Vegan and omnivore diets in relation to nutrient intake and greenhouse gas emissions in Iceland

In this study, those adhering to vegan diets had twofold lower dietary GHG emissions than omnivores. Except for protein intake, dietary habits of vegans were more in line with recommendations for macronutrients³. Compared to omnivores, vegans were also more in line with recommendations for several micronutrients although some indications of insufficient intake of calcium and iodine were observed. Overall, it was observed that both groups could improve their intake of micronutrients. Our results suggest that more explicit guidance is needed for individuals adhering to plant-rich diets within food-based dietary guidelines, which often do not address how to achieve optimal nutrient intake among those consuming little to no animal-sourced foods.

Compared to omnivores, our vegan participants were younger, had higher education, and were more likely to be females. The higher proportions of females in our study reflects the gender distribution within the Icelandic Vegan Association from which most vegan participants were recruited. Interestingly, few vegans reported on some day’s minimal consumption of animal- sourced foods as reflected by the 90th intake percentile (see Table 2). One explanation might be that some processed foods such as confectionary products can contain small amounts of dairy and animal fats, which consumers may not be aware of²³. In addition, some vegans may, on rare occasions, intentionally choose to consume foods of animal origin for practical- or social reasons. In that context, it is worth noting that there is no clear consensus on the definition of veganism²⁴. For example, in the definition of veganism by the UK Vegan Society the consumption of foods of animal origin is partly acknowledged with the formulation “seeks to exclude, as far as is possible and practicable”²⁵ while the definition by The American Vegan society does not allow for such flexibility²⁶. Our results suggest that vegans in our study may have adopted a more flexible approach.

In our study, the total dietary GHG emissions among vegans was twofold lower compared to omnivores. Although such differences have been observed in other studies^27,28, more varying estimates have also been reported²⁹. This may occur due to unaccounted differences in energy intake, use of different environmental databases, or variation in dietary habits²⁹. In our study energy intake among vegans and omnivores was within 10% and could not explain the differences observed. Furthermore, in a separate study, we have assessed the impact of using different databases to assess dietary GHG emissions for all participants in the Icelandic National Dietary Survey¹⁵. The resulting dietary GHG emissions estimates were within 10% in all cases. It therefore appears unlikely that our results are confounded by energy intake or choice of environmental database. Furthermore, our results are also in line with the growing literature which clearly shows that vegans compared to omnivore diets have substantially lower dietary GHG emissions^27,28,30.

Overall, the dietary habits of vegans were more compliant in relation to macronutrient recommendations³. That is, carbohydrate quality was generally higher as reflected by higher proportion of vegans meeting recommendations for total carbohydrates, dietary fiber, and added sugars. The total fat and saturated fat intake of vegans was also more in line with recommendations. In terms of debates on whether there is sufficient evidence to relax the current recommendation for saturated fat intake^31,32, it is noteworthy that even among vegans in our study only 54% complied with this recommendation while the corresponding number of omnivores was only 4%.

On the other hand, fewer vegans complied with the recommended intake for protein³. Lower protein intake among vegans compared to omnivores is also commonly observed in other studies^29,30,31. Despite lower protein intake among vegans, the median intake (62 g/day or 0.87 g/kg body weight) was well above the estimated average requirement for adults (0.66 g/kg body weight), which does not raise any major concern in terms of protein deficiency. However, such intake may still lead to increased vulnerability during sensitive life stages such as during pregnancy and older age when protein requirement increases^3,33. To the best of our knowledge, no studies among older adults adhering to vegan diets are available. However, in the case of pregnant women a study comparing birth outcomes among omnivores, vegetarians, and vegans observed around 200 g lower birth weight of infants among vegan compared to omnivore mothers³⁴. In that study the mean intake of protein among vegans was around 10% of total energy which is slightly lower than the 12% observed in our study.

Concerning food processing our findings that consumption of ultra-processed foods was comparable (~ 46% of energy intake) among vegans and omnivores seems to reflect the growing availability of processed plant-rich foods for vegans³⁵. Although the NOVA classification system has its limitations³⁶, it does provide additional information on the type and source of the food consumed. This in combination with assessment of nutrient intake provides a more complete picture of the participants’ dietary habits.

When quantifying combined micronutrient intake from food and supplements, the median intake among vegans and omnivores only slightly exceeded the average requirement for several nutrients. For vegans, the median intake of calcium and iodine was below the average requirement, which may suggest some risk of nutritional insufficiency¹⁸. Similar indications have also been observed in a small study on vegans from Germany³⁷. In the Nordic countries and the UK, iodine deficiency has been improved either through iodized salt and/or iodine fortification of cow’s fodder to increase the iodine content of dairy^38,39. The indirect fortification of dairy does, for obvious reasons, not benefit vegans but similar measures for other foods could in the same way improve their intake⁴⁰. This also applies to calcium where dairy alternatives are not consistently fortified⁴¹. In the absence of such measures, more emphasis on specific supplemental use among vegans seems warranted.

In terms of limitations, our results on micronutrient intake should be interpreted with some caution as the use of a two-day 24-hour recall may inflate the number of subjects estimated to have both low and high intakes⁴² and accurate quantification of nutrients from food supplements can be challenging⁴³. This combined with the relatively small sample size of vegans in our study (n = 68) is a source of uncertainty, particularly when trying to estimate the prevalence of those within and outside certain dietary reference values. On the other hand, the strength of our study lies in identical assessment and comparison between vegans and omnivores, characterizing their intake in terms of nutritional guidelines and assessing the dietary carbon footprint. Such direct comparisons with official recommendations have been largely lacking, at least in the Nordic setting³.

In conclusion, we observed that those adhering to vegan diets were overall more compliant with meeting macronutrient recommendations compared to omnivores and had substantially lower dietary GHG emissions. When evaluating nutrient intake from food and dietary supplements, possibilities for improvements in terms of meeting recommendations were observed in both groups. Our results suggest that food-based dietary guidelines should differentiate between vegans and omnivores, considering the varying nutritional risks that may be associated with each diet. Our results also indicate that in our omnivore population a shift towards a more plant-rich diet without eliminating all foods of animal origin could lead to substantial nutritional and environmental benefits.