University of California at San Francisco researchers studying experimental breathing support techniques on emergency room patients violated federal guidelines for obtaining their consent, investigators found.

In a study involving patients too sick to give consent for themselves, the researchers improperly persuaded relatives to approve their participation after a mere phone call, without first providing them any written description of the study and its risks, according to two federal reports obtained by The Chronicle.

The findings were made by investigators from the Office for Human Research Protections, an agency within the U.S. Department of Health and Human Services,

who were responding to an anonymous complaint made in 2000, after the UCSF study was completed.

UCSF's practices come to light at the same time the University of California is pressing state lawmakers to pass legislation that would relax state protections for human subjects.

That legislation, which could be approved within the next few weeks, would address the larger issue of whether it is proper to obtain consent from relatives first place. While federal guidelines govern most areas of human subject protection at UC, state law controls who can give proxy consent.

At least three other UC campuses have concluded that state law clearly forbids conducting research on people so ill they cannot give consent themselves, unless permission is secured from the patient's court-appointed conservator -- not simply a relative.

But for years UCSF lawyers have maintained the law can be interpreted to allow such consent by surrogates, and the medical research center continues to permit clinical trials on incapacitated people when researchers can get a relative's consent.

The proposed legislation would open the door to proxy consent not only for the university but to a broad category of research sponsors -- including drug and biotechnology companies -- that are under federal oversight.

Yet few of the lawmakers who have already cast early votes on the UC- sponsored bill were aware of the federal probe of the UCSF study. The investigation was never brought up during hearings in Sacramento.

Assemblywoman Dion Aroner, one of the co-authors of AB2328, said she was unaware that a UC campus had long been using so-called proxy consent from relatives who were not legally appointed conservators.

"Whether or not it's all related, I don't have a clue," said Aroner, who still supports the bill, but said questions raised by The Chronicle about its terms should be examined before its passage.

QUESTION OF CONSENT

While conservators are legally responsible to a court to act in the patient's best interest, say advocates of stricter controls, a relative who is suddenly approached by a doctor may not be equipped to make the best choice -- especially when the person must decide within hours, or even minutes, as in the case of emergency room patients being recruited as test subjects.

The problems that cropped up with getting relatives' consent in the UCSF study illustrate the often unavoidable conflict between the search for medical advances and patient protection.

The 105 UCSF patients were part of a nationwide study on 861 people to find out whether routine ventilator settings to support breathing in patients with pneumonia or other serious lung damage could be harming them by over- stretching their lungs.

The study, which ended in 1999, apparently proved the researchers' hunch: Lower ventilator settings decreased the death rate by 22 percent. Subjects in the nationwide study on average had a lower death rate than the usual 40-50 percent for patients with acute respiratory distress in ordinary medical practice. University officials said they were aware of no claims that subjects had been injured by the study.

But federal investigators found that at least in some circumstances, the scientists, who enrolled patients at UCSF Medical Center and San Francisco General Hospital, violated guidelines for protecting patients' rights.

Family members had nothing in writing describing the clinical trial and its risks when they gave the go-ahead over the phone, the investigators wrote. Neither did emergency room patients who agreed to the study on their own with only a "nodding gesture," since they were too sick to speak or hold a pen.

Consent documents given to relatives who granted permission in person did not adequately inform them of the possible hazards of the experimental treatment, which included breathing difficulties and a potentially life- threatening jump in blood sodium levels, the federal investigators concluded.

By considering the legalization of proxy consent by relatives, state legislators are grappling with one of the most sacred principles that have governed research on human subjects since the end of World War II. After the discovery that captives of the Nazis had been tortured in "experiments," the Nuremberg Code established that no human being should undergo experimentation without giving fully informed consent.

Regulations in the United States have tightened over the years in the wake of troubling research like the Tuskegee experiment in the mid-20th century, which left African American men untreated for syphilis. Congress is now considering reforms after a series of research scandals.

NEED FOR RESEARCH

Aroner said UC San Diego researchers had urged her to support AB2328 because the restrictions on proxy consent were blocking important research on Alzheimer's disease, which afflicted Aroner's mother. The San Diego campus takes the position that, unless the law is changed, only conservators can grant consent for incapacitated people to participate in experimental studies.

University of California spokesman Chuck McFadden said the UC-sponsored bill "arose from a desire on the part of researchers to get a clear set of guidelines or rules so they can achieve some assurance that they are not skirting the law."

Under the bill, researchers could seek consent from surrogates on a prioritized list: First in line would be the agents appointed by patients to carry out their wishes on health care decisions, followed by their conservators. But most people have not chosen such guardians. After that, relatives ranging from a spouse or adult child to a brother, sister or other relative with the closest kinship could be asked.

State Sen. Ray Haynes, who said he was also unaware of UC's history on proxy consent, said he opposes AB2328 because of his experience at a law firm that handled conservatorships. He said relatives can't always be trusted to make the right decision for a vulnerable person.

"I remember hearing family members complain they had to sell Mom's house to take care of Mom," Haynes said. "They wanted to preserve their inheritance."

Vera Sharav of the Alliance for Human Research Protection based in New York said the UCSF violations show that informal methods of proxy consent are prone to abuse.

"I think it demonstrates why you must not have this back-door approval process," Sharav said. "It invites exploitation of incapacitated people."

UCSF STUDY LED TO CHANGES

The UCSF study was run by the Acute Respiratory Distress Network, a consortium of research institutions that study severe lung disease.

Dr. John Luce, a San Francisco General Hospital pulmonary specialist who helped conduct the research, said the new ventilation method tested in the study has been adopted at all the hospitals involved in the research.

Luce acknowledged that errors were made in obtaining consent -- like asking permission from relatives over the phone without first sending them documents explaining the potential risks, or from patients who could nod but could not ask questions. But he said UCSF modified its policies as a result of the federal inquiry, even though the changes are hampering research.

But the federal Office for Human Research Protections, when it closed the inquiry April 11, did not resolve the core question: whether proxy consent is legal under existing California law.

"My impression is that OHRP would not wish to interpret state laws," said Richard Wagner, a UCSF human subject project administrator who worked with Luce on UCSF's reponses to the federal investigation.

Luce said UCSF relied on advice from its attorneys that proxy consent was legal, and he said such consent is also common practice at medical centers throughout the state.

Proxy consent is now forbidden at UCLA and UC Davis, but the bans were imposed only recently. Officials there would not say whether such studies using proxy consent had already been done.

The UC-sponsored bill could resolve the legality of proxy consent, but critics say the bill raises more questions than it answers.

Haynes said nothing in the bill prevents relatives from accepting payment for consigning their relatives to experiments. And Aroner wants the Senate Judiciary committee to mull such questions -- like how hard do researchers have to look for conservators or relatives high on the decision-making list before they move down to the next one?

Haynes said the Legislature should also address UC's level of disclosure before it asked for passage of the bill.

"If UC did this to essentially cover its tail without telling us what they did wrong, then that is a serious breach of ethics," Haynes said.