. What are the risks of third-generation oral contraceptives? Health policy and third-generation oral contraceptives.

In considering whether health policy should be changed based upon the evidence presented by four recent studies which found an increased risk for venous thromboembolism (VTE) with use of third-generation oral contraceptives (OCs) the first fact which must be noted is that the association observed in these studies is of marginal statistical significance. Prescribing bias and health user effect bias which would have occurred before the studies took place apparently had an effect on the results (and may indeed be solely responsible for the results). With an already low incidence of VTE in OC users (2-4 per 10000 woman-years compared to 1-3 for nonusers) the current studies indicate that users of the second-generation OCs have only one-half of the incidence of VTE reported previously. This biologically implausible finding may be explained by the health user effect bias. In order to reduce the risk of VTE more research in the use of screening techniques is required. Switching users from third- to second-generations products may simply move risk-bearing individuals to another group. Independent risk factors such as obesity and smoking increase the risk of VTE in OC users. Current research also provides no indication about the reliability of the interim (and statistically nonsignificant) finding that third-generation OCs may reduce the incidence of myocardial infarction. Since it appears that prescribing patterns led to more at-risk women receiving third-generation OCs only limited evidence is available for the development of appropriate public health policy at this time.