Sun Microsystems Healthcare Mantra: Reduce Cost and Complexity
By John Russell
Sun Microsystems director of Healthcare and Life Sciences, Joerg Schwarz, talks about the need to change financial incentives to spur Health IT adoption, describes mistakes made by the giant U.K. health-IT initiative, outlines Sun’s healthcare strategy, and provides an assessment of Sun’s performance in the race by global IT giants to help improve healthcare and create new markets for themselves. Schwarz has been with Sun for 12 years and held his current position since August of 2005.
JR: Let’s start with one of the biggest problems which is that so few patient records are electronic and those that are can’t be easily shared.
Schwarz: The only place you do have electronic records today with a very high certainty are hospitals, but the hospital records are not linked together. Primary care physicians don’t have access to them in many cases, and they only capture a small portion in the lifecycle, basically, of a patient history. If you want one big topic: the digitization of healthcare, that’s the big transformation that’s going on right now. We see ourselves (Sun) in the midst of this; enabling with our technology, standards, software, hardware, and middleware, this transformation process.
JR: One of problems that persists is that most hospitals and certainly most small docs don’t have money to pay for the technology and don’t really see a direct benefit. In other words, why should they spend money on technology that may or may not result in a boost to their bottom line?
Schwarz: That’s actually a very, very good question and that goes right to the center of what I’m developing right now as our alliance for consumer-centric health. I don’t know if you read the [Michael] Porter book (Redefining Health Care: Creating Value-Based Competition on Results), but Porter is an economist and so am I, and actually one of the key findings is that economic incentives are not set appropriately to basically foster or to drive the transformation process that we’ve been talking about. Plus, a lot of the care providers benefit to a certain extent from some of the problems that we have. For example, if you are you are a government institution, then redundant procedures are a problem because they cost a lot of money and they drive some of the costs that we have in the system; however, if you are a provider, you don’t have an economic incentive to avoid these unnecessary procedures.
To make matters worse, you have even less incentive to invest in an infrastructure that would basically take some of these procedures away from you. It’s counterintuitive to expect a primary care physician to share his information with other primary care physicians or with the hospitals, which might lead to less business with him. That’s a conundrum, which in my opinion is one of the reasons why the RHIOs [Regional Health Information Organizations] have not been as successful as they should be.
How can we resolve this? So we formulated this idea of the consumer-centric healthcare alliance. Today, 80 percent of the claims are filed on paper. How can we incent the primary care physician to file electronically, because if we get the claims in electronic format, we do get a lot of information that we need to build an electronic medical record about the patients. So how do we get an incentive? Well, the incentives for the primary care physician would be that he or she can get the reimbursement when they complete an electronic claims process a lot faster than they can get it today.
From the information I have today, it takes the primary care physician up to four months to get reimbursed by the payer because the payer basically has to process the paper-based claim. Sometimes they upload them even to India to get them processed, and that takes a while. If you file electronically, you can reimburse the doctors within 48 hours, and that’s an economic benefit for them, which would give them an incentive to file electronically. Matter of fact, we learned that they would actually pay for this, similar to a credit card transaction where you say, okay, you have a claim of a $100 but you only get $98, but you file it electronically, and you get your $98 in 48 hours instead of the $100 in four months. So that’s step one.
Step two is that with that claims information at the payer, we can do a couple of things that we cannot do very well today, and the big topic on the payer side is called pay for performance. So they can almost see in real-time what the doctors do with the patients, with their subscribers, and they could map that against best practices. So if you have a patient with a diagnosis of A, you can define what is the best practice for a patient like that, and you can define different gradients of processes and treatments and so on that you define as a best practice. If the doctor matches that you can honor this with a pay-for-performance. So here you get an economic benefit for both the payer and for the doctor. The doctor files electronically. Information’s almost in real-time at the payer organization. Now the payer organization can check not four months later but almost in real-time if the doctor manages against a best practice standard and given economic benefit to the doctor if he or she does.
