It looks like four major private health insurance companies will be releasing data for use by academic researchers. They will create a non-profit institute called the Health Care Cost Institute and deposit the data there. Researchers can request the data from the institute by (I’m guessing) writing a short proposal.
Health insurance billing claims data might not sound all that exciting, but they are a gold mine of very interesting information about population health. In my group, we use billing claims from Medicare Part A to explore the relationships between ambient air pollutants and hospital admissions for various cardiovascular and respiratory diseases. The advantage of using a database like Medicare is that the population is very large (about 48 million people) and highly relevant. Furthermore, the data are just sitting there, already collected. The disadvantage is that you get relatively little information about those people. For example, you can’t find out what a particular Medicare enrollee’s blood pressure is on a given day. Also, it requires some pretty sophisticated data analysis skills to go through these large databases and extract the information you need to address a scientific question. But this “disadvantage” is what allows statisticians to play an important role in making scientific discoveries.
It’s not clear what kind of information will be made available from the private insurers—it looks like it’s mostly geared towards doing economic/cost analysis. However, I’m guessing that there will be a host of other uses for the data that will be revealed as time goes on.