Medical Device Link.

Originally Published MX November/December 2005

INFORMATION TECHNOLOGIES

Industry experts look at the current status of IT adoption for medical products—and for the healthcare system in general—and examine what developments are next in line.

Moderated by Steve Halasey

Throughout 2005, MX has featured a special series on information technology (IT) in healthcare. Operating under broad definitions of both IT and healthcare, the series has explored business systems, manufacturing systems, embedded software and systems in devices, as well as what end-users are seeing in their facilities. All of these elements have an influence on the way medical technology executives develop their product lines.

Over the course of this year-long journey, MX has identified three broad categories of stakeholders in healthcare IT efforts: suppliers of manufacturing and business systems to medtech manufacturers, the makers of next-generation IT-based medical devices and systems, and developers of practice-management systems and the government agencies involved in the federal electronic health record (EHR) initiative. Those within each category have their own pattern of understanding, goals, and capabilities.

For this final installment in the series, MX brought together a panel of healthcare IT experts for a roundtable discussion moderated by MX editor-in-chief Steve Halasey (see sidebar). In this excerpted version, participants discuss the current shape of the field, challenges related to the adoption of healthcare IT systems, and how medical device manufacturers should be planning to do business in the emerging IT-rich healthcare environment.

MX: Does MX's view of IT in healthcare jibe with your own? How do manufacturers and other stakeholders envision current healthcare IT efforts?

David Merritt: I would say that assessment is fairly accurate. In the healthcare industry there are so many different players and uses of technology that it does seem daunting that we are trying to interconnect every single stakeholder.

My boss, former speaker of the House Newt Gingrich, does half of his work in healthcare and half of his work in national security. Having lived in both worlds and having a fairly good knowledge about how each sphere works, he says that healthcare is 30 times more complicated than national security. So I agree that there are disparate stakeholders, and their use of technology is going to be different.

But at the same time, I don't think there has ever been a technological obstacle that we as a country have not overcome at some point. Creating an interstate highway system, building nuclear power plants, or sending a man to the moon each seemed daunting in its day and age, but we overcame the challenges because we have entrepreneurs and we have the smartest people in the world. Building an interoperable and interconnected healthcare system is not going to be any less difficult, but I don't think we are going to be any less successful.

Jay Srini: The groupings you've suggested certainly represent major divisions in the field. But somewhat ironically, in a lot of provider organizations biomedical devices are handled entirely under the umbrella of individual departments. So, for example, the radiology and cardiology departments may have complete control over the devices used in their areas, and the organization's chief information officer (CIO) is not involved with them at all. To create a longitudinal electronic health record system, the data from medical devices and electronic health records will all have to be treated as a single, seamless entity, and this will certainly have organizational implications.

The disparate systems that need to be interconnected are vast and in some ways resemble the financial systems of the past. The fundamental difference in healthcare is that the economic engine works very differently. There is no incentive for a physician to invest in the technology because he has no opportunity to reap the rewards of that technology. Imposing the costs of information technologies on the provider community, when it is the payers and employers that are actually reaping the benefits of those technologies, has created a sort of market gridlock.

Moreover, the difficulties lie not just with the technology, but also in the imposition of the technology into a work flow that must be radically modified to make use of it. A simple example: today a physician drives from one hospital to another and has a nurse call in prescriptions or write out e-prescriptions. That administrative process may cause severe quality problems. But if, to resolve those problems, you require the physician to use a handheld system and to personally write each e-prescription, you are significantly changing the way that he works.

The challenges lie partly in the realm of technology, but more importantly in the work flow, processes, and incentive mechanisms needed to make the economic model work.

David Lansky: I would add a cross-cutting dimension, which is the increasing capability of individuals—consumers, patients, and families—to be players across these three categories. I mean this both in the sense of managing health information by supplying health information into the network or professional environment, and, of course, managing the financial component by becoming increasingly responsible for higher portions of payment.

The Markle Foundation (New York City) interconnects with these issues in two places. One is the interoperability agenda. In the past, the United States has had success with some big initiatives to undertake projects or create networks of various kinds. These have usually resulted from some kind of coordinated federal impetus. But in the healthcare environment, we are not likely to see a major federal investment or mandate. So instead, the Markle Foundation is looking at a variety of strategies, including collaboration between the public and private sectors and collaborative funding models across those sectors, but with relatively little federally directed programming.

As a result, one of the things the foundation works on is bringing together a lot of the parties to sort out how we can all work together toward implementing connectivity. This effort involves the foundation in a variety of collaborative activities to develop standards and models that everyone can regard as constructive, whether they are on the technical side, the policy side, or the financing side.

A second leg of our work is to explore, as the interconnected environment begins to take shape, methods for ensuring that consumers become more actively involved in managing their own health and navigating their own healthcare. This is an area in which the new information technologies—including related device technologies—show a lot of potential, because most of them somehow involve connecting the individual healthcare consumer to an information flow and an information network.

Srini: I couldn't agree more. These systems create an unbounded healthcare environment in which healthcare pervades daily living. One example of this trend is the development of monitoring devices that enable consumers to combat or manage chronic disease, or, even better, to prevent it. Consumers are going to become more and more involved in taking care of these things. And, because of the new financial models, they will be responsible for the financial aspects of healthcare more than they have ever been before.

Thomas N. McCausland: At Siemens Medical Solutions (Malvern, PA), we look at it from the healthcare provider's point of view. Our hierarchical systems generally correspond to end-user applications, such as practice management, ambulatory care, or inpatient information systems. And each of these incorporates structures for handling both clinical and financial data.

Our approach is to try to make sure everything is integrated—interfaceable and ultimately integratable. When it comes to devices on a hierarchical basis, large devices and lab equipment clearly must have user-friendly operating systems. And then, once customers retrieve information from the devices, we have to consider where they put it, how they interface it with the rest of the records, and how they distribute that information to the people who need it to make decisions on diagnosis or therapy. A large part of our approach also deals with real-time collection of information at the patient bedside and, ultimately, discharge orders.

Other than recordkeeping for FDA and process-quality assessment, we do not view manufacturing systems as healthcare components. We look at them as just manufacturing technologies that have some quality twists specific to healthcare. They really are the same, whether you are making jet engines, turbines, programmable logic controllers for industrial use, or healthcare equipment.

Tim Gee: Medical device manufacturers are client focused, although Siemens is an exceptional company in many ways, certainly in the scope and breadth of its product offerings. Most medical device vendors do not have the luxury of having an IT division that can help them understand the technology and its requirements from that perspective.

Lori Hack: What we are hearing from our partners—which are mostly large multihospital delivery systems—is that most manufacturers are not thinking through the systemic issues related to their devices once a hospital obtains them and tries to install them into their facility. When a hospital purchases a device and installs it in a particular unit, for instance, the systemwide implications for the hospital's IT security are often not really well understood, thought through, or documented. From a security standpoint, there certainly are not standards set up for these devices or how they should integrate with other devices the hospital system might already have in place. We are a long way off from the big-picture goals of providing seamless integration and secure systems.

