The Mediterranean diet in a world context

Abstract

To put the debate on the Mediterranean diet in context by highlighting historical and prospective changes in the level and composition of food consumption in the world and key Mediterranean countries.
Data from FAO's food balance sheets are used to illustrate historical evolution. Projections to 2030 are presented from FAO's recent and ongoing work on exploring world food and agriculture futures.
International.
Many developing countries are undergoing diet transitions bringing them closer to the diets prevalent in the richer countries, i.e. with more energy-dense foods. There follows an increase in the incidence of diet-related non-communicable diseases, which are superimposed on the health problems related to undernutrition that still afflict them. In parallel, many low-income countries are making little progress towards raising food consumption levels necessary for good nutrition and food security. Wider adoption of food consumption patterns akin to those of the Mediterranean diet hold promise of contributing to mitigate adverse effects of such diet transitions. However, the evolution of food consumption in the Mediterranean countries themselves is not encouraging, as these countries have also followed the trend towards higher shares of energy-dense foods. Possible policy responses to these problems include measures to raise awareness of the benefits of healthier diets and/or to change relative food prices in favour of such diets (by taxing fattening foods) or, at the extreme, making individuals who follow 'bad' diets, and thus are prone to associated diseases, bear a higher part of the consequent costs borne by the public health systems (tax fat people).

Full text

The Mediterranean diet in a world context
Nikos Alexandratos*†
Consultant to, and former Chief of, the Global Perspective Studies Unit, Food and Agriculture Organization of the
United Nations (FAO), 00100 Rome, Italy
Abstract
Objective: To put the debate on the Mediterranean diet in context by highlighting
historical and prospective changes in the level and composition of food consumption
in the world and key Mediterranean countries.
Design: Data from FAO’s food balance sheets are used to illustrate historical evolution.
Projections to 2030 are presented from FAO’s recent and ongoing work on exploring
world food and agriculture futures.
Setting: International.
Results and conclusions: Many developing countries are undergoing diet transitions
bringing them closer to the diets prevalent in the richer countries, i.e. with more
energy-dense foods. There follows an increase in the incidence of diet-related non-
communicable diseases, which are superimposed on the health problems related to
undernutrition that still afflict them. In parallel, many low-income countries are
making little progress towards raising food consumption levels necessary for good
nutrition and food security. Wider adoption of food consumption patterns akin to
those of the Mediterranean diet hold promise of contributing to mitigate adverse
effects of such diet transitions. However, the evolution of food consumption in the
Mediterranean countries themselves is not encouraging, as these countries have also
followed the trend towards higher shares of energy-dense foods. Possible policy
responses to these problems include measures to raise awareness of the benefits of
healthier diets and/or to change relative food prices in favour of such diets (by taxing
fattening foods) or, at the extreme, making individuals who follow ‘bad’ diets, and
thus are prone to associated diseases, bear a higher part of the consequent costs
borne by the public health systems (tax fat people).
Keywords
Diet transitions
Obesity
Mediterranean diet
Developing countries
Policies
The debate on global issues of agriculture,food andnutrition
has traditionally reflected perceptions of food scarcities–
actual, imminent or potential–under a perceived Malthusian
racebetween foodand population
1
. However, the long-term
historical trends have been pointing in the opposite
direction: at the global level, apparent food consumption
(kcal/person/day as a world average) has been getting more
plentiful and varied, not scarcer, notwithstanding the hefty
increases in world population
2,3,8
(the term ‘apparent’ food
consumption, often also called ‘availability’, refers to the
domestic disappearance for human consumption of food
commodities at the retail level computed in FAO’s food
balance sheets (FBS) for each country‡).
The trend towards increasing world averages notwith-
standing, significant parts of the world’s population (some
800 million people, mostly in Africa and Asia
5
) continue to
have food consumption levels well below nutritional
requirements. This reflects the persistence of poverty, lack
of development, wars, etc. rather than global constraints to
producing enough to meet the nutritional needs of
everyone in the world. However, local constraints to
increasing food production due to natural resource
paucity and associated failures to develop agriculture are
often important factors explaining the persistence of
hunger in several countries which combine high
demographic growth, significant dependence on local
agriculture and few alternative avenues to sustained
development, e.g. Niger. Localized Malthusian situations
are not to be excluded even in a world with plentiful food
supplies and more than sufficient production potential
globally
6
.
q The Author 2006*Corresponding author: Email nikos.alexandratos@fao.org
†The views expressed are the author’s, not necessarily those of FAO.
‡ Apparent food consumption ¼ production þ imports þ beginning
stocks 2 exports 2 non-food industrial uses 2 feed 2 seed 2 waste
(post harvest to retail) 2 ending stocks. Post-retail and household
waste or feeding to pets food designated for human consumption are,
therefore, included in apparent food consumption and can be
considerable in some countries
4
. All data on apparent food
consumption used in this paper are from FAO’s data bank FAOSTAT
(http://faostat.fao.org/faostat/default.jsp?version¼ int&hasbulk ¼ 1)
except where indicated otherwise. The notation 1999/2001 indicates
the three-year average 1999–2001.
Public Health Nutrition: 9(1A), 111–117
DOI: 10.1079/PHN2005932

