Opening the HIT Summit in San Diego, UCSD’s Chris Longhurst Examines the Potential for EHR-Fueled Clinical Transformation

In a wide-ranging opening keynote address, Christopher A. Longhurst, M.D., CIO at UC San Diego Health, offered his perspectives on the healthcare IT-facilitated clinical transformation beginning to take place now in earnest across U.S. patient care organizations, on the first day of the Health IT Summit in San Diego, sponsored by Healthcare Informatics. Dr. Longhurst, a pediatrician who had been CMIO at Stanford Children’s Health before coming to UC San Diego Health as CIO a little over a year ago, told attendees gathered at the Omni Hotel San Diego on Tuesday morning, that now is absolutely the time for healthcare IT leaders to help their patient care organizations turn the corner from the immediate post-electronic health record (EHR) implementation phase, to the vital work of improving patient care quality and outcomes through EHR implementation and clinical transformation.

Longhurst began his keynote address by framing the context of the current moment in clinical informatics in U.S. healthcare. He reviewed what happened when a December 2005 article, “Unexpected Increased Mortality after Implementation of a commercially Sold Computerized Physician Order Entry System,” appeared in the journal Pediatrics, summarizing a documented increase in mortality at Children’s Hospital of Pittsburgh, after EHR implementation there. As Longhurst noted of the article’s core finding, “It was a bombshell result. We gave a lot of thought to this,” he said, “and I wrote a letter to Pediatrics, along with Dr. Classen”—David Classen, M.D. of the University of Utah Health Science Center. Looking at the immediate result of increased mortality at Children’s Hospital of Pittsburgh following EHR go-live, he said, speaking of the analysis of it that he and Dr. Classen undertook, “We thought it was inadequate preparation along with bad order sets.”

Christopher Longhurst, M.D.

As a result of the heightened awareness of the potential for unanticipated negative impacts on patient outcomes in the immediate aftermath of EHR go-live, Longhurst talked about how he and his colleagues at Stanford Children’s Health (at that time still known as Lucile Packard Children’s Hospital) engaged in very careful, thoughtful analysis of outcomes as they were going live with their EHR implementation, and in its immediate aftermath. Indeed, as CMIO, he had asked the hospital’s chief quality officer to monitor very closely all mortality trends in their facility, as they moved forward through post-go-live. As it turned out, their implementation revealed a dramatic decrease in patient mortality: mean monthly adjusted mortality rate for the hospital overall decreased by 20 percent in the 18 months after EHR go-live, with an estimated 36 children’s lives saved during that time, based on an analysis of their outcomes. In an article that he and his colleagues from Lucile Packard published in the May 2010 issue of Pediatrics, entitled “Decrease in Hospital-wide Mortality Rate After Implementation of a Commercially Sold Computerized Physician Order Entry System,” Longhurst and his co-authors attributed the difference between the outcomes at their organization and those years earlier at Children’s of Pittsburgh, to differing implementation strategies, and offered that he and his colleagues had learned a great deal from what their peers at Children’s of Pittsburgh had discovered in their post-implementation; and that a well-executed EHR implementation is indeed fully capable of reducing mortality rates among pediatric patients.

Longhurst also referenced an April 2010 article by Jane Metzger et al in Health Affairs, “Mixed Results In The Safety Performance Of Computerized Physician Order Entry,” which found that, among hospitals implementing six different commercial EHR products, patient safety impacts from EHR implementations varied across every vendor solution as implemented by every hospital, to make the point that the relative success of EHR-fueled patient safety work will depend on what goes on in every patient care organization that attempts such work, not on the commercial EHR product used. In the Metzger study, he noted, “Six different vendors were the safest of six different vendors. So what does that tell us? It’s not the software that you buy. It’s about us. It’s about implementation. Less than 20 percent of outcome was based on the vendor, 80 percent was the implementation,” he said of that study.

