Recently I had a Susan Reverby moment. Reverby is the Wellesley historian best known for unearthing the revelations of the Guatemalan syphilis and gonorrhea studies conducted by the United States Public Health Service and the Pan American Health Organization in the late 1940s. As the now famous story goes, Reverby was doing archival work on the Tuskegee Syphilis Study at the University of Pittsburgh. While going through the papers of Dr. John C. Cutler, a lead investigator of the Tuskegee study, she encountered “Guatemala . . . inoculation . . . ” and immediately worried that a dark tale in the history of medicine had grown darker.
What had been, in Tuskegee, a transgression of omission–the withholding of penicillin from men who had syphilis–had now become one of commission and inoculation. She was to learn that investigators had deliberately infected individuals with venereal disease to test the hypothesis that penicillin could prevent gonorrhea and syphilis. Her discovery led to universal outrage, formal apologies by Secretaries Clinton and Sibelius to their Guatemalan counterparts, and the publication of Ethically Impossible: STD Research in Guatemala from 1946 to 1948 by the Presidential Commission for the Study of Bioethics. (1)
Like Reverby’s, my revelation occurred while conducting a routine task for an academic. I was attending a Canadian conference on developmental disabilities and neuroscience convened by NeuroDevNet in Vancouver. I sit on its scientific advisory board and was to give a talk later that day.
As I settled in for the first session, I did not expect to be jarred by the morning plenary delivered by Dr. Nathan Fox, a professor at the University of Maryland. He was speaking about the Romanian orphan studies, known as the Bucharest Early Intervention Project (BEIP).(2) It seemed to be a wonderful talk, and I was taking notes assiduously.
All was well until Fox started to talk about the details of the study, which I knew about in only a vague way. I knew these children had been institutionalized at a very early age under the Ceausescu regime and that their development had suffered because of profound sensory deprivation and a paucity of human contact.
As I heard more about the later scientific studies conducted by Fox and his colleagues, I became concerned that children had been exploited, even sacrificed, for a scientific end. By the time Fox and colleagues arrived in Bucharest, the regime had fallen–Ceausescu and his wife had been executed by firing squad for mass murder–but the orphanages remained.
Fox showed a postregime video of a bare environment with little toddlers and older children walking around without a purpose, with blank stares. One little girl tries to hug the leg of a housekeeper charged with cleaning up. The child is shooed away, her quest for a sliver of attention denied. I couldn’t help but think that my child gets more affection in a nanosecond than these little ones did in their lifetimes. It was heart wrenching.
As I watched the video, I hoped for something of a happy ending. Here the Western researchers are going in. It would be logical to expect humanitarian relief and a dollop of compassion. I was not entirely disappointed. Later, I was to read that one of the project’s funders, the MacArthur Foundation, created a separate funding stream to set up the Institute of Child Development, described as a “humanitarian institute” by journalist Virginia Hughes, who has written about the BEIP.
The investigators pursued a different approach. They were intrigued by what they saw and were seized, justifiably, by great scientific curiosity. So they created “buckets”– Fox’s word–into which they randomized the children. After screening orphans from many facilities, they randomized 136. Sixty-nine would be put into foster care to see how they would develop. (Because foster care was in such short supply in Bucharest, the investigators had to set these arrangements up with families.) Sixty-seven children stayed in orphanages, even though most of them weren’t even orphans. They had been given up by their parents under the social experiment of a coercive dictatorial regime. They had families to which they could return, but they had to stay institutionalized.
As Fox explained, where the children developed and matured made a difference. If children were taken out of the orphanages before they reached two years of age, they would progress pretty normally. After that there were really catastrophic outcomes for their neurodevelopment, and everything else.
As his lecture proceeded, I became increasingly uncomfortable, deeply worried that something terrible had happened and that no one had noticed. If what I heard was even partially true, this should be the subject of inquietude and inquiry. I have long ago stopped taking such expectations for granted. The possibility for worrisome actions seems endless and the tendency for passivity in the face of these transgressions too real.
