Creating a high-reliability patient care organization inevitably involves great complexity—and an unending amount of work. That reality was made very clear in the opening keynote presentation, entitled “Safety and Unintended Consequences of Interoperability,” that was presented by Luis Saldaña, M.D., and Mary Beth Mitchell, R.N., of the Arlington, Tex.-based Texas Health Resources, on Thursday morning, at the Health IT Summit in Dallas, sponsored by Healthcare informatics. Dr. Saldaña, the chief medical information officer of the 29-hospital THR, and Mitchell, the chief nursing informatics officer of the system, shared with the audience gathered at the Hilton Anatole Dallas, the challenges and advances involved in their high-reliability care initiative, at the Dallas-Fort Worth-area integrated health system.
“We’ve been on a journey towards high reliability,” Dr. Saldaña told the Health IT Summit audience. “Probably a lot of you may be on this journey. We started this journey a few years ago; it would have helped me to know these things as a practicing emergency physician. Interoperability may seem like a technology problem, but it’s really a people, process, and technology problem.”
Saldaña made several references to the book Managing the Unexpected: Sustained Performance in a Complex World, by Karl E. Weick and Kathleen M. Sutcliffe, first published in 2001 and revised in 2015. As an online description of the 2015 edition of the book states, “Since the first edition of Managing the Unexpected was published in 2001, the unexpected has become a growing part of our everyday lives. The unexpected is often dramatic, as with hurricanes or terrorist attacks. But the unexpected can also come in more subtle forms, such as a small organizational lapse that leads to a major blunder, or an unexamined assumption that costs lives in a crisis. Why are some organizations better able than others to maintain function and structure in the face of unanticipated change? Authors Karl Weick and Kathleen Sutcliffe answer this question by pointing to high reliability organizations (HROs), such as emergency rooms in hospitals, flight operations of aircraft carriers, and firefighting units as models to follow. These organizations have developed ways of acting and styles of learning that enable them to manage the unexpected better than other organizations.”
Mary Beth Mitchell, M.D. and Luis Saldaña, M.D.
In addition, an organization called High Reliability Organizing, states on its website that, “In the world today we deal with uncertainty and threat. This can be on the personal level of family or job, on the business level of the economy, or in geopolitics. High Reliability, evolved for optimal performance in an environment of uncertainty and threat, can strengthen a person’s performance, improve the function of a team, and move an organization forward through uncertainty…. The High Reliability Organization is commonly described as an organization that performs high risk work but without rare, catastrophic events.”
In that context, Dr. Saldaña told the audience at the Health IT Summit, “Part of becoming a high-reliability organization is be preoccupied with failure. Years ago, while traveling,” he said, “I met an engineer from Exxon. He was retired, but told me that he continued to carry the same habits of the culture of safety with him, in his personal life, in retirement. He used checklists; he wore compression socks when traveling, to prevent VTE”—venous thromboembolism.
A key point about the Managing the Unexpected book, Saldaña said, was this—a quote from the book: “High-reliability organizations operate under very trying conditions all the time and yet manage to have fewer than their fair share of accidents.”
And, per the book, Saldaña noted that “Risk is a function of probability and consequence. By decreasing the probability of an accident, high-reliability organizations operation to make systems ultra-safe. In that context, he referenced the following principles that patient care leaders need to follow when developing high-reliability initiatives. They need to: “be preoccupied with failure; be reluctant to simplify; be sensitive to operations; commit to resilience; and defer to the experts.” The key in that, he said, is to find a balance between the meeting the need for reliability and “managing the cadence of where the system is moving operationally.”
Saldaña also referenced Nassim Nicholas Taleb, for his concept of “anti-fragility,” as articulated in his 2012 book, Antifragile: Things That Gain From Disorder. “In some cases,” he emphasized, “challenges and failures can actually make you stronger.”
And in all that, Saldaña said, communication is one of the most important critical success factors in an high-reliability initiative. It speaks to the need to “pay attention to detail, communicate clearly, speak up for safety, and partner for accountability,” in order to succeed in this important work. Also, staging has been important: “We started with foundational care”—care delivered to all patients—“moved on to situational care—trigger-driven ‘packages’ of care—and then to condition-specific care,” in working out where to apply high-reliability strategies within the Texas Health Resources health system. What’s more, he said, “Our transformation office is partnering with us, developing workflows and blueprints for us to execute on.”
