- Certain preservatives are often added to foods to make them last longer, but research about the possibly damaging effects is ongoing.
- One study found that preservative food additives are linked to a higher incidence of diabetes, and another found a similar link between certain preservative food additives and some cancers.
- People can take steps to minimize their consumption of food preservatives.
Do preservatives carry health risks? One study recently published in
Another study from the same research team, looking at data from the same original cohort, found that higher intake of certain preservatives was associated with a greater incidence of certain cancer types and overall cancer. This study was published in
The results highlight potential dangers of certain food additives and the possible need for people to minimize these in their diets.
For the first study, the researchers wanted to examine how preservative food additives related to the incidence of type 2 diabetes. The study used the French NutriNet-Santé prospective e-cohort, ultimately including 108,723 participants, excluding individuals such as those who already had confirmed diabetes.
The average age of participants was around 43 years, and they had completed an average of 21, 24-hour dietary recalls.
Researchers identified 17 preservatives that at least 10% of the participants consumed. Citric acid was the additive that was consumed by the greatest percentage of participants.
Researchers focused on these 17 preservatives but also accounted for 58 preservatives in the sum of preservative categories. About one-third of the food additive preservatives that participants consumed came from ultra-processed foods.
The median follow-up time with participants was about 8 years. Throughout the 14-year follow-up, researchers identified 1,131 cases of type 2 diabetes.
They found that greater intake of total preservatives, totals of several additives like total nitrites, and individual additives were linked to a greater incidence of type 2 diabetes. The findings mostly held true for the sensitivity analyses as well, with only total ascorbates reaching a level of non-significance.
The data also identified that exposure to ultra-processed food was linked to a higher incidence of type 2 diabetes, and 17% of this association was mediated by preservative exposure.
For some substances, the data indicated a plateau effect, where likely even higher intake of the substance would not affect diabetes risk once intake reached a certain point.
Similar to the study on type 2 diabetes, the study on preservative food additives and cancer found some possible links between some food preservatives, overall cancer, breast cancer, and prostate cancer risk.
Researchers considered data from the same original cohort as the study on food preservatives and diabetes, ultimately including 105,260 participants in their analysis.
Throughout the follow-up, 4,226 participants developed cancer, with breast cancer comprising 1,208 cases. While the researchers did not find an association between cancer incidence and total preservatives, they found that certain preservatives and groups had a link to higher cancer incidence.
For example, potassium sorbate was associated with a greater incidence of breast cancer and overall cancer, and total sulfites were associated with a greater incidence of overall cancer.
Mathilde Touvier, PhD, Higher Degree by Research, Head of the Nutritional Epidemiology Research Team (CRESS-EREN) at Inserm (French National Institute for Health and Medical Research), and Anaïs Hasenbölher, doctoral researcher in the Equipe de Recherche en Epidémiologie Nutritionnelle: Bobigny, who were authors on both studies, summarized the findings to Medical News Today:
“Widely consumed preservative food additives — such as sodium nitrite, sulfites, sodium erythorbate — were associated with a higher incidence of cancer and type 2 diabetes in this large prospective cohort of French adults, NutriNet-Santé (more than 100,000 participants) […] Out of the 58 food additive preservatives consumed by participants, 17 were consumed by at least 10% of the population study, allowing us to investigate their specific association with cancer and diabetes risks. It turned out that six out of these 17 preservative food additives were associated with a higher incidence of cancer (overall, prostate, and/or breast), and 12 out of 17 with a higher incidence of type 2 diabetes.”
The studies did face certain limitations, even considering the large datasets. For the studies on diabetes, the data on participants’ food intake came from 24-hour dietary records, which may be incorrect, and there may have been be other method-related limitations.
The researchers acknowledge the possibility of classification bias, but the dietary intake estimates for this cohort are very accurate.
It is also possible that researchers missed aspects of preservative consumption or food intake, and the researchers acknowledge that “specific exposure to preservative food additives was not part of these validation studies,” which was the data collection on diet.
Other self-reported aspects, such as sex, could have affected the study results as well. Researchers note that the “cut-offs for exposure categories were sex-specific.”
The researchers acknowledge that the study cannot determine cause and that “residual confounding cannot be fully ruled out.”
They note a caution in generalizing the results because the study had a greater number of women, participants had a higher education level, and participants followed healthier lifestyles than the general population in France.
They also note that these factors, and allowing participants to enroll in the cohort at the age of 15, could explain the low level of diabetes cases. They do believe there is a low risk of missed diabetes cases and thus a lower risk for bias from this, but one aspect of identifying cases was from participant reporting.
Researchers had to use a database from another country to estimate natural occurrences of two substances, and country variation is possible. They also could not quantify “the natural intake of some substances.” Also, not all participants completed all records regarding food intake.
They did not have data on participants’ race, ethnicity, or religion, but “the geographical distribution of the cohort also matches that one of the general population in mainland France.”
They could not examine food additives that participants consumed less of, and it is possible that there are other relevant additives that researchers did not consider.
Finally, researchers showed the results “with and without adjustment for multiple testing,” and note that adjusting for multiple testing carries some problems. However, they report the results as mostly stable.
When examining the study on diabetes and additives, Randa Abdelmasih, MD, an assistant professor of medicine in UTMB’s Division of Endocrinology and Metabolism, who was not involved in this research, expressed some further notes of caution to MNT:
“The use of time-dependent cumulative exposure models fits well with what we know clinically about diabetes as a condition that develops over many years, but some exposure misclassification is still possible as food formulations change… [The study] cannot determine whether the observed associations reflect direct effects of additives themselves, interactions with the food matrix, or correlated dietary behaviors, underscoring the need for further mechanistic and interventional studies.”
Both studies focused on food additives in the European market, so they might not necessarily speak to the specifics in other regions, and since the cohort was French, the results might not be generalizable to other groups.
Any limitations related to the diabetes cohort would have carried over into the cancer cohort, including the self-reporting of data.
The methods used by researchers, such as what they accounted for in the models, could have missed relevant information, and the study does not establish a causal relationship.
They researchers also worked with limited data in some areas, including only a small number of participants having a colorectal cancer diagnosis, which led to limited statistical power.
Overall, however the data from these two studies suggests some possible reasons for limiting the consumption of preservatives.
Karen Z. Berg, MS, RD, CDN, who was not involved in the research, offered some expert tips on how to achieve this:
“Read food labels, and that doesn’t just mean the nutrition facts panel. It’s super important to read the ingredient list as well. If you can’t pronounce or recognize several ingredients, it’s probably something you should opt to stay away from. Try to stick to whole foods as much as possible to help minimize additives in your diets. Go for plain yogurt and add your own fruit or honey, go for homemade oatmeal, and you can add your own maple syrup and cinnamon, etc. When manufacturers make our lives easier (or tastier), there’s unfortunately a detrimental cost that comes with that.”
However, it may also be helpful to look at the broader context as well.
Abdelmasih noted that, “from an endocrinologist’s perspective, the most helpful advice is to focus on overall eating patterns rather than trying to avoid individual additives.“
“Attempting to eliminate specific preservatives one by one is neither practical nor supported by current evidence,“ she emphasized.
“What tends to make the biggest difference metabolically is prioritizing foods that are fresh or minimally processed, which naturally reduces exposure to additives without requiring people to scrutinize every ingredient label,“ Abdelmasih explained.
“Emphasizing home-prepared meals when possible can help reduce reliance on highly preserved convenience foods, but perfection is neither realistic nor necessary,” she advised.
