Risk and Responsibility: Ethics, Grimes v Kennedy Krieger, and Public Health Research Involving Children

Abstract

The legal case of Grimes v. Kennedy Krieger Institute, Inc, has raised concerns in the public health research community regarding the acceptable level of risk in research involving children, parental authority for informed consent, and exploitation of research subjects for the benefit of public health. We provide an overview of the case and discuss the impact of the court's decision and its possible effect on future research protection policies and practices. (Am J Public Health. 2002;92:1073–1076)

THE AUGUST 2001 DECISION in the case of Grimes v Kennedy Krieger Institute, Inc,1 sent shockwaves through the public health research community. The court had challenged the acceptable level of risk in pediatric research studies, concluding that parents in the state of Maryland could not consent to their minor children's participation in research that posed even a minimal risk of harm if it offered no prospect of direct medical benefit to the subjects. Researchers feared that valuable public health research that complied with long-standing federal standards for research on children would be halted altogether or subjected to judicial oversight and intervention.

Two months after the initial decision was handed down, the court indicated that it had not intended to apply a zero-risk standard to “nontherapeutic” pediatric research studies. The court articulated acceptable risk in this context as including the “minimal kind of risk that is inherent in any endeavor,” which appears to be consistent with the federal regulatory standard.2

Despite the welcome clarification, Grimes deserves analysis of the ethical issues it raises and its relevance to future research. Here we provide a brief overview of the case, discuss some of the ethical issues raised by the court (risk–benefit assessment, informed consent, and exploitation), and offer opinions on the impact of the court's decision on research involving children and its possible effect on future research protection policies and practices.

OVERVIEW OF THE CASE

In Grimes, the Maryland Court of Appeals permitted the parents of minor children to bring a negligence action against the Kennedy Krieger Institute (KKI) for leadrelated health injuries allegedly contracted by their children during a KKI research study.1 Researchers at KKI, a Baltimore-based children's health facility and research institute affiliated with Johns Hopkins University, conducted a 2-year study (1993–1995) to measure the effectiveness of differing levels of lead abatement procedures in housing. Identifying the minimal effective level of lead abatement was considered important because of the need to preserve availability of low-rent urban housing that might be abandoned by landlords if they had to pay for the expensive repairs needed to eliminate lead using standard abatement methods. The study was funded by the US Environmental Protection Agency and local Maryland organizations.

In the project, 108 Baltimore rental properties were classified into 5 groups. Three groups of housing received less than full lead abatement (a different level for each group); housing in the 2 control groups either had previously undergone full lead abatement or had been constructed without lead paint. Over a 2-year period, the researchers were to measure and compare lead dust levels collected in the housing with lead levels in blood samples drawn from children living in those homes. Informed consent was to be obtained, and parents were to be notified of their children's blood levels and the results of lead dust collection in their homes. Blood levels would provide evidence regarding the effectiveness of a particular level of abatement. The study was approved by a Johns Hopkins University institutional review board (IRB).

Two families in housing that received less than full lead abatement brought the cases against KKI. One of the families was described as being on public assistance and from a minority group. The parents alleged that KKI designed a study that placed children at unnecessary risk; that KKI had discovered hazardous conditions in their homes and delayed reporting of test results that would have allowed them to prevent their children from being exposed to high levels of lead; and that KKI failed to completely and accurately inform them of all the hazards and risks of the study.

As will be discussed below, in the course of its lengthy opinion, the court willingly stepped into the shoes of investigators, IRBs, and regulators to scrutinize assessments of risk and benefit, oversee informed consent, and raised issues of exploitation. The judges also challenged a parent's authority to act in the child's best interests in consenting to research participation. One of the primary conclusions of the appeals court was that, in Maryland, a parent or other legal surrogate could not consent to a child's participation in “nontherapeutic research or studies in which there is any risk of injury or damage to the health of the subject.”3

This decision would have had the effect of proscribing research on children in Maryland even if the research was in compliance with federal regulations on research using children. However, as noted above, the court's later clarification appears to comport with the federal regulations. It stated that its conclusion referred only to risk in a “nontherapeutic study that promises no medical benefit to the child whatever, so that any balance between risk and benefit is necessarily negative.”2

