Why Predictive Analytics is the Answer to Healthcare’s Big Data Problem

By Scott Hampel

Making improvements in healthcare data analytics has the potential to lead to significant cost savings and improved patient health and wellness. We’ve identified two ways these improvements can be accomplished:

1. Embrace an enterprise analytics strategy to process vast amounts of healthcare data, rather than historical approaches of stitching together dozens of analytics tools resulting in high IT costs and low user adoption
2. Help payers and providers understand what the data means and how to act on the information

By 2020 the amount of healthcare data produced worldwide is expected to reach 2,314 exabytes; in 2013, only 153 exabytes of healthcare data were created.

Healthcare is flooded with data produced by patients, physicians, hospitals, urgent care centers, emergency departments, health fairs and personal wearables. The amount of data created will continue to increase as time goes by and technology advances. Healthcare organizations that don’t adapt will simply cease to exist, crushed under the weight of data, or be consumed by healthcare businesses that recognize the power of data analytics and implement it at an enterprise-wide scale to harness and use the information.

“Big data analytics has the potential to transform the way healthcare providers use sophisticated technologies to gain insight from their clinical and other data repositories and make informed decisions. In the future, we’ll see the rapid, widespread implementation and use of big data analytics across the healthcare organization and the healthcare industry,” according to an article published by Health Information Science and Systems. “Big data analytics in healthcare is evolving into a promising field for providing insight from very large data sets and improving outcomes while reducing costs.”

Enterprise and predictive analytics are the watershed moment in the world of healthcare and Big Data. Recognizing and acting on complex data today—not five or 10 years from now—likely will determine success or failure for many healthcare organizations and businesses.

Analytics for the Enterprise

The Society of Actuaries released a report focusing on the use of predictive analytics in healthcare in 2019 and found 60% of respondents currently use predictive data processing methodologies. This is a significant increase from 2018 when less than 50% said they were using predictive analytics. “(T)he good news is that organizations using predictive analytics are actually achieving their desired results. Our hope is that this trend will continue through additional predictive analytics implementation across the industry,” explains Sarah Osborne, senior vice president, chief actuary and analytics officer at the Government Employees Health Association, in the society’s report.

The results include:

• Reduced costs;
• Improved patient satisfaction;
• Better clinical outcomes; and
• Increased profitability.

Concerning, however, is the 40% who say they have no plan to use predictive analytics ever, may get into it in five years or so, or have another timeline in mind, all of which makes these organizations vulnerable to acquisition or worse.

Healthcare organizations can continue to ignore enterprise analytics, but they won’t survive long in today’s data-flooded environment and likely won’t do well in the future.

The adoption of enterprise analytics today is consistent with where EMRs were 20 years ago. There were many established departmental software vendors and a single organization implemented several different solutions from different sources. This led to excessive spend to try integrating solutions, fragmented workflows across different departments, and difficultly organizing resources across the various specialties and departments of a healthcare provider.  The organizations could have stayed with multiple vendors, but it made sense to adopt the EMR as an enterprise solution. Doing so created many internal synergies, improved patient and physician satisfaction, improved care delivery, and many other general advancements and efficiencies.

Healthcare analytics has the same status today with hundreds of vendors selling a slightly different spin on analytics or focusing on a specific aspect of healthcare or revenue data. Fragmented use of healthcare analytics drives excessive spending and low staff adoption and ROI.

It’s critical for healthcare organizations to embrace an enterprise analytics strategy and centralizing it if possible. Many healthcare organizations are looking at ways to consolidate solutions. They’ve learned—from EMRs—that attempting to stitch together 10 or 20 different solutions becomes extremely expensive and completely ineffective. Adoption and use of these cobbled-together solutions are extremely low. This results in the organization becoming uncompetitive because it creates an environment of:

• Low efficiency;
• High labor costs;
• Ineffective patient care; and
• Inefficient performance.

Those who do adopt effective enterprise analytics strategies generally are better off than their peers in several ways. Healthcare organizations that employ enterprise analytics will create new opportunities in other areas. It frees them up to make investments in new service lines, to serve new populations, to build new facilities or to acquire other healthcare organizations. (Perhaps a few of those in the 40% mentioned earlier, who remain on the enterprise analytics fence.)

As an important part of enterprise analytics, predictive analytics often is the first step in bringing the entire organization on board. Predictive analytics gives organizations a good sense of what’s working well and what’s not. This helps everyone involved in providing healthcare to make better decisions and act with insight and knowledge, not guesses.

Predictive analytics helps healthcare organizations:

• Forecast models for revenue cycle management;
• Analyze population health participants and take action;
• Predict future challenges that drive costs;
• Locate patients with drug-seeking behavior for intervention; and
• Identify anomalies or outliers that cause large-scale health or financial risk.

The more often data insights are elevated and put into the hands of end-users and decision-makers, the better off the healthcare organization. Enterprise analytics ensures uniformity, availability and organization-wide visibility, all of which are crucial to success. In addition, enterprise and predictive analytics save time and resources, and can mean better outcomes for patients and lower costs for healthcare organizations.

Real change in healthcare can be achieved when the adoption of enterprise and predictive analytics is accelerated and brought to those in the business and patient care sides of the organization. The potential to make real improvements to healthcare is significant.

Scott Hampel is President at MedeAnalytics.