Biomedical Research With "Decisionally Incapacitated" Human Subjects:
Legalization of a Defunct Normative Bioethetics Theory

Dianne N. Irving
Currently Professor of Philosophy & Policy
School of Philosophy
The Catholic University of America
(Written on request of the Journal of
Health Care Law, Univ. of Maryland)
Copyright June 30, 1998
Reproduced with Permission

I. INTRODUCTION

Consider this scenario: Martha is legally defined as a "decisionally incapacitated" person. That is, she is at least 18 years of age, and cannot give a valid informed consent for research participation because she cannot sufficiently understand the nature, extent, or probable consequences of the proposed research participation, cannot make a sufficient evaluation of its burdens, risks, and benefits, or cannot communicate a decision. However, she has also been determined to be "competent" by the same psychiatric researcher into whose scientific protocol she would be enrolled and able to give informed consent to sign an Advanced Directive for Research Participation. Martha can also choose a Research Agent who would then use his/her "substituted judgment" that Martha would have wanted to participate in high risk, no direct benefit, "research-in-general" for the benefit of future possible "groups" or "classes" of persons with the same or similar diseases or conditions, for the advancement of scientific knowledge, or for obtaining information which could not be obtained in any other way - were she competent. Other consenters - health care agents, surrogates, and "monitors" appointed by the IRB - could also enroll Martha in other categories of research protocols, including "emergency room research". There would be no civil or criminal liability to physicians, researchers, consenters or IRB members for any harms or injuries sustained by Martha during her participation in the research, nor any compensation for harms or injuries, or follow-up medical care. And unknown to her "normal, healthy" neighbors, all Advanced Directives for Standard Medical Care and Treatment could be convertible to Advanced Directives for Research Participation like Martha's, enabling consenters for standard medical care to enroll them in similar therapeutic and non-therapeutic research protocols if and when they should become "decisionally incapacitated".

Such would actually be the case if the new proposed statute1 in the State of Maryland addressing biomedical research involving "decisionally incapacitated" persons were to pass the Maryland legislature. The rationale (OD, 37) for such provisions is clearly stated in the October 1996 draft by the Working Group under Jack Schwartz, the Assistant Attorney General (although not repeated in following drafts). In that draft, their interpretation of the bioethics principles of autonomy, justice and beneficence (or what is known as "principlism", or the "Georgetown Mantra"), as articulated in the Belmont Report2, and which ground the federal OPRR regulations and Common Rule3 for research involving human subjects, are cited and accepted per se as the grounding of this Maryland State proposed statute. However, I would argue that the interpretation of these bioethics principles by the Working Group seems quite unbalanced. Their interpretation of "beneficence" leans very heavily towards utility, "autonomy" seems to be taken to an extreme, and "justice" is almost obliterated. The Working Group seems to be oblivious that the bioethics principles have had multiple and often contentious interpretations even from their inception,4 that bioethics principlism is a normative ethical theory5 (i.e., it takes a stand on what is right or wrong), that there is no such thing as a "neutral ethics" for public policy or law,6 or that bioethics principlism has been declared inoperative and unworkable by even those leaders in bioethics who helped to create and foster it so many years.7

Regardless of one's interpretation of these bioethics principles, and aside from the very strong possibility of serious concrete harm to multitudes of "decisionally incapacitated" human subjects as a result of certain interpretations, the adoption of this proposed statute would in effect result in the legalization of a very normative non-neutral theory of ethics which would be imposed on all of the citizens of the State of Maryland (and other states, if this were to set a legal precedent). This goal is actually articulated in the October 1996 draft: "The goal of this project is that '[f]or a change, law may be the handmaiden of ethics and ethics served by the law rather than vice versa'" (OD, 3).

The "ethics" they opt for is "secular" bioethics principlism. (It is unfortunate that the on-going drafts do not continue to provide this rationale for the citizens of Maryland.) However, the larger question which needs to be addressed is, should bioethics principlism be enshrined into local, state or federal law - especially now that it is summarily a non-operative normative ethical theory?

Nevertheless, in this limited article the pros and cons of the validity of bioethics, or of its legalization, will not be fully addressed. The focus, instead, will be simply to identify and briefly comment on a short list of ethical and legal concerns associated with this proposed statute which do flow directly from these questionable (and unchallenged) premises of the Working Group.

