The Noblis Center for Health Innovation is a nonprofit science, technology and strategy organization that helps clients solve complex systems, and process and infrastructure problems in ways that benefit the public. HCI Editor-in-Chief Anthony Guerra recently had a chance to talk with Alan Dowling, Executive Director of the Center, about the pertinent issues facing CIOs today.
AG: Why don’t we start with an overview of your work at Noblis, and let’s just keep in mind our core audience, which are CIOs of hospitals, mostly the larger facilities. Tell me a little bit about your work at the center as it relates to the CIO, concentrating on areaas they will find the most useful.
AD: Center for health innovation is the health industry focus group for Noblis. Noblis is a non-profit organization that is involved in advisory services and applied translational research. Our unit’s focus is on the health industry. Most of our attention is paid to the provider community, but not exclusively. Many of us who are in the center have previously lived and worked in telecare delivery organizations.
Early in life, I was the first US Air Force Medical Service Corp Officer that was assigned the responsibility of what today we would call a CIO in a healthcare delivery site. Noblis’ mission is to help improve healthcare and healthcare organizations’ ability to meet their mission.
We focus on several things, one of which is helping an organization formulate their direction and their successful future state as embodied and enterprise plans, service line plans, and other forms of business plans for the health entity.
Next, based on that successful future state, we make an effort to help them innovate and improve, focussing primarily on innovating their work processes and decision processes, so that they can achieve the potential success inherent in their strategies.
The next part of the plan is to help them implement new ways of working. Our intent is not only to develop strategy and make suggestions, but to work with an organization in order to engineer the benefits inherent in their strategy. This way they can focus on a multitude of processes, running all the way down through organizational design, change management, project management, and acquisition support. In doing so, we can help an organization achieve a higher level of success for itself, and help them to believe that people working in healthcare are de facto; which, by definition, means working to improve the national interests of the country. We find that working in healthcare to help organizations achieve success is completely congruent with our non-profit mission.
The other part of our plan is proliferated in our non-profit mission, which also has us involved in applied and translational research – areas that are of significant importance to healthcare. We like to combine our advisory services and applied and translational research – especially in a collaboration of research projects, of which we have a number of going on right now.
AG: Does your non-profit status mean that CIOs should look at you differently than they would a traditional consultancy? Do you think that allows you to bring more credibility to the table?
AD: It may. A number of us have been partners or CM partners in the country. For example, I used to be what's called a B partner in the firm Ernst & Young, when we had a healthcare consulting arm. In my most recent position I was the partner responsible for pulling together global healthcare consulting. So, many of us have deep experience in consultancies that are in the for-profit side, and we have brought this experience to a non-profit environment. But, what we’re doing here is bringing that expertise, and ability to exercise those roles as partners in major organizations, to a non-profit setting for a particular reason.
People who come to work here are heavily interested in outcomes for clients. Our focus on that, and working in a non-profit, means that our attention can be applied primarily to our clients, rather than to our shareholders or what our stock valuations will be the following day. We focus on creating partnerships that are going to create value for our shareholders and our partners. We are an organization that is independent and objective.
When I was E&Y, they were highly objective and highly dependent and highly ethical. So it doesn’t mean that you have to be a non-profit to have those attributes, but having been on both sides, I do know that often I am not encumbered in my decisions. I can simply act in the best interest of our clients. So it may make a difference in the margin. We do not form alliances with commercial entities for the sake of selling things, helping business furtherance, or by fronting products. We’re trying to be a true consultancy, with true objectivity. Ethics and independence are central to the way we do business.
AG: I see that you are involved with the Office of the National Coordinator, working on their projects to define five healthcare IT terms that were put out there a few months ago. When I saw that come out – and I think a few people saw that come out – it seemed as if it was almost silly that we needed to spend time defining five terms. But when we did a poll on our site, that’s not what we found people thought. We found that people thought it was an important project. Tell me why you think it is an important project, and what problems do CIOs and other IT executives face in the field that this could help alleviate?