It is now up to the trial court to apply the conclusions of the appeals court in light of a full presentation of the facts at trial, and further, to determine how the benefit and risk of the study should be characterized. In the interim, 3 additional families seeking $22 million dollars in damages have brought cases against KKI, raising claims similar to those currently under consideration.4

RISKS AND BENEFITS

Research on human subjects must be understood as an inherently risky endeavor, no matter how thoughtfully it is designed and how carefully it is carried out. This understanding is reflected in the long history of federal policies created to protect research subjects from risk, with additional protections for population subgroups deemed particularly vulnerable, including children.5 These policies are anchored in the ethical principles of respect for persons, beneficence, and justice—articulated in the Belmont Report6—which translate to the very issues that the court focuses on in its ruling: informed consent, appropriate balancing of risks and benefits, and avoiding the unjust use of research subjects.

These same ethical principles animate current federal requirements for prospective review by local IRBs to ensure that risks and benefits are appropriately balanced and fairly distributed and for provision of informed consent.7 Additional protections for minor children include parental (or guardian) consent and limits on risk levels. In particular, when research does not offer the prospect of direct benefit to the subjects themselves (a category of research that the Maryland court referred to as “nontherapeutic”), regulations limit participation by children to research that carries “not greater than minimal risk.” In practice, this risk criterion is interpreted as no more risk than the subjects would encounter in their daily lives, as ultimately judged by IRBs. When the proposed research does offer the prospect of direct benefit to the subjects, this minimal risk criterion does not apply, and IRBs are charged with ensuring that risk and benefit are appropriately balanced.

The trial court's characterization of the risks and benefits to the children who were subjects in KKI's research will be key to the final judgment about both the legal and ethical acceptability of the research. In Grimes, the court based conclusions on an assumption that the research had offered the subjects no benefits but a significant risk of lead poisoning, which would have made the risk–benefit balance unacceptable even under federal regulations. Assessment of the risk–benefit ratio on the basis of the facts that will be presented by the parties to the legal action will rest on whether the children would have experienced higher or lower levels of lead in their homes had they not participated in the research—a question that may be impossible to answer. Media accounts suggest that children involved in the research ended up living in homes with lower lead exposure than homes of families not enrolled in the study, as the improvements were made in neighborhoods where about 95% of the homes contain lead hazards.8,9

INFORMED CONSENT

Even if the children were to benefit, on balance, from participation in the research, their recruitment and inclusion would not be ethically acceptable if their parents did not adequately understand the risks and potential benefits of the research when they consented to their children's participation. The Maryland judges expressed their concern that the study's consent forms did not clearly indicate the risk that children would likely ingest hazardous lead dust particles and that the lead levels in their blood, whether increasing or decreasing, would be used as a way of assessing the effectiveness of the lead abatement measures.

The court was also concerned about the timeliness of disclosure of testing results. Although the parents were notified of their children's blood lead levels very soon after collection, they claimed that reporting of the results of dust collection in their homes was significantly delayed. In at least one of the cases, these results were reported 9 months after collection, and only after the child's blood revealed elevated lead levels.

If the researchers did withhold information from parents that could have helped them reduce the risks to their children, thus allowing identified risks to remain unaddressed, this raises significant ethical concerns. The situation would be especially problematic if the researchers knew that the research posed substantial health risks to the children in the study.

EXPLOITATION

Even if the children realized some benefit from their research participation, and especially if they did not, the court was concerned that such studies have the possibility to exploit economically disadvantaged populations, given their limited access to lead-free housing. Comparing the KKI study to the infamous so-called Tuskegee Syphilis Study, the court said:

Otherwise healthy children . . . should not be enticed into living in, or remaining in, potentially lead-tainted housing and intentionally subjected to a research program, which contemplates the probability, or even the possibility, of lead poisoning or even the accumulation of lower levels of lead in blood, in order for the extent of contamination of the children's blood to be used by scientific researchers to assess the success of lead paint or lead abatement measures. . . . [P]arents, whether improperly enticed by trinkets, food stamps, money or other items, have no more right to intentionally and unnecessarily place children in potentially hazardous nontherapeutic research surroundings than do researchers. In such cases, parental consent, no matter how informed, is insufficient.10

The court's comparison of the KKI research to the Tuskegee study may not be entirely apt, but it is important for what it says about how research and researchers are viewed, particularly within communities that may have been exploited in the past and that feel vulnerable to possible exploitation in the future. Real exploitation is obviously unacceptable, and perceived exploitation works to undermine trust in research and researchers. While it is difficult to regulate or adjudicate away mistrust, creating clear policies that ensure respect for subjects fosters trust in researchers, the research oversight process, and the research enterprise generally.

