How Dartmouth-Hitchcock CEO's Personal Experience Led Him to Call for a Health Care Revolution
This article, written by Eric Larsen, originally appeared in the Lessons from the C-suite series on the Advisory Board website.
Welcome to the "Lessons from the C-suite" series, featuring Managing Partner Eric Larsen's conversations with the most influential leaders in health care.
In this edition, Jim Weinstein, CEO and president of Dartmouth-Hitchcock Health System, talks to Eric about empowering patients, why D-H dropped out of the Pioneer ACO program but might consider NextGen, and his belief that transparency is key to improving health care quality.
Question: Jim, you've led the internationally renowned Dartmouth-Hitchcock Health (D-H) system now for five years. In your recent book, Unraveled: Prescriptions to Repair a Broken Health Care System, you've called for nothing short of an industrial revolution of sorts in health care.
I thought it was very affecting how you wrote through the lens of patients, particularly about your own family's experience. How did that encounter with the health care system prompt you to try to reform it?
Jim Weinstein: Our first child was diagnosed with leukemia at about 13 months, which is devastating in and of itself. Your whole life changes. We spent the next 11 years trying to figure out how to go through life, to let our child feel that she could have a normal life, despite carrying around IVs and fanny packs filled with one toxic chemotherapy after another; receiving radiation and not being able to be with anybody because her blood counts were so low. She died at age 12 after five relapses and multiple problems secondary to the treatments.
After that experience, I said, "I've gotta change this system." I'm not sure I've been effective, but I've tried.
Q: One of the most upsetting things you write about in the book was when your daughter's doctor said you risked being sued if you didn't follow the recommended treatment. That's an unimaginable thing to hear. And I would surmise that's one of the things motivating you to seek policy changes and empower patients.
Weinstein: The legal threat in that situation was totally inappropriate. Of course, we were going to follow what we thought was right for our daughter, but we didn't want that dictated to us. I thought that was extremely callous and showed me an attitude that I never wanted to experience from another doctor or a system. I'd read everything about leukemia; I knew as much as any parent did, at least I thought I did. I was only trying to make decisions about the health of our child. Those kinds of things unfortunately seem to happen to patients more often than not.
I think we as a society should embrace the doctrine of "informed choice" where the patient/family are active participants in decision-making based on good information. Many patients don't have the decision-making tools or don't have family members who can help. And yet, they're being asked to make decisions without the help or resources they need.
That's why we opened the first in the nation Center for Shared Decision Making at D-H. I wanted to have a facility where patients could hear actual patient voices and narrative, for those facing similar decisions.
Q: There are some unintended consequences of equipping patients with more information. For instance, there's the RAND study on cost-sharing that showed that patients self-ration (and not necessarily in the most thoughtful manner) when they are more aware of copayments and other coinsurance. How do you solve for that?
Weinstein: Having good information about the risks and benefits of a procedure or treatment is always a good thing and something every patient can benefit from. What the RAND study focused on was giving patient’s information about their own cost and the impact that had on their decisions about care. And you're right, in some cases, people with chronic conditions that would have benefited from intervention avoided care because of the cost.
Today, we're seeing that problem multiplied as deductibles and copays increase. People can't afford to get the routine care that might allow them to manage their condition and avoid a serious health episode.
At D-H, we're approaching that with non-bricks-and-mortar solutions that are convenient for patients and low cost. Our ImagineCare program uses wearable devices and HIPAA-compliant cloud-based technology to provide 24/7 monitoring, management, and intervention for our population. Early results with our own employee base of 10,000-plus have been really encouraging. In the first six weeks alone, we saw a 50 percent reduction in poorly controlled high blood pressure, a fantastic result. And the satisfaction rate is very high. We're now working with other employers and looking at how to move this out to key populations in our service region.
These kinds of solutions, coupled with good information to allow patients to make informed choices, are the approaches we need to pursue.
Alternative payments and capitation
Q: Shifting gears a bit, D-H left the Pioneer ACO Initiative last fall. You're not alone: Many of the original Pioneer ACOs have left the initiative. You were very transparent in explaining D-H's decision to not continue with Pioneer, and I'm curious if you could you talk about why you didn't go forward with the Next Generation ACO model.
Weinstein: We'll still consider joining NextGen, maybe just in the parts of our system where it makes sense. It's hard to run a system with data that comes in too late to act upon and actually make changes. That's been one of the problems with CMS programs that needs to be addressed.
Having used CMS data for a long time now in very complicated ways, I can say we're really good at it at Dartmouth. When we struggle with it, like we did in Pioneer, it's a problem. I can't imagine how others can do it. That said, I think NextGen is a lot better, and APMs (alternative payment models) are essential, if we are ever to get to true value-based payments. Cost is one issue, and quality measures are another. What is really needed is transparent reporting of patient self-reported measures of the impact of treatment.
Q: Over the summer, Harvard Business Review published a couple of contrasting articles between Michael Porter, advocating for bundled payments, and Brent James, advocating for capitation.
You address both of these approaches/issues in your book, and while admittedly this is a bit of an over-generalization, you clearly come down on the side of capitation. I appreciated your critique of the vulnerability of bundles, but also your acknowledgement of the inherent problems with capitation. Could you weigh in on both of those sides?
Weinstein: Bundles are interesting but may be inadequate. And capitation, to me, means you're responsible for a population and have basically a fixed amount of money to care for a population.
