What did we learn from the Oregon HIE?

The Oregon Health Insurance Experiment: Evidence from the First Year

By Amy Finkelstein, Sarah Taubman, Bill Wright, Mira Bernstein, Jonathan Gruber, Joseph P. Newhouse, Heidi Allen, Katherine Baicker, The Oregon Health Study Group
The National Bureau of Economic Research, July 2011

NBER Working Paper No. 17190

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 a unique 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.


By Don McCanne, MD

Although innumerable studies have shown that health insurance provides both health security and financial security, some have contended that insurance is not necessary, especially for low income individuals, since they can find care through our safety-net institutions. As President George W. Bush stated, "After all, you just go to an emergency room." This study, the Oregon Health Insurance Experiment (Oregon HIE), puts an end to that contention. Low income Oregon residents who were selected by a random lottery to be enrolled in Medicaid fared significantly better than those who were randomly excluded.

The Oregon Medicaid lottery provided a great if disconcerting opportunity for policy researchers. It would be unethical for them to have designed and carried out a study that randomly provided health care benefits to one group while leaving uninsured another group serving as a control. This unique opportunity arose when Oregon received more funds for their Medicaid program, but not nearly enough to cover everyone eligible. They decided to randomly select by lottery those whom they could fit into the program while leaving many more out in the cold.

It is ironic that politicians created a study opportunity that is clearly unethical by policy research standards, yet doesn't seem to violate the ethical standards of the politicians. If you need any more proof of that, just look at the Affordable Care Act (ACA) where policy decisions were made that will leave 23 million individuals without any health care coverage, but they can go to the emergency room.

Although much of the media coverage of the Oregon HIE implies that it was a wise decision to include in ACA the expansion of coverage under Medicaid, there are many other factors not covered in this study. Two important considerations stand out.

Having Medicaid is better than having no coverage at all, but access is still impaired under the program because of a lack of willing providers, especially for specialized services. Bringing millions more into the program will certainly exceed the already inadequate capacity. With the low level of funding for this program, it is unlikely that funds will be available in the near future to expand capacity. Adequate capacity for everyone does exist within our entire health care delivery system, but not when providers are allowed to opt out of an underfunded component of it.

Another important element in this study is that Medicaid eliminates most cost sharing by the patients. If the patients are able to find providers who will accept them then they do not have to face financial barriers that would prevent them from receiving recommended care, tests, prescriptions, and preventive services that they should have. This is one of the major advantages of the Medicaid program (though some states are now considering cost sharing as a means of reducing Medicaid spending).

Supposedly the RAND Health Insurance Experiment (RAND HIE) demonstrated that cost sharing did not have a major impact on health care outcomes, except for low income individuals, even though health care utilization decreased. The problem with the RAND HIE conclusions is that they have only intrinsic validity for a healthy workforce and their young, healthy families for only a brief, healthy interval in their lives. It is inappropriate to expect the same result for a cross section of our population including a fair sampling of those with greater health care needs. Had greater cost sharing been used in the Oregon HIE, utilization certainly would have been less, and it is likely that outcomes would have been worse.

Before we start celebrating the fact that Medicaid is better than nothing at all, let's keep in mind the facts that Medicaid lacks the capacity to meet the expanded coverage through ACA, that eliminating cost sharing does improve access, and that expanding Medicaid will do nothing for the 23 million remaining uninsured and the tens of millions more who will be underinsured through low actuarial value plans with very high cost sharing.

There is a far better way - a single payer national health program that provides comprehensive benefits for everyone with no cost sharing barriers to care. Other nations have adopted such programs at a far lower cost than our dysfunctional system. We can do it too. (You've heard this before.)