On December 14, 1995, the first public presentation of results of the Transnational Case-Control Study of Oral Contraceptives and the Health of Young Women was made at the winter meeting of the British Pharmacological Society at Brighton. Four weeks later, those results and further analyses were published in the British Medical Journal. There has been much debate about them in Europe and elsewhere. I feel privileged to have been invited to write a Leading Article about the safety of combined oral contraceptives in the British Journal of Clinical Pharmacology. These are personal opinions about the meaning of all the recent pharmacoepidemiological findings. With an historical context as the background I have added my personal reflections as a scientist and a public health doctor. These views do not necessarily reflect the views of my co-investigators.Oral contraceptives are becoming safer (i.e. desogestrel and gestodene containing) OCs compared with non-users. Compared with the incidence in earlier studies, users of second generation (i.e. levonorgeThe epidemiological studies of the late 1960s and the early 1970s revealed relative risks among current users strel) products have also experienced low relative and absolute risks for MI compared with non-users [25 ]. of what we now call first generation oral contraceptives (OCs), with oestrogenic content typically well above Apart from MI, the other concern about the pill for clinicians and women is stroke. The analyses of 50 mg, to be in the range of six to 10 for major cardiovascular and cerebrovascular events [1][2][3][4][5][6][7][8][9]. The epidemiological work in progress are not complete yet preliminary assessments of the data in the Transnational best news about the recent series of reports on epidemiological investigation of OCs on the market in Study, triggered by recent regulatory actions in the UK, mitigate concerns when we have monitored the frequenthe 90s, predominantly low-dose second and third generation pills, is that the risks are all appreciably cies of thrombotic stroke in users and non-users of OCs and the contrasts between users of second and third lower. The upper boundary of currently reported point estimates coincides approximately with the lower boundgeneration OCs. The burning question today is whether the occurrence ary of the estimates of relative risk observed in the earlier studies . It would mean that prudent of adverse events associated with thrombosis observed in users of the two main classes of combined OCs patterns of practice in family planning, successful research and development by academics and manufacdominating clinical practice in most developed and developing countries warrant selective regulatory action turers and sensible policies enacted by regulators around the world have all contributed to the reduction of the against either class. My view is that nothing has been published in the last few months which has altered the risk of all contraceptive pills over the last three decades compared with 'no use'.benefit-risk balance of...