H.ASTINGS CENTER.
Transparency: Informed Consent in Primary Care Author(s): Howard Brody Source: The Hastings Center Report, Vol. 19, No.5 (Sep. – Oct.• 1989), pp. 5-9 Published by: The Hastings Center Stable URL: http://www.jstor.org/stable/3562634 Accessed: 20/07/2011 10:27
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/pagelinfo/aboutlpolicies/terms.jsp. JSTOR's Terms and Conditions ofUse provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.j stor.org/actionlshow Publisher?publisherCode=hastings.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
The Hastings Center is collaborating with JSTOR to digitize, preserve and extend access to The HastingsCenter Report.
® http://www.jstor.org
While the patient's right to give informed consent to medical treat-ment is now well-established both in U.s. law and in biomedical ethics, evidence continues to suggest that the concept has been poorly integrated into American medical practice, and that in many instances the needs and desires of patients are not being well met by current policies. 1 It appears that the theory and the practice of informed consent are out ofjoint in some crucial ways. This is particularly true for primary care settings, a context typically ignored by medical ethics literature. but where the majority of doctor-patient encounters occur. Indeed, some have suggested that the concept of informed consent is virtually foreign to primary care medicine where benign paternalism appropriately reigns and where respect for patient autonomy is almost completely absent. 2
It is worth asking whether current legal standards for informed consent tend to resolve the problem or to exacetbate it. I will maintain that accepted legal standards, at least in the form commonly employed by courts. send physicians the wrong message about what is expected of them. An alternative standard that would send physicians the correct message, a conversation standard, is probably unworkable legally. As an alternative, I will propose a transpar-ency standard as a compromise that gives physicians a doable task and allows courts to review appropriately. I must begin. however, by briefly identifying some assumptions crucial to the development of this position even though space precludes com-plete argumentation and docu-mentation.
Informed consent is a meaningful ethical concept only to the extent that it can be realized and promoted within the ongoing practice of good medicine. This need not imply diminished respect for patient auton-
Huward Brody is director of th4 Center for Ethics and Humanities in tM Life &iences and associate professor offamilyfn”adice, MWUgan Stid8 University, East Lansing, ML
“lIStings Center Report September/OctDber 1989
Transparency: Informed
Consent in Primary care
by lIoward Brody
Culrent legal standards of informed consent send the wrong message to physicians about their moml and legal expectations. A “tmnsparency” model that sees consent as a conversation process can enhance good medical p18Ctice and patient autonomy without foreclosing appropriatejudicial reu;ew.
omy, for there are excellent reasons to regard respect for patient auton-omy as a central feature of good medical care. Informed consent, properly understood, must be consid-ered an essential ingredient of good patient care, and a physician who lacks the skills to inform patients appropriately and obtain proper consent should be viewed as lacking essential medical skills necessary for practice. It is not enough to see informed consent as a nonmedical. legalistic exercise designed to pro-mote patient autonomy, one that inteITUpts the process ofmedical care.
However, available empirical evi-dence strongly suggests that this is precisely how physicians currently
5
view informed consent practices. Informed consent is still seen as bureaucratic legalism rather than as part of patient care. Physicians often deny the existence of realistic treat-ment alternatives, thereby attenuating the perceived need to inform the patient of meaningful options. While patients may be informed, efforts are seldom made to assess accurately the patient's actual need or desire for information, or what the patient then proceeds to do with the information provided. Physicians typically under-estimate patients' desire to be informed and overestimate their desire to be involved in decisionmak-ing. Physicians may also view informed consent as an empty cha-rade, since they are confident in their abilities to manipulate consent by how they discuss or divulge information.5
A third assumption is that there are important differences between the practice ofprimary care medicine and the tertiary care settings that have been most frequently discussed in the literature on informed consent. The models of informed consent dis-cussed below typically take as the paradigm case something like surgery for breast cancer or the performance of an invasive and risky radiologic procedure. It is assumed that the risks to the patient are significant, and the values placed on alternative forms of treatment are quite weighty. More-over, it is assumed that the specialist physician performing the procedure probably does a fairly limited number of procedures and thus could be expected to know exhaustively the precise risks, benefits, and alterna-tives for each.
Primary care medicine, however, fails to fit this model. The primary care physician, instead ofperforming five or six complicated and risky procedures frequently, may engage in several hundred treatment modalities during an average week of practice. In many cases, risks to the patient are negligible and conflicts over patient values and the goals of treatment ornontreatmentare oflittle consequence. Moreover, in contrast to the tertiary care patient, the typical ambulatory patient is much better able to exercise freedom of choice and somewhat less likely to be
intimidated by either the severity of the disease or the expertise of the physician; the opportunities for changing one's mind once treatment has begun are also much greater. Indeed, in primary care, it is much more likely fur the full process of informed consent to treatment (such as the beginning and the dose acljustment of an anti-hypertensive medication) to occur over several office visits rather than at one single point in time.
