Alan H. Morris talks with DXY.cn on his project of developing computerized decision making e-Protocol-heart failure with Chinese partners at the Third China Heart Failure Symposium in July 2011, at Dalian, China.
In the US and around the world, it is estimated that a significant percentage of the resources spent in healthcare is ineffective, because of suboptimal clinical decision making. This is partly because healthcare providers have only a limited ability to incorporate the tremendous amount of available knowledge in decision making, but the complexity of decision-making is increasing in particular in older patient with multiple co-morbidities. This mismatch contributes to the varying performance and suboptimal clinical decision making.
Dr Morris performs computerized decision-making research, which provides support to physicians. This helps them to make consistent decisions linked to best evidence, and enables reproducible clinical research and clinical care methods and results. These decision-making protocols go beyond clinical guidelines, because they provide detailed step-by-step support in the decision tree. While computerized decision-making protocols standardize clinician decisions, they are driven by patient specific data and produce patient specific decisions. They allow collection of evidenced based outcome data and enable rigorous comparative outcome research, which will be critical for effective allocation of healthcare resources.
Alan H. Morris教授在2011年7月大连召开的第三届中国心力衰竭论坛上，向丁香园讲述了其同大连医科大学第一附属医药合作开发的新项目—计算机辅助心衰治疗方案。
DXY: How long will this e-Protocol-heart failure program last for?
Alan H. Morris：Actually it’s hard to say because we are just getting started. The clinical rules may need 2-3 years to develop, but the whole project of heart failure will last much longer. It depends on how the collaboration goes on, how much time doctors here can devote.
DXY: Besides Dalian First Affiliated Medical School, are you in collaboration with other hospitals or investigators here in China?
Alan H. Morris：No, but I will go to Beijing and Xi’an in next few days to discuss possible collaboration.
DXY:In addition to the protocols for heart failure you are started with, what other protocols have you developed?
Alan H. Morris：We’ve developed computer protocols for mechanical ventilation for patients with life threatening lung failure, protocols for intravenous fluid administration and hemodynamic support, and we’ve developed protocols for the management of blood glucose with intravenous insulin. Those are in operation and have been developed, tried and validated during the past 27 years. We are striving to develop a protocol for heart failure, which encompass management of heart failure in a wide range of settings, including intensive care, in-patients, and outpatients. We are also starting to develop protocols for the interpretation of lung function tests in the pulmonary function laboratory. We could develop protocols in other areas but we need more interested experts willing to participate. That is the challenge!
DXY:In your current program, what different types of scientists are in your team?
Alan H. Morris：Our collaborating centers change, depending upon the study. We have roughly about 20 to 30 centers involved in one study or another, but they are not all involved in the same studies at the same time.
DXY:How many people are there in your group?
Alan H. Morris：It’s very difficult to say because many of my colleagues work part time in our group and they have other obligations as well. So we have 6 to 8 colleagues who work in medical informatics, a number of critical care colleagues - perhaps 10 people in intensive care who work in one way or another. And in the heart failure effort, we have help from colleagues at Jefferson Medical College in Philadelphia, at the University of Massachusetts, at the University of Utah, at Intermountain Healthcare, and at the Heart Hospital here in Dalian. We also have colleagues in Singapore and elsewhere but they work on one project and not on another. They were all part of the group, but not at the same time. There are a large number of people who have participated.
DXY:So there are lots of people from different research areas.
Alan H. Morris：Some are from medical informatics, some from cardiology, some from intensive care medicine, some from pulmonary medicine. Some are from Asia, others from US. In the United States, we have help from colleagues in one way or another - participating in the development of rules, review of rules, in the validating protocols at many different institutions of the United States including Harvard, Yale, and John Hopkins University and so on.
DXY:You are a physician of pulmonary diseases, so how do you transfer to this informatics
research? What interested you?
Alan H. Morris：In a randomized clinical trial in the 1970s, we evaluated heart-lung machine support, for patients with serious lung failure. That was called extracorporeal membrane oxygenation. In 1985, we became aware of some work in Milan, Italy done by Luciano Gattinoni MD, and his colleagues. They indicated that a modification of the heart-lung machine use could save lives. In the original study in the 1970s, we were one of the nine or ten centers that did a NIH study that did not find any benefit from heart-lung machine use, so we stopped using that. But in the 1980sGattinoni and his colleagues reported that they could reduce mortality strikingly and we were so stunned by that report, we decided to do another clinical trial using heart-lung machine. This time, with a method called extra corporeal carbon dioxide removal. But in order to do that, in a scientifically rigorous manner, we all agree that we need to have a reproducible method - which did not exist. So we developed the first protocol between 1985 and 1987 in order to do the heart-lung machine experiment. In the clinical experiment, we also again found that extracorporeal support did not help the patients, but we realized that the protocols might be themselves quite helpful.
