5 reasons to get going with data analytics


Welcome to the data world. Many secrets are hidden in big data, and now, with the computing power to unearth them, analytics promises to deliver transformative power wherever it is put to work. Still, technology is a relative newcomer in the healthcare world. According to Brett Furst, CEO of Arbormetrix, there is nothing to fear – analysis of clinical data has much to offer the medical world. Here, he shares his top five requirements to succeed with, or at least get excited about, the power of clinical analytics.

1. Know the difference between solutions. Analytics solutions vary widely in size and shape. Furst said it is important to know what the different kinds are, and how to apply each one to specific problems, wether they have to do with population health and disease management, episodic delivery or post-acute care. Population health and disease management focuses on "improving the general health of a population and keeping them out of a hospital," according to Furst. Think screening a database to find people who might be at risk for a certain condition and reaching out to them. Episode analytics "focuses on identifying variation in the delivery and associated outcomes of specialty and acute care." This kind of analytics is about looking back and finding ways to improve care in the future based on how it was provided previously, explained Furst. Post-acute analytics centers around "utilization management so patients receive the appropriate level of care after hospitalization," with a focus on cutting down on wasted resources. Essentially, the three flavors Furst outlined could be seen as the analytical equivalents of before, during and after. 

2. What's in the data? Knowing which analytics can be applied to which problems opens the door to immense functionality. With the rise of ACOs and the paradigm shift of reimbursements for quality of care, healthcare providers are scrambling to approach the health of their populations proactively. Analytics has a role in this shift, and Furst said harnessing its power means that organizations will be able to more intelligently identify, solve and manage the challenges that they are beginning to face. Furst said being able to ask questions such as, "Where the spending is, how many readmits do we have every year?" have a massive "effect on clinical performance [that] goes to the actual outcome of the patient." By taking the data generated in a hospital and making sense of it with clinical analytics, Furst said there is a real ability to find and tackle performance issues. "When you combine good clinical data with good accounting data, you can pinpoint what types of conditions might make for readmits," he added.

3. Make data actionable. Furst said there's a common malaise in the industry around the promises analytics and big data seem to offer. He was careful to caution that "just aggregating your data isn't going to lead to big benefits," and that "data is just going to be a reference point." The important thing to remember, he said, is that clinical analytics is a tool first and foremost, and that without knowing which problems need to be solved, their use is limited. Furst said the ideal way to look at it is as if a hospital were like any other type of business: trying to do a top-to-bottom complete overhaul is a tough pill to swallow, and one that may not end up being that effective. Instead, he recommends taking the approach of "Let's start zeroing in on one area, use the data to find where to start." When given a specific area to approach, with clearly defined goals and steps to take, the best results will emerge. "I see the real opportunity..when you apply a higher level of algorithms to make the data more actionable," he said.

4. Understand the additional benefits. Who says clinical analytics is a one-trick pony? By its very nature, analytics is the practice of taking a close look at a large amount of data and then driving outcomes with its findings. Furst said this can be put to a variety of uses in the healthcare world. When the lens is turned in an analytical fashion to the ways doctors work, the results can drive and change the development of best practices. Furst said that in the old fee-for-service world, "surgeons would do what they thought was medically appropriate, but they did it in a vacuum." Now, "when you come to them saying this device is $15,000 and this one is $2,000 -- and the $2,000 device is actually better -- you're improving care and impacting your bottom line." 

5. Provide evidence based on the demographics of the patient. As well as being a powerful tool to drive changes in the operating theatre and the board room, clinical analytics has a role in communicating with the patient. Furst said that a clinician can sit down with his or her patient and pull up treatment files that match that patient's demographics as a way of saying "based on evidence the better procedure is A instead of B." Compared to already existing tools such as WebMD, which Furst feels contain "a deluge of information" that may not necessarily be relevant to the person reading it, clinical analytics has the ability to filter out the unimportant and to provide better information. Furst said analytics enables practitioners to say "our statistical science shows this is the best probable outcome for you."