Recent oral contraceptive use patterns in four European countries: evidence for selective prescribing of oral contraceptives containing third-generation progestogens.

A survey among oral contraceptive (OC) prescribers in the United Kingdom, Germany, Sweden and the Netherlands was performed to investigate OC prescription patterns before and after recent publicity in the media about studies reporting a higher risk of venous thromboembolism with OCs containing third-generation progestogens as compared to OCs containing second-generation progestogens. Before this publicity, most physicians prescribed third-generation OCs as their first-choice formulations for normal healthy women as well as for young girls (< 20 years) and older fertile women (30-35 years). In women presenting with cardiovascular risk factors, third-generation progestogens (desogestrel, gestodene) were considered safer and were five times more often prescribed than second-generation preparations (71% versus 14%). Most prescribers considered ethinylestradiol to be the most important component in relation to the risk of venous thromboembolism (71%), myocardial infarction (66%) and stroke (67%). In addition, for women presenting with cardiovascular risk factors, OC preparations containing 20 or 30 micrograms ethinylestradiol were considered safer than preparations containing 35 or 50 micrograms ethinylestradiol. Although the vast majority of prescribers (78%) stated that their attitudes towards safety of third-generation OCs had not changed since the recent publicity and the regulatory actions in some countries, 56% had changed their prescribing practice, largely due to patient concern about the safety of third-generation OCs. The results from this survey strongly suggest that, prior to the recent publicity, most prescribers considered third-generation OCs to be safer than second-generation preparations. Because of this perceived better safety profile, physicians have selectively been prescribing third-generation OCs to women at increased risk of cardiovascular disease. This pattern of selective OC prescribing may have seriously biased the results of the recently published studies on OCs and venous thromboembolism in favor of second-generation OCs.