 Michael Green, a physician and a bioethicist at the Penn State University Milton Hershey Medical Center. He's chair of the hospital ethics committee and director of the program in bioethics there. In his current position, he cares for patients, teaches medical students in residence, conducts research in bioethics. His research has most recently focused on helping patients to make informed decisions with the help of interactive computer-based decision aids. And that's what he's going to be talking to us about today. So to the podium, welcome Michael. Thank you. Thank you much. It's great to be back here in Chicago at the McLean Center. I want to just start by acknowledging my collaborators on this research, as well as funding that we received from the NIH, as well as American Cancer Society. As everyone in this audience knows, planning for the future is really quite important. And when people don't plan for the future, all sorts of bad things can happen to people, including the receipt of unwanted treatment, unnecessary costs that result from that. People suffer needlessly. There's burdens to family members and even contentious battles between family members. And so we advocate that people fill out advanced directives, but we all know that advanced directives have serious limitations and have been shown in many studies not to accomplish many of the goals that they were intended to accomplish. And they often just don't, they often don't work. We know, for example, that only about a quarter of adults tend to fill out advanced directives, that physicians hesitate to have discussions with patients about these issues because of a lot of reasons, including the fact that having these discussions are time-consuming and they worry that patients lack the knowledge to make sufficiently informed decisions to be meaningful. And particularly in the oncology community, but not only there, there's worry that having these kinds of discussions can diminish hope and increase patients' anxiety so they avoid having the discussions. Now, the alternative to having advanced care planning discussions, of course, is to rely on surrogates to decide. We just learned something about how they go about doing that. But relying on surrogates to make these decisions is problematic. Studies have shown that others can accurately predict patients' wishes only about two-thirds of the time and often get it entirely wrong. So we need a new and better way to have advanced care planning kinds of encounters. And this led us to develop an interactive computer program to fill in some of these gaps and address some of the limitations. So we've been working on this for almost a decade now. The program we developed is an online decision aid. It's called Making Your Wishes Known, Planning Your Medical Future. It was designed over many years with an interdisciplinary group of physicians and nurses and geriatricians and graphic artists and decision analysts. And it's very educational. The program is designed to help people choose a surrogate decision maker. It helps them explore their values and preferences, learn about their medical options and articulate their wishes and discuss and reflect upon their choices in a way that's meaningful. The program includes clinical vignettes and patient testimonials, an example of which is shown here. And perhaps the most innovative aspect of the program is the use of a decision aid that helps people to explore their values. This is based on something called multi-attribute utility theory. And it prompts users to ask, to answer a variety of different questions about their goals and values and preferences. And then it uses the answers to rank and rate the importance of various things for their decision making. And then the computer generates a set of directives that is consistent logically with what people say is important to them. And then the people have an opportunity to say, well, the computer got it wrong or they got it right, but at least it's a starting point for figuring out what it is that they would want. And the program is very educationally oriented. It's sort of designed to emulate what an ideal encounter between a doctor would be in a world where doctors had unlimited time and lots of knowledge, which they often don't. So the question arises, of course, is this effective? Does the thing work? And we've done a number of studies now to try to answer that question. Some of the earlier work we did was using this program with a set of medical students. We did a randomized controlled trial with a class of 150 medical students, where half of them used our computer program, half of them used a standard set of advanced care planning tools, and we measured a bunch of outcomes, and those randomly assigned to the intervention group outperformed the other group across all outcomes that we looked at. We also done a variety of feasibility studies with all these different groups of people outlined here. And we found that the satisfaction with the computer program was quite high on a scale of one to five. They were very satisfied with how it increased their knowledge, provided information, clarified their values, helped them make decisions, put their wishes into words, helped them be prepared to have discussions with families and doctors. And overall, satisfaction scores were high. They rated that the program was very accurate in representing their wishes and was little to no burden. We also did some reliability studies, one of which involved 24 people who completed the program twice, separated by four to six weeks. And we looked at three different aspects of reliability, and there was perfect correlation in terms of the general wishes that the program reflected, but there was much less correlation with very specific wishes about treatment options that they would want or not want, and it was sort of in the middle with regard to quality of life issues. And all this is sort of a preamble to