Coaching with care

During the course of an average workday, Becky Cline sifts through plastic shopping bags full of medication bottles, reviews lengthy post-discharge plans, coordinates follow-up appointments, and acts out various role-playing scenarios with patients in order to educate them on the red flags of their chronic diseases.

Cline is a registered nurse and a transition coach, employed by Physician Health Partners, a management-services organization based in Denver, and she is charged with providing a critical bridge between the hospital and the home.

She works with roughly 25 patients at a time, all of whom are in various stages of the transition program and all of whom have one or more of three chronic conditions: diabetes, congestive heart failure or chronic obstructive pulmonary disease. She meets with patients briefly in the hospital before they are discharged, comes to their homes for an hourlong, in-person visit and follows up with periodic phone calls. The end goal, Cline says, is to empower patients and their families to take a more active role in their care, thereby reducing the rate of hospital readmissions.

While preventable rehospitalizations have been a long-standing problem, providers are now scrambling to find ways to effectively address them, particularly after the passage of the Patient Protection and Affordable Care Act of 2010, which includes a payment penalty in two years for hospitals with the highest rates of readmissions.

The startling statistics related to rehospitalizations are nothing new: one out of every five Medicare patients discharged from the hospital is readmitted within 30 days, and nearly 75% of those readmissions are preventable. And the problem is a serious, costly one, totaling more than $17 billion in additional Medicare spending each year, according to a widely publicized study published in the April 2, 2009 issue of the New England Journal of Medicine.

The prospect of reduced payments, which are set to take effect in October 2012, has prompted increasing numbers of hospitals and payers to turn to solutions that incorporate the use of one person—a coach or advocate—who establishes personal relationships, promotes self-care and guides patients through the thorny period following discharge from a hospital.

“I've had patients who've been prescribed generic and trade versions of medications and are taking both of them, and I've had diabetic patients who don't even know how to test their blood sugar,” says Cline, who has worked as a coach for three years. “There's a lot of confusion so it's very important that patients receive help in an environment where they feel comfortable asking questions and making decisions for themselves.”

Physician Health Partners' approach is based on the Care Transitions Intervention model. Developed 12 years ago by Eric Coleman, a geriatrician and professor of medicine, and his colleagues at the University of Colorado at Denver, CTI is a four-week program aimed at promoting self-management among high-risk patients. The intervention is based on four components, or “pillars”: medication management; follow-up care with a primary-care physician or specialist; use of a paper-based personal health record; and education about the warning signs that a condition is worsening and what to do when they arise.

“Coaches don't fix problems,” Coleman explains. “They model behavior on scenarios such as medication confusion, conflicting advice, follow-up care and what symptoms mean. Adults don't learn by reading brochures. They learn by rehearsal, practice and role-playing.”

Coleman's model has pretty tight guidelines. Coaches visit high-risk patients in the hospital to establish a rapport, meet with patients in their homes—ideally within 72 hours of discharge—and then follow up with them three times by phone. The result, he says, is a relatively short, low-cost, low-intensity intervention that can be deployed in a wide range of settings. As of July, 309 sites in 38 states had implemented the model.

Training is made available to interested sites, Coleman says, and the scope and price tag vary depending on the size of the organization and the number of coaches they want to use.

In a 2006 article in the Archives of Internal Medicine, Coleman and several other researchers presented the results of a randomized controlled trial testing the Care Transitions Intervention model. They found lower rehospitalization rates at 30, 90 and 180 days among patients who had received coaching by advanced-practice nurses. For instance, the intervention group's rehospitalization rate at 30 days was 8.3% compared with 12% in the control group. And at 180 days, the intervention group's readmission rate was 25.8% compared with 30.7% in the control group.

“We were able to demonstrate that, six months down the road, there was a statistically significant difference between those that got it and those that didn't,” Coleman says. “That investment in self-care and education does pay dividends.”

The average cost of a coach, including salary, benefits, cell phone and mileage costs, is somewhere around $75,000 or $80,000, Coleman says. But the net savings, given a standard panel of about 24 to 28 patients per coach at any given time, is roughly $300,000 per year per coach. “And that's based on a very conservative analysis,” he adds.

While Physician Health Partners has one way of implementing the intervention—the company represents primary-care physicians and works with payers to deliver care more efficiently and control costs—other entities, including home health agencies, not-for-profit organizations and hospitals, have also made it work for them.

Facing cash-strapped budgets, some sites have implemented the Care Transitions Intervention program, but with adaptations. St. Joseph Health System-Humboldt County (Calif.) has successfully rolled out Coleman's model at their two hospitals in Eureka and Fortuna using senior-level nursing students as coaches.

“From our perspective, it gave the students the opportunity to gain valuable experience and it also made it easier for us to get started right away,” says Sharon Hunter, coordinator of St. Joseph's care transitions program. “The students get a better understanding of medications and disease processes, but more importantly, they learn a much more holistic approach to patient care.”

Readmission rates among patients who have received the intervention hover around 8.2%, far lower than the national rate of 20%, Hunter says.

Other providers have been even more creative. In April 2009, the CMS chose 14 communities to participate in its care transitions project, a pilot program created to test customized, community-specific approaches to reducing readmissions. One of the 14 regions selected was Lansing, Mich., and its surrounding communities, says Donna Beebe, senior project manager of the care transitions team at MPRO, a quality improvement organization based in Farmington Hills, Mich.

