Take predictive analytics to the next level in healthcare

Tools

Knowing predictive health analytics can improve patient care and contain costs, the University of Pittsburgh Medical Center, which owns hospitals and insurance plans in Pennsylvania, developed prediction models that take analytics beyond patient claims to household data, like shopping preferences, reports The New York Times.

The insurer found some interesting trends: If patients shop online, they may be more likely to use certain emergency services. To determine the correlation between hospital visits and Internet users, the insurer looked at more detailed information, such as a member's household income, education level and number of children at home.

After analyzing the data, UPMC assigns care coordinators to certain high-risk members who may have chronic conditions that aren't being treated properly. That way, the insurer can recommend primary care physicians or specialists to treat the member instead of costly emergency room visits, notes the Times.

As the industry becomes more data-driven, many insurers want to better understand their member population. Blue Shield of California Senior Vice President of Customer Quality Rob Geyer, told CIO Magazine his company recognized the great potential data has to transform the healthcare system and is working with a partner to analyze claims data.

What's more, data analytics could help save U.S. citizens as much as $450 billion in healthcare costs, FierceHealthIT previously reported.

But new innovations often bring skepticism. Many industry experts worry analytics pry too much into their members' lives. "This intensive, intrusive kind of data analytics that leads to differential treatment of customers, even if we are fine with it in the business context, needs to be disclosed in the medical context," Frank Pasquale, a professor in healthcare regulation at the Seton Hall University School of Law, told the Times.

For more:
- here's the NYT piece
- here's the CIO Magazine interview

Related Articles:
Predictive analytics catch fraudulent claims
Big data use: Providers, payers not on the same page
Data analytics to prevent fraud
4 ways to become more customer-centric