00:00 Joseph Betancourt: I think it’s important, very important to understand… This is Joseph Betancourt, it’s very important to understand that systems and providers and patients all relate to each other. It’s very difficult to distinguish specifically system-levels aspects to provider-level aspects to patient-level aspects. And, in fact, we see that systems have a great impact on the way healthcare providers are able to effectively communicate with their patients. We worked very hard to distinguish between words such as “bias” and “stereotyping.” And the evidence led us to the conclusion that in the current healthcare setting, the standard practice in which there’s a medical encounter, there’s three factors that are impinging on that medical encounter due to our healthcare systems.
00:47 JB: We are oftentimes multitasking, so a patient with multiple conditions, and we’re doing a variety of things at once. We’re under time constraints, so productivity requires us to see X amount of patients per afternoon, so we have a limited amount of time there. And oftentimes, we’re stressed due to these productivity constraints and a variety of issues. These three factors we found, from the social… Cognitive literature and social psychology, really contribute to stereotyping. Now, what is stereotyping? Stereotyping is, in fact, taking certain perceptions about groups and applying them to the individual. What we were very interested to learn was that stereotyping is a normal, cognitive process. When you have a bunch of information and you’re involved in trying to put things in boxes to help you think… And in medicine, clearly where you’re doing a lot of these different things all at once, you have to put things in boxes, you may apply them to individuals in ways that might contribute to disparate treatment.
01:47 JB: You may have certain assumptions about that patient, about their ability to comply, to adhere to a certain recommendation. You may be very, very well-intentioned, but these factors may make you susceptible to stereotyping without you even knowing it occurs. And so that’s the way we’ve defined stereotyping. And we think that it’s something that healthcare providers, when they’re able to reflect a little bit on their daily practice, will clearly understand that they may be susceptible to this.
02:14 JB: As it relates to the issue of clinical uncertainty, I’ll give you a very clear example. We care for very diverse patient populations who perhaps have different health beliefs when they come to us. We may not have been taught how to best elucidate that patient’s health belief, how to best understand that patient if they’re speaking a different language. Those uncertainties as well may lead us to treating people differently. Again, among the well-intentioned and, again, without outward knowledge. And that’s the area that I think we focused on there.
02:45 David Williams: I would like to add, this is David Williams, that one of the issues the committee dealt with as we looked at the evidence and psychology from many research studies that have been done on the nature of stereotyping and bias, that is, the stereotyping process is both an unconscious process and an automatic process. It means that many individuals who do not personally express prejudiced attitudes, and who believe in principles of equality, and are committed to egalitarian values will nonetheless, because they hold negative stereotypes, unconsciously treat patients who fit those stereotypes differently. And that’s why I’m saying we cannot rely on the good intentions of healthcare providers since these processes will occur even in the presence of good intentions.
03:37 Martha Hill: This is Martha Hill. I think this is an area in which we try to highlight the importance of the need for research. And the research needs to be on the process of the interaction around patient participation in decision-making and how providers prevent patients with the conditions that they have, and the alternatives for treatment and assessing risks and benefits, and then incorporating patient and family preferences into all of that. Most of the literature to date has been done by audiotaping where there’s an… Different analytic techniques that can be applied, for example, to how often does the provider interrupt the patient? And the average time after asking a closed-ended question is 18 seconds.
04:15 MH: Now, this is very helpful in understanding some aspects of the dynamic, but we need much richer, more qualitative research, and research that will include videotaping and looking at triangulation and a whole series of communication issues. So the communication sciences and the decision-making sciences and the behavioral sciences need to be brought to understanding the process of these interactions. Because you’re right, we don’t know the answer about the relative contribution of these different factors at this time.
04:44 JB: Joseph Betancourt. Just to add, again, it must be understood this occurs within the context of the healthcare system. So it needs to be understood that these different factors fit into the way the systems in which we practice, and so that’s why we stress looking at systems and the medical encounter.
05:00 Speaker 4: Okay, let’s go over to this side of the room.
