If many of our health care facilities don't have EHRs set up today, how will they have time to collect on the EHR-use incentives?
Dare: For hospitals, it's a larger undertaking than for a medical practice. It's not a long window of time for a hospital or health system that hasn't even begun thinking about doing this sort of system.
Honestly, knowing some funding is available fairly soon will help them progress. We would certainly say to hospitals, health providers and physicians, you would want to start now, planning and moving ahead in part because the incentive payments are bigger in the early years. If you wait until 2013 or 2014, whether you're a hospital or a physician practice, the amount of money available to you is less.
At the back end, it's a carrot-and-stick system. For those hospitals and physicians that don't adopt the EHRs, their Medicare reimbursement is reduced over time.
Kennedy: The funds don't start coming online until 2011. I thought that was a reasonable compromise between the need to spend money today for economic reasons versus the need for organizations to get a project in place to begin deploying health information technology.
Secondly, I would define the problem more in terms of market segments. If you have the large integrated delivery systems or large groups, they already have much of the foundation in place. Even if they don't have full-fledged [digital medical records], many have LANs in place. Many have provider authentication processes in place. So it's not like you're starting from scratch for many of these organizations. While I agree it's a challenge, I'm not sure given the restraints you can expect much else.
So should EHR information be shared regionally, statewide or nationwide, and how do you do that?
Dare: Health care organizations of all types ... need to be able to share information in a variety of ways. Public health programs need to have a communitywide view. So the answer to national, regional or local, is yes. So a patient first goes to a physician and then maybe a hospital, and then maybe to a specialty hospital or long-term care setting, and then back home with home care. We need to connect that continuum of care.
Secondly, we need flexible approaches. Honestly, there's no one technology architecture. There's not one technology portfolio. There are big health systems, and small health systems, health care organizations that are payers and providers combined. There's public health, which this bill addresses and sees the need for public health organizations to use IT to transform what they do. I think the bill does a nice job of not being too dictatorial. There's money to support regional and subnational health information exchanges, but that money is available to the broad range of those exchanges out there. It doesn't say you have to use this architecture or this solution set. That's a positive thing.
Kennedy: You're going to have to make room for innovation. Health information exchanges are one approach that can make sense, but we think there will be additional approaches required to get the value out of these solutions. In Dayton, Ohio, we partnered with Kettering Medical Center, and they brought their clinical data sources -- the radiology, lab systems, etc. -- and we brought our administrative resources to the table and that information was integrated into a single patient record that the patient controls and that the doctor uses.
There are federated models, centralized models and hybrid models. This would be a hybrid data-sharing model. You keep the core information necessary to manage the patient on an ongoing basis centrally, and you leave things like the radiology image in the system where it's stored. Those approaches are very scalable and could be rolled out in a relatively rapid and significant way.
EHRs aren't standardized. So how do you share information if it's in different formats?
Dare: The bill does address the data standards piece. It asks the Office of the National Coordinator for Health Information Technology to establish a standards committee. And, of course, we already have the Healthcare Information Technology Standards Panel, which has been working for three or four years now to harmonize standards. It is a challenge for health care. There's continued progress to be made in that realm. We're not there yet.
Kennedy: Should we create a centralized database from which all patient information is shared, or is it more about standardizing the interconnectivity to the various databases? It's some of both. Standards help in allowing data in system B to get into system A, but let's not forget there have been interface engines around for years. So it makes the work easier to do, but it's not a panacea in and of itself.
The more important questions is, what is your data architecture? What we'd like to see is an approach where the patient's clinical status can be represented electronically, not just what the primary care doctor is doing or specialist is doing, but an integrated view of the patient. Once you have that single representation of that patient electronically, then you can begin to run rules off that single electronic representation. That's where we'll see real value. Instead of taking 17 years for clinical research to make it's way into practices, we'll be able to create rules where it may become 17 days. It's that single representation of the patient and applicability of clinical rules to the data in real time that offers us the foundation to deliver on what people hope health IT can do.