A study led by researchers from the University of Antwerp has found that ultra-processed plant-based foods contain high levels of certain chemical contaminants in comparison to their animal-based analogs. However, based on a dietary exposure risk assessment, the contaminants in these products are unlikely to pose a significant health risk for the adult population, even when eating an all-vegan diet.

For the study, published in Environmental Science and Technology, the researchers purchased 52 ultra-processed plant-based food samples from large grocery chains in Belgium, Germany, and the UK, including a wide selection of commercial brands. Samples included plant-based meat alternatives (e.g., burger, chicken, minced meat, sausage, and cold cut alternatives, as well as processed soy products like tofu), plant-based cheese alternatives, plant-based fish alternatives, and miscellaneous other items. The foods were made of ingredients like soy, legumes, vegetables, grains, mycoprotein, oils, and seeds and nuts.

The samples were tested for organophosphorus flame retardant and plasticizer contaminants using liquid chromatography-tandem mass spectrometry. Specifically, the researchers tested for 17 organophosphates, eight phthalates, and eleven alternative plasticizers.

The results showed differences in contamination levels and patterns between food categories. Plant-based cheese-alternatives, which consisted mostly of coconut oil as their base ingredient, showed the highest levels of both total organophosphorus flame retardants, with a mean of 123 nanograms per gram of wet weight (ng/g ww), and total plasticizers, with a mean of 1,155 ng/g ww. When looking at previous studies, the contamination levels of plant-based cheese alternatives were comparable to concentrations measured in commonly consumed oils and fats for both organophosphorus flame retardants and phthalates, but higher than in animal-based cheese for organophosphorus flame retardants. The researchers hypothesize that, since coconut oil was not included in the previous comparable studies, coconut oil may be a possible source of contamination. Additionally, while di-(2-ethylhexyl) phthalate (DEHP) was the main contaminant found in plant-based cheese alternatives, it could not be detected in the only sample that was made of nuts instead of coconut oil.

Processed soy products (i.e., tofu, tempeh, and dry soy) were the second most contaminated food category for both organophosphorus flame retardants (78 ng/g ww) and plasticizers (817 ng/g ww), even though such products are considered to be less processed than other plant-based alternatives. This is possibly due to migration of contaminants from food contact materials, although the researchers could not draw a link between final product packaging and product contamination, suggesting that contamination migration may occur at earlier points in the supply chain, such as during processing. Most of the contamination of the soy products was attributed to 2-ethylhexyl diphenyl phosphate (EHDPHP) and acetyl tributyl citrate (ATBC), both of which are used in plastic food contact materials. The researchers found the high levels of contamination in soy products compared to more processed categories of plant-based foods to be surprising, and hypothesized that the soy content itself may be a contributing factor, calling for additional research into the role it may play in contaminant migration.

The researchers also conducted a dietary exposure risk assessment for organophosphorus flame retardants and phthalates in flexitarian, vegetarian, and vegan diets. For sum organophosphorus flame retardants, the estimated daily intake (EDI) was 2.7 nanograms per kilogram of bodyweight per day (ng/kg bw/day) in the flexitarian scenario and 14 ng/kg bw/day in the vegetarian scenario. The vegan EDI reached 53 ng/kg bw/day, with plant-based meat alternatives being the major contributor (27 ng/kg bw/day). However, calculated EDIs in the vegan scenario were still at least 1,000 times lower than available health-based guidance values, indicating that the estimated exposure is unlikely to pose a significant health risk for the adult population.

Overall, the researchers underlined the differences in contamination levels and patterns between plant-based food categories, and the fact that, in general, the plant-based products assessed in the study showed higher contamination levels compared to their animal-based counterparts (based on previous studies). They call for more research that includes a wider variety of ultra-processed plant-based food products and contaminants in future consumption surveys, market basket studies, and dietary exposure risk assessments. Additional studies are also required to better understand specific sources of contamination along the supply chain and in the production process. Although the researchers’ dietary risk assessment did not indicate an immediate health concern for the adult population, a more robust assessment combining data of multiple compound classes is needed, and the researchers still caution against basing one’s diet entirely on ultra- processed plant-based products.