It might be argued that for all these reasons, the stakes are so low in primary care that it is fully appropriate fur informed consent to be inter-preted only with regard to the specialized or tertiary care setting. I believe that this is quite incorrect for three reasons. First, good primary care medicine ought to embrace respect for patient autonomy, and if patient autonomy is operationalized in informed consent, properly under-stood, then it ought to be part and parcel of good primary care. Second, the claim that the primary care physician cannot be expected to obtain the patient's informed consent seems to undermine the idea that informed consent could or ought to be part of the daily practice of medicine. Third, primary care encounters are statistically more common than the highly specialized encounters previously used as models for the concept ofinformed consent4
Accepted Legal Standards Most of the literature on legal
approaches to informed consent addresses the tension between the community practice standard and the reasonable patient standard, with the latter seen as the more satisfactory, emerging legal standard.' However, neither standard sends the proper message to the physician about what is expected of her to promote patient autonomy effectively and to serve the informational needs of patients in daily practice.
The community practice standard sends the wrong message because it leaves the door open too wide for physician paternalism. The physician is instructed to behave as other physicians in that specialty behave, regardless of how well or how poorly
that behavior serves patients' needs. Certainly, behaving the way other physicians behave is a task we might expect physicians to readily accom-plish; unfortunately, the standard fails to inform them ofthe end toward which the task is aimed.
The reasonable patient standard does a much better job of indicating the centrality of respect for patient autonomy and the desired outcome of the informed consent process, which is revealing the information that a reasonable person would need to make an informed and rational decision. This standard is particularly valuable when modified to include the specific informational and deci-sional needs of a particular patient
If certain things were true about the relationship between medicine and law in today's society, the rea-sonable patient standard would provide acceptable guidance to phy-sicians. One feature would be that physicians esteem the law as a positive force in guiding their practice, rather than as a threat to their well-being that must be handled defensively. Another element would be a prospec-tive consideration by the law of what the physician could reasonably have been expected to do in practice, rather than a retrospective review armed with the foreknowledge that some significant patient harm has already occurred
Unfortunately, given the present legal climate. the physician is much more likely to get a mixed or an undesirable message from the rea-sonable patient standard. The mes-sage the physician hears from the reasonable patient standard is that one must exhaustively layout all possible risks as well as benefits and alternatives of the proposed proce-dure. Ifone remembers to discuss fifty possible risks, and the patient in a particular case suffers the fifty-first, the physician might subsequently be found liable for incomplete disclo-sure. Since lawsuits are triggered when patients suffer harm, disclosure of risk becomes relatively more important than disclosure ofbenefits. Moreover, disclosure of information becomes much more critical than effective patient participation in decisionmaking. Physicians consider it more important to document what
6
they said to the patient than to document how the patient used or thought about that information subsequently.
In specialty practice, many ofthese concerns can be nicely met by detailed written or videotaped con-sent documents. which can provide the depth of information required while still putting the benefits and alternatives in proper context. This is workable when one engages in a limited number of procedures and can have a complete document or videotape for each.' However, this approach is not feasible for primary care, when the number ofprocedures may be much more numerous and the time available with each patient may be considerably less. Moreover, it is simply not realistic to expect even the best educated of primary care physicians to rattle off at a moment's notice a detailed list of significant risks attached to any of the many drugs and therapeutic modalities they recommend
This sets informed consent apart from all other aspects of medical practice in a way that I believe is widely perceived by nonpaternalistic primary care physicians, but which is almost never commented upon in the medical ethics literature. To the physician obtaining informed con-sent, you 1II!VI!T' know when you are jinisW When a primary care phy-sician is told to treat a patient for strep throat or to counsel a person suffer-ing a normal grief reaction from the recent death of a relative, the phy-sician has a good sense of what it means to complete the task at hand When a physician is told to obtain the patient's informed consent for a medical intervention, the impression is quite different A list of as many possible risks as can be thought of may still omit some significant ones. A list ofall the risks that actually have occurred may still not have dealt with the patient's need to know risks in relation to benefits and alternatives. A description ofall benefits, risks, and alternatives may not establish whether the patient has understood the information. If the patient says he understands. the physician has to wonder whether he really under-stands or whether he is simply saying this to be accommodating. As the law
nastings Center Repmt September/October 1989
currendy aptJeo:rs to operate (in the perception ofthe defensively minded physician), there never comes a point at which you can be certain that you have adequately completed your legal as weU as your ethical task.
