This case discusses issues in ethics such as treatment of human research subjects, ethical research methods, use of informed consent, privacy and confidentiality.
From: Graduate Research Ethics: Cases and Commentaries - Volume 1, 1997
edited by Brian Schrag
Dr. Samantha Smith and Dr. Joyce Jones have collaborated on an epidemiologic study to identify determinants of preterm birth (delivery before 37 completed weeks of gestation). Over five years, the project team recruited 10,000 women into the study from prenatal care clinics in South Carolina. The study participants provided genital tract, blood and urine specimens, all of which were frozen and stored. They collected a great deal of other information through interviews with the participants and review of their medical charts, including social, demographic, health behavior and health history information.
All study forms and specimens were coded with identification numbers rather than the women's names. The links between the identification numbers and names were kept locked up and separate from all other study materials.
The consent form signed by the women in the study described in a general way the analyses planned on the biological specimens. For example, genital tract specimens were to be analyzed for the presence of infections, blood samples were to be analyzed for levels of different vitamins, and urine was to be checked for cotinine (a marker of exposure to cigarette smoke).
It has now been five years since the end of data collection for the study, and Smith has become very interested in studying potential genetic causes of preterm delivery. She realizes that this cohort of women, from whom a great deal of information has been collected, provides a unique opportunity to conduct an investigation of her hypothesis with minimal additional work or funding.
Smith is eager to proceed with a genetic analysis using the participants' stored blood specimens. However, Jones is concerned. "Sandy, the consent form that these women signed did not mention the possibility that we might do this genetic analysis. We'll have to contact them again and ask for their permission."
Smith thinks that contacting the women would be too difficult. "But, Joyce, that's unreasonable. You know the logistical challenges we face in trying to find these women -- some of the participants were recruited as long as a decade ago. We should just go ahead with the analysis. It's not that big a deal. It wouldn't make sense to pass up this opportunity."
Smith thinks of a way around having to ask the women for permission to do the additional analysis. "Joyce, how about making the specimens anonymous by destroying the links between the names and the identification numbers before we proceed with the analysis? Then we won't need to ask for consent; in fact, we won't even be able to."
Jones doesn't agree. "No, I still think we need to recontact them to ask for their permission to analyze their specimens. Besides, if we destroy the links between the names and numbers, that would interfere with conducting further follow-up studies. And what if we want to give them the results of our analyses?"
Suppose that when the study was originally conceived, Smith and Jones had foreseen the possibility of wanting to conduct further analyses on the subjects' specimens. Therefore, they had included in the consent form the following statement: "If any specimens remain, additional tests may be done for research purposes such as tests to look at factors in your blood that might affect pregnancy or health."
Posted 12 years ago
Karen Muskavitch Indiana University
This case challenges us to consider one of the fundamental questions in ethics: How ought people to treat each other? Yes, it's a human subjects case that emphasizes the issue of informed consent. However, if we only talk about "human subjects," we can too easily forget what this term really means, and that is people, like ourselves. Here, the subjects are 10,000 women who were initially recruited for the study when they were pregnant five to ten years ago. Those considering the ethical issues in the case are two senior researchers who have collaborated on the study since its beginning. In their discussions in this case, Smith seems to concentrate on the research subject aspect of those who participated in the study while Jones tends to view them more as individual people.
At one level, this case can be discussed as a means to clarify what the regulations are and how one goes about fulfilling them. That is a necessary part of educating researchers in the proper procedures to follow in human subject research, and should not be overlooked. However, this approach leads one to discuss rules and legalities more than ethics and can result in protocols that, while technically correct, do not truly treat the subjects respectfully and/or fail to consider all the possible consequences of the study. To get at the ethical issues, one must move from just considering the letter of the regulations to their spirit; to exploring the ethical principles that underlie all those rules.
For research involving human subjects, these are the three principles delineated in the Belmont Report
If researchers got into the habit of looking beyond the rule itself to consider the reason for the rule, and then asked whether their protocols and procedures were consistent with the underlying principles, a lot of the difficulties that have come up in human subjects research could be avoided.
The question asks whether the researchers are obligated to recontact the participants for their consent before doing a genetic analysis on the specimens donated by the participants for the previous study. The short answer is "yes," both for legal and ethical reasons.
I use the term "previous" study because it has "been five years since the end of data collection," and I assume that the analysis of these data and presentation of the researchers' interpretations at meetings and in journals would have already occurred. Thus, the primary work on the study is probably completed, but the samples are still in the freezer. I make this point because, acting on a point made by Wallace (1982) and noted in the IRB Guidebook (NIH 1993), most current IRBs would require that the experimental protocol specify what would happen to the data and specimens at the conclusion of the study. That does not seem to have been the case here, and probably was not required for most older studies. The fate of the data and specimens should be a concern of all researchers carrying out research using human subjects.
Why all this concern about a few blood samples and a few proposed "look-see" preliminary tests? After all, scientists are trained to keep samples just in case something comes up in the future such as a question about the previous work, or an opportunity to extend the investigation previously conducted. Most scientists have freezers and storage cabinets bulging with materials from earlier research. In this case, it is important to note that those 10,000 blood samples that Smith has in her freezer are not the same as the thousands of slices from a sediment core drilled out of a lake bed that a geologist across campus may have carefully stored in her cabinet. Smith's samples come from human beings, and therefore the issues of respect for persons and confidentiality need to be considered as well as scientific merit when new analyses are contemplated.
The women who participated in the study have a right to have a say in what is done with these specimens taken from their bodies. It is interesting to note that some universities have gone so far as to broaden the official, federal definition of human subjects to include living individuals "as well as human embryos, fetuses, cadavers, and any human tissue or fluids." (Indiana University, 1997) This definition emphasizes that these research materials are different from others scientists may use in their work, like the sediment core noted above, and that these samples must be treated with respect. The concept of respect for persons comes largely out of the writings of Immanuel Kant who asserted that one should Act so that you treat humanity, whether in your own person or in that of another, always as an end and never as a means only. (Kant, 1785) In this case, Smith is in danger of forgetting that the participants in the study are people, not just a means to advance her research. I'm sure that she has good intentions, but she can't skip asking permission of the study participants just because she believes the end will be beneficial for science and society.
Genetic analyses were not listed in the original consent form, and they yield information that is fundamentally different from that yielded by the analyses listed. The original analyses looked at characteristics that were environmental in nature; things that could be changed by treatment or modification of the women's behavior. The genetic analyses could yield information about characteristics of the women over which they have no control, and that could be passed on to their children. This information could cause harm if it became known to either the women themselves, or to others with whom they interact such as employers or insurance companies. Even for oneself, genetic self-knowledge is not always a benefit. It can be a burden if no available corrective treatment is available, and particularly if the knowledge is unwanted and is forced upon a person. What Smith proposes is not just an extension of the previous analyses. The women have a right to decide whether the genetic analyses are to be done on their samples, particularly in light of possible consequences. The issue of confidentiality will be discussed further in the commentary on later questions in this case.
Back to Top
Question 1. In response to Jones' objections, Smith suggests that they just destroy the list that links the study participants' names with the other data collected in the study. That would seem to solve the problem of confidentiality since no one could find out the participants' names after the list was destroyed. In addition, it would seem to qualify their work for an exemption from the need for informed consent based on the federal regulations. This suggestion raises several issues.
First, the proposed course of action is disingenuous. It's not the case that the researchers never had the participants' names. They have simply destroyed them because it seemed the easiest short-term solution. This approach is not the best way to enhance the public's trust of scientists. Also, as Jones points out, those names might be needed later if they decide to pursue a follow-up study.
Second, destroying the names does not really eliminate the linkage between those samples in the freezer and the women who donated them. The relevant regulations state that an exemption from the requirement for informed consent may be granted. . . if the information is recorded by the investigator in such a manner that subjects can not be identified, directly or through identifiers linked to the subjects. (45 CFR 46.101(b)(4)).
However, that is not the situation here because of the other data collected by these researchers. What the regulations refer to as "other identifiers" could still be used to link the genetic results to the women. Information that might facilitate the linkage are such items as maternal age, date of delivery, days preterm, baby's weight, location of interview and delivery hospital. It would not take too much work by a private investigator to identify the women to whom these data referred. Thus, confidentiality must still be a concern of the researchers, and they do not qualify for an exemption from the requirement for informed consent.
