Welcome Dr. Reich and thanks for joining us here today. Can you introduce yourself, please? >> Hi, my name is David Reich, I'm the president and Chief Operating Officer of the Mount Sinai Hospital, I'm President at Mount Sinai Queens. >> Great, and thanks for joining us today to talk about how we're going to use health IT to achieve the goals of the institution. So right now you're President of Mount Sinai Hospital in Mount Sinai, Queens, but you have a long clinical background before that. Can you talk a little bit about that, please? >> Yes, I came to Mount Sinai in 1984 as an Anesthesia Resident. And then I completed that after two years in those days having done preliminary work elsewhere and then was a Fellow in Cardiothoracic Anesthesia. In 1990, I became the Co-Director of Cardiac Anesthesia and then in 2004, Chairman of the Department of Anesthesiology. In 2013, I was asked to be interim, and then permanent President at Mount Sinai Hospital. And so over the course of 2013-14, I transitioned the Department of Anesthesiology's leadership to another individual. >> Now, your use of information technology to further healthcare goes back a long way, and you were really on the forefront of this. Can you talk a little bit about how you got into the business of using information technology to deliver healthcare? >> Well yes, in 1989, there was an article published in one of the IT magazines at the time about the use of information systems to gather electronic anesthesia record data and to put it into a usable form. And at that time, there were only two systems in the country, really in the world, that were capable of doing this. And we adopted the system which we still have in use today, which is the CompuRecord Anesthesia Information Management System, in the end of 1990, the beginning of 1991. And that system, although it has upgraded from a DOS version to a Windows version, live through the transition through Y2K. The product has evolved in a very logical fashion and was very stable over many years. And that gave us a great platform for using that information initially for research. And then later, for quality management and for administrative purposes. >> And just to put this in perspective, 1989, 1990, those were days in which a lot of medicine was done on paper and pen. >> Well, we were, to my knowledge, the only electronic system at that time for electronic record keeping for patients. Of course, there were some administrative systems and radiologists and cardiologists to a certain extent were leaders in using databases, in their particular areas. But in terms of an electronic patient record, this was certainly the first at Mount Sinai and one of the first in the nation. >> Why do you think anesthesia was sort of a natural fit for going with an information technology solution ahead of the curve? >> Well, it was really a bit of a double-edged sword at the time, because anesthesiology lends itself to this area because we have a lot of monitors. And even in the days of the 80s and 90s, those monitors usually had serial outputs, which enabled us to acquire the data from the monitors. And so this vast amount of physiologic data that was largely documented by paper in a retrospective and very inaccurate fashion. Almost at a moment in time, we were suddenly able to collect tremendous amounts of data, typically every 15 seconds. And we could have 40 or 50 separate items of information acquired into the anesthesia record database over the course of moments. And you were able then to display the information that was either averaged or otherwise filtered, so that information would appear on the record every two minutes. >> Now, going back to that time around 1989,1990 when you were starting to use IT, how hard a sell was this to your department of hospital leadership to get this funded and put in place? >> Well, let's talk about the double edge sword for a moment, their practitioners were very much afraid of an accurate anesthesia record. There was a perception- >> That was going to be my second question. How hard of it was [LAUGH] a sell to the clinicians but yeah, I'd like to hear both of those sides. >> Yeah, the administration was in favor of this because they always felt that Mount Sinai Hospital should be on the cutting edge of technology. Although we certainly believe that strongly today, I'm very proud to say that this is an attitude which goes back over two decades at Mount Sinai and probably even before that. And so what did that mean, the clinicians were scared because in those days, malpractice rates for anaesthesiology were very high if you look at it in terms of current dollars. Since certain advances that were made in anaesthesiology, particularly in the mid 1980s relative to adoption of pulse oximetry and capnography, those rates have come down. But the feeling was that if you had an accurate anesthesia record that showed all the peaks and valleys of blood pressure, and all the vicissitudes that occur in an operation, that you were going to hurt yourself medical legally. Well, it turned out that really wasn't the case, and attorneys who were asked about said they would much prefer to see a valid and reliable record, which is an authoritative source. And so really it turned out to be a non-issue, but the anxiety of the clinicians was quite intense. We started with only six operating rooms, only the liver transplant and the cardiac surgical operating rooms. And it wasn't until 1998, really about seven years after the initiation of the first block of ORs, that we extended to all operating rooms at Mount Sinai. >> So what I'm hearing is that when you implement an IT solution like you did with anesthesia, some of it is about the IT, but a lot of it is about the