The compatibility of circularity and national dietary recommendations for animal products in five European countries: a modelling analysis on nutritional feasibility, climate impact, and land use

Abstract

Background
National food-based dietary guidelines (FBDGs) are generally designed from a human health perspective and often disregard sustainability aspects. Circular food production systems are a promising solution to achieve sustainable healthy diets. In such systems, closing nutrient cycles where possible and minimising external inputs contribute to reducing environmental impacts. This change could be made by limiting livestock feed to available low-opportunity-cost biomass (LOCB). We examined the compatibility of national dietary guidelines for animal products with livestock production on the basis of the feed supplied by available LOCB.

Methods
We investigated whether the national dietary recommendations for animal products for Bulgaria, Malta, the Netherlands, Sweden, and Switzerland could be met with domestically available LOCB. We used an optimisation model that allocates feed resources to different species of farm animals. Of the resulting scenarios, we assessed the nutritional feasibility, climate impact, and land use.

Findings
Our results showed the environmental benefits of reducing the recommended animal products in the FBDGs, and that animal products from LOCB could provide between 22% (Netherlands) and 47% (Switzerland) of total protein contributions of the FBDGs. This range covers a substantial part of the nutritional needs of the studied populations. To fully meet these needs, consumption of plant-based food could be increased.

Interpretation
Our results contribute to the discussion of what quantities of animal products in dietary guidelines are compatible with circular food systems. Thus, national dietary recommendations for animal products should be revised and recommended quantities lowered. This finding is consistent with recent efforts to include sustainability criteria in dietary guidelines.

Funding
Swiss National Science Foundation and the Dutch Research Council.

Full text

www.thelancet.com/planetary-health Vol 6 June 2022
e475
Articles
Lancet Planet Health 2022;
6: e475–83
Department of
Socioeconomics, Research
Institute of Organic Agriculture
(FiBL), Frick, Switzerland
(A Frehner PhD, A Muller PhD,
C Schader PhD); Animal
Production Systems Group
(A Frehner,
Prof I J M de Boer PhD,
B van Selm MSc, O van Hal PhD),
Farming Systems Ecology
Group (A Frehner,
R P M Cardinaals MSc,
H H E van Zanten PhD),
and Plant Production Systems
Group (B van Selm),
Wageningen University &
Research, Wageningen,
Netherlands; Institute of
Environmental Decisions,
Federal Institutes of
Technology Zurich (ETHZ),
Zurich, Switzerland (A Muller);
Division of Chronic Disease
Epidemiology, Epidemiology,
Biostatistics, and Prevention
Institute, University of Zurich,
Zurich, Switzerland
(G Pestoni PhD,
S Rohrmann PhD); Nutrition
Group, Health Department,
Swiss Distance University of
Applied Sciences, Zurich,
Switzerland (G Pestoni);
Department of Global
Development, College of
Agriculture and Life Sciences,
and Cornell Atkinson Center for
Sustainability, Cornell
University, Ithaca, NY, USA
(Prof M Herrero PhD)
Correspondence to:
Dr Anita Frehner, Department of
Socioeconomics, Research
Institute of Organic Agriculture
(FiBL), CH-5070 Frick,
Switzerland
anita.frehner@fibl.org
The compatibility of circularity and national dietary
recommendations for animal products in five European
countries: a modelling analysis on nutritional feasibility,
climate impact, and land use
Anita Frehner, Renée P M Cardinaals, Imke J M de Boer, Adrian Muller, Christian Schader, Benjamin van Selm, Ollie van Hal, Giulia Pestoni,
Sabine Rohrmann, Mario Herrero, Hannah H E van Zanten
Summary
Background National food-based dietary guidelines (FBDGs) are generally designed from a human health perspective
and often disregard sustainability aspects. Circular food production systems are a promising solution to achieve
sustainable healthy diets. In such systems, closing nutrient cycles where possible and minimising external inputs
contribute to reducing environmental impacts. This change could be made by limiting livestock feed to available low-
opportunity-cost biomass (LOCB). We examined the compatibility of national dietary guidelines for animal products
with livestock production on the basis of the feed supplied by available LOCB.
Methods We investigated whether the national dietary recommendations for animal products for Bulgaria, Malta,
the Netherlands, Sweden, and Switzerland could be met with domestically available LOCB. We used an optimisation
model that allocates feed resources to dierent species of farm animals. Of the resulting scenarios, we assessed the
nutritional feasibility, climate impact, and land use.
Findings Our results showed the environmental benefits of reducing the recommended animal products in the
FBDGs, and that animal products from LOCB could provide between 22% (Netherlands) and 47% (Switzerland) of
total protein contributions of the FBDGs. This range covers a substantial part of the nutritional needs of the studied
populations. To fully meet these needs, consumption of plant-based food could be increased.
Interpretation Our results contribute to the discussion of what quantities of animal products in dietary guidelines are
compatible with circular food systems. Thus, national dietary recommendations for animal products should be
revised and recommended quantities lowered. This finding is consistent with recent eorts to include sustainability
criteria in dietary guidelines.
Funding Swiss National Science Foundation and the Dutch Research Council.
Copyright © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND
4.0 license.
Introduction
The production of human food aects the environment
in multiple ways, including its associated resource use
that alters global biomass and nutrient cycles, its eects
on climate change, and biodiversity loss.1 Unbalanced
diets that are low in fruit and vegetables, and high in red
and processed meat are a major risk factor for several
non-communicable diseases, such as cardiovascular
diseases, stroke, cancer, and diabetes.2 In high-income
countries, shifting consumption towards plant-based
diets is often recommended, to decrease environmental
impacts of food consumption and to improve human
health benefits of diets.3 This recommendation is due to
the generally favourable environmental eects of plant-
based food compared with animal products,1 as well as
the increased risk for diet-related diseases in the case of
low fruit and vegetable consumption, and high red and
processed meat intake.4 Food-based dietary guidelines
(FBDGs) are key references for healthier food choices.
Although environmental concerns are increasingly
addressed in FBDGs, for example, in the 2019 EAT–Lancet
Commission5 and in several national FBDGs (eg, Sweden
and Germany), most national FBDGs are still primarily
driven by health and nutritional criteria and often do
not include sustainability aspects.6 Compared with
globally applicable guidelines, such as the EAT–Lancet
Commission, national FBDGs take geographical and
cultural circumstances into consideration,7 and are often
well embedded in education and nutrition counselling at
the national level.8
Although the necessity to reduce the consumption and
production of animal products is generally acknowledged,
dierent solutions exist regarding how animal products
could be more sustainably produced, and which animal
products should be reduced and to what extent.9,10 From
a supply perspective, studies suggest that animal

Articles
e476
www.thelancet.com/planetary-health Vol 6 June 2022
production systems should be intensified, which would
result in lower environmental impacts per quantity of
animal products produced but would require higher
concentrate feed inputs given that growth could be
faster.11 From a demand perspective, studies often
recommend reducing consumption of animal products
substantially or to a minimum.1,12 In both narratives, land
suitability and therefore competition between resources
for feed and for food production is mostly not
addressed.13,14 Naturally, when resources are suitable to be
allocated for feed and for food production, choices need
to be made that have consequences for the sustainability
of the food system. In circular food systems, resources
are prioritised for human food first, and animal feed is
allocated as a second priority.15,16 A guiding principle for
this type of system is to close nutrient cycles where
possible and to minimise external inputs, such as feed
and mineral fertiliser imports. Animals would then be
fed with primarily domesti cally available low-opportunity-
cost biomass (LOCB; eg, processed by-products, food
waste, and grass resources), which is also known as the
concept of ecological leftovers in the literature.17
Subsequently, feed–food competition would be largely
avoided and biomass could be used more eectively.
Through this process, animals can contribute to recycling
biomass and nutrients back into the food system, which
would otherwise be lost for human food consumption.18
Considering that recom mendations in national FBDGs
for most high-income countries are currently driven by
health and nutritional aspects and are based on the
current linear food system, the role of animal products in
FBDGs from the perspective of the environment and
ecient resource use is unexplored.
First, we investigated whether it would be feasible to
produce the animal products recommended in national
FBDGs on the basis of LOCB that would be available when
the FBDGs were adhered to, meaning that domestically
available plant-based food quantities would correspond to
the FBDG suggestions, including imports where necessary
owing to insucient domestic production. The LOCB
available for production of animal products would then be
derived from these plant-based commodity quantities,
their processing and waste fractions, and the domestically
available grassland production. By use of a resource
allocation model, which was originally developed by van
Hal and colleagues19 and adapted by van Selm and
colleagues,20 we assessed dierent scenarios for
five European countries—namely, Bulgaria, Malta, the
Netherlands, Sweden, and Switzerland. Finally, we
assessed the climate impact and land use of these
alternative scenarios. We explored the nutritional option
space that fulfils nutritional require ments that animal
Research in context
Evidence before this study
Many studies have investigated the nutritional and
environmental consequences of adhering to dietary guidelines.
We searched the databases Scopus and Web of Science for
studies published between Jan 1, 1990, and Feb 19, 2021 in
English. We used the following search terms: (“food-based
dietary guideline“ OR “dietary recommendation” AND “animal*
food” OR meat AND “environmental impact” OR “greenhouse
gas emissions” OR “resource suitab*” OR “circular*”). A study
published in 2020 assessed healthiness and sustainability of
85 national dietary guidelines. Additionally, a large number of
studies assessed nutritional and environmental consequences
of adopting specific national dietary guidelines. None of these
studies investigated the compatibility of animal-source food
recommendations with circularity principles. Based on
two review articles (one from 2018 and one from 2020) and
their citation record, we identified 20 studies that considered
the circularity principle of feeding only low-opportunity-cost
biomass (LOCB) in scenarios with differing levels of animal-
source foods. However, no study was found that addressed
resource suitability and feed–food competition in national
dietary guidelines, and investigated the potential of limiting
livestock to LOCB for such guidelines.
Added value of this study
To our knowledge, this is the first study that assesses
environmental consequences and nutritional contributions of
national food-based dietary guidelines while considering
circular food system principles. We applied our approach to
five case studies in Europe (Bulgaria, Malta, the Netherlands,
Sweden, and Switzerland), and thereby provide a proof of
concept for contrasting situations with different geographical
and cultural settings. Although we found that all national
guidelines recommend more and different animal-source food
than would be optimal from a resource-use and environmental
perspective, we also reported substantial differences between
the five case study countries. These findings stress the
importance of including environmental considerations in
national guidelines, and provide an estimate for potential
targets for the inclusion of circular livestock in dietary
guidelines as well as in current diets.
Implications of all the available evidence
Meeting the recommended amounts of animal product
consumption currently stated in national dietary guidelines
will not be feasible with only circular livestock systems. When
limiting livestock feed to LOCB, recommended amounts of
animal product cannot be reached. This amount could
become feasible, and environmental impacts could be
reduced, if recommendations for animal products were
lowered on the basis of sustainability criteria, or when the
targets are to achieve a proportion of consumption from
circular systems. The composition and quantity of animal
product recommendations should be revised with regard to
both national resource suitability and specific nutritional
requirements that animal products can provide.

