The US government will back down from a plan to require long-term studies of new psychiatric drugs before allowing them on the market, regulators said yesterday.

The reversal of the recently adopted policy came after a panel of experts unanimously recommended against requiring such studies as a condition of approval. While such studies are needed, the experts said, delaying decisions on new medications would hurt patients.

The panel's vote came after it heard a barrage of complaints from industry executives, academic researchers and patient advocates. All the critics predicted that the policy would lead to delays in bringing new drugs to market while providing little new information that may not apply to most patients. They also warned that the policy would cause drug companies to scale back on developing new drugs because of the potential increase in expense and risk.

The new plan, which the Food and Drug Administration had begun to implement over the past six months, called for companies to conduct studies for as long as half a year before seeking approval of new drugs. Like many other medications, psychiatric drugs are typically approved on the basis of positive results from two short-term studies, each of which may last only eight weeks.

Because physicians routinely prescribe psychiatric drugs for much longer periods, the FDA had started demanding longer-term data, as do regulators in the European Union, said Thomas P. Laughren, director of FDA's Division of Psychiatry Products. After the emphatic rejection by the panel yesterday, Laughren said regulators will "back off."

Criticism of the plan was voiced in all 15 presentations made at the panel meeting yesterday, prompting the chairman of the advisory panel, University of Florida psychiatrist Wayne Goodman, to implore his fellow scientists to mount an argument in favor of the requirement, if only to play "devil's advocate." But all the panelists agreed with the academic researchers, patient advocates and industry executives from Merck and Co., Wyeth Pharmaceuticals, Eli Lilly & Co. and other companies in stressing that the new federal requirement would have adverse consequences.

In the real world, as many as half of all psychiatric patients switch medications after three months of treatment, and as many as 70 percent switch after six months, said David Michelson, executive director for neuroscience medical research at Eli Lilly, which makes Prozac and other psychiatric drugs.

Asking companies to conduct trials that show that medications work for six months or longer will lead to trials that focus on the small subset of patients who do well for such long periods, rather than on the majority that do not, Michelson and others said. As a result, added Gary Sachs, a Harvard University researcher who testified at the meeting, such data will be of little help to clinicians in the real world who usually have to deal with less predictable cases.

"I believe the public interest is not served by this requirement, and it could cause a lot more harm and confusion than benefit," Sachs told the panel. "It would be telling someone with a heart attack that we have a drug that we know works, but we can't give it because we don't yet know whether it would prevent further heart attacks."

Sachs and other experts said "effectiveness studies" that can guide clinicians about which drug to try first, and when patients should stop taking a medication, are very valuable -- but their complex design and requirements mean they are best conducted at public expense by research institutions such as the National Institute of Mental Health.

While that institute does fund such studies, Sachs said, "their commitment to do that is substantially less than we would wish."