fb tracking

In Wake of Controversy Over Unethical NIH-Funded Premature Baby Trial, HHS Seeks Way Forward

Outcome of Aug. 28 HHS Public Meeting Could Weaken Rules for Protecting Human Research Subjects 

WASHINGTON, D.C. – Publicity over an unethical trial conducted on premature infants has sparked an unusual Aug. 28 public forum at the U.S. Department of Health and Human Services (HHS). The meeting is designed to solicit comments from experts and the public about the adequacy of protection of human subjects in clinical trials. Specifically, the department is looking at what risks should be disclosed to participants when the research is focused on the  so-called “standard of care” treatment given patients for a particular condition.

The meeting, to be held from 9 a.m. to 5 p.m. at the HHS Washington, D.C., headquarters, comes amid a growing controversy around the experiment conducted on premature babies, known as the SUPPORT study. It was funded by the National Institutes for Health (NIH) and conducted at about two dozen prominent research facilities throughout the country from 2005-2009, including Stanford University School of Medicine, Yale University School of Medicine, Brown University and Duke University.

In the trial, 1,316 premature infants were exposed to an increased risk of blindness, brain injury and death. One primary purpose of the research was to determine whether extremely premature infants were more likely to die or develop eye disease if treated with high or low amounts of oxygen. The possible risk of a baby dying or suffering eye damage was foreseeable, but none of the consent forms signed by parents mentioned the risk of death and only two noted the risk of eye damage. Long-standing ethical guidelines and the law require that when obtaining informed consent for research, human subjects be provide with, among other things, a description of any reasonably foreseeable risks.

The Office for Human Research Protections (OHRP), an office within the U.S. Department of Health and Human Services (HHS), received a complaint about the SUPPORT study approximately two years ago. On March 7, 2013, the agency sent a letter to the University of Alabama at Birmingham, one of the lead institutions for the SUPPORT study, noting significant deficiencies in the consent forms signed by parents of babies enrolled in the study. The other research centers involved were copied on the OHRP’s letter. But the agency did not demand adequate and meaningful corrective actions from that school or any of the other universities and hospitals involved.

Since Public Citizen publicized the lack of adequate informed consent for the trial in April and called for similar trials to be halted, a controversy has raged in the scientific community over what kind of consent is needed in clinical trials that test the relative effectiveness and safety of  two or more interventions that may be used for the clinical, non-research care for a condition. (Many of these interventions often are mischaracterized as “standard of care.”) The debate goes to the heart of how research is conducted in the United States and could have far-reaching, negative  implications if changes are made to weaken the ethical and regulatory standards by which trials are run.

The Aug. 28 meeting

Many in the academic research community, including the director of NIH, have forcefully criticized OHRP’s actions in the case of the SUPPORT study and sought to undermine OHRP’s independence and authority by  having OHRP’s findings overturned. As a result of this pressure, OHRP was forced to convene the Aug. 28 meeting to discuss issues related to informed consent for research involving interventions that are considered “standard of care” treatment in the non-research setting by the investigators of these studies.

According to the notice announcing the meeting, HHS seeks input on how rules designed to protect people in medical experiments should be applied when the research is focused on what some people characterize as “standard treatments.”HHS specifically wants input about how an institutional review board (IRB) should assess the risks of research that compares one or more “standard treatments” and what risks should be disclosed to subjects during the consent process.

Many critics of OHRP’s actions have sought to blur the line between human subjects research and clinical care. They appear to view the process for obtaining fully informed consent as an unnecessary impediment to conducting clinical trials and advancing medical knowledge. If these critics have their way, the rules for seeking informed consent for research involving experimental interventions that are being misrepresented as “standard of care” interventions and posing significant risks to subjects will be significantly weakened.

