Selection bias in clinical trials and other stories . . .

BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h6248 (Published 25 November 2015) Cite this as: BMJ 2015;351:h6248

Our drug works well for highly selected Martians

Clinical trials are supposed to inform clinical practice. If they don’t, they become ethically questionable. One of the biggest problems is selection bias or lack of external validity—a mismatch between the trial population and real world patients. This is so common that a large literature is dedicated to it, and in a systematic review of 52 studies across a range of specialties, about 70% of studies found evidence of significant and systematic mismatches that make it hard to apply trial findings to typical patient populations (Trials 2015;16:495, doi:10.1186/s13063-015-1023-4).

When “normal” serum sodium predicts death

The “normal” serum sodium range on a typical biochemistry results form is 135-145 mmol/L. But when investigators looked at sodium levels in two London hospitals they began to question …

View Full Text

Log in

Log in through your institution


* For online subscription