I had been thinking about such questions after having just left a conference at Wesleyan University on the fiftieth anniversary of the publication of Hannah Arendt’s Eichmann in Jerusalem and to commemorate the death of her protégé and biographer, the philosopher, Elisabeth Young-Bruehl. Arendt was famously known for speaking up and pointing out the passivity and quiet of others. She said the Holocaust happened not just because the bad guys did bad things but because the good people did nothing. As she put it in her last book, Responsibility and Judgment, it’s not what the Nazis did that disturbs us; it’s what our friends didn’t do. In her view, understanding the breakdown of personal judgment was the key to appreciating what had happened in Germany.
So, inspired by the Arendt conference and with a long-standing respect for her work since being introduced to her by my friend and former teacher, Young-Bruehl, I asked Professor Fox about his work and its moral and ethical implications. In his desire for the methodological rigor of a randomized clinical trial, had he and his colleagues actually harmed children?
I wondered, were there not alternative methods of study available to achieve the same scholarly ends? Might it not be possible to have put a time limit on the duration of each child’s exposure to the orphanage and to follow everyone subsequently in foster care? Each child had been there for varying amounts of time, and one could have stopped the clock right there, taken all those children, followed them progressively, and done a regression analysis to see what variables contributed to the ultimate outcomes. You didn’t need a control group.
I questioned Fox about this, and he said that rigor was necessary. He noted that he would not have been able to have as much influence on the policies of the Romanian government if he and his colleagues had not produced studies with the requisite statistical standing. Romania now has no children under two years old in orphanages because of these findings.
The ultimate statistical significance of what the investigators had identified was so compelling that I think they would have gotten the same results with a regression analysis. Of course this is a post hoc argument, but there was a real time argument that should have prevented the design of their studies: long before the data was collected, a key outcome was knowable. Even as the study began, the investigators were not in a state of equipoise. Although they did not know that taking the children out of the orphanages was going to be beneficial, it was obvious that keeping them there was harmful. It was fundamentally inhumane.
So while the investigators did not know if they could reverse the environment’s developmental effects, it was clear from the start that the environment housing these children was neither the norm nor normative. After all, that was why it was the object of study. They weren’t studying a benign context; they were studying a toxic one.
One of the salient lessons of twentieth-century bioethics is that scientists cannot always do the experiment they would like to do. When you are not in a lab and unable to control all the variables, if you try to control all the variables, people can get hurt. That is what happened in Romania. And it is a double tragedy because investigators could have had the same policy impact if they had done their research in a different way. They could have been more attentive to the fact that some of the children suffered harm from ongoing early exposure to the orphanages that could have been interrupted.
The question remains, how much rigor is worth a child’s life? Might the researchers have reached the same conclusion, in almost as compelling a way, with an observational and longitudinal study and gone without randomization? Fox contested my view that children were randomized. To paraphrase, he argued that he did not randomize children to an intervention. He maintained that for the children who remained in the orphanage, he simply allowed the status quo to continue.
This, to my mind, is either a rationalization or a profound misunderstanding of what he and his colleagues discovered in those orphanages. There was an intervention in place when they arrived, and it was the toxic context that they encountered. The intervention was Ceausescu’s social experimentation, a condition that needed to be interrupted and halted. There was a clear duty to rescue.
Fox defends his study by asserting that they had parental “consent” to keep their children in the institutions, multiple approvals by institutional review boards, and the endorsement of a 2007 Science commentary by Joseph Millum and Ezekiel J. Emanuel,(3) which accompanied the first major presentation of this work. Lead investigator Charles A. Nelson III and his colleagues have also written of the importance of maintaining due deference to Romanian local standards of care.
Each of these defenses responds inadequately to the perpetuation of harm on vulnerable little children. To return to the issue of parental consent, is it surprising that parents acculturated to authoritarian rule and deference to power would agree to allow the children they had given up to remain institutionalized? After all, they had been brainwashed by the regime into believing that the orphanages were good for their kids, and they must have been impressed by Western investigators who asked their permission to allow their children’s participation in the research. Invoking the Nuremberg Code’s language on voluntariness, did the sociopolitical context, power dynamics, and poverty compromise independent decisions by the orphans’ parents? Even that designation, orphans’ parents, suggests that these parents have been stripped of their parental authority and standing.