DRGs were kind of the first bundle, basically a statistical system of classifying any inpatient stay into groups for the purposes of payment. In the case of bundles, I do worry that in a fee for service model we will see indications changed as to maintain volumes necessary to maintain revenues. In that scenario, yes there will be bundles, but more of them.
We have to get away from gaming the system. My fear is that people will regress to the mean, and they'll figure out how to get back to where they were without really changing the system for the benefit of the population they serve.
Q: There's a lot of rhetoric out there right now around reducing clinical variation, and not necessarily a lot of solutions. What's your take?
Weinstein: Sometimes it's easy to put your head in the sand because you're surviving; your system is doing well. And, it seems to work.
We have to have serious conversations by serious people who can legitimately say, "Let's not try to cast a strategy for everybody. Let's test things that actually work and are measurable and report those results openly and publicly. And then let's spend the next five years figuring out how to get to scale." This is a 20-year problem of the next "industrial revolution." Short of that, I think people are putting their fingers in the dike.
I think the way to get at variation it is to transparently and publicly report the cost and the outcomes of care with the variation within and across regions. And if you can demonstrate that everything you're doing is good, with no unintended consequences then, you know, keep doing it.
Collaborations and partnerships
Q: D-H has partnered with Harvard Pilgrim to create a health plan in New Hampshire. How did you find a willing partner? How are you working to ensure the health plan is successful?
Weinstein: We couldn't afford to do a health plan ourselves. We don't have the deep pockets for the risk side. So partnering was a good strategy. Harvard Pilgrim and D-H are alike in that Harvard Pilgrim is a not-for-profit insurance company. They had the highest customer satisfaction in the country from health plan users in their system. And their CEO was willing to talk to me about it.
Q: In the partnership, is it true delegated risk?
Weinstein: No, we have a total cost of care target.
Q: If anybody's prepared to take on full cap, I would argue that D-H is part of a small number of organizations.
Weinstein: Sometimes I think we are, and sometimes I think we're not. Most of our work is still fee for service. In my mind, you don't build something for today, you build something for the future because today is just a mess. You think about what it is you're trying to build toward and make sure that you stay on track for that. So, we're building toward a mostly capitated system. We're not yet there, and much work remains. We'll pull all the throttles we can to stay financially solvent while we move toward a much more sustainable model with more predictable revenues to manage a known population. It's challenging.
Q: Let's talk about one of the more complex and controversial aspects of taking on risk, which is pharmacy (and specialty pharmacy prices specifically). Lots of folks opining on this issue nationally, but not a lot a lot of constructive or implementable solutions. I know there's not a silver bullet, but I'd love to hear your thoughts on this issue.
Weinstein: How I think about it is: the challenge is how do we make sure we offer the very best, evidenced-based medicines for our patients in an environment now where many are facing large percentage increases in the costs of their medications. Though in many instances these drugs are groundbreaking, the costs are in the multitude of thousands of dollars for treatments. Combined with the very high deductibles of many health plans now, families are faced with harder choices.
We are working with others to try to rein in some of this pricing, the most effective levers being larger partnerships and market pressure where we consider appropriate other therapies. A new hope in this area are biosimilars (not a true generic medication but almost exact), which have been used in Europe for a few years are now just coming to the U.S. market. This will provide other choices for medications like Remicade, Neulasta and others that drive a large expense. I see the success of these biosimilar agents as a critical step to help control pricing in areas that never had competition before.
As drugs become more and more costly, I worry about the haves and have nots. I go back to the patient.
Q: Back to the subject of partnerships, you're part of a high-value health collaborative (HVHC) with about a dozen other organizations. How did that come about, and what are you hoping to achieve?
Weinstein: HVHC came out of conversations about variation that I was having with Denis Cortese and Rob Nesse at Mayo, Brent James at Intermountain, and others. We looked at our own organizations and Dartmouth Atlas data and realized that even across some of the best places in country, we had no idea about our own variation, its causes, and how to get to the right rate. We said, "Hey, if we're gonna really change health care, let's do it on the ground and start with our own organizations."
Sepsis is a good example. The national in-hospital mortality rate for sepsis is close to 50%. There are 650,000 to 700,000 sepsis-related deaths a year in hospital. Our own in-hospital mortality rate was just more than 50% when I got the actual data. I had no idea. I was blind. I got our leadership team together, I said, "This is unacceptable. In the next three to six months, it has to be close to zero." And we did get close to zero, and we've sustained it. We've saved hundreds of lives and millions of dollars. Nothing fancy, nothing expensive. It can be done anywhere. We've now started to spread this nationally, through the collaborative.
Besides sepsis, we've been working on joint replacement, heart disease, diabetes, and other common, high-cost, high-variation conditions. The concept is that by openly sharing our processes and outcomes, we can all get better. And then we can move those improvements out to others. We're now publishing the results of our work, transparently, with our organizations named. Part of the requirement to be a member of HVHC is an agreement to share data and outcomes publicly. If you're going to change, then you have to go public.
Q: That level of transparency—when those premier systems are forced to disclose that they have variability from hospital to hospital in the same system, let alone geographically compared to other systems—has to be an uncomfortable moment.
Weinstein: To argue the other side, I think it's a comfortable moment because when you let down your defenses, you can improve. Some people don't like that because they're competing, and they don't want to show that their results may not be what they had hoped. I feel the opposite. I feel that if you don't show your results good or bad, you can't be effective at getting better and both need to be better, continuous improvement in medicine is an essential for any learning health care system, and we are all learning how to be better.