When it funds scientific research, Coca-Cola includes a provision in its contracts with academic institutions that allows the beverage giant to pull its funding for a study at any point, according to a group of researchers who obtained several such agreements.
The policies could pressure recipients of the funding to pursue research that dovetails with Coca-Cola’s goals out of fear of having their project canceled, the researchers said in a paper published Tuesday, though they added that they found no example of that occurring.
The paper, which was published in the Journal of Public Health Policy, comes amid increasing scrutiny of the food and beverage industry’s funding of and influence over academic research. The industry has taken a number of steps to improve transparency and safeguard the independence of studies it sponsors. Notably, Coca-Cola in 2015 started listing on its website the institutions and researchers it funded and the following year outlined principles that would guide its support for scientific research.
But the termination policies the researchers highlighted seem to undermine that commitment to scientific transparency and unrestricted research, they said.
“The ability to quash studies and retain the data — we don’t find evidence that they did that, but the contracts mean that they have the ability to do that,” said Sarah Steele, a policy researcher at University of Cambridge and one of the authors of the paper.
The concern about research sponsored by food and beverage companies is that it might be biased to promote messages about obesity, nutrition, and exercise that align with a company’s business aims. But with limited public resources, corporate money plays a crucial role in research. It’s hard for a university or researcher to turn down an offer of money, particularly because advancing in academia is in part tied to attracting research funding.
In a statement, Coca-Cola noted that it only supports research if other funding sources account for more than 50% of the project’s budget and pointed to its efforts to promote rigorous research and curb obesity.
The new paper relies on thousands of pages of documents from more than a hundred freedom-of-information requests to research institutions in North America, Europe, and Australia. The documents were obtained by US Right to Know, a nonprofit research group, and contained five agreements and related correspondence between Coca-Cola and academic groups.
The paper’s authors noted that it was difficult to draw conclusions from the small sampling of agreements and emails they collected, but said that the documents pull back the curtain as individual examples of how a company interacts with the recipients of its research support.
Overall, the new paper reported, the agreements show that Coca-Cola did not direct or control the research process, but that the company was allowed to review and comment on the research before it was submitted to a journal. The research teams, however, were empowered to reject the company’s suggestions (unless they tied to certain confidential business issues). Coca-Cola also required research teams to disclose they had received funding from the company.
The language of the termination provisions varied among the agreements, but in one 2012 contract for a project at Louisiana State University’s Pennington Biomedical Research Center, it read that once the research team was told that the project was being terminated, it “will immediately discontinue all work under this Agreement and return all copies of Sponsor data, or other materials, and deliver to sponsor all work in progress, including incomplete work …”
“Such termination provisions could, hypothetically, allow Coca-Cola to quash studies progressing unfavorably, or allow Coca-Cola to pressure researchers using the threat of termination,” Steele and her colleagues wrote in the paper, though they added they found no evidence that that had happened in any of the cases they reviewed.
In response to a question from STAT, Pennington Biomedical pointed to a 2016 statement that said: “The funders of our grants and contracts do not determine research outcomes, the science always does. … Results of our research are published in leading peer-reviewed scientific journals, whether the outcome is favorable to the sponsor or not.”
In one case the paper documented, a researcher pushed back on the terms of the agreement and was able to get it changed. At the University of Toronto, a draft of the research agreement said that the university would provide Coca-Cola “the prior right to review and approve (or reject) any communication or other material developed by U of T or its employees” about the research, the funding, and the relationship between the institution and the company.
In an email to a Coca-Cola official, Dr. John Sievenpiper, an associate professor of nutritional sciences at the university, wrote that the language was “very restrictive for being an ‘unrestricted grant.’” Eventually the wording was changed to say that the two parties would “consult with each other in good faith regarding any communication.”
In a statement to STAT on Tuesday, Sievenpiper said: “Coca-Cola was not involved in the design, conduct, analysis, interpretation or the decision to publish any of our research.” He noted he pushed to change the language in the agreement so that it fulfilled the expectations of an unrestricted grant.
Marion Nestle, a professor emerita of nutrition, food studies, and public health at New York University, who was not involved in the new paper, said it showed “evidence that Coca-Cola’s standard agreements with researchers permitted the company to control aspects of the studies that should be entirely up to researchers.” She highlighted how Sievenpiper was able to negotiate the language of the agreement in his case, a sign to her that other universities and researchers should push to ensure the integrity of the research.
“Research is supposed to be investigator-initiated and controlled, from development of the research question all the way to publication,” Nestle wrote in an email. “Research controlled by commercial funders in this way is aimed at marketing purposes, not science. If investigators accept industry funding for research, they need to establish a firewall between the funding and the study. If not, they are doing marketing research, not science.”