Spreading Innovations to Enhance Care for Older Adults: An Interview With Amy Berman of the John A. Hartford Foundation

Innovations Exchange: Tell us about the Hartford Foundation.

Berman: The John A. Hartford Foundation focuses on improving the health of older adults, which in turn helps to address some of the most critical issues facing the nation’s health care system, including spiraling costs, significant quality problems, and the systematic failure to engage patients, families, and caregivers in making health care decisions and managing health. The foundation is especially concerned with improving care for older adults with multiple chronic conditions and functional impairment. These “frequent flyers” often end up being readmitted to the hospital within 30 days of an initial discharge, a sign of very poor quality care that costs the nation an estimated $17.4 billion each year.

What are some of the best strategies and practices for these patients?

The Hartford Foundation supports multiple projects focused on improving care transitions, which are critical for frail, elderly patients seeing multiple providers in different settings. For example, the Care Transitions Intervention focuses on medication self-management, timely followup care, and knowledge of red flags and how to respond to them. A second example is Care Management Plus, which leverages information technology in primary care clinics to help care managers and an interdisciplinary team improve care coordination and identify and address the health issues of individual patients and patient populations, leading to better health and lower health care spending. Other examples include the Transitional Care Model, which uses specially trained advanced practice nurses in the hospital to coach high-risk elderly patients and their families, and Project BOOST (Better Outcomes for Older Adults through Safe Transitions), which leverages hospitalists to promote safe care transitions from the hospital to home. Hartford also sponsors programs that promote more effective medication management, such as the HomeMeds program. With this program, a social worker already caring for a frail elder enters the names of the patient’s medications, how he or she takes them, and the patient’s answers to three simple questions into a technology that then performs geriatric-sensitive medication reconciliation and alerts the pharmacist, who then works with primary care to address any issues. Finally, Hartford supports programs that help primary care providers screen for and treat depression. For example, Project IMPACT (Improving Mood-Promoting Access to Collaborative Treatment) uses consultant psychiatrists who work with primary care physicians to adjust medications as necessary; this program yields twice the effectiveness of usual care.

With so many different programs, to what extent do you support their dissemination throughout the country?

Spreading the use of effective innovations is central to our mission. The goal is to make today’s innovative programs standard practice throughout the country. Of course, the first step is to make sure that the programs being spread are effective and worthy of becoming the standard of care. Consequently, we always insist on rigorous evaluation to establish the credibility and effectiveness of any program, typically through randomized controlled trials (RCTs).

Once the concept has been proven, the “scale-up-and-spread” phase begins. While academic researchers and many innovators tend to view the publishing of RCT findings as the end of the process, Hartford thinks of it as the beginning. The dissemination phase takes much longer than initial program development and testing. Success requires an unrelenting commitment to spreading programs, promoting implementation in a variety of systems and environments, including in different geographical (e.g., rural and urban) and cultural settings and with different workforces. We often partner with other health care foundations to support the spread of proven programs.

What factors contribute to successful dissemination?

Even though we do not actively promote the dissemination of a program until it has been proven to work, consideration of dissemination comes into play from the outset. When developing grant proposals, we explicitly look for people who are passionate about making change occur, and who are in a position to do so by working outside of their own organizations. We’re not interested in developing programs that will be used in just one institution or geographic area; we want to take things to scale so they have a broad impact on quality and costs. We also believe that models need to be developed in concert with potential adopters of innovations. Models must be doable, sustainable, and derive the value that Dr. Don Berwick termed the “triple aim”--improved health, improved care, and lower costs. When testing a concept, we need to understand the underlying business case and how changing incentives may drive new approaches.

It’s not always easy to identify those interested in promoting widespread change. Many brilliant researchers are not focused on or interested in being leaders of change. We try to work with those who are, and we start by evaluating the leadership capacity of those we fund. To that end, we engage in an open dialogue from the outset, not just about the individual(s) involved, but also about whether the leadership of the organizations they work for will support dissemination efforts. We attempt to identify and fund those who are impatient for change and want to make it their life’s work to turn innovations into common practice. You can tell when individuals have a “fire in their belly” to drive widespread change.

A second, critical factor is to pick the right innovations to disseminate. Not every program that works will be a good candidate for widespread dissemination. For example, over the years we’ve funded five different models of care related to use of geriatric interdisciplinary teams. Four of the five had a very positive impact on patient outcomes. Yet only two of those—the Care Transitions Intervention and Care Management Plus—have been implemented in hundreds of organizations around the country. These models are relatively inexpensive to put in place and can be adopted relatively easily by institutions in various settings within the current constructs of how care is organized. By contrast, one of the other programs entails a significant capital outlay and requires more fundamental changes that may be difficult to implement.

How do grantees promote dissemination and how does Hartford support them?

Hartford supports dissemination through multiyear efforts involving millions of dollars. Sometimes we work in collaboration with other national and local funders, including government agencies. For example, Hartford has funded Eric Coleman, MD, MPH, developer of the Care Transitions Intervention, for roughly 10 years. These funds have supported the development of communications campaigns and a technical assistance center, the forming of connections to thought leaders, the identification and targeting of would-be implementers, and the creation and ongoing operation of a national advisory board that offers thoughtful feedback on the program.

With our support, Dr. Coleman has created substantial infrastructure to support organizations in implementing the program, something that more than 400 institutions have done to date. This infrastructure supports all aspects of the process. As a first step, he helps an institution determine its “organizational readiness” for the program. Training individuals before the organization itself is ready represents wasted effort, as the skills will be lost.

Consequently, Dr. Coleman and his team walk organizational leaders through a formal process to determine their level of commitment and readiness. If they are ready, the next step involves training the coaches, something done by trainers in different regions of the country who use standardized training modules. Typically an organization sends anywhere from 3 to 35 individuals to receive the training. These individuals become certified for having completed the training. Following the training, most organizations receive technical assistance (TA), including help in establishing communication mechanisms and associated workflows. TA may include inperson support and/or access to Web-based infrastructure that supports implementation.