The benefit for the payer organization is clear. We know that if you manage a patient against the best practice, the medical outcomes [are better], and you probably do not have a health state that escalates to something that needs severe intervention. So here’s an economic benefit for the payer organization. We now can go and take all information that we collect from the different care providers, from the hospital, from the primary care physician, structure it against these management platforms and use it to basically get that information to the primary care physicians or the hospitals and to the consumers to let them manage their health. Again, we do have evidence that if you actively manage the health of the long-term patients you can actually significantly reduce costs. This involves calling patients if they miss a regular checkup, making sure that they fill prescriptions. You probably know that 50 percent of the prescriptions that are prescribed are not filled.
JR: I think that’s the largest class of prescriptions today.
Schwarz: Right, and matter of fact, if you really closely manage the health, you can know actually when a prescription should be refilled because you know if you have 30 pills, you need to take one a day. After 30 days, you have to refill it. If the patient doesn’t do this, he’s out of compliance, right? So you can start managing the compliance of the patient, and we do have data from several people who do this so there are a couple of BlueCross/BlueShields that experimented with this. If you do that, you can actually reduce the costs of the subscriber base, so this is an economic benefit for the payers. It’s also an economic benefit, and that’s the second group that is very interested in this kind of consumer-centric, consumer-driven health, which are the large employers, either the ones that are self-insured or the ones that pay a significant amount of money to the health insurance plans.
JR: I’m sure you’re aware there’s an Intel-led group planning to insist that providers they do business with at least make personal health records available to their employees.
Schwarz:Clearly, the employers and the payers have an economic benefit, so they are the most incented [to digitize most of the records]. Of course, [for an example of] the way it is done today oftentimes we look at WebMD. The patient inputs the data him or herself. That is unreliable. That is not the quality of data you want. You want data that comes from professional sources. And then, of course, there’s a side aspect of long-term care. Patients that are elderly or sick, and it’s difficult for them to go to the doctor’s office for a regular blood pressure check or for a blood sugar test or whatever, and you know that there are several companies now working on ways how to remotely access this biometric information, and add it to the health record. That’s an issue with the payers as well. You know, how do they reimburse telephone or internet consultation by a doctor, but these are the kinds of topics that need to be addressed if we really want to go to this, you know, tackle the problem of freezing healthcare cost.
JR: So what is Sun’s role in the thick of this challenge and opportunity, and what are the next steps and sort of timeline you envision to see substantial changes in the digitizing, if you will, of healthcare?
Schwarz:If you look internationally, different countries are in different stages in this because, for example, single care countries have a simpler model. I had a conversation yesterday with British Telecom, our partner in the U.K. Spine project and it’s very simple there because you have one payer, which is the government, and they just mandate to every doctor. If you want to get payment, you have to do A, B, C, D. They can just prescribe it, and that’s what they’re doing, so it’s much simpler there. In the U.S., it’s a complicated system, so I just wanted to add that aspect.
JR: The British example is not one without problems. There is a recent poll that suggesting strong patient objections to some medical record proposals. There has also been a lot of funding but not a lot of benefit produced so far.
Schwarz:Right. I think that one of the problems with the U.K. project is it started with a pure infrastructure. Now the infrastructure of the Spine is in place, but you [still] can’t use it for a lot of things. Now they will start adding services basically to the Spine, and as they add services to interconnect people, they will see the benefit. But outside of an IT community, people don’t really appreciate the benefit at this point in time. That goes back to the big discussion that we have here in the United States about RHIOs. Some RHIOs follow the central model. Some follow the federated model. I chose a centralized model, which naturally creates a lot of animosity by privacy advocates, by patients, by people who are just afraid of having all the data concentrated in one place and I don’t want to say who’s right or wrong, but these are the two fundamental models. You centralize everything and use that as a model, or do you have a federated model where you keep the data where it is. You just have to make sure that when you need it you can save it to the aggregate it together.
JR: What is Sun’s view on which is better?
Schwarz:We are Switzerland. We can enable both. You hit on one of the key things. If we digitize everything we need, of course, identity management because data protection to control who accesses information through the entire lifecycle. The best way to do this is building a federated identity management concept so that a doctor that is known and authenticated with one institution can request data from another institution where he is unknown, but that gives him doctor level credentials to access information involving a patient.
