UCLA psychiatrists were reprimanded by the National Institutes of Health on Wednesday for what the agency said was their failure to get proper informed consent from patients enrolled in an ongoing clinical trial of a new anti-schizophrenia drug.

As part of the trial, many of the patients were taken off the drug to determine whether treatment was no longer necessary and, in the process, 23 of the 50 were reported to have suffered severe relapses, including hallucinations and paranoia.

One of the patients committed suicide and a second, Gregory Aller, has said that he threatened to kill both of his parents and attempted to go to Washington, D.C., to assassinate President George Bush at the order of space aliens.

The two young men's parents filed a complaint with NIH and that agency Wednesday sent the physicians involved, Dr. Michael Gitlin and Dr. Keith Nuechterlein of UCLA, a draft of its reprimand, although its contents have not yet been made public.

Sources at NIH confirmed Wednesday that the UCLA researchers were faulted for their handling of so-called informed-consent procedures. "We have been investigating this particular study for some time, and we have sent the draft report to them, as is our procedure," said one NIH source. "They have a month to comment back to us.

"We did find some problems with the informed consent," the source said. "I think that was most of it, the biggest problem. Parents of some of the patients have been quite concerned about this study. This study was peer reviewed in order to get funded, and I think what the parents had problems with was this idea of their getting off the drug for a certain time, and not being informed about the possible consequences."

It is not clear what penalties, if any, might be associated with the sanctions.

Gitlin did not return phone calls Wednesday and Nuechterlein could not be reached. Both have denied repeatedly, however, that they have done anything unethical, and have argued that they could not discuss details of either of the two disputed cases until the parents or the patient sign a release allowing such discussion. The parents have not done so.

"We don't try to produce relapses," Gitlin told Time magazine last year. "They occur even as we try not to have them." The goal of the project, in fact, was to identify patients who could stop receiving the drug--and thereby avoid its side effects--and maintain the successes they had achieved with it.