Part 2 of 4: The foremost booster of the unethical SUPPORT experiment, whose lethal results were predictable from past restricted oxygen study results, is the editor-in-chief of the most influential medical journal.  

“Research that is not conducted in conformity with valid scientific principles cannot produce relevant results. Indeed, such results may be misleading and contribute directly, and indirectly, to harmful clinical practices. Exposure of human participants to the risks and burdens of such research is unethical.” [1]

He notes that the oxygen SUPPORT experiment is the most recent example demonstrating the faulty IRB approval process. Conflict of interest and lack of expertise has led too many IRBs to give the green light of approval to complex clinical trials without proper review.

“The design and ethical review of federally funded research is often undertaken by a homogeneous group of individuals with congruent interests at the same or similar academic institutions. Individuals from the public, advocacy groups and non-academic organizations are often excluded. When people from these groups publicly voice their concerns, their views are vilified in academic publications as impeding future advances in research, or even ignored.”   

He calls for “a far-reaching and comprehensive review of the way US government backed clinical research is funded and approved… The HHS, the NIH and universities must acknowledge that the current research-approval process is flawed and requires an urgent, comprehensive review that should include experts and leaders from outside the academic community.”[2]

Far from acknowledging that the process is flawed, requiring urgent, comprehensive review—in order to strengthen safeguards—the medical research establishment has circled the wagons, defiantly denouncing OHRP for carrying out its public responsibility.  The first round of the SUPPORT defense, following public criticism, was orchestrated by Dr. Jeffrey Drazen, the editor of the NEJM, who led off with an editorial[3] in which he inaugurated the SUPPORT defense campaign: denying the evidence, or claiming no one knew—i.e., “strategic ignorance”[4]:

The OHRP investigation [of SUPPORT] has had the effect of damaging the reputation of the investigators and, even worse, casting a pall over the conduct of clinical research to answer important questions…We are dismayed by the response of the OHRP and consider the SUPPORT trial a model of how to make medical progress.”

“…there was no evidence to suggest an increased risk of death with oxygen levels in the lower end of a range…The study made clear that higher oxygen saturations within the then-recommended range increased the risk of retinopathy but decreased the risk of death…”

Each of Dr. Drazen’s claims is contradicted by the SUPPORT investigators’ report in the NEJM (2010):[5]  

“The results of this large randomized trial to test the effect of lower versus higher target ranges of oxygen saturation, in conjunction with the results of previous studies, add to the concern that oxygen restriction may increase the rate of death among preterm infants. The combined risk difference observed in the trials from the 1950s was an absolute increase in in-hospital mortality of 4.9% [ ] in the oxygen-restricted group, which is close to the absolute increase of 3.7% [ ] in the rate of death before discharge in the lower-oxygen-saturation group that was observed in the current study…”

“Our data suggest that there is one additional death for approximately every two cases of severe retinopathy that are prevented… At the present time, caution should be exercised regarding a strategy of targeting levels of oxygen saturation in the low range for preterm infants, since it may lead to increased mortality.”

Dr. Drazen’s strategic ignorance was accompanied by patently false, unsupportable declarations:

“These findings changed medical practice at many centers…The results of SUPPORT have been critical in informing treatment decisions for extremely preterm infants.”

However, SUPPORT researchers’ follow-up analysis of outcome and demographics published in Pediatrics (2010)[6] and (2012)[7] found “significant differences in outcome” and demographic differences between the mothers and babies in the trial and those eligible who were not enrolled.

They attribute the outcome differences to “enrollment bias”: “Our analysis suggested that underprivileged subjects were less likely to participate.” Wade Rich, Neil Finer et al acknowledge the “lack of generalizability seen in our results”⁷ which they blame on the need for informed consent– even as the SUPPORT documents prove that every ethical and legal requirement for informed consent had been violated.

The accelerated erosion of bedrock medical ethics represents an ominous downward spiral facilitated by a cadre of academic bioethicists who serve as the “accommodating handmaidens”[8] for the biomedical industry, government and academic medical centers—or whoever pays their keep. They provide “ethical permission slips”[9] for research that crosses the moral and legal line—which explains why both the biomedical industry and NIH pours millions of dollars into bioethics centers and bioethics commissions.

