• Expanding Medicaid saved lives

    This post is jointly authored by Aaron Carroll and Harold Pollack

    No matter how many times we refute the idea that Medicaid is bad for health, people keep on saying it. There’s so much evidence to the contrary. Recently, a number of states have used the “questionable” quality of Medicaid to buttress their arguments against the Medicaid expansion contained in the Affordable Care Act.

    Most of those decisions are fiscal. But we shouldn’t ignore the effect of them on patients themselves. There’s a paper in yesterday’s New England Journal of Medicineentitled “Mortality and Access to Care among Adults after State Medicaid Expansions.” It’s worth a read:

    BACKGROUND

    Several states have expanded Medicaid eligibility for adults in the past decade, and the Affordable Care Act allows states to expand Medicaid dramatically in 2014. Yet the effect of such changes on adults’ health remains unclear. We examined whether Medicaid expansions were associated with changes in mortality and other health-related measures.

    METHODS

    We compared three states that substantially expanded adult Medicaid eligibility since 2000 (New York, Maine, and Arizona) with neighboring states without expansions. The sample consisted of adults between the ages of 20 and 64 years who were observed 5 years before and after the expansions, from 1997 through 2007. The primary outcome was all-cause county-level mortality among 68,012 year- and county-specific observations in the Compressed Mortality File of the Centers for Disease Control and Prevention. Secondary outcomes were rates of insurance coverage, delayed care because of costs, and self-reported health among 169,124 persons in the Current Population Survey and 192,148 persons in the Behavioral Risk Factor Surveillance System.

     

    Basically, Sommers, Baicker, and Epstein examined county all-cause mortality rates of working-age adults in three states—Arizona, Maine, and New York–that expanded Medicaid eligibility for childless adults between 2000 and 2005. They compared trends in these states in the five years before and the five years after Medicaid expansion to trends found in nearby comparison states that didn’t’ expand eligibility. These comparison states therefore served as controls. The authors also examined the proportion of individuals reporting that they are in “excellent” or “good” health, as well as those who reported that they were unable to obtain needed care in the past year because of cost.

    Let’s acknowledge that this difference-in-difference design isn’t airtight. It’s not a randomized trial, and we can’t prove causality. But the study is still pretty compelling.  On all fronts, these authors found that Medicaid expansion was associated with reduced mortality rates, improved health, and improved access to needed care. In the preferred regression model, the authors found that annual mortality rates declined by 19.6 deaths per 100,000. That represents a relative reduction of 6.1% (p=0.001). These results imply that the Medicaid expansion prevented 2840 deaths per year in states that expanded Medicaid by about 500,000 adults. That’s not a small change.

    What’s especially impressive is the way this paper’s modest but important findings hang together from both a statistical and clinical perspective.  Mortality reductions were greatest in precisely the groups most likely to benefit from more generous Medicaid policies: Nonwhites, older adults, and those living in counties with more prevalent poverty. The authors found smaller but significant reductions among whites. They found no effects among persons under the age of 35, whose mortality rate is simply too small for such policies to make much of a difference.

    As the authors say:

    Our estimate of a 6.1% reduction in the relative risk of death among adults is similar to the 8.5% and 5.1% population-level reductions in infant and child mortality, respectively, as estimated in analyses of Medicaid expansions in the 1980s….

    A relative reduction of 6% in population mortality would be achieved if insurance reduced the individual risk of death by 30% and if the 1-year risk of death for new Medicaid enrollees was 1.9%… This degree of risk reduction is consistent with the Institute of Medicine’s estimate that health insurance may reduce adult mortality by 25%, though other researchers have estimated greater or much smaller effects of coverage. A baseline risk of death of 1.9% approximates the risk for a 50-year-old black man with diabetes or for all men between the ages of 35 and 49 years who are in self-reported poor health.

    The bottom line is that, according to these findings, state Medicaid programs need only cover 176 additional adults to avert one additional death every year. This allows for a crude but intriguing cost-effectiveness calculation. Annual Medicaid costs for childless adults are roughly $6,000. The cost per averted death (176*6,000) is thus about $1 million. This $1 million figure is easily within the range of acceptable costs based on common, widely-supported interventions to save lives and improve health.

    In 1994, Janet Currie and Jonathan Gruber performed a classic analysis of the health impacts and costs associated with earlier Medicaid expansions for infants and pregnant women. Largely through the financing of NICUs and related care, these expansions reduced infant mortality. Expressing the findings in year-2012 dollars, Currie and Gruber found that early, more targeted Medicaid expansions for relatively high-risk women and infants cost about $1.3 million per averted infant death. Later expansions to relatively lower-risk patients were more costly, with an estimate of about $6.5 million per averted death.

    Now, Sommers, Baicker, and Epstein add to our fund of knowledge by showing that expanded  Medicaid benefits for childless adults can also save lives. Moreover, this Medicaid expansion provides good public value, as it improves many measures of health in addition to preventing death.

    There’s been a wide, often misplaced debate over whether Medicaid helps or hurts its own recipients. We need to stop that. Medicaid helps. As states debate whether and how to expand coverage to millions of childless adults across America, they can focus on how much they’re willing to spend to save lives, but they shouldn’t deny that that’s what’s at stake.

    Share
    Comments closed
     
    • Thanks for this post, and this is a very interesting paper. I’ve only been able to read through it very quickly, but I noticed one interesting point. The authors explicitly state that “We examined three expansion states, and the results are largely driven by the largest (New York), so our results may not be generalizable to other states.” The other two expansion states were Arizona and Maine. So perhaps an alternative reading of this paper would be that the expansion was indeed effective in one state (New York), but was not effective in two others (Arizona and Maine). I don’t see where the authors break out the individual state effects in the paper, but as I said I’ve only been able to give it a quick read. However, I am concerned that the actual variability of outcomes might be quite different than what the aggregate summary suggests.

    • Good analysis. Just to make explicit what is implied by this article: looking at benefits in mortality does not account for the benefits of improved quality of life. Improved mortality rates suggest there will also be significant improvements in quality of life through improved access to care.