Thursday, March 12, 2009

Who Lead Healthier Lives - Democrats or Republicans?

A recent letter to the editor of the International Journal of Epidemiology argued that Republicans lead healthier lives than Democrats.

The authors of the referred study - a doctoral student and a professor of health studies, both at Harvard University - suggested that:

"The observation that [R]epublicans enjoy better health status may reflect the core [R]epublican value of individual responsibility, which could translate into increased adherence to health-promoting behaviours."

Overall, the study controlled the factors of age, sex, race, marital status, religious service attendance, highest educational degree, and total family income (all of which have their own implications towards an individual’s health) and found Republicans were 25% less likely than Democrats to be in poor health.

In making these observations, they analyzed General Social Survey data from 1972-2006. The number one factor they found in contributing to this difference in health and political ideology was smoking. Democrats, they found, were 15% more likely to be smokers.

The study also found that Republicans might enjoy greater well-being because of a greater sense of spirituality "which may lead to health promoting social conditions such as enhanced social ties and networks."

Andrew at took issue with these findings. In a post last night he wrote:

"I think a lot more could, and should, be done here, for example using the full four-point scale for self-reported health and having a better understanding of the transition from raw comparisons to the regression model."

He also mentioned a cause-and-effect issue at hand - as people get sick and need access to health care they may find themselves aligning more with the Democrats - although he was not totally convinced of that notion himself.

So are Republicans really healthier than Democrats?

As we mentioned in our post from Tuesday, there is an interesting new database published by America’s Health Insurance Plans (AHIP) - a lobby group for the Health Insurance industry - that looks into healthiness of Americans by state and congressional districts.

So we compiled some of the data from these "Resource for Well-Being" reports and compared them to the 2008 Election Results on a state-by-state basis.

We looked at four areas along the lines of the Harvard study - Healthy Behavior, Physical Health, Emotional Health, and the overall Well-Being Index (a function of the previous three areas, as well as Life Evaluation, Work Quality, and Basic Access to health care). Here is what we found:

The general trend here seems to be that the residents of blue states tend to adhere to healthier lifestyles. The more Democratic they vote, the better they tend take care of themselves. The top 5 states in this category were 1) New Mexico; 2) Vermont; 3) Montana; 4) Hawaii; and 5) California.

Again, the general trend here appears to be better physical health for states that vote more heavily Democratic. However, the top 5 states in this category were 1) Montana; 2) Wyoming; 3) Hawaii; 4) Colorado; and 5) North Dakota - only one of which is seen as a regular “blue state”. More telling is the bottom 5 states: 46) Oklahoma; 47) Arkansas; 48) Mississippi; 49) Kentucky; and 50) West Virginia - all of which are consistently red.

In terms of emotional health, we didn’t even bother trying to put a line on the graph - it’s too scattered. It seems there is almost no correlation between emotional health and political ideology. The top 5 states in this category were 1) Hawaii; 2) Alaska; 3) Wyoming; 4) Minnesota; and 5) Montana.

What is more interesting is that, of the top 10 states here, six of them were northern Prairie / Mountain states. Hawaii - which was by far the most Democratic-supportive state in 2008 (at 72%) - was also by far the most emotionally healthy state (though probably because people tend to be happy when they live in a tropical paradise).

Finally, we found that the general trend (in terms of the Well-Being function) was that states that were more Democratic-supportive tended to have higher well-being among their residents. However, there were yet again several peculiarities. The top 5 states in this category were 1) Utah; 2) Hawaii; 3) Wyoming; 4) Colorado; and 5) Minnesota.

It is important to mention that, unlike the Harvard study, we analyzed our data on a state-by-state basis rather than at the micro level. This might have important implications. For example, in California the suburban, GOP-voting Orange County resident might be getting more exercise than the inner-city, Democratic-voting, single parent of Los Angeles.

We also did not perform a control on factors such as age, race, gender, etc. There is no question that this accounts for a lot of differences between our figures and those of the Harvard study. It would not be farfetched to say that this is a shortfall in our analyses.

For example, men are more likely to vote Republican than women, and men have a shorter life expectancy than women. African-Americans are generally more susceptible to diseases such as cancer and diabetes, they are by far the most Democratic-supportive race in the United States, and the black population is not as widespread among the states as Whites or Latinos.

However, we doubt that the Harvard study made realistic conclusions. First of all, they used a lot of outdated data. For example, it is prudent to assume that smoking has decreased significantly among Democrats since 1972, and that Republicans may very well be smoking more than Democrats these days.

While we generally find that Democrats lead healthier lives in our own simulations, the most realistic conclusion to make is that political ideology has little-to-no bearing on the well-being of an individual. A more obvious trend is healthiness by region - those living in the Western states tend to be particularly healthy while those living in the Rust Belt and Appalachia are among the unhealthiest of Americans.

That being said, finding correlations between party-affiliation and things like healthiness are very interesting, and we hope to see more studies and discussions like this in the future.

1 comment:

Anonymous said...

so you oppose an individual-based, risk factor-adjusted, made with direct data study with a poblational-based, non adjusted, made with indirect data study.
Either you have no idea about epidemiology, or this post is nothing but cheap, pseudoscientific propaganda.