Mark Leavitt: End-users do not have the resources to integrate their devices, their software, and their hardware with the networks within their healthcare systems. We lag behind other industries. In the auto industry, it simply would not be acceptable if you bought a robot that assembles cars and discovered that it could not talk to your computer-aided manufacturing system. That would not fly for very long.

But in healthcare's case, some of these hospitals and offices are tiny, 10-bed hospitals in Montana or single-doctor offices, which make up a third of all doctors' offices. And these small operations are trying to integrate complex systems that can often require up to $50,000 worth of individual consulting. We have a lot of work to do to create the standards and then have the devices and the software systems comply with the standards. If we don't get closer to plug and play, we are never going to realize the safety benefits of both digital devices for monitoring and treatment and software for managing the information. It will forever remain a safety-risk area if we do not get the standards implemented and the systems working together smoothly.

Rod Piechowski: This is a real opportunity for healthcare to go through a complete redesign. IT is considered by the National Alliance for Health Information Technology (NAHIT; Chicago) to be a tool rather than an end solution. When you have an opportunity to redesign something, you want to know what the vision is and what it is that you hope to gain through the redesign. Then you look at how IT supports that reorganized vision.

The larger picture extends well beyond just the organization that has installed a system for x dollars. There is a cost to the culture. There is a cost to maintaining that system or those interfaces that must be customized so they are interoperable. These relationships with the vendor exist far beyond just the implementation phase, and they have a softer cost to the culture of the institution that has made that implementation. That is one of the things that our members tell us they have a very difficult time measuring: that softer cost to the organization. How will it really change things and how do you measure the return on investment (ROI) for these types of things?

A lot of people agree that technology is in a great position to handle a lot of what we expect it to do. But not everybody has agreed yet on exactly what it is we want to accomplish and what standards we need. We are seeing a lot of organizations that are now starting to share more information and talk about this more openly, however, and that's a great step in the right direction.

Disconnections

There seems to be a disconnect between companies that design and develop IT systems, medical device manufacturers, and end-users. Is this problem real or perceived? Do the differing views and goals of healthcare's various stakeholders represent a problem in their own right?

Piechowski: It is a huge problem, and people are just beginning to grasp the nature and size of the problem. Whenever we talk about who the stakeholders are in healthcare relative to any particular issue, it is easy to come up with a handful of them. But then healthcare people have a natural tendency to segment themselves, and you start getting substakeholder groups, each of which believes that it has a unique way of doing things. That makes it far more complex when you try to integrate all of this and find some method of implementing IT in such a way that everybody benefits equally.

Gee: For both buyers and manufacturers, there is a requirement now to look at a much broader scope than was traditionally considered when making a decision either to define and develop a product or to acquire and implement a product. For a hospital, radio-frequency identification (RFID) systems are a perfect example. You can take the traditional, narrow view that 'It is my department, and I have to solve my problem,' and end up proliferating three or four different RFID systems throughout your hospital. Some of them will consequently overlap one another. Or you can broaden the scope and say, 'OK, we have a need for a general capability, like asset tracking or patient tracking. Let us assess needs on a broader scale and see if we can find a solution that we can leverage not only for asset tracking and medical engineering, but also for the operating room and for patient tracking in labor and delivery.'

There is a converse to that for manufacturers: medical device companies have always thought in terms of embedded systems, and now they are faced with this whole new area of general-purpose computing technology. They aren't fully aware of it yet, and they haven't yet figured out how to interface with it. One of my hospital clients had to send back about eight wireless electrocardiogram (ECG) carts because the manufacturer did not build-in the right kind of wireless local area network (LAN) security. Using those devices would have compromised the hospital's entire wireless LAN security scheme, so the hospital had to pass on them.

Hack: We certainly have heard stories to that effect as well. For example, a healthcare facility might purchase a picture archiving and communication system (PACS) that cannot accommodate the new 64-slice computed tomography (CT) data output. Or the manufacturer might not be prepared to integrate and connect its devices to a physician's office where the level of security is not adequate or the bandwidth is not large enough to accommodate the system.

Taking a systemwide perspective is really the key. From the level of the chief financial officer throughout the whole system, it is critical that the delivery system and the manufacturers understand that innovatively solving the health delivery system's business problem may require fresh thinking. They have to look at a device that typically was used in a single department and think about its use across the system. That leads to the issues of connectivity, performance, and integration, which have never been considered either by the manufacturer or by the end-user.

Lansky: Well, the disconnect among stakeholders is a reality. This is a very large, complex sector of American life and the American economy. To make it work at all there are going to be many moving parts with different business criteria and objectives. It's a challenge to get all these parts moving in a common direction absent very strong coordinated leadership, which we don't have. So there are people of great goodwill in various consortia, associations, and other mechanisms trying to seek out areas of interest in which we can all move in a common direction.

There is a consensus that, absent financing reform, it will be very difficult to override the short-term business incentives of these players—whether they are hospitals, doctors, medical product suppliers, or drug companies. They all have pressing business interests that they need to address. Without some reorganization of their financial imperatives, they all say it will be very difficult to make significant progress on the government's EHR initiative.

Are executives in start-up medtech companies or healthcare providers in a hospital setting aware of all the complexities surrounding these issues, or is the whole EHR initiative just something that their boss or the CIO handles?

Srini: There is an EHR adoption gap. Large institutions can easily absorb the investment required to implement EHR systems. But meanwhile, the sole practitioner down the block is basically just trying to maintain his small practice and keep whatever cash remains and, with ever-decreasing reimbursement rates, is often just managing to keep his head above water.

The independent physician may have heard a lot of noise about EHR systems. But because there are no standards for certifying such systems, the physician doesn't know if the technology he is considering will interoperate with other systems or might rapidly become obsolete. He can't really even be sure that the system's vendor will remain in existence over the next year, because there are so many start-up companies in this area.

Device manufacturers—even large device manufacturers—have been living in a world of their own. Many large companies and their products are susceptible to major problems over even the simplest of things—like virus protection—because the manufacturers do not know how to address such issues as protecting the software on their devices or applying necessary software patches. This is becoming a very critical area, because manufacturers have previously not had to deal with information technologies outside of their own devices' embedded systems. Now, as devices become part of a larger health technology network, they need to consider the impact of their devices on the whole network.

So on one hand you have physicians who are being bombarded with information about multiple technologies, yet are only vaguely aware of what they're dealing with. And on the other hand, you have device manufacturers that are focused primarily on meeting the expectations of their investors, and are therefore not necessarily concerned about the interoperability of their products.

Given this scenario, it's worth asking where the ROI for interoperability is supposed to come from. Who's going to drive the ROI for interoperability?

If manufacturers do not have in-house staff addressing issues of interoperability, are they able to locate software development firms that are sensitive to these issues?

McCausland: Yes, but this is a rapidly moving target. Less than 10 years ago, our devices were not even digital. Getting into the digital world has created a lot of opportunity within customer organizations.