DXY:So you studied informatics since that time?
Alan H. Morris：I’ve had a lot of practical contact with informatics before. We had colleagues in medical informatics who helped us managing information in the intensive care unit, and even do some simple decision support, even back in 1973 and 1974. In 1985, we made a much bigger experiment by developing protocols. We first made the protocols in papers form and had a big book of about 50 pages of paper flow diagrams organized according to what was done to the patient. Nurses, therapists and doctors followed these protocols around the clock in the management of patients for about eight thousand hours. At this point we realized that we might be able to do these protocols and we took the rules in the paper protocol and computerized them in our electronic medical record that has been in operation in our institution since the 1960s.
DXY:Then you continued to develop the protocol?
Alan H. Morris：Then we continued to use the protocols, to advance the protocols and add new protocols. As a part of our work in the NIH Acute Respiratory Distress Syndrome Network - a big critical care research network in the United States. We computerized a number of protocols for several clinical care strategies in the NIH Acute Respiratory Distress Syndrome Network, - modified our mechanical ventilation protocol to make it compatible with NIH studies. We also developed a computer hemodynamic protocol for the NIH Acute Respiratory Distress Syndrome Network experiments. We also developed blood glucose measurement protocol which was widely distributed in the US. We have used this glucose protocol for translating research results to practice. I didn’t show this in my lecture just now but we have published that.
DXY:Are you getting any grants from the NIH?
Alan H. Morris：We have had several from NIH and we are trying to get more, we are trying to get funds for the heart failure work right now.
DXY:What challenges do you find doing this research?
Alan H. Morris：One is getting funds for the research, to make it clear to colleagues that this is both feasible and reasonable to do since many physicians seem to want to behave independently and not work in collaboration with others. It’s very difficult to get not only physicians but human beings in general to abandon their own strong opinions in favor of a reasonable approach that everybody agrees upon. That’s always a big challenge but it’s manageable through more collaboration.
DXY:Just now you mentioned doctors’ opinions as a potential obstacle and in one of your
published papers, you mentioned that nine factors, including self-efficacy or self-control, accounts for 66% whether physicians choose to use protocols or not, right? So what are those factors and what’s the relation with those protocols?
Alan H. Morris：Those have to do with human behavior, how people view themselves as decision agents. It depends on whether people think they are capable of doing the work, having a sense ofself-efficacy, so that they can achieve what they want. If people are concerned about self-efficacy that may have low estimation of what they can achieve. Those differences in self-perception have a big impact on whether they accept to use or decline to use such protocols to manage their patients. Those are in general human behavioral challenges that I am not an expert in. However, as far as I know this is true across cultures and is just as important in Asia as in Western countries.
DXY:So is there anyone who works in psychology area in your team?
Alan H. Morris：Yes, we have a colleague in the statistical department, who has been trained in psychology and qualitative research. He is now writing a grant proposal requesting for support for validating a survey that is based upon psychological principles. We will use this survey (questionnaire) to estimate the burdens perceived by beside clinicians who use our computer protocols. We also have a number of nurse colleges at the nursing school of the University of Utah, with PhD degrees who are quite knowledgeable about human behavior and we have been collaborating with them for years.
DXY:What is the thing that you find is very interesting or exciting about your research, besides the challenge?
Alan H. Morris：If it turns out to be successful on a large scale, it would be possible using these tools to join different hospitals and make a distributed laboratory, a very large human research laboratory that exists simultaneously in many institutions. Using the same web-based copy of the protocol will make possible clinical research that’s clinically not possibly at all now. By producing a reproducible method that is used by everyone we might, just for the sake of discussion, join 5,000 hospitals on multiple continents all accessing the protocol over the internet.We might enroll hundreds of thousands of patients very quickly, and address problems that can’t be possibly addressed now with current tools we have in clinical research.
DXY:How did you update the protocol?
Alan H. Morris：So that’s a very important question. Not only updating. I think your question implies different contexts, different institutions and different environments with different needs for treatments and decision-making. That’s why we need colleges from multiple sites in order to gain insights into the influences of different environments. For example, one simple issue involves drugs. Heart failure management with drugs may be different in China than in North America because there are different drugs available in China. So when we talk about options, which drugs we use first, which drugs we use second and which drugs we use third, it may be important to know where patients are being managed.