Articles
www.thelancet.com/planetary-health Vol 6 June 2022
e477
products could provide without feed–food competition,
and what role animal products could have in balanced
diets.
Methods
Study design and system boundaries
Europe was selected as a case study region because of its
stable food security situation and better quality data,
where discussions on food choices mainly revolve around
issues regarding overnutrition and the search for healthy
and sustainable food choices.21 Within Europe, dietary
habits dier due to cultural habits and resource
endowments. We identified all European countries that
provide detailed dietary guidelines and selected five of
them representing dierent regions with diering
dietary habits as case studies: Bulgaria (eastern Europe),
Malta (southern Europe), the Netherlands (western
Europe), Sweden (northern Europe), and Switzerland
(central Europe). We collected data for the food groups
generally present in FBDGs (appendix pp 1–4).
Plant-based food recommendations
The collected FBDG recommendations were translated
into a daily diet (g per capita per day) by disaggregating
recommendations for food groups (eg, cereals) into food
items (eg, wheat and products, and rye and products),
based on Food Balance Sheets22 (appendix p 1). After
these transformations, we obtained an example FBDG
diet per country, of which we retained only the plant-
based food element for the subsequent analysis. The
availability of LOCB (processed by-products, food waste,
and grass resources) in each case study country was
based on the recommended plant-based food intake
when assuming the whole population would follow the
country-specific FBDG diet (appendix pp 4–5).
Animal product scenarios
We investigated the potential contribution of animal
products to a balanced diet in four dierent scenarios for
each national FBDG. We solely focussed on animal
products and therefore did not aim to provide realistic
alternatives for the FBDGs used, but rather to explore the
full range of options on the basis of LOCB for dierent
nutritional foci. The scenarios all met the circular food
system principle of avoiding feed–food competition,
meaning that animal products only originated from
animals fed on LOCB. The availability of LOCB was
restricted to the production pattern resulting from the
plant products of the respective national FBDGs.
In the first alternative scenario, MaxProt, LOCB was
allocated to the dierent animal production systems such
that human-digestible animal protein was maximised.
Along with protein, animal products contain multiple
essential nutrients for humans, such as essential fatty
acids, vitamins A, D3, and B12, calcium, iron, and zinc.23
To take the specific nutritional functions of animal
products in the diet into account, we used three scenarios
that put dierent emphasis on three main nutrient
groups: omega-3 fatty acids (scenario MaxFattyAcids),
minerals (scenario MaxMinerals), and vitamins (scenario
MaxVitamins). In each of these scenarios, one of the
respective groups of nutrients was maximised instead of
protein. The scenario MaxFattyAcids maximised the sum
of the omega-3 fatty acids α-linolenic acid (ALA),
docosahexaenoic acid (DHA), and eicosapentaenoic acid
(EPA; in g), the scenario MaxMinerals maximised the
sum of the minerals calcium, iron, and zinc (in mg), and
the scenario MaxVitamins maximised the sum of vitamins
A and B12 (in μg). Vitamin D3 was not considered because
ultraviolet B radiation from sunlight is the main source of
its synthesis and only a small proportion is derived from
dietary sources. We applied these three scenarios to show
which nutritional functions of the original FBDGs could
be met with animal products from LOCB, and which
animal products are essential for which nutritional
functions.
Resource allocation model
We used a resource allocation model, for which details
have been published elsewhere,19,20 to estimate potential
animal products on the basis of calculated LOCB. The
model contains a detailed representation of seven animal
production systems (dairy cattle, beef cattle, laying hens,
broiler chickens, pigs, Atlantic salmon, and Nile tilapia),
and allocates feed resources to the dierent animal
production systems while maximising dierent
nutritional contributions (appendix p 5).
Nutritional contribution
We quantified the following nutrient contributions of the
FBDG diets: protein; minerals calcium, iron, and zinc;
vitamin A; vitamin B12; and omega-3 fatty acids ALA,
EPA, and DHA. Nutritional contributions were calculated
with food composition tables (appendix p 1). By
multiplying quantities per food item of the dierent
FBDG diets as well as the scenarios with the nutrient
contents, we derived the total nutritional contribution.
Environmental impact assessment
Greenhouse gas emissions and land use of the scenario
MaxProt were assessed for each case study country. For
comparison with the original FBDGs, the plant-source
element was added to the animal products scenario
MaxProt. Further, to show the environmental impacts of
comparable protein content, the MaxProt scenario with the
plant-source element was scaled to the protein content of
the original FBDG for each of the five countries, resulting
in the scenario MaxProt (scaled FBDG), and to the protein
recommendation of 60 g per capita per day from WHO,24
resulting in the scenario MaxProt (scaled WHO).
Greenhouse gas emissions and land use were assessed
by use of the biophysical mass-flow model SOLm. A
detailed description of the model, including code files, is
available online.25 SOLm represents the relevant mass
See Online for appendix