If you think back to the mid 1990s, the IT department of a hospital handled the financial and administrative sides of the operation. Then as the market began to see more clinical automation, IT departments had an opportunity to do a lot of things. But it was a best-of-breed market where our customers ended up with possibly 30 different IT systems. Then, after trying to manage and support them, somebody would say, 'Well, this all should be linked together.' So then hospitals had to make sure that the Jane Doe who started out in registration is the same one who is receiving treatment in the radiology department. And suddenly all of the equipment that hospitals buy has some kind of digital capacity and it is dumping huge amounts of information into a network that people never thought about designing to handle that.

So we have this explosion of data and information capability. And manufacturers have chosen certain kinds of platforms on which to build their IT operating infrastructure, and with that comes complexities. If you use a particular operating system, how do you handle the bugs that may be in it? How do you handle compliance with the Health Insurance Portability and Accountability Act (HIPAA)? How do you handle hackers? How do you handle viruses?

And on the user side, the IT guys who are running the hospital's system don't know what platform a manufacturer's operating system is running on for its magnetic resonance imaging (MRI) device. And they don't know what will happen if that MRI is hooked into their network. Hospitals could be very vulnerable to viruses that can't get into their normal IT system, but can infect the device operation.

This is probably the most complex industry in the United States, and it is moving rapidly into the IT 21st century. I don't think that anybody on the customer, manufacturer, or software side today has really put it all together.

Siemens and, to some extent, GE Healthcare (Chalfont St. Giles, UK) are wrestling with these issues. But they are very, very complex. The customers have underinvested in this area for years and years. We are all struggling to make sure that everybody is aware of the issues. And we're trying to drive the wisdom of making everything connectable and integratable. No one in a not-for-profit business can sustain the amount of infrastructure necessary to keep operating the way we have done it in the past. It is just impossible.

Standards Progress

Are the organizations that are developing standards moving fast enough organizationally to keep pace with the 'moving target' of healthcare?

Leavitt: Some feel we are moving too slowly, and others feel we are moving too quickly. We have been slow to agree upon standards to solve all of these problems because they are complicated. A bigger reason is the financial incentives are not in place to encourage the adoption and integration of pervasive IT. Physicians only get paid more if they do more procedures. They do not get paid more for accurately transmitting information to prevent an error. Therefore, standards development has moved slowly, and at times there have been internal conflicts and competition between different standards that each claim to solve a problem.

At the same time, in the last two years, we have seen a rapid advancement of the federal government's role in leading both the public and the private sectors to develop and deploy these standards. Besides my work at the Healthcare Information and Management Systems Society (HIMSS; Chicago), I am involved as chair of the Certification Commission for Healthcare Information Technology (CCHIT; Chicago), which is bidding on a federal contract to certify electronic health record systems. Although in the first year that might not include certifying their interface with devices, it certainly would in future years. There is also another federal contract that should be awarded soon under which an organization would harmonize the standards when there are multiple standards and conflicts. So we are seeing rapid progress, but everyone is still skeptical that we can really solve this problem in the next year or two. It is probably going to be five or 10 years before we achieve the ultimate vision.

Srini: Whether industry is moving fast enough depends upon what you are looking at, but I don't think it depends on the size of the company. There is no motivation for large, entrenched companies to look at interoperability standards because they would rather sell customers their surgical suite as well as their emergency room suite. Such companies' products may be interoperable in their own little world, but they are not necessarily compatible with other manufacturers' products. And there is not necessarily a market push, because the stock market has not punished manufacturers for not being interoperable.

Smaller vendors start out with a clean slate. As a result, they are not migrating from older standards, and from the beginning they are able to implement newer, service-oriented architectures built with interoperability in mind.

So the size of the organization is not as critical as how attuned the vendor is to its product functionality, and how much it wants to implement interoperability with other arenas.

Industry is not moving fast enough. That is the reason that the U.S. Department of Health and Human Services' national coordinator for health information technology, David Brailer, MD, PhD, has been working to lay new foundations. The first area of concern is to harmonize existing standards. There are many standards already in existence, and we have to determine which ones we are going to accept or how to meld them together. And once we have adopted appropriate standards, then we have to identify parties that will certify the compliance of EHR systems to those standards.

Interoperability makes the greatest sense when consumers also have access to their own health information. But along with that access come concerns about the security, confidentiality, and privacy of the information. Those are all areas subject to federal requests for proposal (RFPs). Another RFP will look more broadly at how to create an infrastructure whereby all of the nation's regional health information organizations (RHIOs) can be integrated and brought into alignment with the goals of the national healthcare information network.

Consumers want transparency in the healthcare marketplace. They want to be able to choose the best hospital. That judgment needs to be made according to reported data about quality of care. Quality of care is intimately linked to whether physicians have the full information they need when diagnosing and treating their patients. And some of that information has to come from the devices produced by manufacturers feeding directly into the EHR. But while medical products play an important role, the real focus of the national healthcare IT initiative is on consumers. So at this point, healthcare product manufacturers are not taking the lead role in developing interoperability standards.

The healthcare sector as a whole is often cited as lagging behind many other industries in terms of IT adoption and ultimately having underinvested in the field. Hospital administrators will argue that they have invested plenty, only to have to scrap their systems because they did not work, did not meet their needs, did not match current standards, or were otherwise dysfunctional. One of the things that they point to is the absence of standards against which they can match their equipment and practices. Is the healthcare community resistant to adopting information technologies because of its past experience with expensive, unworkable, and obsolete systems?

Lansky: I think that is definitely the case. The interoperability work that many of us are involved in is intended to create widely adopted standards that will reduce the risks for providers who make those investments. They will have confidence that—whatever might happen to their current vendor or technology platform—the underlying data conform to accepted interoperability standards and can be moved to another platform if necessary. We hope this work will also stimulate the development of a more durable EHR industry. So this is certainly one of the reasons that HHS secretary Michael O. Leavitt and others are paying so much attention to establishing a better understood basis for standards.

This isn't by any means the final word on the barriers to the adoption of health IT. The fact that health IT may require a change in work flow is itself a significant barrier. Not many providers have spare time in their day and want to spend it reengineering their practice to support an IT system, and fewer still see the benefits of doing so.

A paradox that the field hasn't dealt with sufficiently yet is the fact that healthcare IT initiatives are not primarily about technology at all. Success is not going to be achieved by merely buying high-end hardware and software and plugging it all in. For example, there is a fear that if we were to simply create an IT exemption to the federal healthcare program antikickback statute and federal physician self-referral law (commonly called the Stark statute) without establishing standards for interoperability, we might end up merely propagating a lot of technology that would still not achieve the important goal of reengineering the healthcare system.

It's essential that we look at the issues of work flow, system redesign, and interoperability—but as part of a larger strategy of changing the healthcare system and not only as a basis for adopting raw technology. There are a lot of people around the country who have adopted health IT and made almost no changes in their work flow or practice, and therefore have introduced no benefits to their patients. To ensure this doesn't happen on a national scale, we have to make sure that the adoption of health IT systems occurs as part of a larger process of redesign.

