Author: Michael Mozdy
Next time you're a patient at our Orthopaedic Center, an Outpatient Services Specialist will likely hand you an iPad at check-in and ask you to complete a quick assessment of how you're feeling. iPads are fun and easy, and an "assessment" sure beats a "survey" or, worse yet, a ton of paper to fill out, so you take it and sit down. Before long, you've told the little device exactly how you're getting along, how active and functional you are, and even how this compares to your expectations for yourself. It's just the type of thing you come to the doctor to discuss.
What comes next is even better: in the exam room, your doctor pulls up the results of your assessment right there on the electronic medical record. She discusses your assessment and any needed changes to your treatment plan, like physical therapy or pain management. Then, she prints for you a Physical Assessment Report where you are assigned a score on a scale of 0-100, and you see how you compare to other patients your age.
If you haven't noticed, this is the new frontier of patient-centered medicine. It's called patient-reported outcomes (PROs), and it's being implemented at the University Orthopaedic Center.
From the patient perspective, it feels like medicine should feel: you are an active participant, and the care team is focused on how you feel and your goals for yourself. But in reality it's a far cry from our traditional focus on medical outcomes like how an incision is healing and whether or not patients develop secondary infections. Not that those aren't important, but they aren't necessarily focused on a patient's expectations and functional abilities.
Ten Years in the Making
Physicians have asked patients these questions for centuries, although with modern medicine, patient conversations and perspectives often do not get into the medical record. Capturing this information is spotty, and there has been no data bank of either standard questions or historical answers.
In September 2004, the National Institutes of Health initiated a cooperative group, referred to as the Patient Reported Outcomes Measurement Information System (PROMIS) at six primary research sites. By 2009, the PROMIS program yielded several sets of statistically validated questions for adult and pediatric populations that can assess their functional abilities. Since then, NIH researchers have been developing web-based tools or "instruments" for physicians and hospitals to use. These assessment instruments are grouped in over 100 domains like "pain - behavioral," "emotional support," and "physical function."
Meanwhile, by 2009 the Chair of Orthopedics, Charles Saltzman, MD, was ready to get serious about measuring functional outcomes. "The goal of Orthopedics is to improve function," says Saltzman. "If a patient's function is not improved, even if we are nice and provide prompt, quality care, we have failed in our primary mission as orthopaedists."
Saltzman hired Man Hung, PhD, MEd, MStat, a statistician and psychometrician specializing in developing these types of tools. In addition to being a part of the U of U Study Design and Biostatistics Center, Dr. Hung is probably one of few people who actually enjoys finding needles in haystacks and making sense out of the most complex sets of data.
In 2010, Hung and Saltzman published the first of many scholarly papers describing their efforts. It received a lot of interest in the orthopaedics department, but most people were unwilling to try the tool until a solid boty of evidence was presented. By late 2010, they had developed a PROMIS-compliant, web-based computerized adaptive testing (CAT) questionnaire, complete with a complex logic that presents patients with different questions based on their previous responses. Patients might only have to answer three questions, or they might have to answer 10, but the process generally takes less than three minutes for everyone.
The next step in the process was to store this data in our data warehouse, which Saltzman accomplished in 2012 by engaging the IT experts from the original NIH-funded projects with our own IT experts in the data warehouse as well as faculty and staff in Orthopaedics. After this was accomplished, we were poised to make the data more available to more applications, including the electronic medical record.
Rolling Out PROs throughout the Orthopaedic Center
By late 2012, Hung knew that despite a number of well-received papers being published and some interest being drummed up in the department, true uptake and use of patient-reported outcomes would not happen without something changing. Some physicians at UOC had been collecting functional data for a decade by this point, but it was collecting dust on the shelf and was not being used in any meaningful way. She conceived a plan that she unveiled to Saltzman: consolidate all measures to one tool, and by standardizing the instrument across the orthopedic center, provide a richer pool of comparative data to be analyzed.
Saltzman wholeheartedly supported this plan and moved quickly to make it a reality.
In early 2013, Saltzman received funding from the University of Utah Medical Group (UUMG) to proceed with a pilot project at the orthopedic center. He hired Rebecca Johnson to manage the rollout of this new tool to all of the physicians and areas at University Orthopaedic Center. Johnson put together a plan by which each specialty was engaged to understand the effort, develop new questions or retrofit their legacy assessment tools to the new PROMIS system, complete the programming to create a new, custom set of questions for their patients, and to change the processes of each clinic so that all patients would not only receive the assessment tool but talk about it with their physician during the visit. Six months of persuasive coordinating work later, Johnson succeeded in implementing the system UOC-wide.
"Dr. Saltzman is a visionary," says Johnson. "This significantly improves the way we provide care and opens a future of possibilities now that we are collecting this data."
Data Collection Feeding Research
The Institutional Review Board (IRB) first approved the collection of this patient data as part of the standard course of patient care. Subsequently, it was approved for secondary analysis by researchers. To Dr. Hung, however, these two pieces are inextricably linked. "We should only collect data that is beneficial to both standard of care and research," she says. A pragmatist at heart, Hung's outspoken philosophy is: "If you're not using it now or don't have a clear plan to use it later, don't collect it. It's a waste of patient time and resources."
Hung and Saltzman's long-term goals are to answer questions about the effectiveness of medical and surgical interventions, and the differences between the techniques that are used in helping patients live better lives. But to answer these questions, Hung says "the instrument must be sensitive and precise." So, they are working hard to ensure that the patient-reported outcomes tool is as finely developed as possible before combining this data with other metrics like Press Ganey patient satisfaction scores or quality outcomes measures.
Saltzman, in particular, is excited about how this data might transform the perspective within orthopaedics: "This will give us for the first time the ability to rigorously analyze our data on a large scale. Our field has been fairly narrow minded in that surgeons think about the outcomes of the patients who come back to their own clinics over a fairly short amount of time. Now, we'll be able to see how our patients are doing compared to a much wider and deeper set of data--and, most importantly, from the patient's perspective."
The Future: "Nice" to Have or "Need" to Have?
Surprisingly, payers are already asking some services like orthopedic joint specialists to provide patient-reported outcomes in order to receive reimbursement. According to our Associate Vice President for Clinical Affairs, Sean Mulvihill, MD, "Patient-reported outcomes are complementary to traditional medical outcomes and our patient satisfaction data. We see great potential in improving our ability to select the right treatment for the right patient at the right time by better understanding PRO. I'm excited by the opportunity for the University of Utah to maximize our value equation by incorporating PRO: improved medical quality, improved patient experience, improved cost effectiveness, and improved functional outcomes."
Saltzman and Hung are presenting their work nationally. Not a week goes by that Johnson doesn't get several emails from colleagues around the nation asking how we've done it. In fact, even the Epic corporation has visited with the team to get a better understanding of how they are administering the assessment tool and incorporating the data in the EMR. Clearly, PROs are the new frontier, and the University Orthopaedic Center is on the leading edge.
The Huntsman Cancer Institute and the University Neuropsychiatric Institute are next in line to incorporate the assessment tool across the physician specialists practicing at those facilities.
But Dr. Hung points out that future investments must be made on the analytics side of the equation. "Ideally, an equal amount of money should be devoted to analytics as IT. Right now, we're at about 95% IT and less than 5% analytics." She points out that the mountain of data they're collecting can't effectively be understood and analyzed by her team alone. A greater investment in data analysts and statisticians will be needed for the University and its patients to realize the full benefits of this program.
There are many more lessons to be learned as we embark on this journey. The tool must be continuously monitored and adjusted, as must the IT systems, the research questions, and the analytics. Thankfully, continuous improvement is exactly the type of work we embrace.