News: Incomplete data stalls predictive analytics effectiveness

The majority (57%) of healthcare payers and providers believe predictive analytics will save their organizations 15% or more over the next five years, according to a 2017 Predictive Analytics in Healthcare Trend Forecast by the Society of Actuaries, Revenue Cycle Advisor reported. Though there’s a great potential for savings, however, several significant roadblocks stand in the way.

Of the 223 healthcare executives surveyed, 45% of respondents were payers and 55% were providers. Respondents said the three biggest hurdles for predictive analytics implementation in the next five years are:

• Lack of budget (16%)
• Regulatory issues (13%)
• Incomplete data (12%)

Despite the anticipated speedbumps, though, 93% of respondents said that predictive analytics is important for the future of their business. Even more indicative, though, 57% of respondents using predictive analytics believe it to be the tool that saves their organization, Revenue Cycle Advisor reported.

Though more payers (63%) currently use predictive analytics than providers (47%), nearly 90% of each group say they either use predictive analytics currently or plan to use them within the next five years.

Trends, however, reveal that payers and providers will use the analytics for different services. While providers most value predictive analytics for patient satisfaction (53%) and hospital readmissions (48%), payers find the most valuable use related to cost (51%) and profitability (47%).

“Predictive analytics don’t give you the definite answer of what might be missing [in the documentation and coding], or what might be wrong. But, the analytics give you clues to investigate,” Felicia Ziomek, MBA, BSN, CHFP, nurse auditor and CDM coordinator at Stanford Healthcare-ValleyCare in Dublin, California, told the National Association of Healthcare Revenue Integrity.

Editor’s note: This article originally appeared in Revenue Cycle Advisor. To read a case study about Stanford Healthcare’s use of predictive analytics, click here.