NDC: IT has impacted the delivery of health care for many years. There are examples going back to the 1950s and 1960s where people began to use computers to record data that historically had been only on written paper charts. What’s been particularly interesting recently is that there has been enough experience to get more and more people excited and convinced that there needs to be more investment in health care IT, because they now see the direct link to improvements in patient care as distinct from where it’s been used more historically around administrative services in the health care field. There are a number of areas that I think are becoming very well accepted. The first, given the focus on patient safety particularly since the IOM report in 1999, is the ability to make the process of medication administration be more error-proof than it has been historically. The second is computerized physician order entry (CPOE), which allows less likelihood of making inadvertent errors and allows a much more efficient system for submitting orders and providing those services to the patient. As more and more hospitals implement electronic health records or other types of systems that digitize at least some part of the care process, you have more machines producing electronic data, such as imaging machines, EKGs or other types of diagnostic devices. These data streams are now providing us a far more robust and extensive body of data that can be analyzed for opportunities to improve the care process, really as a byproduct of installing these systems in the hospital.

NDC: I think medication administration is an area that everyone believes we can do better in, and where information technology has shown to provide value in making sure that the right therapeutic, at the right dosage, is provided to the right patient at the right time – whether it has to do with bar-coding drugs or managing the data on a hourly or daily basis to ensure that the process is performing as well as it should. CPOE has been proven to reduce the prevalence of inadvertent adverse drug events and other challenges in prescribing the right medications, as well as providing more decision support to physicians and other caregivers. All of these areas have high returns to both patients and caregivers, and there is now quite a body of literature that shows the value of instituting systems such as that. Another area has to do with analyzing meta data in order to provide insights into care processes that otherwise aren’t available. I think the work that’s now happening at institutions like the Mayo Clinic, University of Pittsburgh Medical Center and others to take their databases and look for trends from the extensive patient histories that they have, combined with information in other databases, allows physicians and other caregivers to provide ever more value in both diagnosis and treatment.

NDC: The fundamental challenge, generally, is culture – not only pertaining to health care, but to all organizations in general. In the case of new and more sophisticated uses of IT, there are only so many people who are experienced enough to apply it as well as everyone might like, so we have a diffusion challenge, and an education and training challenge, much like in the case of new surgical procedures. We’re going to increasingly have a challenge to understand how to best train caregivers, and ultimately patients, to take advantage of some of the more sophisticated approaches to diagnosis and treatment that health care IT will help us bring to bear. From a caregiver’s perspective, the challenge of using technology and process improvement to improve health status is a question of whether financial incentives encourage them to do that. The more we can move to a pay-for-performance type of arrangement where caregivers and institutions are compensated for investments in health care information technology that allow them to improve the care process, the more we’re going to be able to overcome the incentive challenge – and the general cultural challenge of doing things differently within these complex institutions.

NDC: I think it’s going to differ by their environment. If someone’s a rural physician and they’ve been treating the same people and families for many years, and they have another five or ten years left to practice, I don’t imagine they’re going to be put in a position where they have to use electronic health records. The more germane question is whether they will increasingly feel that the benefit of making that change outweighs the disinterest or dissatisfaction they perceive in making that change, not to mention the cost. American health care is a very large and complex business and we aren’t going to see massive changes in one or two years, so those physicians can probably continue to use paper charts for the next five or ten years, and retire. The question will be during that time, will they feel that they can provide better services to their patients, whether preventative or the treatment of chronic disease?