The point is not simply that phy-sicians are paranoid about the law; more fundamentally, physicians are getting a message that informed consent is very different from any other task they are asked to perform in medicine. If physicians conclude that informed consent is therefore not properly part of medicine at aU. but is rather a legalistic and bureau-cratic hurdle they must overcome at their own peril; blame cannot be attributed to paternalistic attitudes or lack of respect for patient autonomy.
The Conversation Model
A metaphor employed by Jay Katz, informed consent as conversation, provides an approach to respect for patient autonomy that can be readily integrated within primary care prac-tice.' Just as the specific needs of an individual patient for information, or the meaning that patient will attach to the information as it is presented, cannot be known in advance, one cannot always tell in advance how a conversation is going to tum out. One must follow the process along and take one's cues from the unfolding conversation itself. Despite the absence of any formal rules for carrying out or completing a conver-sation on a specific subject. most people have a good intuitive grasp of what it means for a conversation to be finished, what it means to change the subject in the middle of a conversation, and what it means to later reopen a conversation one had thought was completed when some-thing new has just arisen. Thus, the metaphor suggests that informed consent consists not in a formal process carried out strictly by protocol but in a conversation designed to encourage patient participation in all medical decisions to the extent that the patient wishes to be included The idea of informed consent as physician-patient conversation could, when properly developed, be a useful analytic tool for ethical issues in informed consent. and could also be
AcconIing to the transparency modeL the key to reasonable disclosure is not adherence to existing standards of other practitioners, nor••••to a list of risks that a hypothetical reasonable patient would want to know. Instead, dJaclosure is adequate when the physician's basic thinking has been rendered transparent to the patient.
a powerful educational tool for highlighting the skiUs and attitudes that a physician needs to successfully integrate this process within pati(mt care.
H primary care physicians under-stand informed consent as this sort ofconversation process, the idea that exact rules cannot be given for its successful management could cease to be a mystery. Physicians would instead be guided to rely on their own intuitions and communication skills, with careful attention to information received from the patient. to deter-mine when an adequatejob had been done in the informed consent proc-ess. Moreover, physicians would be encouraged to see informed consent as a genuinely mutual and participa-tory process, instead ofbeing reduced to the one-way disclosure of informa-tion. In effect. informed consent could be demystified, and located within the context of the everyday relationships between physician and patient. albeit with a renewed empha-sis on patient participation.·
Unfortunately, the conversation metaphor does notlend itself to ready translation into a legal standard for determining whether or not the physician has satisfied her basic responsibilities to the patient. There seems to be an inherendy subjective element to conversation that makes it ilI-suited as a legal standard for review of controversial cases. A conversation in which one partici-pates is by its nature a very different thing from the same conversation described to an It is hard to imagine how a jury could be instructed to determine in retrospect whether or not a particular conver-sation was adequate for its purposes. However, without the possibility for
7
legal review, the message that patient autonomy is an important value and that patients have important rights within primary care would seem to be severely undermined The ques-tion then is whether some of the important strengths of the conversa-tion model can be retained in another model that does allow better guidance.
I propose the transparency stand-ard as a means to ope rationalize the best features of the conversation model in medical practice. According to this standard, adequate informed consent is obtained when a reason-ably informed patient is allowed to participate in the medical decision to the extent that patient wishes. In tum, “reasonably informed” consists oftwo features: (1) the physician discloses the basis on which the proposed treatment. or alternative possible treatments, have been chosen; and (2) the patient is allowed to ask questions suggested by the disclosure of the physician's reasoning, and those questions are answered to the padent's satisfaction.
According to the transparency model, the key to reasonable disclo-sure is not adherence to existing standards of other practitioners, nor is it adherence to a list of risks that a hypothetical reasonable patient would want to know. In&read, disclo-sure is adequate when the physician's basic thinking has been rendered transparent to the patient. If the physician arrives at a recommended therapeutic ordiagnostic intervention only after carefully examining a list of risks and benefits, then rendering the physician's thinking transparent
requires that those risks and benefits be detailed for the patient. If the physician's thinking has not followed that route but has reached its con-clusion by other considerations, then what needs to be disclosed to the patient is accordingly different. Essentially, the transparency standard requires the physician to engage in the typical patient-management thought process, only to do it out loud in language understandable to the patient.9
To see how this might work in practice, consider the fonowing as possible general decision-making strategies that might be used by a primary physician:
1. The intervention, in addition to being presumably low-risk, is also routine and automatic. The physi-cian, faced with a case like that presented by the patient, almost always chooses this treatment.