AD: Let me preface it with the fact that any of my comments are not speaking on behalf of ONCHIT or of this effort. There is a communications channel that I don’t wish to violate. So I would be talking to you from my own perspective and my own experiences, if that’s alright with you. They can officially speak on their own behalf.
The reason I was interested in this is because I have been around healthcare and informatics for awhile, and I've watched terms that once meant something become almost meaningless because of the multiple voices in the marketplace trying to use those terms for particular and different purposes. For example, what does artificial intelligence really mean anymore? What does decision support really mean anymore? People use those terms, sometimes correctly, sometimes incorrectly and sometimes for other objectives, for example, marketing and sales.
So the issue of semantics I think is really kind of important. If you read my background, you know that my doctorate is from MIT. I’m interested in a cross of things, very practical and also conceptual. At first, the thought of defining is a bit on the conceptual side. But my belief is that it’s a highly pragmatic issue, and it’s central to our ability to communicate and understand, and that is important within a healthcare organization.
When I saw electronic medical record or electronic health record in my own organization, I thought, what does that mean? As we work with other organizations like ourselves (I’m imaging myself to be back in my role at a hospital), and with other hospitals and providers, if we say this, are we really meaning the same thing? For example, if our integrative delivery system is going to provide electronic health records capability to our referring physicians, what does that mean? What should people expect? I find the definitions of personal health record – I was involved very early with this technology in a very deep way in the 90s – to be sufficiently ambiguous at this point. Things that I do not consider personal health records are being labeled such, for marketing purposes.
So my feeling is that there is far more confusion about these terms than people realized. The way I think about this is if you put 30 people in a room and ask them to write down the definition of electronic medical record or electronic health record or personal health record, you're going to get 35 different answers. While that may be kidding around a bit, I’m also very serious about that. And if in fact you don’t have a common semantic, then you don’t have a common language, and you don’t have a common way to reach agreements or understandings of what's in and what's out.
I feel the work of HL7, and all of the other organizations that are helping us come to standard definitions and standard terms, can enable us to create some kind of common or standardized way of approaching healthcare. This is really instrumental to the building of a common infrastructure, and will enable us to implement (nationally or regionally) a much stronger informatics, which I believe are central to the improvement of health and health management of the country.
So when it first seems kind of trivial – define electronic health record – it turns out if you peel the layers off of this, it’s actually a very complex undertaking that gets you into a lot of issues – who owns, what is, what are the privacy issues associated with this, who has responsibility for constructing it, feeding information into it, managing it, assuring its existence over time? It gets you into a lot of very fundamental and very important issues. So it’s not just a semantics issue, although that’s important; it gets you into the fabric of what we’re building in this country. That’s important.
AG: Where do you think we are in this process, in terms of where we need to be to really help people that are out in the field working on implementing these systems? Are we just at the beginning of getting our house in order so everyone can go out and do good work? Where are we in the process and how far do we need to go?
AD: I feel we’re much further along than that. I started getting involved in this and researching why certain systems succeeded and why some health organizations failed. That led very early on into my research at MIT when I was doing my doctorate work on the creation of something called macroergonomics, which essentially has been generalized by lots of organizations, but it basically says our scope – when we talk about a system – is really the integration of mission based processes, people and their capabilities, and the technology they use to execute on those processes, or trivialize people processes and technology, which is now a very widespread concept. But I believe that it’s true.
As a result of that, I think that we've matured greatly, especially in the last 10-15 years, in our appreciation, understanding and agreement as to what healthcare informatics can actually do to improve healthcare in the United States. I believe that in the early days, it was like pushing the rock up the hill, trying to get people to understand and appreciate and give us a chance to try systems. Now you’ve got to remember in this business, I go back to the 70s. Now, those of us in healthcare informatics are kind of on the other side of the hill trying to hold back the rock from rolling down and rolling over all of us as the demands from many constituencies, including clinicians, are overwhelming our ability to deliver on the promise of informatics.
That is a problem, but it also shows the progress that’s been made by many people in the country, the degree of acceptance, the degree of expectation, the reduction and beliefs that informatics is a threat to the practice of medicine. If you’ve been around for awhile, you can see these major trends. I actually think we’re in a very good environment to make substantial progress with the creation of certain standards that enable intercommunication to occur with a greater emphasis on governments and non-health organizations, which will help people understand the importance of informatics.
