That’s what John P. A. Ioannidis, an epidemiologist at the University of Ioannina School of Medicine, Ioannina, Greece, and Institute for Clinical Research and Health Policy Studies at Tufts University, believes, and has stated in an essay in PLoS Medicine:
Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias.
The Wall Street Journal has an article.
Disturbing if true, especially (to be ridiculously self-centered about it) considering how much I write about research findings in my science column and on this blog!