II. SHORT LIST OF CONCERNS

The following is merely a very short list of some of the major legal and ethical concerns presented in this proposed legal statute as expressed in the October 1996 and the May 1997 drafts:

1. Equating standard medical care with therapeutic research: One of the major concerns with this proposed statute, and one from which much of the other concerns below flow, is its unique equation of "standard medical care or treatment" with "therapeutic research":

If particular research has the potential for direct medical benefit to the individual, as determined by the institutional review board, and therefore could be considered a type of "health care", then the HCDA [Health Care Decisions Act] applies to participation in the research protocol as if the research were a treatment option like any other. If, however, a research protocol has no potential for direct medical benefit, the HCDA does not address the matter." (OD, 2) (insert mine)

Along the same lines, and included in the back of the October draft is a letter by Jack Schwartz to the Working Group to further clarify this equation:

"The term 'health care' [as used in the HCDA] is not defined. However, other provisions in the Act make its meaning clear. It is synonymous with a procedure or course of treatment that relates to the disease state of the particular patient"... This decisional framework, requiring a "health care" judgment in terms of the patient's assumed decision about a treatment, works well enough for therapeutic research. So long as there is an articulable link between the research and a possible improvement in the patient's condition, then a "health care" decision is possible, and the patient's hypothesized wishes would be the basis for it." (OD/S, 2)

Thus the logic of the proposed statute seems to go like this: If a treatment or procedure is "medically beneficial", then it is a form of "health care". Both standard medical care and therapeutic research can possibly be "medically beneficial". Therefore both standard medical care and therapeutic research are forms of "health care". In effect, this equates standard medical care with therapeutic research for purposes of this proposed statute.

That is why in the proposed statute a "health care agent" - or any substitute - who is allowed to make standard medical "health care" decisions for an incompetent person, can also make "health care" decisions concerning the person's participation in therapeutic research. That is also why, when a health care agent - or other consenter - makes decisions about "medical best interests" for a person, those "medical best interests" can include possible medical benefits from participating in therapeutic research. This in effect also makes an Advanced Directive for standard medical care equivalent to an Advanced Directive for Research Participation (the on-going interpretation in a policy for the "cognitively impaired" at NIH since 1985),8 and both can be used to enroll "decisionally incapacitated" persons in non-therapeutic research for groups or a "class", or in non-therapeutic research for the advancement of scientific knowledge or to obtain information which cannot be obtained otherwise, etc.

Although the Belmont Report is needlessly vague on the fundamental distinction between standard medical care and research,9 and on other key related points as well, most international codes of medical research ethics, adopted by the United States government, have always articulated a real distinction between standard medical care or treatment, and research. For example, the Declaration of Helsinki (1964, 1975, 1983, 1989)10 divides these activities into three basic categories: (1) standard medical care or treatment; and, (2) research, which itself is further divided into (2a) medical research combined with clinical care (clinical research), usually referred to as "therapeutic research"; and (2b) non-clinical biomedical research, usually referred to as non-therapeutic research. These distinctive kinds of activities are distinguished primarily according to their different aims, goals or purposes -and according to whether or not there is direct medical benefit to a particular patient or not. With reference to defining "research", the Declaration states:

"The purpose of biomedical research involving human subjects must be to improve diagnostic, therapeutic and prophylactic procedures and the understanding of the aetiology and pathogenesis of disease... In the field of biomedical research a fundamental distinction must be recognized between medical research in which the aim is essentially diagnostic or therapeutic for a patient, and medical research, the essential object of which is purely scientific and without implying direct diagnostic or therapeutic value to the person subjected to the research [Intro.]... The physician can combine medical research with professional care, the object being the acquisition of new medical knowledge, only to the extent the medical research is justified by its potential diagnostic or therapeutic value for the patient [Medical research combined with clinical care - clinical research]11 (inserts mine)

Note that in the first kind of research - i.e., "clinical research" - the goals of performing biomedical research to improve diagnostic, therapeutic and prophylactic procedures and the understanding of the aetiology and pathogenesis of disease are referred to as research, not as standard medical care, or "health care"; and the possible benefits must accrue to the individual patient - not to "others". Both of these elements are changed in this proposed Maryland statute. The second category of research is clearly defined as "non-therapeutic biomedical research" or "non-clinical biomedical research", where the research would not be for the direct benefit of a particular patient, but rather for the advancement of scientific knowledge, to obtain information that cannot be obtained in any other way, etc.

It is this blurring of the fundamental distinction between "standard medical care" and "therapeutic research" which is used to justify much of what follows.

