CDC's study on alcohol use is misleading

By Raymond Scalettar

The Centers for Disease Control has weighed in very publicly in the debate surrounding the privatization of state liquor stores in Pennsylvania, suggesting that privatization would harm public health. The studies CDC cites as evidence, however, do not support that claim.

Robert Brewer, who leads the alcohol program in the CDC's National Center for Chronic Disease Prevention and Health Promotion, has repeatedly pointed to a review by CDC's Community Task Force. It found a 44 percent median increase in per capita sales of privatized alcoholic beverages within jurisdictions that underwent privatization of retail alcohol sales.

Unfortunately, Brewer has never explained what the 44 percent estimate really means. This data, which has been presented out of context, is misleading and useless in the Pennsylvania privatization discussions.

The 44 percent figure was derived by analyzing 17 studies that looked at the impact privatization had on the privatized beverage. The 44 percent growth estimate is not an estimate of total alcohol consumption, nor is it an estimate of alcohol-related harms.

Of the 17 studies analyzed, six showed no increase in consumption, and four showed only moderate increases. This fact alone would give most researchers pause with regard to any kind of sweeping conclusion.

Importantly, the Community Task Force's review found no pattern of increased alcohol-related harms in the studies it analyzed, which ultimately is what the public is most concerned about.

What really drives Brewer's "the sky is falling" 44 percent figure are studies of wine privatization from more than 30 years ago.

Between 1970 and 1981, six U.S. states (Alabama, Idaho, Iowa, Maine, Montana, and New Hampshire) privatized state wine sales. And these states did see significant increases in wine consumption. But so did the rest of the country.

Wine was much less popular in 1970 and per capita consumption levels were only around half of what they are today. However, between 1970 and 1980, total wine volume sold in the United States went up by 85 percent, and per capita wine consumption increased by 66 percent.

As with the six privatized states studied, large percentage gains are easily achieved when growth is from a small base. It should be noted, however, that while wine consumption was increasing dramatically, total alcohol consumption was flat.

While the numbers reported sound big, when they are spread across the adult population, the actual increase in consumption was minimal. The largest increase, in Idaho, would have caused each adult to increase consumption by 16/100's of a drink per day - about three-fourths of an ounce of wine.

Brewer also used the 44 percent figure to predict the impact of privatization in Washington state, where residents voted overwhelmingly to privatize the state's liquor control system.

While it will take a full year to determine just what the impact has been, initial results show spirits volumes (including cross-border sales) up 11.6 percent (only spirits were sold in state stores in Washington). But, because beer sales were down and wine sales flat, total alcohol consumed was up only 2.7 percent - similar to the U.S. total of 2.5 percent for 2012. Thus, no 44 percent increase in spirits or total alcohol consumption is in sight.

In the clinic of the real world, the 44 percent calculated increase in consumption has already been shown to be wrong. As crafted, the CDC-funded study was never designed to predict consumption in a privatized Pennsylvania market, and it certainly says absolutely nothing about alcohol related harms.

In the interest of honest public debate, Brewer should explain what his 44 percent figure really means, and be upfront about the key fact that the task force could find no solid evidence that privatization will lead to increased alcohol-related harms.

The CDC has the imprimatur of a respected, science-based government organization. Brewer has the responsibility to honestly present research in an unbiased, forthright manner so the public and elected officials can make decisions based upon the best available evidence.