More than two dozen people are expected to speak at the meeting. They include Dr. Michael Carome, director of Public Citizen’s Health Research Group; Dr. Sidney Wolfe, founder and senior adviser to Public Citizen’s Health Research Group; Alice Dreger, Ph.D., professor of clinical medical humanities and bioethics at the Feinberg School of Medicine at Northwestern University; and George J. Annas, JD, MPH, the William Fairfield Warren Distinguished Professor and Chair in the Department of Health Law, Bioethics & Human Rights at the Boston University School of Public Health, School of Medicine, and School of Law who support OHRP’s conclusions that the study lacked informed consent and many others, closely allied with the SUPPORT study researchers and NIH, who are expected to attack OHRP’s position and endorse efforts to roll back long-standing standards for obtaining informed consent for human subjects research.


Because their lungs often are poorly developed, extremely premature babies routinely require three forms of treatment: 1) oxygen; 2) the drug surfactant, which helps the lungs stay open and is administered through a tube place in the trachea (windpipe; the main airway leading to the lungs); and 3) mechanical ventilation (artificial breathing machine). These treatments have been shown to be life-saving.

As part of routine care, the amount of oxygen given to such babies is adjusted based on a measure of blood oxygen level called oxygen saturation, which can be continuously checked with a probe placed on the skin. Routine care of premature infants typically involves giving enough oxygen to maintain oxygen saturation levels between 85 and 95 percent. The exact oxygen target level for a particular infant at any particular time is based on the baby’s individual medical needs.

The SUPPORT study involved two simultaneous, very complex experiments using a combination of procedures that were not consistent with “standard of care” for any premature baby.

In one experiment, the babies were randomly divided (similar to the flip of a coin) into two groups, each receiving a different treatment to assist breathing. One group received the usual standard of care procedures that had been shown to be life-saving: surfactant and mechanical ventilation. The other group received experimental interventions: use of a face mask to assist breathing and aggressive attempts to avoid using surfactant and mechanical ventilation. In this experiment, the researchers primarily wanted to see which group of the babies were more likely to die or develop lung disease.

For the second experiment — which has been the focus of most of the attention during this controversy — babies in both mechanical-ventilation and face mask groups were further randomly divided into a low-oxygen group and a high-oxygen group (the oxygen experiment). In contrast to usual care, the researchers tried to maintain oxygen saturation levels in a restricted low target range (85 percent to 89 percent) for babies in the low-oxygen group, and in a high target range (91 percent to 95 percent) for the high-oxygen group, rather than adjust each baby’s oxygen levels within the broader range of 85 percent to 95 percent — identified as the “normal” range by the investigators — to meet the baby’s individual needs, as would have been the case had the baby not been in the study. Of note, in the protocol for the study, but not in the consent forms the researchers characterized the high target range as being “more conventional.”

In the oxygen experiment, the researchers primarily wanted to see whether the babies were more likely to die or suffer retinal damage, which can lead to blindness, if their oxygen levels were targeted to the low or high range. It has long been known that, in general, giving high levels of oxygen to premature infants can lead to retinal damage, blindness and lung injury, while giving too little oxygen can cause brain damage and death.

For the oxygen experiment, the oxygen monitors used by doctors were deliberately miscalibrated to display either falsely low or falsely high reading when the babies’ actual oxygen levels were between 85 percent and 95 percent. This was done so that medical personnel caring for the babies would not know to which oxygen group the babies were assigned. Because oxygen levels are one important factor used to make critical clinical decisions about premature babies – such as when to start or stop mechanical ventilation – the use of miscalibrated oxygen monitors represented a remarkable, unequivocal  deviation from standard of care and exposed the babies to significant risks of harm. 

As documented in the March 2013, letter from OHRP, the consent forms for the study used at all study sites did not disclose to the babies’ parents the risk of death, and most did not disclose any of the other risks of the experimental oxygen management interventions, including risks of severe retinal damage, possible blindness, and brain injury. Further analysis by Public Citizen revealed that the consent forms also failed to address several other critically important pieces of information, including the following:

  • That one purpose of the research was to determine whether extremely premature infants were more likely to die if treated with the higher versus lower oxygen target;
  • The serious risks related to use of the intentionally miscalibrated pulse oximeters; and
  • An explanation of how the infants would be treated if they weren’t in the study compared to how they would be treated by participating in the study.

