The Patient Protection and Affordable Care Act (PPACA, a.k.a. Obamacare) has been an utter mess. From passage in 2010 with procedural gimmicks to implementation in 2013 with unworkable software, from the loss of doctors and health plans millions wanted to keep to escalating premiums and insurers dropping out of the market, Obamacare has fallen short of nearly every conceivable goal of health-care reform.
There’s one single exception: Obamacare has dramatically expanded health insurance coverage. This single remaining reason explains why it retains the support of progressives and a significant chunk of the electorate. All other considerations are secondary, if not irrelevant. More people have health insurance, so more people are benefitting from improved health outcomes and access to care.
There is only one simple flaw in this reasoning. It does not appear to be true.
Does Having Insurance Mean Better Health Care?
Public health experts have long accepted the inverse correlation between increased insurance coverage and reduced mortality. Multiple studies have been published showing this connection.
In December 2009, the American Journal of Public Health published an important study. Dr. Andrew Wilper and five colleagues from the Cambridge Health Alliance updated a 1993 study using data from the Third National Health and Nutrition Examination Survey. It found that private health care insurance was associated in 2005 with a 40 percent mortality risk reduction among the pre-Medicare U.S. adult population (age 18 to 64). This association was robust after controlling for numerous co-variables.
To put numbers on it, reducing the risk of death by 40 percent among the 2005 uninsured population would eliminate up to 45,000 premature deaths of adult Americans. This study had a huge effect on the political debate surrounding Obamacare. A search of “Harvard study 45,000 deaths” reveals more than 500,000 hits.
PPACA advocates frequently repeated this number, and exerted considerable pressure on lawmakers during debate and adoption of the bill. Rightly so. What good-hearted American would be unwilling to expend a great deal of effort to save 45,000 fellow citizens from premature death? So Obamacare became the law of the land in March 2010.
Congress Duly Expands Insurance Coverage
One lovely feature of science is that it can reliably predict outcomes of actions. So, as of 2009, we had a firm prediction of the effects to expect on U.S. adult mortality if taxpayers expanded health-care insurance coverage to more people who couldn’t or wouldn’t buy it. In 2010 Congress passed Obamacare, which expanded insurance coverage.
While the PPACA had a few small early coverage expansion programs, activating the insurance exchanges in October 2013 implemented the primary Medicaid and private health care insurance coverage extensions. Despite well-known problems with the exchange roll-outs, by the first quarter of 2014, millions of Americans had new health insurance coverage.
The coverage expansion increased in subsequent years, with more than 20 million people enrolled in Medicaid or exchange plans by the end of 2015 open enrollment. While some of these people had insurance before 2014, estimates indicate that more than 15 million Americans were newly insured in 2015.
Fifteen million newly insured Americans is a big change in the U.S. insured population and, using Wilper’s numbers, population-level mortality statistics should clearly convey a reduction. Specifically, using Wilper’s 40 percent mortality risk reduction with 15 million newly insured people means that approximately 21,000 fewer adult Americans should die in 2015 relative to the pre-Obamacare status quo.
How Is that ‘Settled Science’ Working Out for Americans?
The Centers for Disease Control collects U.S. mortality statistics and publishes them in a database called WONDER. The database is indeed a statistical wonder, allowing researchers to slice and dice U.S. mortality data into segments by age, gender, location, year, cause of morbidity, and many additional criteria.
With WONDER, it is a short exercise to attempt to confirm Wilper’s predictions. Examining U.S. adult mortality in the decade prior to Obamacare’s insurance expansion (2004-2013), the all-cause mean death rate for ages 15 to 64 is 310.4 people per 100,000. The rate is fairly steady over the decade, with a low of 306.8, a high of 313.5, and a standard deviation of 2.2. If extending taxpayer-sponsored insurance to 15 million people since 2013 has resulted in 21,000 fewer annual deaths, then the mean death rate should decrease from 310.4 to approximately 300.
Returning to the WONDER database for 2014-15 numbers, one finds the mean death rate is … 320.4. Well, that is unexpected. Since Obamacare provisions extended insurance coverage, the death rate has substantially increased, by more than 20,000 deaths per year.
A correlation does not prove causation, of course, and since we believe health insurance reduces mortality, there must be a coincident event causing the spike in deaths since 2014. And there is an apparent scapegoat. An opioid crisis has gripped the United States since Obamacare insurance expansion was implemented.
Opioids have caused thousands of early deaths, enough to distort mortality statistics in adult Americans, and the crisis worsened noticeably in 2014-2015. Assuming the opioid crisis is independent of Obamacare insurance expansion (for analysis purposes only, since some work has suggested these two phenomena may be causally linked) may eliminate the excess deaths and show the expected reduced mortality from health care insurance.
Fortunately, WONDER allows researchers to separate causes of morbidity, so it is a simple matter to repeat the analysis, excluding drug-related and other external causes of death, and clear up the confusion about the increased U.S. mortality.
What happens when we calculate the death rate after excluding all external causes of morbidity (ICD-10 codes for deaths caused by drugs, alcohol, assault, suicide, and accidents—in short, anything that is not due to an internal illness)? For the decade 2004-2013, the death rate is 247.4 people per 100,000 population. It is more stable than the all-cause death rate, with a low of 244.7, a high of 249.9, and a standard deviation of 1.7.
With Obamacare extending insurance to 15 million more people, this death rate should fall to 238 per 100,000. The 2014-15 data show the actual reported death rate among U.S. adults, excluding external causes, is … 252.9.
This is equivalent to an excess 11,000 annual U.S. adult deaths relative to the pre-Obamacare steady state trends, and more than 32,000 annual deaths greater than predicted by academic studies quantifying health benefits from improved insurance coverage. It is more than three standard deviations higher than the pre-Obamacare mean mortality, and it has persisted for the two full years, 2014—15, for which mortality data have been compiled. It is not a statistical aberration. Figure 1 shows the data. Whoa.
Figure 1: Time Series of U.S. Adult Mortality, Excluding External Causes
How to Account for the Death Spike?
In short, we know much less than we think. We know Obamacare became law, and millions of individuals who were previously uninsured gained new insurance policies, through subsidized private insurance or through Medicaid. We know that academic studies predicted large reductions in U.S. adult mortality following the insurance expansion.
We know that the same year Obamacare’s insurance expansion provisions took effect, there was a pronounced, and statistically significant, surge in U.S. adult mortality. We know the surge in mortality remains after removing drug-related deaths, and other external morbidity causes, from the statistics. That is all we know. The rest is speculation. But it is fascinating speculation.
Has Obamacare, or some of the secondary effects of Obamacare, actually caused the negative impact in U.S. adult mortality so evident in the statistics? Is the improvement in public health that was assured turned out simply to be another false Obamacare promise, like being able to keep our doctors and health plans, or reducing our health costs?
If any causal relationships are discovered between Obamacare and mortality, there will be profound policy implications. As Sen. Tom Cotton has said, the objective of further health reform is “to help those who were hurt by Obamacare while not hurting those who were helped by it.”
Assuming for the sake of argument that the PPACA has negatively affected mortality, meeting Cotton’s objective is not likely to be as simple as repealing Obamacare, or even the specific provisions correlated to increased mortality. Most of the changes wrought by the PPACA in the insurance and medical markets are irreversible; a hospital is not going to de-merge if Obamacare is repealed.
Nevertheless, examining Obamacare’s relation to U.S. adult mortality will prove invaluable in designing new policy ideas to improve health outcomes and reduce mortality. Such policy ideas are likely to employ free-market incentives, and be much more patient-oriented than the PPACA. Regardless of the specific fate of Obamacare, that is enough.