For the past few months, I keep reading and hearing about medical bankruptcies. Newspapers, NPR, politicians, even President Obama. First it was “half of all bankruptcies are caused by medical spending.” Then it was seventy percent. And it didn’t matter if you had health insurance – medical bankruptcy was always just around the corner. The Democrats even want to cap out of pocket payments in private health insurance so that Americans don’t go bankrupt. Some think we need a single payer system to eliminate the threat of medical bankruptcy.
Stuff and nonsense, that’s what this is. Of all the myths that have permeated the health reform debate, this may be the worst. And my new research shows the dangers of myths that originate in academia.
Some background: A few years ago, a lawyer and a couple of doctors from Harvard were dabbling in empirical social science research and managed to publish a rather poorly designed and executed study of bankruptcies. Confusing cause and effect – well, not really establishing either – they concluded that half of all bankruptcies were “medical bankruptcies” and that health insurance didn’t prevent them. The authors, notorious supporters of a single payer system, have since updated their work and increased the medical bankruptcy rate to 70 percent.
Many social science researchers have repudiated these studies, and for good reason. The Harvard researchers found that folks who go bankrupt spent thousands of dollars annually on medical care and used this as the basis for their policy prescriptions. But they cannot determine how many of these folks would have gone bankrupt if they had spent nothing on medical care. Perhaps all of them, perhaps none. There is no way to know. And thus, there is no way to know how many bankruptcies would be averted if we increased health insurance coverage, capped out of pocket payments, or moved to a single payer system.
I was one of the leading critics of the Harvard study. Working with Michael Millenson, an adjunct professor at Kellogg, I published a critique in the same journal that the original Harvard study appeared. (Full disclosure: In the process of writing this critique, Michael and I accepted small honoraria from an insurance industry lobbying group. The group paid for our time but not our opinions. We insisted that our work be peer reviewed and we did not allow the lobbying group to edit our work.) It seems that Chicken Little got all the attention. The highly questionable findings of the Harvard study have become the stuff of legend. The critiques have been all but forgotten.
It is well enough to criticize someone else’s research. It is far better to tackle the same study with better methods and set the record straight. This is what I have been trying to do. A new study by Keziah Cook, Andrew Sfekas and myself, funded by the independent Robert Wood Johnson Foundation, uses the methods that I outlined in my critique of the Harvard study. Our study has just been published by the journal Health Services Research. Examining household assets (comprehensive data on bankruptcies that would be usable for this type of study were not available), we find that it really does stink to be uninsured and get sick. Uninsured Americans who have major illnesses stand to lose 40-50 percent of their life savings. This is reason enough to try to expand coverage to all Americans. Of equal importance, we also find that insured Americans who have a major illness do not dip into their savings. In other words, private health insurance is doing its job and protecting Americans from financial catastrophe. If our goal is to prevent bankruptcies, we don’t need a single payer system. And we don’t need to cap copayments. In fact, capping copayments will eliminate one of the most effective tools we have for containing costs.
I realize that the concept of medical bankruptcies is captivating and my research confirms that the uninsured can face severe financial hardship when illness strikes. But the Harvard studies are so poorly designed that it is impossible to tell from their work just how serious the problem is, and the conclusion that private health insurance does not protect against bankruptcy appears to be totally misguided. Even worse, the Harvard studies are leading to bad policy. Other than outright fraud, I cannot think of a worse thing to say about academic research.
I have said it before: It is vitally important that academics get the numbers right.