NDC: It’s happening today: there are plenty of systems in your average hospital where the data is taken from one system which is in a fundamentally different format than the system that needs that data. We’ve had that technology for many years – so-called interface engines – that translate data. What’s envisioned in the current interoperability discussion has to do with the degree of data that’s interoperable – clinical data, physician notes, things that are more difficult to figure out a way to translate on a consistent and systemic basis from one system to another with adequate and appropriate security and privacy to respect patients’ and caregivers’ rights. I think it’s a significant challenge, but how far off that would be, I believe, is actually within the hands of industry, government and other constituencies to help decide. The area that needs increasing amounts of work and attention – and is getting it – has to do with standards. We have an opportunity in the health care industry to proactively establish a series of open standards in areas where they don’t yet exist, in order to allow the information to be shared securely and with appropriate privacy among the different entities, including patients and their families who need it. However, there is a lot of legitimate discussion about security, privacy and appropriate use of the data more than, frankly, any particular technology issue, per se. So, both the opportunity and the challenge is to create the right forum which the federal government and other organizations have been trying to do in order to create consensus as rapidly as possible around these standards so you don’t have islands of automation, which is often the case today. I’m sure that in the next two to three years we will see advances in interoperability far beyond what we’ve seen in the last five to ten years. We will see that because there’s leadership on this issue from the federal government, from the physician community and from the hospital community, driven in many cases by the demands of patients, employers and other payors in the health care system. The question is, will it be as pervasive and valuable to the individual patient as it could be? That’s where we still have a lot of work to do.

NDC: We largely are driven by the services we provide to health care insurers and providers, which consists of consulting, systems implementation, infrastructure services, and what we call business performance and transformation services – improving processes in a hospital or health plan. Those are supported by our software and hardware offerings, to help our clients improve patient safety, caregiver satisfaction, or revenue cycle – things that are part of their business objectives, not only a more robust IT infrastructure and faster response time, which are classic information technology services. We have a methodology called the Safe and Lean health care provider, where we take simulation and modeling technologies that have been used in industry for many years – including Lean Six Sigma – as well as clinical information systems from partners of ours such as Seimens, Cerner and Epic, and work with hospitals and provider organizations to enable an information technology implementation that accomplishes improvement objectives, whether it’s administrative – such as patient throughput time – or, as is often the case, it improves patient safety and clinical quality outcomes.

NDC: There have been a lot of error reduction approaches in health care for many years, sometimes called continuous quality improvement (CQI), also called total quality management (TQM). Obviously JCAHO has had many improvement methodologies integrated into their accreditation process for many years. Six Sigma is a particularly comprehensive approach to error and variation reduction that’s been used to great effect in many industries, including health care. We’ve linked that methodology to a process that’s been used in industry, made most famous by Toyota, called Lean Manufacturing. What that does is look at the process that’s being evaluated – which steps actually add value to the process, and which are just holding patterns or aren’t having any value. It seeks to take wasted steps out of the process. It allows you to look at the process across a hospital for a certain condition and not only reduce unnecessary variation and errors, but also improve the efficiency of the process, thereby addressing the joint objectives in most health care institutions of improving patient quality while at least moderating or reducing costs.

NCD: Even some of the smallest physician offices have some sort of practice management system or other way of billing for their services. If we move beyond that, you have physicians even in smaller practices who believe that information technology can substantially improve the way they practice medicine. There’s been a challenge to invest aggressively in information technology around the care process in the small physician’s office. If the financial incentives were there, that would help overcome one of the main barriers in what is already a very difficult financial environment for many physicians. Most studies have estimated that it’s going to cost between $20,000 and $25,000 for a physician to install electronic health records in his or her office. If physicians implement a technology that keeps people from making unnecessary office visits, unnecessary ER visits or unnecessary hospitalization, that benefit fundamentally accrues to the person paying the bills: the health plan and/or employer. Some states are looking at ways that those benefits can potentially be shared with small physician offices so that they have an appropriate incentive to implement this technology.

NDC: We’re looking at solutions where we can actually work with them to improve different processes in the health care environment. We’re not going to be developing health care applications, such as electronic health records, like many companies are out there doing today. The relationship is basically combining their world class health care expertise with our process and IT expertise. One of the projects we’ve starting working with them on is a strategic incident management system, which addresses concerns about public health surveillance and bio-security and bio-surveillance. The other area that we’ve talked about is what they call the intelligent hospital. There are a number of technologies benefiting from the greater digitization of health care that can help improve the process of care – one of them is a process of tracking things in hospitals called radio frequency identification. They’re building a new Children’s Hospital in Pittsburgh and one of the things we’re looking at is using technologies like radio frequency identification to better understand where things and people are in the process of care, not only just to track – which has a benefit of understanding how to locate things quickly – but also over time to have a set of digitized data to look at increasing levels of process improvement. IBM has a computational biology center with two dozen people with Ph.D. or M.D. backgrounds who work with our clients in order to design information technology approaches that allow them to take data that historically hasn’t been utilized in the health care setting until recently, such as genomic information and the work that their physician scientists do in a research environment, and apply that more directly to diagnosis and treatment in the hospital setting. Those are two important areas that we’re working on with UPMC.