Hack: National RFPs and Katrina are going to drive some significant movement toward standards harmonization. We are also responding to the RFP on privacy and security to discuss the laws and regulations that would support data transfer, as entities start to develop and certify standards. That is an important factor in driving our industry to respond a little bit more quickly than it has in the past.

With regard to hospitals purchasing IT systems that don't last, that has often occurred because the traditional purchasing approach has been to build an IT solution that works only for a specific facility or department. Now that we are starting to see capital planning and strategic models that identify IT as a backbone solution for the whole organization, we are finding that some smarter decisions are being made about technology and its use in some of the more advanced groups, especially those working with vendors like Siemens, GE, and Cisco Systems Inc. (San Jose, CA).

There is some movement. The fear of the past is there, but we are starting to see a clear understanding of the ripple effect: how introducing either an electronic health record or an RFID product into a system might affect the entire system, as well as the cost implications and return on investment.

Gee: There is a strong concern, particularly around the integration points that are required to put together a complete solution to address a broader work flow. Cardiovascular information systems (CVIS) are a good example right now. It is a relatively immature product category with vendors coming from a number of different perspectives, such as those handling a registry database, the imaging modality, or data analysis. They all claim to offer a CVIS, but they all have a different spin on it and no one has a product that meets a cardiology service line's full needs. So hospitals are forced to consider the trade-offs between a single-vendor solution and a best-of-breed solution. And in many cases even the single-vendor solution has been put together through acquisitions and has as many integration issues as putting together your own best-of-breed solution.

Does that picture represent an opportunity for medical device manufacturers?

McCausland: Imaging companies have long been working on standards for interoperability and transmitting information so that hospitals can connect various manufacturers. Systems by Siemens, Toshiba (Tustin, CA), and Philips (Bothell, WA) can all be integrated into a PACS, for example, through the digital imaging and communications in medicine (DICOM) standard developed by the National Electrical Manufacturers Association (NEMA; Rosslyn, VA). It is not perfect from a standards standpoint, but it is a far cry from what many other device sectors have implemented. DICOM has been a long-time part of Siemens's standards-building for the internal communication of images.

There is a lot of progress being made. Having sat in on a lot of discussions going into the year 2000, I think all the Y2K issues woke up the healthcare system to the fact that IT folks have to be involved in building the entire network. They have to be responsible for the entire communication network within a healthcare institution. There were so many individualized, departmental, isolated islands of solutions that were out there. And when healthcare facilities started to take a look at all of the pieces to be sure that they were in compliance with Y2K, I really think it shocked most of our customers. The good news was that Y2K was a nonevent because many companies and organizations had at least individually built fixes into their systems or had prepared for it. So it was not the calamity that everybody thought it would be. But it exposed the issue, and you wouldn't find a health system today where the information group and the CIO are not involved in the decision process for adding major capital equipment that has a substantial IT component.

All of the manufacturers of monitors, imaging equipment, and diagnostic or therapy equipment are well aware of what the issues are, and we are rapidly trying to change the architecture under which we develop our systems. Newer systems are not workstation oriented. They use traditional laptops, and the intelligence is put into servers and distributed like the modern IT architecture. We use open systems and try not to be tied into any platform like an old-style UNIX or a Microsoft operating system. We're all trying to go to modern, open-system languages so that we can get around a lot of the hardware-specific or proprietary systems problems that existed previously.

It is evolving, and for us it is an easy problem because we make equipment that costs anywhere from tens of thousands to millions of dollars. Siemens can afford to put some real intelligence into how it builds a system. But some companies' devices are pretty small in value, and sometimes the software content can be more expensive than the hardware. Those folks are still on the learning curve of where to go, how to interface, and what kind of standards they should use. They are still defining what information they need for their products to be able to connect to healthcare records.

Affordable IT?

The concern is commonly voiced that devices with high-technology components increase the overall cost of healthcare, and certainly increase the expenses that hospital systems and practitioners have to outlay to have those products. How do solutions that are in the pipeline or that already exist look in terms of affordability?

Srini: Right now, the pricing of healthcare IT systems is so high—and the risk is so great—that neither employers nor third-party payers are willing to invest in them. If demand were to rise to the point that IT system suppliers could recoup their costs and make reasonable profit margins by sheer sales volume, system sales would likely increase.

Large independent delivery networks (IDNs) such as the University of Pittsburgh system, which includes 20 hospitals and 5000 physicians, can have a significant influence by making sure that all of their product manufacturers come to the table. At Pittsburgh, we have one IT system of every kind—GE, Siemens, Cerner (Kansas City, MO), Epic (Madison, WI), and the entire gamut—and we are not going to rip those apart. Instead, we are going to make sure that they all interoperate with one another. So we have taken the approach, in making deals with our IT vendors, that we are insistent about our intent to integrate all of our systems and make them interoperable while maintaining a broad base of suppliers. So new companies simply have to give thought to the issues involved in becoming interconnected.

But in the long term, merely ensuring that local systems can be interconnected is not a viable approach. An individual organization tends to take on the connectivity constraints of a particular vendor, and this can limit its ability to connect to other systems. But once you rise to the level of regional health information organizations and begin to look at the types of interconnectivity that each requires, it becomes immediately apparent that it will be essential to be able to connect multiple systems from multiple vendors.

Pricing is certainly going to be a factor. But if the restrictions of the Stark and antikickback statutes are modified, a lot of the larger provider organizations will be able to provide IT services on a pay-as-you-go basis to smaller provider organizations. In this way, large organizations could achieve economies of scale and provide services at cost levels that small physician groups could afford.

Leavitt: I am not hearing concerns about the affordability of the devices. What has happened with high-tech devices like imaging systems is that manufacturers have encapsulated the complexity. This is part of the reason they do not interface as easily. Just to make sure that they would work, manufacturers incorporated all of the software pieces within the product, treating it as an encapsulated device. That was good in terms of simplicity of installation and use, but it can create challenges for adapting the device to some kind of common data transfer protocol within the hospital.

We are not hearing concern over the cost of devices, but we definitely hear screaming about the cost of IT. That brings me back to the payment system. The way we reimburse is the underlying reason for this distinction. The reason people have been buying systems for isolated departments is that isolated departments generate revenue by doing procedures. And so long as we continue to have reimbursement that pays based on volume of procedures, we are going to buy systems based on that money. When we start to see reimbursement linked to lowered risk of errors, or to improved outcomes and quality, then the hospital, clinic, or integrated delivery system will have to consider where it can put its dollars to reduce errors. That is the way that we will begin sending resources into integration and standards efforts.

Piechowski: It costs top dollar to rely on faster, shinier, and newer as points of market differentiation, and so long as organizations follow that path their money will be invested just to achieve those goals. To compete for funding, IT systems must start to show payback in patient safety and other areas that are more difficult to measure. But how do you put a value on how many people you did not injure, or on events that did not happen? As soon as IT systems start to address these other issues and can start to demonstrate their benefits that way, they will become more accessible to a greater range of people.

