Secondary Logo

Journal Logo

Invited Commentaries

Why Health Insurance Matters—and Why Research Evidence Should Too

Sommers, Benjamin D. MD, PhD

Author Information
doi: 10.1097/ACM.0000000000001723
  • Free


Since the Affordable Care Act (ACA) was signed into law in 2010, it has been a lightning rod for controversy. Conservative policy makers have decried it as a costly and ineffective government intrusion into health care, while some liberals argue that it did too little to expand coverage and too much to enrich private health insurers. Now, with the fate of the ACA in serious doubt, taking stock of what we know about the effects of the law thus far—and more generally, the impact of expanding health insurance coverage to previously uninsured patients—is critical to informed policy making.

Over the past six years, my colleagues and I have been conducting a range of studies designed to provide evidence that would improve our understanding of the impacts of health insurance on patients. I care about this both as a health policy researcher and as a primary care physician. As any clinician knows, there are so many factors that affect our patients’ health outside what occurs in the office. How long did my patients wait before coming in to be seen? Can they pay for the medications I prescribe? Can they see the specialists I refer them to? And do they get better? Our work has aimed to answer some of these questions at the population level, and we have identified several key lessons from this work.

How Does Health Insurance Affect Patients?

Lesson 1: Coverage can be a matter of life or death

The first lesson is that health insurance coverage matters to patients’ lives. Some of the most useful evidence in support of this observation comes from expansions in health insurance that occurred prior to the ACA. Studying expansions of Medicaid in several states in the early 2000s by comparing them to neighboring states without expansions, we found large reductions in the uninsured rate, improved self-reported health, and a drop in mortality of 6% over the following five years.1 These changes were largest in lower-income areas and among racial and ethnic minorities. We also studied Massachusetts’s 2006 health reform, the model for the ACA, and found that the coverage expansion led to a significant reduction in mortality for the state compared with what was happening in demographically similar counties outside the state.2 Most of the deaths prevented were due to causes potentially more amenable to health care, such as cancer, heart disease, and infections. Overall, we concluded that one life was saved for every 830 adults who gained coverage. Most recently, I examined the costs of Medicaid expansion in relation to these mortality changes, and found that the increase in spending was a good investment compared with how much we as a society spend on other public policies that affect survival.3

Now, with Congress debating a potential repeal of the ACA, there has been renewed interest in this work as a gauge for how many deaths might occur if the law is repealed. Policy makers and analysts have proposed various extrapolations from these studies, including the White House Council of Economic Advisors, which estimated that the law could be saving as many as 24,000 lives a year.4 This is a challenging calculation to make with precision, and no one can know the exact numbers for sure, but our research indicates that these are indeed life and death decisions. Taking coverage from people will likely lead some to forego medical care that could have saved their lives.

Lesson 2: The ACA has succeeded in expanding coverage and access to care

These studies of pre-ACA expansions in coverage are illuminating, but what lessons can we draw about the ACA’s effects in particular? Using a variety of data sources, our team and other researchers have documented that not only has the ACA lowered the uninsured rate to its lowest level in U.S. history but that coverage has also produced meaningful benefits for patients. The earliest ACA studies examined the “dependent coverage” provision that allowed young adults to remain on their parents’ plans until age 26 starting in 2010. Studies show that this policy was more successful than even the law’s drafters had hoped, with two million to three million more adults covered. The coverage helped young adults better afford their care, reduced their use of nonurgent emergency department care,5 and improved their perceived physical and mental health.6

Next came the ACA’s 2014 expansions. Medicaid expansion in the roughly 30 participating states and new subsidized Marketplace coverage led to about 20 million more Americans with insurance. From 2010 to 2014, as policy makers scrambled to implement the law, researchers scrambled to figure out how to study it—and, in particular, how to study it rigorously and quickly. Standard data sources from the federal government sometimes take a year or more to become available. For a policy as large and consequential as the ACA, we needed results faster. Working with colleagues at the U.S. Department of Health & Human Services, we obtained and evaluated a new data source—the Gallup Healthways Well-Being Index. With it, we published some of the first journal articles showing—within months of real time—how the law was increasing coverage and also improving trends in rates of having a primary care doctor, ease of access to prescription medications, affordability of care, and self-reported health.7

Following up on this, my colleagues at Harvard and I then conducted our own rapid-turnaround scientific survey to evaluate the Medicaid expansion in several southern states. This work showed that in Kentucky’s traditional Medicaid expansion and Arkansas’s private insurance expansion, low-income adults saw major improvements in health care, compared with those in Texas, which did not expand. Adults in the two expansion states reported more primary care visits, better chronic disease care, more preventive care, less ER use, and again—better self-reported health.8

Given this pattern of findings, some have questioned whether policy makers and society in general should care about changes in self-reported health. I would argue that we should for two reasons. First, self-reported health turns out to be a strong predictor of survival; people who say they are in poor health die younger.9 Second, subjective well-being is a key part of health. If you’re a doctor and you don’t care whether your patients feel better, you should quit. We should hold our policy makers to the same standard.

