Texas A&M University is defending a study that recommends Americans continue their current consumption of red and processed meat.
The study has drawn criticism from nutrition experts, including some Harvard University faculty members. The faculty members and others publicly criticized the research as detrimental to public health and influenced by industry.
Last week, A&M University System Chancellor John Sharp sent a fiery open letter to Harvard University President Lawrence Bacow calling for an ethics review of several Harvard faculty members, mentioning some by name. Sharp accused them of mischaracterizing scientific research by Texas A&M AgriLife professors and of falsely alleging that Texas A&M scientists sold out to industry interests.
Texas A&M AgriLife, which calls itself the nation’s largest comprehensive agriculture program, said the study's authors were interested in the quality of science that informs public health.
Controversial New Guidelines
The research in question is a study on red and processed meat published last year in the Annals of Internal Medicine. Reviewing extensive evidence from prior research, an international team of physicians and scientists generated new dietary guidelines suggesting Americans not cut back on consumption of red and processed meats.
According to the World Health Organization’s International Agency for Research on Cancer (IARC), processed meat is a group one carcinogen. An IARC press release from 2015 said researchers found that each 50-gram portion of processed meat eaten daily increases the risk of colorectal cancer by 18%.
“Two of the most settled matters in modern nutrition are that the typical American diet is harmful largely because of excesses of added sugar and excesses of processed meat,” David Katz said.
Katz is the founding director of the Yale-Griffin Prevention Research Center and president of the True Health Initiative (THI). THI focuses on reducing preventable diseases and correcting misinformation about dietary and lifestyle science. It advocates for plant-based diets, among other practices. Katz does not receive financial compensation for his work with THI. Sharp's open letter mentioned Katz by name.
Katz said the authors of the Annals study violated the Hippocratic Oath.
“Guidelines to the public in a major medical journal advising people to keep eating current levels of processed meat, despite data — on which those very guidelines were purported to be based — showing higher levels of death, heart disease, cancer and diabetes with higher intake of processed meat, violate every bedrock principle of medical and public health ethics,” Katz said.
Gordon Guyatt, a physician and distinguished professor of medicine at McMaster University in Canada, co-authored the Annals study. The new study made the recommendation that it did, Guyatt said, because the health risks are outweighed by the benefits, including Americans’ affinity for red and processed meats.
“So at the end of the day, maybe red and processed meat does cause cancer or bad heart outcomes, but we’re really not sure,” Guyatt said. “Secondly, even if it does, the effects were very small.”
The team weighed the benefits by analyzing the public’s health-related values and preferences towards red and processed meat. They found that Americans have a strong affinity for red and processed meats for cultural reasons, that the public believes meat is hard to replace with other food items and that most Americans would be unwilling to cut meat from their diets.
Although many nutrition experts and health organizations may disagree, Guyatt said the science around the health impacts of red and processed meat remains undecided.
“I am the leader in the academic and scientific community,” Guyatt said. “So to make a statement that it's already been decided is simply inaccurate.”
The Annals study looked at both observational research and randomized clinical trials, but the bulk of the evidence came from observational research.
“For most dietary practice, especially for red meat and other food groups, it’s not possible — or even ethical — to conduct large, long-term randomized clinical trials to look at whether overconsumption of red meat will cause diabetes, heart attack or some cancers,” Frank Hu said.
Hu is the chair of the Department of Nutrition and a professor of nutrition and epidemiology at Harvard University, and he sits on the THI council. He does not receive financial compensation for his work with THI. Sharp’s open letter also mentioned him by name.
Instead of randomized clinical trials — like the kind used in pharmaceutical drug research — most nutrition science is observation-based. This observational research tracks people’s diets over time and measures health outcomes. It can be affected by several variables, like smoking and other unhealthy lifestyle choices.
The authors used a system called “Grading of Recommendations, Assessment, Development and Evaluation" (GRADE) to evaluate the quality of evidence that red and processed meat consumption leads to poor health outcomes. Using GRADE, the authors rated that evidence as low to very low certainty.
Hu said the GRADE evidence-rating system is inappropriate for most observation-based nutrition research because it is designed for analysis of randomized clinical trials. He said an independent team in Germany developed a different system specifically to evaluate evidence in dietary research.
According to Hu, when the German team’s system was applied to the research analyzed by the Annals study, the evidence for adverse health outcomes tied to red and processed meat consumption was rated as moderate to high certainty.
