Soda Ban Science: Doing the Right Thing?

Weighing the evidence behind the proposed restrictions on sugar-sweetened beverages in New York City

Published on September 10, 2012by Dana March

Today, the New York City Board of Health will vote on the Bloomberg administration’s proposed limits on sugar-sweetened beverages. Will Bloomberg’s public health paternalism, which delivered smoke-free bars and restaurants, calorie counts, and a trans fat-free foodscape also dictate size restrictions on sugary drinks for New Yorkers?

The policy, summarized by a brief, proposes a maximum size of 16 fluid ounces for sugary drinks offered in Food Service Establishments regulated by the Health Department—any establishment with Health Department letter grades. “Sugary drinks” are defined as those “that have added sugar, contain <50% milk/milk substitute by volume and have more than 25 calories per 8 fluid ounces.”  Free refills are permitted, so sugar lovers can purchase as many 16-ounce cups as their hearts desire and wallets permit.

Bloomberg’s current proposed restrictions are only the most recent efforts to dabble politically in drink. A proposed penny-per ounce soda tax in New York State in 2010 failed, as did a subsequent proposal to the USDA to nix sugar-sweetened sodas from the approved goods list for New York City recipients of the federally-funded Supplemental Nutrition Assistance Program (SNAP).

Is Bloomberg’s targeting of sugary drinks scientifically justified? What is the evidence linking sugar-sweetened beverages to obesity? Do studies demonstrate that structurally shaping the food landscape vis-à-vis portion control is the way to go? Has the New York City Department of Public Health leveraged this science, and if so, how?

“Solid” Science on Sweet Liquids?

Scientists and other experts, advocates, concerned citizens, heads of community-based organizations, chef Jamie Oliver, and even Spike Lee have suggested, in statements published on the website—albeit under the tagline “More Medical Experts Weigh In on Importance of Mayor Bloomberg’s Anti-Obesity Campaign”—that Bloomberg is indeed doing the right thing.  (I love Spike Lee’s films and Jamie Oliver’s attempts to seed a neoculinary revolution one organic ingredient at a time, but “medical experts” they are not.)  Together, these expressions extolling Bloomberg’s bravura create a kaleidoscopic view of why the proposal should pass. Few, however, focus on the actual science.

One exception is Dr. Frank Hu, Professor of Nutrition and Epidemiology at the Harvard School of Public Health and Professor of Medicine at Harvard Medical School:

“It is entirely justifiable to single out sugary beverages as a target for policy interventions.  First, obesity has become the most important public health problem of our time, and sugary beverages are the major source of excess calories and sugar in our diet. Second, there is solid scientific evidence supporting that regular consumption of these beverages contributes to obesity, diabetes, and heart disease, and a larger portion size has been linked to an increase in calorie consumption. Third, studies have shown that changing the food environment by limiting access to large portion sizes of soda is effective in reducing consumption, while education alone is not sufficient to change people’s behaviors.”

Daniel Engber recently published an interesting analysis of the “contested science” in Slate, sketching the contours of some recent systematic reviews and meta-analyses, pointing out that some science is potentially soured by industry ties (for the uninitiated, a systematic review is one with a replicable method, and a meta-analysis is a quantitative means of summarizing a body of scientific literature).

So what is the “solid scientific evidence” to which Hu refers? A 2006 systematic review of studies in both adults and children and adolescents, on which Hu himself was the senior author, gives us some indication.

In ascending order of causality-establishing credibility, the types of studies reviewed were:

  • Cross-sectional studies, which measure the exposure and the outcome simultaneously.
  • Prospective cohort studies, which follow exposed and unexposed groups of people forward in time to determine the impact of a given exposure on subsequent outcomes.
  • Experimental studies, which manipulate exposures to isolate their contribution to an outcome.

Although cross-sectional studies constitute the bulk of the literature reviewed, they cannot actually distinguish whether consuming sugary drinks makes people obese or whether obese people just consume more sugary drinks.

From prospective cohort studies in adults, the evidence is mixed. There is little consistency with respect to exposures, such as measures of sugary beverage intake, and outcomes—obesity, overweight, BMI changes, and so forth.  This is a particular challenge with summarizing bodies of scientific literature.

Some specific findings were compelling, however, with respect to soda consumption and obesity over time. For example, one longitudinal study in Norway failed to find a statistically significant difference in overweight and obesity between men and women who reported long-term high soda intake (three to four per week) and those who reported long-term low soda intake (two or fewer per week).

