Death by lack of insurance: statistical trickery

If past cycles are any indicator, we are about due for another round of studies showing that being uninsured kills. Headlines will abound, commentators will wring their hands, and anyone who opposes ObamaCare will be portrayed as an insensitive killer.

The good news is that the majority of the supposed deaths due to a lack of health insurance are statistical artifacts. Unlike deaths from government health care, which has been extensively documented to cause harm to the seriously ill, the elderly, and the newborn, the deaths due to a lack of health insurance are usually conjured out of data sets using a variety of statistical tricks.

Here’s a simple checklist for cutting the number of deaths down to size:

1. Does the study interview people only once and then assume that they remain uninsured? If so, it will likely overstate the fraction of people who die because they don’t have health insurance. Relatively few people remain uninsured for more than a year.

2. Does the study track how people died? If not, it probably overstates the number of deaths due to a lack of insurance. Younger people are more likely to be uninsured. Younger people are also more likely to die from homicide, something that health insurance has nothing to do with. If homicides or accidents contribute to the higher mortality in the uninsured group, it is not reasonable to claim that their higher mortality is due to a lack of health insurance.

3. Does the study control for different the differences in social and economic status between the insured and the uninsured groups? All work to date suggests that people who are uninsured differ from the privately insured in ways that are likely to independently increase mortality even in people with coverage. For example, the uninsured are more likely to be high school dropouts, more likely to smoke, less likely to have ever worked, and less likely to be married. All of these characteristics are associated with higher mortality whether or not someone has health insurance.

So, does having health insurance reduce mortality or not?

Current estimates suggest that if it does, the effect is small, likely smaller than the deaths from the lower incomes created by the higher taxes needed to ensure everyone in American using the Massachusetts-like plans in the ObamaCare bills.

June and Dave O’Neill, in a paper on the uninsured written for the Employment Policies Institute, conclude that “lack of health insurance is not likely to be the major factor causing higher mortality rates among the uninsured.”

Doyle, in a comparison of survival for Wisconsin motor vehicle accident victims who were uninsured or covered by a private plan found that the uninsured were more likely to ride a motorcycle and crash in the middle of the night. Controlling for that, he found that their mortality rate was 1.5 percent higher, except in hospitals with a religious affiliation.

(Via StateHouseCall.org. See Linda Gorman’s other posts there.)

Advertisements

1 Comment

Filed under myths & fallacies

One response to “Death by lack of insurance: statistical trickery

  1. Pingback: Health Care Bill has passed - Page 37

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s