Indeed, Nobel Laureate scientist, James Watson, (an avowed eugenicist) who headed the Human Genome Project, assured a panel at an economic conference in Davos, Switzerland, that “nobody should worry about the morality of what [the Genome Project] was doing because “the project had allocated millions of dollars to get the best ethicists that money can buy.”[8]

Dr. Drazen’s declaration—“We consider the SUPPORT trial a model of how to make medical progress” —and his equally absurd, promotional claims were accompanied by a series of letters and Opinion pieces by SUPPORT study group participants, and by prominent representatives of the bioethics establishment, culminating in a letter co-signed by 47 pediatricians and bioethicists,[10] “the best that money can buy.”

These bioethicists identify themselves as “a group of scholars and leaders in bioethics and pediatrics.” The letter was paid for with a grant from NIH[11] and was published in the same issue of the NEJM (June 5) as the NIH “Support of SUPPORT.”  The NEJM is the epicenter of high impact research reports whose unethical methods and claimed findings are violating legal requirements and pushing medical ethics down the slippery slope where babies, children, disabled persons, and eventually all hospitalized patients will be enrolled into medical experiments without their knowledge or voluntary, informed consent.

“In Support of SUPPORT—A View from the NIH,”[12]
published in the NEJM (June 5), is part of an organized effort by the major co-dependent stakeholders to provide institutional shields for unethical, deviant medical research practices. Kathy Hudson, Deputy Director for Science, Outreach, and Policy, NIH; Dr. Alan Guttmacher, Director of the Eunice Kennedy Shriver Institute of National Institute of Child and Human Development; and Dr. Francis Collins, Director of the NIH—the public officials who head the agencies that had approved and funded the SUPPORT experiment—make numerous false claims, such as:

“The SUPPORT researchers and institutional review boards (IRBs), practicing clinicians, and the AAP had no scientific evidence to expect a difference in mortality between the two treatment groups.”

On the contrary, they had ample evidence from multiple studies focusing on oxygen management in premature infant; the scientific literature spanning 60 years confirms that that restricting oxygen from premature babies increases the risk of death.[13]

The largest randomized trial testing the safety of restricting oxygen levels in premature babies, in an effort to prevent retinopathy was the NIH-sponsored Cooperative Study conducted at 18 medical centers (1953-1954).  In that flawed but highly influential experiment, oxygen was withheld from all but 52 premature babies born at those 18 centers during the first crucial 48 hours after birth[14]—resulting in the death of 634 (45%) babies.

But their deaths were not included in the study’s reported findings; [15] only the outcome of 786 infants who had survived 48 hours and who were then randomized to receive restricted oxygen or standard care was reported. The study findings were misinterpreted; the conclusion drawn by the neonatal establishment failed to consider that babies in one arm of the experiment received more oxygen, and received it for a much longer time;[16] and the fact that some babies in the restricted oxygen arm still developed eye damage. Indeed, 33% developed ROP, 6% developed severe (stage 3 to 5) ROP, and 2% became blind[17] in the restricted oxygen group. Nevertheless, the Cooperative study was promoted as having indicted oxygen as the sole cause of retinopathy.

The misleading claimed results of the Cooperative trial were uncritically accepted, and widely broadcast as having eliminated retinopathy. 

That claimed finding was highly politicized, and led to an expansive increase in Congressional appropriation for medical research;[18] but worst of all, that claim was the catalyst for a disastrous nationwide policy restricting the oxygen supplementation level to <40% for premature babies. Though the study’s methodology and the Draconian public health policy restricting the level of oxygen supplementation for premature babies have since been discredited, [19] the policy was in effect for more than 25 years; the extraordinary death toll resulting from that policy has been estimated at a ratio of 16 deaths for 1 case of prevented ROP.[20] And it was cited in the published SUPPORT report [5]   demonstrating that the risk of death from restricted oxygen had been well known to the SUPPORT researchers prior to their trial.

Given the extensive body of evidence demonstrating a foreseeable risk of death for premature babies whose oxygen needs are not met by a restrictive oxygen protocol, the specious claim made by Dr. Collins and Dr. Guttmacher that the SUPPORT researchers, the IRBs and NIH reviewers—“had no scientific evidence to expect a difference in mortality between the two treatment groups”[13] suggests that they had all failed to perform due diligence to prevent causing harm—as would have been  their professional responsibility.