DXY:So there will be different choices?
Alan H. Morris：There might be. We will try to get everybody to agree to use the same drugs, so we have to understand for all institutions. But perhaps the drugs that people in US want to use first are not available in China - then we couldn’t do that. But I don’t think that would be a problem because I think in China, virtually all the drugs we have in the US are available.But the converse is not true - there will be drugs in China that may not be approved by the FDA in the United States. Then I will try to convince Chinese colleagues not to use these but to use drugs that are available everywhere, if possible.
DXY:So it’s standardized, internationally standardized.
Alan H. Morris：I hope so, because if you want a distributed laboratory that is international you have to standardize it internationally, and we have done that. When we refine our rules, we first start with a small group in one institution then we go to more local institutions and get their critiques and input and then we go to more distant institutions in different parts of the United States and get their critiques. And for heart failure, we are also going to Asia for critiques.
DXY:What do you think is the major issue that is not answered yet in your research?
Alan H. Morris：We shall fill a book with major issues that are unanswered.
DXY:You say it is a quite new area.
Alan H. Morris：A basic foundation of what we do with healthcare delivery is doing more good than harm.However, many health care interventions do more harm than good, and therefore should be removed. You can imagine how many issues that involves. It would fill more than a book and each of them will have to be tested carefully, those that don’t work discarded and those that work implemented widely to increase the quality of care. So this will be is a long ongoing work.
DXY:So which hospitals are you in collaboration with? Is Dalian in China is the first one.
Alan H. Morris：From the First Affiliated Hospital of the Dalian University, Dr Xia, Dr Ke, the chief of heart failure and Dr. Yang, the Vice President of the Hospital and the President of the new Cardiovascular Center, have all agreed to collaborate. Perhaps we will be collaborating with people in Xi’an but we haven’t met them yet - that will happen later in this trip, and possibly in Beijing there will be someone who wishes join. So I don’t know how many there would be.
DXY:Yesterday I talked with Mr. Burnett, and he said this time the trip to China is a mission trip,
because he is finding collaborations through three ways, so are you also looking for some collaboration in China? If yes, then what is it?
Alan H. Morris：We have two major domains in which we are seeking collaborators. One is the development of the rules for heart failure, called knowledge engineering, where we capture the way clinicians make decisions, and put them down as rules that can be carried out in a computer protocol. The second domain is in intensive care (ICU medicine). We already have three ICU protocols developed and implemented. We want to attract colleges or institutes that want to use the ICU protocols for initially research purpose in quality improvement and then later perhaps continue to use them in clinical care in general. And now we are just starting a program to develop interpretation rules for lung function test. We have a college in Xi’an who wants to participate in that but that’s not yet underway and we are just preparing. One of my colleagues, a young physician, is leading this effort and she has a local grant to do this.
DXY:You said the project of the heart failure may last at least for two to three years.
Alan H. Morris：Well, the development of the rules. The research project, if the rules are successful and work well, may go on for decades. As you know there are so many questions in heart failure so we may not know how long we’ll go on. I don’t know whether 50 years is a reasonable statement but there are many questions that need to be addressed and a lot depends on how successful the rules are, whether the protocols are effective and safe, and how many sites we can get to join. If we get a large number of sites and we can deal with a large number of patients quickly, then we can answer many questions in a short period of time. If we only have a small number of sites, we have to do it very much more slowly. So it’s a guess about how long this will go on. It will go for a long time unless it turns out that we are not successful but I think the chances are good that we will be successful - but there’s no guarantee.
DXY:Right now, in addition to the heart failure project and the intensive care project, are you also involved in any other projects that use the protocol?
Alan H. Morris：One of my colleges is doing a randomized clinical trial in pediatrics examining two different blood glucose targets for children. Both groups of children in the randomized clinical trial use our eProtocols (computer protocols) to standardize bedside clinician decision-making. So that’s underway and we have a few other studies underway.
DXY:The last question, how do you foresee the collaboration with Chinese colleagues?
Alan H. Morris：For two years colleagues in Dalian and I have been talking and collaborating a bit. But this time I come with three other members in our group from Salt Lake City, Utah - so we have a group of four of us and we are making a big effort to get everything started in a very serious manner.We will meet colleagues tomorrow in intensive care at the First Affiliated Hospital of Dalian Medical University, and then on Monday in heart failure.Then we will go Beijing for three days, and then we will go to Xi’an for three days before I return to the United States on August 8th.
DXY:So best wished to your research project and hope everything goes on very well and thank you for your time for our interview!