Letters Hormonal contraception and thrombosis

Authors’ reply to Costescu and Waddington

BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f2369 (Published 16 April 2013) Cite this as: BMJ 2013;346:f2369
  1. Frits R Rosendaal, professor in clinical epidemiology1,
  2. Frans M Helmerhorst, professor in clinical epidemiology of fertility1
  1. 1Leiden University Medical Center, 2300 RC Leiden, Netherlands
  1. F.R.Rosendaal{at}lumc.nl

Costescu and Waddington think that there is no reason for a European Medicines Agency review or any concern because oral contraceptives containing third generation progestogens, drospirenone, and cyprotereone acetate are all equally safe.1 2

Their main point is that all retrospective studies are fraught with bias. Undoubtedly, any study can be biased—we dealt with this point more than 10 years ago, when we dissected all conceivable biases that had been brought forward.3 None held true. Since then, many more studies have been published, with very different designs, and all have confirmed the increased risk of the oral contraceptives mentioned above over those containing levonorgestrel.4 In addition, new laboratory methods enabled us to warn about a thrombotic risk of oral contraceptives containing drospirenone even before epidemiological studies found evidence.5

The main difference between studies that did and did not show a higher risk is that those that did not were paid for by the manufacturers.6 Even the same database (general practice research database) showed an increased risk of third generation oral contraceptives when analysed with public funding, but the absence of such risk when bought by one of the manufacturers and reanalysed. This evidence led us to warn: “Competing interests and controversy about third generation oral contraceptives. BMJ readers should know whose words they read.”7 Not much has changed.

The choice is simply between a safe and an unsafe oral contraceptive. The risk of fatal pulmonary embolism due to the use of an unsafe oral contraceptive should not be minimised. There is a crucial difference between proving effectiveness and safety—effectiveness needs to be proved but harm does not. Reasonable doubt about harm is sufficient to render prescribing a drug irresponsible.


Cite this as: BMJ 2013;346:f2369


  • Competing interests: None declared.