Happy New Year to all our readers! Today marks the completion of 5 years of SBM and the beginning of year 6. My contributions, at one a week, have now reached a total of 260. My first post on this blog, 5 years ago, was a review of an important book about science and alternative medicine, Snake Oil Science. This year I’d like to start with an important book about communicating medical science to patients, Critical Decisions, by Peter A. Ubel, M.D.
I was wrong about informed consent. I thought informed consent was a matter of explaining the risks and benefits of treatments to patients so they could decide what they wanted to do. That was naïve, simplistic, and misguided. Ubel’s book has radically changed my thinking about how doctors should interact with patients.
Paternalism in medicine is dead. Patient autonomy rules. We respect the right of patients to determine their own treatments, even if their choices seem unwise to us. Patients should do what they want. But there’s a problem: patients may not know what they really want. Emotions and unconscious and irrational forces influence their medical decisions. Preferences can change from one moment to the next, and they can shift with subtle changes in how treatments are described and how the issues are framed. Doctors need to develop a better understanding of what is going on in their patients’ minds, of how the way they present treatment options can inadvertently influence patients, and of how they can participate with patients in a process of shared decision-making. It’s possible to provide direction without paternalism.
The biopsy shows prostate cancer. The doctor explains that the risk of death in this particular case is small and the options include watchful waiting instead of surgery. The patient may hear that as “I have cancer and you’re not going to do anything about it?!” Or be so alarmed (“Omigod, I’m going to die”) that he is unable to process any further information. He is told the risks of surgery, but he can’t imagine what it would really be like to live with impotence or incontinence. He may over- or under- estimate his risk of side effects and how they would affect his life. The doctor typically uses medical words that the patient doesn’t understand, and overwhelms him with abstruse details about Gleason scores and atypical cells and non-aggressive appearance on biopsy. The urologist often doesn’t recognize the emotional impact of the diagnosis and doesn’t respond to the patient’s feelings. He asks if the patient has any questions, and when the patient doesn’t ask any, he is reassured that the patient has understood. But the patient may simply be confused, scared to death, and unable to express his concerns.
What Would It Be Like?
Young people dread the limitations of old age but research shows that happiness actually tends to increase with age. People often think they would rather die than be paralyzed, go blind, etc. because they can’t imagine what it would really be like. They think they would be terribly unhappy. But in reality, people who face those challenges soon adapt to them and are just as happy as they were before. People mis-predict their own long-term emotional reactions to everything from losing a leg to winning the lottery. Winning millions makes them ecstatically happy at first, but as time passes, they end up no happier than people who didn’t play the lottery. The initial emotional impact of an injury or illness is strong, but it wanes over time. People underestimate their emotional resilience. Talking to others who have faced the same challenges may help them to judge more accurately what it would be like for them. In one anecdote in the book, a patient categorically rejects surgery involving a colostomy but doesn’t think it through to realize that he would rather live with a colostomy than die. Making a quick judgment based on an initial emotional reaction is unwise
Don’t Make Families Regret Their Decisions
Families may be reluctant to authorize “Do Not Resuscitate” orders when the doctor tells them the patient is dying and resuscitation would be futile. The current practice is to explain the situation to the family and let them (make them) decide. But Ubel says:
Resuscitating a dying patient is not the worst thing we doctors can do. Making a family regret their decision is far worse… in the vast majority of these conversations, my recommendation [for a DNR order] was met not with resistance but relief. They didn’t want their loved one to suffer, and they knew that CPR would only prolong their pain. But they didn’t want to be the ones who made that choice. My recommendation helped them by sharing the decision-making burden.
Experiments in Decision Psychology
Research has illuminated how irrational people are at making decisions. If they are trying to get a red jelly bean, they prefer to pick from a bowl of 100 jelly beans with 10 red ones than from a bowl of 10 with one red one. The odds are exactly the same, but somehow the larger bowl seems like it ought to offer more chances, even though they understand on a rational level that it doesn’t. In a hypothetical scenario where a virulent flu will kill 10% of the population and a vaccine will prevent infection but carries a 5% risk of death, people prefer inaction with a higher risk of death to taking an action that results in half the risk. On the other hand, if patients are asked to choose between watchful waiting for a cancer than has a 10% chance of eventually killing them and taking chemotherapy that will eliminate the cancer but kill 5% of patients from side effects, almost no one chooses to forgo the treatment that reduces the risk by half. Apparently there is some kind of unconscious bias against vaccines as compared to chemotherapy (?!). In a scenario where watchful waiting led to a 5% chance of death and surgery carried a 10% risk of death, most people preferred the surgery, preferring to take action even when that action was more likely to harm them. They just wanted that cancer out of their body.
