The height of health IT

January 29, 2010 by Managed Healthcare Executive Magazine Online
Filed under Healthcare IT, Managed Healthcare

Even health insurance giant WellPoint—with more than 35 million members and arguably enough reach to change the system with sheer volume alone—is taking few chances on the future of healthcare delivery. Like most plans, it’s testing new programs with cautious optimism, while aiming for large-scale implementation.

Charles Kennedy, MD, WellPoint’s vice president for health information technology, has a vital role in the plan’s innovation because few initiatives these days can be accomplished without the backbone of health IT.

Specifically, WellPoint’s emerging Individual Health Record—a simultaneously patient-facing and physician-facing electronic record—is “almost an air traffic control system to manage disease,” according to Dr. Kennedy. It’s probably one of the most promising efforts to control costs among members with chronic conditions. Pulling claims and clinical data through complex algorithms to arrive at a functional health summary differentiates the Individual Health Record from the typical EMR system.

“If you’re a hospital or institution, you have a variety of clinical data sources that have information on the patients that you see,” he says. “If you haven’t deployed an interface engine or some way of pulling those various clinical data sources together, you’re late to the party, and you need to do that ASAP.”

With more than 20 years of experience comprised of clinical practice and health IT implementation, Dr. Kennedy began his career in internal medicine. When he was a resident at Highland General Hospital in Oakland, Calif., he noticed how the patients’ needs far outstripped the hospital’s resources, and that experience solidified his vision of where medical care and information should meet.

“We tried to treat each patient regardless of who they were or their ability to pay,” he says. “It had the unfortunate side effect that we never thought about cost. We only thought about what was right for a patient. But that created a system where people are actually being hurt because they can’t afford care. I began to realize that the very laudable and applaudable approach of not caring about cost—only the patient—is right, but that doesn’t mean you can become cost unconscious. Cost unconsciousness has its own set of bad outcomes. That’s what’s led me into thinking we need to be more efficient. We need health IT.”

Earlier this year, Dr. Kennedy was named by the Government Accountability Office as a member of the new Health Information Policy Committee, which was established by the American Recovery and Reinvestment Act. Serving a three-year term, he and other committee members are creating policy framework for the development and adoption of a nationwide health IT infrastructure, including standards for the exchange of patient information. The committee will also make recommendations for handing out the $38 billion in health IT funding earmarked in the reinvestment act.

WHAT ARE SOME OF THE HEALTH INFORMATION POLICY COMMITTEE’S GOALS?

A:We’re trying to make sure the Obama health reform strategy becomes real. What people don’t realize is the number of things the industry and the government agree on. For instance, the government invested $1.1 billion in comparative effectiveness research.

The stimulus bill has $38 billion in it for health IT, and we’re trying to help the government develop policies to spend that money wisely. Our function is to say, ‘How do we take this incredible resource that Congress and the President have given us, and how do we turn it into an investment that creates healthcare value for the whole country?’ It’s a massive undertaking.

Our first objective was to ensure that the money from the stimulus package paid out over five years created value. We asked ourselves where we wanted to be five years from now, and then we worked backwards from there.

Deploying computers is not the goal. Having physicians and patients use computers to create better care at a lower cost is the goal. To do that, we have to set the bar high for the care system. Not only must you use the computer, you must use it in a meaningful way for better care. These are the ‘meaningful use’ criteria that we’ve published.

If we distribute a substantial number of computers, and physicians don’t use them, we won’t be successful. We didn’t want to focus on technical measures. We created the meaningful use criteria, and every single one is clinical.

We want physicians to achieve a clinical result, and we want information technology and the money in the stimulus package to be a contributor to that improved clinical result. For instance, one of the criteria is to avoid 1 million heart attacks and strokes by 2015. Another is to make cardiac disease no longer the leading cause of death in the United States. Those are stretch goals. That is not something simple and trivial.

It would have been much easier to say, ‘Our goal is to make sure 90% of physicians have computers.’ But we consciously didn’t do that because we recognize that health IT is a tool and that other changes need to happen.