2. Redefining categories of research: In the May 1997 draft of this proposed statute, even new and different categories of "research" are used: "expected benefit research", and "no expected benefit research" (MD, 10-15). "Expected benefit research" now explicitly includes research which, essentially, investigates diagnostic, therapeutic and prophylactic procedures and the understanding of the aetiology and pathogenesis of diseases, but which can be beneficial only to a group or "class" of possible future persons with the same or similar diseases or conditions. However, research performed for these purposes and for a group or a "class", is unequivocally non-therapeutic research, since there is no direct medical benefit to a particular individual patient. The category of "no expected benefit" research in the May 1997 draft essentially retains the traditional meaning of non-therapeutic research.

Nevertheless, "expected benefit research" now includes, by definition, "non-therapeutic research" for a group or a "class", instead of only for a particular individual person. The two categories of therapeutic and non-therapeutic research are commingled and confused, and both are redefined under "expected benefit" research.

The logic of the proposed statute seems to go something like this: Research that is "medically beneficial" is therapeutic research. Research for a group or "class" of possible future persons with the same or similar diseases or conditions is "medically beneficial". Therefore, research for a group or a "class" is therapeutic research (which has already been equated with "health care"). The critical phrase, "beneficial only to a particular patient", has been left out of the equation.

If some are confused as to where to draw the line between benefits to a particular individual, and benefits to groups, "classes", or society, etc., the Declaration of Helsinki and other codes offer clear guidance:

Concern for the interests of the subject must always prevail over the interests of science and society [Basic principles]... The physician can combine medical research with professional care, the objective being the acquisition of new medical knowledge, only to the extent that medical research is justified by its potential diagnostic or therapeutic value for the patient [Medical research combined with clinical care - clinical research]... In research on man, the interests of science and society should never take precedence over considerations related to the wellbeing of the subject (non-therapeutic biomedical research involving human subjects - non-clinical biomedical research].12 (inserts mine)

3. Redefining risk categories: Risk categories have never been successfully defined, including those offered in the Belmont Report or in the OPRR regulations. The bioethics literature13 does contain numerous works which at least attempt to articulate three basic categories for purposes of calculating a valid risk/benefit ratio: minimal risk, more than minor risk (or slight increase over minimal risk), and high risk. Obviously human subjects and their consenters are going to want to know what they are getting themselves into.

In both the October and May drafts it is not made clear that "greater than minimal risk" (OD, p. 7-15), or "more than minor increase over minimal risk" (MD, p. 12-16), includes "high risk" research. However, at least in the October draft these vague risk categories are applied to both "direct benefit research" and to "no direct benefit research". In the May draft, however, there are NO CATEGORIES OF RISK articulated for "expected benefit research"; the risk categories are used only in reference to "no expected benefit" research. This, too, is unprecedented. Both the lack of attempting to articulate a comparable "high risk" category, and the total lack of any risk categories for "expected benefit research" in the May draft, would make it difficult for any one to know or understand what they were getting into, and therefore to give a valid informed consent.

However, codes such as the Nuremburg Code, the Declaration of Helsinki, and others, consistently confirm the need for accurate and sound estimates of research risks:

"Every biomedical research project involving human subjects should be preceded by careful assessment of predictable risks in comparison with foreseeable benefits to the subject or to others. Concern for the interests of the subject must always prevail over the interests of science and society." [Basic principles] ... "Physicians should abstain from engaging in research projects involving human subjects unless they are satisfied that the hazards involved are believed to be predictable. Physicians should cease any investigation if the hazards are found to outweigh the potential benefits. [Basic principles] ... In any research on human beings, each potential subject must be adequately informed of the aims, methods, anticipated benefits and potential hazards of the study and the discomfort it may entail. [Basic principles] .... "The potential benefits, hazards and discomfort of a new method should be weighed against the advantages of the best current diagnostic and therapeutic methods. [Clinical research]14 (inserts mine)

Obviously, an accurate determination of risks is critical for purposes of protocol designs, informed consents and "substituted judgments", knowing when to remove patients from protocols, or determining whether patients should remain on standard treatments. Nevertheless, the May draft would allow a "decisionally incapacitated" person to participate in "expected benefit research" and "no expected benefit research" without it being clear to the researcher, the IRB, "decisionally impaired" patients or their consenters that high risk research is involved. Informed consent or "substituted judgments" are essentially precluded.

