Designing pathways to better health
"What's the current process? What's the needed future process? What's the decision-support for that? What are the data reports and data to support it? What data elements do you need to make this all work?"
These were some of the key questions that Dr. Eugene Nelson, an expert in using outcomes measurements to improve health care, was asking his colleagues when he began creating a workbook for Dartmouth-Hitchcock's Clinical Pathways Project. Launched in September of 2009, the project is designed to directly support D-H's vision—which is to "achieve the healthiest population possible, leading the transformation of health care in its region and setting the standard for the nation."
"To start with," says Nelson, "we identified a handful of high-volume, high-variation, and high-health-impact patient populations—total knee replacement, diabetes, heart failure, spine, and primary care and prevention—which will serve as models for many others that will follow."
What that means in practice is evident in just one among many examples from the project's primary-care arm. "A typical scenario might involve a 65-year-old diabetic who is at home but no longer has the ability to take care of themselves," explains Dr. Don Caruso. He is the project's operations director for primary care; a family practitioner based in Keene, N.H.; and the physician leader on the diabetes/preventive care part of the project. "This is where we need to work with community partners," continues Caruso, "whether it's Meals on Wheels to deliver diabetic meals or with the local pharmacies to provide low-cost medications—to make sure the resources are in place to help the patient."
Caruso, one member of the team working to implement the project throughout Dartmouth-Hitchcock's 23 primary-care practices, sees diabetes as an ideal chronic-care model. "What we've done with the workflow is we redesigned it from being all office-visit-based, to a pre-visit, office visit, and post-visit [basis], so the work gets done in all three of those areas," he explains. "That construct can be applied to any chronic disease, and we think we can have a greater impact on outcomes by looking at it that way."
As the project was getting off the ground, Gene Nelson, who is the director of population health and population health measurement for both D-H and The Dartmouth Institute for Health Policy and Clinical Practice (TDI), gathered together a number of interdisciplinary teams—like the primary-care team that Caruso is a member of—to map out the best possible clinical pathway processes for each of the high-impact areas they'd identified. Since 2009, the teams have been working in concert with IT staff, patient advisors, and care teams who are implementing a new electronic health record system, known as eD-H, that is slated to go live in April 2011.
"When you boil it all down, this work is about achieving better outcomes and health, and achieving better value, and determining what the best way to get there is," Nelson explains. "At the end of each pathway, we've created what we call the Value Compass, a representation of health outcomes and care experience in relationship to cost. It has four dimensions: important clinical outcomes like surgical complications or infection rates; functional or quality of life outcomes; satisfaction—how satisfied a patient is with the process of care and their health results; and costs of care."
These health outcome measures are also at the heart of a new national initiative set in motion by Dr. James Weinstein, copresident of D-H and director of TDI. Announced in December of 2010, the Collaborative is aimed at improving care and lowering costs. Six of the nation's leading health care systems—Dartmouth-Hitchcock, Cleveland Clinic, Denver Health, Geisinger Health System, Intermountain Healthcare, and Mayo Clinic—will share data on outcomes, quality, and costs across a range of common and costly conditions and treatments. TDI will coordinate the data-sharing and analysis and report the results back to Collaborative members to inform the development of best practices. The group will then work to replicate them across the country.
One of the major goals of D-H's own Clinical Pathways Project is to reduce unwarranted variation in care, which undermines quality and safety and drives up costs. "In many cases, we have very good scientific evidence that tells us what the best guidelines for care are, but historically we haven't been very good at moving that evidence quickly and reliably to the bedside," says Dr. Carolyn Kerrigan, a professor of surgery and of TDI and a physician leader on the project.
The fault lies not with individual practitioners, but with the way health-care delivery systems have been organized in the past, says Kerrigan, who also serves as director of the plastic surgery residency program at Dartmouth-Hitchcock. "Everybody means well, but we have different training, we've been trained to be autonomous and make our own decisions, and we tend to believe the way we've been taught is the best way," she says.
A major challenge, then, is to transform how care teams work together. "Our job as project leaders is to coach and guide the frontline teams—those who need to be heavily engaged in the work of redesign—through the process of sitting down together, looking at the evidence, getting a consensus, and then implementing the change," explains Kerrigan. "It's a very complex process, and we've been trying to keep everybody's feet to the fire and move quickly. I think people are doing a fabulous job; we're really proud of what we've been able to accomplish so far."
Running in parallel with the Clinical Pathways Project is the IT work needed to build eD-H, which will support the flow of care. "The electronic medical record is really a tool that can be used to enhance the reliability and safety of the process of care delivery," explains Scott Berry, who as D-H's senior director of clinical population health is responsible for implementing the project. "There are a number of strategies and functionalities we're using—things like checklists, cues, and best-practice alerts to bring more standardization to the care process while enabling providers to still customize around the critical points of care that meet their patients' needs."
Soon, Berry and his colleagues will be developing a program to teach modern improvement science—an approach adapted from many academic disciplines and applied in industry—to frontline teams across the organization. "Teaching improvement science will help us to ensure that we're building and sustaining pathways that are the best and safest for the patients and for the staff, that we're maximizing our resources, and that we're providing care in an efficient and cost-effective fashion," he says.
One of the key benefits of the new electronic medical record system (the electronic system it's replacing was implemented in the 1980s) is that it will allow care teams to capture patient information that can be used for measurement. "We have this great vision as an organization to achieve the healthiest communities possible, but we've got to be able to measure our performance so that it's not just nice words," Berry explains. "So we're starting with something we're calling Vitals Plus 3. The idea is that every time a patient is in a setting where they would have their vitals checked, we'll also be asking them a set of questions having to do with pain, physical functioning, and mental-health status. It will be the first of many opportunities we'll have to expand our view of the patient journey, and by compiling data understand if we are achieving our vision."
The clinical pathway/population health approach is, at its core, a proactive mechanism to make sure that patients don't fall through the cracks, explains primary-care team leader Don Caruso.
"To me, it means that I'm not just responsible for the patient when they're in front me, but also after they leave the clinic, and that I'm taking care of everyone, not just that individual," he says. "You know, I think one of the best things we've done as we've built this is we've brought patients in to be part of the process. The better we understand their needs, the more successful we'll be."