HOW WILL THE INDUSTRY ACTUALLY ACHIEVE MEANINGFUL USE AND OTHER MILESTONES?

A:The law is actually quite specific in defining what a qualified system is, and we have a subcommittee that’s identifying the actual entity—such as the Certification Commission for Healthcare Information Technology (CCHIT)—that will assess systems as to whether they qualify or not. The bigger challenge is data integration.

Everyone recognizes that healthcare is horribly fragmented, that there are silos of care. We know that there’s massive inefficiencies, and there are significant quality concerns because information is not shared as people move across silos.

The challenge with data integration is that we really haven’t figured out how to do it correctly. If you’re an integrated delivery system and you buy one EMR, that’s fine, and that works. But 70% of physicians practice in a community setting, solo and small group practice. You have this tremendous problem that all of these systems are different. They call things by different names, and they even capture different sets of data.

WELLPOINT HAS CREATED THE INDIVIDUAL HEALTH RECORD SYSTEM THAT USES ALGORITHMS. HOW WILL THAT MAKE A DIFFERENCE?

A:Algorithms, also known as decision support, are going to be the key to getting value out of these systems. Let’s say the federal government funds a comparative-effectiveness study that identifies a new drug is great for certain people. In today’s world, we know it can take up to 17 years for that to be commonly found in a physician’s paper record. With this approach, you can create an algorithm as soon as physicians or specialty societies have decided on certain best practices. Now you’ve created an infrastructure to get that message to every doctor, but only when there’s an appropriate situation for that rule to be applied. That will take that 17 years down to 17 days. That’s a huge advance.

Let’s say we have noticed that there’s a lot of inappropriate use of PET scans. In today’s world, a doctor would have to call us for preauthorization every single time he orders a PET scan. In the future, the algorithms will be running, and they will only alert the doctor if there’s an issue with a PET scan. Today, they call 100% of the time, and we generally approve the scan more than 90% of the time. Algorithms will take hassles, administrative costs and bureaucratic burdens out of the system.

The right kind of health IT allows us to use new knowledge from our outcomes research subsidiary [HealthCore] and any gaps in a member’s care identified by our informatics company [Resolution Health] in much more effective ways. The right kind of health IT allows these advances to be applied real time at the point of care while the doctor is treating the patient or helping the patient at home.

IS WELLPOINT’S INDIVIDUAL HEALTH RECORD WORKING? HOW IS IT ANY BETTER THAN OTHER EMRS OR PHRS?

A:We’ve run a pilot in Dayton, Ohio. The idea was not just to create interoperability—don’t just allow System A to talk to System B. When you connect systems together, what you create is just a data dumpster. It’s like putting a jigsaw puzzle on a physician’s desk.

That information has to be organized to just the summarized information that the doctor needs…You don’t take all of the information out of these various systems, you only take the information necessary for the ongoing management of the patient.

Many EMR implementations have failed to show value. About 30% of the time, physicians will actually turn them off because they are incredibly time-intensive and will reduce a physician’s productivity. That will hit them in the pocketbook. We’ve looked for solutions that wouldn’t be so intensive from a physician’s data-entry perspective and would do more sorting of information and presentation of information.

Physicians are not data generators. They’re data consumers. Their orders create significant amounts of data, but the physicians themselves usually just scribble a relatively brief note. The problem with many EMRs is they will require physicians to become data-entry clerks.

In Dayton, Ohio, we have a very significant market share. We’re Anthem Blue Cross Blue Shield of Ohio, and we also have a strong partnership with Kettering Hospital Network.

Kettering had already installed an application integration solution, so even though they had 120 different clinical sources, many of those clinical sources could be accessed through infrastructure they had already built. That made it easy for us to collect all of the clinical data out of their systems. We built feeds to the application from Anthem’s claims systems. We were able to get this application up and running in a little over three months, which is incredibly rapid. We made it available to the patient in the form of a PHR and to the doctor in the form of a CCHIT-certified EMR with e-prescribing.