This is consistent with a 2004 report in JAMA, using data from the Nurses Health Study II—a prospective cohort study of 116,686 female nurses initiated in 1989 and followed every two years—of which Hu was again senior author. There was no significant difference in weight change between women who maintained a high soda intake (one or more per day) and those who maintained a low soda intake (one or fewer per week). Notably, in this study, women who increased their soda intake from low to high over the course of four years put on an average of about 10 pounds. However, these women did not lose a significant amount of weight when they then reversed their soda intake from high to low in the subsequent four years. Also, women who decreased their soda intake from high to low still gained weight.

Hu and his colleagues noted that the benefits of curtailing consumption were limited, as is the literature:

“The lower weight gain associated with reduction of sugar-sweetened soft drink consumption compared with stable intake suggests that women do benefit from decreasing consumption but that weight trajectories do not continue to diverge with time. Long-term effects of sugar-sweetened beverages on body weight have not been studied in experimental settings so far, and further research is warranted.”

Which leads us to the handful of experimental studies reviewed. These studies offer somewhat more convincing data that links sugary drinks to weight gain and obesity. One randomized controlled trial that followed adolescents for about six months demonstrated that decreasing intake of sugary drinks could nudge people from being obese to overweight (from a BMI of 30 or higher to a BMI of 25-29.9).

But are these findings overwhelmingly in support of the proposed restrictions on sugary drinks?

Hu notes in the review that “further research, particularly from large prospective cohort studies with long follow-up and repeated measures of both diet and weight, is needed to provide more convergence in the data.”

It seems the jury is still out, even though the judgment has been issued.

Sipping Science… or Drinking the Kool-Aid?

A 2011 report issued by the New York City Department of Health and Mental Hygiene (DOHMH), which examines consumption of sugar-sweetened beverages in the city, selectively leverages the literature in its background. The report states “[c]onsuming sugary drinks is directly related to weight gain and obesity,” citing Hu’s review, but not others, such as a 2007 review by Kelly Brownell and colleagues at Yale, which concluded that “available data indicate a clear and consistent association between soft drink consumption and increased energy intake,” although not weight gain and obesity per se, as well as a dissenting 2008 review.

The report presents findings from a survey of adults in five New York City neighborhoods, documenting residents of East and Central Harlem, North and Central Brooklyn, and the South Bronx regularly consume many more sugary drinks than residents of the Upper West Side and Flatbush. They also happen to be more overweight and obese.  Again, cause and effect cannot be determined.

According to the report, drinking water is an acceptable alternative to sugary drinks to nearly all people surveyed, but over a quarter of them indicated that they were leery of New York City tap water, which, it should be noted, consistently wins taste tests.

So instead of pouring resources into promoting the consumption of our Evian-trumping city tap water, the Bloomberg administration has focused on limiting sugary drinks.

Consider the Source

If the science underlying the Bloomberg Administration’s proposed restrictions on sugary drinks seems fluid, it is. It depends on who’s stirring. Bruno Latour and Steve Woolgar remind us in Laboratory Life: The Construction of Scientific Facts that “scientific facts” are socially shaped and necessarily contested by scientists and non-scientists alike:

“…[B]oth scientists and observers are routinely confronted by a seething mass of alternative interpretations. Despite… well-ordered reconstructions and rationalisations, actual scientific practice entails the confrontation and negotiation of utter confusion. The solution adopted by scientists is the imposition of various frameworks by which the extent of background noise can be reduced against which an apparently coherent signal can be presented.”

The proposed limits on sugar-sweetened beverages are borne of the progressive paternalism that keeps New York City at the vanguard of public health action in the United States. However, there is a lot of noise around the science. Should the proposed restrictions on sugar-sweetened beverages pass today, almost certainly to immediate appeal, the DOHMH would best serve New York City residents and scientists alike by deploying a sound scientific strategy to determine the success—or failure—of this particular policy intervention. Moreover, the scientific evidence clearly indicates that the causes of obesity are many. This liquid lever on the epidemic alone is inadequate, and must be part of a coordinated plan to address the weight and magnitude of this problem.

The Bloomberg administration is constructing its own scientific narrative on sugar-sweetened beverages, and just as “medical experts” are featured on the website, so should eloquent (or perhaps just not belligerent) detractors. Still, the coherent signal is that the Bloomberg administration—whither its reconstructions, rationalizations, and negotiations of confusion—seems to be trying to do the right thing in the interest of public health. Individual liberties notwithstanding, we can all drink to that. At least in 16-ounce cups.