“The word ‘unethical’ doesn’t even begin to describe the egregious and shocking deficiencies in the informed-consent process for this study,” said Dr. Michael Carome, director of Public Citizen’s Health Research Group. “The failure to disclose such critically important information about the research undoubtedly directly affected parents’ decisions to enroll their extremely premature infants in this study. It is highly likely that had they been appropriately informed about the nature of the research and its risks, many, if not most, parents would not have allowed their babies to be in this study.”

Carome would know; for 10 years prior to joining Public Citizen, Carome was on the OHRP staff, serving most recently as its associate director for regulatory affairs and previously as its director of the Division of Compliance Oversight.

Not surprisingly, the study found differences between the two groups of infants in rates of death and retinal damage. In the lower oxygen group, 41 babies (nine percent of those evaluated) developed severe retinal problems and 130 (20 percent) died. In the higher oxygen group, 91 babies (18 percent of those evaluated) developed retinal problems and 107 (16 percent) died.

Because there was no control group that received the usual care with individualized oxygen management using oxygen monitors that displayed accurate readings, and because the babies were so sick to begin with, it is impossible to determine how these rates of death and retinal damage compare to what otherwise would have occurred had  a simultaneously randomized group the babies been managed according to usual clinical care based solely on their individual needs.

As a result of the informed-consent deficiencies, the investigators of the SUPPORT study failed to obtain the legally effective informed consent from the babies’ parents, and the conduct of the study was highly unethical, Public Citizen maintains.

Public Citizen has called on HHS to:

  • Direct OHRP to require each institution that participated in the SUPPORT trial to submit corrective action for avoiding such serious deficiencies in the consent process for future research;
  • Launch an independent investigation into how the HHS system for review and oversight of clinical trials failed so miserably. This should include an assessment of all entities within NIH or other HHS agencies that played a role in the review, approval and funding of the study.
  • Determine what system-wide actions are necessary to prevent similar future failures; and
  • Identify and suspend any similarly unethical ongoing research involving premature infants that is funded by NIH or any other HHS agency.

More information

Notice of Aug. 28 meeting at HHS in D.C.: https://www.citizen.org/sites/default/files/2013-15160.pdf

June 27, 2013 – Bioethics and Medical Experts Join Chorus of Criticism of Inadequate Consent Procedures in NIH Baby Study


June 13, 2013 – NIH Censoring Expert to Prevent Criticism of Ethically Questionable NIH-Funded Study on Premature Infants


June 5, 2013 – The Office for Human Research Protections Reaffirms Main Findings of Inadequate Informed Consent for Premature Baby Study, But Bows to Political Pressure and Puts Enforcement Actions Hold


May 8, 2013 – Public Citizen Report to Secretary of HHS: New Information About Baby Trial Shows It Was Even More Unethical Than Originally Thought


April 10, 2013 – HHS-Funded Experiment Exposed Babies to Risk of Death and Blindness Without Informing Parents


April 11, 2013 – Letter to Dr. Richard Marchase at the University of Alabama at Birmingham Criticizing His Misleading Statements to the Media Regarding the Unethical SUPPORT Study


April 15, 2013 – Public Citizen to Secretary of HHS: Suspend Enrollment in Baby Trials, Release Details of Ongoing and Earlier Experiments


May 8, 2013 – Public Citizen Report to Secretary of HHS: New Information About Baby Trial Shows It Was Even More Unethical Than Originally Thought


May 10, 2013 – Public Citizen Commentary on the SUPPORT Study Involving Premature Babies


Public Citizen’s petition to HHS Secretary, demanding apology: http://action.citizen.org/p/dia/action/public/?action_KEY=12172