people, and getting people to believe in your cause big and people to champion what you're trying to do. You were able to find those partners to work with you? >> Well, it was interesting. Over time, a few individuals showed some interest, but starting with the beginning of the project in 90, 91 through the final adoption in 98 of all ORs, I only had one or two people that I recruited and hired to support me. In general, if there was a Daylight Savings Time change, I had to wake up and go to the server and adjust it. >> That's true leadership there. >> Yeah, as a Cardiac Anesthesiologist, there were a lot of transplants, so I was often present at that time of the day or night, I should say. But the moving along and getting to that next level, it didn't really happen until I'll become chairman of the department, and I changed the system such that it drove quality and most importantly, position compensation. And that time when people were depending upon the record not just for documentation but also as a source for quality and for the compensation, which was their livelihood, there was a much greater adoption and much greater interest of people in the department in working with me to help manage the data. >> So let's talk about that a little bit more, if you don't mind. So when you use information technology for healthcare, you're thinking about quality, you're thinking about costs, you're thinking about your patient experience, you're thinking about your business objectives. So let's talk first about the impact of putting in the IT systems for anesthesia on quality and then about the compensation. So did you look at it like a quality initiative? Well certainly we did, and also a research initiative related to analyses of what's now called big data, but back then we just called data mining. And for example, an early project looked at the failure rates In a cardiac cases proposal as chemistry and in other sub specialties in the OR. And the reason that was important was a State of New York came in and they did one of their inspections as they want to do and State health department do this through the nation. And they came and then they said you Mount Sinai are violating State laws. Because at Mount Sinai you have anesthesia records which have missing pulse oximetry data. And we went to Cornell, and we went to Columbia and NYU. And their pulse oximetry numbers read 100, 100, 100, 100, 100, 100. And yours read 98, 98, 99, missing, missing, 98, 99. And I said, well, we're not lying, we're presenting the actual truth, and there are times when pulse oximetry doesn't work And then they said, we don't believe you and we're going to fine you $5000. And so I did a research project which demonstrated quite nicely in those days, doing a relatively effective data mining in an era before it was relatively common. There was very predictable causes of pulses symmetry failure. Hypothermia, use of cardiac pulmonary bypass and the pre-embossed post bypass period. And when I published this everyone applauded it. It was a great publication but the state did not refund the $5,000 fine. >> That was going to be my question. You didn't get your money back. >> We didn't get the money back but we learned a lot. And they never came back again and questioned us. Because they knew that we had proven our point that these technologies are imperfect. And the paper documentation that implies that they're perfect is just a fiction. Because people don't want to incite controversy in their paper documentation. >> Back in the late 1980's, and early 1990's when you started to electronify the anesthesia systems that you were using. Did you come up with any predictions as to what the impact on quality was going to be? >> Well there are a few things that we did related to quality. Some of them were in the form of research papers again Where we looked at the impact of intraoperative hemodynamic abrasions on outcome. So for example, intraoperative hypertension, high pulmonary artery pressures. In the days where we used PA catheters in many patients and other variations that occur during the intraoperative period. We're strongly associated with that first outcomes. But we also used it for administrative purposes. And so we were able, for example, to start looking at periods of time between cases. That was one of the first systems that we were able to use for management. And also sometimes we could look at periods of delay. For example, if we had backups in getting patients out of the operating room, we could see when the operation ended and when the patient left the OR. So it was the beginning of the beginning of using electronic systems for administrative purposes for OR management. >> And the other thing you mention is that you started using the anesthesia record as a way to come up with a compensation model. >> Well when I became chairman of the department in 2004 I felt very strongly that the incentives of the physicians were not aligned with the incentives of the institution and relieve the department as a whole. And so, I worked with a small group of individuals to come up with a point system to drive compensation within the department. Now, some of that was outside of the ORs such as writing papers Or providing service and leadership in a particular area, such as running a division of obstetric anesthesia, for example. But within the OR, the vast majority of the points that drove people's compensation was related to providing clinical care. And it isn't enough with anesthesia to just look at the amount of time you spend in a case and the complexity of the case. But also it's important to look at aspects related to concurrency. How often in individuals covering two cases at the same time. So without going into