From The Editor | March 19, 2014

What's Next For Health Data Analytics?

ken congdon

By Ken Congdon

As referenced in my HIMSS14 post show analysis, big data, population health management, and clinical analytics were among the hottest topics at this year’s premier health IT conference. However, the surface-level excitement surrounding these topics quickly revealed underlying provider uncertainty about how to leverage solutions in this category most effectively. Moreover, it exposed the relative immaturity of analytics technology and applications in the healthcare industry. In an effort to gain a deeper understanding of the current state of analytics in healthcare and the technology developments on the horizon, I recently spoke with leaders from two analytics vendors (Philips Healthcare and Truven Health Analytics).

Dr. Anthony Jones, CMO of Philips Healthcare’s Patient Care and Clinical Informatics business unit agrees that health data analytics is in its infancy. “When it comes to big data, healthcare is currently facing what other industries faced years ago with data warehousing,” he says. “Providers need to not only collect data in a usable digital fashion, but also pull it together into a single location where analytics tools can be applied to provide actionable information. The healthcare industry is still in the early stages of this evolution — providers are collecting and beginning to realize the value of the data, but they aren’t yet integrating it and making it more usable.”

According to Dr. Michael Taylor, CMO of Truven Health Analytics, one of the main reasons for the slow adoption of analytics tools in healthcare is due to the recent revelation that the technology was even necessary. “Initially, I think there was an expectation among many health providers that the EMR by itself would provide the analytics tools required to improve efficiency, performance, and outcomes,” he says. “Providers now realize that the EMR is not the end-all, be-all. They are recognizing the need for analytics, and are starting to make the necessary technology investments.”

Cost Analysis, Risk Assessment Key Analytics Applications To Date

According to Dr. Taylor, most initial applications of analytics technology by health providers have been focused on understanding and managing cost pressures. In other words, they are collecting and analyzing claims data to identify problem areas and cut costs. Even small victories in this area can be huge for hospitals now forced to do more with fewer resources.

“Hospitals are typically very low-margin operations,” says Taylor. “Even a 1% or 2% savings in specific areas can have a significant impact on their overall cash position.”

While the market to date may be dominated by analysis of claims data, Dr. Taylor does see a new trend beginning to emerge. The rise of risk-based contracts is driving many hospitals to begin introducing more types of data (especially clinical data) into their analysis efforts to gain a deeper understanding of their patient population and better manage risk.

“Providers are just beginning to combine inpatient and outpatient data to gain more insight into the full range of what’s going on with their patients,” adds Taylor. “This is a step in the right direction, but most providers just incorporate historical clinical information, not real-time clinical data. This needs to be the next phase of analytics evolution in healthcare.”

Another analytics trend Dr. Jones sees taking shape is an increased demand among health providers for better visualization tools. “Many of our customers are actually more interested in visualization than they are in the analytics themselves,” he says. “They don’t actually need to incorporate more data, they just need better ways to visually process and digest the data they already have so they can act upon it.”

One of the ways Philips is trying to address this need is via a new graphical interface it developed for its IntelliSpace eCareManager software. The new interface is called orb and it’s a dashboard designed to help clinicians quickly prioritize a large number of patients with complex needs based on strong visual cues. With orb, each patient in an ICU (and their different body functions/vitals) is represented by circles of different colors and sizes. A circle that gets larger or changes color is an indication of a deteriorating condition. A clinician’s eye is naturally drawn to this graphical change, and prompts them to quickly take the proper action.    

Hospital-To-Home Initiatives Are The Future Of Health Analytics

Dr. Jones and Dr. Taylor both agree that the future focus of health data analytics will be on helping hospitals and health systems not only assess patients in inpatient and outpatient settings, but throughout the entire care continuum — even extending to the patient’s home.

“In the past, it was never a big issue for hospitals to keep tabs on patient conditions outside the four-walls of their facilities,” says Dr. Jones. “They were paid for the services provided onsite. Once the patient went home, their obligation ended. Now, with value-based reimbursement, there is increased focus on predicting and preventing hospital readmissions. Doing this effectively will require providers to monitor and analyze patient data collected remotely.”

According to Dr. Jones, analytics vendors will need to expand their algorithms and incorporate new types of data to provide clinicians with a complete longitudinal view of the patient and deliver the real-time information necessary to prompt intervention and keep patients out of the ER.