Oregon Health Insurance Experiment

I have been teaching an advanced microeconometrics course specialising on estimation of treatment effects. One of the lectures was devoted to examining the analysis of randomised controlled trials. A particularly interesting recent randomised trial is the Oregon Health Insurance Study. A description of the study is below:

"The Oregon Health Study is the first randomized controlled trial examining the impact of insuring the uninsured. In April 2008, the Oregon Health Authority determined that it had enough funds to provide health insurance to an additional 10,000 uninsured low-income adults through the Oregon Health Plan Standard (OHP) program– but 90,000 people wanted in. To be as fair as possible, the state held a lottery to determine who would receive an OHP application and who would not. The OHP lottery randomly assigned people to treatment (those who “won” the lottery and received the opportunity to apply for OHP) and control (those who did not receive the opportunity to apply) groups.  The Oregon Health Study follows participants over time to measure the impact of access to OHP."

 A key paper from this study was published in the QJE in 2012.

"In 2008, a group of uninsured low-income adults in Oregon was selected by lottery to be given the chance to apply for Medicaid. This lottery provides an opportunity to gauge the effects of expanding access to public health insurance on the health care use, financial strain, and health of low-income adults using a randomized controlled design. In the year after random assignment, the treatment group selected by the lottery was about 25 percentage points more likely to have insurance than the control group that was not selected. We find that in this first year, the treatment group had substantively and statistically significantly higher health care utilization (including primary and preventive care as well as hospitalizations), lower out-of-pocket medical expenditures and medical debt (including fewer bills sent to collection), and better self-reported physical and mental health than the control group."

As can be seen above, granting access to medicaid had effects on both utilisation of healthcare services and on health (see also their Science paper and NEJM paper). These results relate to the early stages of the study and it will be very interesting to see what happens as participants are tracked over time.

The study is a very good example of the use of randomised trials to answer large-scale policy questions. It also demonstrates the potential for linkage of administrative and survey data in examining the effects of public policies. The low take-up of health insurance (albeit high enough to generate strong first stage predictions) among those offered might be a particularly interesting question for behavioural economists.