Articles
e478
www.thelancet.com/planetary-health Vol 6 June 2022
and nutrient flows of agricultural production, allowing
assessment of the consequences of large-scale changes
in the food system on resource use and emissions
(appendix pp 5–6).
Role of the funding source
The funders of this study had no role in study design,
data collection, data analysis, data interpretation, or
writing of the report.
Results
The compositions of the FBDGs of the case study
countries diered by agroecological and sociocultural
context (figure 1). The protein contribution of the overall
FBDGs of the countries investigated ranged from 60 g
(Malta) to 98 g (Netherlands) per capita per day (table).
All countries except Malta issued recommendations with
higher protein contributions than the average daily
protein requirements recommended by WHO
(50–60 g),24,26 factoring in population groups with higher-
than-average protein requirements (eg, for pregnant
individuals, or for older people). Moreover, the share of
animal protein from total protein was remarkably high in
all FBDGs assessed, ranging from 0·46 (the Netherlands)
to 0·66 (Sweden).
The scenarios MaxProt, MaxFattyAcids, MaxMinerals,
and MaxVitamins show the option space of dierent
nutritional foci (figure 2). Through the optimisation
process used, results were driven by relative eciencies
(ie, nutritional contribution, such as protein, in relation
to feed requirements and availability). Generally, the
scenarios revealed trade-os between the dierent
nutrients. When fatty acids were maximised, supply of
minerals and vitamins (eg, calcium, iron, and vitamin A)
was reduced in most countries. When maximising the
three minerals calcium, iron, and zinc, mainly the fatty
acids DHA and EPA showed a substantial decrease.
Increased supply of vitamins came at the expense of fatty
acids. Overall, these trade-os per scenario were most
pronounced for the fatty acids DHA and EPA, whereas
for calcium, protein, vitamins A and B12, and zinc, the
signals were less strong. These results emphasise the
potential nutritional contributions of animal products,
which are embedded within a balanced diet.
No alternative animal product scenario of any country
was able to meet the protein contribution of the animal
products recommended in the original FBDG diets
(figure 2). The maximum achievable protein contribution
based on LOCB (scenario MaxProt) ranged from 15·9 g
protein per capita per day (Malta) to 38·9 g protein per
capita per day (Switzerland). The Netherlands could
provide 21·6 g protein per capita per day, Sweden reached
25·4 g, and Bulgaria reached 37·4 g.
In some scenarios and countries, ALA, calcium, zinc,
vitamin A, and vitamin B12 of the animal products of the
original recommendations could be met, whereas DHA
and EPA were always deficient (figure 2). For Malta, no
nutrient contribution at the same level as in the original
FBDG could be reached. For Bulgaria, all scenarios were
able to cover ALA, calcium, zinc, and vitamin A and
vitamin B12 intake. For the Netherlands, only vitamin A
reached the original contribution. For Sweden, ALA,
calcium, and vitamin A could be fulfilled, whereas others,
specifically DHA and EPA, were strongly deficient. Of all
countries, Switzerland met the most animal nutrient
contributions of the FBDG diets, with only slight
deficiencies for protein and iron, and more pronounced
deficiencies for DHA and EPA.
The dierences in nutritional composition stem from
changes in the composition of the animal products in the
scenarios (figure 3). Compared with the MaxProt
scenario, increases in fatty acids were mainly reached
with increased fish, and for Bulgaria and Switzerland
this outcome was shown with increased pork. When
minerals were the focus, eggs were substantially
increased, mainly at the expense of pork and fish.
0
25
50
75
100
Bulgaria Malta Netherlands
Country
Sweden Switzerland
Food group
Cereals
Potatoes
Vegetables
Oils and fats
Fruits
Legumes and nuts
Milk
Beef
Pork
Chicken
Eggs
Fish
Proportion of total fresh weight (%)
Figure 1: Composition of the original food-based dietary guidelines (in primary product equivalents)
Total protein
contribution
(FBDG)
Animal protein
contribution
(FBDG)
Share animal
protein / total
protein FBDG
Bulgaria 93 g 44 g 0·47
Malta 60 g 34 g 0·56
Netherlands 98 g 45 g 0·46
Sweden 85 g 56 g 0·66
Switzerland 83 g 44 g 0·52
FBDG=food-based dietary guideline.
Table: Protein contribution of FBDGs in five European countries per
capita per day, by population average

