Over the years there has been a push to limit the consumption of sugar-sweetened beverages. Various jurisdictions have attempted to levy a tax on these products, including Mexico, Finland, Hungary, France, and parts of the United States. Recently, provincial governments in Canada were kicking the tires on a potential pop tax, including the Northwest Territories and Alberta.

Any efforts to throttle mass consumption of sugar-sweetened beverage must be backed by irrefutable evidence that these drinks actually contribute to the development of obesity, diabetes, and related co-morbidities. Although I had just taken for granted that this was well established, it turns out there is much noise within this area of research.

Investigations into potential health impacts of consuming sugary drinks is often sponsored by the very same companies who manufacture these drinks, and whose very profits and existence relies on increased consumption. Since this smacks of bias, it makes one wonder how much of the conflicting results in the literature are an artifact of funding source, rather than the natural messiness of science.

A recent study by Schillinger and colleagues at the University of California, San Francisco, asked a brilliantly simple question:

Are studies showing an absence of relationship between sugary drink consumption and detriment to health more likely to be sponsored by the pop industry?

Their design was also delightfully straightforward:

They surveyed the scientific literature looking for experimental studies and systematic reviews or meta-analyses of experimental studies assessing relationships between sugary sweetened beverage drinking and obesity, diabetes, or other health outcomes. Then they looked through the mouse font of the articles, or through clinical trial registries to ascertain who funded each study.

In total, they found 60 such studies (28 trials and 32 systematic reviews/meta-analyses of trials).

Highlighting the disparate results in this area, they found that 34 articles (20 trials and 14 systematic reviews/meta-analyses) described positive associations between drinking pop and obesity or diabetes, while 26 articles (8 trials and 18 systematic reviews/meta-analyses) found no such association.

In other words, there is a 50/50 chance that a study may find a link between pop drinking and disease. No wonder so much confusion abounds.

But here is the fascinating part: 25 of 26 studies (96.2%) that found no association had funding ties to this industry. Conversely, only 1 of 34 studies (2.9%) showing an association declared ties to the pop industry.

Sometimes in science, performing a statistical analysis seems unnecessary when the observation is as stark as this. (Not to worry; the authors did the analyses, and unsurprisingly, the stats support the observation).

The authors of this pithy but poignant study hold no punches in summarizing the results:

“In conclusion, clinical trials and systematic reviews of trials in which the conduct of research or investigators were supported by the sugar sweetened-beverage industry were much more likely to find no association between their products and metabolic outcomes than those that were independently funded. This industry seems to be manipulating contemporary scientific processes to create controversy and advance their business interests at the expense of the public’s health.”

Peter