Wrong conclusions drawn, againBMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c2523 (Published 11 May 2010) Cite this as: BMJ 2010;340:c2523
- Joel Brind, professor1
Hannaford and colleagues again made headlines in the popular press with their claim of a 12% decrease in the rate of death from any cause with oral contraceptive use,1 echoing the similar reduction in overall incidence of cancer claimed for their 2007 paper.2 However, once again,3 detailed analysis suggests no such benefit, and even an opposite trend, given current trends of oral contraceptive use.
The lack of any significant life saving benefit is easily discerned from the difference between results for the full dataset and the general practice observation subset. Common sense suggests relying on the results from the smaller, better, dataset whenever statistical trends differ materially. However, the authors state: “The pattern of relative risks was different when we used the smaller general practice observation subset. In this subset, the adjusted relative risk for any death between ever users and never users was very close to unity (0.98, 0.88 to 1.10).”
Troubling trends of increased morbidity and mortality are also apparent in young nulliparous women, who constitute only a small minority of users in this study cohort but predominate among current users. For example, there was an almost threefold increase in death from any cause among women under 30, and a threefold increase in death from breast cancer for women under 45 between five and nine years after stopping use of oral contraceptives. Moreover, despite the decreased risk of death from breast cancer 10 or more years after stopping use of oral contraceptives in this latest report,1 the 2007 report showed a dramatically increased incidence of breast cancer (relative risk 2.45) persisting up to 20 years after stopping use.2 Such trends, however, are easily diluted by data from the majority of women in this study who overwhelmingly used oral contraceptives only after bearing one or more children and for a few years, contrary to current patterns of oral contraceptive use.
Cite this as: BMJ 2010;340:c2523
Competing interests: None declared.