Skip to main content

Q&A: Geisinger's CEO on Cost, Quality, Data Sharing

 |  By Philip Betbeze  
   June 21, 2013

Glenn Steele, MD, discusses the need for healthcare organizations to collaborate and learn from each other with shared data that can offer real-time insights as healthcare leaders re-engineer the industry to provide real value.



Glenn D. Steele Jr., MD, PhD

Some time ago, I had a wide-ranging conversation with Glenn D. Steele Jr., MD, president and CEO of Geisinger Health System in Danville, Pa., about the need for healthcare executives to share information as the industry transforms. Geisinger, you may remember, is a staple of the effort to change the way healthcare works, to make it more interactive, safer, and less costly.

The $2.9 billion system, which serves more than 2.6 million residents in central and northeastern Pennsylvania, has long been held up as an example of high-quality, low-cost healthcare; even the President of the United States touted the system as an ideal to which other healthcare providers should aspire. Steele himself was widely rumored to be Barack Obama's pick for CMS administrator before Donald Berwick was appointed. I spoke with Steele on a range of topics, mostly dealing with what often obstructs the goal of coordinated and less wasteful healthcare.

HealthLeaders: Are senior healthcare executives working together more than in the past to solve the big problems in healthcare?

Steele: They are but there are a number of ways to focus on this. What we do in our own system can serve as a good example. We have three very different demographics we're dealing with: the hospital, the payer, and the doctor group. We have a remarkable interaction between physician leaders and finance people and senior administrative leaders, and the key is we don't expect everyone to have expertise in every area but we expect them to combine their expertise to approach problems. … We're facing an oncoming transformation that is one of the most remarkable in 100 years. There's a growing understanding that no single template is going to work, so I'm not arguing for our specific template.

HealthLeaders: What are some of the factors encouraging or enabling that increase in communication and collaboration?

Steele: I'm biased as the chairman of Premier, but organizations like that represent a huge amount of enabling for systems getting ready for a new payment system. They're effective conveners where we're essentially trying to learn from each other through shared data. When I go to my board it's more convincing when what I'm proposing is backed by metrics that come from 200 of the best, most aspirational organizations in the country.

HealthLeaders: In what areas does it make sense for senior executives to network with each other? For example, on strategy, utilization, costs, efficiency, quality?

Steele: Our assumption is that with so many changes going on in how healthcare is practiced and delivered, thanks to payment reform (whether by government, employers, or payers), this kind of networking increases the knowledge base among leaders, and they are able to quickly experiment among themselves and reach conclusions on best practices in many of these areas. A few organizations have moved into this convener role.

HealthLeaders: Why is it important to have a neutral "convener," as you put it? Is more data by definition better data?

Steele: Yes. We've got a ways to go not only on EHR functional initiation but also with data warehousing. Organizations like Premier are creating their own data warehouse that can be used as outsourced capability for those who don't want to or can't build their own. When we do all our road tests, whether in hospital care re-engineering or focusing on a different way to provide ambulatory care for those who have chronic disease, our most important metrics are total cost of care and whether we have achieved lower cost of care without sacrificing near- and long-term quality.

HealthLeaders: Why is it so difficult to come up with usable information on the cost of care and whether the healthcare dollar is well spent?

Steele: You would think that looking at total cost of care would be an easy task. It's not easy at all. Being able to use data that is aggregated and understanding how to define precisely what goes into the calculation of cost of care has more credibility if you can do it with other organizations that are aspiring to same goal. Ideally you could do that from CMS data. But they're under such duress on analytics that often there's huge lag time between what you do and getting data fed back. Medicare Advantage data, which we have because we have a Medicare Advantage plan, is more accessible. We use that data as a surrogate to determine the effects of the changes we make. For physician demonstration projects, the data coming out of CMS is 15–18 months after the fact. If you're looking at actually getting docs and nurses and community resources to do things differently in caring for patients, that data has to be fed back to them almost in real time to be usable.

HealthLeaders: Are technological innovations making it easier? In what ways?

Steele: At Geisinger we started with Epic in 1995. At the time, we realized that decision was either very silly or very wise, but it's turned out to be wise. Starting almost 12 years ago we began this data warehouse effort. Not only on healthcare provision but also on the cost and transaction side. We spent 4.5% of our revenue per year on maintaining and improving our data warehousing capability. That's a huge amount, and we're constantly upgrading. But the fact is, we couldn't do what we do in value re-engineering without having real-time feedback. The fact that now other systems in the country have this capability gives us a scaling and generalizing capacity that wasn't available before.

HealthLeaders: Are there competitive or antitrust concerns that must be addressed in these types of collaborations?

Steele: As we do this, we have to make sure we're in the regulatory compartments that don't allow us benefit from the things that Econ 101 says is possible in terms of colluding. Secondly, one of the major compelling reasons for consolidation is to allow us to integrate in ways that are only possible in the same fiduciary. But you have to prove that you're bringing value to a population of patients—that you're not ripping them off. How we actually regulate what is becoming a consolidating payer and provider market—there will have to be a different functional metric to show they bring value to a population.

Philip Betbeze is the senior leadership editor at HealthLeaders.

Tagged Under:


Get the latest on healthcare leadership in your inbox.