4. Informed consent by "decisionally impaired" human subjects: The issue as to whether "decisionally incapacitated", "decisionally impaired", "cognitively impaired" or any other category of "incompetent" persons are truly capable of giving valid informed consents to participate in research protocols, sign any kind of Advanced Directives, or appoint consent agents of any sort are still controversial issues.15 Theories abound; but solid studies or consensus do not. In fact, the literature contains a great number of arguments to the contrary.

Even the definition of a "decisionally incapacitated individual" in the May draft would seem to preclude any participation of these persons in any of the proposed categories of research: "'Decisionally incapacitated individual' means an individual who is at least 18 years of age and who cannot give a valid informed consent for research participation because the individual cannot sufficiently understand the nature, extent, or probable consequences of the proposed research participation, cannot make a sufficient evaluation of burdens, risks, and benefits of the proposed research participation, or cannot communicate a decision." (MD, 2) (emphasis mine).

5. Informed consent determined by the psychiatric researcher: "Decisionally incapacitated" patients are to be judged "competent" by the same psychiatric researchers in whose research projects they will participate. This could reasonably be considered the source of a clear conflict of interests. There is also the possibility of more subtle pressures on "decisionally incapacitated" human subjects when the researchers are also the subjects' own psychiatric physicians. As this concern is articulated in one of the codes:

When obtaining informed consent for the research project the physician should be particularly cautious if the subject is in a dependent relationship to him or her or may consent under duress. In that case the informed consent should be obtained by a physician who is not engaged in the investigation and who is completely independent of this official relationship. [Basic principles]16 (insert mine)

But if "decisionally incapacitated" persons are really competent to give informed consent, as testified to by the psychiatric researchers, then why is there any need for all those consenters to begin with? Why can't a "competent decisionally incapacitated" person simply then just give his/her own informed consent to take part in these research protocols?

6. Informed consent for "research-in-general": At least in the October draft there is reference to participation in "a particular" future research protocol. However, the May draft refers only to future unspecified "research" protocols or, what one might term, future "research-in-general". Informed consent would essentially be precluded if human subjects, or their consenters, are not given the specifics of a research protocol in which they would consider participating.

7. "Substituted judgments": "Decisionally impaired" competent human subjects will also be able to give informed consent to choose "research agents" who would then give their own "substituted judgments" that the patients would have wanted to participate in research-in-general had they been competent to consent. If the "research agent" is not available to consent, then a "health care agent", a "surrogate" or a "monitor" appointed by the Institutional Review Board (IRB) of the institution can then consent (another possible conflict of interests). Depending on the level of "consent agent", this could include high risk, no direct benefit research for groups or "classes" of possible future persons who might have this or similar diseases or conditions, or even for no direct benefit non-therapeutic research with utilitarian goals such as the advancement of scientific knowledge, the greater social good, for national security, or because there is no other way by which scientists can obtain the information. The issue of the use of "substituted judgments" within the context of even Advanced Directives for Standard Medical Care is still very controversial, and has been specifically rejected by many courts.17

8. Distortion of the principle of "autonomy": Allowing "decisionally incapacitated" patients to give informed consent is done in the pretext of "respecting the 'decisionally incapacitated' patient's autonomy". In the May draft this includes the patient's participation in high risk non-therapeutic research which would have no direct benefit to the individual patient. Such an interpretation of "autonomy" is a distortion of that bioethics principle, now being redefined and applied as an absolute right regardless of anything and without limitations. This was hardly the Belmont Report's original meaning of the bioethics principle of "autonomy", which was first referred to as "respect for persons" - a term which then included both autonomous and non-autonomous persons (who, according to this principle, are respected by protecting them from harm),18 and which initially was neither understood nor presented as an absolute bioethics principle. In fact, the bioethics principles were declared to be "prima facie", i.e., no single principle could out-weigh any other principle.19 Of course, the problem then became how to resolve conflicts among the principles, conflicts which bioethics theory itself never could resolve - a major reason for the internal collapse of bioethics principlism.

This distortion of the principle of "autonomy" would allow for considerable potential abuses of "decisionally incapacitated" patients. "Respecting" a patient's supposed "absolute autonomy" is understandably appealing to those who advocate for the "decisionally disabled" and to others who want to preserve their "rights". But in reality, it could obviously lead directly to great physical and psychological harm and injury precisely to those same vulnerable patients in areas in which they should actually be protected from abuse.