Question 2. This question is largely addressed in the commentary for Question 1 above, but there is an additional reason why the researchers should contact the participants for their consent before doing the genetic analysis: The genetic analysis poses potential risks that differ from those presented by the earlier analyses. If confidentiality is broken somehow, the women could be burdened with self-knowledge that they did not want; they may lack support in dealing with this information; and their new knowledge could have a negative impact on them and their children. Depending on the genetic markers that Smith selects for her work, the information derived may simply enable molecular geneticists to estimate the risk of premature delivery for the participants and their daughters.
However, if genetic markers with known associations with other diseases or conditions were used, the information could be even more damaging. For instance, if Smith checks for the alleles at the BRCA1 locus, one of the so-called breast cancer genes, she may or may not gather data that will help predict if a woman is in danger of giving preterm birth, but she will have information that bears on the current health of these women and their blood relatives. This information could be devastating to a woman, especially if she did not expect to receive it, and could be detrimental to her future employment and insurability if it became known to others. Because of these considerations, the new informed consent form would need to be different from the original one in order to address the potential risks associated with genetic analyses. The women should also be asked to indicate if they want to be informed of any information the researchers may gather that could have an impact on the participants' health. They have a right to determine what information will be gathered about them, and to control who will have access to the information, including themselves.
Question 3. Asking whether the researchers have an obligation to inform the women of the results of the genetic analysis assumes that the researchers went ahead with the analyses without obtaining additional consent from the study participants. Of course, it would be best to avoid the dilemma posed by discovering that Mary Brown, for instance, has an 80 percent chance of breast and/or ovarian cancer, but not knowing how she will react if you send her a letter containing this information. This scenario could be avoided by getting additional consent, before doing the analyses, in which one also asks the women if they want to be informed of the results of the genetic tests.
Assuming that they didn't ask, Smith and Jones must now consider the possible consequences of their decision to contact Mary Brown with the bad news. Here it is valuable to take a closer look at the principle of beneficence.
In the Belmont Report, the committee discussed what they meant by the principle of beneficence as it applies to research involving human subjects. In this document, beneficence is understood . . . as an obligation. Two general rules have been formulated as complementary expressions of beneficent actions in this sense: (1) do not do harm and (2) maximize possible benefits and minimize possible harms. (NIH, 1993)
Note that the first rule listed is "Do not do harm." Many philosophers assert that one has a greater obligation to avoid doing harm than to attempt to do good (Frankena, 1973), and that is an important consideration in this case. What "goods" might the researchers do by carrying out the analyses and then contacting a participant if they discovered something they felt she should know? What "harms" might occur as a result of this course of action? In the "goods" column, a list might include advancing scientific knowledge for the benefit of society, and aiding in the health care of study participants. In the "harms" column would go such things as psychological harm to the participants who are told bad news, and risk to future employment and insurability if confidentiality is broken. Many of the potential benefits are to science, society and the researchers, while almost all of the potential harms are to the study participants themselves. Thus, one must consider both the principle of beneficence and the principle of justice in answering this question. In my analysis, the potentials for harm and injustice were so much greater than the potential benefits that I would argue that the woman should not be informed of the results of the genetic analysis unless they have said they wanted to know.
In a situation like this, a vague statement on a consent form is of benefits to no one, although it might be technically in compliance with the regulations. One must question whether the researchers are really demonstrating a respect for the participants when the consent process is so vague. In addition, the vague wording, while sanctioning the genetic analyses Smith wants to do, does nothing to help her resolve the dilemma of how to respond if data are found that could have an impact on the participants' health. For the benefit of the participants, the researchers and the enterprise of science, it is best to be as through and forthright as possible in the consent form and in the recruiting interviews with the potential subjects.
From: Graduate Research Ethics: Cases and Commentaries - Volume 1, 1997 edited by Brian Schrag
This case concerns the appropriate use of stored biological specimens from research with human subjects. Parts 1 and 2 describe a scenario in which the investigators have finished their originally planned analyses on the stored samples and now wish to conduct further analyses that were not described when they obtained consent from the participants in the study. This scenario may happen more often today because of the development of new technologies for analyzing biological specimens that were not anticipated when samples were first collected.
