|By the Innovations Exchange Team, based on interviews with Dr. James B. Battles and Dr. Stephen Hines|
About Stephen Hines, PhD
What were the goals for AHRQ with respect to the CUSP program?
Dr. Battles: The Comprehensive Unit-Based Safety Program (CUSP) represents a different kind of endeavor for AHRQ, one that reflects a shift in the agency’s focus that has taken place over time. For many years, AHRQ operated under the assumption that “if we publish, people will adopt.” Consequently, the agency focused on supporting research designed to test the effectiveness of interventions and then publishing the results. The underlying assumption was that organizations would read about, become interested in, and adopt the best programs. Over time, however, it became clear that this model was insufficient to promote the implementation of effective interventions.
The agency’s former director, John Eisenberg, MD, was instrumental in transitioning AHRQ to focus more on implementation, in response to what became known in the agency as “The Porter Question.” At a 1998 hearing of a House Appropriations subcommittee, Rep. John Porter questioned the impact of AHRQ-sponsored research. After hearing Dr. Eisenberg’s testimony, he noted that the key question is not “how many reports have been done, but rather how many people's lives are being made better” because of the research.
To that end, AHRQ funded the initial CUSP Keystone project as part of its Partners in Patient Safety or PIPS project, which focused on implementation of proven tools and resources to promote safety. The project, which ran from 2003 to 2006, involved funding of just under $1 million ($300,000 a year for 3 years). Participating hospitals significantly reduced central line-associated bloodstream infections (CLABSIs), proving that it might be possible to move to zero preventable infections. The question then became—how can AHRQ take this idea and expand it throughout the nation?
What were the critical success factors for spreading the innovation?
Dr. Battles: We first considered the mechanisms available to AHRQ to embark on such an endeavor. Although the initial Keystone project had grant funding, agency leaders questioned whether a grant could work for a national rollout. The task at hand seemed more amenable to a “work-for-hire” contractual approach.
The second consideration related to a national infrastructure to support hospitals in implementing the program. AHRQ does not have such an infrastructure, as the agency has no field offices or other “on-the-ground” resources. Because the Michigan Health & Hospital Association (MHA) had been critical to the success of Keystone, AHRQ leaders began looking for a way to bring other state hospital associations into the effort. Health Research & Educational Trust (HRET) represented a natural partner for doing so, as it had long-standing relationships with the various state hospital associations. Consequently, AHRQ entered into a sole-source contract with HRET to spread the program, first as part of a $3 million initiative targeting 10 states, and later as part of a $20 million initiative targeting all 50 states, Puerto Rico, and the District of Columbia.
Dr. Hines: There’s no question that the state hospital associations have been critical to CUSP’s success, as they provide a quality improvement (QI) infrastructure for member hospitals. Most organizations that succeed in spreading innovations use an established infrastructure. The Centers for Medicare & Medicaid Services (CMS), for example, works through established Quality Improvement Organizations (QIOs) in each state, and through existing renal networks for dialysis care. The Centers for Disease Control and Prevention (CDC) works through state health departments. AHRQ needed the same kind of network to spread CUSP, and the state hospital associations represented a natural partner.
Tell me more about the hospital association network and its QI infrastructure.
Dr. Hines: At this point, between 45 and 47 associations have begun working on CUSP-CLABSI. Naturally, some variability exists in the effectiveness of these efforts. Some state associations have very strong quality improvement infrastructures and can effectively lead the CUSP-CLABSI initiative with limited support from the project’s national leadership. Other state hospital associations have more limited resources available for QI and less experience leading statewide improvement initiatives. Over the last 3 years, however, there has been a general strengthening of the QI infrastructure across the associations, thanks to CUSP and growing linkages between hospital payments and quality. The key challenge for AHRQ and HRET is to lay out a clear, general improvement strategy that every association can understand and lead, while still allowing flexibility for states to tailor the initiative to fit with their resources, skills, and with other statewide improvement activities. Naturally, the national project team will do more in those states that do not have the internal capacity and resources to do more things on their own. Over time, however, more improvement activities can be owned at the state level as the local infrastructure evolves and strengthens, and as state associations increase their investments in QI, supporting hospitals as they transition to an era of value-based purchasing that features incentives for high-quality, efficient care.
