Processed snacks, drinks and desserts may be associated with a higher risk of cancer, according to a new study.
The research, published Wednesday in The BMJ, focused on ultra-processed foods, which tend to be high in fat, saturated fat, sugar and salt. (Past research has found that Americans get 61% of their calories from highly processed foods.) In the new study, researchers found that, among almost 150,000 French adults, a 10% increase in the proportion of ultra-processed foods in a person’s diet was correlated with a 12% higher risk of cancer. The paper is the first to explore the link between cancer and ultra-processed foods — that is, industrial formulations that typically contain many ingredients, including some not found in the typical kitchen — says study author Mathilde Touvier, an investigator at the Sorbonne Paris Cité Epidemiology and Statistics Research Center.
The research only uncovered patterns in the data, meaning it’s impossible at this point to say that ultra-processed foods cause cancer, Touvier cautions. Nonetheless, the correlation was striking.
TIME Health Newsletter
During the five years of study follow-up, about 2,200 people were diagnosed with cancer. In addition to the 12% increase in overall cancer risk, the researchers found that eating processed food regularly was linked to an 11% increase in the odds of getting breast cancer and a “borderline significant” increase in the risk of colorectal cancer. (Only 153 people got colorectal cancer, and Touvier says she believes there weren’t enough cases to prove an association.) No strong connection was observed between ultra-processed foods and prostate cancer.
Not all processed foods appeared to be equal. Drinks, sugary products, fats and sauces were most strongly associated with a heightened cancer risk, according to the paper, while sugary processed foods were most strongly linked to breast cancer.
The poor nutritional quality of ultra-processed foods likely contributes to their negative effects on health. For one, processed foods are often high in inflammatory constituents, like sugar and certain fats, and chronic inflammation can increase cancer risk. But Touvier says the study suggests that other elements may be responsible for their association with cancer. Other possible triggers could be food additives—”some of which are quite controversial, with animal studies suggesting potential carcinogenic properties,” Touvier says—as well as materials found in processed food packaging. Other possibilities include potentially cancer-causing compounds, such as acrylamide, which are produced during the industrial preparation process, she adds.
It will take further research to understand which of these hypotheses, if any, is valid, Touvier says, so it’s too soon to panic.
“We don’t want to be too alarmist,” she says. “We need other studies to confirm the results, and to increase the understanding of the mechanisms. We need to further elucidate before saying to people, ‘You’re going to die with cancer.'”