vendredi 24 août 2018

We cannot escape from the uncertainty zone using unreliable tools: pill and mortality

https://www.ncbi.nlm.nih.gov/pubmed/25361731

https://www.ncbi.nlm.nih.gov/pubmed/26333690

When the difference between treatment and placebo is low, only the RCT can come out of the uncertainty zone. You can try large numbers, cooperative studies, large-scale studies and long durations on huge databases, statistics and software and their software, you can not counteract known and unknown biases. Funding for epidemiological studies is a waste of money, except for researchers' careers.


Despite other criteria of causality these studies remain unreliable at least for the quantum of the risk. It is possible that the risk is larger, contrary to the opinion of autors and of the editorial piece of NEJM becaus e of unknown confounding factors which are non adressed in this paper. We need a randomized controlled study which will have difficulties to pass the ethics committee.
In another paper, authors minimized the risk. This it is very clear as smoker women in this series had no increased risk of all cancers ! In the discussion chapter of the paper, authors explain the multiple biases that they have noted.
http://www.lakemedelsvarlden.se/wp-content/uploads/2017/03/Iversenfinal-paper.pdf


"Counseling on Contraception The study authors agree that it’s imperative that physicians are upfront with patients about the possible risks of hormonal birth control so that women can make an educated choice based on their individual risk factors, including their genetics, age, weight, and lifestyle habits such as smoking or alcohol use. “Our study shows that it’s particularly important, if you want to limit the risk of breast cancer, to limit the total amount of years on hormonal birth control because the risk increases with increasing duration of use,” Mørch said. For instance, women in their 20s and 30s who have been taking hormonal contraception for several years may want to reevaluate their birth control options as they approach their 40s. Mørch also suggested that based on the data, women who have had breast cancer should not use hormonal contraception. Additionally, women who have a family history of breast cancer might consider looking into nonhormonalmethods of birth control, suchas thecopperIUD,diaphragm, tubal ligation or a vasectomy in a partner—or confining their use of hormonal contraception to their early years."
Peachman, R. R. (2018). Weighing the Risks and Benefits of Hormonal Contraception. JAMA, 319(11), 1083. doi:10.1001/jama.2018.0448 

Finally the risk of BC is not uniform and those studies brought nothing about counselling women. After reading those papers we are just in the same position: watch your own risk of BC and don't trigger it.

https://www.ncbi.nlm.nih.gov/pubmed/24458845

And as expected by rational analysis the risk is NOT the same with the different hormonal prep!

https://www.ncbi.nlm.nih.gov/pubmed/25085875

Aucun commentaire:

 
Paperblog