Medicaid Improves Health of People Who Are Low Income and Uninsured, RWJF Investigator Finds
Amy N. Finkelstein, Ph.D., M.Phil., is a 2003 Robert Wood Johnson Foundation Investigator Award in Health Policy Research recipient and a professor of economics at the Massachusetts Institute of Technology (MIT). Finkelstein gave the following interview to the Human Capital Blog as part of our ongoing Voices from the Field series. Read more about her research.
Human Capital Blog: How did you come up with the idea for this study?
Amy Finkelstein: In early 2008 I heard a story on the radio about how the state of Oregon was conducting a lottery for access to Medicaid. It was one of those once-in-a-lifetime moments in which I thought "I must drop everything and look into this right away!" and so that's what I did. This was, literally, the chance of a lifetime: the opportunity to bring the gold standard of medical and scientific research—a randomized controlled trial—to an important social science and policy question.
HCB: Why hasn’t this kind of randomized control trial study of Medicaid been done before?
Finkelstein: There have been two major impediments to doing a randomized control trial of the effects of being uninsured relative to having insurance. The first is ethical concerns regarding doing such a randomization for research purposes. In our case this wasn't an issue because the state of Oregon had decided that a lottery was the fairest way to allocate a limited number of Medicaid slots. It was doing it for policy, not research purposes. Fortunately however the state also saw the enormous potential to learn from this opportunity and generously collaborated with researchers at Harvard, MIT, Providence Health & Services, and the National Bureau of Economic Research to make this possible.
The second major impediment of course was funding. Here we were extremely fortunate to have such generous funders, including of course the Robert Wood Johnson Foundation!
HCB: What were some of the most surprising findings?
Finkelstein: I’m not sure I can say what was "surprising" per se, because I didn't have strong opinions before the study of what we would find, which is of course exactly when doing such a study is most valuable.
I will say that before we produced any results, there seemed to be genuine uncertainty in the public policy discourse about whether Medicaid would make much or any difference in the lives of low income uninsured. The "glass half full" version of this story I heard is that the existing social safety net of public health clinics, charitable care, and uncompensated care functions reasonably well for the low income uninsured. The "glass half empty" version I would hear is that having Medicaid didn't mean much, because it could be very hard to find a physician who would accept Medicaid.
Our results indicate that in fact Medicaid does make a difference to the people it covers. Using a randomized controlled trial, we find that people who are enrolled in Medicaid use more health care, have less out of pocket medical expenses and medical debt, and report themselves to be in better health and overall well-being than otherwise similar individuals who are not on Medicaid.
Our results suggest that expanding Medicaid has real benefits. They also indicate that expanding Medicaid has real costs, both in terms of increased overall health expenditures and increased government spending.
HCB: How do you think your study will impact the policy debate on Medicaid?
Finkelstein: I think, and hope, that objective, rigorous scientific evidence such as that provided by our study can (and will) contribute to informed public policy discourse and decision making.
HCB: What kind of reaction have you gotten so far?
Finkelstein: We've been very gratified by all the interest and attention our study has garnered. I think it reflects the appreciation of the strength of the evidence from randomized control trials, and suggests that there is great value to doing more of these in social science and policy research.
HCB: Do you plan to follow-up on this study and how so?
Finkelstein: Most certainly! My co- principal investigator Kate Baicker has led the effort to conduct approximately 12,000 in person in-depth interviews and physical health exams in the second year of the experiment (the current paper reports results only from the first year). This will provide the opportunity to study the effects of Medicaid on physiological measures of health such as blood pressure, weight, cholesterol and blood sugar. The in-depth interviews also will provide a richer picture on such topics as medication use and disease management.