• Is it unethical for physicians not to consider costs?

    The American College of Physicians has published their updated manual on ethics for physicians. I’m still wading my way through it, but on the whole, I’m impressed. Many of these types of documents get watered down by committee. This one, though, manages to make a fair number of declarative statements. Ezekiel Emanuel noticed the same in his accompanying editorial:

    However, what is truly amazing is that the Ethics Manual contains so many clear and specific recommendations. On execution, interrogation, and torture, the ACP Ethics, Professionalism, and Human Rights Committee is definitive and absolute. Indeed, it specifies not only negative duties but positive ones as well:

    Participation by physicians in the execution of prisoners except to certify death is unethical.
    Physicians must not conduct, participate in, monitor, or be present at interrogations. . . .
    Physicians must not be a party to and must speak out against torture. . . .

    I’m a little less impressed with their hedge when it comes to relationships and gifts from industry. Again, Emanuel agrees with me:

    Another section of the Manual that will probably generate debate involves physician–industry relationships. The Ethics Manual states that “[t]he acceptance by a physician of gifts, hospitality, trips, and subsidies of all types from the health care industry that might diminish, or appear to others to diminish, the objectivity of professional judgment is strongly discouraged.” Many will question why the ACP strongly  discourages such gifts rather than prohibits them outright. Indeed, if the gifts diminish or appear to diminish judgment, they must be prohibited to maintain the integrity of the physician and the profession.

    But the part that raised my eyebrows the most, and where I’m not sure I am in total agreement has to do with a physician’s responsibility to society:

    Physicians have a responsibility to practice effective and efficient health care and to use health care resources responsibly. Parsimonious care that utilizes the most efficient means to effectively diagnose a condition and treat a patient respects the need to use resources wisely and to help ensure that resources are equitably available.

    I’m sure many of you are shocked by my concern with this statement. Let me explain.

    I talk a lot about the fact that we, as a society, need to think about cost-effectiveness if we are going to get a handle on the cost of health care. This means saying no to some treatments and tests, because we have to use health care resources responsibly. When I say such things, inevitably someone counters me by questioning whether I would feel the same way if my child’s life was on the line.

    The answer is, of course not. If my child’s life were at stake, I would fight tooth and nail to get anything – and I mean anything – to save him or her. I’d do it even it it cost a fortune and might not work. That’s why I don’t think you should leave these kind of decisions up to the individual. Every single person feels the way I do about every single person they love, and no one will ever be able to say no. That’s human.

    Similarly, I don’t think that it’s necessarily fair to make it a physician’s responsibility. I also want my child’s doctor to fight tooth and nail to get anything that might save my child. Many times, physicians have long-standing relationships with patients. Asking them to divorce themselves from the very human feelings that compel them to do anything that might help their patients is not something that I think will necessarily improve the practice of medicine. They also should be human.

    So whose job is it? Well, mine for instance. That’s what I do as a health services researcher. That’s what policy makers should also do. That’s what we, as society should do. There are people who should have the responsibility of debating and deciding what is and is not cost-effective. They should have to make decisions that may be unpopular, and they should have to face the wrath of those whom the decisions impact. But there’s no good way to make it an individual’s responsibility to determine what is cost-effective for their child. That hardly seems “ethical”. I’m not sure asking doctors to do it is such a good idea either.

    But ethics is something we can and should debate. I hope we can have a robust discussion about this.

    UPDATE: Paul Kelleher thinks the ACP and I are more in agreement than I thought. I’m still not sure. I think some, if not many, will read the manual as I did. If the ACP and I are in full agreement, I with they had been clearer.

    UPDATE #2: I’m surprised by a few people who are really, really working to say that I’m twisting the ACP’s words. I’m not. And I don’t have to be “careful”. The ACP chose to work the words “cost-effectiveness” and “resources” into a manual of ethics. I’m bringing up a point of concern with that. I’m not condemning the ACP, nor the people who worked on the document. It’s their job to make things clear when they are telling physicians how to be ethical. If I’m misreading it, they might want to reconsider their language. If they agree with me, there were, perhaps, better ways to say what they were trying to get across.

    • Aaron
      interesting post. I think the document is quite strong, but I would probably would fall toward the side of thinking that docs do have some broader responsibility beyond just the person in front of them, esp if making decisions that spend public money. However, that is easy for me to say because I haven’t had the experience of being a doc and trying to answer such questions, of course.

      As an aside, what got me interested in end of life care was when I became guardian of a very ill person ~ 15 years ago (terminal cancer). I had power of atty and at some point had to make some difficult decisions and was very distressed when this persons doc couldn’t answer basic questions about likelihood of success or how much a procedure cost. This was driven in part because of the persons insurance situation and she had instructed me to not spend her finite resources on “hopeless” care b/c she was in the process of orphaning minor children.

      Very difficult stuff and it gives me a pit in my stomach to remember being in the middle. However, I recall being very unsatisfied by the inability of her docs to answer these questions, and was surprised at the time of their inability to even imagine trade offs related to cost and effectiveness.

    • Aaron,

      I understand your point, but I think you need to be a little more careful about the precise words they chose (which puts me basically in agreement with Paul). The particular sentences you extracted are more narrowly constructed than I think your characterization allowed–they refer to the “most efficient means to effectively diagnose and treat…”To me, that actually imposes a two-sided duty. The first, clearly, is to eliminate ineffective treatments and tests, and to avoid unnecessarily expensive tests when cheap ones will work just as well.

      The second, though, is to be aware of when a less expensive option has failed to effectively diagnose or treat, and to consider more expensive options at that point. Essentially, I think the guideline reiterates that the primary responsibility is to be effective–and at a certain margin, that means letting someone else be responsible for thinking about cost. That interpretation also seems consistent with the non-rationing role of the bedside physician that Paul mentioned.