On the question of IRB approval, I have not had the opportunity to review these protocols or the review process in detail, but I do think that they should be subject to outside impartial review. While a regulatory standard might have been met in a pro forma manner, from an ethical point of view, these approvals seem lacking and a breach of the categorical imperative.
There are the dark shadows of Willowbrook and Tuskegee here. People, especially vulnerable innocents, are not a means to science. Science is a means to promote human goodness, human growth, human flourishing. It’s not the other way around. We are not here to help science–we are here to help humanity. What was done was a clear “disrespect of personhood” and in violation of the Belmont Report.
Millum and Emanuel disagree with my assessment of the question of equipoise. In a commentary in Science, they write, “[T]he welfare of institutionalized children depends on choosing correctly between further institutional care or switching to foster care. Prior data focused on adoption and did not directly address this comparison, had selection biases, and lacked a definitive randomized trial. Thus, the study appears to fulfill equipoise. Moreover, although both institutional and foster care can sometimes result in maltreatment, study participation was unlikely to cause net harm to the children; no child was put at additional risk to obtain the results, which reduces the ethical reasons for worrying about equipoise.”
To be blunt, I am worried both about the logic of Millum and Emanuel and about their understanding of equipoise in this context. Certainly a comparison between institutional care and foster care would be reasonable if both met minimal standards and were not the products of a brutal regime. The institutions in Romania were never at a minimal standard of care and thus were never worthy of a comparison. And, yes, while both foster and institutional care can result in maltreatment, it is inconceivable that anyone could write that “study participation was unlikely to cause net harm to the children.” Indeed, the very study under question shows that children who were not removed from these barren places before age two sustained irreparable damage. The youngest of children in the study could have been rescued. They were not, and they were therefore harmed.
Finally, there is the question of contextualizing the study in light of existing standards of care. Nelson et al. wrote in Science, “Additionally, given that the study was invited by the Romanian authorities and conducted there, with the aim of guiding child welfare policy in Romania, it made sense to assess the study in view of the local standard of care, which was institutional care. The study also presented no more than minimal risk to the subjects; specifically, children assigned to the IG (institutional group) continued to receive the same care as if the study had not been conducted . . .”
I find this justification highly problematic. How could the same investigators who found a lifetime developmental risk from institutionalization claim that the study presented no more than minimal risk? The investigators will say that the toxicity of the environment was proven only by randomization and thus that maintaining this “standard of care” was not a known increment over the usual risk. But I would counter that what they sought to prove was already known. A contemporaneous editorial in Nature Neuroscience (4) commenting on this study trenchantly made the point, “Unsurprisingly, cognitive development in children who remained institutionalized was markedly lower than that of never-institutionalized children or of children who were removed from institutions and placed into foster care.” Thus, by the standards of the day, and like their predecessors in Guatemala, those who investigated the development of the Romanian orphans did something that was “ethically impossible.”
I do not mean to recapitulate what was addressed in Science six years ago. But in light of Henry Richardson’s brilliant work on the ancillary-care obligations of investigators (see his book Moral Entanglements: The Ancillary-Care Obligations of Medical Researchers), I do think these issues need to be reconsidered. Under an ancillary-care obligation, I believe that the researchers had an obligation, at the very least, not to utilize a toxic environment in pursuit of data. Although they did not create the institutions, they should not have perpetuated them for the sanctity of study design. And if they were unsure about that toxicity, and confused the orphanages with conventional institutional care, as Millum and Emanuel seem to do, then the investigators should have had the benefit of an independent data safety monitoring board (DSMB) to monitor the consequences of ongoing institutionalization and deinstitutionalization on childhood development. Properly assembled, with clinical and research personnel attuned to the signs of developmental delay and pediatric neuropsychiatric conditions, a DSMB would have likely called a prompt halt to the study.
Fox mentioned that he and his co-investigators are writing a book about their work. I will look forward to learning more about how the investigators considered their ethical obligations. Perhaps I will be persuaded by their arguments. I am open to the evidence and contrarian points of view, which I have sought to represent in this essay. But until then, I am convinced that I was right in asking that first question and voicing a strong objection to what happened, making ethical concerns explicit, as Arendt would have urged.
