A statement from the agency yesterday confirmed that the expert team had been dissolved, a week after the Guardian revealed that two of the four members held shares in GlaxoSmithKline, which manufactures Seroxat, and that an expert witness at the first hearing had not disclosed his connections with it and other drug companies making similar anti-depressants.

The "intensive review" of the drugs known as SSRIs (selective serotonin reuptake inhibitors) was announced in the Commons in December by the health minister Hazel Blears after public concern about withdrawal effects experienced by people who wanted to stop taking Seroxat.

Claims that a small minority of patients became suicidal soon after starting a course were also on the agenda.

Recently the Brecon coroner called for Seroxat to be withdrawn pending an investigation after he recorded an open verdict on a retired headteacher who had killed himself soon after being prescribed Seroxat.

The statement from the medicines control agency (MCA), which is part of the Department of Health, said: "We can confirm that we have now decided to dissolve the original group and appoint a new expert group to conduct the review.

"The membership of this new expert group has yet to be decided, and individuals' interests in the pharmaceutical industry will be taken into account when considering the appropriate membership. The timescale for completion of the review is uncertain."

The original group was drawn from the committee on the safety of medicines, which advises the MCA.

It is routine practice at CSM meetings for scientists to declare their shareholdings, consultancies and other payments from drug companies whose medicines are being discussed. In the case of shareholdings, they must leave the room.

When the group met for the first time to review the SSRIs in November, before Ms Blears announced its existence, the two Glaxo shareholders had to leave the room when Seroxat was discussed, but they were able to be present during general discussions of the SSRIs.

The minutes show that an expert witness with many drug company links did not make a full disclosure, although his recollection is that he did.

"We would emphasise that the members of the Medicines Act advisory committees are required to follow a code of practice relating to declarations of interests in the pharmaceutical industry," the statement said.

"We seek to ensure that the code is applied where members attend meetings of expert working groups such as those in the present case."

The Seroxat Users Group was unhappy about the conflict of interest on the review team and its apparent reluctance to look at evidence of the side-effects suffered by its 4,000 members. Sarah Venn, of the group, said: "I think it is a decision that had to be made. The review members that they had were put in an untenable position faced with personal conflicts.