Building on the American College of Physicians Ethics Manual

Abstract

Physicians face ethical issues in their daily clinical practice. Consider a 67-year-old man newly diagnosed with low-risk localized prostate cancer after prostate-specific antigen screening. Radical prostatectomy does not decrease long-term mortality in randomized trials involving such patients, but increases patient-reported adverse effects of incontinence and sexual dysfunction (1). How should physicians present complex information during informed consent discussions? Should the physician give a recommendation? How should the physician take into account uncertainty in evidence and the weight the patient places on the benefits and adverse effects of various options? The American College of Physicians and the authors of the Ethics Manual have provided helpful, specific guidance on many difficult ethical issues that internists face in everyday practice as well as providing an explanation of the ethical principles and professional obligations that justify these recommendations. There are, however, several areas where the manual could be improved, which we illustrate using the above vignette. Improved Philosophical Framework for Clinical Ethics Autonomy, beneficence, nonmaleficence, justice, and the duties of professionalism are indeed important fundamental concepts for the ethical practice of medicine. But how we understand those concepts matter for concluding what actions they permit or encourage. As defined in the manual, some of these ideas lack conceptual clarity, which may lead to inconsistency or confusion when applying them in cases. Take, for example, the idea of beneficence. The manual repeatedly urges physicians to act in the best interest of the patient. Indeed, it reminds physicians that doing so is their primary responsibility (2). But, what do we mean by interest or best interest? We could mean at least 4 different things by interests, which would give a physician different answers to how a case should be handed. Objective interests are objectively good for a personfor example, health, privacy, relationships with family and friends. Objective health-centric interests prioritize avoidance of death, disability, pain, suffering, unnecessary hospitalizations, and adverse outcomes of interventions. Subjective interests are defined by a patient s particular values, preferences, and goals. A patient may, for example, value the pleasant feeling associated with smoking or overeating, regardless of long-term health effects. Subjective informed interests are what a patient would want if she were fully informed, rational, and self-controlled. What is in the best interest of the man with low-risk prostate cancer: immediate treatment or active surveillance? In terms of objective health-centric interests, active surveillance might be in the patient s best interest because it has a similar mortality risk as immediate treatment, without the risks for impotence and incontinence. From the subjective interests view, men often have a strong preference to get the cancer out to minimize anxiety, which favors immediate treatment. From the informed interests view, there is evidence that men tend to overestimate the anxiety reduction associated with immediate treatment and underestimate the impact of impotence and incontinence, perhaps favoring active surveillance (3). Thus, different interpretations of best interests will lead to different recommendations for action. The specification of best interests can also help in other examples, including decisions regarding surrogate decision making for incompetent patients, counseling patients about genomic testing, and caring for minors with sexually transmitted infections or seeking contraception. Specifying justice and respecting autonomy would also shift recommendations. The manual defines justice as the equitable distribution of the life-enhancing opportunities afforded by health care (2). This, however, is an egalitarian view of justice. Other views of justice (for example, basic decent minimum, libertarian, desert-based) would generate very different answers to justice questions. In sum, more clarity regarding which account is being appealed to can illuminate the disagreements in particular casesan important step toward resolving them. Integration of Behavioral Science Another opportunity for improvement involves integrating behavioral and decision science with ethical decision making. For example, the manual recommends that under the doctrine of informed consent, the physician should provide enough balanced information so that the patient can make an informed judgment about how to proceedand that a recommendation accompany that presentation. Yet, behavioral science calls into question the very idea of a neutral presentation of information, shows that recommendations themselves are highly biasing, and demonstrates that dozens of decisional heuristics and biases may interfere with informed decision making (4, 5). What should a physician do? Understanding how and why patients make certain decisions can be used to help them make better decisions, fulfilling physicians professional obligations of beneficence while still respecting autonomy. The idea of choice architecture is that arrangement of the context of choice matters and cannot be avoided (4). Thus, it should be done reflectively and responsibly. A physician treating the man with low-risk prostate cancer may recognize a commission bias (a drive to act, even with worse outcomes) steering him toward surgery and may counter it by using framing effects to discuss the risks of immediate treatment first (information heard first is more influential) and in a manner that makes them vivid (for example, through a narrative rather than statistics). As a second example, the manual recommends that physicians disclose potential conflicts of interest to patients, such as ownership in imaging facilities and ambulatory surgical centers where patients with prostate cancer receive care. But behavioral science has demonstrated a backfire effect, whereby disclosure results in patients trusting physicians more rather than increasing their scrutiny (6). The value of disclosure may not be to inform individual patients or research subjects, but rather to provide evidence to change professional norms and eliminate unacceptable conflicts of interest. Acknowledgment of Controversial Issues The manual could do more to acknowledge and analyze controversial topics where disagreements among physicians, ethicists, and the general public are common. Instead of futile treatments, critical care professional societies frame the issue as potentially inappropriate treatments and recommend a fair process of conflict resolution rather than unilateral decisions by physicians (7). Other controversial topics include brain death (8), physician aid in dying (9), and care of unauthorized immigrants (10). Acknowledging such disagreements is an important first step in respecting patients and colleagues and practicing in a culture where many issues are deeply contested. In sum, the American College of Physicians Ethics Manual is a tremendous resource to physicians. As professionals who are dedicated to lifelong learning, physicians should build on the manual to address the ethical issues confronting them in clinical practice.