2. The decision is not routine but seems to offer clear benefit with minimal risk.
3. The proposed procedure offers substantial chances for benefit, but also very substantial risks.
4. The proposed intervention offers substantial risks and extremely questionable benefits. Unfortunately, possible alternative courses of action also have high risk and uncenain benefit.
The exact risks entailed by treat-ment loom much larger in the physician'S own thinking incases 3 and 4 than in cases 1 and 2. The transparency standard would require that physicians at least mention the various risks to patients in scenarios 3 and 4, but would not necessarily require physicians exhaustively to describe risks, unless the patient asked, in scenarios 1 and 2.
The transparency standard seems to offer some considerable advan-tages for informing physicians what can legitimately be expected of them in the promotion ofpatient autonomy while carrying out the activities of primary care medicine. We would hope that the well-trained primary care physician generally thinks before acting. On that assumption, the physician can be told exacdy when she is finished obtaining
informed consent-first, she has to share her thinking with the patient; secondly, she has to encourage and answer questions; and third, she has to discover how participatory he wishes to be and facilitate that level of participation. This seems a much more reasonable task within primary care than an exhaustive listing of often irrelevant risk factors.
There are also considerable advan-tages for the patient in this approach. The patient retains the right to ask for an exhaustive recital of risks and alternatives. However, the vast ity of patients, in a primary care setting particularly, would wish to supplement a standardized recital of risks and benefits of treatment with some questions like, ''Yes, doctor, but what does this really mean for me? What meaning am I supposed to attach to the information that you've just given?” For example, in scenarios 1 and 2, the precise and specific risk probabilities and possibilities are very small considerations in the thinking of the physician, and reciting an exhaustive list ofrisks would seriously misstate just what the physician was thinking. If the physician did detail a laundry list of risk factors, the patient might very well ask, ''Well, doctor, just what should I think about what you have just told me?” and the thoughtful and concerned physician might well reply, 'There's cenainly a small posSlbility that one of these bad things will happen to you; but I think the chance is extremely remote and in my own practice I have never seen anything like that occur.” The patient is very likely to give much more weight to that statement, putting the risks in perspective, than he is to the listing of risks. And that emphasis corresponds with an under-standing of how the physician herself has reached the decision.
The transparency standard should further facilitate and encourage useful questions from patients. If a patient is given a routine list of risks and benefits and then is asked “Do you have any questions?” the response may well be perfunctory and automatic. If the patient is told precisely the grounds on which the physician has made her recommen-dation, and then asked the same question, the response is much more
8
likely to be individualized and meaningful.
There cenainly would be problems in applying the transparency stand-ard in the courtroom, but these do not appear to be materially more difficult than those encountered 'in applying other standards; moreover, this standard could call attention to more important features in the ethical relationship between physician and patient. Consider the fairly typical case, in which a patient suffers harm from the occurrence of a rare but predictable complication of a proce-dure, and then claims that he would not have consented had he known about that risk. Under the present “enlightened” court standards, the jury would examine whether a rea-sonable patient would have needed to know about that risk factor prior to making a decision on the proposed intervention. Under the transparency standard, the question would instead be whether the physician thought about that risk factor as a relevant consideration prior to recommend-ing the course ofaction to the patient. Ifthe physician did seriously consider that risk factor, but failed to reveal that to the patient, he was in effect making up the patient's mind in advance about what risks were worth accepting. In that situation, the physician could easily be held liable. If, on the other hand, that risk was considered too insignificant to play a role in determining which interven-tion ought to be performed, the physician may still have rendered his thinking completely transparent to the patient even though that specific risk factor was not mentioned In this circumstance, the physician would be held to have done an adequate job of disclosing information.1o A ques-tion would still exist as to whether a competent physician ought to have known about that risk factor and ought to have considered it more carefully prior to doing the proce-dure. But thatquestion raises the issue of negligence, which is where such considerations properly belong, and removes the problem from the con-text of informed consent. Obviously, the standard of informed consent is misapplied if it is intended by itself to prevent the practice of negligent medicine.
Transparency in Medical Practice
Will adopting a legal standard like transparency change medical prac-tice for the better? Ultimately only empirical research will answer this question. We know almost nothing about the SQrts of conversations primary care physicians now have with their patients, or what would happen if these physicians routinely tried harder to share their basic thinking about therapeutic choices. [n this setting it is possible to argue that the transparency standard will have deleterious effects. Perhaps the physician's basic thinking will fail to include risk issues that patients, from their perspective, would regard as substantial. Perhaps how physicians think about therapeutic choice will prove to be too idiosyncratic and variable to serve as any sort of standard Perhaps disclosing basic thinking processes will impede rather than promote optimal patient partic-ipation in decisions.
But the transparency standard must be judged, not only against ideal medical practice, but also against the present-day standard and the mes-sage it sends to practitioners. [ have argued that that message is, “You can protect yourself legally only by guessing all bad outcomes that might occur and warning each patient explicitly that he might suffer any of them.” The transparency standard is an attempt to send the message, “You can protect yourself legally by con-versing with your patients in a way that promotes their participation in medical decisions, and more specif-ically by making sure that they see the basic reasoning you used to arrive at the recommended treatment” [t seems at least plausible to me that the attempt is worth making.
The reasonable person standard may still be the best way to view informed consent in highly special-ized settings,where a relatively smaIl number of discrete and potentially risky procedures are the daily order of business. In primary care settings, the best ethical advice we can give physicians is to view informed con-sent as an ongoing process of con-versation designed to maximize patient participation after adequately revealing the key facts. Because the
IIBstlngs Cent« Report. Septsmber/OctDber 1989
conversation metaphor does not by itself suggest measures for later judicial review, a transparency stan-dard, or something like it, may be a reasonable way to operationalize that concept in primary care practice. Some positive side-effects of this might be more focus on good diag-nostic and therapeutic decisionmak-ing on the physician's part, since it will be understood that the patient will be made aware of what the physician's reasoning process has been like, and better documentation of management decisions in the patient record If these occur, then it will be clearer that the standard of informed consent has promoted rather than impeded high quality patient care.
Acknowledgments I plan to develop these ideas at somewhat
greater length, with special emphasis on the duty to disclose remote risks, in a volume to be titled The Healer's Puwer (in prepara-tion). I am grateful to Margaret Wallace and Stephen Wear for their insightful comments during the preparation of this manuscript.
References • Charles W.lidz It a1., “Bmiers to Infonned
Consent,” Annals of Internal Medicine 99:4 (1983),539-43,
• Tom 1.. Beauchamp and Laurence McCul-lough, Medi&al Etmcs.: The Mom! Resporuibil-i&s of Physiciam (Englewood Cliffs, l'{J: Prentice-Hall, 1984).
, For a condse overview of empirical data about contemporary infonned consent practices see Ruth R. Faden and Tom 1.. Beauchamp, A I:IisIcr] and '1'IItiory o/In,{ormttd Comenl (New York: Oxford University Press, 1986), 98-99 and associated footnotes.
• For efforts to address ethical aspects of primary care practice, see Ronaldj. ChriSlie and Batty Hoffmaster, Ethiall'-in FamU, Medicine (New York: Oxford University Press, 1986); and Hannon 1.. Smith and Larry R. Churchill,Professimud &hies andPrimo.ry Care Medicine (Durham, NC: Duke University Press, 1986).
, Faden and Beauchamp, A I:IisIcr] and '1'IItiory ofInform«l Comenl, 2S-49 and 114-50. I have also greatly benefited from an unpublished paper by Margaret Wallace.
• For a specialty opinion to the C<1l1trary, see W. R Coles el a1., "Teaching Infonned Consent," in Further DeueIopments in A.tse.!siIIg Clinical Ct>m.petma, Ian R. Han and Ronald M. Harden, eds. (Montreal: Can-Heal Publications, 1987), 241-70. This paper is interesting in applying to specialty care a model very much like the one I propose for primary care.
t jay Katz. T1IeSilent World oflJodorandfbtienl (New York: Free Press, 1984).
• Howard Brody, Stones of Sidmess (New Haven: Yale University Press, 198'1), 171-81.
• For an interesting study of physicians' practices on this point, see William C. Wu and Roben A. Pearlnian. “Consent in Medical Decisionmaking: The Role of Communicali.on,”Joumal of Genemllntemal Medit:ine 3:1 (1988). 9-14.
.. A court case that might point the way toward this line of reasoning is Prtcoun v. Frederidt, 395 Mass. 689 (1985). See Williamj. CumIn, “Infonned Consent in Malpractice Cases: A Tum Toward Reality,” New England Juu7TUJl ofMblicine 314:7 (1986). 429-31.
The Hastings Center has compiled a directory of state and regional ethics committee networks and an extensive bibliography on the structure and function of ethics committees; both are available at cost through the membership department.
To order please call or write:
The Hastings Center 255 Elm Road
BrlarcliffManor. NY 10510 (914) 762.$500
9