So what makes for a good “scale-up-and-spread” network?
Dr. Hines: At a macro level, we need a multistate infrastructure that can accommodate the needs of hospitals and hospital associations that vary tremendously in terms of their capabilities. For example, some associations have been collecting data for many years, while others have not; some have a firmly entrenched QI culture, while others do not. Some have been more successful than others in securing significant long-term funds from foundations and other sources to support their improvement efforts. Consequently, the national infrastructure must account for these variations.
Visionary leadership also remains important, particularly an understanding of the growing importance of providing QI support to member hospitals. Finally, a certain amount of creativity is necessary. For example, some state associations have found ways to embed consultants within their organizations to provide more support to hospitals; others have partnered with QIOs or other statewide organizations to expand capacity. Such creativity can make a huge difference whenever resources and staff remain limited.
Dr. Battles: To understand how best to spread an initiative like CUSP, one must first remember what CUSP is—i.e., an organizational change package that promotes clinical practice redesign at the unit level by creating a learning community and ownership for change at that level. This approach stands in contrast to traditional QI, which has focused almost entirely on measurement. Traditional QI assumed that simply informing providers about their performance—i.e., measurement—would be enough to get them to change. Now the QI field recognizes the need for an organizational change package that focuses on both the cultural environment and practice at the unit level. Just feeding back information cannot magically make things better. Rather, organizational and practice-based changes must occur, and often these changes can be quite disruptive within a community of practice.
Tell me about the disruptive nature of QI and CUSP. How does one overcome resistance from those being disrupted?
Dr. Battles: CUSP has been a disrupting force within the infection control community, which initially felt threatened as the locus of control moved from an external watchdog agency (CDC) to the unit itself, and as the focus shifted from measurement to QI. This disruption has been felt inside individual hospitals and at the national level, as existing power structures become threatened.
Dr. Hines: To overcome resistance, one must understand and try to accommodate the interests of those whose work activities may be disrupted. CUSP shifts the traditional responsibility for monitoring infection rates away from clinicians focused on infection prevention to the entire care team. That could be threatening, but it also could be very liberating. Rather than feeling bogged down in data collection, monitoring, and reporting, the infection prevention professional can work with the CUSP team to focus on what they have been trained and are uniquely qualified to do, preventing infections.
With the CUSP-CAUTI (catheter-associated urinary tract infection) project, we’re working with relevant professional associations to develop an integrated, multidisciplinary program to equip infection preventionists, nurse executives, unit managers, hospitalists, and others involved in infection control to collaborate effectively to prevent infections. Understanding each others’ roles, responsibilities, and concerns and coordinating their efforts will better enable them to support their peers in participating hospitals. Breaking down silos between different groups of clinicians and treating infection prevention as a shared responsibility is essential to a successful and sustainable infection prevention effort. Development of this program is just getting started, but it’s already clear that there are large benefits to facilitating conversations that enable each group to better understand the perspectives of others.
How can one effectively manage the relationship between the original innovator and those involved in spreading the innovation?
Dr. Battles: The original innovator will naturally be critical to spreading the concept. Johns Hopkins, for example, has proved to be a wonderful partner in spreading CUSP nationwide. Not surprisingly, however, pioneering organizations that innovate on a local level often do not have the resources or mechanisms available to spread something throughout the nation. The innovator, therefore, provides the intellectual property. However, someone else often needs to serve as the scaleup organization, which is where MHA came in for Michigan and HRET for the national rollout. These organizations have the infrastructure—data systems, management, and other resources—to spread the concept broadly.
