what they said (Hover the mouse cursor over underlined words for more info)

Associated Press
Cholesterol lowering drugs known as statins can increase the life expectancy of elderly patients by an average of two years, a new study by a doctor at the University of Arkansas for Medical Sciences claims.
The study by Dr. J.L. Mehta tracked nearly 1.5 million patients in the U.S. Department of Veterans Affairs' South Central Healthcare Network. Doctors prescribed statins to about 350,000 patients, almost half older than 70. Another 1.2 million patients did not receive the drugs...

The original article can be found at: http://www.ctv.ca/servlet/ArticleNews/story/CTVNews/20061010/statins_drug_061010/20061010?hub=Health

The original article can found in the Media Doctor archives.

how did it rate? (more information)

what we said (Hover the mouse cursor over underlined words for more info)

This is a relatively flawed report of a cohort study of people who take statins. The problem with these kinds of studies is that there are many potential ways bias can creep in. There is prevalence/incidence bias, selection bias, healthy workers bias etc. Which is to say, the study could simply reflect the relative healthier state of the responders as these are the ones who are still alive -thus overestimating or biasing the effect size.

Reports like this urgently need journalists to supply appropriate caveats that give the reader a sense of the strength of the evidence. A properly done randomized controlled trial can tell so much more about the effect of a treatment than any cohort study such as this and the reporter should be ready to warn readers that there may be many other explanations for the apparent greater life expectancy of the statin users, which may have nothing to do with the fact those people took statins. There is a reason why properly controlled studies strive for randomization, to eliminate many sources of potential bias and thus be able to conclude that the effects seen are due to the treatment and not other factors. Without randomization we get an interesting hypothesis, but certainly no cause and effect conclusions.

public forum

voice your opinion in the forum

• All comments and feedback submitted to Media Doctor are subject to editorial approval before being made viewable by the public. It may take up to a week for your comments to be approved. Additionally, no response will be given to questions posed in public comments. Media Doctor does not provide medical advice, or answers to medical questions posed by the public.
• If you provide your email address it will not be displayed to the general public.
• Comments may be edited by Media Doctor to remove defamatory or sensitive statements, and brand names.
• Required fields are marked with an asterisk (*).