all of the details of this, using the anesthesia record that was already in place And giving points for service for various times of day so giving more credit for people who are working in the evening hours who are on on call, giving the appropriate mind of credit for complexity of cases and time worked. There were suddenly a system were anesthesiologist to every incentive To keep the turn over time between cases as short as possible. To perform every case that was possible to perform because the productivity which is measure by many physicians, this are for used was measure from anesthesiologist base upon the points that were generated. From the intraoperative record. >> Did you find that by using the information technology platform for determining compensation, did the physicians find that to be a fair system or did they feel like it was somewhere living under the tyranny of the computer? >> Well I'll make a couple points about that. Number one is when I started the system, I did involve several key members from the department. I'd say about 20 in total in forming interest groups to talk about how we should award those points, both for difference of specialty areas including obstetrics and pain management, And for the intraoperative portion. And once that group had gotten together, we had some consensus among the department in terms of how things should work. However, I learned quickly that calling something fair and transparent was a bit of a problem because people generally define something fair is something that paid them more money and something as unfair is something that paid them less money >> And important leadership lesson. >> Well, yes, but the transparency piece, that was okay. There was no question about the transparency because people received reports and they got those reports as often as weekly. I think we settled finally on bi-weekly. And so, if people felt somehow that a case had been missed or there was some inaccuracy There was a period of time where they could go back and talk about that. And then in addition, the system morphed gradually over time to include more quality metrics. And so you can, as a chairperson, you can do a minor amount of experimentation on your faculty without research consent. And so, for example, adding a reminder button to give antibiotics before the button that shows up that indicates that surgery has begun. Just the implementation of that sort of soft alert created improvements in quality. And the ability to measure how often individuals were appropriately documenting the improvements and quality. >> I'm taking you up into the directions now of when you became President of the hospital in 2013 at which point your responsibilities expanded. But let's talk for a moment about the period of time when you were chair of the department of anesthesia in the period from 2004 onwards. At that point, anesthesia at Mt. Sinai hospital was certainly ahead of the The IT curve, compared with most of the rest of the hospital, maybe radiology, maybe the emergency department were some of the areas where they were farther along than others. But a lot of the hospital processes were still on paper. Tell me a little bit about what the environment was like to be sort of ahead in the IT curve. And working in a hospital where they were so many of your colleagues and coworkers were not that fall. Well the challenges of being an early adopter department, is finding and developing the relationships with IT, which in those days was subdivided into both a school IT and a hospital IT. A group that weren't integrated so well. And the other departments that had bits and pieces of information that we felt was important to share over time. And so I've often joke that I didn't have any gray hair before I built my first interface and I'm sure there's a direct correlation between the two. But the point of that is that creating those alliances with IT and with the other departments that had information systems was an important lesson for me and I hope for others in what it takes to really be successful in the world of information technology, where a lot of it is about sharing, and a lot of it is about citizenship. And so I said, those were the lessons learned in the early days of this. And also expanding and improving our individual systems required the IT group to develop interface gateways such that we could receive demographic information, so that we could share information with their billing company so that the electronic anesthesia record could drive the electronic voucher, which drove the creation of a patient bill. And then other people needed other bits of information, such the hospital would need to know when patients had come back for surgery within 30 days, so they could file reports for New York state. And so that's how this group came to coalesce and I'd say it also was a great way to build camaraderie and collegiality by making partners and friends in other departments who care passionately about IT, and its value for the institution. >> One of the sort of rude awakenings and I'm sure happens when you move from being chair of anesthesiology to president of the hospital is the number of things that all of a sudden became potentially your problem, whether or not they were problems. All of a sudden your IT responsibility's magnified. Tell me a little bit about that experience. >> My first roommate in medical school said the secret to respectability was a clean kitchen floor. Now we were not all that respectable by that criterion. But what I take away from that for IT systems is a secret to grade IT systems is elegant downtime procedures. And so, I learnt in the anesthesiology days and learned much more in the hospital presidency days about the issues