A position paper dated October 1998 shows that managers at SmithKline Beecham - now GlaxoSmithKline - were concerned at the commercial implications of two clinical trials in which their drug Seroxat was given to children and adolescents with major depression.

The results of both trials, known as protocols 329 and 377, showed that the drug was no better than a placebo - an inert pill - in alleviating the children's depression. An internal unit at the drug company called the Central Medical Affairs team prepared a strategy.

The target, says the document - leaked to the BBC's Panorama team, which has made two programmes on Seroxat - was "to effectively manage the dissemination of these data in order to minimise any potential negative commercial impact". About 500,000 adults were at the time taking the drug in the UK. Seroxat was licensed for their use, but not for use in children. Even so, some 8,000 to 10,000 children were also on it because doctors can prescribe an unlicensed drug on their own responsibility.

The paper says that following consultations within the company, SmithKline Beecham would not submit any data to the regulators to get a statement on the efficacy or the safety of the drug. It says: "It would be commercially unacceptable to include a statement that efficacy had not been demonstrated, as this would undermine the profile of paroxetine [Seroxat]." It adds, however, that "positive data" from the first and bigger study, protocol 329, would be published in abstract form at a psychiatric meeting the following month and that a full manuscript of the trial "will be progressed".

It was eventually published in July 2001 with the conclusion: "Paroxetine is generally well tolerated and effective for major depression in adolescents."

It was only last spring, after Glaxo submitted the full data from protocols 329 and 377, together with a third study of depressed children on Seroxat and data from trials of the drug in children with obsessive compulsive disorder and social anxiety disorder, that the UK regulator, the Medicines and Healthcare Products Regulatory Agency (MHRA), realised that not only was Seroxat ineffective in children but that a bigger proportion of those taking it in the trials thought about killing themselves than among those on placebo. The drug was then banned.

Yesterday Alastair Benbow, GSK's head of European clinical psychiatry, said the document "draws inappropriate conclusions and it is inconsistent with the facts".

He insisted that there was no sign of a suicide problem until all the trials were put together.