TDWI | Training & Research | Business Intelligence, Analytics, Big Data, Data Warehousing

Healthcare Analytics in the Face of Heavy Volume

These three ideas for leveraging data can help healthcare companies face an uphill battle with the high costs of handling overwhelming data volumes.

• By Jake Dolezal
• September 9, 2016

A few years ago, I wrote an article about business intelligence in the face of the healthcare volume-to-value shift. This article focused on shifting healthcare data management and BI to become more patient-centric, leverage big data, and prepare for using data science in prevention.

Although following these trends is still valuable, heavy volume continues to be the Achilles’s heel of modern healthcare in the U.S.. With the recent news about the Aetna-Humana merger and Aetna leaving the Affordable Care Act exchanges in 70 percent of markets, it seems even the largest insurers are shying away from shouldering huge volumes of healthcare claims.

The sources of these heavy claim volumes are the subject of much debate. Regardless, healthcare companies and organizations must embrace intelligent, data-driven decision making unlike ever before. They face an uphill hike of continual process improvement if they want to manage this high volume, and many are wisely looking towards data for solutions.

To handle these volumes, healthcare organizations should use their data to drive improvement design and measure the effectiveness of implemented solutions every step of the way. I suggest they start by looking under the following data “rocks.”

1: Uncover the Root Cause of Inaccurate Payments

Claim payment adjustment and recoveries due to over- or underpayment are costly activities. Everything from the call from an incorrectly paid provider to follow-up work involved in recovery eats into the bottom line.

The first step is to zero in on the payment inaccuracies that are under your control -- this requires defining and identifying which inaccuracies are due to internal errors and which happen as just a part of everyday business. Find the underlying causes of internal errors and correct them, then implement targeted and automated key performance indicators to measure your improvements (in terms of both volume and dollars).

2: Turn Compliance Reporting from Reactive to Predictive

Each state has compliance regulations regarding the turnaround time of claims and other requirements; these targets are often tricky to hit and even trickier to manage. Unfortunately, most compliance reporting is “post mortem,” meaning the awareness is raised after the compliance limit is breached and it’s too late.

Regression analysis of the claim life cycle from receipt to payment for those that fail to hit compliance standards may provide underlying causes and early warning signs. It may also help pick out problematic claim needles from the haystack of claims that can be easily adjudicated and paid on time.

3: Triangulate Claims Data with Member Enrollment, Encounters, and Call Center Data

Just like a dot-to-dot drawing, the full picture is not realized until you connect the dots. Triangulating claims data with information from other sources and breaking down data silos could involve a variety of data integration, federation, or big data solutions.

Architecting a solution to analyze data with high volume and variety is a challenge, but it may allow you to discover patterns in the volumes of claims, member enrollment growth, and call center or appeals and grievances burdens. Identifying those patterns will allow you to set up automated alerts to trigger in any systems and processes that show a spike in inbound activity.

Healthcare companies certainly have their hands full, and anyone near the industry can describe the problems much more easily than they can envision the solutions. Leveraging data in new ways will bring these companies one step closer to sufficiency (and profitability) in spite of overwhelming volumes and regulatory challenges.