“The Albatross of Neonatal Medicine” 

Much can be gained from reading “Retrolental Fibroplasia: A Modern Parable ” (1980) [21] and “A Cautionary Tale About Supplementary Oxygen: the Albatross of Neonatal Medicine,” (2004) [22] by Dr. William Silverman, one of the most prominent neonatal researchers whose strong advocacy for that 1950s NIH trial led to its approval, who reassessed the trial and its consequences decades later. In a seldom cited article, “Oxygen Dogma Challenged” in the Archives of Disease in Childhood (1982) [23] ,

Dr. Silverman called those 25 years of an uncritical acceptance of oxygen as the sole cause of ROP, “dogmatic slumber,” stating, “It was some time before a balance was struck [  ] whereby premature infants who need extra oxygen to survive without brain damage receive it, but in concentrations that do not seem to be giving rise to blindness.” Dr. Silverman further acknowledged that the theory blaming oxygen as the single cause for premature infant blindness is largely in doubt: “The watertight case in support of ‘hyperoxia is the cause of ROP’ has been leaking for years; the leaks are now so large they can no longer be dismissed.” [24] He called for other avenues of research to seek better resolutions.

In “Lessons Learned from Randomized Trials Involving Newborn Infants,” Dr. Silverman stressed the importance of using concurrent control arms: “The precaution can always bring about a substantial reduction in the number of patients maimed and killed as the result of inevitable surprises!”[25]  Dr. Silverman died the year that the SUPPORT experiment began enrolling its tiny, unprotected subjects.

Rather than learn from the mistakes of the past ill-advised attempts to reduce the risk of ROP by withholding needed oxygen from premature infants whose lives were sacrificed, the SUPPORT experiment sought to compare two protocol-dictated opposite supplemental oxygen levels without the safety precaution of a current practice control arm. The risks for extremely vulnerable babies in this flawed experimental design were further increased with the use of miscalibrated pulse oximeters used to monitor the babies’ oxygen saturation rates.

It is astonishing that the highest ranking NIH officials would misrepresent a published report claiming that:  “more recent studies showed no increased risk of death or neurodevelopmental impairment at saturation levels as low as 70%“, citing a report by Tin et al (2001) as the source. But that claim is contradicted by the actual published report;[26]  clearly states:

 “Staff always aimed to maintain saturation in the top half of the target range (particularly when the lower limit of this range was less than 85%). No formal attempt was made to document how often saturation fell outside the recommended management limits, but review of a random sample of case notes quickly showed that narrowly set limits were broached much more often than wider limits…”

Thus, the trial had a theoretical target range of 70 to 90% oxygen saturation, but according to the authors, it is unlikely that the staff ever allowed a baby’s blood oxygen saturation level to reach a low 70%. With an annual budget of more than $30 billion, one would expect that the Director of the National Institutes of Health and his deputies would at least be able to accurately describe the published findings of a study instead of spinning them to their defensive agenda. These officials acknowledge that: “a major public debate has arisen regarding [the SUPPORT] study …And the ramifications go well beyond this one study the outcome of this debate could affect how we conduct and communicate about critical research on interventions that are within the standard of care for all diseases and conditions…” [12]

These medical “authorities”–Dr. Collins and Dr. Guttmacher— who authorize billions of dollars in research grants need a crash course in ethical and legal parameters of research beginning with the definition of research in the Code of Federal Regulations:

“Research means a systematic investigation, including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge.” [45 CFR46.102(d)]

In standard practice in intensive care neonatal units, treating physicians assess each baby’s physical condition, and carefully titrate oxygen levels within an unrestricted, full range of oxygen saturation levels—85%–95%–according to each individual baby’s needs—not according to a predetermined restricted dose. In standard care, physicians’ judgments are guided by an accurate pulse oximeter that monitors each baby’s arterial blood oxygen saturation levels, the condition of the baby’s lungs, and their lungs’ ability to extract oxygen from the air they breathe.

