Search This Blog

Tuesday, November 16, 2010

Research Studies Are Often Invalid

David H. Freedman's article, "Lies, Damned Lies, and Medical Science" recently published in The Atlantic calls into question the validity of scientific studies that doctors rely on to make important medical decisions.

Freedman highlights the work of meta-researcher Dr. John Ioannidis, one of the world's leading experts when it comes to the validity of medical research. Ioannidis believes that "as much as 90 percent of the published medical information that doctors rely on is flawed."

Even randomized trials, the scientific community's gold standard, can be wrong. Several problems are at play, including study design and duration, subject selection, measurements and analysis, and financial conflicts of interest.

But maybe even more importantly, the big questions that studies set out to answer in the first place could be the real issue. After all, research results that make media headlines are more likely to receive funding than those that confirm what other studies have already "proven" to be true.

What is the solution?

For doctors, understanding which studies have the potential to yield valid results and applicable information is essential.

For everyone else, Ioannidis suggests adjusting our expectations.


No comments: