Are 4 in 10 Americans Really Underinsured?

In its September 2007 issue, Consumer Reports wades into America’s health care dialogue with a study that finds “four in ten Americans can’t depend on their health insurance.” The study is the most recent example of research trying to prove that many Americans are underinsured.

Incredibly, it begins by pointedly asking in its title, “Are you really covered?,” and never once seriously discusses what adequate and affordable coverage should look like for the families it surveys. Instead, the report labels some survey respondents as “well insured” and others as “underinsured” without discussion. The fine print does hint at some framework for defining the underinsured, but never explains it. Apparently we’re supposed to take their underinsured label at face value and trust that they know their stuff.

But there’s a lot of controversy over what actually constitutes underinsurance and some researchers who would like to steer the health care debate their way appear to be misusing the underinsured label to stigmatize different types of insurance policies they just don’t like. Typically, these are consumer-driven policies with higher cost-sharing, including higher deductibles, co-payments, and co-insurance.

Consider a 2005 study by the Commonwealth Fund and published in Health Affairs, entitled “Insured But Not Protected: How Many Adults Are Underinsured?” It estimates that 16 million Americans are underinsured and finds that the underinsured are more likely to forgo needed medical care, which, not surprisingly, leads the authors to question current marketplace trends toward more cost-sharing. While the study asks important questions, it’s strong rhetoric against cost-sharing demands a skeptic’s eye.

Upon closer inspection, at least one problem stands out. The study partly follows the traditional income-based approach academics take to define the underinsured based on the percent of out-of-pocket medical expenses relative to income (10 percent generally, and five percent for low-income families), but the study inflates the number of underinsured by adding another economic factor to the mix. The added factor defines anyone with a deductible equal to five percent or more of income as underinsured and it accounts for 24 percent of the study’s underinsured adults. Clearly, the factor casts too wide a net and assigns the uninsured label to many people who believe their coverage to be adequate. Why should a single adult making $50,000 be presumed underinsured if he has a deductible of $2,500 or more? Moreover, if the average family can spend 5.7 percent of its income on eating out, can’t that family also afford a deductible that might consume 5.7 percent of income?

Likewise, the Consumer Reports study uses the presence of a high deductible plan as a factor to identify the underinsured. However, in a more inflationary move, the study appears to ignore income as a factor. Thus, someone making $100,000with a $1,500 deductible might be considered underinsured.

Studies indentifying medical bankruptcies have similarly used questionable methodologies to infer inadequate coverage and criticize health plans with higher cost-sharing. Another Commonwealth Fund-supported study by Harvard professors David Himmelstein, Elizabeth Warren, and Steffie Woolhandler and Ohio University professor Deborah Thorne that found a large (about half) and rising (over 2200 percent since 1981) proportion of bankruptcies are being caused by medical debt, even for those with health insurance. Himmelstein et al. conclude that “many health insurance policies prove to be too skimpy” and that medical expenses bankrupt some people in high deductible plans long before they reach their deductible. They use this finding to support “broad reforms” more in line with health care delivery in Canada and western Europe.

But their findings have been roundly criticized. In a testimony before Congress last July, George Mason University School of Law professor Todd Zywicki stated that he has “been unable to locate any independent researcher unaffiliated with any of the authors of the study who has endorsed the methodology or findings of this study.”

Apparently, the study’s expansive definition of medical bankruptcy included debts caused by gambling and adoptions, and the threshold for medical debt as a cause was set at anything above $1,000 regardless of the extent other debt contributed to the bankruptcy. Thus, as Zwicki has pointed out, a bankruptcy caused by running up $50,000 on a Bloomingdale’s card would be considered a medical bankruptcy if medical debt accounted for at least $1,001.

All that is not to say that the underinsured label is bunk and intended only to mislead. Certainly some insured people become unable to afford their lifestyles, some incur substantial debt, and some avoid needed medical care when faced with unexpected illnesses.

But can we really identify the actual rate of underinsured Americans just the same as we report the rate of uninsured Americans? Some certainly think we can and should. Consumer Reports sure tried and so have a number of academics. (Lynn Blewett and Andrew Ward of the University of Minnesota and Timothy Beebe of the Mayo Clinic College of Medicine published an excellent review of the academic literature last year — well worth the read if you‘re interested in the topic.)

However, establishing standards that define the underinsured population will necessarily be subjective and, thus, heavily influenced by ideology and life experiences. To define underinsured, at least two questions must be answered. First, what is affordable coverage? And second, what is an adequate level of health benefits? But “affordable” and “adequate” are open to all sorts of personal interpretations.

It’s difficult to believe that experts or anyone else could ever settle on any one benchmark definition that measures underinsurance when ideology from both the left and the right seem to have such a large stake in the outcome.

— Peter J. Nelson is a Policy Fellow with Center of the American Experiment in Minneapolis. 

This commentary originally appeared on on August 31, 2007.
Permission to reprint in whole or in part is hereby granted.