Physicians treating premature infants try to keep the blood oxygen saturation at greater than 92% and they try to get the oxygen concentration to room air (or 21%) where they believe to be no risk of death or blindness. Adjustments are made according to the babies’ ability to extract oxygen from the air they breathe and other physiologic data, weighing the risks versus benefits of the oxygen level the physicians set for the baby. But this precautionary standard of care is not the standard that the babies in the SUPPORT experiment received.

In their editorial in the NEJM, Experienced practicing neonatologists, Dr. Richard Polin and Dr. David Bateman,[27] NY-Presbyterian Columbia University, caution against efforts to pinpoint an ideal saturation range:

“the saturation ranges that are associated with morbidity or mortality are not synonymous with pinpointing safe and unsafe saturation levels, because the ideal saturation range may differ among infants at various gestational ages… Furthermore, maintaining an infant in a given saturation range can be difficult and does not guarantee an optimal outcome.”

The SUPPORT experiment was not designed to reduce risks for premature babies; it was designed to compare the rate of death vs. the rate of ROP by increasing risks in violation of Federal requirements.
In a follow-up OHRP letter (June 4^th) [28] to the University of Alabama, OHRP notes that

“every child in the SUPPORT trial experienced some change …for at least some children participating in the SUPPORT trial, the effect of such participation was to specifically increase the likelihood of being assigned to oxygen levels…they would not have been assigned by their individual physician, had they not been in the study.”

The findings reported by SUPPORT investigators merely confirmed what was already known:

“Like the meta-analysis and most non-randomized studies, our trial confirmed that lower target ranges of oxygenation result in a large reduction in the incidence of severe retinopathy among survivors…there is one additional death for approximately every two cases of severe retinopathy that are prevented.”⁶

• What logic do the Directors of NIH and NICHD use when they describe the SUPPORT experiment as “interventions that are within the standard of care?”
• Their effort to deny reality-based evidence by resorting to circular reasoning that defies logic is astounding.
• If the oxygen SUPPORT experiment was not research, then under what authority did Drs. Guttmacher and Collins fund it as research?

“What’s in it for you?”
When asked by Todd Neale, MedPage Today[29] reporter, about their conflicts of interest, these high ranking NIH officials had the gall to claim they “had no conflicts of interest.”

Their response absolving themselves from their obvious conflict of interest—having approved and funded the SUPPORT experiment, they now insist on being the arbiters “of how the regulations should apply to the state of scientific understanding.” This demonstrates the arrogant, insulated mindset under which NIH leadership operates as if in a cocoon.

However, it is not within the purview of NIH to re-define research so as to help researchers and their institutions avoid having to comply with the Common Rule.  If Drs. Collins and Guttmacher, and their battalion of bought bioethicists were to succeed in redefining research (such as SUPPORT) so that the Federal protections and informed consent requirements do not apply, then they will be responsible for frightening away most potential volunteers and bring to a halt almost ALL research involving human subjects.

Indeed, in their article, “In Support of SUPPORT,”¹³ Drs. Hudson, Guttmacher and Collins acknowledge that:

“This controversy has alarmed some of the parents of infants who were in the study, confused the biomedical research community, and befuddled IRBs. Several other studies seeking new insights to improve care for these vulnerable infants have been put on hold as the field tries to understand the OHRP findings.”

If “the field” doesn’t understand the OHRP findings, they had better study the Common Rule requirements with greater attentiveness— taught by someone outside “the field.”

SUPPORT was one of a series of five oxygen supplementation experiments using a similar protocol under the NeOProM Collaboration protocol,[30] an umbrella for participating medical centers. Approximately 5,230 extremely premature babies were involved in the experiments According to the prospective NeOProM meta-analysis protocol, it “would have a 80% power to detect a 4% difference in the primary outcome: death or severe disability in survivors.”

