Introduction

The Oregon Medicaid experiment[1] is referenced frequently by policy advocates in support of their positions on Medicaid, health insurance, and a number of other issues. Specifically, it has been used on both sides of the aisle to both support and condemn the Affordable Care Act’s (ACA) Medicaid expansion. Since it is so regularly referenced by advocates and analysts on all sides, it is appropriate to step back and look at what the study actually found.

Background

In 1994, before passage of the ACA, Oregon expanded its Medicaid program, known as the Oregon Health Plan (OHP), to all citizens and legal residents below the poverty level. To pay for this expansion, Oregon prioritized certain benefits that the state determined to be of high-value, which were then covered by OHP. Oregon then moved its enrollees into Medicaid managed care plans. Enrollees who were not included in the Federal Medicaid population were also required to pay an income-adjusted premium and copayment to participate in OHP.[2]

Despite the attempts at cost saving, OHP was too expensive and in 2004 Oregon closed OHP enrollment to individuals who were not Medicaid eligible under the Federal standards.

In 2008, Oregon had saved enough money to again open OHP enrollment to all Oregon residents below the poverty level. However, there was only enough money to cover 10,000 new enrollees out of 85,000 applicants.[3]

The ensuing ‘Medicaid lottery’ used to choose the 10,000 participants accepted into the program gave researchers a rare opportunity: they had before them a naturally occurring, randomized, controlled study of the impact of Medicaid as health insurance for this expansion population. For ethical reasons, no study of this kind had ever taken place before, nor is it likely to occur again post-ACA. The results of this experiment have been of great interest to health policy scholars.

Results

The big take-away after two years of the Oregon Medicaid experiment is that having Medicaid coverage makes enrollees feel better without actually making them any healthier.[4]

Two years after Medicaid eligibility was expanded in Oregon, researchers found that new enrollees were significantly more likely to self-report that they are in good health. These same new enrollees were also less likely to report symptoms of depression than the control group that was not enrolled in the program.

The researchers also found that being enrolled in Medicaid increased an individual’s likelihood of being hospitalized by 30 percent; it increased the likelihood of using prescription drugs by 15 percent, and outpatient treatment use by 35 percent. In fact, contrary to the expectations of many, in the first years after this Medicaid expansion, Emergency Room use increased by 40 percent for new enrollees.

Despite more use of emergency and other health care services, the researchers found that Medicaid enrollment had no statistically significant impact on objective measures of health. The diagnosis and treatment rates for high blood pressure and cholesterol remained the same. Measures of blood pressure, cholesterol, and blood sugar remained the same. Smoking and obesity rates remained the same. Future risk of cardiovascular events was not reduced. But new enrollees were less likely to report symptoms of depression when screened by the researchers.

Although the study found that Medicaid enrollment had no discernable impact on employment or annual income for enrollees, it did, unsurprisingly, reduce the risk of catastrophic health care expenses by 50 percent. Additionally, it reduced the likelihood of unpaid medical bills being sent to collections agencies by 23 percent.

Analysis

Policy analysts have been fascinated by this study not just because it provided such a rare opportunity to test our understanding of a major federal program, but also because its results were so surprising for many. The accepted wisdom that having Medicaid will make people healthier and keep them out of expensive emergency departments has been proven to be inaccurate. The theory that Medicaid coverage makes it difficult to find health care providers is brought into question by the fact that health care utilization rates increased markedly. However, the perceived value of that care is also thrown into question by the fact that objective measures of health remained the same.

Conclusion

On balance, this study may have raised as many questions as it answered. The results of the Oregon Medicaid experiment shows that there are too many unknowns, too many unexpected results, and too much at stake to simply ignore inconvenient truths.

[1] http://www.nber.org/oregon/5.documents.html.

[2] http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=141.

[3] http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=141.

[4] http://www.nber.org/oregon/5.documents.html; http://www.nejm.org/doi/full/10.1056/NEJMsa1212321.