In regard to underinvestment, that goes hand-in-hand with how we reimburse and where our focus is. This long-term underinvestment will come home to roost. What we hear from a lot of people is they cannot afford to put in the type of systems that are available today, knowing that on the day they turn them on they become legacy systems. The current system of reimbursement and the margin on which hospitals are operating do not allow them to view IT as a normal cost of doing business, like most other industries do. In the financial services industry, about 15% of gross income is reinvested in IT. In hospitals it is much, much lower than that. It is under 5%.

Once we get over the cost hurdle and begin to see systems being implemented, we expect to see an evolution as new systems come on and people buy newer products with more capabilities. But you will not see everybody being able to do it all at once. Something has to break the bottleneck.

It is not all about the IT. IT is a tool, and we have to look at how to move the pieces of the puzzle around in such a way that we do not bankrupt the hospitals as we try to make these systems more accessible to them. Then they can start aiming for improved patient safety, and we can start reimbursing for that and changing the whole way we do this operation.

Reimbursement Incentives

Where has the ball been dropped on reimbursement and what now needs to be done?

McCausland: I am an advocate of market economy. When you talk about the biggest influencers of what gets done in healthcare, from a payment standpoint, it is the Centers for Medicare and Medicaid Services (CMS; Baltimore) and the insurance companies. They can single-handedly create the rules that can change the system. Some 60% of healthcare payments are made through CMS, but the insurance companies follow what CMS does 90% of the time. So CMS is the big influencer in what we do.

CMS is aware of all of the issues that healthcare is going through, and it knows that there is no amount of government money that can be allocated to solve this problem. It has to be accomplished through incentives that CMS creates, and that is why officials are beginning to talk about measuring performance and linking differentiations of payment to those measurements. It is an effort for CMS to take the first step in creating quality awareness and payment differentiation. It's an effort to say, 'Yes, we can make a difference, and the only way that you can conform to these performance measures is through the use of good processes.'

IT is just a tool to facilitate the process to achieve better quality and to make healthcare more efficient. Both Siemens and AdvaMed (Washington, DC) would like to see an open system that rewards healthcare systems and physicians for quality and good practices.

How are healthcare stakeholders responding to some of the pay-for-performance proposals that are coming forward now?

Lansky: Ultimately, improved performance and outcomes are the desired goals of the healthcare IT transformation. So it is important that any new financing incentives reward the provider for benefits that ultimately accrue to patients. We want to see health outcomes improved by virtue of using these technologies; we don't want to buy technology for technology's sake.

There is a balance to be struck between encouraging the initial adoption of technology and overcoming the barriers of capital and adoption costs. But we must also keep our eye on the prize of improving health outcomes and making sure that financial rewards are used to stimulate such improvements.

Hack: At the Health Technology Center (San Francisco), we are working closely with CMS on two fronts. One is to help the agency identify early on the high-value technologies, such as medical devices, that will be influencing the way that the healthcare sector addresses business problems and clinical-care issues. In this way, in the two-to-five- and five-to-10-year time frames prior to the release of those devices, CMS can prepare to make quicker decisions about reimbursement for the medical device components used in treating clinical conditions.

Secondly, we see pay for performance as the carrot for installing some of these IT systems that our organizations need. In five years, this is really going to be the way that business is conducted. Personal health records are coming, and patients will expect their electronic medical records to be available. That is added incentive for an organization, and that is the way that reimbursement should occur. The details of how we get there are going to take some work.

Gee: For both medical device manufacturers and for hospitals, it comes down to making wise investments in existing technology. There are many opportunities for medical device companies to add value and differentiate their products by adding connectivity today without having to worry about reimbursement and all of these other longer-term structural issues. There are a lot of opportunities for hospitals to accrue benefits by investing in technologies that give them operating advantages and increase patient safety and satisfaction—even if they don't reduce full-time equivalents or other hard-dollar costs.

What is the role of organizations like NAHIT and HIMSS in forwarding policies like this and helping CMS to devise and implement them?

Piechowski: At NAHIT, we hear frequently from our members, and we do have a policy representative in Washington, DC, who acts as a resource to the government. That is how we are positioning NAHIT: as a resource rather than a lobbyist for any one particular piece of legislation. We bring our members' concerns and views to those discussions.

Leavitt: HIMSS has been very active in this area. The associations are really the only way you can take a large competitive industry and develop a collective voice that people agree is representative. So HIMSS has been involved with standards development. It is actively participating in the federal RFP for a standards-harmonization contractor. And NAHIT is actively involved in defining interoperability.

HIMSS and NAHIT, along with the American Health Information Management Association (Chicago) basically founded CCHIT and got it launched, and the commission has many work groups, steering committees, and task forces on standards and interoperability. We created an organization specifically for EHR vendors called the EHR Vendor Association. So the associations are responding to this movement in a very healthy way.

But when it comes to payment for healthcare, it has been said, and I agree, that 'As CMS goes, so goes the country.' The balance between public-sector regulatory or legislative approaches and private-sector industry-based approaches is still very delicate. Getting that balance right is going to be key. It is going to be a very interesting story to watch and to be involved in.

Coordinating Efforts

Very few medical device manufacturers are active members of the organizations that are involved in developing policies and standards for healthcare IT systems. Is there an opportunity for device manufacturers to have greater influence in this area? What would you see as their role?

Merritt: Siemens is associated with our center, and it is certainly a leader when it comes to both standards and adoption issues. Siemens and other companies like it can certainly play a huge role in both the device and electronic health records areas. Some such companies already do a lot of work in both areas, and at some point all of their systems are going to have to interconnect. Interoperability should include not only getting information from one repository to another, but getting information from one device to another, or from a device to a record of some sort.

With the combination of the smart folks that we have in this industry and federal leadership, I am confident that we will achieve all of this at some point.

On a more general note, some folks want to prioritize our work to overcome the two main obstacles to healthcare IT, the lack of standards and the lack of adoption. They suggest that we should work out the standards first and talk about adoption later. But we feel that it is a better approach to tackle these questions simultaneously. If we have to wait for standards to be developed, we are going to be waiting a very long time for this technology to actually be filtered out into the healthcare system in a meaningful way.

Realistically, counting from the discussions of HHS secretary Leavitt's health IT commission to the introduction of an actual electronic health record onto the market, I think you are talking a minimum of five years. That is quite a long time. If we simply wait for standards to be developed before we decide to purchase a system, we are going to miss a huge opportunity and lose a lot of time.

We have the technological know-how, the entrepreneurial spirit, and the expertise to allow us to purchase the system today and upgrade it to adhere to any technological standards that are developed tomorrow. We should approach this the way that the software market does: a version is released, bugs are identified, and patches are created to fix them—and the next version incorporates those changes. Where there is incentive in the market, somebody is going to fill the need. But I don't think the market will allow all of the systems that have been developed to this point to simply be rendered obsolete and replaced entirely.

Not everybody in the manufacturing community is aware of what is going on in the other associations of interest in this field. Is there a need for better coordination among the various sectors that are involved in healthcare IT? How would that help move things along?

