I am going to give you a bit of an overview and use several cases as examples. First, I want to convince you that ethics is not just about life and death. It is more than that. I think that it matters a lot in rheumatology, where we are not worried about taking people off ventilators, we are not talking about abortion. There is a large range of ethical issues.
The second thing I want to give you is a little bit of vocabulary for how people in ethics try to talk about ethics. As you hear these words in the future, youíll have a little bit of help in framing and understanding that vocabulary. I remember when I was first doing cellular immunology years ago, and later molecular things, that I was in a fog that had to gradually clear as I was trying to read the papers and learn new things. It is the same with this discipline as well.
I am going to talk about some examples of ethical distress in patient care, clinical research and some things facing the profession, including an example of something that recently came up at the medical school. This is from a recent paper on the ecology of medicine. It is a follow-up on a study of four years ago that was also published, I think, in the New England Journal of Medicine. It points out that, of 1000 people, in one month 800 report some type of symptom and over 300 pursue medical care. Over 200 will actually go to an outpatient doctorís office, half of them to primary care. Another 60 seek an outpatient alternative clinician of some type, and 20 of them go to an outpatient clinic. So weíve got over 300 who are seeking care in an outpatient setting. So if you work your way down to the bottom of the slide, less than 1 out of 1000 are hospitalized in an academic medical center where most of the focus is on clinical care and clinical research, and that is true of ethics as well. I think what happens to those several hundred who are typical is what is interesting.
This study done in the late 1980s suggests that one-quarter to one-third- of visits or patients in an outpatient general medical practice, by some criteria that they developed for the purpose of their own study, had some type of ethical problem for the physician seeing the patient. There were issues of costs of care, psychological problems, including problems with consent that they framed in that way, problems with decision-making capacity, confidentiality, and patients refusing treatments. And several people -- LaPuma, Scheidermayer and later Howard Brodyís group in Michigan -- tried to organize this as a way of spurring study in this area. They defined problems with: dual loyalty; bed-side rationing; communication, like informed consent; professional and social responsibility -- for example, how we relate to the pharmaceutical industry; behavior and sexual issues; personal space; patient/guardian conflicts; and others.
This is just to give you an idea that if you make a list and look at it closely, a lot of ethical conflict can occur in rheumatology. Now the words that are used by ethicists to talk about these issues come out of the social and cultural changes during and following World War II. Religious authority, local community and paternalism waned as the way that people made decisions about what is right and wrong. The secular society became very pluralistic in this country, and we needed some other way of framing these discussions, because if you look from one religious group to another, there isnít even always shared vocabulary about what the right ethical decisions are and how to make them.
In the early 1970s, the Beaumont report was issued providing principles to protect patients in research. It came out as a logical development of the Nuremberg code addressing the horrors of what had happened during the Holocaust, as well as revelations about the Tuskegee experiment, in which black men in the South were left untreated for the observations of the natural history of syphilis even after antibiotics became available -- although there are some subtleties to that which make it more complex than it sounds on the surface -- and then observations that only a small fraction of patients in clinical research studies published in major medical journals had ever been asked for their consent to participate in any detailed fashion.
So these principles of research were issued, and participants of that group included Beauchamp and Childress from Georgetown and the Kennedy Institute, who wrote a landmark book called Biomedical Ethics. They talk about four principles. : first, we recognize respect for autonomy as a major principle, and over the last 40 years that really has been the dominant issue in ethics, although some of that is shifting a little bit. Second, do no harm -- non-maleficence. Ė And third, beneficence, is the other side of that. And then in some ways in contrast to individual autonomy, justice or distributive justice is the fourth issue. Most medical problems at least in medicine, or biomedicine, have aspects in which these values are in competition with each other.
Now there are other issues that inform ethics. Some people think that relying on these four principals, which sometimes, with a little bit of a negative connotation, they refer to as principalism, is too simplistic. Clearly religious and cultural values, legal constraints, social factors, and, as we know, economic constraints are enormous in their impact. Different institutional practices, a "Weíve always done it this way" sort of ethic, also play a part. Respect for human dignity and other profession virtues - compassion, trustworthiness, integrity - that we view as important.
We go from this vocabulary to some cases. This is a case of Mrs. Doe, a 51-year-old factory worker with a recent onset of mild inflammatory arthritis. You are seeing her for an initial consultation, and you recommend several measures, including a nonsteroidal antiinflammatory drug. So my first question to you is do you need to obtain informed consent? Letís look at thresholds here. Anybody have any thoughts? The lawyer says yes.
