Lies and Statistics

There is an old saying, “There are lies, damned lies, and then there are statistics!”  With the current debates in the US about health care reform, this has never been more true.  The problem is actually quite simple, however.  Statistics is viewed as “science” or “math”.  It is the embodiment of “truth” and part of all that we hold dear.  It is not to be questioned!  Unfortunately, a large portion of the US population knows next to nothing about it.

In order to get some “reliable” statistics about health care in the US, I went to Michael Moore’s statistics information for his new movie, Sicko.  It can be read here.  Note, for those brave enough to read it, there is a HUGE difference between “health care” and “health insurance”.  Also note that filing bankruptcy does not, in general, destroy your life.  One quote is: “countries spending substantially less than the US have healthier populations.… The infant mortality rate for the U.S. is now higher than for
many other industrial countries” which is a quote from The Impact of Health Insurance Coverage on Health Disparities in the United States, Human Development Report, UNDP, 2005.  Are you scared for your babies yet?

So, why does this not tell us everything it seems to?  Let’s go to the CDC website for birth facts and find out.  Two facts pop out: 1) 8.3% of babies are born with low birth weight.  2) 38.5% are born to unwed mothers.  This can’t be good.  How does it compare with other countries? has some information here, but it doesn’t tell us a whole lot.  The CDC has more facts here and here.  The short version is, low birth weight is linked to high infant mortality.  So, if the US has the highest rate of low birth weights among industrial countries, then the high infant mortality rate would make sense.  Looking at the site, this seems to be the case.

As you start digging, you find references to the age of the birth mother as a risk factor for low birth weights.  I recall hearing a lot of discussions about teen pregnancy rates in the US over the past decade.  Could behavior be a health risk factor?  Ask the obese people suffering from heart disease!  After digging, what we can find is that a “health statistic” is not necessarily an indictment of the US health care industry.  It may be an indictment of the behaviors US citizens engage in.

There is a valid statistic that indicates the US has an unusually high infant mortality rate.  The problem is: that statistic doesn’t tell us anything.  What needs to be examined is infant mortality rates among married women in various age ranges and unmarried women in various age ranges.  Unfortunately, while the CDC has some excellent statistics for what’s happening in the US, it’s harder to get the same data for other countries.  Hint: being a teenage mother is a major risk factor for a low birth weight baby.  25-35 is the optimal age range.

Statistics are worse than lies, the tell the truth, but not the whole truth.  Unless you dig a little, you are unlikely to have the necessary information to draw a meaningful conclusion.  This gets worse when people have an agenda.  If someone presents statistics about insured and uninsured people, be suspicious.  You don’t need insurance to get health care!  The emergency room has to accept you.  If you are uninsured, have a heart attack, get saved, and file bankruptcy, that is better than being dead and financially solvent.  When someone presents a statistic, please dig a little deeper.  Ask for details, especially when they aren’t being provided in the argument.  It could make a huge difference in understanding what’s actually going on.


Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: