In particular, doctors have attacked the way trials for a new drug aimed at preventing vomiting after operations have been carried out.
In studies of whether the drug ondansetron worked, 8,806 patients had taken part in the trials by July 1994. But 2,620 of these were given placebos and denied existing anti-nausea drugs which "though not completely effective or without side effects" do bring some relief, according to Dr Rebecca Aspinall and Dr Neville Goodman, anaesthetists at Southmead Hospital in Bristol.
The doctors acknowledge that when new drugs are first produced, placebo trials are needed because it is known that any medical intervention can appear to benefit some patients. By testing the new drug against a placebo, doctors can be sure of its real effect - and not just the effect of patients being given at least some apparent "treatment".
Once it is known to work, however, new drugs should be tested against existing products to find out which works best, rather than having drug companies sponsor yet more trials which involve placebos - 18 in this case - in order to build up an apparent weight of evidence in the drug's favour.
In the case of ondansetron, "it is difficult not to conclude that this was an example of the industry failing to seek information that would allow true comparison against rival products", the doctors say in this week's British Medical Journal.
To gain a licence, drug companies have only to prove that their drug works - not that it works as well or better than others. In future, comparative data should be mandatory before a licence is given and the NHS may need itself to run such trials, the doctors argue.Reuse content