What I want to emphasize just with that question -- Iím going to give you a little data -- is that informed consent is a process of shared decision-making. Some audiences less sophisticated than this will say no, that is a waste of time, I donít need to get anyone to sign anything for using a nonsteroidal, and I always explain to them what is going on with it and get their agreement. It is a two-way conversation about the patientís goals and values and the clinicianís goals. And, in some legal opinions, it has to do with reasonableness, giving information that a reasonable person would need to make a decision.
Now there are several components to this, obviously. There is the nature of the decision. For example, I am going to send you for some blood tests to check your liver. There are alternatives that you could present, especially when you talk about medications. There are the risks and benefits which, with blood drawing, would seem excessive perhaps. There are related uncertainties -- this test may not help us or this drug may not work. Assessing the patientís understanding, that is not all that hard to do. You could say "Does that make sense to you"? or "How does that sound?" and elicit the patientís preference.
Now what information is likely to be communicated to this patient? Do you think we are going to go through all six of these risks, three or four of them, not mention them at all? Anybody want to venture a guess? How many would you guess that you do when you see a patient? Two of them? Four of them? So you think four or five, even when you are starting a drug like a nonsteroidal.
Well, this was one study done in general medical clinics and family practitioners, where there is much larger volume as you alluded to. And 80% of the time they were told the nature of the decision, "I want you to get a blood test" or "We are going to start methotrexate" or whatever the drug is. None was told of any alternatives. Only 9% (this is out of the University of Washington affiliated hospitals in the northwest) were told the risks and benefits. This is not just starting a nonsteroidal; it involves various drugs and recommending that people go to surgery, etc. None was told of related uncertainties; 2% assessed whether the patient understood what they were telling them, and none formally elicited the patientís preference. That sounds trivial and you say "If the patient goes and does it, then they must agree," but the patient doesnít always appreciate that it is their decision. So, in acknowledging that it is the patientís decision simply by saying "Is that okay with you?" youíve told the patient that they have permission to say no and ask further questions. In this group, that didnít happen; in fact, the mean number of criteria that this group met was 1.7. Now Braddock and his co-workers were concerned that maybe they made it too hard.
If youíre asking someone to do something simple, maybe you shouldnít ask them to do all those six things. And they did another study and they created a sliding scale that for simple things -- "Iím sending you for labs" -- all you would have to say is "Iím going to send you for these labs to be sure that the drug is not hurting your liver or interfering with the kidney. Is that okay with you?" -- but only 17% of the time did that happen.
Giving a new medication, where you say "There is this medication that we can try, but I think this other one is more likely to work and stop your pain, does that make sense?" -- none did that much. Sending someone for a procedure -- there were more stringent criteria typically to document that the patients agree to whatís on the consent form. But 90% of the time, even with easy criteria, when you really look at what doctors do, patients donít get informed consent.
What about rheumatologists? This is a study at the Brigham and Womenís. It was done in the early 1990s by Jeff Katz, Lauren Daltroy, Matt Liang and an attorney Troy Brennan. They audiotaped 46 encounters, and they told the doctors and the patients that they were looking at the doctor/patient interaction. They didnít tell them it was about consent. 24% of these esteemed colleagues of mine did not mention any side effects at all to the patient they started on nonsteroidals. Less than 16% of the time did they mention anything other than it can upset your stomach. 74% did say that it might not work, but a quarter didnít indicate that, so the patient didnít necessarily have a clue if it didnít work or in what period of time it didnít work to call the doctor and say "Hey this didnít work, what do we do next?" Again, similar to Braddockís study -- 1.7, but here it is the mean side of side effects that were mentioned. It didnít matter whether they were rare or common. Only 3% of the time were the rare side effects or 11% of the time were common side effects mentioned other than epigastric distress, and only 7% of the time were severe and 8% less severe side effects. It didnít matter whether it was severe, less severe, common or uncommon, that didnít dictate what people did.
So even in a rheumatologic clinic, even in an academic center where we are training people how to talk about these decisions, we donít always do as well as we think we are doing. Just looking into this study has helped me, because when Iím sitting in the room, I take a little bit of a step back, and a voice in the back of my head asks if Iíve really addressed, in those little simple phrases in less than two sentences, what I need to do this right.
An issue Iím not going to address in detail today is the capacity for making decisions, which arguably may be different for different decisions, but is something that needs to be assessed when interacting with your patients. We often only assess it when patients disagree with us, but we should be assessing it when patients agree. I mentioned the Connelly and DalleMura study about when they thought ethical problems arose, and it included decision-making capacity. The problems arose much more often in patients over 60. That is a population that we are seeing increasingly, and I think it is something you should think about.
Let us go on to a second case, Ms Jones is a 59-year-old woman with moderately severe osteoarthritis of her knees. Her medical history is otherwise unremarkable. Acetaminophen and non-drug measures have been of limited value. So she asks for that brand new COX-2 she saw on a commercial. So how do you respond? How many want to give her that brand new COX-2 in the commercial? Is that it -- only three or four? How many are going to give her the old tried and true COX-1? Okay, well I have three and three in an audience that is much larger than 6. So Iím not sure, but we will take that. I will assume that the latter answer is the one. Why would you give her the brand new COX-2? Okay and you know why the samples are there. Why wouldnít you? So you have more time. When I used to talk about nonsteroidals at the drug company talks, I used to talk about the rule of the first three patients, so I think that illustrates it. There is an observation about how our initial experiences shape our prescribing practices. She is insured by a capitated managed care plan, and she is referred by a primary care physician in the community who is trying to decide whether to refer his patients to your organization or to another competing network. This happens in New York, I believe, as well as other parts of the country. So now how might that affect your response? So if they are equally effective, but there is a measure of safety then it is clearly an advantage. It is not economic for the patient, but it is worth it. You are going to advocate for your patient, that is your job. Is that what you are saying? Okay -- your organization is under a little bit of financial pressure here and you do have a formulary and the managed care plans have formularies that define some of these things. Would you prefer that the decision was easier in the sense that the health system formulary or health care company gave you very clear guidelines, if it was someone up there that told you rather than you and the patient? No. All right.
Can you see 15 to 20 patients a day doing that with every patient? One last comment. This is a situation that certainly Iíve faced, the groups are pretty tough. They have withholds; they want their money back; they guard their patients. And they want to do well economically. And they are worried when they send someone to a big center, and a lot of money is being spent that they donít want spent. Okay. Now you recall having seen this patient somewhere, and then you realize it was in a television commercial advertising her tenacity and skill as a malpractice attorney. How would this effect the recommendations of those who want to have the lengthy discussion of old drugs work well, we know their side effects, it is cheaper, donít be duped?
So Prilosec -- in a few months hopefully will get cheaper when the generic price drops and so there are other approaches.
What is bed-side rationing, that may occur in this case? It is where you are making the decision to spend more money on a patient or not. It occurs when a physician withholds, withdraws or fails to recommend a service that in the physicianís judgment may be in the patientís best medical interest. That happens sometimes. Formularies say that you canít use this drug until you have failed two others with side effects. The formulary says you canít have it unless youíve had a bleeding ulcer or are on steroids or something else. So the physician here may make that decision because their hand is forced. In addition, to call something bed-side rationing, the physician has to act primarily to promote the financial interests of someone other than the patient. So that could be you withhold to protect a contract, to protect your organization, which has a limited formulary. Then the third criteria to cause bed-side rationing is that the physician has to have control over the use of the service. So if you have no choice, it wouldnít be rationing, you wouldnít be making the decision.
Now people have different views about bed-side rationing. Some view it as really abhorrent and others view it as the only way to do things. One says "We donít want to ration at all; we have to advocate for our patients". According to Steve Pearson, who tries to come to some sort of in-between view that some people are comfortable with and others are not, "rationing, broadly defined as resource allocation" and that is something that we recognize has to happen, "is necessary and inevitable in some form and at bed-side rationing the doctor and the patient making that decision is usually preferable to centralized rule-based rationing because it individually tailors the spending decisions to the circumstances to the individual patient and through incentives rather than through some bureaucracy". So he argues that keeping it one-on-one, where you can individualize the decision is better. He describes what he calls "ethical, proportional advocacy", where he says that there is open communication, rationalizing going on; you have to justify it. There should be some determination by the group, by the organization or a larger group of what a high-value or low-value service is, but you apply this approach across all patients, across all physicians in a fair manner. But by doing this there is minimal conflict of interest, he sees, by individual physicians. Then of course there would be recourse for challenging those decisions.
Emma Pellegrino, a former chairman of medicine at Georgetown, a distinguished ethicist, says with regard to bed-side rationing that the gatekeeper role is simply morally dubious. That is many ethicistsí view.
I will sometimes ask the patient "This is an expensive drug, This is what it will cost, Does this matter for you? Will this make a difference?" But I try not to make too many assumptions. One patient I had several years ago, who I would say was the most aggressive in seeking samples, was a man who owned a very large construction company, who had built large mega-complexes and was quite wealthy himself. He wanted samples.
I ran into a peculiar situation. We are seeing that the uncompensated care pool formularies and Medicaid formularies, at least in Massachusetts, are getting more restricted. The decisions arenít always rational. I have one formulary which is part of the uncompensated care pool where there is reimbursement from a pool of money that is state-wide for patients that donít even qualify for Medicaid, where the four choices are naproxen, ibuprofen, celecoxib and rofecoxib. There is no other nonselective nonsteroidal. If they have problems of some sort with those two drugs, I have to give them a more expensive drug. So I think we are going to see those formularies evolve.
So then we have Pearsonís view contrasting with Pellegrino, saying you have to have proportional advocacy, but then we have a view that the esteemed ethicist Dan Callahan has begun espousing from the Hastings Center. Callahan says we have to accept rationing. Rationing has to occur from social and political institutions, and it has to be widely talked about. He feels that rationing is inevitable. We have limited resources. We cannot all have the most expensive car. Some people will have less expensive cars. We cannot all have the most expensive drugs, and so far what technology has done, he argues, is to raise the cost of medical care. We havenít seen very many technological advances that have been lowering the costs, although I think there may be an argument about that. But he says this is inevitable. He said that if you look at Europe, the healthcare systems there that have national health are struggling to keep it, in part because the rationing has made it difficult. There is not enough open discussion about it because people donít want to face these decisions. He feels that to get to universal healthcare in this country we need first to accept and understand and begin to agree on rationing. It was very unusual to hear him speak recently, that it was the philosopher who was the most down-to-earth and the idealistic physicians in the audience, some of whom are advocating for national universal healthcare, who seemed most often in a philosophical netherworld at times.
These are the facts that Callahan argues that we have to face. So the question is where is it going to happen. I donít know the answer, but I think your arguments are important. This gets to some other issues, in part related to the TNF question, so you can twist this case around a little to ask it in a fairly clinical way.
Mr. Smith is a 46-year-old engineer with RA who is receiving some benefit from methotrexate 12.5 mg a week. He tolerates this without side effects, but still has a number of tender and swollen joints -- itís more than 6, more than 9 -- what the magic number is you need to have. What are the preferred treatment options right now? He is on 12.5 mg, no side effects but still active.
Increase the methotrexate, that is a big voice. This side of the room, do you agree with that side of the room? Does anybody want to add a new drug? Okay -- add leflunomide. Do you want to add them now? Discuss side effects? Talk about demyelinating disease risks?
Now at your department meeting, one of your colleagues reports that one of her clinical trials for a new biologic for RA, typical trial, methotrexate plus the new drug or methotrexate plus placebo, is enrolling slowly. But there is an additional financial incentive if you get three more patients in the next few weeks. The research associate has suggested that Mr. Smith, whom you will see tomorrow, appears to quality for this study. 12.5 is often the lowest point at which you could qualify, although you plan to increase his methotrexate. Like this side of the room, he trusts your clinical judgment. You have been taking care of him for a long time and, probably if you said "Letís do the study," he would follow your lead. What would you recommend at this point and why?
[The audience agrees that this situation would not occur for several reasons, primarily that no one on the study would know about this patientís qualifications for the study, for HIPAA and other basic confidentiality reasons.]
Well, one way Iíve seen colleagues address their concern is to say to the patient, knowing that the study is out there, but wanting the patient on a little more methotrexate and knowing that they have to be on it for three months, letís say, but thinking of the study, raise the dose or stop the Plaquenil, but make some change in anticipation for the study to sort of prep them for the study. Does that make anyone uncomfortable?
Anyone want to talk about the financial incentives of this study? Some might argue you should, and some might argue that it is not your study. Yes? So it is not just a choice between raising the dose or the study. There are other alternatives, and that would really be critical if you are going to call this informed consent.
How would this situation change if you werenít scheduled to see the patient soon enough to get him enrolled in the incentive but your colleagues say "Well, I know you canít, but I will see the patient." This colleague, who is the PI, knowing that you were reluctant to add him on during a busy day because this patient always seeks a lot of detailed information. . So what if your colleague says "Iíll see him, Iíll talk to him." What do you think of that? What is going to happen? So you are actually highlighting a difference between you and the investigator here. What if you were also the site PI of this study and the income for the study was actually important to you because it allowed you to do fewer clinic sessions, this was a portion of your income? Then what would you tell the patient other than the alternatives, risks and benefits? How far are you going to go in disclosing your interest to the patient?
Well, here this allows you to do two fewer clinics a week, so you are enthusiastic. I just want a show of hands. How many of you would disclose that you have a personal financial work-time etc.-related relationship to this study, that this is your study? How many will disclose that your study would do economically better, and so will you, if the patient enrolls?
This is an important point Ė the therapeutic misconception Ė the patient assumption in the context of clinical trials that the trial is much more likely to be a clinical benefit than it is, at least statistically when you analyze that likelihood. We are conflating the role of clinician, the treating physician who has this long-standing relationship that I pointed out, with the investigator who has an entirely different goal. The investigator is interested in the equipoise, the clinical equipoise of this decision that there is an unknown and wants to do a good study. That is a very different role from that of the clinician. Patients are relying on their clinician, who clearly sets up -- "Iím your doctor, you are going to talk to this guy about the study, and you can come back to talk to me, and we can make the decision together if you still have questions, because Iím advocating for you and not for the study."
How long are you going to carry out the trial as described? We know in three months that we are not going to know much except the very initial response. So is it six months from now, is it a year, two years? How many want to do this study for two years? How many a year? How many want to do it for six months? A good size of the audience. How many want to do it for three months? Very few. So there is a consensus around six months.
Ted Pincus has written about this problem of study design, and there is a lot of discussion about placebos in clinical trials. He points out that if you give a placebo for longer than three months in a rheumatoid arthritis trial, you are allowing for permanent damage to occur in some of those patients. If you are going to do that, it may be part of the informed consent, but maybe it is something that needs to be addressed in study design. He argues that just as we have data safety monitoring boards looking at toxicity, we should have just as rigid criteria for inefficacy of drugs in a trial so that we can pull patients out when we know they are in danger of getting irreversible injury to their joints. He also argues that payment for trials should not be issued as a per visit calculation so there is a motivation to keep people in trials who are being harmed and that consent should include this risk of irreversible damage. This study design might not be the right design for lupus or scleroderma or some other study. So when we talk about placebos, you canít really argue with them that placebos are either good or bad. Helsinki says you have to have the best possible treatment to compare with the modeled treatment, and this study would be virtually impossible by that criteria, which would require disease-specific guidelines.
Let me get to one more comment about conflict of interest and then the last case. "Iíve never been bought," says C. Everett Koop. "I cannot be bought. I am an icon and I have reputation for honesty and integrity. Let the chips fall where they may." "It is true that there are people in my situation who cannot receive a million dollar grant and stay objective. But I do." And this is from the Nofreelunch.org website that quoted Koop with a wonderful picture.
A study of residents and their relationship with pharma showed that most residents didnít think that they would be influenced by the reps. Less than 1/3 would, but 2/3 thought their colleagues would be influenced by the reps.
I think we all feel the same problem when we have a conflict. I just want to talk about conflict of interest where there is financial conflict with the study.
A conflict of interest is a set of conditions in which professional judgment concerning a primary interest tends to be unduly influenced by a secondary interest (such as financial gain). Conflicts of interest are different than other ethical conflicts of interest in that the conflict is not just a choice between competing values, such as patient autonomy, justice, cost of new drugs. Rather, it is an asymmetric choice between a primary interest, such as a professional duty in the care of your patients or a professional duty to do your research correctly, and a secondary interest, which is usually legitimate. It is legitimate to do things for economic gain; it is legitimate to want to do good research, so it is very different from other types of ethical conflicts. You have to keep in mind what the primary interest is, because the aim is to prevent the secondary factors from dominating or appearing to dominate that primary interest in making your professional decision. There is sort of a sliding scale starting with disclosure, which people debate the sufficiency of through mediation, abstention, judicial recusal, or even prohibiting certain activities.
Now, the last case, which happened to me recently. After the conclusion of a second year musculoskeletal pathophysiology block, you get a call from a colleague in the Division of Medical Ethics of the school, asking you to participate in a panel discussion in a meeting of second- and third-year medical students. Some of the second-year students chose to decline, and in fact chose to purchase at full cost, a free copy of the Primer of Rheumatic Diseases, because they perceived it as an improper gift from the pharmaceutical industry. What are the first words out of your mouth at that point? Mine was something like "Oh my god, youíre kidding." Something like that. Now I want you to raise your hand if you know who is recognized in this addition of the Primer for their donation. You donít need to name the company. I just want to know if you know who the company even is. One of you. How many of you have seen this edition of the Primer? I conveyed to him that I donít even know who is supporting this edition, but it was more complicated, because a second company had donated the Primers because a second company had paid for the shipping and handling. So there was not only the first company with an imprint in the book but there was the second companyís letter saying we are giving you this book.
