Holly Prigerson, a sociologist, is the Irving Sherwood Wright Professor in Geriatrics and co-director of the Cornell Center for Research on End-of-Life Care at Weill Cornell Medicine. Valerie Reyna is the Lois and Melvin Tukman Professor of Human Development in the College of Human Ecology and is director of Cornell’s Human Neuroscience Institute and co-director of the Center for Behavioral Economics and Decision Research.
Together, they have combined forces to study end-of-life communication between patients and their physicians and clinicians, with the goal of improved prognostic understanding and decision-making and, ultimately, better end-of-life decisions. They have received a Cornell intercampus seed grant that has sparked additional collaborations and new NIH-funded studies.
How did you begin working together?
Prigerson: My husband, Paul Maciejewski, and I had moved from Boston to Weill Cornell Medicine to co-direct the new center for research on end-of-life care. Our results demonstrated that patients and family members had little understanding of their prognoses – the course of their illnesses, how sick they were and, bluntly, how close to death they were. It pointed to the need for a new approach to improve medical communication.
We were searching for a theoretical basis for our empirical approach, and Paul found one of Valerie’s articles. We thought there might be insights for medical decision-making, so we sought Valerie out.
Reyna: And I’m so glad you did. Holly is really an expert on these end-of-life issues, and she and Paul have done some tremendous work in a clinical context. I’ve done research on things like cardiovascular decision-making by doctors in the emergency room and patient decision-making around a number of things like arthritis medication.
Valuable resources are expended for health care during the end of life, sometimes to achieve goals that patients would not want if they understood their options. What we’re about, really, is not constraining those resources; what we’re taking is a choice approach, a decision-making approach, so that patients have the ability to make informed choices about their own lives. That’s why the decision-making component is so important, empowering people to have the information to make those informed decisions.
Prigerson: Applying what Valerie’s done in her work and her insights, we believe, could transform medical communication. For example, oncology programs have been looking at ways for physicians to improve empathic communication. We examined feelings as integral to improving communication not only to address emotion but to improve understanding about one’s illness and treatment options.
Reyna: The key way to do that is to focus on “getting the gist” – and it’s not an accident that “gist” also works perfectly as an acronym for “get information strategically and transparently.”
Prigerson: It’s not just about using simple words, and it’s not just responding to emotion. For the patient, it’s “What is the meaning of this for me and for my values and preferences?” Everyone wants a cure, and we understand that. We’re not trying to disparage hope. The vast majority of patients we’ve studied want to know their prognosis, they want to hear it from their oncologists, and they want their oncologists to talk to them in ways that they understand.
How does your approach play out for doctors and patients?
Reyna: So doctors are trying to give a lot of detail, and patients don’t necessarily have the background knowledge to absorb that detail. Lost in the detail is that really important bottom line – for example, that your latest scans suggest that you have metastasized cancer, and your prognosis is likely months, not years.
“Months, not years” was something that Holly and her team examined in prior studies and found to be important to end-stage decision-making. We discussed this and we came to the conclusion that’s really the gist sometimes of, say, a stage 4 diagnosis with scans that show metastasis. So, that’s the essence of it. It’s literally a phrase. To capture where the patient is.
Often, people feel you need to give the numbers to the patient so that the patient can decide. I think that’s probably a good idea. But numbers, just like words, are not meaningful by themselves. Statistics about outcomes and probabilities are important, but they have to be interpreted, just like words have to be interpreted.
Say, for example, I said you had a 20% chance of prostate cancer. Do you feel informed? What does that 20% mean? Should you be relieved? Is that below the mean? Or, should you be really, really worried?
And something like “months, not years” really serves as the start of the real conversation.
Prigerson: We published some studies showing how well advanced cancer patients understood four basic facts: one, that they had incurable cancer; two, that they were terminally ill; three, that they were at a late-end stage of their illness; and four, that based on best estimates of recent test results, they had months, not years, to live. We created a scale based on those four areas to determine the patients’ prognostic understanding. The studies showed that following a discussion of scan results with their oncologists, only 5% of the patients had an accurate understanding of their prognoses.