Moving into a new era in clinical transformation, Longhurst emphasized, will require some enterprising ingenuity, and above all, leadership and a passion for leveraging health IT effectively to improve clinical outcomes. He spent some time recounting a clinical case at Stanford Children’s Health that broke new ground in that regard, and which was recounted in an article entitled “Evidence-Based Medicine in the Clinical Era,” in the Nov. 2, 2011 edition of New England Journal of Medicine, with Jennifer Frankovich, MD., the lead author, and Dr. Longhurst and Scott M. Sutherland, M.D., as coauthors.

As the coauthors wrote in the NEJM article, “We recently… admitted to our service a 13-year-old girl with systemic lupus erythematosus (SLE). Our patient’s presentation was complicated by nephrotic-range proteinuria, antiphospholipid antibodies, and pancreatitis. Although anticoagulation is not standard   practice   for   children with SLE even when they’re critically ill, these additional factors put our patient at potential risk for thrombosis, and we considered anticoagulation. However, we were unable to find studies pertaining to anticoagulation in our patient’s situation and were therefore   reluctant   to   pursue   that course, given the risk of bleeding. A survey of our pediatric rheumatology colleagues—a review of our collective Level V evidence, so to speak—was equally fruitless and failed to produce a consensus. Without clear evidence to guide us and needing to make a decision swiftly, we turned to a new approach, using the data captured in our institution’s electronic medical record (EMR) and an innovative research data warehouse. The platform, called the Stanford Translational Research Integrated Database Environment (STRIDE), acquires and stores all patient data contained in the EMR at our hospital and provides immediate advanced text searching capability. Through STRIDE, we could rapidly review data on an SLE cohort that included pediatric patients with SLE cared for by clinicians in our division between October 2004 and July 2009. This “electronic cohort” was originally created for use in studying complications associated with pediatric SLE and exists under a protocol approved by our institutional review board.” Using that data set, Frankovich, Longhurst, and their colleagues ultimately chose the path of anticoagulation for that patient.

As Longhurst explained it to his audience on Tuesday, “In the fall of 2009, I was on service, and there was a patient with lupus. She was well-known to us, and had been served many times before. She was flown by LifeFlight, had multiple systems failure. I was standing at our bedside with our intensivist and one of our pediatric rheumatologists [Frankovich]. And the rheumatologist said, she’s losing a lot of protein in her urine, might she be at higher risk for clotting? So she looked it up in the literature. And how many papers have been published about teenagers with lupus and clotting issues? Zero. So she went to our subspecialists. And the first one said, I would absolutely anticoagulate. The second said, I would absolutely not. Both had had cases like this.” What Dr. Frankovich was able to do was to use information from the STRIDE data warehouse, to research past cases, including identified cases, of lupus, with complications. She and her fellow physicians found 98 lupus cases from the previous five years, and divided them into various cohorts based on different clinical elements. Ultimately, based on finding a significantly higher risk of clotting among patients with lupus whose cases most closely approximated this one, they pursued anticoagulation. “What we know,” Longhurst told his audience, “is that in the absence of published data, we made the based and most evidence-based decision we could under the circumstances. And if that were my daughter in the hospital, I would hope that she would get the same care.”

Longhurst shared with his audience that the key point of all of these articles and analyses is this: the potential for EHR-facilitated clinical transformation is huge—but healthcare IT leaders must help lead clinical transformation efforts strategically and thoughtfully, while also sorting very carefully through all the intended and unintended consequences that might emerge. “I think that electronic health record implementation is exactly the right thing to do; it improves the care we provide,” he said. “But we all know that it leads to unintended consequences. And we’re not yet really using the aggregate data in the system to improve care.”

Among the points he made included references to recommendations from the Institute of Medicine that urge healthcare leaders to address both the anticipated and unanticipated consequences of healthcare IT implementation, and also urge healthcare IT leaders to architect clinical decision support alerts that are clinically helpful. Longhurst referred to IOM’s having noted that most CDS alerts are around excess utilization.

In the end, Longhurst told his audience, “Every quality improvement project involves some IT component. But we [healthcare IT leaders] need to be leaders. We need to lead change. We need for IT implementation to support the overall strategy of operating better as a system, and we need to be thoughtful about it.”