Logo of schbulOxford JournalsContact UsMy BasketMy AccountSchizophrenia BulletinAbout this JournalContact this JournalSubscriptionsCurrent IssueArchiveSearch
Schizophr Bull. 2006 Jan; 32(1): 37–41.
Published online 2005 Sep 28. doi:  10.1093/schbul/sbj004
PMCID: PMC2632190

The Relevance of Empirical Research in Bioethics


Empirical research related to ethical issues in clinical research has grown dramatically in recent years. However, little attention has been devoted to the ethical relevance of the findings from this research. In order to examine the value and limitations of ethics-related empirical research, we discuss 3 case studies involving research with stored biological samples, placebo-controlled trials, and the idea of the therapeutic misconception.

Keywords: research ethics, empirical ethics research, stored samples, placebo-controlled trials, therapeutic misconception


Empirical research has been part of ethical reflection on clinical research for at least the past 30 years. For example, sociologist Bernard Barber and colleagues published in 1973 the results of extensive survey research concerning the practice of medical experimentation relating to “two key issues in the treatment of human research subjects: informed consent and the proper risk-benefit ratio.”1(p11) The volume of empirical research on ethical issues relating to clinical research has grown dramatically in tandem with the growth of bioethics as a discipline, especially its development within academic medical centers and the increasing tendency for bioethics research to be published in medical journals. The demands for professional advancement largely explain the increasing volume of ethics-related empirical research, as publication of empirical research in peer-reviewed high-impact journals is a key to professional success. Nevertheless, the ethical significance of this research—the way in which it can contribute to understandings of ethical issues in biomedical research—is not transparent and has not received the attention it deserves.

Ethics-related empirical research faces a basic philosophical challenge. Since at least the time of Hume (1711–1776), moral philosophers have argued that there is a distinction between descriptive propositions stating matters of fact about human conduct and normative propositions stipulating how people ought to behave.2(p521) As a matter of logic, normative conclusions cannot be derived from purely descriptive premises. Since empirical research yields descriptive propositions about practices, beliefs, and attitudes, what (if any) relevance can it have to pressing ethical issues?

Ethical inquiry and debate often rely, at least implicitly, on empirical propositions that frame the context of moral judgment and contribute to the justification of these judgments. For example, informed consent is a basic requirement of human subjects research, grounded in the principle of respect for persons. Empirical research can assess various aspects of informed consent for clinical research, including whether subjects have the requisite understanding of the research in which they are participating. Such research may be conducted to determine if participants in randomized controlled trials (RCTs) comprehend the fact that their treatment will be selected by a random process and not a medical judgment of what is best for them.

Suppose it is found that 75% of subjects in a variety of RCTs understand random selection of treatment. This poses further ethical issues. Should we feel satisfied that most subjects understand randomization or dismayed that a substantial proportion do not? Empirical research generates this question but is not equipped to answer it. The answer depends on the normative issue of what is considered necessary to give informed consent to an RCT, which may vary depending on the nature of the study in question. Nevertheless, those who are dissatisfied with these survey results may undertake valuable empirical research to determine whether innovative methods of informing prospective subjects of RCTs can improve their understanding of randomization and other elements of informed consent. This inquiry might itself take the form of an RCT to evaluate the innovative method in comparison with standard practice.

Ethical assessments of the design and conduct of clinical research by means of survey research and evaluative research aimed at improving the ethical conduct of clinical research are important examples of potentially valuable ethics-related empirical research. In what follows, we will examine 3 case studies with the aim of illustrating the value and limitations of empirical research as it relates to the ethics of clinical research.

Research With Stored Biological Samples

As mentioned above, respect for persons is a basic norm governing research with human subjects, and respect for persons implies that investigators typically should obtain individuals’ informed consent before enrolling them in research. Yet it is a truism of research ethics that no individual can possibly understand all that there is to know about even the most basic research study. This fact raises the question of what information individuals should be provided in order for them to give valid informed consent. Most commentators agree that investigators should begin to answer this question using the “reasonable person” standard, that is, investigators should provide to prospective subjects the information that the reasonable person would want to know to decide whether to enroll in the study in question. Empirical research can help to determine what information the reasonable person wants, in effect narrowing the is/ought gap. Research with stored biological samples provides an illustrative example.

With increases in genetic technology, research with biological samples has become increasingly popular and important. What information should investigators provide individuals prior to obtaining their sample to be used for future research? Some commentators argue that as long as the risks are sufficiently low, for example, when samples are anonymous, informed consent is unnecessary.3 Others argue that prospective subjects should be provided a list of various options, such as which diseases will be studied and how the samples will be stored, so they can make a meaningful decision whether to enroll in the research.4–6

Current standards indicate that the determination of which side is right in this debate depends, in large part, upon what information the reasonable person would want to make this decision. Allowing individuals to control the use of their samples shows respect for them as persons. However, the ethical principle of respect for persons alone does not determine how many or which choices individuals should be offered. To make this determination, it is important to assess which choices individuals want to make. For instance, the widely endorsed recommendation to offer individuals multiple check-off options for different diseases is based on the assumption that individuals find this choice meaningful.

Numerous studies have been conducted to assess individuals’ views regarding informed consent for research with biological samples.7–12 Using hypothetical scenarios, 1 study suggests that individuals want to control whether their samples are used for research but do not want to decide which diseases can be studied using their samples.10 A study of oncology research participants found that the vast majority would allow their biological samples to be used to study any disease.11 A study of over 1,000 research subjects found that more than 85% authorized unlimited future research use of their biological samples when given the opportunity to do so.12

These data reveal that the vast majority of individuals do not want to control which diseases are studied using their samples; nor do they want to control which researchers are allowed to use their samples for future research. These empirical findings have been cited in support of the proposal that it would be permissible to offer individuals a simple binary choice of allowing or refusing future research on their stored biological samples.12 Essentially, it seems that this is the relevant choice according to the reasonable person standard.

Like all empirical findings, these data do not definitively settle the issue. First, as with any meaningful area of human endeavor, empirical data will never show that absolutely everyone agrees on anything. Hence, empirical data, no matter how convincing, will leave us with the question of how we should treat the minority who do not, according to our data, agree with the views of the majority. This is not an issue that can be answered by the data themselves. Instead, ethical analysis is needed to assess the importance of the competing claims, and these will vary depending upon the issue in question. Is a fundamental right of the minority in question? Or is it more a matter of convenience?

Placebo-Controlled Trials

In the past 10 years a lively debate has developed over the ethics of placebo-controlled trials. There are sound methodological reasons for placebo controls for the evaluation of new treatments in certain disorders, despite the existence of proven effective treatment for the disorder in question.13 Placebo-controlled trials are common in psychiatric disorders for a variety of methodological reasons, including the fact that available treatments are generally aimed at ameliorating symptoms; they are assessed according to subjective outcomes; and RCTs demonstrate high rates of placebo response.14 Moreover, active-controlled trials designed to evaluate the equivalence or noninferiority of an experimental treatment compared with a standard treatment are subject to problems of internal validity or assay sensitivity.15 Nonetheless, patients randomized to placebo must forgo medically indicated, proven effective pharmacologic treatment. Many commentators consider this to be unethical.16–17 Other commentators assess the use of placebo controls in trials of treatment for depression and anxiety disorders (and a variety of chronic conditions in other areas of medicine) as no different in principle than the use of other research procedures that carry risks to subjects with no prospect of benefit to them.13, 18

One issue relevant to this debate is whether research participants randomized to placebo controls are exposed to undue risks of harm. The risks of placebo assignment can be assessed by empirical research. The risks of greatest concern in the case of depression are suicide and attempted suicide. Also important is whether patients randomized to placebo fare worse than those receiving experimental or standard treatment with respect to symptoms of depression. A meta-analysis of antidepressant trials in the U.S. Food and Drug Administration database, encompassing thousands of patients, found that those depressed patients receiving placebo were not at significantly greater risks of suicide or attempted suicide.19 With respect to symptomatic outcomes, patients randomized to placebo in these antidepressant trials experienced a mean 31% symptom reduction during trial participation, as compared with 41% symptom reduction for patients who received investigational or active comparator drugs. Thus, it appears that, in the aggregate, depressed patients receiving placebo controls in short-term trials are not made worse off or disproportionately disadvantaged as compared with those receiving pharmacologic treatment.

These data inform the debate, but they do not settle it. Knowing the consequences of placebo assignment does not determine whether the observed differences between placebo and active medication are ethically acceptable. This difference in risks and benefits is open to assessment on ethical grounds. Here, again, we face the gap between is and ought, between descriptive data and normative conclusions. More important, from an ethical perspective, these data relate to only a single dimension of the ethical dispute: risks of harm from placebo assignment. A key issue in contention is whether it is wrong per se to randomize patients to placebo in the face of proven effective treatment, owing to the therapeutic obligation of physicians and the rights of patients to receive competent medical care. Some commentators, invoking the principle of clinical equipoise, condemn such placebo-controlled trials.17 For example, Weijer has recently claimed that “placebo-controlled trials in the context of serious illnesses such as depression or schizophrenia are ethically egregious precisely because no competent physician would fail to offer therapy to a patient with the condition.”20(p10) Others argue that this position is fundamentally mistaken because it confuses the ethics of clinical research with the ethics of medical care.18, 21 This dispute concerns the basic ethical norms that should govern clinical trials. Empirical research cannot determine whether it is wrong to conduct placebo-controlled trials despite the existence of proven effective treatment, though it may play a subordinate role in the ethical analyses and arguments in support of the contending positions.

The Therapeutic Misconception

Empirical research may serve bioethics by posing issues of ethical concern deriving from the interpretation of research results. Therapeutic misconception has become a term of art in research ethics, denoting the propensity of patient/subjects enrolled in RCTs to confuse treatment provided in these studies with the situation of routine medical care, in which physicians recommend treatment based on a judgment of what is best for the particular patient.22 Appelbaum and colleagues coined this term to describe the findings of empirical research on the informed consent process for psychiatric patients enrolled in RCTs.23 It is worth noting that the initial formulation of the idea of the therapeutic misconception (TM), prompted by reflection on ethics-related empirical research, was guided by ethical thinking relating to RCTs. The authors appealed to the work of Fried, who contrasts the design of the RCT with the standards of “personal care” characteristic of clinical medicine.24 The former activity is designed to generate scientific knowledge about the efficacy and safety of treatments; the latter is designed to do what is best for individual patients. Recognizing the fundamentally different contexts of clinical trials research and medical therapy raises the question of whether patient/subjects understand that they will be treated differently in research than in medical care. Appelbaum et al. offer the following example of a patient/subject who manifested the TM:

In the same study, another subject was a twenty-five-year old woman with three years of college. At the time of the interview, she had minimal psychiatric symptoms and her understanding of the research was generally excellent. She recognized that the purpose of the project was to find which treatment worked best for the group of patients. She spontaneously described the three groups, including the placebo group, and indicated that assignment would be random…. When asked directly, however, how her medication would be selected, she said she had no idea. She then added, “I hope it isn't by chance,” and suggested that each subject would probably receive the medication she needed.23 (pp 22)

Identification of the TM, based on empirical research, leads to a number of vexing ethical questions. What difference does it make that some patient/subjects harbor TMs about research participation? Does evidence of the TM imply that informed consent was not obtained? Appelbaum et al. raise another key ethical question about their findings that continues to bedevil the ethics of clinical research: “Should we do anything about the therapeutic misconception?”23 (pp 22) It is a basic logical principle of moral philosophy, propounded by Kant, “that ought implies can.” It makes sense to argue that we should attempt to dispel the TM only if there is reason to think that we can do so. Appelbaum and colleagues have contributed to an answer to this question by building into their initial research an “augmented informational process” using a “neutral discloser,” not involved in the RCTs under examination, who instructed participants prior to enrollment about “key methodologic aspects of the research project, especially methods that might conflict with the principle of personal care.”23 (pp 23) It was found that subjects exposed to this neutral and augmented disclosure had a better understanding of important aspects of the research design, including randomization, use of placebo, and protocol-defined limitations on treatment.

From a policy perspective, whether we ought to require an informed consent process that is designed to obviate therapeutic misconceptions depends, in part, on an empirical issue: How prevalent is the TM? Although bioethicists appear to have presumed that the prevalence is high, it is only recently that any systematic data have been available beyond the small studies conducted in the 1980s on psychiatric patient/subjects. Appelbaum and colleagues have attempted to answer this question by reporting data on interviews with 225 research participants in 44 studies across a wide range of conditions.25

A fundamental methodological issue related to assessing the prevalence of the TM is how to measure this phenomenon. There is no standard, operational definition of what constitutes a TM; nor are there standard assessment tools for determining its presence or measuring the extent to which a given research participant manifests the TM. With respect to definition, Appelbaum et al. identify 2 variants of the TM, which they call TM1 and TM2. TM1 occurs “when participants express an incorrect belief that their individualized needs will determine assignment to treatment conditions or lead to modification of the treatment regimen.”25 (pp 2) TM2 occurs “when participants offer an unreasonable appraisal of the nature or likelihood of medical benefit from participation in the study, due to a misperception of the nature of the research enterprise.”25 (pp 2) Whether these 2 forms of the TM tap the same phenomenon is an important methodological issue with ethical implications. In their research, Appelbaum et al. found that 31% of the subjects manifested TM1 and 51% manifested TM2, with 62% judged to have a TM on 1 or both variants. It is obvious that how the concept is defined makes a big difference in determining its prevalence. TM1 certainly captures the core concept of confusing research with therapy. It is not as clear that TM2 constitutes the same misunderstanding.

More work is needed to refine and standardize the definition of the TM, to develop assessment tools to reliably identify this phenomenon, and to conduct systematic studies across different types of clinical trials with different groups of patient/subjects. Nevertheless, the policy question of whether methods should be developed and implemented to dispel the TM remains pressing. Focusing only on TM1, the finding that 31% of subjects manifested the TM appears ethically troubling. But the ethical significance of this concern remains open to inquiry. Joffe and Weeks assert that “the therapeutic misconception arguably constitutes the most important threat to the validity of informed consent to research.”26 (pp 1847) Does it follow that any manifestation of the TM among research participants invalidates their consent to research, thus making their enrollment unethical? In other words, is accurate understanding and appreciation of the differences between participating in clinical trials and receiving medical care necessary for informed consent?27 Since the TM appears to be a matter of degree, what level or extent of this misconception undermines informed consent? How should investigators in the informed consent process, or consent monitors, assess the presence of TMs among prospective subjects and intervene to dispel them? There is no consensus on how to answer these ethical questions. Further empirical research can inform the answer to this last question, but further conceptual research on the ethical significance of the TM and, more broadly, the meaning of informed consent to clinical research will be needed to address the other questions.


Empirical research can contribute importantly to ethical inquiry concerning clinical research. The potential value of this research depends on the ethical significance of the questions asked, the suitability of the methods to answer the questions, and interpretation of the ethical import of the findings. Ethical analysis can and should inform each of these aspects of empirical research. However, regardless of the quality of ethics-related empirical research, there are limits to what such research can accomplish. At the end of the day, the gap remains between what is the case and what ought to be.


Our work on this article was supported by the Clinical Center of the National Institutes of Health and the Intramural Research Program of the National Institute of Mental Health (FGM). The opinions expressed are ours and do not necessarily reflect the position or policy of the National Institutes of Health, the Public Health Service, or the Department of Health and Human Services.


1. Barber B, Lally TL, Makarushka JL, Sullivan D. Research on Human Subjects: Problems in Social Control in Medical Experimentation. New York: Russell Sage Foundation; 1973.
2. Hume D. A Treatise of Human Nature. Harmondsworth, UK: Penguin Books; 1969. [1739]
3. American Society of Human Genetics. Statement on informed consent for genetic research. Am J Hum Genet. 1996;59:471–474. [PMC free article] [PubMed]
4. National Bioethics Advisory Commission. Research Involving Human Biological Materials: Ethical Issues and Policy Guidance. Vol. 1. Rockville, MD: U.S. Government Printing Office; 1999. Available at http://bioethics.georgetown.edu/nbac.
5. National Action Plan on Breast Cancer. Executive Summary: Model Consent Form for Biological Tissue Banking: Focus Group Report. Available at www.4woman.gov/napbc/catalog.wci/napbc/model_consent.htm.
6. Clayton EW, Steinberg KK, Khoury MJ, et al. Informed consent for genetic research on stored tissue samples. J Amer Med Assoc. 1995;274:1786–1792. [PubMed]
7. Schwartz MD, Rothenberg K, Joseph L, Benkendorf J, Lerman C. Consent to the use of stored DNA for genetics research: a survey of attitudes in the Jewish population. Am J Med Genet. 2001;98:336–342. [PubMed]
8. Jack AL, Womack C. Why surgical patients do not donate tissue for commercial research: review of records. Brit Med J. 2003;327:262. [PMC free article] [PubMed]
9. Stegmayr B, Asplund K. Informed consent for genetic research on blood stored for more than a decade: a population based study. Brit Med J. 2002;325:634–635. [PMC free article] [PubMed]
10. Wendler D, Emanuel E. The debate over research on stored biological samples: what do sources think? Arch Intern Med. 2002;162:1457–1462. [PubMed]
11. Malone T, Catalano PJ, O'Dwyer PJ, Giantonio B. High rate of consent to bank biologic samples for future research: the Eastern Cooperative Oncology Group experience. J Natl Cancer I. 2002;94:769–771. [PubMed]
12. Chen DT, Rosenstein DL, Muthappan PG, et al. Research with stored biological samples: what do research participants want? Arch Intern Med. 2005;165:652–655. [PubMed]
13. Emanuel EJ, Miller FG. The ethics of placebo-controlled trials—a middle ground. N Engl J Med. 2001;345:915–919. [PubMed]
14. Miller FG. Placebo-controlled trials in psychiatric research: an ethical perspective. Biol Psychiat. 2000;47:707–716. [PubMed]
15. Temple R, Ellenberg SE. Placebo-controlled trials and active-control trials in the evaluation of new treatments: part 1: ethical and scientific issues. Ann Intern Med. 2000;133:455–463. [PubMed]
16. Rothman KJ, Michels B. The continuing unethical use of placebo controls. N Engl J Med. 1994;331:394–398. [PubMed]
17. Freedman B, Glass KC, Weijer C. Placebo orthodoxy in clinical research. II: ethical, legal, and regulatory myths. J Law Med Ethics. 1996;24:252–259. [PubMed]
18. Miller FG, Brody H. What makes placebo-controlled trials unethical? Am J Bioethics. 2002;2(2):3–9. [PubMed]
19. Khan A, Warner HA, Brown WA. Symptom reduction and suicide risk in patients treated with placebo in antidepressant clinical trials: an analysis of the Food and Drug Administration database. Arch Gen Psychiat. 2000;57:311–317. [PubMed]
20. Weijer C. When argument fails. Am J Bioethics. 2002;2(2):10. [PubMed]
21. Miller FG, Brody H. A critique of clinical equipoise: therapeutic misconception in the ethics of clinical trials. Hastings Cent Rep. 2003;33(3):19–28. [PubMed]
22. Lidz CW, Appelbaum PS. The therapeutic misconception: problems and solutions. Med Care. 2002;40(Suppl):V-55–V-63. [PubMed]
23. Appelbaum PS, Roth LH, Lidz CW, Benson P, Winslade W. False hopes and best data: consent to research and the therapeutic misconception. Hastings Cent Rep. 1987;17(2):20–24. [PubMed]
24. Fried C. Medical Experimentation: Personal Integrity and Social Policy. New York: American Elsevier; 1974.
25. Appelbaum PS, Lidz CW, Grisson T. Therapeutic misconception in clinical research: frequency and risk factors. IRB. 2004;26(2):1–8. [PubMed]
26. Joffe S, Weeks JC. Views of American oncologists about the purposes of clinical trials. J Natl Cancer I. 2002;94:1847–1853. [PubMed]
27. Sreenivasan G. Does informed consent to research require comprehension? Lancet. 2003;362:2016–2018. [PubMed]

Articles from Schizophrenia Bulletin are provided here courtesy of Oxford University Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...