A recent study reported that people newly enrolled in the Oregon Medicaid program through a unique lottery system experienced better access to care, increased detection of depression and diabetes, and reduced out-of-pocket costs, compared with a control group who remained uninsured. However, the study did not find significant improvements in measures of health such as blood pressure, cholesterol, or blood sugar levels. Some have said these findings suggest that Medicaid—and health insurance coverage generally—may not appreciably improve health. This conclusion is incorrect, in our view. To understand why, one need only place the study within the context of the literature on the association between insurance coverage and health.
That literature leaves little doubt that access to coverage is associated with an array of beneficial effects: having a regular doctor; receiving timely preventive care services; better management of chronic health conditions; improved health status, particularly among people with chronic health problems; greater workforce participation; and longer life expectancy. Many studies have focused on the effects of Medicaid coverage in particular. For instance, a classic study of California’s termination of the state’s Medicaid program in 1982 found deterioration in both access to health care and health status among people who lost coverage. A more recent study by Benjamin Sommers, M.D., and colleagues compared the effects of Medicaid expansions in three states with neighboring states that did not expand their programs. It found that the expansions were associated with significant reductions in mortality and increases in self-reported health status.
How then do we explain the results of the Oregon Medicaid study? First, the study examined the effects of coverage over a relatively short period of time: 17 months on average. In contrast, Sommers and colleagues examined the effects of state Medicaid expansions over a 10-year period. Similarly, several studies have used the longitudinal Health and Retirement Survey to examine the effects of gaining Medicare at age 65 among people who were insured continuously prior to that time, compared with those who spent time uninsured before enrolling. Most of these studies found gains in health over multiple years among people who had been previously uninsured, compared with those who had always had coverage.
Second, the benefits of health insurance may take the form of changes in trends in health status, rather than absolute improvements, even over longer periods. For example, the study by J. Michael McWilliams, M.D., and colleagues at Harvard Medical School and Brigham and Women's Hospital found that prior to age 65, health status among older adults who were intermittently uninsured declined at a faster rate than among the continuously insured, and was significantly worse at age 65. However, this trend changed after age 65: previously uninsured adults who gained Medicare reported stabilization in their health status.
Third, the sample of adults examined in the Oregon study may have been too healthy and the number of outcome measures too small to discern immediate health effects from gaining coverage, an issue noted by the authors. Only 16 percent of study participants in the control group had high blood pressure, 14 percent had high cholesterol, and 5 percent had elevated blood sugar levels. In contrast, the McWilliams study examined a larger set of measures and focused on people with diagnoses of hypertension, heart disease, stroke, or diabetes. More than three of five (61%) adults in the study cohort had at least one of these diagnoses before age 65. The gains in health among adults in the study were concentrated among those with these conditions.
The results from the Oregon study do not contradict previous research, but do show that our expectations concerning the immediate effects of gaining health coverage should be tempered, especially for relatively healthy populations. The Congressional Budget Office projects that 14 million people will become newly insured next year under the Affordable Care Act and 25 million will gain coverage by 2016. These historic gains in health insurance coverage are likely to have variable effects on the health of newly insured people over the next several years, with the sickest and the oldest realizing more immediate changes in health status. But even those changes may not show up as dramatic improvements in health, but rather as the preservation of good health, or a slowing of its deterioration. Certainly, these are outcomes that society should highly value.