So you participate in this panel. There is a student who initiated the whole thing, and there is a student in your own tutorial group who turns out is the rep to AMSA. There is an MD/PhD candidate who has studied the history of the relationship between physicians and the pharmaceutical industry. He is working with the Chairman of Medical Ethics who is also the chair of the Science and History Department at the university right now, and several colleagues from the Division of Medical Ethics including a couple of former editors with very recognizable names from the New England Journal of Medicine, and also your mentor in ethics whom you had done a visiting fellowship with. After a vigorous discussion of the issues pro and con, the session comes to a close. You do disclose that you are the chair of the Medical and Scientific Committee of the Arthritis Foundation for Massachusetts and your other relationships. You call Jack Klippel the previous day after that phone call, and you say "Jack, do you know who supported this? You were the editor, the medical director." And he says "I donít even know how to check." And you ask him "Is there any conflict?" He says, "People donít even know who is running it." So after a vigorous, intellectually challenging discussion, the course administrator comes up to you afterwards and says "Well, should I order the books for next year". Well, some of the first-year students were there, they came. They are into preventive ethics, which I will not discuss today, but preventive ethics is what you then do after you face an ethical dilemma and you think about how to prevent ethical problems.
Some say, "Well, a gift is a gift, why as students should we receive any gifts?" Some say, "We donít want the pens. You can do a pen exchange. You get a No Free Lunch pen if you send them pharma pens. We donít need to start this relationship now, we are just medical students. We donít want to be bought."
This is a problem. My understanding, in fact, is that a former chairman of yours has been asked to chair a committee for the Arthritis Foundation to look into its relationship with industry.
So the discussion was raised "Is the Primer influenced by pharma? People wanted to know when they saw this imprint whether the company is involved in the editorial decisions on the book. They didnít know one way or another. I know some medical schools, when they give away the Primer, send a letter that describes exactly how the Primer is put together, sort of as a proactive step to limit concerns. Could it exist without pharma support? Some people in the Arthritis Foundation believe it might not be able to. Is this a gift from pharma to the students or the Arthritis Foundation to the students? I donít know the answers to all these questions, but I do know why the AF supports its publication -- it is for education and for the field, But pharma has a different reason for supporting it and they are not stupid.
How does the Arthritis Foundation, without going even more to industries, get the support its needs? Should students accept gifts from pharma? Some students say no.
There are a lot more ethical issues that we could be talking about in rheumatology. There is cost of treatment. Phase 1 trial consent. There is vast therapeutic misconception. Is informed consent even possible? Some experts say, truly, it isnít, but we should do our best. There are dual roles. Non-financial conflicts of interest, things we do for promotion, time conflicts, whether rheumatologists should be primary care doctors. There are ethical dimensions, disclosure of trial data, IRBs, managed care, gifts, alternative therapies, disability, fibromyalgia, pain management, and the list goes on.
Here is a summary to take home. Rheumatology is filled with ethical dilemmas. That is the first thing I wanted to convince you, and it is only likely to get worse. In times of shortage, ethics only becomes a bigger issue. Informed consent is a process, not a piece of paper. That is the key thing. If you just remember that and live that, I will be very happy with what you got out of this. Rationing is a reality. It is just here and doing it right is going to be very hard, and I have laid out for you some of the choices in how to do it.
In conflicts of interest, looking out for number one, it means something a little different. It means look out for the primary interest, which must come first and also appear to come first, or you will be in a weak situation.
And, finally, that a gift begets a relationship. That is what gifts are about, and that a relationship begets an obligation and that in accepting gifts, we have to keep that in mind and understand the nature of the gift and the nature of the obligation. That is where the question of whether the Primer is a gift from the Arthritis Foundation or a gift from pharma becomes important.
Then I know that you have got a lot of interest in legal issues. Ethics and the law brings to mind two quotes for me. A commencement address up in Boston more than 20 years ago says that a society that is based on the letter of the law never reaches any higher. It is taking small advantage of the high level of human possibilities. Whenever the connective tissue of life is woven of legalistic relations, there is an atmosphere of mediocrity paralyzing manís noblest impulses. So as George Annas, an attorney and ethicist says "Good ethics begins where the law ends." And I thank you.