Further, Saldaña noted, “A big part of success with this is around the sustainability issue. And a big part of that is around how you set up your governance structure, develop KPI alignment, and develop performance improvement.” In all that, he said, clinical informatics is heavily implicated. “Physician orders are important. Alerting is important—alerts and prompts. Documentation is important. As we do analytics, a lot of times, we’re still dependent on discrete documentation points. So the analytics point will be important as well.”
Saldaña, Mitchel, and their colleagues have consistently followed six design steps in the design process in applying the high-reliability principles to care delivery areas within the organization. They make sure to define the goals and outcomes for each module; define the clinical specifications for every patient to achieve desired outcomes; identify the enablers required to reliably deliver the specifications; designate people, process, and technology in an integrated workflow; finalize the functional requirements for creating the required enablers; and identify the process measures needed to ascertain levels of success.
Mitchell told the audience that the lived experience of putting any high-reliability principles in place in patient care settings is inevitably complex and extremely challenging. What has helped everyone move forward steadily, she said, has been a focus on the benefits, including the benefits of electronic health record (EHR)-medical device integration for accuracy of data entry, timeliness of data, ease of clinician use, and improved patient outcomes.
At the level of bedside care, the complexity is inevitable, Mitchell said. For example, she noted, “We now require nurses to take vital signs every four hours, and in critical care areas, every hour,” in order to ensure a higher level of reliability of care delivery quality. “And how do you do that without a lot of interoperability?” she asked.
Still, when it comes to executing high-reliability organization principles in patient care, in practice, and integrated into workflow, things immediately become challenging. “We’re having to put a lot of effort into managing interoperability around device integration, she said.
And this work has been part of a longer, broader journey. “We began EHR integration in 2006, and it took us six years to get 14 hospitals implemented,” Mitchell told the audience. “In 2009, we implemented barcoded meds administration. And we have about a 97-98 percent rate of barcoded meds administration now. Then, in 2015, we accomplished the device integration of our IV pumps. In that context, interoperability means that the physician order is automatically brought to the pump’s function, through the physician order. The system verifies that physician order, and the nurse only needs to do a verification at the bedside. So the nurse scans the patient, scans the medication, scans the pump, and hits go. That’s in theory, at least.”
The challenge is that there remain many patient care areas that are still “out of scope” of the integration, for example when a patient is transferred from a PACU (post-anesthesia care unit), the infusion involves an out-of-scope drug, or the patient is moved to a recovery room. And, she said, “When a patient who is on multiple IVs, moves into the ICU, the nurse is now having to manage a variety of pumps—to manage pain, manage antibiotics, manage the ventilator, though we do have a bit of ventilator integration. But it is complex for the nurse to manage at the bedside, and sometimes the barcode is on the same side of the bed as the medications, but on the wrong side from the nurse.” So the lived reality of this remains complicated for nurses, in practice. “So,” she said, “the scan the patient, scan the meds, scan the nurse’ theory ends up inevitably being complex in practice.”
Still, all the effort is worth it, Mitchell said, as the organization is decreasing variation, increasing standardization, and increasing reliability.
Things to watch for? Mitchell cited several. “Changes in communication patterns and practices—providers are monitoring the EHR for remote locations, and acting on validated data; but is that data valid for that patient?” Another: “New kinds of errors end up being introduced,” because of the complexity of processes. And there can be an over-reliance on technology.
More significantly, she said, “We’ve ended up increasing the amount of work that nurses are compelled to do. We’ve added to their stress and burden, and haven’t yet seen initial time savings. But I think it’s worth the effort, and our nurses think it’s worth the effort.”
In the end, Mitchell said, leveraging technology to improve care delivery processes is “part of nurses’ clinical picture now. Remember when nurses used to say, ‘I don’t do computers’?? Well, you wouldn’t say, ‘I don’t do IV infusion,’ or ‘I don’t do ventilators,’ would you? The reality is that this is part of their workflow now.”