Dr. Hines: Part of the challenge relates to the ability to reflect on what has and has not worked effectively. Those involved in Keystone know that CUSP worked in reducing infections, but not necessarily which components were integral to success and which were less central. For example, did we really need to have weekly team calls, or would biweekly calls have worked just as well? And we still don’t know whether and how aspects of the local environment contributed to success. Those involved in taking this innovation to the next level—whether they are the original innovators or others—must remain reflective about why something worked. Sometimes—although not always—it can be easier for an outside organization to engage in this type of reflection. Given that environmental factors are in a continuous state of change, the interventions themselves must continually morph over time as they are rolled out to organizations facing new challenges.
How have CUSP and its infrastructure evolved? And what are its key functionalities and capabilities today?
Dr. Hines: We began with a leadership group that coordinated with both Johns Hopkins and MHA as part of the Keystone project. Originally, Johns Hopkins had separate funding from a private source to embark on parallel projects in other states—over time, these projects were merged under a single, AHRQ-funded plan. Also, it took some time to clearly define the roles for the hospital associations and hospitals. As the CUSP-CLABSI rollout moved to other states, it became clear that the same plan could not work everywhere. Consequently, we started with a standard plan, but then worked with states to tailor and adjust it to meet their local needs and circumstances.
When we first started targeting CLABSI, infection rates were much higher than they are today. In fact, when we go into hospitals now, many have already made the “easy” improvements and often have CLABSI rates that are half the level of those of Michigan hospitals at the start of the Keystone Project. So we are adapting strategies to focus more attention on the aspects of implementation that are harder to accomplish. We also have to take steps to avoid “QI saturation,” as multiple QI projects are occurring at the state, hospital, and national levels.
Dr. Battles: As the project evolved, we adopted lessons learned from other AHRQ projects, such as TeamSTEPPS®, which seeks to optimize patient outcomes by improving communication and teamwork skills among health care professionals. At the national level today, we have a strong faculty capable of teaching both the clinical and cultural side of CUSP. Johns Hopkins takes the lead in this area, supporting kickoff and midcourse meetings, Web events, and conference calls. Johns Hopkins and MHA work together to analyze and make sense of the data, and to help struggling units improve their performance over time.
A separate effort focuses on promoting sustainability by equipping those at the state and hospital level to understand and support the core CUSP projects. Over time, we have increased support for the leaders of state hospital associations, particularly their efforts to promote CUSP to executives and other key stakeholders within individual hospitals. More recently, we’ve begun to target professional societies, with the goal of promoting professional collaboration and overcoming any turf issues raised by this disruptive innovation. For example, we’re talking to the leaders of the Association for Professionals in Infection Control about how CUSP can help their profession. As we begin to target surgical-site infections, we will do the same with leaders of the American College of Surgeons.
So how would you summarize your progress in spreading CUSP to date, and what is the plan going forward?
Dr. Battles: AHRQ has had significant success in spreading CUSP. With CLABSI, we’ve gone well beyond the initial million-dollar project in one state, expanding first to a $3 million project in 10 states, and now to a $20-million initiative that spans the nation. The project has also “grown arms and legs,” with an additional $20 million investment in reducing CAUTI; a $10-million, 3-year initiative targeted at safe surgery; an approximately $6-million program for end-stage renal disease (ESRD); and a $700,000 program aimed at ventilator-associated pneumonia (VAP). We’re adapting these programs as necessary—for example, for ESRD, AHRQ works with the existing CMS network rather than the state hospital associations (which remain the primary dissemination vehicle for CAUTI, VAP, and safe surgery). Thanks to the agency’s strong focus on the Porter question and the success of Keystone, Congress has provided AHRQ with funding to tackle other important safety problems. For example, AHRQ has a $34 million annual budget targeted at eliminating health care–acquired infections (HAIs).
