Disclosure of the egregious ethical, methodological and legal violations outlined in the letter of determination by the Federal Office of Human Research Protections (OHRP) regarding the so-called SUPPORT oxygen experiment conducted on more than 1,300 especially vulnerable premature babies at 23 major medical centers, sponsored and endorsed by the NIH with taxpayer funding, is garnering critical reactions.
The New York Times editorial, “An Ethical Breakdown ,” calls the experiment “startling” and “deplorable.”
But even in light of the number of babies who were killed in the experiment, Dr. David Mangus, Director of the Stanford for Biomedical Ethics, defends the experiment in a NY Times article calling it "research that’s not risky."
We are particularly encouraged by the Times’ editorial board’s endorsement of one of AHRP’s recommendations: “if the institutions do not offer strong reforms, the agency can suspend their ability to conduct federally financed research on human subjects.”
In yesterday’s Infomail, we recommended: “the revocation of the Federal licensure from academic institutions that have been the site of unethical medical research whose subjects suffered permanent injury or death.”
And we also recommended holding medical researchers accountable.
An article in the Science section of the Times by Sabrina Tavernise (below), provides the context within which this experiment—and others like it—are being conducted and the ethical problems of doing research on especially vulnerable premature babies whose life hangs on whether the oxygen level they receive is right for their particular condition.
The trial design was unethical: researchers randomly assigned infants to two groups, with half receiving an oxygen concentration on the low end, 85% to 89%, and the other half to receiving 91% to 95%, on the high end—with no regard for the infants’ individual clinical condition.
The researchers violated legal requirements for informed consent; they did not explain to parents that the risk of an eye disease, retinopathy of prematurity, was greater in the higher oxygen group. Nor did they disclose that low oxygen concentration was associated with increased risk of death.
The results of the experiment found that infants in the upper band developed eye disease at more than twice the rate of those in the lower band — 18% compared with 9%. And the death rate was 20% for infants in the lower oxygen group.
But the study was designed without a proper control arm to detect either increased retinal damage in the high oxygen range or increased mortality in the low oxygen range.
The article reports that “the study designers agreed that the risk of blindness should have been more clearly explained,” but insisted that the infants were within the standard of care, not at greater risk than any premature infant. That is a misstatement: being randomly assigned to one of two fixed treatments—without regard for the babies’ clinical condition—is not standard of care.
The principal investigator, Dr. Neil Finer, Professor Emeritus at the University of California, San Diego, claimed “When we designed that study, there was no information from anywhere to suggest that you’d have increased mortality if you stayed within the range.” He did not comment about failure to disclose the risk of blindness to parents.
Dr. Jerry Menikoff, director of OHRP, said that it was “deeply troubling” that 23 institutions did not catch the failure to warn parents about the risk of blindness. In terms of punishment — which can range from institutions offering corrective action to their being forbidden to use government grants for research with human subjects — he said “everything is on the table.”
Dr. Menikoff added: “My goal is to make sure the parents are appropriately protected. It’s not too much to expect.”
To appreciate the nefarious contribution made by Bioethicists who lend support to unethical research practices–including the sacrifice of human subjects of research–note the reaction of Dr. Mangus to the OHRP disclosures:
“Dr. Magnus argued that the move by the agency would have a chilling effect. Such clinical trials are so difficult to get off the ground and can add such valuable scientific knowledge that raising what he argues are unnecessary alarms will hold back innovation.”
Even in light of the number of babies who were killed in the experiment, he calls this "research that’s not risky."
“Making research that’s not risky sound risky means we’ll be doing medicine based on hunches…I’m very worried about what this will mean for research that tries to answer the question: Of the things we are now doing, which one is best?”
April 15, 2013 An Ethical Breakdown
Despite reforms to protect patients from being harmed by medical research in recent decades, 23 academic institutions authorized a research project that failed to meet the most basic standard: providing an informed consent document to parents that accurately described the risks and benefits of the research to be conducted on extremely premature babies.
This failure was startling, and deplorable. Federal officials have rightly demanded that the University of Alabama at Birmingham, the lead institution, and Stanford, Duke and Yale, among others, take corrective action to prevent a recurrence. Such actions, which must be approved by federal officials, could include requiring investigators to undergo additional training in the ethical conduct of research, enhancing oversight by institutional boards that monitor research and notifying parents of the ethical breach.
The study involved more than 1,300 premature babies, born between 24 weeks and 27 weeks of gestation. The underdeveloped lungs in such babies are often unable to extract enough oxygen from the air to nourish the brain, so doctors often supply extra oxygen.
The danger is that too much oxygen can cause severe eye damage and blindness, whereas too little can lead to brain damage and death. The current standard of care calls for the blood to be saturated with 85 percent to 95 percent oxygen, but the researchers sought to pinpoint a more exact range that would minimize eye damage without increasing the risk of death.
The study, financed by the National Institutes of Health, was conducted between 2004 and 2009 and published in 2010. The researchers randomly divided the 1,300 babies into two groups; one got oxygen at the low end of the range, while the other got oxygen at the high end of the range. The parents were asked to sign consent forms that had been approved at all 23 medical centers by the institutional review boards that are supposed to ensure ethical conduct of federally supported research.
But, in this case, the federal Office for Human Research Protections concluded that the consent forms failed to reveal that there was a greater risk of dying in the low-oxygen group and a greater risk of severe eye damage in the high oxygen group. The form also stated that, because both groups would receive oxygen within the standard of care, there would be no predictable increase in risk no matter which group a baby was in.
