An odd dream, to be sure. But health insurance data is crucial to understand how health care dollars get spent. It shows how people use health care, what’s changing and, in some cases, why. Health insurers, however, have tended to keep that data private, as it could tip competitors off to how they handle business.
That all, however, changes today. This morning a new nonprofit called the Health Care Cost Institute will roll out a database of 5 billion health insurance claims (all stripped of the individual health plan’s identity, to address privacy concerns).
Researchers will be able to access that data, largely using it to probe a critical question: What makes health care so expensive?
The Health Care Cost Institute database is massive. With data voluntarily supplied by Aetna, Humana, UnitedHealth and Kaiser Permanente, it covers 33 million Americans with employer-sponsored health insurance. In order to store just one of those insurers’ data in paper form, you would need 60 million four-foot filing cabinets.
“It’s the first time that this data has ever been assembled like this,” says Martin Gaynor, who chairs the Health Cost Care Institute Board, adding that the non-profit’s data people “have been working like crazy to make this happen.”
The database has been in the works for about two years. “At various junctures, it would look like this wasn’t going happen,” says Gaynor, who is also a professor of economics and health policy at Carnegie Mellon University. “There was a lot of effort to make all the data line up.”
Now that all the data is in one place, researchers can start to tackle questions like: Where is health care expensive? Are certain procedures driving up prices? Is health care becoming more costly because of higher prices or volume?
HCCI’s own economists have tackled that last question, in a report out today, the first to use the new database.
It finds that higher prices charged by hospitals and other prices have driven health care cost growth during the recession, rather than Americans using more medicine. Medical prices grew three times faster than the Consumer Price Index, a measure of price inflation, between 2009 and 2010.
This confirms similar trends seen in the National Health Expenditures report as well as in Medicare data, both of which show people using less health care as the economy slowed.
Researchers have usually relied on Medicare data to answer these questions. The Dartmouth Atlas, which has spearheaded much of the country’s health cost research, relies on that data, largely because the federal government makes it available. And while the 30 million Medicare patients make for a large dataset, they do have the distinct statistical disadvantage of all being elderly.
With the HCCI database, researchers can understand better what’s happening to the 87 percent of the population under 65. Preliminary research suggests that there are indeed different things happening in each age group: While health care costs grew 4.5 percent for the under-18 population between 2009 and 2010, the rate of change was half that for older adults between 45 and 54.
This is the first study to use the HCCI data, although more are in the works. Gaynor has been inundated with about 130 requests from health policy researchers to use the database. While his team sifts through those, three approved studies are already tackling big health policy questions.
Researchers at Northwestern University are using the data to probe the effect of business cycles on health insurance. That could get at one big question in health policy right now, as to whether the recession is responsible for a slowdown in health care costs. Carnegie Mellon-based economists are teaming up with researchers at the London School of Economics to look at variations in hospital pricing, which may provide data similar to that of the well-respected Dartmouth Atlas, except for the non-Medicare population.
“There is immense interest in gaining access,” says HCCI executive director David Newman. “We’re having trouble keeping up with that.”