By Renee Boss and Ruth Faden

(Updated June 5, 2013 – Originally posted April 25, 2013)


OHRP issued a new letter, dated June 4, 2013, to the University of Alabama at Birmingham (UAB), clarifying that that they do not and have “never questioned whether the design of the SUPPORT study was ethical”. In addition, all compliance actions against UAB have been put on hold, with no plans to take further action.  OHRP acknowledged their “obligation to provide clear guidance” on disclosure of risks in similar trials, where treatments fall within the range of standard care, and announced plans to develop guidance using a process that is as open as possible, with the usual notice and comment process and an open public meeting.

The New England Journal of Medicine (NEJM) published a letter to the editor, serving as an open letter to OHRP, authored by scholars and leaders in bioethics and pediatrics urging OHRP to “withdraw its notification to the institutions involved” in the SUPPORT trial that they failed to meet informed consent requirements regarding reasonable foreseeable risks.  The authors express their belief that the “this conclusion was a substantive error and will have adverse implications for future research.”  Authors include Berman Institute faculty members Ruth FadenRenee BossJeffrey KahnNancy KassJeremy SugarmanHolly Taylor, and Yoram Unguru.

NEJM also published a perspective authored by Kathy Hudson, Alan Guttmacher and Francis Collins, “In Support of SUPPORT — A View from the NIH“.



The New York Times editorial board recently excoriated a randomized clinical trial investigating optimal oxygen levels for premature babies, describing its consent form as “startling, and deplorable.” The advocacy group Public Citizen described the trial as “highly unethical.”  The ethical issues raised by the study, the Surfactant, Positive Pressure, and Oxygenation Randomized Trial, or SUPPORT, are far more subtle, far less dramatic, and much more complex than that editorial suggests. Here we consider two of them.

First, did the SUPPORT study place these extremely premature babies at additional risk, and if so, was that additional risk within ethically acceptable limits?  In this case, ‘additional risk,’ means risk above what could reasonably be expected from the care they would otherwise receive in the neonatal intensive care unit (NICU). In standard practice, NICU monitors that continuously analyze blood oxygen level are programmed to alarm each time an infant’s level falls outside 85%-95%, a range set by the American Academy of Pediatrics.

Some have suggested that the infants in the SUPPORT trial were deprived of individualized medical care, implying that goal oxygen levels in regular practice are uniquely tailored for each individual infant.  In routine medical care, oxygen levels are not uniquely tailored to individual infants, with certain exceptions, like those born with heart malformations.  Most infants are treated—as were all of the infants in the SUPPORT trial—to keep oxygen levels within the standard range.  It is true that SUPPORT study investigators had the hypothesis, based on prior research, that treating babies in the lower portion of the range might result in fewer cases of severe retinopathy, a potentially blinding eye disorder, and lower rates of death related to oxygen toxicity. But there were also reasons to be concerned that too little oxygen might be problematic for at least some babies.

It was this uncertainty about how best to adjust the oxygen range that prompted both the SUPPORT study, and also the clinical practice of keeping babies within the 85%-95% parameters.  As the SUPPORT study operated within this standard, the risk was ethically acceptable and arguably no greater than what the babies were otherwise facing.  Although some might argue it was reasonable at the time to suspect that the children in the higher range faced higher risk than if they received standard care in the full 85% to 95% range, it was equally reasonable at the time to view any such suspicion as being on tenuous grounds. And as it turned out, and this is commonplace in clinical research, the results of the study were just the opposite of that suspicion – babies in the higher range turned out to be at lower risk than babies in the lower range.

The second ethical concern is whether parents were adequately informed before consenting to the SUPPORT trial. It is standard practice in NICUs across the country not to ask parents’ permission before their babies are exposed to oxygen levels within the standard of care of 85-95%. The question then seems to be whether standards for consenting to participate in a study of standard medical care should be more stringent than standards for consenting to receive standard medical care.  This is a crucial question that must be answered in order to proceed with comparative effectiveness research, an area of inquiry that focuses on weighing the pros and cons of treatments that are already approved for use and being used for many patients.

The SUPPORT trial and the controversy surrounding it may not be the best example for analyzing this critical question, however.  What is at stake in an ICU is nothing short of life and death.  Desperate parents are often just learning what their babies, and they, are up against.  Involving parents in decisions affecting their babies’ prospects is very difficult to do and must be undertaken with the greatest care, in research and in clinical practice.  In such contexts, it is critical that there be a high degree of confidence that the research (or the clinical course of action outside of research) is ethical and the risks are acceptable.  In the case of the SUPPORT study, OHRP raised no questions about the ethically acceptability of the research design or its riskiness. OHRP’s censure was restricted to the consent form, which the agency found to inadequately describe “foreseeable risks and discomforts.”

Treatments in the intensive care unit always entail risk.  To treat life-threatening infections, we may need to use an antibiotic that could cause deafness.  Ventilators can save babies born too prematurely to breathe on their own, but over time may cause damage to lung tissue that could persist for years.  Oxygen can protect the developing brain, but may damage the developing eye.  ICU patients receive dozens of simultaneous therapies that have the potential for benefits and risks.  ICU clinicians are trained to perform the complex calculus to balance these risks and benefits in light of the fact that the patient may die without these therapies.

Careful review of the science and ethics of medical research is essential to the process of maximizing knowledge while minimizing risk, especially for particularly vulnerable populations.  Evidence-based medicine depends on information gained from large medical research studies.  Reports and editorials in the press that sensationalize and misrepresent the facts of medical research unfairly paint the entire field with a broad brush of suspicion and distrust.  The result is a disservice to us all.  Everyone needs medical care at some point; without well-designed research, the medical care we receive will be entirely subjective.  Clinicians make better decisions with results of large studies, carefully analyzed, to guide them.  Patients are more likely to survive.

renee_bossRenee Boss, MD is an assistant professor in the Division of Neonatal-Perinatal Medicine at Johns Hopkins Hospital and a faculty member of the Johns Hopkins Berman Institute of Bioethics. She is a member of the Johns Hopkins Hospital Ethics Committee.

RuthFadenSqRuth Faden, PhD, MPH is the Wagley Professor of Biomedical Ethics and Director of the Johns Hopkins Berman Institute of Bioethics.  She is the co-author of Social Justice: The Moral Foundations of Public Health and Health Policy.

4 people like this post.


Renée Boss
Ruth Faden

Tags: , , , , , , , , , , , , , , , ,

2 Responses to “Sensationalizing SUPPORT Harms Us All”

  1. […] Boss R, Faden R. Sensationalizing SUPPORT Harms Us All. Johns Hopkins Berman Institute of Bioethics … […]

  2. […] evaluate whether the balance between the benefits of research and the risks to vulnerable subjects is reasonable; and what comprises appropriate informed consent in randomized studies of standard of care. […]

Leave a Reply