A new study underscores the persistence of medical harm based on a review of patient records in 11 Massachusetts hospitals. With the SafeCare Study, researchers revisit the foundational Harvard Medical Practice Study I and II that used health records to identify adverse events in hospital care more than 30 years ago. Although the results are not directly comparable, due to changes in care delivery and technology, the SafeCare study sheds new light on the current state of patient safety.

Three of the study’s authors talked with Patient Safety Beat about their research and how it might inform future safety improvement efforts. David W. Bates, M.D., is Chief of Internal Medicine at Brigham and Women’s Hospital. Elizabeth Mort, M.D., is Senior Vice President of Quality and Safety at Massachusetts General Hospital and the Massachusetts General Physicians Organization. And Patricia Folcarelli, Ph.D., R.N., is Vice President of Patient Safety at CRICO, the professional liability insurer for the Harvard medical community and its affiliated institutions and the sponsor of the SafeCare Study.

What was the goal of the SafeCare study?

David Bates: We wanted to understand how much progress has been made since 1991, when publication of the initial Harvard Medical Practice Study galvanized the patient safety movement. We have not had a similar study since then. National data cover some but not all of the safety issues. In the SafeCare study, we tried to be comprehensive and assess the frequency of all types of adverse events.

Pat Folcarelli: We wanted to get a current overview of patient harm: What are the drivers? How prevalent is it? We’re fortunate to have been able to do an extensive study with such an experienced team, right here in our community. We were also able to be agnostic to the medical records. Our hospitals use many different electronic record systems, including brand name and bespoke systems. It’s unusual to be able to fold all of them together for use in one study.

In the study, you emphasize the need for ongoing evaluation of patient harm and the benefits of electronic review of records in real time. How would that change future work on patient safety?

David Bates: Automated surveillance systems are the way things can and should be done. They will enable us, at a reasonable cost, to have a sense of exactly how much harm occurs across the broad array of the types of harm that we know can happen. Organizations will be able more effectively to manage these systems and the safety of care they are delivering.

Liz Mort: We know that our electronic health records and other systems, including natural language processing and artificial intelligence, can give us a heads up and allow clinicians to jump in and reduce the risk of harm. In my mind, signals are a better investment than measures. We can now begin to leverage electronic information to reduce harm from happening at all.

Pat Folcarelli: Having been a nurse leader, I know we push nurses to keep reporting safety events, but we’re still not learning enough from those types of systems. I would love to see us apply more intelligence to the medical record for surveillance, with great benefits for all clinicians and patients.

Do any of the hospitals in the study currently use automated surveillance systems?

David Bates: Several hospitals in the study do, but not at the level needed to really do a great job with this in the long run. I will say, however, that several hospitals we evaluated have been very forward thinking in this regard.

Liz Mort: In addition to the technological advances, which all have tremendous promise, it’s important to remember that people are still foundational to health care as an industry. Step one is to ensure our workforce feels safe and trusts its leaders. Then we prioritize safety, helping everyone understand their role in making care safe and high quality. 

How did you go about defining the adverse events you were looking for in the study?

David Bates: We used a method called “World Café” to tap the community’s expertise and think about which types of adverse events to look for and how to classify them once we found them.

We brought 80 experts from CRICO institutions together for an all-day event in March 2018. Working in small, clinically focused groups, they reviewed lists of domain-specific triggers and measures we had developed through literature review. Each group reviewed those metrics, looking at their validity for assessing safety and other considerations.

Pat Folcarelli: We had one group each of intensivists, nurses, physicians, infection control practitioners, generalists, nurse leaders and more, all discussing current measures of harm. What’s the value of each one? Is it available electronically? It really helped us make meaning for measuring how we’re doing and developing standardized ways to look at harm.

There is wide variability in the incidence and types of harms reported across peer institutions. Can that be fixed?

David Bates: My hope is that organizations will move toward using a standard set of metrics to assess the frequency of harm. That would help organizations judge, “Is our rate of harm higher than expected or not?” Right now, most hospitals are just comparing themselves with themselves. In a few areas — hospital acquired infections, for example — you can make comparisons across institutions, but for most types of adverse events, that’s not practical today. I think it will be very valuable for organizations to know how they compare and learn where they could do better.

Pat Folcarelli: With so many leading organizations in the state, we have an opportunity to be more standardized. Right now, everybody’s dashboard looks different, how they describe what’s preventable is different, and so on. There’s a lot of non-standard work.

Liz Mort: At the management level, before settling on specific initiatives, I’d like to see organizations begin to prioritize by looking at their strategic plans. In health care, our job is to give every patient the opportunity to have their best possible outcome. Where in your strategic plan is that articulated? Beyond that, what are your goals for the year?

In the early days of the pandemic, our main goals were to take care of the COVID patients and all the other patients the best we could. Now it’s time to restructure and ask ourselves, “What’s goal number one? What’s goal number two? What are we going to do this year?” It’s time to embed the goal of high quality and safe care in the entire management structure. I think that’s the challenge for leaders in 2023 — to get back to that structured approach to management.

You also looked at safety in ambulatory care settings. Can you share any of those results?

David Bates: While we can’t yet talk about specific findings in other settings, I can say that we look forward later this year to sharing what we’ve learned. We thought it was very important to study the outpatient setting because we are pushing lots of care outside the hospital today, and many of those patients are quite sick. That’s making the outpatient environment riskier than it was before.

At the same time, our tools for measuring the frequency of harm in ambulatory settings are much less mature than they are in hospitals. Undoubtedly, there are a lot more things happening than we’ll see in the data. I recall that when we did our first big adverse drug event study in the outpatient setting and asked patients if they had had a problem with one of their drugs, we found eight times as many events as we found recorded in the chart. People were having lots of problems with medications and often just stop taking them without telling the doctor.

How should executive leaders be thinking about patient safety today? What can they learn from your findings to apply to their own practice settings?

David Bates: In an editorial that accompanies our study, Don Berwick notes that patient safety has fallen off the radar screen a bit and needs to be given more attention.

I do think it's time for leadership to refocus on patient safety, which our study shows is still a major issue. COVID has obviously been a huge challenge and consumed much of the mind space for leaders for the last few years. But we're now largely through that. We need to take a hard look at how many people are being harmed by the care that they get and how we can make things work better.

Pat Folcarelli: We also have to think about our workforce. We need to double down at the leadership level and provide an environment where it's safe for people to deliver safe care. It’s time to prioritize both the quality of care and safety of staff members and devote sufficient resources to do both.

Liz Mort: I think the SafeCare study really elevates the importance of the need to work on patient safety. There's a lot going on right now. Everyone talks about the challenges — financial, workforce, burnout and the end of the pandemic — and it's important of course to address these challenges. But we can't do that at the expense of continuing to focus on reducing preventable patient harm. That has to be at the top of the list.

Boards, senior management teams, clinical leaders, nurse leaders — we all need to ask, “What is my role in the organization? How can I personally, with my job description, authority and staff, advance the reduction of preventable patient harm today? And how do I put it into my work plan? How do I hire and train people for that? How do I celebrate their successes? How do I help them learn when they don't succeed? And how can I make it clear to all staff that their performance on this important mission of reducing preventable patient harm matters to the patients. So, for that governance, board-to-bedside management piece, I think we can restructure how we're prioritizing things and help advance the work.