In His Words: Pronovost on CUSP

Peter Pronovost, MD, Ph.D., senior vice president for patient safety and quality at Johns Hopkins Medicine, was the lead on the Comprehensive Unit-based Safety Program (CUSP) initiative. CUSP reduced central line–associated bloodstream infections (CLABSIs) in intensive care units by 40 percent, according to September 2012 preliminary findings from the Agency for Health Care and Research Quality (AHRQ). Wing of Zock Associate Editor Sarah Sonies sat down with Dr. Pronovost to learn more about the initiative.   

CUSP was first used in Michigan, under the leadership of the Michigan Health & Hospital Association. What then brought about the idea to utilize CUSP at Johns Hopkins? How did you and your team identify the need to use a toolkit such as CUSP in the hope of reducing CLABSIs?

After a little girl named Josie King died from a central line infection at Johns Hopkins, we were inspired to implement a unit-based program to reduce infections. CUSP grew out of this need for an infrastructure at that unit level, to help learn from local mistakes and improve local teamwork and culture.

Different types of safety problems require different methods and approaches. We looked at virtually every quality measure there was, from teamwork to safety culture and hand-washing. We noticed enormous variation among units in the hospital. There are what we call “common problems,” like bloodstream infections — things for which there is empiric evidence on prevention and metrics. We realized that there are also “local problems” — things that might be unique to the unit. The unit is the microsystem we wanted to tap into, because the units are the heart of where the work gets done at a hospital. When the people in the unit are connected and have pride in their unit, there is an organizational cohesiveness. We really tried to leverage that element of pride.

How long did you work on CUSP before implementation?

CUSP is iterative; we are constantly learning things. One of the real gifts for our team is that we are not wed to any model, so there is lots of room for developmental flexibility. The program began with eight steps, then went down to six; now it’s down to five. It evolved because its purpose is to serve the needs of patients and clinicians, so we need to have lots of room for mobility. It’s certainly been highly effective, but we are continuously working on improving it.

How did you partner with the Agency for Healthcare Research and Quality (ARHQ) and the American Hospital Association (AHA) on this initiative?

After implementing CUSP in Michigan, we were given some philanthropy money and ARHQ money to spread it across the country. ARHQ funded our implementation of CUSP in order for us to implement it in hospitals nationwide. Additionally, we partnered with AHA as the national coordinator, then we partnered with each individual state hospital association, which recruited teams from a hospital in each state.

There is currently a big emphasis placed by those in the clinical community, such as yourself and Atul Gawande, on systematic treatment systems to reduce the rate of error and infections in patients during hospitalization. Based on the success CUSP has seen, what are other ways these types of initiatives could work well at various academic medical centers?

One thing we have found: If you do just the evidence-based practice piece, it’s in many ways still just externally driven; it works as long as you push. On the other hand, if you simply stick to local quality work with culture, the risk is that it becomes a little too feelings-driven. Clinicians say, “Show me the money. Where are the hard outcomes?”

Combining evidence-based practices and local culture change together is key. We need department-level and hospital-level infrastructure. The flexibility of initiatives like CUSP is great for other areas because they are constantly in flux. You can change the common problems you are working on. Looking at the major causes of preventable deaths: healthcare-acquired infections and DVTs, which kill about 1,000 people a year each; blood clots and so forth — these are great places to use CUSP teams as the enduring structure.

What CUSP does is encourage co-creation development. Programs like CUSP are based on the separate resources of academic medical centers, which are all unique. Academic medical centers should be the vertical cores of these initiatives, with the technical skills to do this work. These safety efforts are going to have to be led by clinicians. We should be driving the science of health delivery with these initiatives and then sharing it with the world.

You work with the Armstrong Institute for Patient Safety and Quality, which is dedicated to eliminating preventable harm to patients and to achieving the best patient outcomes at the lowest cost possible. How can clinical faculty incorporate programs like CUSP not only to improve care during hospitalization, but in ambulatory and primary care as well, to provide the best possible care at more efficient costs?

We are putting CUSP teams in our primary care regimens, creating a quality management infrastructure. There is a quality infrastructure at our health system level, then each hospital has a quality infrastructure. Each higher level holds the lower levels accountable. At each level, they need skills, resources, and accountability. Until now, what we’ve done to reduce health care costs has not been improving productivity, but just paying doctors and hospitals less. There is less money to fund this quality management infrastructure. The risk is that our health reform may very well be taking patient safety and quality in the wrong direction.

By utilizing CUSP and, in your words, “putting trust in the frontline of your physicians,” you and your team were able to institute a major culture change in reducing infections. How do you think clinical faculty and faculty at medical schools can work together to impart this type of collaborative, ground-up knowledge to students so they are better prepared for the workforce?

The culture change that is CUSP did not happen overnight at Hopkins. Our academic medical faculty must view health care delivery as a science every bit as important as basic and clinical research. Previously in many hospitals, including my own, care delivery was something nurses and administrators did, but faculty did not see it as official. To make it as a legitimate science, these academic medical centers are likely going to have to build capacity of these scientists. The leaders have to drive the message home that this is a legitimate and important science.

Health care is the only industry where frontline operators rather than safety experts investigate the harms. You wouldn’t think of a pilot or flight attendant investigating a plane crash. We need more physicians trained with advanced degrees in safety and health services research. We have to make sure the faculty has the skill set for quality and patient safety training so they in turn can teach it to students.

We have to empower our frontline doctors and nurses to trust that we really believe in them. For many doctors, we feel like we have absolutely no authority or power to change the system. We really believe that it is frontline clinicians who have to own this. That is the only way CUSP and similar initiatives will work.

What are the top three things you would recommend to those who want to implement CUSP within their organizations?

  1. Get commitment from senior leaders to build that CUSP infrastructure. It requires executive, departmental, and unit buy-in. Ideally, the program would begin with a couple of units, but there certainly needs to be the infrastructure to do that.
  2. Pick some common causes of harm to work on with the CUSP team that can be targeted directly, such as infections, teamwork failures, or blood clots.
  3. Make it your own. CUSP is really just a shell. We put a lot of social theory into it and made it suitable for customization. I would adhere to the principles, but how that all plays out is up to the institution. We encourage others to make it better and to share back with us what they have learned.On a final note, academic medical centers should be leading these health care delivery initiatives. For too long it’s been The Joint Commission, the Center for Medicare and Medicaid Services (CMS), or regulators driving the bus for us. With CUSP, we have shown what is possible for AMCs across the country to do; I hope academic medicine stands up and works together to help push us forward.