Wednesday, January 23, 2013

If It Smells Like Junk Science, It Probably Is

The news of the latest study about the 21 drinking age now appears to be going viral.   According to this study, which is now available (only to subscribers) online ahead of print, folks who were young adults in states that allowed them to drink legally before age 21 at that time were statistically more likely to become more frequent "binge" drinkers later in life compared to those who were not allowed to drink legally until age 21.  The research, which used data from surveys in 1992 and 2002 taken by those who were born between 1949 and 1972 (i.e. were young adults in the 1970s and 1980s), interestingly found no difference in overall alcohol consumption or frequency between the two groups, but apparently found that those allowed to drink before 21 had more "binge" days and fewer "non-binge" days per month compared to those who were not allowed to drink until 21.  The former were 19% more likely to "binge" more than once per month compared with the latter, and the differences were largely (if not entirely) driven by men and those who never attended college.  So what should we make of this study, which is not yet available for the general public to read?

First of all, we at Twenty-One Debunked always put the term "binge drinking" in scare quotes when we are referring to the 5+ or 5/4+ drinks definitions, as we believe that such definitions are grossly inaccurate measures of the very real problem of truly dangerous drinking, and can potentially mask actual trends in the latter.  (More information about this issue can be found in our previous posts here and here)  And we know based on the article's summary that a 5+ drinks threshold is the one used in this study, as is the case in virtually every other pro-21 study out there.  Strike one.

Secondly, no information is provided about which, if any, confounding factors are controlled for.  This is crucial because there are numerous other differences between people who grew up in different parts of the country and/or at different times.  One should also note that the effect size is fairly small as well, with a relative risk (or odds ratio) of 1.19 overall (1.31 for men who never attended college).  In epidemiological research, relative risks below 2.0 are especially likely to be due to a combination of chance, bias, and/or confounding, and thus should be taken with at least a grain of salt (if not a whole pound).  Strike two.

Finally, the study really adds nothing else new to the scientific literature beyond what was mentioned above.  Zip, zilch, nada.  And nothing about whether there were any between-group differences in actual problem drinking.  The authors (as well as MADD member Ralph Hingson) refer to other past studies (including a 2009 study which we had debunked years ago) by other authors in an attempt to connect the dots.  But given enough dots, one can pretty much connect them any way to form any picture one chooses.  Strike three, you're out!

Thus, our preliminary analysis of the study (to which we were unable to gain full access--stay tuned for updates!) suggests that the study reeks of junk science, and clearly should not be used to set public policy.  However, let us be clear that even if it (and the 2009 study about increased risk of alcoholism) somehow were 100% true, which we seriously doubt, we at Twenty-One Debunked would still support lowering the drinking age to 18.  Why?  The 21 drinking age is nothing less than a hate crime against young people, plain and simple.  In our society we know, for example, that certain ethnic groups are statistically more prone to alcoholism than others, yet we do not arrest, jail, revoke privileges, or publicly humiliate members of such groups for the simple act of drinking alcohol in the name of "public health."   That, of course, would be illegal discrimination since it violates the 14th Amendment's guarantee of equal protection of the law, and no amount of "scientific" research can justify it.  And even known alcoholics over 21 are not jailed simply for being alcoholics--they simply hold too much political power for that.  But 18-20 year old men and women, despite being legal adults in virtually every other way, are apparently a much more acceptable target for "public health" fascism run amok, no matter how responsibly they drink.

The injustice must end NOW.  Let America be America again, and lower the drinking age to 18.  If you're old enough to go to war, you're old enough to go to the bar.  'Nuff said.

1 comment:

  1. Junk science shouldn't pass as real science. The study in this article is like a study that has bad input all around it. This study shows that the group which had the right to drink at 18-20 are close to the group which didn't have the right to drink at 18-20. Along with the similarity, I think that this study is junk science. The drinking age should be lowered to 18 in states and territories where the age of majority is 18. The drinking age of 21 is ageist. A drinking age of 18 with reasonable restrictions would create a culture of responsible alcohol consumption. This would mean that most young women and young men would drink responsibly. My primary reason in support of a drinking age of 18 is civil rights. Closely tied to this are civil liberties. Internal possession laws violate civil liberties and those laws are at the height of oppression against young women and young men.