I feel that we’re in very fertile ground. I think an enormous amount of incredibly good and innovative work is being done, not only in this country, but in other countries. So I feel very good about the situation.
The things that still inhibit it, as far as I’m concerned, are the fact that while there are a great deal of very good systems out there, there are also systems that aren’t as good. As a result, I feel that a lot of healthcare organizations, once you move out of the IT shop, are understandably in a state of confusion as to what is good and what's not. What's the best approach to take when talking about functional architectural and the management of informatics – not just the technical architecture?
I find that there is an increased readiness and increased intelligence to increase the availability of capability and good systems out there. But there is still a kind of natural and understandable muddling as to what's the right thing to do in a very complex environment. What are the right priorities, and what's the right timing, and who do I trust and who do I not trust – that all still needs to be worked through. Actually, that’s one of the areas that we as a non-profit help with, because we see ourselves as an absolutely unbiased ‘friend’ of the healthcare organization. We enable them to make a decision that’s going to be successful for them. That’s one of the reasons that we’re really insistent on our having an unbiased objective and independent stance.
AG: One of the other things I have seen that you were interested in is helping people measure their success. Tell me a little bit about how important you think benchmarking is for it, and do you mean for people to see where they are visa vie other hospitals, and what do you think that you can do in terms of allowing them to move forward?
AD: There are a couple of components to that. I started life as a hardcore computer scientist. Early on, I was fairly naïve, and I felt that if I could build a good system, that’s all that was important.
When I was responsible for informatics in my own hospitals (this is in the Air Force), I very rapidly realized that the technology is critical, but it’s not the only part of the equation. Getting people to understand the value that it creates is important to enable them to build trust in informatics’ ability to support them – allowing them to see the benefit and have them come on the side of constructive technology and change.
How do you do that? I’ll give you an aside – I remember very early on, I was in a situation bringing up a fairly sophisticated system in my hospital. The clinicians were very upset about the loss of results from the lab system, and they saw it one way and I saw it another way. They felt that the system had come up and they were losing about eight percent of their results. So we got together and I agreed that that is absolutely not acceptable. They were starting from not losing any results and then we bring in this computer system, and suddenly they're losing eight percent. Actually, I believe that we need to measure progress for a number of reasons. I had done a fairly comprehensive and scientifically quite valid measurement of lost lab results before the system went live, and it was 23 percent.
So in just 10 hectic days of implementation, we had improved from a 23 percent loss to about an eight percent loss. So we’re driving it down to zero, moving in the right direction. However, the clinicians who were my coworkers (and quite frankly, my friends), very bright and very capable people, thought that things were getting worse. And the reason was because before the system came in, they had developed coping mechanisms. And we measured that they were actually ordering about 125 percent of the labs that they really needed. With a 23 percent loss rate, it still meant that they were getting all the information they needed. So they just had embedded this coping mechanism, and didn’t even realize they were over ordering, which obviously increased costs.
When the new system came; the over ordering stopped and they suddenly started noticing a loss. So they became sensitive to it, but because they hadn’t done the pre-measures, they thought things were getting worse when actually they were getting dramatically better.
I was able to take this data to them and convince them that this, in fact, was correct. It turned their attitude around entirely and brought them back on the side of supporting the information system, and the change that it brought to the organization. It turned out to be a successful implementation. Had I failed to do that evaluation, that pre-measurement, I would not have had any data to give them to show them things were getting better.
We developed something I call an expectation curve, and I can share that with you. It shows that once a system goes live, the expectation for the success falls, and it’s not just in the health industry. It can go into a critical range when you're seeing even good systems being thrown out of hospitals if influenced leaders do not support it.
One of our jobs in healthcare is to make this curve flatter, so there is less trauma and mixed expectations by being, I think quite frankly, honest with everybody upfront as to what they can expect in the adoption of the technology. And then after it goes live, have them realize that yeah, there is a lot of work in an implementation, but we’re going to come out okay and this is going to be good in the long run. Inherent in that is the measurement, and a demonstration of the progress that you're making to achieve the outcome you wish to desire. So that’s one thing.
