I am surprised we've not seen more comments regarding the new FL law regarding patient data and offshore data storage…
HIStalk Interviews David Lareau, CEO, Medicomp
David Lareau is CEO of Medicomp Systems of Chantilly, VA.
Tell me about yourself and the company.
I’ve been here since 1995. I discovered Peter Goltra and Medicomp when I was in the billing business and had a customer who wanted electronic health records.
We provide the MEDCIN engine and software. Our sole purpose in life is to present the relevant clinical content to a clinician at the point of care so that they can treat the patient, get their documentation, and have all the billing and Meaningful Use stuff happen in the background so they can focus on the patient, find the information they want quickly, treat the patient, and get on to the next patient. Not slow them down or get in their way.
You have quite a few physicians developing content and helping design the Quippe product and you’ve recently hired Jay Anders as chief medical officer. How do those physicians drive product direction?
We have a knowledge engineering team that, going back to our start in 1978, sits with the physicians and says, you’re treating somebody with asthma — what are you thinking about? What are the symptoms, history, physical exam? What are the tests and therapies? What are the other things that intersect with asthma? What are the co-morbidities? What would differentiate asthma from something else that has a similar presentation? It’s an endless series of peeling away the onion.
The questions that we have for the providers are, what would you want to see given this presentation? Some people think we’re trying to tell the docs, "Here’s what you should do." We’re presenting back through software what the doctors have told us they would want in that case. As you might imagine, it’s an iterative process. It never ends. Things are always changing.
We have anywhere from 15 to 20 physicians active at any point in time. They work with our knowledge engineering team. Jay Anders has joined us recently, because as you add more content to anything, it can tend to complicate life for the user. The more concepts you add to things like SNOMED, to other terminology sets — you’re seeing it with ICD-10 now — more content puts more pressure on the provider of software to make it usable at the point of care.
Jay Anders came on board because he represents the clinical end user for us. OK, Jay, I want your input on everything we do. Are we making it easier or are we making it harder? What should we be showing? What options do we need? How do we let the users control what they’re doing without slowing them down and getting in the way?
Putting more information in front of a user is not always the answer. It’s the right information at the right point in time. Does the engine have the content? Do the UI tools that we’ve built around it to help people deploy it provide for a proper presentation that the docs like, allow them to focus on the patient, and get all that other stuff in the background? There’s a lot more interaction.
One of the things Jay is doing for us is defining ways to do better work flows at the point of care and also recruiting our clinical advisory user group of physicians, not just a knowledge engineering group. We have two teams that work in parallel.
How many employees does the company have and what do they do?
We have 20 employees now. The last time we talked, we probably had about 10 or 11. We have seven people who are developers. We have three people who do terminology, stuff like mapping to ICD-10, SNOMED, LOINC, Meaningful Use, etc. We have three people who do testing. We have three people in product management. We have three or four people in knowledge management.
Then we have the clinician advisors. We have two-full time physicians on board doing that. Then we have about 10 or 15 who are on staff at major medical centers. They’re not employees, they’re contractors. They do a lot of work with our knowledge engineers on the knowledge base using our knowledge editing tools.
Peter Goltra had a great idea in MEDCIN, but it didn’t feel like a real business early on. What does it take to turn a great idea into a great business?
Focus. Absolute core focus on what you do, what you do well. Any time you’re in the kind of business that we are in — development of intellectual property, development of content, development of techniques to present things — your sole asset is your people. You’ve got to find the best people and you’ve got to keep them.
I consider salaries the only expense I will never cut because those are the people who produce what we have that is of value. Everything else is negotiable. I can move into smaller office space. I can do less travel. I can have non-fancy furnishings, which we do. But we want the best people. We never want to lose anybody, because when you lose somebody, it slows you down. You lose their energy and other people have to make up for it.
One of the things that happens in other companies is that they don’t focus on one thing. As they get successful, they start doing things they shouldn’t be doing. When bad times hit, they cut their head count. Our head count is our asset. That’s it. I’m not talking about in terms of numbers, I’m just talking about terms of quality. We pay people very well. We treat them very well. We contribute six percent to their 401(k) whether they do or not. We recognize them. We listen to them. We empower them. They love working here. We don’t lose people. That gives us continuity. That allows us to build in successive versions of what we do, on what we had before.
