Sorry, there are no polls available at the moment.

Romanian Orphans Study: A Bioethicist Responds to Ethical Concerns

Last month, Joseph J. Fins published a commentary on this blog criticizing the ethics of the Bucharest Early Intervention Project (BEIP)–a randomized, controlled trial of Romanian children who had been in orphanages, comparing those who were placed in foster care with those who remained institutionalized. The principal investigators of this project have already published a comprehensive reply. Here, as one of the bioethicists Fins criticizes, I add a couple of points to their response.

Fins claims that the children in the control arm of the study– who remained in the orphanages– were wrongfully treated. He is “concerned that children had been exploited, even sacrificed, for a scientific end.” In response to a commentary by me and Ezekiel J. Emanuel, he asserts that it is “inconceivable” that anyone could write, as we did, that “study participation was unlikely to cause net harm to the children.” Rather, Fins states, “[T]he 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.”

Were the trial participants harmed? It is true that institutionalized children in Romania were harmed by their institutionalization. No one is denying that. But the important question here is whether the children were harmed by being in research. The children had all been placed in institutions by Romanian government authorities prior to the study. Moreover, the investigators did not intervene to preventanychild from being removed from institutional care. The minimal risk study procedures used to measure their physical, cognitive, and social development did not lead to harms, and, indeed, likely benefited at least some of the children. Thus, none of the children in the control arm were harmed by their study participation. The children in the intervention arm, given their better developmental outcomes, do not appear to have been harmed by their placement in foster care. So, the study did not harm any participants.

It might be that Fins did not know some of the details of the study design. He writes, for example, that although the researchers “did not create the institutions, they should not have perpetuated them for the sanctity of study design.” As noted above, the investigators did not place any children in institutions; nor did they interfere with any participants in the control group being transferred out.

Independent of the question of the investigators causing harm, Fins is concerned that they did not prevent harms that they were in a position to observe. This would explain his reference to the “duty to rescue.” The duty to rescue applies in cases when someone can prevent a great harm at a low cost to herself. Assume these conditions were met in this case. Assume further, contrary to fact, that the BEIP investigators could have spent the funding used for the BEIP as they saw fit. Their research study directly benefited the 69 children placed in the foster care they designed. Indirectly, the knowledge gained from the study appears to have been beneficial in building evidence for the case against institutionalization in Romania and around the world and thereby changing child care policy. Would more children have been rescued overall if the researchers had been able to devote the entire research budget to foster care, so that more children would have been fostered by the BEIP, but no data collected? It seems unlikely.

Consider, next, the charge of exploitation. Did the investigators take unfair advantage of the situation of the participants or wrongfully use them as means to an end? The children in the study contributed to knowledge that will primarily benefit people like themselves. If there is any end for which it would be acceptable to enroll them in research, that is surely one. At most, there remains a question of whether participants in trials like these deserve to receive further benefits because of their contribution to generalizable knowledge. Fins does not answer this question.

Finally, Fins contends that the same social benefits could have been realized through alternative study designs that would not have involved an institutionalized control arm. However, he accepts that his judgment is post hoc (given what we now know are the results of the study, we can, of course, predict the children’s developmental outcomes). Moreover, he acknowledges that he has not read the scientific protocols and appears at times to be unclear on the aims of the BEIP (which, as clarified by Fox et al. in their response to Fins here and elsewhere,  was primarily concerned with whether foster care “could remediate the deficits incurred by previous early experiences”).

What set of study designs would have yielded valid data is a matter that must be judged by scientists who are experts in child development and according to the state of knowledge at the time the study was carried out. The correct answer to this question may be disputed. However, the BEIP investigators did not have any suspect motives for carrying out a randomized controlled trial rather than using some other study design. They wanted clear data that would be compelling to Romanian policy makers and designed their study on that basis. Why, then, suppose that they chose a study design that would expose subjects to unnecessary risks? The most plausible answer is that they did not.

I will close with a remark about the use of analogy. Fins compares the BEIP with the United States Public Health Service studies on venereal diseases in Guatemala and Tuskegee, and even invokes the Holocaust. Like “their predecessors in Guatemala,” he writes, “those who investigated the development of the Romanian orphans did something that was ‘ethically impossible.’” Likening himself to Hannah Arendt, Fins explains that he is speaking up because “the Holocaust happened not just because the bad guys did bad things but because the good people did nothing.” I find it hard to identify the relevant points of similarity between the BEIP and these other acts. To reiterate: the BEIP caused no study participants harm, provided significant benefits to at least some participants, led to socially valuable knowledge to help other institutionalized children, and received painstaking scientific and ethical scrutiny from a wide range of experts.

The purpose of an analogy is to illuminate. Given the lack of similarity here, Fins’s analogies do the opposite. They simply associate the BEIP with evil. This is problematic for wider reasons than just the rights and wrongs of a single study. First, they trivialize the wrongs committed against the subjects in these egregious studies and the victims of Nazi Germany. Second, no researcher wants to be compared with the Nazis, or have her work labelled as a new Tuskegee. Unfounded or exaggerated criticism of carefully designed and rigorously scrutinized studies with vulnerable populations risks having a chilling effect, deterring the very research that is needed to help such populations.

As a professor of medical ethics, a Hastings Center Fellow, and outgoing president of the American Society for Bioethics and Humanities, Fins has great authority on bioethical issues. It is particularly important, therefore, that what he writes is meticulously checked for factual accuracy, charitably interprets those he criticizes, and is not needlessly inflammatory.

Joseph Millum, PhD, is a staff scientist at the Clinical Center Department of Bioethics and the Fogarty International Center of the National Institutes of Health.The views expressed are the author’s own. They do not represent the position or policy of the National Institutes of Health, U.S. Public Health Service, or the Department of Health and Human Services.


Posted by Susan Gilbert at 11/22/2013 03:53:49 PM

Published on: November 22, 2013
Published in: Clinical Trials and Human Subjects Research, Professional Ethics

Receive Forum Updates

Recent Content