Lesson 3: Challenges in coverage and access remain

The final lesson from our research is that all is not perfect. Yes, health insurance matters, and the ACA has helped expand coverage and improve access to care. But as many as 30 million Americans are still uninsured, and millions more find themselves switching between various types of coverage each year. Some of this is related to the ACA, but much of it is due to the United States’ underlying patchwork health insurance system. This switching in coverage—sometimes called churning—has real impacts on patients. In one recent study, we found that roughly one in four low-income adults experienced a change in coverage each year. Although this is lower than many had predicted would occur under the ACA, these changes in coverage were harmful—patients reported reduced continuity with providers, disruptions in medication regimens, and negative effects on overall quality of care and health.10 In part, this research has been useful to states and federal policy makers as they try to streamline some of the transitions in coverage. But now, with ACA repeal on the table, our findings have another implication. While taking coverage away from people would clearly be quite harmful, even transient disruptions in coverage from dismantling parts of the law could also cause significant distress.

An Uncertain Future—and the Need for Evidence-Based Policy

Despite this body of evidence, the political future of the ACA’s coverage expansion remains uncertain. In fact, the rhetoric in the debate over the ACA raises fundamental questions about what role research evidence plays at all. Despite hundreds of high-quality studies by researchers across the country probing the law’s successes and shortcomings, many people still don’t know the basic facts about the law. For instance, one recent survey by National Public Radio found that only 49% of Americans knew that the ACA had reduced the number of Americans without health insurance; 27% didn’t know or said it was unchanged, and a stunning 24% thought the uninsured rate had gone up.11 Meanwhile, some physicians, politicians, and pundits continue to argue that the ACA expanded coverage but did not meaningfully improve access to care. But these claims are flatly contradicted by the research evidence. That so many might still not know the basic facts about this law is, at least in part, the fault of researchers like me in health policy and academic medicine. We certainly are not alone in this—there are other factors too, including some media outlets’ desire for an evenhanded cross-fire debate rather than a focus on facts, and rhetoric from some politicians that ranges from misleading to simply wrong.

But already there is some indication that the facts are starting to sink in and affect the policy debate. The steady drumbeat for repeal among congressional Republicans has met the reality of millions of Americans who have gained coverage under the ACA and are benefiting from it. Republican governors from expansion states have stepped forward to say that a repeal and large cuts in federal funding for Medicaid would not be good policy.12 As of this writing, the debate continues and is unsettled. For those of us in academic medicine who believe evidence should guide what we do for our patients, now is the time to bring a similar mentality to the policy discussions that will affect our patients just as surely as the next prescription we write or the next test we order. The future of health care is too important not to have the major policy decisions in it driven by evidence.


1. Sommers BD, Baicker K, Epstein AM. Mortality and access to care among adults after state Medicaid expansions. N Engl J Med. 2012;367:10251034.
2. Sommers BD, Long SK, Baicker K. Changes in mortality after Massachusetts health care reform: A quasi-experimental study. Ann Intern Med. 2014;160:585593.
3. Sommers BD. State Medicaid expansions and mortality, revisited: A cost–benefit analysis. Am J Health Econ. In press.
4. The Economic Record of the Obama Administration: Reforming the Health Care System. 2016.Washington, DC: White House Council of Economics Advisers.
5. Akosa Antwi Y, Moriya AS, Simon K, Sommers BD. Changes in emergency department use among young adults after the Patient Protection and Affordable Care Act’s dependent coverage provision. Ann Emerg Med. 2015;65:664672.e2.
6. Chua KP, Sommers BD. Changes in health and medical spending among young adults under health reform. JAMA. 2014;311:24372439.
7. Sommers BD, Gunja MZ, Finegold K, Musco T. Changes in self-reported insurance coverage, access to care, and health under the Affordable Care Act. JAMA. 2015;314:366374.
8. Sommers BD, Blendon RJ, Orav EJ, Epstein AM. Changes in utilization and health among low-income adults after Medicaid expansion or expanded private insurance. JAMA Intern Med. 2016;176:15011509.
9. DeSalvo KB, Bloser N, Reynolds K, He J, Muntner P. Mortality prediction with a single general self-rated health question. A meta-analysis. J Gen Intern Med. 2006;21:267275.
10. Sommers BD, Gourevitch R, Maylone B, Blendon RJ, Epstein AM. Insurance churning rates for low-income adults under health reform: Lower than expected but still harmful for many. Health Aff (Millwood). 2016;35:18161824.
11. Ipsos Public Affairs. NPR survey of healthcare knowledge and perception. Published January 9, 2017. Accessed March 14, 2017.
12. Sommers BD, Epstein AM. Red-state Medicaid expansions—Achilles’ heel of ACA repeal? N Engl J Med. 2017;376:e7.
Copyright © 2017 by the Association of American Medical Colleges