We do not become unfocused by saying, “They said we should do dental software.” Somebody else says, “Why don’t you guys do a drug database? or somebody else says, “Why don’t you do this?” No. We provide an engine to present relevant information to clinicians at the point of care. That’s what we do. If it doesn’t have to do with that, I don’t want to do it. That’s it. Great software gets produced in small, very collaborative, highly productive teams of experienced people who know what they’re doing and are very committed to it.
I assume Medicomp has been around long enough that you don’t have impatient investors demanding that you do something that sacrifices long-term success for short-term profits. Do you see that happening with other companies?
Absolutely. Anybody that has to answer a quarterly conference call is under that pressure. Any time there’s a blip in earnings or revenue, they really can’t do long-term investment at the expense of short-term results, so they cut people.
We’ve seen some of our own licensees — I won’t mention any names, but the news shows up in HIStalk all the time — they cut and then they hire and then they cut and then they hire and they rearrange and they right size and they downsize and they expand and they cut. It’s a tough way to build a business.
Our advantage is that Peter got into this because he loves what he’s doing. He carried the company through from 1978 to about 1992. We are owned by employees, Peter, and some family members. Everybody here is a stockholder. We’re all invested in the same thing. We all have a long-term vision because we believe that eventually, these systems have to be usable by clinicians at the point of care, and right now, they’re not. That’s why we’re starting to make some inroads.
Are EHR vendors are concerned about usability issues given that most of their development agenda is sucked up by ICD-10, Meaningful Use, and quality measures?
They’re not concerned about it yet. There’s a couple of reasons. The government has just pumped $30 to 40 $billion into HIT. They said that in order to qualify for this money, these are the things you have to do. That’s been a great boon to the sales, revenue, stock price, and valuation of the big vendors because here it is — just do this and you’ll get it.
At the same time, ICD-10 CM is no picnic and neither is Meaningful Use. Those things are so challenging that many physicians have said, well, to heck with this — I’m going to sell out and become an employee. Then they become disempowered by the organization. It’s happened here in northern Virginia. There’s one health system that dominates. They’re buying up practices left and right. They don’t have to listen to the docs right now. They haven’t had to listen to them for a few years because they’re doing great. They’re addicted to this money, which has let them do what they do, not have to adjust, etc. The docs aren’t really empowered, so usability, schmoozeability, we don’t care. It’s not a factor yet.
We think it will be, which is why we have folks like Phoenix Children’s Hospital coming to us and saying, our vendor’s not delivering on usability. Our docs need something they can use. Can we give it a shot with your software and put it in? And they did. Their vendor, Allscripts, tried to talk them out of it, but ultimately cooperated with them. They put it in. Within nine months, their docs love it. They’re seeing 30 percent more patients per day and they’re leaving early.
I believe that once the tsunami of money coming in dries up, they’re eventually going to have to turn back to, how do we make doctors more productive? Particularly given that with the new health insurance laws, there are more patients to treat and possibly fewer primary care docs to do it. As we go to outcomes-based reimbursement, they’re going to be paid for how well they care for patients. We still think that’s going to happen one patient, one clinician at a time. You need to be able to efficiently provide care, so at some point, you’ve got to make this usable by the providers.
That’s what we’re hoping. We’re starting to see that. We’re starting to get some traction in that. And as you said, we’re a more patient company than most.
I thought your business was working through EHR vendors who signed up to embed your product into theirs, but Phoenix Children’s went their own direction. Will you offer Quippe or the MEDCIN engine directly to customers without their vendor’s involvement?
Well, possibly. I don’t want to do that. I want to go through the vendors. But Phoenix came to us. They asked their docs, “What do you want for documentation?“ They did about a six-month analysis with the docs. The docs found it. They presented it to Allscripts. Allscripts said we’re not going to do that yet — we might have something in two years.
Then they came and played our Quipstar game at HIMSS and said, you guys have what we want — can we try it? We said no, we don’t do that. They said, we’re a co-development site for Allscripts. We have access to their code. We’ve convinced them to let us do it. They think we’re going to fail, but they said they’d let us do it.
They had a great team, which is why I don’t want to do it with many other people. They had the best team I’ve ever seen. There’s a reason they got that award a few years ago as the best IT department. David Higginson is a demanding visionary leader, I’d call him. He had one programmer work on this part time for about nine months. They did the full integration with SCM. It went very well, beyond their and our wildest dreams. We had to back up and say, hold on, what are we going to do now?