9. IRB incompetence: The reality of "unchecked research" and of the ineffectiveness of IRB's to competently review the research and/or to protect the patients from harm and abuse has historically been demonstrated to be problematic, and is under serious scrutiny now by both Rep. Shays and Sen. Glen20 who are conducting hearings on this and related research issues. If IRB's don't work, or are so compromised by bias and incompetence, then what justifies their continuing use in any research using any human subjects? Two interesting new phenomena concerning IRB's are: "IRB shopping", where drug or device companies go to a different IRB until they find one which will not reject their research protocols; and "deferred consent", where incompetent patients are enrolled in research protocols without their informed consent, and only later their informed consent (or proxy consent by a surrogate) is obtained after the fact.21

10. Appendix B: Obscured way in the back of the May 1997 draft, Appendix B would appear only to increase the list of surrogate consenters in an Advanced Directive for Standard Medical Care in the present Maryland Health Care Decisions Act. They could then consent for standard medical treatment for a once-competent patient who later becomes incompetent if none of the other consenters are available. This list includes spouses, siblings, adult children, other relatives, friends, etc. However, if the new proposed statute becomes law, these additional surrogate consenters for standard medical care could also consent to enroll any patient - "decisionally incapacitated" or otherwise - in medical research protocols, since the new proposed statute, in essence, would allow Advanced Directives for Standard Medical Care to be convertible with Advanced Directives for Research Participation.

11. Distortion of the principle of "beneficence": Given that one could find in the Belmont Report justification for defining "beneficence" as including "benefits to society", etc.,22 as well as to individuals, the Working Group seems to lean quite heavily on a purely utilitarian interpretation of it. For example, under "expected benefit research" in the May draft, the "benefit" part of the risk/benefit ratio can refer only to benefits to a group or "class", rather than referring only to a particular individual patient. As the May draft states:

Expected benefit research: (a) This section applies to research that: (1) presents a reasonable prospect of direct medical benefit to the class of decisionally incapacitated individuals who have been authorized by an IRB to be enrolled in the research; or (2) pertains to a disorder or condition of a decisionally incapacitated individual and presents a reasonable prospect of direct medical benefit to that individual. (MD, 10) (emphasis mine)

When a consenter is to consider "medical best interests", this beneficence could also refer only to "society", for the advancement of scientific knowledge, etc. The inclusion of purely non-therapeutic no direct benefit research in this proposed statute can be justified as "medical benefits" and considered as "best medical interests" to society:

But suppose there is no scientific evidence that participation is reasonably likely to offer benefits to the patient. The Act's "best interest" calculus does not include potential benefits to society as a whole, or even to those who might suffer from the same disease in the future. Participation in research of that kind, even with minimal risk, is not a "health care" decision within the meaning of the Act... If ultimately we think that health care agents or surrogates ought to have the authority to consent to nontherapeutic research under limited circumstances, and if researchers ought to enjoy immunity for acting upon such a consent, we should recommend a change in the law. (OD/S, 2)

And hence this new proposed statute. One has to wonder how one can even calculate "direct medical benefits" to groups or "classes", or to society. But it is interesting that here the Working Group was clearly aware that research for a group or "class" is considered as non-therapeutic research, and that it would not be considered "health care". However, the important point is that from now on, in all drafts, in "expected benefit research" the terms "direct medical benefit" and "medical best interests" can refer only to a group or a "class", or to an individual human subject. The terms "direct medical benefit" and "medical best interests" can also refer only to "no direct benefit" non-therapeutic research for "society", for the advancement of scientific knowledge, etc. The citizens of Maryland should understand that.

In sum, up to this point, the traditional distinctions used in biomedical research using human subjects have been blurred by redefining and equating: standard medical care, with therapeutic research; non-therapeutic research for a "class", with therapeutic research; beneficence for a particular individual patient, with beneficence for a "class"; "slight increase over minimal risk" research, with "high risk" research; an advanced directive for standard medical care, with an advanced directive for research participation; consent for a particular detailed research protocol, with consent for "research-in-general"; a risk/benefit ratio that refers to benefits to a particular individual patient, with a risk/benefit ratio which refers to benefits to groups or a "class". To put it mildly, such purely equivocal definitions of so many of these key medical research terms alone could render any and all informed consents and "substituted judgments" invalid. Clearly, the citizens of Maryland will have to be quite sophisticated about the controversies swarming within bioethics and with these deconstructed definitions, in order to understand exactly what they are getting into or voting for.

Next Page: 12. Invalid risk/benefit ratios
1, 2