A prominent case of scientific fraud is being seized on by critics of the pharmaceutical industry to highlight their calls for a crackdown on the use of scientific studies for marketing purposes.

Starting in 1996, Scott Reuben of the Baystate Medical Center in Springfield, Massachusetts, published a series of trials testing whether painkillers, including Pfizer's Celebrex and Merck's Vioxx, relieve post-operative pain. Now 21 of Reuben's papers have been shown to contain fabricated data, after he was investigated by Baystate officials. Many have already been retracted.

Reuben's studies were part of an array of small clinical trials funded by Pfizer and Merck after Celebrex and Vioxx were approved for market by the US Food and Drug Administration (FDA). The trials investigated the drugs' use in a variety of different medical situations. The firms were not aware of Reuben's fraud.

Because such trials are usually too small to get the FDA to add new uses to a drug's label, critics allege that they are run for marketing reasons. Doctors are free to prescribe how they see fit, and sales reps often give doctors reprints of papers describing the results of such trials. Drug companies say that these studies are legitimate preliminary investigations of new uses for their products, and reject suggestions that they are undertaken for marketing reasons.

Now the Reuben case has focused attention on calls for tougher controls on the way the pharmaceutical industry uses the results of small, post-approval trials. Merrill Goozner of the Center for Science in the Public Interest in Washington DC hopes the FDA will introduce tougher controls on the distribution to doctors of scientific papers describing the use of approved drugs in new settings. The current guidance is "very permissive", he argues.