I Am Sick: A flimsy excuse to use the phrase "evidence-based."

As the title says, I'm sick. Not deathly sick, but merely the massively inconvenient sinus problems one can get with the changing weather. As a part of trying to treat myself, I went down to the local drug store and bought a 50-dose bottle of pseudoephedrine (totaling 3 g of the active ingredient).

Had this been in the United States, the comically self-righteously named Combat Methamphetamine Epidemic Act of 2005 would have meant that I was dangerously close to the 3.6 g/day purchase limit. Exceeding CMEA05 limits is criminally punishable, as some people have had the misfortune to find out by accident. Even if one does not exceed the given limits, CMEA05 requires that retailers maintain an archive of purchaser IDs for at least two years, establishing a worrying paper trail about one's OTC purchases.

The ostensible purpose of the law is right there on the tin: to combat what is seen as an "epidemic" of methamphetamine usage and production. The questions for any supporter of evidenced-based public policy, then, are whether this epidemic exists, and whether draconian measures against allergy and cold medication purchases help. I will leave the first question for now, as there are significant arguments to be made about what constitutes an "epidemic" of addictions.

Rather, I would like to focus on the question of whether or not such draconian measures are effective. Were they effective, we would expect that the rates of meth abuse in Canada, with its comparatively more lax laws regarding pseudoephedrine and other means of meth production, would be significantly higher. To truly argue this point, one would need to control for as many other potentially confounding factors as possible (different economic and cultural conditions, availability of other drugs, degree of enforcement of existing laws, population density, etc.), but we can at least take a first look at the data. Thanks to the help of Saver Queen, I was able to find statistics on 2009 meth abuse rates amongst middle- and high-school students in the US and Canada, respectively.

Only the Canadian source comes with error margins, so a formal hypothesis test is impossible, but we see that to the 95% confidence level on the Canadian statistics, between 0.8% and 1.7% of students in grades 7, 9 and 11 reported past-year usage of meth, while data from the estimated US rates lie within the 1.0% to 1.6% range (again, no error bounds on the US data). Since all of the US data linked to from here is consistent with Canadian statistics, we have no basis to even suspect that US rates are in fact lower than Canadian rates, in lieu of hypothesizing significant confounding factors.

My point with this is not, of course, to do a full analysis of the comparative rates of meth abuse, as that would require significantly more research than my stuffed-up head can handle right now. Rather, I simply wanted to give an example of how an evidence-based conversation about public policy might look. In particular, it has to start by deferring to reality and not simply to rhetoric.