One of the biggest challenges that I see is we have not had a lot of standardization in terms of the data formats and data architectures. I think if we had more CDA2 (Clinical Document Architecture) compliant applications out there going forward it would make things a lot easier to combine information from different records, and actually that’s what we discussed yesterday with the people from the Spine project. That’s one of the things that they regret now that hasn’t been done right from the get go that you define the data architecture so you can easily map data from different sources together. You want to know if what basal temperature in one application and basal temperature in another application really means and how you map that together. That’s what CDA2 of HL7 version 3, but there are not many applications out there at this point in time that really follow this standard.
We define five focus areas where we basically bring in our technology. Number one is the health information network. That’s where the RHIOs and the client projects of the world go in, and that’s where we bring in our expertise and federated identity management in business integration and application integration. The second focus is identity-driven data management. That’s where we bring expertise and information lifecycle management. See, right now several organizations, countries even, are trying to build a countrywide archive for medical image data. We discussed that yesterday with a client. It’s actually an interesting approach. Instead of building the infrastructure first and then starting to add services, what I see now is that several institutions, several countries are approaching other clients with a particular service in mind
JR: Number three?
Schwarz:Number three would be the initiative for consumer-centric health that I mentioned before. We’ve collected a number of partners that contribute different services, for example, for the claims processing, for the claims data assessment, the mapping of this data against the disease management platforms, the actual management of the long-term care patients and so on and so forth. It’s a very broad alliance that can implement all these elements from claims data back to the personal health record.
JR: These are ISVs you partner with?
Schwarz: Exactly. ISVs and there’s a company in there called Wipro that does also a lot of the services, even call center activities. And we’re talking right now with one or two SIs to join the alliance as well.
The fourth area is the mobility security. That’s a specific topic for the provider sector where we have a concept that is based on ultra thin client and secure global desktop software. We basically want to address the fact that you should not have any device in provider institutions with the data footprint. You should keep the data safe in the data center and provide access to it on stateless devices, which could be the ultra thin client, would be a PC, could be a mobile device, and we have the infrastructure in the middleware to make that happen.
You see a lot of interest right now from hospitals for that because they just cannot afford the model of having hundreds of thousands of PCs in the wards where they have to support a very complex software spec with virus scanners, popup blockers, Citrix clients and so on, and spend a lot of money in people time on maintaining this. The Hospital Corporation of America had an incident where ten PCs got stolen. We had an incident here in Stanford where 5,000 medical records were lost. These are things that can be avoided if you keep the data in the data center.
JR: Okay, and number five?
Schwarz:Number five is compliance. Basically what we can do to help payers and providers to achieve and maintain HIPAA or FDA compliance with audit trails, secure email, identity management, access management, these kinds of things.
JR: Whose job at Sun is it to conceptualize, productize, and organize offerings into the healthcare space?
JR: How do you characterize Sun’s healthcare initiative? Are you early to the game, late to the game, and what will be milestones of success going forward?
Schwarz:Early or late depends on the segment. In consumer-centric health I think we are early and right on time. I think it was just beginning to take off, and we’re right there. In other areas like information lifecycle management, it’s definitely not a new topic and we have to gain market share. If you give me just the binary choice early or late, then I would say we’re more on the late side. In caregiver mobility, I think right now is the right time because people have this pain point in the cost.
The health information network [area] is actually something where we have a very strong position. We are involved in the two largest projects that are in existence right now. As far as I know, the U.K. project and we’re in the contract negotiation phase with Canada infoweb project in British Columbia. So we have some representation in the largest projects in the world and are involved in several smaller RHIO projects in the U.S. I think we’re right up front with this. Compliance, as you know that’s a tough cookie right now because a lot of the provider institutions are actually not as concerned about HIPAA compliance as you would think. You know, I recently read a statistic that only 30 percent claim to be HIPAA complaint and obviously 70 percent don’t have that big an issue with not being HIPAA compliant. But I think we are ready for this when the market picks up on that.
JR: If there were two points you would emphasize, what would they be?
Schwarz:Number one is we want to address cost and complexity. That is an impediment right now to the digitization of healthcare. You know, some of the initiatives that I mentioned address more the complexity point of business processes, business integration. Caregiver mobility addresses, for example, complexity and cost so that these are really two things that we try to differentiate ourselves. The second thing is standardization. I would encourage every CIO as they make investment decisions and no matter who they invest with to follow some standards like HL7 version 3 CDA2 that’s going forward. If CIOs are not asking for it the vendors, the ISVs, are not going be forced adopt standards.