In recent years, skepticism about research has become ever more pronounced and ethical standards for research ever more ambitious. A visitor to America’s institutional review boards soon discovers that their members have become concerned about increasingly slight and subtle threats to research subjects’ autonomy. Do subjects truly, truly understand what they’re agreeing to? Have they truly, truly been unswayed by anything that could possibly be called pressure?
Standards for informed consent to medical treatment have also become more exacting. The scope of legal standards for informed consent has expanded. Ethicists are realizing that if patients are really to make their own medical decisions, they will need more than the “just the facts, ma’am” that ethicists and courts once imagined. And even as it is becoming clear that true autonomy is elusive, ethicists increasingly feel that patients not only have a right to make their own decisions, but have duty to do so as well.1 And why not? Don’t free people want to control their own lives, to decide for themselves what sort of people they want to be, and thus to make decisions — not least medical decisions — that affect them?
These developments reflect a larger confidence in American life and law: If you tell people enough about their situation, they will analyze that information and use it to make their own decisions. The Miranda rule, for example, requires that criminal suspects be read their rights so that they can decide whether to waive them. Product-liability law insists that consumers be warned of a product’s risks so that they can decide whether to buy it and how to use it. And borrowers must be told what interest rates really are.
And what could be wrong with these plain examples of social and moral progress? Perhaps nothing — except that there is so much evidence that so many of these programs do not work as they are supposed to. Even the defenders of the Miranda warning contend that most suspects ignore it. Notices about products’ risks go largely unread and unheeded. Few people actually use loan disclosures.
Nor has informed consent fared wonderfully better. Consent to research has been bedeviled by, for example, the persistent failure of subjects to understand, even when they are told, that phase I studies do not test the efficacy of a drug, only its toxicity. In clinical situations, even intensive and expensive efforts to inform patients regularly disappoint.
In the real world of limited resources, informed consent withers. For example, Braddock et al2 examined transcripts of discussions between physicians and patients, looking particularly at (1) the patient’s role in decision-making, (2) the nature of the decision, (3) alternatives, (4) benefits and risks of the alternatives, (5) uncertainties associated with the decision, (6) the patient’s understanding of the decision, and (7) the patient’s preferences. They concluded that “the completeness of informed decision making was low … Few decisions (9.0%) met criteria for completeness of informed decision making. Completeness of discussion of decisions varied by decision complexity. Whereas 17.2% of basic decisions were complete, none of the intermediate and only 1 (0.5%) of the complex decisions were complete.”
The conventional response to such data is to say, “Try harder.” But when a reform has long been widely advocated and still falls short, we should ask why reform is unachieved and whether hopes for it are realistic. Reforms often fail, even when the goal is worthy and the people implementing it want it. (How long have we been seeking prisons that rehabilitate criminals?) In fact, conventional continuing medical education (eg, lectures) regularly leaves physicians unchanged, and even more sophisticated programs to persuade physicians to adapt quite sensible and straightforward improvements in their practices frequently have no measurable effect.
Even if physicians devoted themselves lavishly to educating patients, many patients would spurn the gift of autonomy courts and ethicists want to give them. Why? First, people ignore disclosures because they can’t read or read poorly. People think that they already know what they need to know or have better ways to spend their time. People misinterpret what they read because they see it against a background of misinformation and ill-founded theories. Detailed disclosures are impenetrable; simple disclosures are misleading.
But there is an even deeper reason informed consent has disappointed its advocates: Many patients do not wish to make their own medical decisions. However improbable this may seem, the evidence is overwhelming. Consider this example from many: Ende et al3 asked patients whether they wanted information. They did. On a scale on which 100 is “Tell me everything” and 0 is “Tell me nothing,” the average response was 88. So far, so good. But Ende and his colleagues also asked patients whether they wanted to make their own medical decisions. On a scale on which 100 is “I want to make them all” and 0 is “I don’t want to make any,” the average answer was 33. What’s more, the sicker (and, in some studies, the older) people are, the less inclined they were to make decisions.
How can patients be so passive? Patients often have excellent reasons for declining to seize the gift of autonomy. For one thing, people avoid troubling decisions, and medical decisions, because they demand perilous and complex choices based on uncertain data, are especially repellent. What is more, we sense that we make decisions badly. We systematically misperceive reality; we systematically fall prey to failures in reasoning. These problems of perception are exacerbated by the special difficulties of medical decisions. Few of us know how our bodies are built, how they work, how they break, and how they can be fixed. Because we are so systematically ignorant, it is hard to teach us enough to make our own decisions.