“And if that system were used in the Annals papers, then the conclusions would be dramatically different from what they draw in their papers,” Hu said.
The GRADE system also rates the strength of recommendations based on the certainty of evidence. Using GRADE, the Annals authors rated their own recommendations — that Americans continue current consumption of red and processed meat — as weak.
Hu and Katz both took issue with the study’s use of what the authors themselves described as low-certainty evidence and weak recommendations to generate new guidelines. They argued the word “guidelines” should require a higher threshold of evidence.
Jason Morrow is a physician and clinical ethicist at the University Health System and UT Health San Antonio. He said the authors’ use of low certainty evidence to generate weak-rated recommendations is common.
“But when they are citing weak evidence and giving weak recommendations, they usually don't conclude with ‘And therefore carry on with what you're doing.’” Morrow said. “Usually they'll say, ‘There's no evidence to stop what you're doing,’ or something along those lines.”
In other words, the explicit recommendation to continue consuming red and processed meat was unusual, considering the uncertain evidence on which the guidelines rest.
Texas A&M Responds To Criticism
On New Year’s Eve, the Annals of Internal Medicine published additional disclosures that were not included in the initial article. The disclosures revealed that Johnston received a grant from Texas A&M AgriLife for separate research into saturated and polyunsaturated fats. At the time of the Annals study, he was employed by Dalhousie University in Canada. He is now employed by Texas A&M AgriLife as a tenured associate professor in the department of nutrition. Stover is the vice chancellor and dean for Agriculture and Life Sciences at Texas A&M AgriLife.
This is a moment in history to make a difference in science. Let’s put aside ad hominem attacks and deal with the science. https://t.co/Vyaq9BKZNR
— Patrick J Stover (@patrickjstover) January 16, 2020
A spokesperson with Texas A&M AgriLife told TPR the study wasn’t funded by industry sources and that Johnston has never received industry funding. AgriLife did not provide funding for the Annals study.
In 2019, the Texas A&M AgriLife Research program received more than $10 million in funding for research on beef, lamb and pork products — that’s under 6% of the program’s total revenues.
AgriLife's financial sources include the United States Department of Agriculture, the National Institutes of Health, the state of Texas, agriculture industry groups, corporate entities and nonprofits.
Morrow — the University Health System physician and clinical ethicist — said funding from industry groups for research issues relevant to that industry is standard across science.
“These folks are tied to industry throughout their careers. They work for state institutions that get funding from various levels, including — in some part — industry,” Morrow said. “And that's just ordinary. We live in a time where funding for research is the name of the game. Research won't get done without some sort of funding source. So, we can't say that science is bad because it takes money to do so, and whoever pays the piper might call the tune.”
In his open letter to the president of Harvard University, A&M Chancellor Sharp said Stover, Johnston and other A&M researchers were independent.
“I can assure you that Texas A&M’s research is driven by science,” Sharp wrote. “Period.”
Harvard University confirmed reception of the letter but declined to comment on it.
One of the study’s coauthors — Chirag Patel — is an assistant professor of biomedical informatics at Harvard Medical School. He also declined to comment.
Guyatt was the only Annals study coauthor contacted by TPR who agreed to be interviewed. He said the majority of the authors had no financial conflict.
Hu said the issue was broader than perceived financial conflicts on the part of the study’s authors. He said public trust in science was at stake.
“Scientific research is, of course, an evolving process, and contradictions between published studies are inevitable, and that's a healthy part of this process,” Hu said. “However, these contradictions can cause confusion, and especially considering the media reports and the journals and internet bloggers tend to oversimplify nutrition studies.”
In this case, Morrow explained, the Annals study included no new evidence. And as part of its recommendations, three people on a panel of fourteen — which included eleven experts and three non-medical community members — actually voted against the final guidelines that were issued. But he said many members of the public who heard about the study through the media were likely unaware of those nuances.
Morrow said academic and medical journals should be careful in the framing of new health research. But, he added, the public also bears some responsibility for interpreting new scientific findings.
“There is a sense of accountability that editors and journals must have in owning how information is presented to the public, so the public can be informed consumers, which also entails a sense of active consumerism that the medical establishment and the public at large has to have,” he said. “We have to be active interpreters of the data that is presented to us.”