Medical bankruptcy in the USA: the data behind the headlines

As President Obama continues his push for major healthcare insurance reform in the US, a recent study in the American Journal of Medicine on medical bankruptcy is grabbing headlines.

The headlines read something akin to, ‘Medical costs contribute to more than 60% of US bankruptcies’ – a startling statistic, indeed. The study, published by Dr. David Himmelstein, a primary care physician and professor of medicine at Harvard Medical School, and 3 other authors, does report that illness or medical bills contributed to 62.1% of all bankruptcies in 2007 in the US. Three fourths of those with ‘medical bankruptcy’ had health insurance and 92% had medical debts over US$5000 or more than 10% of pretax family income.

As always, the devil is in the detail. The headlines do not tell the entire story from the study.

Disclosure of potential conflict of interest — It first should be noted that two of the four study authors are co-founders of Physicians for a National Health Program (PNHP), a “single-issue organization [claiming more than 16,000 members] advocating a universal, comprehensive single-payer national health program [in the USA].” Moreover, another co-author (Elizabeth Warren, JD) is the chair of the Congressional Oversight Panel for the Troubled Assets Relief Program (TARP) – the US government’s financial bailout program. At least some of the authors have an agenda to promote.

Defining medical bankruptcy — The authors defined persons as being medically bankrupt if any of the following applied:

• illness or medical bills were cited as a reason for bankruptcy (although not necessarily the reason);
• medical bills totaled more than US$5000 or more than 10% of pretax family income;
• the debtor(s) lost at least 2 weeks of work-related income due to illness/injury; or,
• the debtor(s) mortgaged a home to pay medical bills.

With the exception of mortgaging a home to pay medical bills (5.7% of all bankruptcies who were recent homeowners), these are rather loose definitions of medical bankruptcy. They also do not address the role of personal responsibility in saving for medical or other emergencies. As Dr Himmelstein readily concedes (in a phone interview), ‘Well, most Americans don’t have much in the way of savings.’

Distinguishing medical bills from cost of illness — Such liberal bankruptcy definitions also blur the lines between the truly debilitating effects of underinsurance or lack of insurance, and the simple cost of illness due to lost wages (which health insurance does not address), although 95% of lost-income debtors also had ‘high medical bills’ (presumably more than the US$5000/10% of income threshold) and almost all health insurance in the US is obtained through employment. In this study, 70% had health insurance at the time of bankruptcy filing.

In the published article, the data are broken down by cause of bankruptcy, but these specific figures and analyses are rarely reported in the general news media. In fact, the data are used by the media and Dr Himmelstein to promote a single-payer health insurance system in the US. As Dr Himmelstein noted in our interview, ‘Other people can reach different conclusions [based on the data]…You can quibble around the edges. Is it half of all bankruptcies that are caused by medical problems or is it two thirds? …We think 62% is probably the best estimate, but other people are free to use our data to reach different conclusions. But even if it’s only 50%, that’s still an indicator of a huge problem.’

Health economists should be aware of these studies as select data are quoted and referenced in the news media and by politicians. Irrespective of whether medical bankruptcies are increasing or whether single-payer health insurance is the correct path for Americans, the discussion should be clear about the data supporting a particular argument.