This Week In The Bmj

Results of clinical trials are not generalisable

BMJ 2004; 329 doi: https://doi.org/10.1136/bmj.329.7456.0-d (Published 01 July 2004) Cite this as: BMJ 2004;329:0-d

The selection process for participation in clinical trials means that participants may not represent average patients, and that extrapolating data can be misleading. Dieppe and colleagues (p 31) reviewed trials on non-steroidal anti-inflammatory drugs in osteoarthritis and compared the participant profile to information from the medicine monitoring database on 131 410 patients taking the drugs in Scotland. People at risk of adverse effects, minority groups, and older people were often excluded from trials. Drugs tend to be used for a wider range of indications than those for which they are trialed, and the harms observed in practice are different from those reported in trials. Databases that link prescribing to hospital data and other health records are needed to assess the relative benefits and harms of drugs, say the authors.

Sign in

Log in through your institution

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial

Subscribe