Leavitt: The industry is going through a growth stage right now from adolescence to early maturity. It is in the stage that everyone says is highly fragmented. There are hundreds of manufacturers, and the purchasers are confused about not just which product to buy, but what defines a product category. 'When I buy an electronic medical record, what should I be getting?' There has been this youthful exuberant stage of the industry, and we are maturing now to a stage where we all agree on what an electronic record is and what an imaging system is and how they connect. We agree on what you can expect when you buy these. That is a very healthy thing, and the associations are doing a good job of helping the industry move to that next stage.

Piechowski: What you have yet to see is the private sector really organizing as a market. Once private-sector IT suppliers realize that they are all in the same boat, they will begin to have a much stronger voice on these issues. CMS is certainly a major market mover, but there is also a large market among all the sectors not controlled by CMS. Eventually, those potential consumers of EHR systems may begin to see themselves as a common market with common interests. That will change the landscape a lot and help move things along.

Lansky: This is a multilayered environment, and the top layer is national policy. We all hope that Secretary Leavitt's creation of the American Health Information Community (AHIC) will provide a platform for coordination of some of the broadest issues—particularly the issue of standards, together with the privacy and security policies that relate to them. There is draft legislation to require that federal agencies conform to AHIC recommendations, and this will establish a very powerful source of coordination at the national level. Also, we imagine that a lot of private-sector actors will look to federal commitments as a guide for how they want to proceed in their own environments, so there is a hope that AHIC will become a major driver.

Trade associations and professional societies provide another mechanism for coordination. Our own network, Connecting for Health, now has about 80 associations, and associates regularly attending meetings.

An emergency version of much of what we're talking about is the KatrinaHealth Web site (www.katrinahealth.org), which was set up to help evacuees from regions affected by Hurricane Katrina get access to as much of their medical information as possible. In the larger scope of things, this site is taking on just one small challenge, but it represents in microcosm many of the issues we have talked about here, including industry cooperation, the coordination of federal action, and the struggle to deal with issues of incentives, privacy, and standardization. As we review what have we learned from the tragedy of this hurricane, I think there are lessons we can apply to the next steps of building an interoperable health information environment.

Reengineering the Infrastructure

What is the relationship between the IT supplier community and the medical device manufacturing community on the issue of coordinating the development of EHRs? Do they know what one another are doing?

McCausland: Yes and no. If you had asked that question three years ago, the answer would have been no. But the field is really moving along rapidly, and anybody who makes a device today knows that somebody wants to take the information that the device is generating and do something with it electronically. They want to put it into some format that goes into a record, whether it's a short-term record for a specific visit or part of a long-term permanent patient record.

Software manufacturers, along with the healthcare institutions, are defining what that record really is. Unfortunately, right now that record is whatever an individual hospital decides it is going to be. We all know the basic ingredients of what one wants in a health record, but there is not currently a standard set of information for every physician and hospital.

Getting all of the physician offices to define the standard transaction set for health records in their particular practice is pretty tough. We all know that there is a certain defined set of information that we work with. All of the device manufacturers today are building their products with the expectation that they have to generate that information, send that information somewhere, and then let the IT system worry about whether it stores it, displays it for a period of time, or discards it. This is still not a very clear and concise process, and it varies a great deal.

What is the best way to get the healthcare sector's incentives aligned with the goals for an EHR system that everyone seems to have in mind?

Merritt: This is most definitely a public policy question, simply because the Medicare and Medicaid programs drive so many of the reimbursement fee schedules, even in the private sector. Those programs frequently set the standard of what doctors are going to be reimbursed for. But at the same time, this is also a political question, and that adds a whole new complexity to the situation.

Doctors do need a financial incentive to adopt health IT. Most doctors are interested in moving from a paper-based system to an electronic system. But they're not doing so, because it either would cost too much money or would upset their work flow too much. So doctors automatically present barriers that need to be overcome, and I think financing is the first and foremost of these. At this point, I think, the best approach is to find a way to properly incentivize doctors through Medicare and Medicaid, simply because the private-sector health plans will follow suit. The creation of pay-for-performance programs that would rely heavily on health IT in order to report quality information is certainly an approach to look at.

A related policy question that has received a lot of attention involves the Stark and antikickback statutes, whose restrictions on financing, in this context, constitute a barrier to the adoption of health IT. Those statutes were originally drafted to root-out corruption between a physician and anyone he might refer his patients to—be that a hospital, a long-term care facility, a pharmaceutical manufacturer, or what-not. The intent is to eliminate any financial incentive for a doctor to steer patients to facilities in which he has a financial interest.

Today, small clinical practices are simply not adopting any technology, and many times it is because they don't have the financial wherewithal to do so. Nearby large hospitals often have fairly robust health IT systems, and they might be able to provide assistance to smaller clinics. But the Stark and antikickback statutes essentially prohibit hospitals from reaching out to their doctors and equipping them with health IT.

Both hospitals and providers are fully behind reforming the Stark and antikickback statutes to address this issue. One hospital system on the West Coast told me that if these reforms were made—and if they are clear-cut and workable—within 12 months it would equip 6000 physician offices that do not currently have any health IT systems.

So resolving this issue on Capitol Hill could have an immense impact on financing the investment in health IT systems and expediting their adoption.

Those who discuss electronic patient records sometimes seem to have a scant idea of how patient data might actually enter such a repository. Given the cost constraints on the healthcare sector, is it practical to build a system that would require a massive workforce of data-entry personnel?

Merritt: If the use of an electronic health record—or any sort of health IT—is simply to digitize a paper record, then we are missing the boat. That is where the work flow question comes in. If, instead of having a doctor write something down in a file, you simply have someone else typing it into a system, then you are not really gaining any efficiencies. However, the ability to transfer that information from its original input into a hospital record, or from one health plan to another, does offer productivity advances, because the information can be entered just once and can follow the patient wherever he goes.

A sophisticated system should automate a lot of the processes that are now manual, things like having someone available just to fax something to another office. The pharmacy community can certainly attest to the difficulty of trying to reach a physician's office via phone calls and faxes just to check a simple thing like a prescription refill. Once we have a sophisticated system in place, you will see a level of automation that will probably blow most people's minds.

Are medical device companies or their systems headed in that direction?

Srini: Absolutely. Asking someone to key in information is not only resource intensive but also a source of errors. New products such as smart infusion pumps and an increasing number of wirelessly enabled devices can coordinate their own information with data stored in a healthcare information system. For example, a smart infusion pump can be triggered according to rules and decision points based on the patient's weight. This is definitely the way that product development is going.

The practice of setting aside critical patient data to be input into the patient record at a later time can create latency issues that are highly detrimental. If a patient receives a prescription during an outpatient visit, and later winds up in acute care where the physician doesn't know about that earlier prescription, the potential for harmful drug-drug interactions is very real.

The best resolution for this problem is to implement seamless, time-dependent automated entry of medication data at the point of care. This can be accomplished in hospital settings by using a computerized physician order entry system, or on the ambulatory side by using an e-prescription system.

