11-15-2009 - Vol. 26 / No. 11An international team of researchers tried to determine the effect of relatively stringent alcohol control policies on adolescent drinking patterns in 30 countries. They "scored" each country's strictness based on 16 policy topics in 5 "domains": alcohol availability, drinking context, price, advertising and motor vehicle issues like random breath testing. Then they correlated the scores with measures of adolescent drinking. They hypothesized that "countries with more comprehensive and stringent alcohol control policies will have a lower prevalence of alcohol consumption, heavy drinking and related harm among youth." On first blush, they seemed to be correct. Overall scores on their "Alcohol Policy Index" (API), advertising and availability were linked to less drinking. The advertising index was also related to fewer heavy drinking sessions and less likelihood of having a first drink by age 13. The authors cited no specific link between price and adolescent consumption. But importantly, most of the relationships lost statistical significance "after controlling for per capita consumption," including any link between ads and consumption. All that remained statistically significant was the "inverse relationship between the API score and drinking 6+ times in the past 30 days" and the association of availability measures and 2 of the 30-day measures.
These findings led to a remarkably hedged conclusion. They "suggest that more comprehensive and stringent alcohol control policies could, potentially reduce the prevalence and frequency of alcohol consumption by youth," the authors ventured. But the findings also suggest that "alcohol use in the general population (per capita consumption) may mediate or confound at least partially the observed relations between alcohol control policies and prevalence of youth alcohol consumption and having first drink by age 13." That is, "stricter policies may reduce adult consumption which, in turn, influences consumption by youth."
But there's also another common-sense possibility: "More conservative adult drinking styles may lead to the adoption of stricter drinking policies and independently to lower rates of drinking among youth." At the same time, beyond per capita consumption, "there may be other cultural or socio-economic factors (i.e. political ideology)" that could "influence" both policy and adolescent drinking. The available data "precludes making causal inferences." The authors also noted that levels of enforcement of specific policies could not be determined and can "vary considerably." On the specific hot topic of alcohol marketing, again the findings suggested stricter policies "may be related" to less drinking. But "it is also possible that the observed relationships between advertising policy and youth alcohol consumption were attributable at least partially to drinking norms in the general population." In the end, this study doesn't appear to advance the ball very far and the authors are left with the same fairly thin reed with which they began: "It seems plausible" that stringent policies, "especially policies that limit availability, can reduce consumption by youth." Ref 2
