Plan Watch

Researchers Mining HCCI Trove of Private Payer Data for Pricing Insights

The info comes compliments of Aetna, Humana, and UnitedHealthcare.

Joseph Burns
Contributing Editor

Since December, researchers have been releasing study after study showing how health insurance works—and doesn’t work, in some cases—because of unprecedented access to data from three large national insurers.

For four years, the Health Care Cost Institute (HCCI) has been acquiring data from Aetna, Humana, and UnitedHealthcare and in the past year or more has added staff, engaged in partnerships with academic researchers, and received $1.5 million in grant funding from the Laura and John Arnold Foundation. All of this has allowed HCCI to produce research on health care spending and prices and how states are implementing the ACA and other reforms.

Previously, researchers relied almost entirely on CMS data, which represents only 16% of total spending but has skewed much of the nation’s health policy research, according to Zack Cooper, an assistant professor at Yale who published one of the first reports based on the insurers’ data.

The HCCI data from the three insurers include 92 billion claims filed from 2007 to 2011. The claims came from 88 million Americans with employer-sponsored coverage, a little more than a quarter of the American population.

In December, Cooper and colleagues used the data to document wide variations in hospital prices within and across geographic areas, and to examine how hospital prices for the privately insured vary widely. They showed, for example, that when hospitals have a monopoly in their referral regions, prices are 15.3% higher than in regions with four or more hospitals, even after researchers controlled for cost variations among regions. Their findings about market power and price is significant because it gives health insurers evidence to refute a common argument among hospital administrators (“our costs are higher”) when explaining price variation. The National Bureau of Economic Research published its research, and Cooper and his colleagues posted it, along with other supporting data, on a website called the Health Care Pricing Project.

More recent research using the HCCI treasure trove showed that consumers in some states spent twice as much for some health care services than consumers in other states and that the price of health care services varied by threefold within states. HCCI Executive Director David Newman and colleagues published their findings in the May 2016 issue of Health Affairs. When they looked at price variations for 242 common medical services, they found that Alaska, Wisconsin, North Dakota, New Hampshire, and Minnesota had the highest prices, while Arizona, Florida, Maryland, and Tennessee had the lowest. They also found large variations within states; for example, in California, the price of a knee replacement ranged from $57,504 in Sacramento to $30,216 in Riverside.

A month earlier, HCCI researchers reported findings showing that health care spending for children with private health insurance has risen sharply in recent years because prices have gone up. For example, the price of brand-name prescriptions more than doubled from $7 per filled prescription per day in 2010 to $16 in 2014.

Confirming conventional wisdom

In February, HCCI released six research reports based on the health insurers’ data. In some cases, they confirmed conventional wisdom: Using nurse practitioners in primary care settings cuts costs, and consolidation among oncology providers drives up the cost. In others, the HCCI gave an evidentiary boost to some fashionable developments in health care. HCCI’s data crunch showed, for example, insurers’ payments for telehealth services are 40% lower than care that does not use telehealth.

“We believe that this data-driven research will build a knowledge base about the changing health care landscape, and serve as a resource for policymakers and consumers alike,” said Amanda Frost, a senior researcher at HCCI.

She might have added that the data are likely to prove useful to insurers as well.


10 Most Important Numbers in American Healthcare
By Robert Calandra
10 Most Important Numbers in American Healthcare
By Timothy Kelley
10 Most Important Numbers in American Healthcare
By Joseph Burns

A blueprint for high-volume, high-quality lung cancer screening that is detecting cancer earlier—and helping to save lives

Clinical Brief

Multiple Sclerosis: New Perspectives on the Patient Journey–2019 Update
Summary of an Actuarial Analysis and Report