A sophisticated healthcare IT system will enable us to create and analyze information, but it should also make it possible to better and more quickly assimilate existing information. It has taken about 20 years of clinical trials for a simple procedure like a lumpectomy to become adopted as a standard form of treatment. But if we are ever to implement a system of evidence-based medicine, that kind of lag time is way too high. A sophisticated healthcare IT system will make it possible to convey the guidance of evidence-based medicine to the clinician at the point of care, so that the latest advances can be rapidly incorporated into his practice pattern without the extra legwork involved in searching for the results of various clinical trials.

David, are these issues something that the Markle Foundation is wrestling with?

Lansky: We recognize the great difficulty of changing the practice environment of 300,000 small practice physicians as a precondition to having portable electronic health records. Most of today's discussion is based on the idea that we must first convert every physician to an electronic record, and then we can talk about giving patients more access and control. But if you invert the paradigm, starting instead with the capture of patient information that physicians and others use to monitor and improve care, we are actually farther along than it might otherwise seem. Over the past 20 years, a lot of the information that you might think of as constituting the health record is already in digital form. It has already been captured, processed, and distributed across a variety of networks in laboratory computers, imaging computers, and diagnostic systems, in healthcare systems and institutions. All of these systems are now using digital data streams, though they are almost entirely unintegrated.

The fact that there is a tremendous flow of information coming across the wire now creates some significant opportunities, especially for companies in the device industry that are developing tools of various kinds. But right now we are not adequately integrating and applying the information we already have. We shouldn't have to think of the adoption of healthcare IT as requiring an army of clerks to rerecord data that is already available in digital form. Instead, it is really a reengineering of the information infrastructure.

Broadening Involvement

What is most important for each of the sectors to be doing to drive the future of IT? Is there a need for a broader meeting that would involve a wider range of these sectors?

Hack: There is a need for broader communication than what has taken place in the past, and we are starting to see that occur through the development of state initiatives and through the RHIOs. That seems to be a way to pull together all of the disparate parties to sit down and try to tackle some of the healthcare industry's big business problems, from the EHR to basic connectivity and what needs to be housed within an individual physician's office.

Our partners want organizations to facilitate a dialogue between the IT suppliers, the device manufacturers, and the end-users. They want products codeveloped so that they can integrate with one another. And they want these groups to understand what the end-users' needs are and not have to go back and repeat the product development process until they finally get it right. So we are hopeful that new product development will increasingly make use of codevelopment methods. And this would, of course, influence the way that the supplier community is working.

Gee: You could use point-of-care work flow automation as a model of what is going on in the market. The caregiver and the patient are in the middle, and they are surrounded by all of these different devices, ranging from infusion pumps, patient monitors, and nurse-call systems to order-entry systems, results-reporting systems, and the EHR. The customer wants one solution, and obviously there is no single vendor that supplies all of those products.

This is not a situation that is going to be solved by top-down standards being imposed on the industry, but it does create new business opportunities for innovative companies like Vocera Communications (Cupertino, CA) or Emergent Information Technologies (Newport Beach, CA). And it will require a lot of business development and alliances among vendors that are not necessarily direct competitors, but that have never worked together before.

Merritt: For healthcare professionals and hospitals, the most important thing is to get their hands on this technology. Study after study has concluded that using health IT will not only improve quality of care but will also save money. Nevertheless, last spring a Centers for Disease Control and Prevention study concluded that information technology is nearly nonexistent in ambulatory offices and even in many hospital systems. According to the study, only 30% of hospitals have health IT systems for patient care. In ambulatory settings, where physicians practice most of their medicine, the adoption rate was more like 15%. And in smaller practices the rate is even lower—down into the single digits. So the most important thing for the provider sector is to actually make the investment and get the technology.

Manufacturers like Siemens and EHR vendors such as IDX (Burlington, VT) and Allscripts (Chicago) should continue to roll out products that anticipate the standards now under development and are able to be upgraded in the future to meet the needs of an interoperable system.

Finally, consumers absolutely have to get engaged in this debate. The word interoperability is probably foreign to 99% of the population, and so is the concept of being actively engaged in one's own healthcare. Largely because someone else is paying for it, the vast majority of Americans believe that the extent of their responsibility for their health and healthcare is to visit the doctor and get a prescription.

I couldn't disagree more with that attitude. Consumers should become more engaged on both the financial and technological sides. A great place for consumers to start is by identifying an online site where they can enter their own information into a personal health record, and can begin managing their own health and healthcare.

Lansky: I agree that providers have to get their feet wet in IT. However, I would emphasize that they need to be selecting products and services that are interoperable and conform to standards. And they should work with vendors that are actively participating in the standards process, so they have a sense that those suppliers are committed to being part of a larger information stream. There are still a number of products and services out there that are very proprietary and aren't really oriented toward interoperability.

In terms of the consumer sector, I agree that pursuing a personal health record is important. But patients should also ask their doctor if he is keeping their information in an electronic health record and, if so, request online access to that record. Consumers should also ask their doctor to communicate via e-mail. This should be simple, since e-mail has permeated every other aspect of our lives; but unfortunately it is still almost foreign to the patient– provider relationship. Patients could become drivers of technology adoption simply by saying that they expect to communicate with their doctor via e-mail. This will put some demand-side pressure on doctors to participate in this change of technology.

What forces are currently guiding the development of healthcare IT systems? Is the field being driven by market forces, directed from the top down by the development of standards, or something else?

Srini: In the device arena there are two forces that are driving product development. The first is the aging Baby Boomer population. The second is the diminishing number of available doctors and nurses.

Together, these forces have made it very expensive to carry out patient care in traditional healthcare facilities, and the entire healthcare system is therefore struggling to move care out of those settings and into patients' homes or other alternative-care settings. Health plans have initiated wellness campaigns to help their members maintain their health, and they have adopted a wide range of home-use medical devices that provide remote monitoring functions. In effect, market economics have forced the nation's health plans to become very specific in identifying a patient's medical needs and determining the best and least-expensive setting for treatment.

But while this type of remote healthcare monitoring is becoming pervasive, there is still a disconnect in the way that the resulting information is collected and added to the rest of a patient's record, which may originate in a doctor's office or hospital. There is a need for new products that will combine those two streams of information at once.

In the EHR arena, the force that is driving product development is the sheer amount of patient information that is now being provided to the physician. Already, doctors are becoming overloaded. To avoid breakdowns, payers will have to put in place health coaches—people who are trained to triage patient information, handling some mundane items themselves and sending only critical items to the physician. Products that facilitate this sort of integration between the functions of the payer and those of healthcare professionals will meet with a ready market.

Are most device companies hearing this kind of varied input from their customers? Is there a need for some more orderly way for that input to get to the community at large?

McCausland: The only way that we can get to a reasonable exchange of information is by defining some broadly accepted standards. By standards, I mean the type of data people want and the communication protocols that everybody must conform to. Those agreements are essential if we are to be able to send data to some sort of repository that collects them and puts them into a record format.

We have to have formats. People have to know what it is we are trying to transmit so that they can populate formats that are transferable from institution to institution. We do need some fair consensus on what information people need to have and how they can get that information. But I think all that is happening.