But the federal agency found that many infants could have faced greater risks by participating. For example, if a baby whose clinical needs might ordinarily have led doctors to deliver a relatively high level of oxygen was enrolled in the study, the infant might be randomly assigned to receive lower levels of oxygen. The Department of Health and Human Services needs to investigate how this breakdown occurred. And if the institutions do not offer strong reforms, the agency can suspend their ability to conduct federally financed research on human subjects.
April 15, 2013
Crucial Studies, Fragile Subjects
WASHINGTON — How much oxygen should a premature baby be given in the first days and weeks of life? Neonatologists have been trying to answer that question since the 1940s, sometimes with disastrous results.
Premature infants need oxygen because their lungs are not fully formed, but early attempts to save lives by turning up oxygen inadvertently caused blindness in many babies.
“They did a lot of damage before they realized what was going on,” said Arthur L. Caplan, head of the division of medical ethics at NYU Langone Medical Center.
That experience has cast a long shadow over subsequent efforts to pioneer medical interventions for newborns. And the issue arose again last week when it came to light that a federal watchdog agency, the Office for Human Research Protections, had formally notified a network of 23 major research institutions that they had failed to warn parents about the risks of their infants’ participation in a large oxygen study.
The underlying ethical question remains: How do researchers balance protecting these most vulnerable patients from the risks of medical studies with the potential benefits of such research for all premature babies?
The purpose of the study, known as Support, was to find the sweet spot for oxygen concentration in an infant. The American Academy of Pediatrics had set the standard treatment at a band between 85 and 95 percent, and researchers were trying to determine what part of that was optimal.
The goal was laudable, the federal watchdog said, but the method was not. Researchers randomly assigned infants to two groups, with half the children to get an oxygen concentration on the low end, 85 percent to 89 percent, and the other half to get 91 percent to 95 percent, on the high end. But the researchers did not explain to parents in consent forms that the risk of an eye disease, retinopathy of prematurity, was greater in the higher oxygen group. Based on past studies, there was reason to expect that infants in the upper band would be at higher risk. Indeed, the results found that infants in the upper band developed eye disease at more than twice the rate of those in the lower band — 18 percent compared with 9 percent.
“Based on their very hypothesis, they were thinking that there might well be a difference,” said Dr. Jerry A. Menikoff, the director of the Office for Human Research Protections. “Being in the higher end should have put you at greater risk of developing eye disease.”
The study’s designers agreed that the risk of blindness should have been more clearly explained, but said that the infants were within the standard band of care, and therefore facing the same steep odds as any premature infant not in the study.
Dr. Menikoff disagreed.
“To be told that this was all standard care — it wasn’t,” he said. “It was taking a child and flipping a coin and giving them 50 percent chance of being at the higher end and 50 percent chance of being at the lower end. They were changing what happened to all of the children.”
Some experts contend that there is a lot of guesswork in oxygen care because there still is no established evidence for a particular level of concentration and that the risks were effectively the same for infants in the study as those outside it.
“A really honest clinician just flips a coin in his head,” said David C. Magnus, a professor of pediatrics and the director of the Stanford Center for Biomedical Ethics. “Physicians could have their own views, but there was no evidence supporting any given spot on the band.”
In the late 1990s, neonatologists began calling for systematic studies to establish that evidence. Dr. Ola Didrik Saugstad, a professor of pediatrics at the University of Oslo, who is an expert in neonatology and oxygen, said there had been about 12 studies since. Among them were three large clinical trials involving thousands of babies, including the Support study.
The Support study’s results, published in 2010, surprised researchers. Infants in the lower oxygen band had a higher mortality rate — 20 percent — than those in the higher one — 16 percent. The finding has had a profound effect on medical practice.
“We changed our practices immediately, almost overnight,” said Dr. Saugstad, who did not participate in the study. “We increased saturation targets again.”
The letter from the watchdog agency said that brain injury and death were clear risks for the low-oxygen group, and that in addition to the risk of blindness, parents should have been notified of those too.
But Dr. Neil Finer, the study’s principal investigator, who is a professor emeritus at the University of California, San Diego, said there had been nothing to suggest that the lower end of the standard band would cause higher mortality, and therefore it was impossible to warn parents.
“When we designed that study, there was no information from anywhere to suggest that you’d have increased mortality if you stayed within the range,” he said.
Still, the risk of blindness was not unknown, and Dr. Menikoff called it “deeply troubling” that 23 institutions did not catch the failure to warn parents about it. In terms of punishment — which can range from institutions offering corrective action to their being forbidden to use government grants for research with human subjects — he said “everything is on the table.”
Dr. Menikoff added: “My goal is to make sure the parents are appropriately protected. It’s not too much to expect.”
But Dr. Magnus argued that the move by the agency would have a chilling effect. Such clinical trials are so difficult to get off the ground and can add such valuable scientific knowledge that raising what he argues are unnecessary alarms will hold back innovation.
“Making research that’s not risky sound risky means we’ll be doing medicine based on hunches, who the doctor trained with, and which drug company gives away the best gifts,” Dr. Magnus said. “I’m very worried about what this will mean for research that tries to answer the question: Of the things we are now doing, which one is best?”
Dr. Saugstad, the Norwegian expert, said that the warning about the consent forms was concerning, and that scientists and neonatologists would be discussing it at a meeting in Washington this month.
“I was really surprised when I read about it,” he said. “I thought that everything had been done in a very careful way. And apparently it had not. We really have to think once again about how to do this in a proper way.”