My point is a simple one: protecting the best interests of the child, a point forcefully and persuasively made by Elisabeth Young-Bruehl in her last public lecture weeks before her death in late 2011. Speaking to the New York Institute for the Humanities at NYU’s Deutsches Haus, Young-Bruehl spoke of our collective ethical mandate:
It is the great task of human beings–the essential task–to understand what adults should give children; what is–to use a legal phrase–“in the best interests of the child.” The basic needs of all children are the same; there are universal needs. And it should be the task of any and all adults to understand those needs and meet them. Children depend upon adults for this understanding, and if it is not applied, not translated into the actions of child-rearing and education, children cannot grow and develop freely and become adults who, in turn, give such understanding and action to their own children.(5)
Young-Bruehl’s charge makes the work in Romania all the more challenging, for no one doubts that Fox, Nelson, and their colleagues, their funders, or commentators on the BEIP had anything but the best interests of children in mind when the team undertook their research. In the aggregate, they have been advocates for children and for reform of policies in Romania that relied upon institutionalization of child care as a means of social control. Their work has debunked the official notions of the regime that children entered the orphanages already damaged, demonstrating convincingly that nurture, not nature, was the culprit for developmental delay. Their work has contributed to what we know about human development. It will likely help us better understand autism and autism spectrum disorders and conditions like fetal alcohol syndrome.
And this is precisely why it is so difficult to be critical of what they have done. Despite pristine motives, a desire to advance study and to promote improvements in childhood learning and development, their actions raise the significant question of whether Young-Bruehl’s essential task, to understand “what adults should give children” and what is “in the best interests of the child,” can be understood in the plural or just the singular? That is, can it be exemplified in a utilitarian fashion, as would justify the orphan study, or would damage to a single child serve as a moral condemnation of the entire enterprise?
Young-Bruehl’s admonition is ambiguous. She writes of both children and the child, placing this judgment upon us–and we should not shirk from declaring our views. I remain deeply haunted by the vision of the children wandering about, dependent upon adults for their care and safe-keeping. Without that vision in mind, without the proximity of each little face looking for love, for guidance, it is all too easy to fall prey to moral blindness and justify what was done with good intentions. If this argumentation is inconsistent with randomization, then the method should be sacrificed, not the most vulnerable amongst us.
Acknowledgements: Dr. Fins acknowledges the partial support of a CTSC grant to Weill Cornell Medical College, # 1UL1 RR024996.
- Presidential Commission for the Study of Bioethical Issues, “Ethically Impossible”: STD Research in Guatemala 1946-1948 (PCSBI: Washington, D.C., 2011).
- C. A. Nelson III, C. H. Zeanah, N. A. Fox, P. J. Marshall, A. T. Smyke, D. Guthrie, “Cognitive Recovery in Socially Deprived Young Children: The Bucharest Early Intervention Project,” Science 318, no. 5858 (2007):1937-40. http://www.ncbi.nlm.nih.gov/pubmed/18096809
- Millum J and Emanuel EJ. The ethics of international research with abandoned children. Science 2007;318:1874-1875. http://www.ncbi.nlm.nih.gov/pubmed/18096792
- The ethical neuroscientist. Nature Neuroscience 2008; 11(3):239. http://www.nature.com/neuro/journal/v11/n3/full/nn0308-239.html
- Young-Bruehl E. Untitled Talk. New York Institute for the Humanities, Deutches Haus, New York University. November 18, 2011. See also: Young-Bruehl E. Childism: Confronting Prejudice Against Children. New Haven: Yale University Press, 2012.
Joseph J. Fins, M.D., M.A.C.P., is The E. William Davis, Jr., M.D. Professor of Medical Ethics; Chief, Division of Medical Ethics; Professor of Medicine; Professor of Public Health; and Professor of Medicine in Psychiatry at Weill Medical College of Cornell University. He is a member of the board of directors of The Hastings Center, a Hastings Center Fellow, and chair of its Fellows Council.