In Part 1, Smith's desire to pursue this research area, which could be beneficial to her career, and society's interest in increasing knowledge about the causes of preterm birth are conflicting with the right of the participants to make informed decisions concerning the types of research activities in which they participate. Smith wants to take advantage of a unique opportunity to conduct important research with little additional effort or funding. She fears that a requirement to obtain permission from the participants to do additional analyses of their specimens will make it impossible to conduct the research at all.
The women who participated in the study probably assumed that the biological specimens they provided would not be used for any purposes other than those specified on the consent form. The informed consent process should allow prospective participants in a study to make informed decisions about whether or not to participate in research activities, based on knowledge of the nature and purpose of the study and of its risks and benefits. Some women may not have agreed to participate in the study if they knew it would contribute to knowledge of genetic causes of adverse pregnancy outcomes. They may feel that this type of research could lead to discrimination against groups of people or to less attention to prevention programs if preterm birth is perceived as primarily a genetic problem rather than the result of modifiable factors.
If Smith proceeds with the genetic analysis, she will presumably contribute to knowledge of causes of preterm birth, and she may expect that her reputation as a researcher will be enhanced. However, she may have doubts about whether her decision was ethical. There is a risk that results of the genetic analysis could be linked to the participants individually and threaten their ability to get or maintain health insurance or employment if it is discovered that they are genetically susceptible to preterm delivery. Also, there are possible adverse consequences for Smith's relationships with women who agreed to participate in the original study. If the women find out that their specimens were used for such a purpose without their permission, some of them may feel betrayed by the researchers and may develop a mistrust of scientists and the scientific process. Other members of the public who learn the details of the consent process may also become less trusting of the scientific enterprise, thus jeopardizing public support for science.
Smith has an obligation to fully disclose to women who agreed to be in her study what she intends to do with the samples she collected from them. She should respect the women's rights to make informed decisions regarding their participation in research studies. However, she might argue that the need for informed consent is not the same for research on stored samples as it is for research that more personally involves the participants. In Part 2, Smith raises the possibility of avoiding the need to obtain consent for the genetic analyses by destroying identifiers. Making the specimens anonymous would alleviate many of the concerns about confidentiality. However, even if the specimens are "anonymized," as long as each specimen is linked to other information obtained from the women (for example, age, race and date and site of recruitment), the danger of deductive disclosure remains. When specimens are currently linked to names, the issue also arises of whether it is appropriate to pass up the opportunity to obtain informed consent (by destroying the links). Additionally, as Jones points out, making the specimens anonymous means that the investigators will be unable to provide specific information on the results of the genetic analysis to individual women in the study, or to obtain more information about the women for use in follow-up research.
Several points must be considered regarding whether the investigators should inform women of their individual results. First, it is important to consider the accuracy of a genetic test and how predictive a genetic marker is. In other words, how likely is it that a woman who is determined to have increased susceptibility to preterm delivery will actually deliver prematurely if she becomes pregnant? The women should be asked, preferably before they agree to participate in the research, whether they would want to receive their specific results. Some women may prefer not to know. If results are shared with the women, counseling would be important. In the specific case of studies on preterm delivery, any specific individual results will not be particularly relevant to women who are past reproductive age or who do not plan to have more children. However, this information may be relevant if the women's children are likely to be at higher risk of preterm delivery.
In Part 3, the scenario is changed to suppose that the investigators anticipated the desire to do genetic research and included a statement in the consent form to that effect. Although this statement is technically accurate, it is quite vague, and it is likely that most participants will not understand that their samples may be used eventually for genetic research. Even if the consent form had informed the participants that the investigators would "look at genes," this phrase may still be inadequate without further explanation of the potential consequences of genetic research. It may be desirable to ask participants whether they agree to have the links between their names and the identification numbers preserved, and, if links are kept, whether they would want to know any specific results concerning genetic analyses. Alternatively, if the specimens were made anonymous, it may be appropriate to ask participants if they agree to have the links destroyed and to inform them that in that case it would not be possible to provide them with specific results.