University hospitals exec: ‘Healthcare is riddled with mindless variation’

By Christopher Cheney, HealthLeaders Media

To advance value-based care, health systems and hospitals need to eliminate “defects in achieving value,” a top executive at Cleveland-based University Hospitals says.

Peter Pronovost, MD, PhD, is the chief quality and clinical transformation officer at University Hospitals. The critical care physician has a global reputation as a patient safety champion, including life-saving work developing checklists to reduce central line–associated bloodstream infections. He is a prolific researcher, with more than 800 articles published in peer-reviewed journals.

HealthLeaders interviewed Pronovost recently to discuss a range of topics, including checklists, value-based care, workforce safety, and his vision for telehealth after the coronavirus pandemic has passed. The following is a lightly edited transcript of that conversation.

HealthLeaders: What are the primary elements of a good checklist for clinical care?

Peter Pronovost: The main element of a good checklist is that the checklist helps to ensure that the task that it is intended to occur actually occurs. In other words, the check list is a means to doing something.

There are a few things that guide a good checklist.

One, is that you are crystal clear about what you are trying to accomplish with the checklist. You need to be clear about the behavior you are seeking to change. Two, is that you co-create a checklist with the people who will be using it. In other words, if you impose a checklist on people, it is often wrong and not adopted. Three, is that you ruthlessly try to reduce ambiguity. If you look at one of the main reasons why checklists are not used, it is unclear who is to do what, where, when, and how. Finally, you need to have a mechanism to review and evolve a checklist because, like any tool, the initial draft of a checklist is often wrong. As you use a new checklist with patients, within a short period of time you should have a well-vetted checklist.

HL: What are the primary opportunities to improve the value of clinical care at health systems and hospitals?

Pronovost: Everybody talks about value from a theoretical perspective. Much like checklists, if we cannot get down to the defects in value, we will never be able to achieve it.

I have developed a checklist for value that has three domains: defects in helping people stay well such as not performing cancer screenings, not helping people get well such as poorly managing chronic diseases, and defects in helping people to get better such as mistreating an acute illness. Defects in helping people to get better include making sure that inpatient care is coordinated with primary care, making sure that recommended care is optimal because a significant percentage of every procedure is not needed based on objective criteria, making sure the site of care is optimal at achieving the best outcome at the lowest cost, and eliminating harm such as using checklists to eliminate infections.

At University Hospitals, we have done research that shows defects in achieving value cost the healthcare system $1.4 trillion—it is a third of the annual healthcare spend. But most of these defects are invisible to healthcare providers. At University Hospitals, we have been able to reduce the annual healthcare spend for Medicare patients by 9% in 2018 to 2019 and another 13% in 2019 to 2020. We achieved this by systematically making defects in value visible, then designing systems to eliminate them.

HL: What are the primary opportunities to achieve high reliability in clinical care?

Pronovost: Healthcare is riddled with mindless variation and has too little mindful variation.

Mindless variation is variation that exists either because we have not taken the time to create a standard for high reliability or because someone in a position of power just wants to do something their way. In healthcare, we must make enhanced efforts to reduce mindless variation because it is harmful to patients.

Mindless variation exists in all three domains of defects in value-based care. It exists in preventive care, where we are not making sure people get the right kind of preventive care. It exists in management of chronic disease—there is variation in diagnosis, there is variation in how we treat people, and there is variation in controlling chronic illnesses. And in acute illness there is variation in several areas such as having care in an emergency department that could be provided in a lower cost setting.

A lot of what I just talked about were the technical aspects of high reliability, but there are also tremendous opportunities to achieve high reliability in our teamwork. We do not have highly reliable ways of communicating with each other. For example, most hospitals have people suffering harm because a code will go off and there is an order to call anesthesia, but the code does not say who should make the call. Predictably, five minutes later, somebody says, “Where’s anesthesia?” Nobody took responsibility for calling anesthesia. This is a simple example of low-reliability systems in healthcare that need to change.

HL: What are the main areas of workforce safety at health systems and hospitals?

Pronovost: Right now, with COVID, we are seeing a tremendous uptick in assaults and aggressive behavior among patients. It is quite alarming. It is likely due to the impact of COVID and the stresses on people. The issue of assault—especially in our emergency departments and behavioral health units—is troubling and an opportunity for improvement.

A second issue is burnout and the emotional safety of healthcare workers. The past year has been enormously stressful, and staff have witnessed immense suffering. We need to make sure that we heal our healers.

In addition, needlestick injuries are a big issue along with back injuries.

HL: What is your vision for telehealth after the coronavirus pandemic has passed?

Pronovost: The most exciting piece of telehealth is the idea of combining technologies to enhance value.

For example, in our health system, we are combining several individual technologies that generate enormous value. We are doing what we call triggering, which means if you are a patient and you have gone a while without being seen or not having a required test, we can automatically notify you and your provider to come in. With that outreach, we include smart chatting, so there is a chatbot that can educate the patient about diabetes or the importance of mammograms. The outreach includes self-scheduling. And the outreach includes navigating, so you can speak to a nurse or schedule a telehealth visit.

Telehealth is great, but if the people who most need in-person visits are not coming in, we will have a defect in value. By combining technologies, we are creating what I call a web of well-being, where we can identify people in need of health services and provide them with the services that best meet their needs. A chatbot might meet someone’s needs, or scheduling a primary care visit might meet someone’s needs, or a telehealth visit might meet someone’s needs, or speaking to a nurse on a call line might meet someone’s needs.

HL: How can leadership drive change at health systems and hospitals?

Pronovost: Most importantly, it is by the narratives that we tell. Stories are the most potent source for change because they define how we act. I learned this in my central line–associated bloodstream infection work.

After we saw a 90% reduction in CLABSI across the country, we interviewed clinicians to ask what was different about this CLABSI effort. We wanted to know if it was just the checklist. Our hunch was it was not the checklist—it was something deeper. When we interviewed clinicians, what we found was they all said, “We changed the narrative.” They started the effort thinking that these infections were inevitable, but they got to zero when they thought these infections were preventable and they were capable of doing something about it.

Christopher Cheney is the senior clinical care​ editor at HealthLeaders. This story first ran on HealthLeaders Media.