Device companies are very aware, and the hospitals are well aware, of what is necessary. I do not think the physicians are aware yet. It is 550,000 entrepreneurs out there, anywhere from a one-man office to a hundred-person office. That is a very difficult set of cats to herd, and there's no point in trying to get them to vote on issues because you would get 550,000 different answers.

Somebody has to decide what goes into a patient record, what we are going to collect, and what we are going to be able to transmit back and forth across the country. Somebody has to decide what we are going to be able to transmit to and from hospitals, nursing care facilities, and nontraditional places. But a lot of progress is being made. We just need to continue to help this evolve so that standards groups like NEMA and HIMSS weigh in, and we have to help standards get developed in the other places they are needed.

Are organizations moving in the right direction to broaden the involvement of others outside of their traditional member bases?

Piechowski: You are seeing that in a lot of organizations. I know NAHIT was set up originally to reflect that. Our members are IT vendors, manufacturers, payers, and providers. And we are seeing a growing component of people that you would not normally consider to be healthcare organizations. For example, General Motors (GM) just joined NAHIT, because healthcare is a huge expense for GM. It is sitting at the table with other stakeholder groups trying to help something move forward.

The key here is trying to break this cycle. It sounds like a lot of the information that we need comes from the very people who do not have the time to participate in this process. Physicians are currently paid by the procedure. For every day they spend sitting in meetings to help IT suppliers decide how systems need to operate, they lose money. And we are squeezing them out of the equation. Maybe there is some way that they can be compensated to participate in this process. We hear this all the time from our provider members: they would love to do something, they just do not have the time to do it.

Leavitt: The last frontier for IT is the physician office, and part of the reason is the fact that because physicians are paid on productivity, they lose money sitting in meetings, whereas almost everyone else in the healthcare industry is paid when they sit in the meetings. But we have seen a change. A lot of physicians are realizing that if they do not take some leadership in this, the systems could be used to dictate orders to them rather than their writing orders into the system.

We can reach out to them. For example, HIMSS has launched a series of conferences for physicians on buying and implementing an EHR. We choose to do them on Saturdays because that way the physician is not losing a normal patient care day. The physician is sacrificing some family time, but not patient care time and revenues. Those conferences have been very well attended, and every time we have asked, 'Is Saturday the right day?' they have said, 'Yes, this is the right day to do this.' So we have to meet these audiences and new constituencies halfway.

Physician professional associations have also stepped up to the plate. A few years ago, they focused mainly on not having an unfunded mandate for electronic records dropped on them. Now they are absolutely at the table and very actively engaged in the development of standards and certifications. I see positive signs everywhere.

Take-Home Messages

What is your best advice for medtech company executives on doing business with the IT supplier community, making the right products for the emerging IT-rich environment, and working with clinical partners?

Hack: My best advice, and the advice that our partners are looking for, is for manufacturers to work early on with the delivery systems and end-users to help them solve the business problems. That goes far beyond producing a device. They have to look at work flow, culture, and change-management issues that result from the implementation of a device. That is a huge role for them to step into, but that ultimately will lead to the success of the manufacturer. And it will certainly move our industry forward in bettering connectivity.

Gee: Medical technology executives need to recognize that there is a shift in the market, a value migration, where requirements from the marketplace are changing and evolving. They really need to open the gates and look in a much broader way at the customers' needs and the best way to address those needs.

Piechowski: Think big, be bold, and participate as part of a group of leaders, not as leaders of individual companies. We need a group of leaders to put this together and to see the healthcare industry through this transformation.

Leavitt: Medical technology manufacturers that do not have an industry-affairs or government-affairs leader need to get one now. And if they have one, instead of having them focus on regulatory issues and reimbursement for new technologies, start sending them to the healthcare IT meetings hosted by associations or by the government. They need to find out what is going on at the HHS Office of the National Coordinator for Health Information Technology. That individual or team of individuals is going to become increasingly important to companies, so I would pay a lot of attention to what is happening there.

McCausland: Siemens is spending a lot of money to educate its people about the business of healthcare. What is the business of being a family practitioner? What is the business of being a cardiologist? What is the business of running a hospital? What processes exist throughout an entire organization? This knowledge will help us understand all of the issues that healthcare professionals face.

We say that we want to be a solutions company. But you cannot be a solutions company unless you know the problems of your customers. So you have to know as much about healthcare processes as the physicians, nurses, and other caregivers who use them. That is where we are putting a lot of emphasis: on education to understand the work flow. How the patient flows or how the paperwork flows through the system generally determines what the work flow is. And finally, we must consider how to adapt our products and systems to support that work flow from a quality and a cost standpoint.

Srini: What is most important for patients is not just the recording of discrete pieces of information but how the longitudinal health of the patient can be measured and recorded. Whether the technology used is a medical device or an independent electronic health record, and whether the information comes from the ambulatory sector or the acute-care sector should not be issues. The only important issue is how to maintain the wellness of the patient.

To get the full picture, device manufacturers should evaluate how their products contribute to the overall wellness of a patient over time. The data recorded by a device should have value to patients in terms of their understanding of maintaining wellness, preventing illness, and guiding treatment. And if a patient succumbs to a critical problem, devices should be designed to enable the patient to have a hand in recovering or improving their own health in addition to alerting the patient and physician right away.

Lansky: It is important for everyone in this field to recognize that now, in a way that was not true five years ago, everyone is a node in the network. Everyone is contributing to a larger fabric of information exchange that ultimately the patient will control but is also indispensable to many providers and professionals across the network. So instead of thinking about proprietary approaches—whether on the financing side or the application side—everyone needs to think of their work as part of a network environment.

Many of us have not given enough thought to the public's confidence in this network. We must be sensitive to the issues of privacy and security, protecting confidentiality, ensuring that patients control where their information goes, and protecting against secondary uses of data that weren't anticipated by the patient. We will have unpredictable problems if the public loses confidence that we are handling their very personal information with great care and trustworthiness.

These ideas suggest two strategies that will become very important as we move forward. First, we are all part of the network and we must commit ourselves to cooperation in the service of the patient. And second, every one of us is a vulnerable node on that network, and therefore we all have a responsibility to protect the patient's privacy and ensure public confidence.

Merritt: I agree that it is a key priority to maintain public confidence that healthcare IT systems will secure information and protect patient confidentiality and privacy.

Congress has been kicking around several bills to expedite adoption as well as the debate over standards. I would urge everyone—whether they are an employer, a vendor, a practitioner, or a member of the general public—to contact their member of Congress and exert some pressure for the passage of a good health IT bill that will boost both adoption and the standards debate.

Folks should also take a look at some of the initiatives that are popping up across the country. According to some estimates, there are currently between 600 and 800 different local and regional efforts to connect their community. Some are citywide, some are county based, and others cover entire states. Most are being driven by hospitals, doctors, and technology companies. But health plans, state and local governments, and the public also need to get involved in these initiatives.

Device companies should also consider how they can play a role in getting these initiatives off the ground, whether through financing, technical support, expertise, or simply participating in the process of developing these networks. These could truly be the beginning of something that eventually blossoms into the fully interconnected and interoperable national health information network that we are all envisioning.

Copyright ©2005 MX