05:02 Peggy Eastman: Okay. I’m Peggy Eastman with Oncology Times, and I have two major questions. First, there was a suggestion that sometimes minorities refuse treatment, and I’m wondering if you found that this could possibly be due to the way in which those treatments are communicated. In other words, there might be some feeling of fear or some feeling that, “Maybe this is not really the best treatment for me as a minority person.” And so that might be why the treatment is refused. There have been some incidents, as a matter of fact, in the District of Columbia, that suggest that… Regarding some immunization programs, for example, which were well-intentioned but went astray. The second question is, could you give us an example of a provider incentive that might promote disparities in care? Dr. Nelson, you mentioned the need to limit provider incentives that may promote disparities, so I’m just wondering if you could give us an example of that kind of provider incentive.
06:09 Risa Lavizzo-Mourey: Let me begin. This is Risa Lavizzo-Mourey. You are quite right. We found that, in a small number of studies, there was a difference in the refusal rate among minority patients and non-minority patients. However, if you look at those findings, it’s very clear that much of the time the minority patients will say that they did not fully understand the treatment choices, and that that may have contributed to their refusing the treatment. And in addition, we know from our reading of the social and psychological literature, that the way questions are posed can have an impact. So I think that the key finding that this does not fully explain the disparities and that there may be some uncertainty in the patient’s mind is also important. On your second question, and then I’ll let Dr. Nelson also respond, one factor that actually has already been referred to is the amount of time that a provider has to interact with the patient. Under stressful conditions, as is created when you’re in a very time-limited situation, these kinds of stereotyping can be exacerbated.
07:29 Alan Nelson: I think that… This is Alan Nelson, I think that Risa provided the answer, that where productivity incentives may shorten the amount of time that is available for a better understanding between the patient and the physician, we should go in the direction of providing incentives that may reduce incentives not tend to create them.
07:55 MH: Martha Hill. To follow-up on the first question, I think we have a series of recommendations that speak to both patient education and empowerment. And I would include families in that, particularly in some of the cultural subgroups in the country where family participation in the medical encounter and in the decision-making is very, very important. And also in our cross-cultural education of health professionals, we need to much better understand communication and how people make decisions so that we can better appreciate what is thought to be refusal of treatment, it may in fact be a need for more time in decision-making or a different way of reaching that decision. So we recognize that this is an area of important work that needs to be done.
08:40 PE: Just a very quick follow-up. Does the system tend to favor somebody who is more educated and more articulate and a better communicator, as a patient?
08:51 RL: The studies that we reviewed are very clear in showing that physicians and other healthcare providers are more comfortable interacting with people that are like themselves, that is, highly educated, articulate, and that when you ask them to, in an experimental design, reflect on characteristics of patients, they often show that in their answers.
09:21 PE: Thank you.
09:22 S4: Okay, we have some questions from our web audience that were emailed in.
09:24 Craig: We have a question from a web listener, John Hodgson, who’s a reporter with Nature Biotechnology. He asks, “Can you explain how you know that the outcome differentials do not segregate better with poverty than with race? Poverty may be easier to define precisely than race, given the rather political and social definitions of race.”
09:46 DW: I would say that we are very well aware that there are large racial disparities in health status in the United States, and we’ve documented these racial disparities in care. We do know that it’s a complex process, and that race and socioeconomic status or economic status do go together. Our charge was to look at, “To what extent are there racial disparities in the quality of care received after persons have entered into access?” Several of the studies adjusted for the quality of insurance that individuals had, as well as adjusted for education or income as measures of proxies for their economic status, and we still find these disparities exist. So I think the data is quite overwhelming, that there is an independent contribution of race and ethnic status to these disparities, although part of the complex, multifactorial context in which they occur is the lower levels of economic status among minority groups.
10:44 S4: I’m sure you recognize that distinctive voice of Dr. Williams.
[laughter]
10:49 S4: Let’s take one more question from the email audience. Craig, do you have another? Okay, Sheryl?