Going forward, AHRQ’s approach will vary depending on the problem being targeted. CLABSI rates have come down substantially across the nation. Bringing them down further will require a more detailed understanding of individual infections that occur, and the disease etiologies that lead to them. For AHRQ, this case-based approach represents a different way of thinking. However, other agencies, such as CDC, have the capacity to engage in this type of analysis, and AHRQ has already begun preliminary discussions with CDC about how to work collaboratively to further reduce CLABSIs. AHRQ funding for CLABSI will end in September 2012, and the agency’s role beyond this time remains unclear. Outside of CLABSI, work will continue on CAUTI, VAP, ESRD, and surgical site infections. The work in surgery will be expanded to include common surgical complications, as limiting the focus to infections ignores a significant part of the problem.
Dr. Hines: The core of CUSP relates not to reducing infections, but rather to changing culture and implementing a set of clinical activities that can rectify a frequently occurring harm, be it an infection, surgical complication, or a fall. The goal is to create a standardized change package at the unit level. CUSP represents an organizational framework, not a specific, unique mix of activities. In fact, many organizations already use various elements of CUSP, including root-cause analysis. In some cases, these organizations may use different language to describe these activities. Our goal is not to convince them to adopt our language, but rather to provide a “translation guide” that helps them understand how the terminology and activities that they are doing can be integrated into a comprehensive, unit-based safety program.
At a broader level, CUSP has been and continues to be a catalyst for change. It began with innovative work that uncovered strategies for significantly reducing infections within a single organization. That effort morphed over time to larger-scale initiatives targeted at a particular state, then 10 states, and then the nation as a whole. During this time, other agencies, such as CMS, developed valid measures for reporting infection rates, and began mandating that such data be collected and reported publicly (something that will start next year). Going forward, publicly reported data will become an additional impetus for hospitals to identify and address problem areas, with support from the QIOs. Over time, Medicare, Medicaid, and other insurers may not pay for infections and other safety problems that should not have occurred (as they do with “never events” today). This evolutionary process represents an unscripted, informal trajectory that turns isolated innovations into standards of care over time. AHRQ’s role in this process may end fairly early—as noted, AHRQ funding for CLABSI will end around the time that mandatory public reporting begins. After that time, QIOs will work with hospitals that still have high rates, and further improvements should occur. In essence, the evolution of QI efforts within CLABSI represents a good example of how this process can work effectively over time. Other areas may or may not fit perfectly into this trajectory, but the overall prospects for continued, meaningful improvement appear to be bright, with CUSP playing a critical, catalyzing role in making it happen.
Stephen Hines is the Vice-President for Research at the Health Research and Educational Trust (HRET) and a senior adviser on the CUSP-CLABSI and CUSP: CAUTI projects.
About James Battles, PhD
James Battles is a Senior Service Fellow for Patient Safety at the AHRQ Center for Quality Improvement and Patient Safety (CQulPS).
Disclosure Statement: Dr. Hines and Dr. Battles have not indicated whether they have financial interests of business/professional affiliations relevant to the work described in this article.
Dixon-Woods M, Bosk CL, Aveling EL, et al. Explaining Michigan: Developing an Ex Post Theory of a Quality Improvement Program. Milbank Quarterly. 2011;89(2):167-205.
AHRQ Innovations Exchange. Intensive Care Units Participating in Hospital Collaborative Implement Multiple Improvement Strategies, Leading to Fewer Deaths and Lower Costs. Available at: http://www.innovations.ahrq.gov/content.aspx?id=2668.
AHRQ Innovations Exchange. Unit-Based Safety Program Improves Safety Culture, Reduces Medication Errors and Length of Stay. Available at: http://www.innovations.ahrq.gov/content.aspx?id=1769.
AHRQ Innovations Exchange. Standardized Communication Process Guided by Daily Goals Form Improves Nurse–Physician Communication and Increases Nurse and Patient Satisfaction on Surgical Unit. Available at: http://www.innovations.ahrq.gov/content.aspx?id=2699.
Eliminating CLABSI A National Patient Safety Imperative Second Progress Report on the National On the CUSP: Stop BSI Project. Available at: http://www.onthecuspstophai.org.