of having elegant ways of dealing with periods of time when your systems are dysfunctional. And it is a continuing struggle and it's a path we can all hopefully move along in the right direction of our time but we're so dependent now upon electronic systems. And we have so many disparate systems that are dependent upon so many little widgets and bits and pieces that hold them together that we have to plan for failure. Failures will occur. They'll sometimes occur when there's scheduled maintenance, sometimes when there's unscheduled maintenance. Sometimes just because they fail and so, what I learned as hospital president is you always have to be ready using the tools you learned in your emergency management courses for managing the situation where the systems that you depend on are less dependable than you like them to be. >> Some of the things I will be talking about with the students during this week of the curriculum are the principles of high reliability organization. Preoccupation with failure and the resilience to recover from those failures obviously are things that are important to your role and responsibility here. >> If I could amplify in the previous answer one of the other things that I'd say is that having a close working relationship with IT leadership and having those individuals that you can reach out to literally on a 24 7 basis is what makes those situations where you're challenged. Where a system isn't working as it supposed to work or whether there's been a partial or complete failure. Coming up with ways of dealing with those problems in a collegial fashion and limiting the time when your organization is at risk, those are based upon great relationships between IT and a hospital operational team and the clinical departments. And in all cases, there need to be champions, champions who get along well with one another and who understand the value of communication. And how we, we're not in it to cast dispersions on one another. We're in it to work on it together and to put our heads together to come up with great solutions. >> In terms of the use of IT for healthcare leadership there sometimes you can look at the two ways. Sometimes you have a healthcare goal you want to be able to do a sort number of surgeries prior. You want to be able to open up the new area the hospital and so IT will be supporting whatever your idea is, but in some ways, IT potentially gives you the ability to lead in a new direction. Tell me a little bit about the importance of the two forms, IT as a way of supporting what you were going to do anyway, and IT as a way of leading in a direction that you couldn't otherwise go. Well, it's hard to come over the exact examples but IT supports us everyday in every way. The changes in being able to access in electronic patient record remotely from overseas even, has changed the way physicians and nurses interact with the clinical data that can hopefully always lead to better patient care. And one of the great things about the path we have as an organization, is that we have focused ourselves on the patient as we've implemented all these IT systems. And I believe that most of us always try to think about the patient as the center of the strategy for IT. And so I'd say that's very much the present, is how do we do this? Yes, there's also the business of or in the hospital. We need administrative data. We need quality data. We need financial data. And the IT systems support all of those supports that help us. But where does it take us for leadership? Well, one of the things that we need to do is create better decision support systems. Physicians, nurses, all the other healthcare professionals that work with us, all the mid level extenders, everyone is prone to error. And one of the challenges that the IT vendors have is that based upon their own liability structure and the licensing that they have from the FDA, it limits their ability to truly go into the world of decision support in the way that I hoped they would have done. That keeping the patient at the center of everything that we do. And so, what does that mean right now? It means it's upon us as healthcare leaders to use whatever tools they give us to develop the best possible decisions support. To develop a best reports that will enable us whether it's on a moment by moment or daily or monthly basis to review our data. An intervene is appropriate with teams or with individual patients, and then finally to automate as much as possible in the future, in the present and in the future. The use of decision support to truly improve patient care, and the rest of what we do, the financial, the quality, all of the outcomes that we measure in the hospital. >> And I know from working with you that you've been championing some ways in which to potentially leverage IT to find that sickest patient, to. Find that patient risk for falling, to see how to get messages of the patient who is not yet in critical shape but is heading in that direction to the right people. Is there anything that you want to highlight from that work that you are going in the direction? >> Well, I have to preface it by saying that you've been my biggest partner in that, so thank you Dr. Darrel. But the vision that we share is one where if a patient has, for example, an episode of hypercalcemia, that we use relatively frequently checks on that patient's status over the next minutes and hours to see that the team has selected appropriate therapy, implemented appropriate therapy, and the therapy has been successful. For example, at Mount Sinai hospital, when we looked, we found that if a patient had an episode with a potassium level reported as greater than 6.0. Real equivalence per liters, that their risk of mortality during