Health Insurance and Statistics
This month the U.S. House of Representatives Committee on Energy and Commerce presented a report offering the findings of an investigation into the practice of denying individual health insurance coverage to people with preexisting medical conditions. The committee, chaired by Representative Henry Waxman, the Democratic Congressman from of Los Angeles, reported:
From 2007 through 2009, the four largest for-profit health insurance companies…refused to issue health insurance coverage to more than 651,000 people based on their prior medical history. On average, the four companies denied coverage to one out of every seven applicants based on a preexisting condition.
The Wall Street Journal ran an article on the report titled “Insurers Denied Coverage to 1 in 7.” It came as no surprise to me that the congressional committee published a report critical of private health care insurance and supportive of the national health insurance overhaul passed by Congress in 2009 and signed into law this year, but I it was a bit surprised when I compared The Wall Street Journal to the report and found that the reporter, Janet Adamy, had repeated the committee’s claims without challenging them.
For example, the report equated “people” with “applicants,” but I am not convinced that the two were the same. Did anyone check with the committee to make sure there was no duplication of applicants between the four health insurance companies? The period of the study—three years—was certainly sufficient time for an individual denied health insurance coverage by one insurer to apply again and be denied by a different insurer.
It is possible—I would say likely—that 651,000 denials of coverage do not represent denial of coverage to 651,000 individuals. Since the committee studied four health insurance companies, there is a possibility that the number of denials is off by as much as a factor of four, if each person denied by one company reapplied to all of the others. Instead of 651,000 people denied coverage over three years, there might be as few as 162,750.
The possibility also exists that individuals denied by one company might have applied to the same company a second time in the three years. I have seen this happen in my own business: A person denied coverage because of weight, high blood pressure, high cholesterol, or other factors might make lifestyle changes and reapply to the same health insurance provider, sometimes for the same plan. That person could be denied a second time for one of the earlier factors or for something completely different. Any duplication of applicants decreases the final tally.
There is another possibility, along the lines of the example I just gave: The person who made lifestyle changes could be accepted upon reapplication. Their denial still shows up among the 651,000, but that individual is actually covered now.
Any journalist should be wary of partisan committee chairs wielding statistics. As the great American humorist and novelist Mark Twain observed, “There are three kinds of lies: lies, damned lies, and statistics.”
Another questionable aspect of the report relates to the time frame. Why were the number of denials grouped over three years? Not only does the time span allow for the possibility of duplication, but it also skews the magnitude of the problem vis a vis the size of the general population. In other words, 651,000 people represents two-tenths of one percent of the population. However, that 651,000 is an accumulation over three years. On average, fewer than 220,000 people were denied coverage in any one year, according to the committee’s findings.
If you accept the possibility of duplication of applicants by a factor of four, then only 54,250 people were denied coverage per year—a truly miniscule number in a country with 310 million people, working out to less than two-one hundredths of one percent of the population.
It is clear to me that the authors of the congressional study grouped the three years to inflate the number. 220,000 is just not as impressive sounding as 651,000. What was the logical reason to group three years? Why not two years? Or four? Or fifteen?
There can be no doubt that the committee’s report, coming just three weeks before a national election, was politically motivated to make the highly unpopular health insurance reform legislation more appealing. The authors of the report wrote:
The insurance company practices described in this memorandum are those that exist in today’s market. In all likelihood, they would continue unabated in the absence of federal health reform legislation. One of the major benefits of the Affordable Care Act, which was signed into law on March 23, 2010, is a ban on the practice of denying coverage based on preexisting conditions.
The authors of the report tried to extend their findings beyond the four largest health insurance providers and into the general population, writing “approximately 15.7 million adults under 65 received their health care coverage through individual health insurance policies.”
As a side note, 15.7 million people sounds like a large number, but represents merely 5 percent of the total population of the United States. The other 95 percent of the population do not face denials based on preexisting conditions because the law prohibits group health insurance plans and governmental health insurance plans from denying coverage based on a preexisting condition.
Although the report references the 15.7 million Americans in the individual health insurance market, it does not state that all 15.7 million applied for health insurance coverage during the time frame of the report. In fact, it is a certainty that they did not. Yet the number of applicants nationwide is the crucial number to consider when analyzing health insurance denials. Only then can the number of people denied coverage be placed in context.
Instead, the report stated that on average approximately 1 out of 7 applicants was denied coverage. Since 651,000 people in the study were denied coverage, that would mean 4,557,000 people applied for coverage to the four major companies over three years (651,000 x 7).
We have already seen, however, that the number of people denied coverage could be four times smaller than the number of applicants, due to duplication of applications. Therefore, the more accurate statement would be that 1 in 28 applicants was denied coverage. Since not all people in the individual market applied for insurance in the last three years, but 4.5 million did apply to the largest health insurance companies, a reasonable estimate of the number of applicants over the past three years would be half of the market, or 7.6 million people. If 1 in 28 was denied coverage, that would total 271,428 individuals, or 90,474 per year. Even this is a minuscule portion of the population, just three-one hundredths of one percent of the population.
The compassionate thing might be for taxpayers to pay the medical bills of those who are denied individual health insurance coverage. Perhaps they should be allowed to join Medicaid, or we could set aside a federal superfund to pay for their care. This would be much smarter than burdening the private health insurance industry with mandated coverage that is actuarially unsustainable, as the new health insurance reform law does.
![]()