Reducing Hospital Readmissions, One Health Care System at a Time

Dr. Amy KindDr. Amy Kind and colleagues developed a system used at Veterans Administration hospitals to help manage high-risk patients transition from hospital to home. Now, it’s being implemented at other locations.
Being discharged from a hospital stay can bring a range of emotions for patients, from relief at returning to familiar surroundings to apprehension over continuing their recuperation away from the hospital and its personnel. For some, though, their progress is thrown into reverse and they end up returning to the hospital within a short period of time. Unfortunately, preventable readmissions are common; nearly 1 in 5 older adults in the US is rehospitalized within 30 days of discharge, at a total combined cost of more than $25 billion per year. 
Changing this situation for the better is a key focus of Amy Kind, MD, PhD, associate professor, Geriatrics and Gerontology, whose research has shown that readmission rates are correlated with healthcare disparities in many regions of the United States. “I'm really interested in improving care in the most vulnerable and resource-strapped areas of the country. How can we create real change?,” said Dr. Kind. 
In 2010, Dr. Kind and colleagues launched a transitional care program at the William S. Middleton Memorial Veterans Hospital to help bridge hospital and home during the pre- and immediate post-discharge period. The Coordinated-Transitional Care (C-TraC) Program is a low-cost, primarily telephone-based, protocol-driven transitional care program carried out by a nurse case manager. An analysis published in 2012 showed that it reduced 30-day rehospitalizations by one-third, leading to significant cost savings while being well-received by veterans enrolled in the program. Which immediately brought up the question: Could this be implemented elsewhere?
Over the next few years, Dr. Kind and colleagues across the country did exactly that. “C-TraC is a program that's quite scalable. It's appropriate for both small hospitals and large hospitals,” said Dr. Kind. Moreover, the telephone-based program made it appropriate for rural areas and hospitals serving a large geographic region. 
In July 2013, C-TraC was launched at UW Health's UW Hospital (now known as University Hospital) in Madison. A year earlier, University Hospital had piloted a transitional program that used home visits for high-risk medical inpatients. But with many patients living beyond the reach of home visits and others declining to allow medical staff to enter their homes, the program experienced low enrollment. As a telephone-based program, C-TraC was introduced as an alternative. 
At this point, the story—detailed in a newly published article in the Journal of American Geriatrics Society—took an interesting turn: rather than attempting to replicate the program from one hospital facility to another without modification, the implementation team decided to assess whether flexibility would be needed to ensure success. “Our goal is to achieve long-term success as defined by local stakeholders. Achieving reduced rehospitalization may be a common goal across different institutions, but how each site defines success aside from reducing rehospitalization may differ,” said Dr. Kind. 
The team leveraged a concept from the field of implementation science that had been used by the Centers for Disease Control to guide effective dissemination of human immunodeficiency virus (HIV) interventions at different locations. The model, called the Replicating Effective Programs (REP) Implementation Theory, provided a framework and process for adapting the core features of a health care systems intervention (e.g., activating patients), while engaging key stakeholders at a new site and determining additional desired outcomes that are unique to that institution. “Different stakeholders have different goals, and we try to help our stakeholders tailor their intervention plans to meet their objectives for success,” said Dr. Kind. The REP model also provides a framework for identifying program champions, ensuring a robust launch phase while carefully monitoring outcomes and refining as needed, followed by steps to ensure long-term maintenance and further dissemination (for example, to additional units or services.) 
At University Hospital, representatives from the executive team, inpatient nurse managers and physicians, outpatient primary care providers, and patients formed a group that worked with the implementation team to follow the steps of the REP model, including clearly and specifically defining which outcomes and goals would be indicative of local success for a new program. Some of these goals included mitigating confusion about the discharge plan and identifying and correcting medication discrepancies. 
The results of the launch were gratifying for Dr. Kind, Maria Brenny-Fitzpatrick, MSN, FNP-C, GNP-BC, APNP, director of transitional care, UW Health, Beth Houlahan, DNP, RN, CENP, senior vice president and chief nursing officer, UW Health, and many others involved in the effort. “Outcomes of C-TraC at UW Health were assessed by their internal business office, and it was very quickly shown that the program was successful in reducing re-hospitalizations,” said Dr. Kind. In its first 16 months, C-TraC successfully enrolled 1,247 individuals with 3.2 full-time nurse case managers. 
As C-TraC begins to roll out at other hospitals including the VA hospital in Boston and locations in rural Colorado, Dr. Kind hopes that the value of both the program and the REP implementation model continues to make an impact. “I believe that a lot of the field of dissemination has focused overly strongly on fidelity—in other words, replicating something just the same as it was in the place where it originated. Fidelity is critically important for some programs. However, for others, a strong and formalized approach to adaptation is equally, if not more, important for the program’s long-term sustainability,” said Dr. Kind. While a singular focus on fidelity might make sense for bio-physiological procedures such as standards for inserting a central line, C-TraC and many similar programs involve more social variables, so flexible adaption to new systems, settings, and cultures is critical. “The idea that we could take a social- and psychologically-based program designed in a Midwestern VA medical center and launch it without adaptation in a very different location, system or culture doesn't make sense to me,” said Dr. Kind.
In the end, it’s important to understand what will help programs succeed. “You have to have respect for the fact that you can't just take a seed and plant it without tilling the soil first,” said Dr. Kind.