10:55 Sheryl: Yeah, I’d like to go back to the issue of bias. You say in the report that this exists in the larger context of American society, where racial and ethnic biases and stereotypes persist. So is there any evidence that in the healthcare profession these biases are more prevalent or more pronounced? And also, is there any suggestion that one of the problems is a lack of minority physicians?
11:25 AN: Let me… I’ll start. No, we found no evidence that these attitudes are more prevalent within the community of caregivers. As a matter of fact, without solid evidence we reached the conclusion that people in the healthcare system are members of our society and reflect many of the same attitudes and biases that exist in society. Parenthetically, we reviewed a study from the UK that shows that stereotypical behavior, bias and prejudice aren’t unique to our country, that it exists in other countries as well, although we didn’t do an international review. The second part of your question?
12:13 Sheryl: The second part was, is it a problem that there are too few minority physicians?
12:20 RL: Go ahead.
12:21 DW: I wanted to add to that. Let me say a comment on the first part of the question. This is David Williams speaking. Yes, there are very careful audit studies done in the areas of employment, in the area of housing, that demonstrates that there is systematic and pervasive discrimination that still exists in our contemporary society today. We don’t have that same kind of evidence in the healthcare field, but it suggests that the patterns observed in the healthcare field are not inconsistent with what occurs in other sectors of society. In terms of minority healthcare professionals, that is certainly an area where we have made an explicit recommendation. We do not know, from our evidence base today, whether these disparities are less likely to occur among minority healthcare providers. But we do not two things: One is that patient satisfaction tends to be higher when there’s a provider that is similar to them; and we also know that minority healthcare providers are more likely to work in underserved areas.
13:23 DW: We also know that it’s very important that we have not had much success as a society in increasing the proportion of minority healthcare providers. So for example, in 1968, 3.5% of American physicians were black or African-American. In 1999, 3.9% of American physicians were black. So we have made progress but minimal progress in increasing the number of under-represented minorities in medicine and other healthcare professions, and that is certainly one of the recommendations that we make.
13:51 Sheryl: And just a short follow-up, I wonder if you think that the issue of bias is maybe more pressing in the healthcare field. If you get turned down for a job or an apartment, it’s not a life or death matter, but in healthcare it might be. Is that the committee’s view?
14:08 RL: I think one of our initial statements and one that we feel very strongly about is that in the healthcare arena, there is a moral imperative to correct this. I think that we know that there are disparities in the health status of minority patients, often related to these same conditions where we have been able to find clear evidence of disparities and unequal treatment. And it is critical, therefore, that the healthcare providers, systems, and stakeholders take this on.
14:47 AN: Very briefly, Alan Nelson, it was clear that the streams of evidence that led the committee to conclude that bias and stereotyping may contribute to disparities, those streams of evidence were indirect. That in housing and employment, testers can be sent in and directly measure the degree to which those attitudes occur. Obviously, for ethical reasons, you can’t have a patient go into a hospital and have an operation to see if they’re treated differently based on their ethnic group.
15:25 S4: Okay, let’s go to this side of the room.
15:26 Rachel Jones: Rachel Jones with National Public Radio. I’d like to get you to revisit this issue of public awareness, and the need for the public to know about this information and, I would assume, to act on it in some way. I think that when these sorts of stories come out, you’re getting the same sort of reaction among the public in terms of drawing conclusions about why these disparities exist based on their own stereotypes, and not many of them are going to read a 500-page report and try to get at the root of why this. So if you had to speak to people in a non-clinical, across-the-kitchen-table kind of way and explain to them why they need to know this information, understand what’s going on and what their role in solving it might be, could you let me know what that is?
