UCSF doctor cites bias in health studies linked to soda industry

By Heather Knight

Published 5:00 pm, Monday, October 31, 2016

Of the 60 scientific studies conducted since 2001 to determine whether drinking sugar-sweetened beverages contributes to obesity or diabetes, 26 of them found no link. All 26 of those had something in common: They were funded by the sugar-sweetened-beverage industry or conducted by people who have financial ties to the industry.

Of the 34 studies that did find a link between drinking sugary drinks and developing diabetes or becoming obese, just one was funded by the beverage industry. The rest were conducted independently.

These are the new findings of Dr. Dean Schillinger, a professor of medicine and chief of the UCSF Division of General Internal Medicine, whose report will be published Tuesday in the Annals of Internal Medicine. Schillinger, a strong supporter of the soda taxes on the San Francisco, Oakland and Albany ballots next week, called his findings “a smoking gun” in the longtime question of whether bias exists in industry-funded scientific studies.

“They’ve got the gun in their hand, it’s smoking and the patient is dead,” Schillinger said.

Schillinger was a paid expert for the city of San Francisco in a lawsuit defending an ordinance mandating health warnings on soda advertisements. The city paid him to write a report on what is known about the association between sugar-sweetened drinks and diabetes and obesity, and his new study stems from that work.

The American Beverage Association, which has so far spent $21.3 million in San Francisco to defeat Proposition V, released a statement calling it ironic for Schillinger, the city’s paid expert, to charge the soda industry with bias.

“This paper is the latest in a trend of pro-tax forces writing speculative opinion papers to influence voters a week before a vote on several ballot initiatives to tax beverages,” the statement read.

The statement also said that the beverage industry has a right to engage in scientific research and that the research is sound.

“The research we fund adheres to the highest standards of integrity for scientific inquiry based on recognized standards by prominent research institutions,” the statement reads.

Schillinger’s study is just the latest to knock industry groups for trying to sway the outcomes of scientific studies. In September, UCSF researchers revealed the sugar industry in the 1960s paid two Harvard scientists to conduct a literature review showing eating a low-fat diet, rather than a low-sugar diet, would reduce cholesterol and prevent heart disease.

For his study, Schillinger searched the PubMed database, a service of the U.S. National Library of Medicine, to find all soda-related studies published in English-language journals from January 2001 to this July.

He excluded studies funded by competitors of the soda industry, including bottled water companies and the dairy industry.

There was a 100 percent likelihood that studies that found no link between sugary drinks and diabetes or obesity were funded by the beverage industry or were conducted by authors who had financial ties to the beverage industry. There was a 2.9 percent likelihood that studies that found a positive link had the same financial ties to the industry.

Schillinger called that difference “ridiculously high” and far greater than has been found when it comes to industry-funded studies related to food or pharmaceuticals. He said the findings were so stark, he wanted to ensure they were reported before the election.

“The general public by and large believes the scientific method is immune to that sort of bias,” he said.

“The industry is going all out, and they have invested significant amounts of money and intellectual energy into creating study designs, study settings, study investigators and study outcomes that are favorable to their interests.

“We in the public are not aware,” he continued. “It’s largely been invisible to us.”