Yesterday John Graham of the Pacific Research Institute outlined a number of problems with the Commonwealth Fund’s just released study that updates their 2003 survey indentifying the underinsured. Here's another issue: Using a high deductible as an indicator exaggerates the number of underinsured. (I touched on this point in an earlier post, but it's more problematic than I originally thought.)

The study’s methodology classifies someone as underinsured if they meet one of three indicators of financial exposure, including: (1) Out-of-pocket medical costs equal to or greater than 10 percent of income; (2) medical costs equal to or greater than 5 percent of income for those with low-incomes under 200 percent of the federal poverty guidelines; and (3) a deductible equal to or greater than 5 percent of income.

Lynn Blewett and Andrew Ward of the University of Minnesota and Timothy Beebe of the Mayo Clinic College of Medicine recently reviewed the academic literature related to the concept of underinsurance and found “great variation in the measurement and subsequent estimates” of what constitutes underinsured. Commonwealth Fund studies were the only reviewed studies that used a high deductible as an indicator of underinsurance. The deductible indicator hopes to measure "potential risk" of serious financial hardship.

Using the deductible as a measure of potential risk raises three issues that exaggerate the number of underinsured. First, not everyone with a high deductible has the same risk of reaching the deductible because of the wide variation in health risks among people and so the deductible, by itself, is a poor measure of potential risk. For instance, a healthy 23-year old with a high deductible will be at a much lower risk than a similarly financially situated 54-year old with high blood pressure. By ignoring this distinction, the high deductible indicator likely includes people with incredibly low risks of ever reaching the deductible. 

Second, as the authors admit, some employers contribute funds to savings accounts matched with HDHPs to offset the deductible, which can eliminate any financial risk related to the deductible. Because the study fails to remove people with employer funding, it overestimates the number of people at financial risk from high deductibles. 

The authors dismiss the problem and claim that “recent studies find that most employers do not contribute.”  But the study they reference makes the opposite point: most employers do contribute something. The authors must have mistakenly only considered employees with health savings account (HSA)-qualified HDHPs. It’s true that, according to their referenced study, only one-third of employers offering HSA-qualified HDHPs contribute to the savings accounts. However, HSA-qualified HDHPs only represent half of the HDHPs that enroll employees.  Another type of HDHP, offered in combination with a Health Reimbursement Arrangement (HRA), enrolls the other half and, according to the study, all employers contribute to HRA accounts when they offer them.  Consequently, the study cited by the authors actually finds that 66 percent of employers that offer HDHPs linked to an HSA or HRA contribute to their employees’ accounts.

Third, without using some top-end income threshold, the methodology draws in a number of high-income people that are unlikely to be at risk.  In the study, 9 percent of the underinsured have incomes between $60,000 and $99,999 and another 7 percent have incomes higher than $100,000.  Over on quarter, or 26 percent of those added to the ranks of the underinsured by a high deductible had incomes higher than $60,000. Consider too that a 5 percent deductible would include someone with a $7,500 deductible who makes $150,000.  It’s hard to believe that anyone with an income over $100,000 is truly underinsured.

Despite these problems with the study’s methodology, we should not lose sight of the fact that many people struggle to afford medical care even when they have insurance. Some people are, in fact, underinsured and this group’s set of issues deserve serious attention. As a group, they will benefit most from policies that tear down barriers to affordable health care.