When given a statistical prognosis like a 10% chance of survival, patients tend to think it doesn’t really apply to them, to think there is something about their case that ratchets their chance up to 20% or even more. Doctors, too, tend to over-estimate the chances of the patient’s survival in their unconscious desire for optimism.
The way information is framed is influential. Patients prefer an operation when it is presented as having a 90% survival rate to the same operation when presented as having a 10% mortality rate. And the power of a doctor’s recommendation is very strong. In another experiment, patients were told that the risks of a hypothetical vaccine outweighed the benefits. Then they were told that their doctor recommended it, but that their doctor didn’t know any more about the vaccine than they did. They chose the vaccine because the doctor recommended it, unconsciously feeling that the doctor knew more than they did even when they had been assured he didn’t.
Then there is the power of stories. For a surgery with a 75% success rate, telling one anecdote about a success and one about a failure seemed to cancel each other out, but 3 success anecdotes and one failure anecdote reinforced the 75% statistics and made people more willing to get the surgery, even though the anecdotes were about specific individuals and did not change the information about the actual percentage of people who would benefit.
Telling patients there is a 17% risk of side effects or showing them pie charts or graphs tends to confuse them. Pictographs with 100 squares and 17 filled in with a different color to represent side effects are far better at reminding them of the 83% who don’t get side effects.
These decision psychology experiments provide clues, but it is still necessary to test every approach. A number of decision aids have been developed, such as videos showing what it is like to live with a colostomy; some patients are reassured to see how real people have adapted and how little it has affected their lives, others are repulsed by seeing a colostomy stoma. We need to test whether each component of the decision aid does more harm or good. We want to inform, not influence.
Decision aids give patients the information they need, but they don’t always change behavior. Informed patients must also be prepared to interact successfully with their doctors. Ubel provides helpful tips and guidelines for patient empowerment and better joint decision-making.
What Does Patient Autonomy Really Mean?
Every doctor has been asked “What would you do if you were me?” “What would you do if it were your mother/wife/daughter?” “What do you recommend?” Sometimes patients really want the doctor to decide for them, either because they just don’t feel confident about their knowledge and judgment or because they don’t want to feel responsible if something goes wrong. Doctors are told not to make those decisions, but to give patients the relevant information and leave the decision to them. But here’s a thought: what if the patient autonomously decides he wants the physician to choose for him? Does that count? Can he delegate and still be autonomous? What are the legal implications? Would he have to sign over a power of attorney for health care to the physician? Would that even be legally acceptable? We need to grapple with this and other thorny issues.
Oliver Wendell Holmes said:
Man’s mind, once stretched by a new idea, never regains its original dimensions.
This book stretched my mind, and it feels good to know that I have learned something and have improved my way of thinking. Admitting that one has been wrong should not be a source of shame, but a source of pride because one is now more right. It is one of the great joys of being a critical thinker.
Patient autonomy is good, but shared decision-making is better. The goal of truly informed consent with no hint of bias or coercion is a worthy one, but it’s not enough. The doctor and patient should cooperate in a joint decision-making process that incorporates the doctor’s medical knowledge and the patient’s values and preferences. It’s difficult for a doctor to understand those values and preferences when the patient himself may not consciously understand them or be able to articulate them. If the doctor can develop a deep understanding of the patient’s values and preferences, then he can tell him what he thinks he should do without being paternalistic or coercive. It’s complicated, and our efforts to improve the situation must be tested with the scientific method to see if they are as effective as we would like to think. If doctors and patients could all read this thought-provoking book, we would be off to a good start. Highly recommended.
This article was originally published in the Science-Based Medicine blog.