We’ve done a couple of things. We learned a lot in that process. We’ve made it much easier to integrate Quippe with an existing system. As evidence of that, the next thing we did was when Bangkok Hospital in Thailand came to us. They have an IT subsidiary called Greenline Synergy. We’re getting some really good traction in Asia, in the Asia Pacific region, but we’re not implementation and training people. That’s not what we are and that’s the danger for us. When I talk about focus, I don’t want to do that.
They came to me and said, we want to do a little pilot. We want to see how quickly we can take Quippe and stand it up in one of our ambulatory clinics, and if it goes well, we will consider becoming your implementation, training, support, and distribution partner in Southeast Asia. Because we already have Bumrungrad Hospital live on the nursing stuff and we’re getting a lot of traction in Malaysia, I said, OK, let’s try it. They came here on April 27. They sent three people — two developers and their clinical lead. They spent two weeks with our team. They went back. They got back to Bangkok on May 15 and they are now live in their ENT clinics with Quippe for physician documentation.
Is the product the same no matter where it’s installed? There’s nothing that needs to be localized?
It needs to be localized. We’ve had to build in some options for people that allow for localization. We did a project about four or five years ago where we said, if they really want to present it in local language, we will never get caught up, because there are 300,000 concepts, positives and negatives for each, and multiple presentations of each. But we did a little study and found that about 10,000 meds and concepts constitute about 95 percent of all documentation activity. Common things are common.
We did that. We did a translation into Thai, Chinese, and Spanish just as a test. When it came back, people said, we don’t really care about that because we operate in English a lot, but we use different forms of things in English. In Australia, they say "nappies" instead of diapers. We had to build in some additional tools to say to people, you can replace things by user, by site, by specialty, by country, etc. It’s sort of a localization pack.
We also have made it much easier for people to change the way that the engine behaves, because infectious disease things in Singapore are taken much more seriously than they are here because of the density of the population — they don’t want hand, foot, and mouth disease getting loose in even one building because it’s so contagious, so they want the software and the engine to work a little bit differently. They want to promote those things and get them right in front of users. It’s similar to what happened here when everybody decided we had to ask a few questions about Ebola. Think about that as a massive localization at every hospital in the United States for a while, although it’s kind of died down now. We’ve had to put in tools that make it much easier to localize our content and localize the operation of the engine.
What will be the biggest factors impacting healthcare IT over the next five to 10 years?
The concrete is poured, in the United States at least, for people who have spent the time and money to put in the systems that they have, which are heavily based on transactions, billing, and organizing admission-discharge-transfer stuff. This is our hope and this is our plan — that attention will turn back to, what are we going to do? How are we going to make all of this big data that everybody’s talking about actionable at the point of care?
People are going to take a couple of different approaches. People like IBM with Watson, people working on all the natural language processing stuff, big data, all that. They’re going to approach it from the standpoint of, we can analyze all this information on the population and we can detect trends. Now whether they can do just correlations or causation, I don’t know, but at some point, if you’re going to improve outcomes, that stuff has to come back and be usable at the point of care.
We think that’s our opportunity. That’s what we provide. If we’re wrong and nobody cares about that, I probably won’t be around for you to talk to me in 10 years. But we think it’s turning because we’re getting more and more people come to us and say, can we do what Phoenix Children’s did? We spent all this money, our doctors still can’t use this stuff, we’re not getting the data we need, and we’re not pushing it back to the point of care – please help us do that. I think after the dust settles with ICD-10 and Meaningful Use, the industry is going to turn back and say, we’ve poured the concrete, now how do we build a road that these docs and nurses can use?
Good interview. I particularly liked his comment about staff. The real investment in software is in the people and it is called ‘intellectual capital’, but you never see it as an asset on a balance sheet. In health care it takes years for a system designer /analyst/ developer to really understand the problem to be coded in all its permutations. When you downsize, for whatever reason, then upsize you have to repeat that big intellectual capital investment. Its months, if not years, more than a quick rehire with a month or two learning curve.
I have seen and experienced way to many software companies fall in to this trap when they get the ‘wall street you got to make the number’ pressure. Give into that pressure and it’s all downhill from there!