We suggested that, rather than talk at length about tumor sizes and growth rates, oncologists should describe scan results as indicating whether their cancer is “better,” “worse” or “the same.” No numbers. They can add numbers later, and a lot of patients want detail and should get as much detail as they desire. But we’re concerned that details can get in the way of comprehension and that the way these details are presented can be misleading. What patients really need to know is, is the cancer growing on treatment, remaining stable or shrinking (getting better)?
If it’s the case that things are getting worse and they have months, not years, left to live, that can lead to a conversation because for those patients, “months, not years” means a different course of treatment than they might have thought if they were curable.
Reyna: Part of the reluctance to have these conversations, based on the work I’ve done with physicians, is that physicians don’t want to tell patients something that’s wrong, that’s overly pessimistic, that removes hope, and then it later turns out that it was wrong. That is an admirable motivation.
So the whole point of risk and uncertainty has to be taken into account. That’s one of the reasons why hyper-precision about prognosis is not favored by physicians – they realize that these treatment outcomes are uncertain. However, they do feel compelled to accurately inform patients so that they and their families can make plans about treatment and about the rest of their lives. So this “gist” approach allows you to do both of those things.
We just received funding from the National Institutes of Health to train oncologists in communicating the gist to advanced cancer patients. Our Oncolo-GIST study instructs oncologists to focus on basic bottom-line points related to the meaning of scan results for the patient (better, worse or stable), what those results mean for their prognosis based on data, and last, what this prognosis might mean for medical decision-making.
There have been many attempts to improve end-of-life communication with patients, but what makes this approach different is the grounding in evidence-based theory, which is about emphasizing meaning.
You’ve had several published studies already; what are some of the most interesting findings?
Prigerson: We found a few interesting things: one, which confirmed a vast literature on what is called depressive realism, was that depressed people can hear bad news pretty well. They’re almost anticipating it, and they find it validating. Depressed advanced cancer patients more accurately heard that their scan results were worse than the nondepressed. Those who were anxious heard it significantly less well, and they heard good news better than they heard bad news. For them, bad news was particularly hard to hear.
We also have shown gender differences in communication. Male oncologists were more likely to talk about prognoses and treatment options than females. However, it seemed the female oncologists had figured out which patients could hear it better and who couldn’t. The rate of talking about prognoses and curability in these discussions was threefold higher in the male oncologists than the female oncologists, the rate of patients accurately hearing what was discussed in the conversations was threefold higher for those with female oncologists.
Strong religious beliefs can also influence information processing because the patient might say, “My doctors are great, I trust them. They’re very educated, they treat me well; I respect them. But it’s really not up to them. It’s up to God when I go.” We respect that, and our theoretical approach actually makes it easier for people to connect medical information to their values.
Reyna: Yes – we also have an additional part of our theory that has to do with moral and social values: What’s important to you? For example, if someone says, “The most important thing to me is to be able to take care of my family,” that would be a gist-based value. Or, “The most important thing for me is to be in this trial so that I can leave a legacy of hope, because I know that I’m going to die. But this is my contribution, and it matters to me to help others.”
One of the things we’ve discovered in looking at different cultural groups, African Americans, Latinos and other people is that we have a lot of common values, but each individual does not have the exact same order of priority.
What have you learned personally through this work together and from each other?
Reyna: Instead of doing more work on things that I’ve done in the past, it helps me really think in new ways and form new partnerships. Our work and this whole new crop of collaborations that are leading to new studies are connected to the seed funding we received from the Office of the Vice Provost for Academic Integration.
Holly is an inspiration, and I think we’re very complementary in our approaches. It is good to be reminded, in such an evidence-based way, and in such an empathetic way, of some of the downsides of being aggressive in treatment, and I think that’s important. It’s always part of what we inform patients about, but really, I think the two of us provide a kind of a very balanced and complementary perspective on these end-of-life decisions.
Prigerson: To the limited extent that medical researchers apply psychological science to clinical care, it is often without the direct involvement of the scholar who developed those ideas. But we have been able to work with Valerie from the outset; we have remained true to her conceptual framework, applied it, and now we are eager to see its impact. If all goes according to plan, we should be able to produce a large family of studies that can apply a new and improved approach to medical communication.