16:14 MH: This is Martha Hill. Yes, I would say that we want people to understand that they need to become informed and to advocate for themselves, their family members, and their friends in a way that says, “We need the information and the facts. We need to be heard, and we need to hear, and to understand and participate in decision-making about care so that it is an informed decision with time and cultural competency.” So it’s in terms of the language and the vocabulary, and as best possible, an understanding of the risks and the benefits and the extent to which preferences can be incorporated into that decision-making. It’s in a sense, people need to own this and take it upon themselves. And then in an assertive way, get the information and have some influence over the process in which the care decisions are made. It’s gonna take all of us to do this, it can’t just be on the side of the system, on the side of the providers, or on the patients, it’s all of us together.
17:15 RJ: I just wanted to get you to expand on that, maybe several of the other ones could, because I think in this atmosphere, that’s backlash against political correctness or multiculturalism or whatever, I think that there’s a tendency to hear these sorts of stories and go, “Not again,” or, “Why are we still hearing this sorta thing?” Because there is maybe the perception that these gaps don’t exist among some people or, again, they’re gonna use stereotypes to sort of process what they’re hearing.
17:45 MH: I’m sorry. You’re asking me for a more succinct response?
17:49 RJ: No, I was hoping one of the other ones might speak to it.
17:50 RL: I think it was just a comment. Yeah.
17:50 AN: You wanna comment?
17:51 DW: Yeah, I thought it was just a comment.
17:53 S4: Okay, let’s go, sir.
17:54 Gary Dennis: Yes, I’m Gary Dennis. First of all, I’d like to thank you for doing such an outstanding job. I was the President of the National Medical Association when Congress actually funded this study, and I lobbied for it. And I would say you are a distinguished group, you are scholars, you are intellectuals, and you have really done an outstanding job. I thank you very much. What I’d like to do, though, is comment on a couple of your recommendations. First of all, recommendation 4-1. Even though there seems to be an overwhelming body of evidence that you’ve accumulated, that healthcare disparities are really contributed to by racial bias, your recommendation 4-1 uses the word “may” instead of “do,” and I just wanna know why you decided to go soft on that after you’ve compiled such a powerful report.
18:48 GD: The second is that you mentioned the use of accrediting bodies to help reveal some of the data. In fact, there was a paper that was published last week by a group of… Written by a group of researchers from Harvard, using HEDIS data, which looked at the Medicare Plus Choice population, and was very revealing in terms of healthcare disparities even in a situation like that. So clearly, there is a value to it, but why haven’t we made recommendations under either 5-1, where we deal with the shortage of healthcare providers, or 6-1, where we deal with the transcultural education process? Especially for Dr. Nelson, I think that maybe we go to the accrediting bodies, like LCME and ACGME, to make sure that we are actually addressing racism and include members on those bodies that will be more independent and may actually be reviewing that at those institutions and programs because there seems to be a lot of stagnation in terms of increasing the number of healthcare professionals…
19:55 S4: Dr. Nelson, do you have an answer to the first or one of the…
19:58 AN: The why the use of the “may contribute to disparities,” it simply reflects the fact that the lines of evidence, while they were very powerful, were by and large indirect. And the second part of that recommendation speaks to the need for more focused research that will help answer that more directly. Yet I think I can speak for the committee in saying that the committee acknowledged that the evidence didn’t always completely… Let’s see. There was not enough evidence that acknowledged the other variables in this complex environment for us to be as precise as we would be if we didn’t have a qualifier in there. The accrediting bodies, LCME and others, we did speak to the Joint Commission and NCQA, and my guess is that the accrediting bodies for the education process will be very aware of our report, and I imagine that they will put it on their agenda. Certainly, we can write a letter that lets them know about your comment.
21:09 S4: Dr. Betancourt.
21:10 JB: Yes. Dr. Betancourt. Just a quick add-on. In fact, the LCME has already moved forward in terms of an accreditation standard for medical schools. Brief, but we’re trying to expand on it now, that says that medical schools should begin teaching cross-cultural education and understanding of bias and discrimination, not only based on race, but based on gender and some of these other factors. So the LCME has already stepped up. We think that the ACGME will follow. The Association of American Medical Colleges have been very active in this area. So I think that we’re developing a groundswell here, and we hope that our findings will add to that…