Articles
www.thelancet.com/planetary-health Vol 6 June 2022
e479
Moreover, focusing on vitamins A and B12 generally
increased milk and the associated beef production. From
a nutritional perspective, the quantity and composition
of recommended animal products is incompatible with
circular animal production.
Resource endowments largely drove the final
composition of animal products. Milk and beef supply
were mostly driven by available grass resources, which led
to minimal supply of milk and beef in Malta, where grass
resources are scarce. Grass resources were the main driver
for milk production in Bulgaria, the Netherlands, Sweden,
and Switzerland, and therefore contributed most to total
protein. Thus, although milk recommendations in the
original FBDGs were substantial, they were in agreement
with the quantity of animal products resulting from LOCB
for Bulgaria, Sweden, and Switzerland. However, chicken
and fish recommendations could never be met, but pork
and eggs recommendations could be met for Malta, in
some scenarios.
Greenhouse gas emissions and land use were
calculated for the MaxProt animal scenarios and the
plant-based food of the original FBDGs, with two scaled
versions. With the exception of Bulgaria (+25% for
greenhouse gas emissions, +5% for land use) and
Switzerland (+2% for greenhouse gas emissions), the
MaxProt scenario led to a reduction in environmental
impacts (figure 4). For Bulgaria and Switzerland, the
high amount of milk and milk products probably
contributed to the slight-to-moderate increase. The
MaxProt scenario for Sweden and the Netherlands
showed a notable decrease (Sweden: –12% for greenhouse
gas emissions, –22% for land use; Netherlands: –24% for
greenhouse gas emissions, –24% for land use). The
strongest decrease was observed for Malta, for which
greenhouse gas emissions were reduced by 56% and
land use by 40%. When the MaxProt scenarios were
scaled to the isoprotein basis of the original FBDGs, the
reductions for the Netherlands and Malta were less
0
20
40
60
0204060
Protein content of FBDG (g)
Protein content of scenario (g)
0
0·2
0·4
0·6
00·2 0·4 0·6
ALA content of FBDG (g)
ALA content of scenario (g)
0
0·1
0·2
0·3
0·4
0 0·1 0·2 0·3 0·4
DHA content of FBDG (g)
DHA content of scenario (g)
0
0·05
0·10
0·15
0 0·05 0·10 0·15
EPA content of FBDG (g)
EPA content of scenario (g)
0
250
500
750
1000
0 250 500 750 1000
Calcium content of FBDG (mg)
Calcium content of
scenario (mg)
0
1
2
3
01
23
Iron content of FBDG (mg)
Iron content of scenario (mg)
0
2
4
6
0
246
Zinc content of FBDG (mg)
Zinc content of scenario (mg)
0
200
400
600
0200 400 600
Vitamin A content of FBDG (µg)
Vitamin A content of
scenario (µg)
0
2
4
6
02
46
Vitamin B12 content of FBDG (µg)
Vitamin B12 content of
scenario (µg)
Bulgaria Malta Netherlands Sweden Switzerland MaxFattyAcids MaxMinerals MaxProt MaxVitamins
Country Scenario
Figure 2: Nutrients of the animal products in the original food-based dietary guidelines versus nutrients of the animal products in the scenarios and countries
per capita per day
Diagonal line indicates equal nutritional contributions in the scenarios and dietary guidelines. For values in the upper triangle, nutrient contribution of the scenarios
exceeds those of the dietary guidelines, and in the lower triangle (grey shading), nutrient contribution of the scenarios is lower than those of the dietary guidelines.
Horizontal differences show differences between countries and vertical differences show differences between scenarios. ALA=α-linolenic acid. DHA=docosahexaenoic
acid. EPA=eicosapentaenoic acid. FBDG=food-based dietary guideline. MaxFattyAcids=maximised fatty acid scenario. MaxMinerals=maximised mineral scenario.
MaxProt=maximised animal protein scenario. MaxVitamins=maximised vitamin scenario.

Articles
e480
www.thelancet.com/planetary-health Vol 6 June 2022
strong, whereas for Sweden, the direction of change
switched to an increase. These eects were driven by the
higher protein contents of the original FBDGs. Given
that the MaxProt scenarios of all countries in combination
with the plant-based element of the original FBDG
resulted in a higher protein content than protein
requirements of 60 g protein per capita per day,24
greenhouse gas emissions and land use decreased for all
countries in the MaxProt (scaled WHO) scenario.
Discussion
Considerations for environmental impacts in national
FBDGs are currently limited to side-notes and sug-
gestions, but do not drive the actual recommended
quantities. For example, multiple FBDGs in Europe
recommend reducing overall consumption of animal
products while increasing plant-based foods in the diet
(eg, the Netherlands, Germany, Iceland, and Denmark).
Four out of the five national FBDGs assessed in this
study recommended amounts of animal products that
exceeded what would be required to meet protein as well
as other nutrient requirements in a well balanced diet.
Our results showed the benefits of reducing animal
products in the FBDGs for greenhouse gas emissions
and land use, and showed that animal products from
LOCB could cover a substantial part of the nutritional
needs of the studied populations, or all protein needs,
depending on dietary shifts. With dietary shifts, the
reduction in consumption of animal products would be
compensated by an increase in consumption of plant-
based food, which also contributes to reach nutritional
requirements. Our quantitative estimates for the
potential of animal products with dierent nutritional
foci can directly contribute to discussions on recom-
mended targets for the inclusion of animal products
produced via circularity principles. Such animal products
could contribute between 22% (Netherlands) and 47%
(Switzerland) of total protein contributions of the original
FBDGs. Using domestic availability as a proxy for the
national consumption of animal products, a reduction in
current animal product consumption of 24–69% in the
case study countries would be required.11 In other words,
at least a third of the average daily protein needs could be
provided in all countries.24,26 Notably, the recommended
0 250 500 750 010203040010203
04
0
0
250
500
750
Milk content of FBDG (g)
Milk content of scenario (g)
0
10
20
30
40
50
40
40 50
Beef content of FBDG (g)
Beef content of scenario (g)
0
10
20
30
40
Pork content of FBDG (g)
Pork content of scenario (g)
0
10
20
30
40
010203040
Chicken content of FBDG (g)
Chicken content of scenario (g)
0
10
20
30
010 20 30
Eggs content of FBDG (g)
Eggs content of scenario (g)
0
20
10
40
30
50
010 20 4030 50
Fish content of FBDG (g)
Fish content of scenario (g)
Bulgaria Malta Netherlands Sweden Switzerland MaxFattyAcids MaxMinerals MaxProt MaxVitamins
Country Scenario
Figure 3: Quantities of the animal products in the original food-based dietary versus quantities of the animal products in the scenarios and countries per
capita per day
Diagonal line indicates equal nutritional contributions in the scenarios and dietary guidelines. For values in the upper triangle, quantities of the scenarios exceeds those
of the dietary guidelines, and in the lower triangle (grey shading), quantities of the scenarios are lower than those of the dietary guidelines. Horizontal differences show
differences between countries, and vertical differences show differences between scenarios. FBDG=food-based dietary guideline. MaxFattyAcids=maximised fatty acid
scenario. MaxMinerals=maximised mineral scenario. MaxProt=maximised animal protein scenario. MaxVitamins=maximised vitamin scenario.