      (This is perhaps already established, but I think it’s worth pointing out the words you used in your hypothetical: you would push for “anything that might save my child.” There are treatment options that exist which, according to evidence, don’t actually fall into that category–they simply haven’t been proven effective. You might differ with the ACP on the proper physician advocacy role in that case, but other than that I think you’re substantively in agreement.)

    • I would look at your paragraph from a bit different POV. There are many billing agencies that offer there services telling docs how they can be sure to not leave any money on the table. If your goal is to maximize income from the patients you care for, I think that words like parsimonious dont enter your everyday vocabulary. Instead, you end with very different utilization rates for similar patients. An emphasis on ethics could help, but I think that we really need to find better ways to align incentives.


    • Most, if not all, of the ethical conundrums that trouble you about making physicians rationers-for-the-greater-good go away or are at least significantly attenuated under cost sharing regimes.

      When patients are spending their own money, at the very least it forces doctors to explain the costs and benefits associated with each treatment, and forces patients to consider them.

      So long as the funds in question have alternate uses that the patient has discretion over – either for other care down the road, for other family members, for donation to medical charity, etc the patient will have a strong incentive to at least consider these questions in a meaningful way.

    • Aaron
      As I read your post, my thoughts were similar to Don’s, but unlike Don, and like you, I see patients.

      On delivering care: negative patient autonomy (the right to refuse a recommended treatment) and positive patient autonomy (the right to demand a treatment) are NOT morally equivalent.

      The patient’s right to refuse is nearly inviolable, but a patient’s right to demand a specific treatment is subject to physician discretion and veto. Were this not so, patients could demand virtually anything.

      These demands often fall along the lines of futility (lets leave that definition for another day), or waste. Ergo, if not us, then who?

      Often times, I look over my shoulder for the “other guy” to say no–as I did not sign up for this component of the job. Its me, unfortunately.

      When our profession accepts this role, and we have not, we can begin to approach the problem differently and cope more optimally with the stress. We are still recoiling from it, expecting others to step into the breech. It wont happen, and we must accept this fact.


    • I’d love to see a concrete example of how they feel the “parsimonious” principle should be applied. It seems obvious that we should try to save money **for the patient**–finances are a part of well-being after all– but is the suggestion here that doctors should become financial advocates for hospitals and insurance companies?

      Put another way: If I’m a patient making a decision, the first thing I’d like to know is what course of action is in my best interest. And by default, that’s what I expect to hear from my doctors and the committees that advise them.

      After that initial evaluation, I’m happy to consider making small concessions in my care in order to save the system money, but if I want to incur an extra X% risk of Y complication (perceived or real) in order to save my insurance company $Z, that’s an entirely charitable act on my part. And whatever else happens, I need to be able to absolutely trust that my doctor’s initial “my best interest” recommendation was untainted by other interests, or else trust is lost and I’m going to start looking for some other source of information on which to base my health decisions. My gut reaction (from the quoted sections) is that the ACP’s statement acknowledge the need to preserve that trust.

    • “So whose job is it? Well, mine for instance. That’s what I do as a health services researcher. That’s what policy makers should also do. That’s what we, as society should do. There are people who should have the responsibility of debating and deciding what is and is not cost-effective. They should have to make decisions that may be unpopular, and they should have to face the wrath of those whom the decisions impact. But there’s no good way to make it an individual’s responsibility to determine what is cost-effective for their child. That hardly seems “ethical”. I’m not sure asking doctors to do it is such a good idea either.”

      It’s far from clear that the capacity to do the right thing, or even understand what the right thing is increases as the specific knowledge of and responsibility for bearing the consequences of such a decision decreases.

      It’s also far from clear that the algorithmic rationing board will have perfect incentives even if it has perfect information about a patient’s particular information and perfect clinical insight into what’s effective.

      If nothing else, the payment schedule cooked up by the RBRVS should give advocates of centralizing authority substantial pause.

    • Very interesting discussion. One question appears to be whether ACP meant for parsimony to imply a choice function for tests and procedures that would reduce a patient’s outcomes (through reducing beneficial tests/procedures) versus reducing only those that are pure waste or serve the physician without helping the patient more than cheaper tests/procedures. I think Paul Kelleher concludes it’s the latter. As an economist, I’d think of it as minimizing the cost function subject to the constraint of health reaching the maximum level, as opposed to maximizing health subject to some or no cost constraint (or perhaps as opposed to maximizing health and profit).

    • A more interesting question Aaron raises is whether a doctor should be put in the position of limiting treatment that’s potentially beneficial or whether this should remain with society / health systems researchers / managed care. There are very legitimate reasons for cost effectiveness limits – think of the mentality of cancer drugs that probably won’t work on a specific patient, but hey you never know. The reason for docs as gatekeepers is that they observe patient-specific characteristics managed care simply cannot. For example, only my allergist can practically look at my outcomes on an expensive asthma drug and say it’s working substantially better for me than cheaper ones. Under the current system, managed care sets broad policies and cost-sharing mechanisms, but it cannot observe specific characteristics. Under this model a physician could achieve substantially better health outcomes than managed care as the gatekeeper.

      However, my experience with physician-gatekeepers in France leaves me wary. There, physicians wouldn’t go the extra mile in tests to figure out an illusive condition. This could be my preferring the higher amount of testing in the US, but it creates incentives for patients to be strategic in information they provide physicians and incentives to change physicians without providing histories, potentially leading to worse outcomes than a managed care gatekeeper.

      Does anyone have thoughts about a physician as an advocate for the patient, trying to get tests/procedures that are probably not that effective (especially given cost)? I don’t get why we’d want that other than avoiding strategic behavior by patients.