Another thing is – this is weird because I have an undergraduate degree in physics and I have an MIT PhD. – I don’t believe the adoption of information systems should be a scientific process. I believe it should be an engineering process. What I mean by that is in science, we are observing phenomena, we’re measuring phenomena, we’re understanding phenomena. But I believe that’s not why we put information systems into our healthcare organizations. We actually want to engineer a positive outcome that says we need to approach implementations from the view that we are going to engineer a successful outcome.
To do that, I argue, you need to measure your progress all the way through the adoption of the technology and use that hard data to show how well we’re doing or where we might be stumbling, to be able to reprioritize our implementation efforts so that we can retune the system. We can retune our work processes so we can train our people better, and so that as a result of measuring all the way through, using that data to improve and therefore engineer a better outcome for everybody involved by the time we’re finished. Those are just two areas that I think measurement and evaluation are important for.
AG: How do you reconcile all these duties, responsibilities, and obligations – necessary things you're saying, and I assume we’re talking about the CIO role as the person who would prove out the fact that you are getting much better than you were on paper, even though there is a loss, it’s a far less significant loss. And you talked about all this reengineering work. I wonder if in the real world CIOs aren’t faced with the pressures of a checklist mentality from the board of directors and from their CEO in terms of okay, you rolled out a prescribing, check, now let’s make closed loop medication management, check. Did you roll out your EHR, check. So how do you reconcile what I suspect may be the real world pressures of a checklist mentality with all the deep, time consuming and difficult work you're talking about to really make these systems effective?
AD: I think you're right, first of all. And I think sometimes information officers do it to themselves by creating their own checklist, and we all do it to make sure that we’re getting things done. I think there is a deeper cultural issue here. I see informatics information systems as really not being an anomaly to life, but part of life, part of work life. And if you take a look at it that way then I disagree that all of the responsibility should be solely placed on the shoulders of the information officers. I feel that informatics enables the healthcare mission to be achieved. And as such, it’s a very powerful tool, a very important part of life.
If that’s the case, other executives bear heavy responsibility to enable the information officer to help them be successful. As a result of that, I believe from the board, through the C suites on down to the organization, that the adoption of informatics is really a shared responsibility. And that means people must become aware, must become more knowledgeable, must become more collegial and see this as a team thing that the information staff may be the spearhead. No CIO I know cares about simply adding more technology to the institution; they care about effecting the outcome for the organization, so the organization is intrinsically successful in its mission. And that goes from patient care to financial stability.
So I see this as a much more shared responsibility that is going to require other officers and board members in healthcare organizations to become better educated, have a better understanding of the technology, and ensure that we form partners with our information systems staff. If they do that, they're going to start understanding the incredible complexity in this country as far as informatics goes. It is not easy to deal with legacy systems as you're trying to move the organization forward. That with that increased knowledge, and if we can create an atmosphere of partnership, I think there is going to be far more understanding, far more patience (as there should be) with the evolution of the information systems environment.
All too often, I think a minority today, but still, some healthcare managers think that bringing an information system is bringing in a black box thing that does something, rather than seeing it as a change in the basic infrastructure and fabric of the organization. So there is a disconnect. I've lived through that disconnect. I think it’s improving, but I wish in our industry it could improve more rapidly. If you go into the banking industry, airline industry, or other industries, I think there is a much stronger appreciation of the integral role information systems and the information officers play in the missions of those industries. I think it’s even more true for us because I feel (and I’ll argue this at any cocktail party) we are the most information intensive industry, aside from the intelligence services, in the world. I don’t denigrate the fact that banking is entirely relying on information systems, but I argue that we are too. If my ATM goes down, okay I can wait until tomorrow. But if a critical information source is not available, we are dealing with peoples’ lives here.
So when you multiply the information intensity times the seriousness of the impact, I argue that we are the most information intensive industry. It’s important, I think, for our executives to understand that it’s part of their work life, and they need to absorb the responsibility of knowing and cooperating and collaborating with the introduction of information systems.