MPRO, which leads Michigan's participation in the pilot, chose to implement the CTI and received on-site training from Coleman and his team, Beebe says.

Sixteen local providers eventually got onboard, including acute-care hospitals, critical-access hospitals, skilled-nursing facilities and home health agencies. But none felt they could employ additional nurses, Beebe says.

“What we discovered is that none of the providers were in a financial position to dedicate resources to hire a coach,” Beebe says. “One facility used therapists, and another lengthened the hospital visit and relied more on follow-up calls.”

Other sites used social work students from nearby Michigan State University who received some specific training, she adds. In some cases those students acted as coaches and at other sites, the students filled in for case managers, who then had time to perform coaching duties. And although they have used a range of deployment strategies, they have seen readmissions plummet by about 50% among patients who received coaching, says Diane Smith, MPRO's director of care transitions.

Coleman says he strongly urges people to maintain fidelity to the model whenever possible, but he also acknowledges that the intervention has worked well with both nursing and social work students. He did, however, warn against using lay volunteers as coaches because of their lack of formal training in clinical care or counseling.

As Coleman's model and others like it began to make measurable dents in rehospitalization rates, Sen. Michael Bennet (D-Colo.) took notice. Bennet crafted a bill called the Medicare Care Transitions Program Act of 2009, using components of the Care Transitions Intervention as well as some of the approaches used by Rocky Mountain Health Plans, a not-for-profit health benefits provider based in Grand Junction, Colo.

In lieu of formal coaching, Rocky Mountain incorporates its care transitions program into its coordinated care within the hospital and case management following discharge, says Sandy Dowd, director of case management. The level of follow-up is based on patients' illness acuity, and includes phone calls, health literacy assessments and referrals, she says.

According to a Dartmouth Atlas Project report released in 2006, the Grand Junction region achieved some of the lowest Medicare costs in the country and was also one of the most efficient regions for end-of-life hospital use.

Introduced by Bennet in May 2009, the bill proposed the creation of a national network of transition coaches that would be managed by community-based organizations, thus removing some of the care coordination burden from providers.

Included in the health reform law was the Community-based Care Transitions Program, a provision modeled after Bennet's bill. The program allots $500 million over five years, beginning Jan. 1, 2011, for community-based organizations to target high-risk Medicare beneficiaries with histories of multiple chronic diseases and past hospitalizations. According to the text of the provision, organizations chosen to participate in the program will provide medication management, self-management support and help arranging follow-up care. Priority will be given to organizations that target small, rural and medically underserved communities.

Interested applicants will also have to demonstrate that they are actively involved in collaborations within the community, Coleman says. “The idea is that you can't come to the dance by yourself,” he says. “You have to create and show partnerships.”

And there are plenty of other interventions that would fit within the parameters of the provision, says Mary Naylor, a professor of gerontology in the school of nursing at the University of Pennsylvania at Philadelphia.

Two decades ago, Naylor and her colleagues began work on the Transitional Care Model, a comprehensive, high-intensity intervention that targets older adults with two or more risk factors for hospital readmission. Naylor's model uses nurses with master's degrees who act as “transitional care nurses.” Nurses support patients through regular home visits and telephone calls, and they also accompany them on doctor's appointments. Unlike other programs that measure success with 30-day readmission rates, Naylor says her goal is to stop the downward trajectory that many patients are on and reduce rehospitalizations in the long term.

“In our most recent trial, we've been able to demonstrate that if you make this investment, you'll see improvements in satisfaction, cost savings and reduced hospitalization through 12 months,” Naylor says.

Naylor also responded to arguments that her approach is too expensive, citing research that shows $5,000 in mean savings per Medicare beneficiary.

“Investing in this kind of intervention gets these people and their families in a position to deal with their chronic health problems in a very different way,” she says. “I don't think it's too intensive; I think it matches their needs.”

Other sites have employed different approaches, including some that use virtual, nonhuman coaches. Project Re-Engineered Discharge is a program at the Boston University School of Medicine that stresses patient education before discharge from the hospital. Led by Brian Jack, associate professor and vice chair for academic affairs in the university's department of family medicine, Project RED uses several tools including an electronic “coach” named Louise for providing post-discharge instruction.

Created in collaboration with Tim Bickmore, a professor of computer science at Northeastern University in Boston, Louise is an animated character displayed on a touch screen mounted on a cart near the patient's bed.

Louise, or the “virtual discharge advocate,” as she is also known, talks to patients and reviews orders, and they respond using the touch screen. The Louise system also tests competency by asking questions such as, “What medications do you take?”

“Our data show that twice as many people prefer Louise to a clinician because she's not in a hurry and she will go over instructions again and again,” Jack says.

For Mary Shankle, an 87-year-old woman living in Temple, Texas, the impact of the home visits and phone calls she received from her transition coach could not be more profound. Shankle suffered from various chronic conditions including hypertension and had been in and out of the hospital several times. After her last hospitalization in September 2009, she was paired with Jamie Jones, a transition coach employed by Scott & White Healthcare, based in Temple, which uses Coleman's CTI model.

Jones worked to help Shankle learn to manage her care and reach her goal of remaining in her home and living independently. Nearly a year later, Shankle has stayed out of the hospital and has learned to spot signs of trouble.

“My life is better now at 87 because my health is better,” Shankle says. “I'm feeling good; my blood pressure is down. I'm eating good food and I'm active. It's amazing, this knowledge, it has kept me on my feet.”