Articles
www.thelancet.com/planetary-health Vol 6 June 2022
e481
protein intake could also be achieved by increasing
the protein contribution from plant-based products.27
Parallel to the indicated reduction of animal product
consumption, moving towards a more circular food
system would require substantial adjustments of animal
production. As well as investing in breeds that are better
suited to value LOCB, the guiding principle to close
nutrient cycles where possible and to minimise external
inputs, such as mineral fertiliser and feed imports, would
result in substantial changes in trade patterns. Only by
consistent transformations of food systems in this regard
can the estimated environmental improvements be
reached.
In line with our findings, the study by Springmann and
colleagues,28 which used similar quantities to represent
diets meeting the FBDG, reported that recommendations
for animal products in most FBDGs are too high from a
human health and environmental impact perspective.
van Zanten and colleagues11 defined a land boundary for
sustainable livestock consumption and concluded that,
on a global average, 9–23 g of protein per capita—which
covers around a third of the daily protein requirement24—
could be derived from animal products solely from
LOCB. However, the recommended amounts of protein
from animal products in the five national FBDGs
assessed in this study ranged from 34 g to 56 g.
Remarkably, this amount could almost cover protein
needs without considering plant-based foods. Cultural
aspects can possibly explain these relatively high shares
of animal products: in many countries, cattle had (and
still have) an important role in converting grass resources
from marginal areas into valuable animal products.
Consequently, consumption of dairy—and the associated
beef products and fats—proved an essential source for
protein and fats for these populations. However,
currently, grass resources are partly grown on land that
could be used for human food consumption, and not all
of this land is temporary grassland with an agronomic
function in crop rotations.13 Thus, part of these grasslands
come with higher opportunity costs for alternative use in
food production. These findings call for a revision of
dietary guidelines assessed in this study and beyond,
which would help in meeting nutrition recommendations
based on LOCB-sourced animal products. Similar
considerations should be made for other animal products
and countries, and redesigning current food systems at a
more regional level should be a priority.27
The four scenarios showed that, to fulfil a diverse set of
nutritional requirements, diversity in animal product
consumption is important. Fish and seafood substantially
contribute to the dietary omega-3 fatty acid intake, and
our findings support this idea. Our analysis showed that,
in some countries maximising fatty acids even led to an
increase in pork production, with the by-products
available as feed for salmon (Bulgaria and Switzerland).19
When focusing on minerals—specifically zinc, calcium,
and iron—egg production was increased, and the
associated meat as well. Further, when vitamins A and
B12 were focussed on, milk and the associated beef
showed a slight increase. In the selected FBDGs, the
nutritional function of animal products beyond protein
supply was rarely mentioned in the recommendations,
and therefore, the reasoning behind the recommended
quantities of animal products was not further clarified.
For chicken, results diverged most between our
scenarios and FBDGs. Although chicken is the meat type
most often recommended in FBDGs,6 it was rarely
selected in the proposed scenarios, independent of the
nutritional focus. Chicken is often promoted as a
relatively sustainable source of meat owing to its
favourable feed conversion ratio, resulting in ecient
production and low environmental impact intensities per
kg of product.12 However, the high eciency of chicken
comes with a downside—namely, the required high
quality of feed that cannot be provided by standard
circular feed production methods.19 Consequently,
currently widespread chicken breeds are not able to feed
on lower quality feedstus (to which part of LOCB
belong), and are therefore not competitive in scenarios
with LOCB. In current production systems, feed for
chicken is often of high quality, and its production
Greenhouse gas emissions Land use
–60 –40 –20
Deviation (%) Deviation (%)
02040 –40 –200 20
Malta
Netherlands
Switzerland
Sweden
Bulgaria
Scenario
MaxProt MaxProt (scaled FBDG) MaxProt (scaled WHO)
Country
Figure 4: Greenhouse gas emissions and land use of the scenarios MaxProt, MaxProt (scaled FBDG),
and MaxProt (scaled WHO)
Deviation is relative to the original FBDG per country. FBDG=food-based dietary guideline. MaxProt=maximised
animal protein scenario. MaxProt (scaled FBDG)=maximised animal protein scenario scaled to the protein content
of the original food-based dietary guideline. MaxProt (scaled WHO)=maximised animal protein scenario scaled to
the protein recommendation of the WHO (60 g/capita/day).