When we looked at who was using the tool, we found that patients who had a higher illness burden actually made preferential use of the tool. For many of the tools we’ve deployed, the ‘worried well’ have been the type of people who used it, not the people with the chronic disease that we really need to reach.

We noticed the people who used the tool and had the higher illness burden, their cost increase year over year was actually less than the people who didn’t use the tool, even though those people who didn’t use the tool were healthier.

We built algorithms in the system that exactly correlated with various HEDIS measures and every time the doctor or the patient logged on, they could see their exact compliance. By giving the patients and the doctor the same information in a simple red light, yellow light, green light format with algorithms enabled us to see quality improvement scores of anywhere from 10% to almost 40%.

WHAT’S THE BUSINESS CASE FOR A HEALTH PLAN TO CREATE A SYSTEM LIKE THAT?

A: Our strategy is maximizing healthcare value, and healthcare information technology is really a tool to get you there. But it has to be the right kind of health information technology. It has to influence doctor’s decisions, and you have to present sufficient clinical data—not mountains of data but the key things the doctor needs to know so that you can influence his decision to do something that’s consistent with the evidence base, or to prescribe a drug that will cost the patient less but has the same likelihood of creating a good patient outcome.

If you look at why healthcare spending is out of control, it’s chronic disease, not health plan profits and not health plan administrative costs. We are seeing an explosion of chronic disease in this country, and chronic disease is managed largely by the patient at home. They’re managing their diabetes 99% of the time at their home, not in the physician’s office. If you don’t make your health IT solutions patient-centric and if they don’t address chronic disease, I don’t think that you’re going to get the kind of value that you want.

HOW ARE THE PHYSICIANS EMBRACING THE INDIVIDUAL HEALTH RECORD?

A:We have 300 physicians using the system now. We’re planning for a broader rollout to the greater Dayton area in 2010 to virtually all primary care physicians.

What we’re focusing on is chronic disease management, and there’s not huge debate about many of the things that need to be done to take care of these patients. That’s not the problem. The problem is actually getting it done. The physicians in general have been positive and are beginning to see how their lives could be easier.

We also added all of our pay-for-performance rules. We pay physicians more if they practice medicine consistent with the evidence base, and we took the existing measures and turned them into algorithms in the system. As long as the physician follows all the alerts, he can be sure that he’s going to maximize his pay for performance incentive. That’s convenient for the doctors because what they usually have to do is identify the patients who haven’t had certain interventions and then reach out and call them.

We’re just starting to incorporate our utilization management rules. If we can begin to move those algorithms to the point of care, then physicians might not have to call except for when there’s a real reason to discuss something, which might be 5% of the time.

DETERMINING THE EFFECTIVENESS OF TREATMENTS IN ORDER TO BUILD THE ALGORITHMS IS AN EXPENSIVE PROCESS. HOW CAN IT BE DONE?

A:This is the beauty of health information technology…if you bring it together in a repository that’s reflective of the patient’s clinical condition and how they’re being managed, you can begin to do database-driven studies rather than very expensive prospective clinical trials where you’re enrolling patients and following them over time. You can begin to do database driven studies that are a fraction of the cost. No, they’re not the gold standard, which will always be a randomized perspective-controlled clinical trial, but there’s a lot of information we’re going to be able to glean out of database-driven studies that are more observational and more retrospective.

BE A VISIONARY. WHAT DO YOU SEE AS THE POTENTIAL FOR HEALTH IT?

A: I hope that every time a patient needs information when they’re home or need to take care of their chronic disease or want to stay well, that they have that information at their fingertips, it’s actionable, and they don’t even have to think about it. If we can make it that easy—and there is a path to get there—we could actually fix the healthcare system.

Charles Kennedy, MD, has held strategic health IT positions with a variety of organizations. He also served as the medical director of a California health center in addition to other clinical service. He earned an MBA from Stanford University, an MD from the University of California at Los Angeles, and a bachelor’s degree in genetics from the University of California at Berkeley.
” Physicians are not data generators. They’re data consumers.”