what I wanted to talk to you about, which is the results of some research that we did in an American Cancer Society study, which was a randomized controlled trial of making your wishes known versus standard advanced care planning materials among patients who had advanced cancer. And these are people who had life expectancies less than two years and were not depressed. And we had hypothesized that compared to standard advanced care planning use of the computer program with patients who had advanced cancer would result in greater knowledge without a decline in hope or an increase in anxiety and greater sense of self-determination as well as greater satisfaction. And we looked at pre and post as well as between group changes and these outcomes here, knowledge, hope, hopelessness, anxiety and self-determination. You know, and one of the reasons for this hope and anxiety is what I said before about physicians often using that as a rationale for not having the discussions and nobody actually had looked empirically as whether these discussions had an impact on hope and anxiety. And then post intervention we looked at satisfaction. So the intervention group used our computer program, the control group used educational materials and a standard living will form that we actually chose from among the best in the country as rated by the report on dying in America. And then we did some statistical analyses, which I won't get into. But this is sort of the flowchart of the study. And one of the interesting things about it is that doing this kind of research with very ill patients is really quite challenging. And we recruit, we randomized 200 people into the study and in order to get 200 participants we started with 2,000 referrals. And that's just what it takes to do this research. It's very labor intensive and time consuming and that's what it took. And then we randomized 100 in each group and we had almost everyone complete the project. Now we're not done yet. We're actually following all these people until death and then we're going to do all sorts of measures after patients die in terms of what it is that they receive in terms of medical care and whether their wishes were followed and what the doctors knew and all that. But only about half of the patients have died so far so we've got a couple more years to follow these people. So we're going to report on some more short term outcomes. But what we found is of the 200 people who were enrolled, mean age was 60, most were male, white and computer literate and there was no difference between the groups. In terms of knowledge, we had a 27 item knowledge test. They got about half of the questions right at baseline. After using the decision aid as well as the other materials, their knowledge increased in a statistically significant manner but the change in knowledge was significantly greater in the decision aid group than it was in the standard group. Anxiety was surprisingly low at baseline for both groups despite the fact that these people had severe illness and had a very poor prognosis. Their baseline level of anxiety was low and after engaging in advanced care planning discussions in either group, the anxiety did not go up and in fact it went down a bit in the decision aid group. It stayed the same in the standard group and the change in anxiety was greater in the decision aid group than in the standard group. Hopfulness, despite being very ill, people had pretty high levels of hope at baseline and their hopefulness did not change after engaging in these very difficult discussions about their end of life wishes as we had hypothesized and there was no difference between the groups. We also looked at hopelessness which is a similar construct to hopefulness and people did not have a lot of hopelessness which is good and it didn't change after having these conversations in either group. Self-determination was looking at how much people felt empowered to affect the kind of treatments that they would receive at the end of life and people felt moderately self-determined and it went up in a statistically significant manner after either intervention whether that's clinically significant I'm not quite sure but it did go up and the change in self-determination scores was no difference between the groups and then we looked at satisfaction with the advanced care planning method that they engaged in and we found that they were more satisfied in the decision aid group than in the standard group and that was statistically significant. These results were not affected by age or gender or self-reported health or prior computer use or prior completion of advanced directive or performance status and so despite the fact that this has some limitations I mean this is only one study site it's a population of limited ethnic diversity there are many people who decline participation and unfortunately we don't know if their wishes are different than the people who did participate and this is a narrow set of outcomes despite this we have some conclusions that compared to standard advanced care planning patients with advanced cancer who use the decision aid have greater increases in knowledge no adverse changes in anxiety and hope they express greater self-determination and are more satisfied and in ongoing research we want to continue to investigate this population to see whether use of the decision aid actually has an impact on the medical care that people receive at the end of life which is really what matters and whether it increases the likelihood that their wishes are known by their physicians as well as followed and then I'm also pleased that we just got a five-year R01 grant from the NIH to study the impact of this decision aid on family caregivers and we'll be doing that over the next number of years if you're interested in checking this out you can find the program online it's at www.makingyourwishesknown.com and I will stop there and take questions later