NDC: That was a very innovative and important development in our relationship with UPMC. They’ve managed themselves quite well and are fortunate to have the financial wherewithal to invest in areas that they think are important for the provision of health care. We agreed to put up $25 million starting out, and they’ve agreed to put up $25 million as well, into a joint investment fund to start working on some of these innovative solutions, and we’ll therefore have a joint investment and a joint interest in their success, both within UPMC and hopefully through commercializing some of these solutions beyond the UPMC environment. When we decide on what projects to work on with UPMC, it’s with a dual purpose. It has to be valuable to their capability to provide better and more efficient patient care, and to us in terms of advancing our knowledge of what could be valuable to other players in the health care industry.

NDC: What we continue to learn is that you can never be close enough to the ultimate beneficiary of the work you do, which in the case of health care is the patient and the individual. What are the needs that they have that our services and information technology can better assist them with? The people who quite naturally are closest to understanding those needs are clearly the caregivers, such as people at UPMC. Last year, we started an effort called a global innovation outlook, and we focused on three areas: government, health care, and work life balance. We created forums around the world, in Europe, Asia and the U.S. where we invited patient advocacy groups – organizations that you wouldn’t typically consider part of your market research, necessarily –to understand how and where health care was evolving and how we could be better partners in improving the evolution of health care to benefit patients. One of the insights that came out of that effort was that, as markets are evolving, it’s sometimes best to collaborate with what ultimately would be your competitors in order to provide a faster on-ramp to improving the marketplace. We’ve been very active collaborating with other parties in the IT industry to more rapidly create open standards around the interoperability challenge. We also discovered that there are solutions being implemented in developing economies that, because of their resource constraints, are potentially more efficient and effective than the way we’ve approached challenges in providing care for the underserved in this country. One organization, for example, was trying to figure out how you collect data in environments where there are no cell phone connections – there are no phones, never mind any computers. When the World Bank or others have been applying funds in order to help people in these rural communities, how do you actually collect data to know whether it’s working or not? They developed a system that uses land line phones situated in specific villages in order to provide data essentially in the way you bang in codes to get your voicemail system on your telephone.This kind of thing may not seem like rocket science, but it’s proven extraordinarily valuable in terms of how we keep track of whether interventions are working in communities that do not have a lot of connectivity. Some of these solutions we discovered need to be integrated into what we do in the U.S., for example to address problems we have with the homeless, who don’t have an address that you can send things to, and don’t have a fixed telephone number that can you can necessarily contact them at.

NDC: We’ve tried to be very careful to make sure that we link our services in our strategic partnerships to the core objectives of each institution, and therefore we have a unique approach and relationship with each institution. Mayo Clinic has done a tremendous amount of work, and us with them, around the use of their five million consented patient records. We’ve done work with the University of Utah Hunstman Cancer Institute looking for markers for certain oncology trials to better advance the work they do around cancer care. The Cleveland Clinic is quite prominent in their cardiac care and research, and we’ve done work with them on abdominal aortic aneurysms in order to look at gene markers in order to better identify patients at risk for aneurysms before they have one. We’ve tried to design our relationships so that they match the strategic objectives that the specific institution has, and in areas where we can add value in helping them advance those aims. To some degree academic medical centers have similar missions, but because we’ve been very careful to have the relationships be well-structured and specific to both our objectives and our partners’, I believe we’ve avoided having unnecessary overlap between the work we’re doing or the work institutions are doing with us. IBM has a $5 billion a year research budget, seven labs located in different countries and researchers who have been fortunate to win things like the Nobel prize, the National Academy of Sciences designation and national technology medals. We have a very deep and long standing investment in advanced and pure research, as well as applied research, so a lot of the reasons why we’re such a good partner for these academic medical centers is that we truly understand their research mission as well as their care mission.