After years of taking on responsibility after responsibility, John Halamka, M.D., CIO, Beth Israel Deaconess Medical Center, found the increasing demands of his many roles meant one would have to go. Considering he’s a doctor fascinated by the challenges of accountable care, it’s not surprising the CIO role of Harvard Medical School wound up on the chopping block. In this interview with healthsystemCIO.com — during which Halamka begins to caution government organizations that the industry is being asked to change too much too fast — he discusses Meaningful Use, ACOs, ICD-10, CIO leadership, and much more.
- Resigning as the CIO of Harvard Medical School
- “In 2011, it’s a different world for CIOs than it was 10 years ago”
- Why some brakes must be applied to the rate of change
- Does MU need some brakes too?
- Channeling the 200-bed community hospital CIO
- A call to postpone ICD-10 — “As I survey the country, CIOs are overwhelmed”
- ACO — “Nobody knows what they are, but everybody wants to be one”
… you look at the complexity, the regulations, the compliance, the infrastructure, the cloud, the mobile devices, et cetera. And boy, 2011 represents a different world for CIOs than it was 10 years ago.
If you think about being a healthcare CIO right now, you have the 5010 deadlines (Jan. 1, 2012), ICD 10 (Oct. 1, 2013), you have Meaningful Use Stages, 1, 2, and 3, you’ve got a swirl of compliance and privacy issues, and then you have healthcare reform. It’s not as if all of this is happening tomorrow, but it is certainly happening in the next 24 months. And so, yes, I better have some increased flexibility in that next 24 months to deal with these challenges.
The question will be what is the scope of Stage Two? And that’s where there may be some controversy because, as you look at some of the proposals around the level of patient and family engagement, it’s all extraordinarily forward looking, good stuff, it’s just how much can you do and how fast.
I really believe that, at the moment, there is too much change, too fast, and ICD-10 is the straw that breaks the camel’s back.
Guerra: Good morning, Dr. Halamka. Thanks for joining me to talk about a lot of things going on in your life. And I think we’ll start of by talking about your departure or resignation from the CIO position at Harvard Medical School. But thank you for joining me, first off.
Halamka: I’m absolutely happy to be here.
Guerra: Well, let’s talk about that. I wrote a column about it. It sounded like a very interesting decision. And the way you came about it was, to me, very revealing and instructive of how other people may arrive at these types of decisions. But take me through it — I know you’ve been at Beth Israel since ’97, when did you take on the CIO role at Harvard Medical School?
Halamka: It was in 2011, so I’ve been there for ten years. And I recognize that when I took that role in 2011, it happened in an interesting fashion, I had been asked by the dean of medical education to help out with the automation or courseware and to think about ways in which mobile devices and the Web could be used for all types of educational technologies, from courseware delivery to simulation. And so we did some really interesting innovation, and that worked out well.
I was then asked to become the part time CIO at Harvard Medical School. And that part-time CIO role was largely focused on education, administration and infrastructure. At that time, in 2001 and 2002, it was pretty straight forward. I mean, we didn’t really have huge storage or computing demands. But you can imagine how things have changed in the last 10 years, with an explosion of technology types and complexity, and then the research world, the human genome, new ways of doing image analysis and modeling and data mining.
So, what went from a half-time CIO position focused on a couple of academic areas really shifted to a more of a full-time position focused on supporting the research environment and increasingly complex infrastructure. So, that makes you scratch your head and say, “Well, let me think, should a $600 million highly complex organization have a full-time CIO?” Probably it should.
And, of course, my choice was do I come on as that full time CIO or, given my other full-time job on the hospital side, do I begin to focus on accountable care organizations, patient-centered medical home and what is going to be the future of healthcare delivery? And so, as you say, tough, tough decision. But I said, you know, “This accountable care thing is pretty scary and these are probably questions no one can answer. Let’s go the accountable care, healthcare delivery route because I’m a doctor. Let me find my own successor at Harvard Medical School over the next couple of months.
Guerra: So it was not clear cut that you were going to leave Harvard Medical School. You actually were going to do one position and you had to decide which one.
Halamka: And that’s true because it was clear, you know, I am a Gemini, so I should have two sides and always be able to do two projects equally well. But there are 168 hours in a week and you can imagine if you have 60 hours on one job and 60 hours on another job and you’re trying to serve your Federal Government and state and be an author and have a family, oh, gee, you’re running into the 168-hour-a-week limit.
And then your choice is either you shrink your breadth or you shrink your depth. And my worry was that given accountable care organizations and the future of healthcare reform will require unbelievable depth the only choice was to shrink breadth.
Guerra: Some people might not be able to let go of such a prized position. Was it very difficult?
Halamka: Absolutely. And so, one of my habits, if you look at my CV, is that as I have evolved in my career, I never give up anything. I just keep adding and adding and adding, and that worked probably in the early 2000. But you look at the complexity, the regulations, the compliance, the infrastructure, the cloud, the mobile devices, et cetera. And boy, 2011 represents a different world for CIOs than it was 10 years ago. And it’s really challenging just to meet the customer needs and the compliance requirements of one job, let alone two.
Guerra: You mentioned about how many hours there are in a week, but I hear you use more of them than most. Do you sleep significantly less than others?
Halamka: I’ve slept from, I think, three to four hours a night, every night since I was 18. And I don’t use any kind of stimulants (laughing), just stay even all the time. I drink green tea, which is low in caffeine, and have attempted to do a fair amount of exercise and have a vegan diet. And so, in between no alcohol, no caffeine and exercise and vegan diet, it helps. Maybe it’s genetic as well. The need for sleep is modest.
Guerra: Right. Now, this is just for personal interest. I’ve read a lot of historical biographies. And a lot of people who have made great successes of themselves are able to sleep on command. Now, do you ever have trouble falling asleep?
Halamka: So, I’m an emergency physician, and one of the things you learn when you become a physician, especially one that has to change your hours around the clock, is how to sleep on command. So, what you say is very true — I was in Japan two days ago and am able to change my Circadian rhythm by sleeping on command.
Guerra: So, you don’t have any wasted time lying there looking at the ceiling? You’re very efficient?
Halamka: That’s true.
Guerra: Well, that’s – I mean, people who are not good sleepers, they need a lot more time because they’re not getting that deep good sleep. So, it is a gift to be able to sleep deeply on command.
Halamka: Well, it certainly worked so far for me and my daughter. My daughter is 18 and she starts at Tufts University in three weeks. And she sleeps about five hours a night.
Guerra: So we said 50% of your time was at Harvard Medical School. Did they have someone doing some work over there? I mean, is there someone in that CIO-type role that was helping you out?
Halamka: The way that my management teams are organized and all of my organizations is that I serve as the person responsible for all strategies, human resources, budgeting. But I have a strong operations partner so that if somebody’s PC doesn’t boot at 3 AM on a Sunday, there are processes to deal with that, and I don’t generally need to get involved. Obviously, if there were something of a catastrophic nature or a privacy breach, of course, I’m right there. But you’re exactly right. If the CIO can be more of a people person and delegate the day-to-day operations of technology, it works much better.
Guerra: Have you actually concluded your duties over there?
Halamka: So, at Harvard, no, because I will be finding my own successor. The process is just beginning. And so, I imagine knowing that we want to find a world-class person, and especially someone who can deal with petabytes of data, thousands of servers and what is the emerging science of genomics, it will take us a couple of months. So, I imagine continuing to serve in the Harvard Medical School role till the end of the year.
Guerra: Obviously, you’re doing this to reduce some duties that are on your plate, some stress. So, you’re okay with keeping that up till the end of the year, but you’re looking forward to, I would imagine, offloading it?
Halamka: Well, what I see is healthcare reform is going to be a challenge by itself. If you think about being a healthcare CIO right now, you have the 5010 deadlines (Jan. 1, 2012), ICD 10 (Oct. 1, 2013), you have Meaningful Use Stages, 1, 2, and 3, you’ve got a swirl of compliance and privacy issues, and then you have healthcare reform. It’s not as if all of this is happening tomorrow, but it is certainly happening in the next 24 months. And so, yes, I better have some increased flexibility in that next 24 months to deal with these challenges.
Guerra: If I were to say that I’ve noticed a change in your writings, from being more of a champion of change to a voice of caution, would you say that’s true?
Halamka: So, we have three levers on our plate: the scope of the work we do, the resources we have to do that work, and the time we have to do that work. So here is why my writing has changed: Our time is accelerating. If you look at the compliance requirements, if you look at all the deadlines we’re facing, if you look at the changes of healthcare reform, it’s all happening day to day and not year to year. So, time is fast.
Resources, well, all we have to do is pick up the paper today and I hope your stock market portfolio is doing okay (laughing). But from 2008 to the present the economy has not been so robust. So wage growth and position growth has been pretty limited.
So, if time is fast and resources are scant, your only choice is to reduce scope. And so, we really have to deal with governance bodies, ask what are our highest priorities and what can we do well and what should we not do. And hence, the reason why I’ve said, “Let’s temper the amount of work we’re doing.”
Guerra: And I’ve seen you’ve written about ICD 10, you believe that should be postponed…
Guerra: You’ve written about the disclosure rule and really ripped that one apart, but just about everyone else did too (laughing). But you have not been outspoken against Meaningful Use. Will that change?
Halamka: So, when I think of Meaningful Use Stage 1, it’s all pretty reasonable stuff. And that is, if I think as a doctor, do we want problem list? Absolutely. Do we want medications and allergies, labs, do we want to begin public health and data exchange? They’re all incredibly reasonable. The question will be what is the scope of Stage Two? And that’s where there may be some controversy because, as you look at some of the proposals around the level of patient and family engagement, it’s all extraordinarily forward looking, good stuff, it’s just how much can you do and how fast. And so, I will watch quite closely to see how the Meaningful Use Stage 2 final regulations are produced. But, for the moment, I’m actually quite happy with the compromises on Stage 1.
Guerra: You are at a very advanced institution and you’re obviously a very advanced individual, is it important for you as someone in a position of national leadership to get into the mind of the average community hospital CIO? Is it important for you to understand what they’re dealing with when you are influencing national policy?
