- BIOETHICS FORUM ESSAY
Autonomy, at least in the context of our framework for patient decision making, is under siege. The attack is coming from at least two quarters. Some believe that patients have taken the autonomy model too far, by demanding treatments that physicians would not recommend. This has occurred in legal cases like Wanglie, Gilgunn, and Baby K, where physicians believed that continued life-sustaining treatment for the patient was “futile,” or medically inappropriate, and patients or their family members demanded continued treatment. But it also occurs in other medical settings, for example, where pregnant women demand C-sections or assertive parents demand an antibiotic for their child’s sore throat.
The second criticism, wholly different from the first and less frequently discussed, is that patients, far from demanding too much, are actually uncomfortable making complex medical treatment decisions for themselves, especially when their choices are fraught with uncertainty. These patients, or their family members, are overwhelmed by the information and the responsibility.
There is considerable support for the view that patients who are ill and fragile may not want to make their own treatment decisions. In his book The Practice of Autonomy: Patients, Doctors and Medical Decisions, law professor Carl Schneider amasses considerable data on this point. For example, he cites a study 312 patients in a primary care clinic who were presented a series of vignettes representing various levels of illness severity and asked to indicate their desire for making their own treatment decision in each scenario. The mean score for the study population was 33.2 on a scale from 0 to 100, indicating that patient preferences for decision making were quite weak. In addition, they found that as patients were asked to consider increasingly severe illnesses, and as they got older, their desire to make decisions themselves declined.
Schneider also presents data that patients are willing to cede some of their autonomy when they are incompetent, even if they have expressed preferences for or against various forms of treatment. Here he cites the work of Ashwini Sehgal and colleagues (published in the Journal of the American Medical Association in 1992), who asked 150 dialysis patients “how much leeway their physician and surrogate should have to override [their] advance directive if overriding were in [their] best interests.” The patients varied greatly in their responses: 39 percent said “no leeway;” 19 percent said “a little leeway;” 11 percent said “a lot of leeway,” and 31 percent said “complete leeway.”
Articles confirming this perspective have also appeared in the popular press. A year ago, a New York Times article titled “Awash in Information, Patients Face a Lonely, Uncertain Road,” discussed the anguish and abandonment patients feel when forced to make difficult medical decisions. The article described the case of a 39-year-old patient with ovarian cancer that had metastasized to her liver. She was asked to decide whether to undergo a novel chemotherapy regimen about which five oncologists disagreed. When she asked her doctor what she should do he said he didn’t know, that she would have to make the decision based on her own values. The patient, “bald, tumor-ridden and exhausted from chemotherapy was reeling. ‘I’m not a doctor!’ she shouted, ‘I’m a criminal defense lawyer! How am I supposed to know?’” The story illustrates the frustration, anxiety, and loneliness of being a “modern patient” attempting to cope with medical uncertainty.
When patients are competent, ideally they and their physicians engage in shared decision-making, where there is give and take between both and neither the autonomy nor the paternalism model dominates. However, once a patient lacks competency, the opportunity for shared decision-making may be gone (at least as between the patient and the physician). In those circumstances, how should we respond?
If we are persuaded by the studies and anecdotes indicating that at least some portion of our population is not comfortable with the “autonomy model,” should we change our legal framework for health care decision making, in particular our framework for making decisions about life-sustaining treatment for patients lacking decision-making capacity?
Certainly, we cannot abandon autonomy; for many people the autonomy model is still sacrosanct. Rather, we need a model that allows for flexibility – for both autonomy and paternalism. There are, however, obstacles to choosing paternalism once a patient lacks decision-making capacity. One reaction would be to give patients an option to “undermine” their own autonomy, i.e., to autonomously choose paternalism. At least one state, Alabama, has made it easier for patients to give up some autonomy by modifying the state’s advance directive form so that patients can indicate whether they want their wishes strictly followed or prefer that their proxy do what he or she thinks is best, even if it means doing something different from what the patient has specified in the form.1 The Maryland legislature recently made a similar change to its advance directive form.2
Another response would be to make it easier for patients to appoint their physician as their health care agent. Many states actually prohibit this, arguably due to concerns about physician paternalism and possible conflicts of interest – although in so mandating, ironically, these laws replace physician paternalism with state paternalism.3 In New York, for example, health care proxy instructions provide that you can appoint your physician as your proxy but then he/she cannot also be your treating physician.4 In Massachusetts, a physician can be appointed as a health care proxy, but only if the physician is not the administrator, operator, or employee of a health care facility where the patient is receiving care, unless the physician is also a relative of the patient.5 Whether or not we should loosen our laws to allow for this option raises normative questions about whether physicians should be given the authority to both treat and act as agent.
Unfortunately, our current framework for health care decision making seems to be one in which we expect that one size will fit all. In the 1960s, we rejected the paternalism model and adopted autonomy. Perhaps, at least in the context of decision-making for patients who lack capacity, we need a scheme that allows for both – permitting the patient to choose autonomy or paternalism.
1. Ala. Code § 22-8A-4(h) (LexisNexis Supp. 2005).
2. S. 369, 2006 Leg., 421st Sess. (Md. 2006). The Department of Veterans Affairs is also revising its Advance Directive form to include a provision that would allow patients to specify how strictly they want their preferences followed. Personal communication from the VHA National Center for Ethics in Health Care, March 31, 2006.
3. See Arti Rai, Mark Siegler, & John Lantos, The Physician as a Health Care Proxy, Hastings Ctr. Rep., Sept./Oct. 1999, at 14.
4. N.Y. Pub. Health Law § 2980(3)(c) (McKinney 2002).
5. Mass. Gen. Laws Ann. ch. 201D, § 3 (West 2004).