that hospitalization was approximately 6%, which was about four times the expected mortality for a normal patient in the inpatient beds at Mount Sinai hospital. In the regular medical surgical enterprise outside of psychiatry obstetrics. And so we worked on that, and working with IT and a great group, we've implemented some best practice alerts within Epic and created a notification system. But we need to go beyond that. And the other things that we need to do are to take the original prediction models that we had from years ago, the system that predicts the likelihood of falls within the hospital. The systems that predict whether or not patients are likely to have sepsis, or likely to be having a hemodynamic decompensation, such as the medical early warning system. To go beyond those, and use more information that is available to us, to hone in and create more immediate and more predictive models and to implement interventions in a more of a just in time basis. >> As a healthcare leader who has been a champion for IT for your career, how hard is it to sell this vision to your hospital leaders, and to the clinicians who are providing care, that IT is the way to provide better care? >> Well I think that one of the important things is not really to ask for permission in certain cases like this. One of the things you sometimes do as a leader is you understand that there are tools that you're disposal that can help you get to a new place. And that, if you ask too many individuals about what the challenges are of getting there they will provide enough challenges that you want to give up the project. So I think someone should embrace enthusiasm and optimism but something beyond very cautious and overly measured approach. It's what is required. You have to have the vision and you have to drive that vision. For example, we want to have a patient itinerary at Mount Sinai where patients will have information about what is scheduled for them that day in the way of tests and medications and perhaps surgical procedure, and update on whether the timing has changed. Or whether anything else has change about their care. And to give them information about their medications that they are taking. Creating these systems will take individuals who already very busy to devote additional time and resources to this. And it's only by inspiring them and it showing them the way that they are going to follow that particular lead. because we can't hire hundreds of additional individuals to come up with every brilliant programming idea that we have, we have to leverage our existing limited resources and set the priorities. And I think that's the challenges of the leaders to say of we have 20 things that we would like to do and we can accomplish through this year just what those three should be. >> When do you think the I.T and I'm speaking about the technology itself? Where do you think that it is not meeting your expectations, or it's failing? Where would you like to see IT have capabilities that it just hasn't been able to deliver yet? >> Well, coming from someone who spent a lot of time in the operating room, where a lot of information is acquired very much automatically. I think that what are the challenges that we have is that, when we get out of the OR, out of the Intensive Care Unit to patient floors that we have much less in a way of monitoring systems and I think that one of the beauties of IT is it's ability to have unlimited vigilance. This is an anesthesiologist speaking vigilance is often the motto of our societies. And I think that that vision of using IT to improve vigilance is where I would like to go. And I don't believe that we are close to that vision yet, and I don't actually believe that some of the IT vendors understand what the problems are of clinical medicine. They are business and they exist to be successful. They share our goal at a high level, in terms of improving patient care, but I think it's our job as hospital leaders, as physicians, as nurses, as other dedicated healthcare professionals to drive the IT towards that solution. So, what are we missing? We're missing a lot of adverse events that happen both in the patient's floors and even more so in the ambulatory setting. So whether it is your watch, or some device that's around your wrist, or some other device that will sense your physiologic data or other aspects of your health at home. I think that the big challenge here is to find a way to affordably and reliably gather more information on patients with their permission in settings that are have much less intense monitoring and the vigilance of the IT solutions and the ability of brilliant scientists to mine massive data sets and create information out of big data and that's where I think the challenge is for all of us to use the big data to make people's lives better. >> Dr. Rich thank you so much for your time and for talking with us today. Was there anything else you wanted to make sure that our students had a chance to hear? >> Well I think the take away that I'd like all the students to have is whether or not you were a computer nerd in high school or in college. The IT is part of our futures in health care leadership and so you should either have some great resources available team in a way of colleagues you can support you and tell you about what are the latest and greatest development society or find other resources whether it's by reading or by telling course work such that you can really learn about what IT brings to the table. We will not be successful as healthcare leaders unless we leverage the best and brightest in the IT world and to bring them into our management teams. And so, don't worry if you didn't start as a computer nerd just become one.