Halamka: Well, absolutely. So, if you understand my job, which is interesting, it’s that I oversee Beth Israel Deaconess. But Beth Israel Deaconess also has a 21-bed community hospital in the western suburbs of Boston called the Needham Hospital. And they have an affiliation with a 50-bed hospital to the south, Milton Hospital. So, I actually have direct reports at community hospitals who I have to help achieve Meaningful Use. And so, I actually live the community hospital experience every day. And you’re exactly correct, it is a very different situation. I have 300 staff, they report to me at the academic health center, but Needham has two and Milton has 10. And so, when you say, “Oh, I needed to speak with your network team,” the response is, “Oh, he’s on vacation this week.”
And so the kind of change, the speed of change and the complexity of change that an academic health center can do is quite beyond the community hospital. That’s why at the Standards Committee we say over and over, “Let’s engineer for the little guy.” I’m very, very sensitive to those issues.
Guerra: So Beth Israel was one of the first organizations to attest to Stage 1. One of the things that impacted me was the post you wrote about doing the EACH self-certification program. I think it read as if you were saying this is pretty easy, but what came across to me is that it’s pretty hard, especially when you said you recommended folks set aside five FTEs for two weeks.
Halamka: Right. So, the certification through CCHIT was great because CCHIT was such a helpful organization, but the problem was the scripts that were written were very challenging, and they didn’t need to be. So, what we’ve done at the Standards Committee is establish our implementation workgroup to review all future scripts that are written for clarity and ease of execution, because you don’t want to create, as you suggest, a burden that is so high when it doesn’t need to be.
So, yes, we got through it. But I think if you talk to people who have done EACH or talk to vendors who have just done the standard certification, they’ll tell you, “Boy, if you could have just re-written three or four of those scripts, you could have reduced the burden by 50%.”
Guerra: Regarding ICD-10, you’re very strongly in favor of postponing, what do you think are the chances of that being postponed?
Halamka: I have no inside information, whatsoever. It’s just that as I survey the country, CIOs are overwhelmed and Meaningful Use has taken the bulk of their attention, along with lots of new compliance issues. So, when you say, “Well, where are you on the ICD-10 project?” they’ll say, “Oh my god, you know, we haven’t started. We aren’t really planning. We just can’t find the resources to do it.”
I really believe that, at the moment, there is too much change, too fast, and ICD-10 is the straw that breaks the camel’s back. So, I just hope there’s a 2016 delay. And logistically, it just makes a lot more sense. ICD-10 procedure codes go up at such a factor of complexity that we need better clinical documentation if our coders are going to actually code accurately. So, wouldn’t we want our front-end EHRs to be fully rolled out and full featured before we change the coding system on the backend? It makes no sense to require coding complexity when you don’t have the clinical data to inform that coding complexity.
Guerra: I interviewed a CIO who said that at his organization, ICD-10 was going to be a governance problem because it hadn’t been assigned to IT or the CIO and it hadn’t been assigned to medical records. And it was just kind of hanging out there waiting for someone to grab it. I wonder if the nature of the project means that may happen at other organizations.
Halamka: What we did is establish a multi-disciplinary team across IT finance, medical records, nursing, lead by a steering committee that is co-chaired by the CFO and the CIO. So, you’re correct, it’s a governance challenge because it is not clearly under the purview of any one department. And we have really had to work together as a multi-disciplinary team in order to decide a scope, bringing in some external project management assistance and look at the tasks in front of us.
Guerra: Who decided to form the team? Who initiated that?
Halamka: The CFO and I.
Guerra: Regarding ACOs, I’ve spoken to CIOs who said that they really can’t even think about that right now because there is so much on their plates.
Halamka: Well, the problem with ACOs is no one knows what they are, but everybody wants to be one. So, what I’ve done over the last two years is based on recognizing that whatever the future will bring will require data and that we only need narrow tools and narrow registries. So, starting two years ago, we built a quality data warehouse for all of our ambulatory physicians — owned and non-owned — and we stipulate that your EHR must be able to send a summary of care at every transaction to the quality data center so that we will have the capacity to look at care processes across the entire community. So, that’s live and we have millions of records in it, and we have a set of tools that will allow us to now do pay-for-performance and outcomes analysis and process analysis.
So, at least on the ambulatory side, every physician in our entire community of referrers is aggregated in a single data set, and the hospitals have their own individual data set. So, we’re reasonably prepared for whatever ACOs will be.
Guerra: One of the things which CHIME took issue with regarding the ACO NPRM was the way they were dovetailing it with Meaningful Use. Does that bother you as well?
Halamka: Well, the challenge, I think, is that Meaningful Use is something that 80% of hospitals seem to be signing up for and 50% of practices seem to be signing up for. To the extent that you can control whether or not an individual achieves Meaningful Use, that is, in our case, we have these owned practices which are highly affiliated. In those cases, coupling may make sense. Where it doesn’t make sense is where you have groups that are beyond your control. So, one of the things I always think about is alignment of responsibility and authority — you’d never want to make me responsible for someone I have no authority over. That’s the problem with coupling, and it just depends structurally how you’re organized.