DISCUSSION

It is important to recognize that the case has not run its course through the judicial system. However, even at this stage of judicial inquiry, the case clearly highlights the need for true partnership in the research enterprise, particularly when proposed research involves vulnerable communities. Collaboration and open communication should begin with the development and design of research and continue throughout the processes of subject recruitment, data collection, and eventual disclosure of research results.

Further, community partnerships and collaborations provide an important test of the ethical propriety of the research. If researchers cannot convince the community in which they propose to conduct the research that it is acceptable, then they must reevaluate and revise their proposal until the community's concerns are adequately addressed.

Public health researchers already recognize the value of community partnerships in preventing misunderstandings. The extent of community participation in this particular research project is not indicated in the court's opinion, but given the court's views, a participatory approach can be seen to have a secondary benefit: as a tool to prevent judicial second-guessing of researchers' motives. In the absence of community participation, researchers can be perceived, as they were in Grimes, as placing the investigators' and institution's interests above the interests of research subjects and the communities to which they belong.

This is not to suggest that community partnerships are a substitute for improved oversight of research—both are important priorities. Certainly no child or parent should be taken advantage of in the context of research, and significant policy protections have been designed to prevent exploitation. However, no matter how well written, policies cannot prevent poor implementation in practice. We can always improve research practices, and we need only look to historical deficiencies to understand the need for doing so.

In the research at issue in Grimes, we do not know yet whether shortcomings were inherent in the research protocol (which ought to have been caught by peer review or the IRB) or occurred in the implementation of the research (meaning the investigators deviated from the protocol or there was a failure to adequately communicate with the research subjects). Whatever the eventual findings, protection of subjects would benefit from improved education of researchers, ongoing oversight of research in addition to prospective review, and mechanisms for assuring that research protocols are faithfully followed, all of which could be part of the IRB's responsibility.

Throughout its opinion, however, the court expressed its concern that IRBs are in positions of conflict of interest because they are committees of the research institution, with most of their members drawn from the institution they are charged to oversee, and therefore that the peer review system cannot adequately police itself. Our view is that the problems with oversight are less related to conflicts of interest than to the fact that IRBs have not received adequate priority and attention, financially and otherwise; therefore they have increasingly unmanageable workloads and inadequate resources to carry out their efforts. Clearly, IRBs can do a better job—and the court's strong words should provide an impetus for IRBs to redouble their efforts to carry out their task in ways that are above reproach—but they are only part of the overall research subject protection process.

With pointed references to many of the worst research scandals in history, such as the Tuskegee Syphilis Study, the ruling in Grimes seems to conclude that public trust in research is not warranted. While researchers and institutions can and must do a better job of respecting research protection policies, instances of failure do not indicate that investigators or the system of oversight are corrupt, or that intervention by the courts is the most effective remedy. Such interventions will lead only to greater obstacles to carrying out research, not to greater protections for subjects.

The court's decision in Grimes is not so much a call to rethink the ethics of research involving children as it is a warning to make sure that existing protections are adequately applied and overseen.

Wherever the source of research protections and oversight, creating and fostering a culture of ethical research among researchers and research institutions will lessen the need for emphasizing compliance and will go a long way toward warranting the trust of the public. Public trust may be the key, since without it there can be no research—no subjects for research projects, and no funding to support them.

However the case is ultimately decided, the court's involvement in Grimes should serve as a wake-up call for all those involved in research and its oversight—investigators, IRBs, peer review panels, and research institutions themselves. The age of judicial oversight is upon us, and responsible research protocols and scrupulous oversight are the way not only to conduct research in the most ethical manner, but to avoid judicial intervention as well.