Proponents of informed consent shrug off this problem by supposing it can be solved by “educating” patients better. They shouldn’t. I know they shouldn’t because I am a teacher. Every semester I try to teach material no harder than the material doctors teach patients. My students have all excelled in tests of general intelligence and aptitude to learn law. They have nothing to do but study. Even during the first year of law school, the pressures on them are trivial compared with those endured by patients. I have been striving to teach well for many years, but as I read my exams, I am regularly chagrined at how easily even straightforward facts and black-letter law can be mangled in the learning. I always emerge with a fresh and chastened awareness that teaching and learning are humblingly difficult.
But this problem of conveying information is only the start. For example, even if you have a remarkable thirst for facts and uncommon powers of learning, sickness might still disincline you to make your own medical decisions. Sickness enfeebles you. It drains the energy you need to learn enough to make good decisions, and it saps the confidence hard choices demand. When you have the flu, do you want to be empowered to make decisions, or do you want to crawl into bed and have done with the world? For me, the latter course every time.
Once informed, patients are supposed to ask which treatments best promote their “values.” This assumes that patients have relevant values, some set of beliefs that is coherent and considered enough that a decision can be deduced from it. But people have better things to do than formulate principles for problems they hope will never arise. In fact, for most of us much of the time, we find out what we value by observing what we chose. It is hardly too much to say that our “values” are the explanations we give for our choices, not their source.
It’s not just that people’s values are often undeveloped. It’s that people have trouble predicting even relatively simple preferences, have trouble anticipating what they will want and how they will react to what they get. People mispredict (to keep the list short) which snacks they will want over the next three weeks, how happy election results will make them, how much better living in California instead of Michigan will be, how much personal criticism will wound them, and how painful visiting the dentist and other tormenters is.
How can this be? Consider a more consequential situation, that of people who have just become paraplegic. How accurately will they foresee their future? The question they will ask is “How will I live as a paraplegic?” This directs their minds where they will already be — on the misery of being paralyzed and all the things paralysis will deprive them of. They will not ask themselves what will be left in their lives that paralysis leaves unaltered. They will not ask how they will seek happiness in new activities. They will not realize that we all overestimate the intensity and duration of our emotions. These problems in anticipation help explain why patients generally think themselves happier than the healthy (including their doctors) think they are.
Even if all the problems I have so far described could be solved, patients would still need to assess both their information and their values carefully to make good decisions. Unhappily, the evidence we have suggests that this does not happen. First, patients frequently make decisions so quickly that they cannot be evaluating those factors accurately. Second, patients seem most commonly to seize on a single aspect of a problem and make it the basis for their choice. And third, once they have chosen, any new evidence is taken to confirm that choice.
Can things be this bleak? If people have so much trouble acquiring information, analyzing it, and using it to promote their goals, how do they stumble through their days? In fact, people can be quite shrewd about decisions they make frequently — but most consequential medical decisions are made infrequently.
In sum, the theory of informed consent in law and ethics is that patients and research subjects need thorough information so that they can reach an autonomous and wise decision. Courts, ethicists, and institutional review boards set increasingly stringent standards for evaluating the adequacy of that information. Yet the empiric evidence increasingly shows that patients are not making such decisions. I believe we are chasing a chimera if we suppose we can ever secure from patients and research subjects the truly informed consent that is now envisioned.
These thoughts are unfashionable. I think many scholars will agree about the data I’ve adduced, but they are loath to discuss them because they are so anxious not to return to the bad old days of medical paternalism and imperialism. I am old enough to remember those days, so I don’t want to resurrect them. Who does?
If the past worshiped false gods and the present’s gods have failed, where do we turn? Perhaps we need a skeptical reconsideration of informed consent. That reconsideration would demand honesty, realism, and modesty. We need to be realistic about how far the theory of informed consent can be implemented. We need to be modest enough to ask what patients want for themselves rather than what we want for them and not waste scarce resources on hopes that can only be disappointed. But we will set sensible goals for improving the lives of patients only if we confront the evidence before us with honesty.
FULL TEXT