The participants included the US-NICHD SUPPORT trial (involving 1,316 babies) conducted between 2005-2009; BOOST II conducted in Australia, New Zealand, and the UK between 2006-Dec. 2010 (involved 2,448 infants); and the Canadian-sponsored COT trial (involved 1,201 infants) conducted at 25 hospitals in 6 countries between 2006 and 2010.  Barbara Schmidt, the principal investigator for the COT trial, is from the University of Pennsylvania, one of three participating COT centers in the US, who has published articles claiming that newborn infants benefit from participation in randomized clinical trials,[31] a claim resoundingly contradicted by the preventable fatalities resulting from the experiments conducted under the NeOProM Collaboration protocol.[32]

The SUPPORT and the BOOST II trials were published in the NEJM,^{6 [33]} and the Canadian trial (COT) was published in the Journal of the American Medical Association (JAMA)[34]—the two journals with the greatest impact. As might have been expected, the results of the SUPPORT and BOOST II experiments confirmed that premature babies whose oxygen is restricted are at greater risk of dying during hospitalization than babies whose supplemental oxygen is less restricted. These study results differ only in the percent of babies killed:

• the SUPPORT trial report disclosed that prior to discharge from hospitals, 107 of 662 babies died (16.2%) in the high oxygen group; and 130 of 654 babies died (19.9%) in the low oxygen group (a 3.7% difference);
• the BOOST II trial reported “the rate of death was significantly higher in the lower-target group than in the higher-target group (23.1% vs. 15.9%).” This alarming 7.2% difference in the death rate resulted in the suspension of the study.³⁴ 
• The COT trial reported no significant difference in deaths or ROP between high and low oxygen groups—15.1% vs. 16.6%.³⁵

The COT investigators sought to explain the difference in their results and those of SUPPORT and BOOST II:  “Study limitations included lack of perfect adherence to the target saturation ranges and that caregivers may have tolerated saturations approaching the upper limit more than the lower limit.”

This suggests that they had expected the death rate in the low oxygen group to be higher. It also suggests that the nurses’ failure to adhere to the target saturation ranges (most likely) because they could not bring themselves to deny needed life-saving oxygen to the tiny babies in their care—is regarded as a “study limitation.”

The first randomized experiment comparing low and high oxygen supplementation in premature babies, was conducted in 1950 over the objections of the NIH pediatric grant reviewers who rejected the proposal for scientific and moral concerns: “these guys are going to kill a lot of babies by anoxia (inadequate oxygen) to test a wild idea…”²²

The NIH awarded the grant anyway, after the researchers promised “to keep the babies a healthy pink at all times.” However, the night nurses, who were convinced that the babies in the low oxygen group were going to be killed, turned on the oxygen at night for any baby who was not receiving enough oxygen.[35] The authors failed to disclose the mortality rate in their experiment—even though the suspected high death rate in the restricted oxygen group was the main reason for the NIH reviewers’ objection.  A reasonable assumption is that the death rate was alarming—why else would they have failed to disclose one of the two primary endpoints of the study?

The safety of the babies in these trials was subordinated to protocol-dictated procedures:

Failure to include a control arm of standard care in the SUPPORT, BOOST and COT trials undermined the ability to assess whether more babies were killed or seriously harmed by being randomly restricted to a protocol-dictated low or high oxygen intake. Risks were further increased because physicians charged with their care were misinformed about the babies’ blood saturation oxygen levels. The primary outcome that the SUPPORT experiment sought to measure was a composite of the combined risks of death and ROP which they did not find “statistically significant”: 28.3% in the high oxygen group and 32.1% in the low oxygen group. But the risk of ROP—which is treatable— does not equate with death or brain damage.

The flawed experimental of design of SUPPORT failed to meet the standard of external scientific validity:[1]

Lacking a standard of care control undermined whatever scientific value the study findings may have had. Indeed, the SUPPORT researchers acknowledged the “lack of generalizability seen in our results.”  So, the only lesson from these trials is that providing physicians with inaccurate pulse oximeters will have dire consequences; physicians will be fooled into prescribing more or less oxygen than they would in standard care. As a result, babies who received too little oxygen were exposed to increased risk of death and brain damage, and babies who received too much oxygen were at increased risk of ROP and the need for surgery.

The most detailed critical analysis of the SUPPORT protocol and 22 consent forms was issued by Michael Carome and Sid Wolfe for Public Citizen (May 8):[36]

“The combined experimental procedures of randomly assigning infants to interventions using narrow low or high oxygen saturation targets plus the use of miscalibrated pulse oximeters represented a clear departure from the standard of care that these critically ill infants would have received had they not been enrolled in the study. Yet, the majority of the consent forms included statements assuring subjects’ parents that all research procedures used in the study were ‘standard of care.'”

End of Part 2 of 4