Articles
e482
www.thelancet.com/planetary-health Vol 6 June 2022
For more on the project
associated with the NWO-veni
grant see
www.
circularfoodsystems.org
competes directly or indirectly with human food
production.14 Notably, results could look dierent if, for
example, greenhouse gas emissions would be considered
in the scenario definition process, which might favour
chicken over other animal production systems. Moreover,
chicken features a more favourable nutritional profile
than other types of meat. In several epidemiological
studies, no correlation with increased risk for non-
communicable diseases was found for white meat, but it
was found for red and processed meat types.29,30
Geographical circumstances shape the availability of
LOCB. Here, we used a national geographical scope, and
did not allow trade of LOCB. This approach led to large
imbalances between countries regarding available LOCB,
and thus regarding available animal products. This
imbalance was particularly pronounced for Malta, for
which available LOCB was so low that a large nutritional
gap resulted. Previous assessments of circular food
systems took a global or regional perspective, and assumed
that within these geographical contexts part of the
produced LOCB can be traded freely.19 Thus, countries
with low levels of LOCB could import LOCB from other
countries, based on the assumption of an equal distri-
bution of LOCB across the geographical scope assessed.
Moreover, although LOCB availability is determined by
the geographical level of assessment, it is also determined
by the available share of landings from fisheries.
Sustainable landings from fisheries could be an important
source for animal products in circular food systems.
LOCB and fish landings might not be ideally allocated in
the country they are produced. Therefore, it is important
to investigate suitable and equitable distribution
mechanisms of LOCB, animal products, and fish landings.
Thus, the consequences and eects of distribution have to
be weighted against the benefits that occur when resources
are allocated optimally across larger scales.
There were some limitations to our study. First, we solely
varied animal products in FBDGs, while keeping plant-
based food constant. However, the proposed reductions in
animal products would need to be compensated with
increased or specifically diversified plant-based food,
which would also contribute to meeting nutritional
requirements. For this consideration, a land use model is
needed that includes both plant-based food and animal
products, while capturing resource use eciency as well as
flows between the dierent production systems.
Further, we acknowledge that environmental sustain-
ability encompasses much more than greenhouse gas
emissions and land use. The production of our food
aects the environment in many ways (eg, by altering
the global nutrient cycles, adverse impacts on
biodiversity, and fostering soil erosion). Moreover, the
environment is only one dimension of total sustainability;
social and economic eects of food production and
human health implications of our food consumption are
also important factors. Although we only considered the
two environ mental indicators, our proposed scenarios
would probably aect many other dimensions of
sustainability. For example, nitrogen surplus could be
reduced substantially, resulting from the reduction in
animal farming and omission of imported feed. This
omission would lead to a reduction of the whole nutrient
throughput in the system and, therefore, potentials for
losses are smaller. Further research could, for example,
focus on the implications on nutrient flows and soil
health of following such scenarios, as well as economic
and social consequences.
In conclusion, the proposed approach can transparently
contribute to discussions on recommended targets for the
inclusion of animal products produced via circularity
principles. Although the animal product recommen-
dations in the FBDGs of Bulgaria, Malta, the Netherlands,
Sweden, and Switzerland are neither in their composition
nor in their total nutritional value achievable with animal
products from LOCB, 45–88% of the protein from animal
product recommendations could be met with the proposed
circularity principles. This result comes with major
implications for the five national FBDGs assessed, and
applies to most others in high-income countries.28 To make
the dietary guidelines of these countries compatible with
principles from circular food systems as well as protein
requirements, animal product recommendations would
need to be substantially reduced. Such a reduction would
lead to a substantial reduction in greenhouse gas emissions
and land use. Clarity regarding the nutritional function of
the recommended animal products in the diet could help
to target the animal products composition, and to decide
how to allocate LOCB resources in the optimum way.
Contributors
AF, RPMC, IJMdB, AM, CS, OvH, and HHEvZ designed the research.
AF, RPMC, and BvS conducted the research. AF and RPMC analysed
and verified the data. AF, RPMC, IJMdB, AM, CS, BvS, OvH, GP, SR,
MH, and HHEvZ contributed to writing the paper. All authors had final
responsibility for the decision to submit for publication.
Declaration of interests
We declare no competing interests.
Data sharing
Data described in the manuscript, code book, and analytic code will be
made available upon request. Requests can be directed to Anita Frehner
(anita.frehner@fibl.org).
Acknowledgments
We thank A Hayer (Swiss Society for Nutrition SSN) for inputs
regarding interpretation of the Swiss dietary guidelines. The
contribution of AF, AM, and CS to this work was supported by the Swiss
National Science Foundation (project 4069- 166765). HHEvZ received
funding from a personal grant (NWO-Veni).
References
1 Poore J, Nemecek T. Reducing food’s environmental impacts
through producers and consumers. Science 2018; 360: 987–92.
2 GBD 2017 Diet Collaborators. Health eects of dietary risks in
195 countries, 1990–2017: a systematic analysis for the Global
Burden of Disease Study 2017. Lancet 2019; 393: 1958–72.
3 Stylianou KS, Fulgoni VL 3rd, Jolliet O. Small targeted dietary
changes can yield substantial gains for human health and the
environment. Nat Food 2021; 2: 616–27.
4 Clark MA, Springmann M, Hill J, Tilman D. Multiple health and
environmental impacts of foods. Proc Natl Acad Sci USA 2019;
116: 23357–62.

