OR WAIT null SECS
Sponsored: How FHIR can help your health system do more with data.
Russ Leftwich, M.D., senior clinical advisor for interoperability at InterSystems, describes how SMART-on-FHIR solutions can help health systems make better decisions. This branded article is sponsored by InterSystems.
In an article in 2011, Dr. William Stead at Vanderbilt University estimated that in medical practice, there were roughly 10 facts or data points needed to inform one complex decision. At the time, pulling those 10 facts together was feasible for two reasons: First, there were fewer total data points to comb through to get to the ones you needed. Second, most of them lived in just two locations — the patient’s medical record and the clinician’s well-trained mind.
As wondrous as a doctor’s mind can be, research suggests that the average among us can only wrangle about five facts at a time, or about half of the facts required to inform a complex medical decision. If you are exceptional, you may be able to handle 7 or 8 facts. Stead’s estimate states that the number of facts continues to increase exponentially. In 2018, there are roughly 700 data points needed to inform one complex medical decision, and by 2020, there will be about 1,000.
Not only are there increasingly enormous amounts of data related to individual patients, but the systems in which the data are recorded continue to grow too. A key actionable data point could be siloed in a distant genomic sequencing lab, a personal mobile device or a government database, to name just a few. But often, you need all of these data points to see the true picture and inform the best decision.
So, as data about individuals grows, health systems are left with two pressing questions: How can I make sense of the data I have, and how can I get my hands on the missing data I need?
According to Russ Leftwich, M.D., that’s where interoperability comes into play.
“There’s just too much data out there, yet in clinical care our workflows are still very much the way they were back in 1960,” Leftwich, the senior clinical advisor for interoperability at InterSystems, said during a presentation at HIMSS 2019 in Orlando, Florida. “We still practice in hospitals and clinics as if we could know in our minds every data point needed to inform the complex decisions we have to make.”
When people first started using the term “interoperability” in clinical care in the 1980’s, it mostly referred to the ability to connect the increasing number of departmental systems within a single hospital. The idea of exchanging data electronically between providers is more recent.
Thus, in the last 40 years, interoperability has evolved alongside a rapidly changing healthcare ecosystem. For example, it’s estimated that the average hospital now has more than 80 distinct IT systems within its walls before accounting for mergers and acquisitions. “It’s remarkable that we’ve leveraged standards from the 1980s to do what we do, but they don’t really meet our current needs,” Leftwich said.
In short, modern interoperability means being able to see all the data relevant to a patient or a population in real time. To take that a step further, we can look at HIMSS’ newly published definition:
“Interoperability is the ability of different information systems, devices or applications to connect, in a coordinated manner, within and across organizational boundaries to access, exchange and cooperatively use data among stakeholders, with the goal of optimizing the health of individuals and populations.”
Based on that definition, we can see that while health systems are constantly working toward interoperability, it remains out of reach today in medicine. That can be dangerous, Leftwich explained.
“Medical errors are the third leading cause of death in the U.S., but I would argue that most of these are not truly medical errors. Rather, they occur when the wrong decision was made because all of the data wasn’t available to make the best decision, and because there wasn’t sufficient clinical decision support in place,” he said.
The recognition of that reality is what led to Health Level Seven’s (HL7) creation of Fast Healthcare Interoperability Resources (FHIR), a standard for exchanging healthcare information electronically designed to simplify implementation without sacrificing information integrity and leveraging the technology in use in the web in other domains.
FHIR acts as a REpresentational State Transfer interface, or REST interface. These types of interfaces are the basis of what we do every day on the internet, and support the operation of sites like Google, Twitter, Facebook, and many e-commerce sites, Leftwich said. Their value lies in representing the data on one server on another computer in its current state.
Travel websites can help us understand the concept, Leftwich added. When you enter your departure and arrival locations and a travel date, in a few seconds you see a list of flights on different airlines. This works not because the travel website has downloaded all the airline schedules, but because all of the airlines have agreed on how to represent the data in an airline flight — how to ask for it, and what we will receive in return.
FHIR is the same technology and it is the meaning of the data in healthcare, Leftwich said. This is where it gets complicated. The data in healthcare is a lot more complex than the data in an airline flight, and there is a lot more of it. Not only is the meaning of the data complex, but the relationships of the pieces of data is also very important.
“Suppose you print out a patient’s history and physical exam and then cut out each data element with scissors. Now you have all these pieces of paper. They have words that you understand, but you don’t know the relationship of those words because they’re no longer attached to the rest of the document,” Leftwich said. “If you pick up a snippet and it says ‘diabetes,’ you don’t know if that’s the condition the patient has, the condition they used to have or the condition their mother had in their family history. As you can see, maintaining the relationship of data in a healthcare scenario is essential to interoperability.”
While many enterprises see FHIR as a draft standard from HL7, the organizations that have been early adopters of FHIR for internal uses are already starting to see benefits, according to Leftwich. With FHIR, they’re finding it easier to access data they previously couldn’t and interface with new systems and display data in insightful new ways that were not possible without it. They can do things they couldn’t do before, he said.
And while FHIR standards will continue to evolve, the product’s core functionality is now normative, which means adopters can begin to build new strategies and expect long lifecycles. The core part of FHIR will remain stable and future versions will be backward compatible.
“We have a new concept that I call ‘the iPhone model’ of standards maturity,” Leftwich said. “We know that the iPhone n+1 will always be better and have more features than the current version, but the core functionality is the same. And like early iPhone adoption, some people are using FHIR because they can do things with it that they never thought possible. Other people are waiting because they haven’t yet appreciated the value of FHIR.”
Learn more about how InterSystems helps healthcare organizations extract value from data.