IHI Vice President Carol Haraden, PhD, is executive lead for the IHI Patient Safety Executive Development Program. In this interview with IHI Communications Specialist Jo Ann Endo, Dr. Haraden talks about changes in patient safety in the last decade and what organizations of every size need to do to improve patient safety.
Q: What are the most significant changes you have seen in patient safety in the last 10 years or so?
A: Technology has been a strong enabler of some important changes in patient safety. It’s allowed better communication, particularly across sites. It has facilitated better documentation, added some reminders, and put in place a number of important forcing functions.
Technology has also added a lot of complexity. People now spend inordinate amounts of time taking care of the computer system instead of the patient. There are also limits to what one can write in some of the fields [in the electronic medical record], so it’s hard to record some important impressions of the patient if they don’t match the predetermined boxes you have to check.
Health care is still unsophisticated in many ways, particularly around the complexity involved in diagnosing a patient and the complexity of patients themselves. A larger number of patients are getting older and have more complex diseases and more co-morbidities. Every time there is an added disease or complication, there is an added specialist involved in the patient’s care. The combination of more doctors, more nurses, and more pharmacists and others caring for a single patient means the handovers are complex and difficult, and we don’t manage them well. It is one of the chief reasons for unintended harm to patients when we look at safety events that have occurred.
Another change in the past three years or so is the decision by the Centers for Medicare & Medicaid Services (CMS) to not pay for conditions like central line infections or ventilator-acquired pneumonia for Medicare patients. The attention this change has brought to these issues is great, but there is also a downside. Health care organizations in the US are working to reduce “defects” they don’t get paid for, which has narrowed people’s focus. In some ways, narrowing your focus can be a good thing, but in some places it is too narrow. The non-payment list is the safety agenda in a number of organizations. While that’s understandable — improving safety for these specific conditions is important to patients and the organization — it is also worrisome that the conditions on the list seem to have become almost the sole focus of safety efforts in some organizations.
In the past 10 years, so many issues in health care are competing for our attention. Patient safety is certainly one of them, particularly with the increased focus on issues related to payment reform, staffing, and working across the continuum of care requirements for accountable care organizations. But it makes it more challenging to give safety the necessary attention given the competition for our focus.
Watch Carol discuss patient safety issues on YouTube
Q: What are the keys to improving patient safety?
A: One key is to understand that most everything in health care is linked to patient safety. In our minds, many of us have patient safety segmented out somewhere in this hinterland, but we need to think of patient safety as an integrated phenomenon.
Payment reform is definitely linked to patient safety and that’s not just related to CMS non-payment issues. When you consider the cost of defects, we are paying every day for waste and that’s money that we really need for more important things.
Care across the continuum is related to patient safety because we have very convoluted and difficult systems to ensure a patient is safely transitioned from one setting to the next. When something bad happens, everyone suffers, most particularly the patient and their family.
Flow and waiting times have a lot to do with patient safety. For example, if early sepsis isn’t recognized in the emergency department and a patient is left to wait without appropriate interventions, they can become critically ill quickly and we move from management to survival. When patients are placed in off-service beds because of flow failures [e.g., a patient with a cardiac issue is placed on an orthopedic unit], the ability to safely deliver care can be undermined.
We generally parse out these issues — we work on waiting times and flow here, and on payment reform here, or on frail elderly here — and every one of these issues has big patient safety considerations. It is an important promise we make to patients that we won’t harm them when they come into our organization. We have to think about patient safety in everything that we do.
Q: What has been the most profound change in your thinking about patient safety in the last 10 years or so?
A: Moving from defect reduction to a more proactive prevention approach, and thinking about patient safety as preventing the patient from incurring any number of risks. When a patient comes into the health care system, there are risks from the environment and risks from the various interventions used to treat the patient. So, how do we think about the environment we are in as inherently risky and focus on prevention? Patient safety is mostly about defect reduction, which includes prevention as well, to some degree. However, we need to move way upstream to ask, “What are the conditions that are creating the likelihood of these defects occurring?” For example, it’s important to locate and identify those patients who are ill with sepsis and treat them. However, we should spend more time preventing sepsis. That is the biggest change in my own thinking.
Q: What are the most promising approaches IHI has used to help organizations, systems, and countries improve safety?
A: It really doesn’t matter whether you are talking about improving patient safety in a small hospital, a large academic center, or even an entire country. To change from a defect reduction system to an identification and prevention system, you have to do four key things:
- Create a learning system
- Use diagnostic tools and data for understanding
- Build capacity and capability to improve
- Develop a leadership system.
1. Create a Learning System
There is so much being learned every second in each hospital in the world. What’s learned, however, may only be learned by one person or maybe the few people around them who are involved in a safety event. I’ll often ask folks when an adverse event occurs, “Do you do root cause analysis?” They’ll say, “Yes, everyone does root cause analysis.” Then I’ll ask, “How much time do you spend on it?” It is really quite remarkable given that everyone involved in the event has to be part of the root cause analysis (RCA). For example, for a surgical event, the surgeon, the anesthetist, and the operating room nurse have to be engaged in the RCA to fully learn about what happened. When I then ask, “How do you make sure that every single person who this could happen to know what happened and how to prevent it?”, almost no one has a system in place to guarantee that reliably happens. We fix it for that particular person or family and the immediate surgical team involved learns, which is very important, but what about preventing such errors from happening to anyone else?
2. Use Diagnostic Tools and Data for Understanding
Using diagnostic tools to identify patient safety vulnerabilities is all part of having a learning system. What are our most important safety issues and where do they occur? How are they different from location to location? It is hard to fix something you don’t deeply understand.
Bring data to the forefront so that everybody in the organization understands the depth of the safety issues. Lucian Leape refers to “the tyranny of small numbers” — that is, any given clinician will see very few catastrophic safety events firsthand in their lifetime. However, looking at an entire health care system with 2,000 doctors, for example — if each doctor has one preventable adverse event in a year, that means the organization has 2,000 events in one year. In other words, from the individual clinician’s point of view, the numbers don’t seem that significant. As a system, however, these numbers add up and are enormously important to understand and manage.
Being more transparent with data and about safety issues is so important because we have to build a shared understanding of the scope of the problem, and we have to build the will of clinicians and leaders to improve the processes and systems that lead to such issues. Clinicians and those who see the larger system implications of safety issues need to have that conversation. In many organizations — especially without data to help tell the story — you might have one group saying, “I don’t know why we are working on that, I hardly ever see that problem,” and another group saying, “Yes, but that problem is important.” Clinicians are very busy. Why would they take on something if they don’t recognize the severity of the problem?
3. Build Capacity and Capability to Improve
Even if you know what your key safety issues are and you have a great learning system, you still need the capacity to improve. It does little good to have a great diagnosis with no ability to fix the problem. Capability can be understood as “I know what to do” and capacity means “I have the time and tools to do it.” IHI uses the Model for Improvement. Some organizations use Lean principles and others use a combination of various models available to them to guide their improvement work. No matter what methodology you use, you have to address issues around capability and capacity so people can say, “When I see a problem, I know how to fix it.”
4. Develop a Leadership System
Leaders need to understand and own the success of all important endeavors in their organization, and patient safety has to be among those endeavors. Leaders can delegate patient safety to the patient safety officer or to somebody who has “safety” in their title. I don’t think they do that because they don’t care; it’s more the case that leaders think these individuals are the most knowledgeable about safety, so they should lead the work. However, you cannot delegate patient safety if you’re the leader. Leaders need to ask themselves, “What system do I have in place to deeply understand what is going on in my organization with regard to patient safety?” I am sometimes surprised at how little leaders truly understand the depth of safety issues in their organizations.
Q: What is your best advice to people who feel overwhelmed by the various improvement initiatives and requirements vying for their time and attention?
A: There are probably more requirements and demands now than there have ever been. My advice is to write everything that is required up on the wall and, once it’s all up there, put it to one side. Then build a system for prioritizing your work.
For example, start by specifying the aim: “Our aim is to improve patient safety by ‘x’ amount by ‘x’ date within ‘x’ hospital (or it could be an entire health care system).” What are the safety issues that need to be addressed? Where are those issues occurring (e.g., a medical unit, a pharmacy, an operating room)?
After you identify what and where you think the issues are occurring, then go back to that original list of all the requirements to look at the big picture: “We have to be working on pressure ulcers, so we have to be working on them here, here, and here.” Then ask, “Can we put some of the work together so we can, for example, respond to a requirement from CMS and one from the Joint Commission in the same way?”
This approach is not the same as trying to organize work according to the mishmash of various requirements. That will make you crazy. You have to assess, “What are our vulnerabilities? What are our issues?” Lay out those issues and develop the system to address them — look at the unique contributions of a medical unit, an outpatient ward, or day surgery, for example, and think about how they impact patient safety as a whole. Then, go back and look at those issues and ask, “What might we have missed? What work are we doing that could satisfy those requirements?” It’s not the other way around.