Articles
www.thelancet.com/planetary-health Vol 6 June 2022
e483
5 Willett W, Rockström J, Loken B, et al. Food in the Anthropocene:
the EAT–Lancet Commission on healthy diets from sustainable food
systems. Lancet 2019; 393: 447–92.
6 Herforth A, Arimond M, Álvarez-Sánchez C, Coates J,
Christianson K, Muehlho E. A global review of food-based dietary
guidelines. Adv Nutr 2019; 10: 590–605.
7 Fischer CG, Garnett T. Plates, pyramids, and planets: developments
in national healthy and sustainable dietary guidelines: a state of play
assessment. Rome: Food and Agriculture Organization of the
United Nations, 2016.
8 Brown KA, Timotijevic L, Barnett J, Shepherd R, Lähteenmäki L,
Raats MM. A review of consumer awareness, understanding and
use of food-based dietary guidelines. Br J Nutr 2011; 106: 15–26.
9 Billen G, Aguilera E, Einarsson R, et al. Reshaping the European
agro-food system and closing its nitrogen cycle: the potential of
combining dietary change, agroecology, and circularity. One Earth
2021; 4: 839–50.
10 Theurl MC, Lauk C, Kalt G, et al. Food systems in a zero-deforestation
world: dietary change is more important than intensification for
climate targets in 2050. Sci Total Environ 2020; 735: 139353.
11 Van Zanten HHE, Herrero M, Van Hal O, et al. Defining a land
boundary for sustainable livestock consumption. Glob Change Biol
2018; 24: 4185–94.
12 Frehner A, Muller A, Schader C, De Boer I, Van Zanten HH.
Methodological choices drive dierences in environmentally-
friendly dietary solutions. Glob Food Secur 2020; 24: 100333.
13 Schader C, Muller A, Scialabba N-H, et al. Impacts of feeding less
food-competing feedstus to livestock on global food system
sustainability. J R Soc Interface 2015; 12: 20150891.
14 Mottet A, de Haan C, Falcucci A, Tempio G, Opio C, Gerber P.
Livestock: on our plates or eating at our table? A new analysis of the
feed/food debate. Glob Food Secur 2017; 14: 1–8.
15 Muscat A, de Olde E, de Boer IJ, Ripoll-Bosch R. The battle for
biomass: a systematic review of food-feed-fuel competition.
Glob Food Secur 2020; 25: 100330.
16 de Boer IJ, van Ittersum MK. Circularity in agricultural production:
Wageningen University & Research, 2018.
17 Röös E, Patel M, Spångberg J, Carlsson G, Rydhmer L. Limiting
livestock production to pasture and by-products in a search for
sustainable diets. Food Policy 2016; 58: 1–13.
18 Van Kernebeek HR, Oosting SJ, Van Ittersum MK, Bikker P,
De Boer IJ. Saving land to feed a growing population: consequences
for consumption of crop and livestock products. Int J Life Cycle Assess
2016; 21: 677–87.
19 Van Hal O, De Boer I, Muller A, et al. Upcycling food leftovers and
grass resources through livestock: impact of livestock system and
productivity. J Clean Prod 2019; 219: 485–96.
20 Van Selm B, Frehner A, De Boer IJ, et al. Circularity in animal
production requires a change in the EAT–Lancet diet in Europe.
Nat Food 2022; 3: 1–8.
21 FAO, IFAD, UNICEF, WFP, WHO. The state of food security
and nutrition in the world. Building resilience for peace and food
security. Rome: Food and Agriculture Organization of the United
Nations, 2017.
22 FAO. FAOSTAT: food balance sheets. Rome: Food and Agriculture
Organization of the United Nations, 2001.
23 Adesogan AT, Havelaar AH, McKune SL, Eilittä M, Dahl GE.
Animal source foods: sustainability problem or malnutrition and
sustainability solution? Perspective matters. Glob Food Secur 2020;
25: 100325.
24 WHO. Joint WHO/FAO/UNU Expert Consultation. Protein and
amino acid requirements in human nutrition. Geneva:
World Health Organization, 2007.
25 Muller A, Frehner A, Pfeier C, Moakes S, Schader C. SOLm model
documentation. 2020. https://orgprints.org/id/eprint/38778/1/
mueller-etal-2020-SOLmV6_Documentation.pdf (accessed
Feb 2, 2022).
26 EFSA. Dietary Reference Values for nutrients: summary report.
Parma: European Food Safety Authority, 2017.
27 Springmann M, Wiebe K, Mason-D’Croz D, Sulser TB, Rayner M,
Scarborough P. Health and nutritional aspects of sustainable diet
strategies and their association with environmental impacts:
a global modelling analysis with country-level detail.
Lancet Planet Health 2018; 2: e451–61.
28 Springmann M, Spajic L, Clark MA, et al. The healthiness and
sustainability of national and global food based dietary guidelines:
modelling study. BMJ 2020; 370: m2322.
29 Etemadi A, Sinha R, Ward MH, et al. Mortality from dierent
causes associated with meat, heme iron, nitrates, and nitrites in the
NIH-AARP Diet and Health Study: population based cohort study.
BMJ 2017; 357: j1957.
30 Cocate PG, Natali AJ, de Oliveira A, et al. Red but not white meat
consumption is associated with metabolic syndrome, insulin
resistance and lipid peroxidation in Brazilian middle-aged men.
Eur J Prev Cardiol 2015; 22: 223–30.