The rise in the world average food consumption in
recent decades reflected primarily the gains made by many
developing countries, with China dominating the scene.
Many of these countries have been gradually attaining
consumption levels and patterns approaching those
prevalent in the industrialised world, in particular as
regards changes in diets towards energy-dense ones high
in fat, particularly saturated fat, sugar and salt and low in
unrefined carbohydrates. In combination with lifestyle
changes, largely associated with rapid urbanisation, a
corresponding increase in diet-related chronic non-
communicable diseases (NCDs) has been observed. In
many countries undergoing this transition, the obesity-
related NCDs appear when health problems related to
undernutrition of significant parts of their populations are
still widely prevalent
7
. The two problems coexist and
present these countries with novel challenges and strains
in their health systems.
The increases in per capita food consumption and the
change of diets towards more livestock products are set to
continue. While it is beneficent in many countries with still
inadequate diets, these further changes will likely be
accompanied in many cases by enhanced risks of
increased incidence of diet-related NCDs. In the next
section, we present projections of the possible evolution
of world food consumption (levels and structure) to the
year 2030 from recent
8
and ongoing work in FAO.
Evolution of food consumption in a diverse world:
past, present, future
Table 1 presents the historical data and projections to 2030.
The massive improvement that occurred in the average
food situation (kcal/person/day) in the developing
countries in the past three decades is evident in these
data. It can be further appreciated by noting that in the early
1970s three-quarters of their population of 2.6 billion lived
in countries with under 2200 kcal/person/day. Both India
and China, with a combined population of 1.37 billion in
1970, belonged to this class. The situation had changed
dramatically by the end of the century: only 12% of their
much larger population (4.7 billion) lived in such countries,
while 50% of the population lived in countries with over
2700 kcal, up from only 4% 30 years earlier. This general
progress reflected principally the growth of apparent food
consumption in some of the most populous developing
countries, foremost among them China, but also Indonesia,
Brazil, Mexico and, to a smaller extent, also Nigeria, India,
Egypt, Iran and others.
Structural change in favour of energy-dense foods
(livestock products, vegetable oils, sugars) is evident in
the historical evolution. Only half a dozen developing
countries (in South America but also pastoral Mongolia) had
over 50 kg of meat per capita three decades ago, while the
great bulk of the population (75%, including China and
India) lived in countries with under 10 kg. The proportions
had changed dramatically by the end of the century,
reflecting above all China’s rapid growth, with the country
approaching 50 kg. Change would have been even more
pronounced were it not for India, which has so far proved
very resilient in resisting penetration of meat-eating habits–
a phenomenon reflecting both persistent massive poverty
and cultural/religious factors
9
. The gradual shift of more and
more countries towards diets having structures exceeding
recommended limits of ‘bad’ nutrients, highly correlated
with the increase of the relative shares of livestock products,
is also seen in the data provided in Table 2 (reproduced
from Schmidhuber and Shetty, 2005
10
).
However, very pronounced differences in diet struc-
tures continue to exist among the individual countries. For
example, food consumption of cereals is 220–250 kg/per-
son/year in some countries (e.g. Egypt and Morocco,
mostly wheat; or Burkina Faso, mostly millet and
sorghum) and as low as 40 kg/person/year in others, e.g.
Democratic Republic of Congo, Rwanda and Burundi,
where roots, tubers and plantains predominate as sources
of food energy.
According to the FAO projections (Table 1), the trend
towards higher levels of per capita apparent consumption is
setto continue, albeitata slowerpace than in the past,andso
will the structural changes in favour of the energy-dense
foods. However, the prospects for growth and further
structural change differgreatly among countriesand regions.
Income growth plays a role in all cases, but its effects are
mediated through complex interactions of a multitude of
other factors (ecological, social, religious, demographic,
urbanisation, health, policy, etc.) and, of course, the stage in
the nutrition transition countries find themselves in at
present determines how far they have still to go. Some
countries have made the transition to more or less fairly high
and stable consumption levels and livestock-rich diets (e.g.
several industrialised countries) while others are at various
stages in the transition. Still others have barely started the
transition out of the traditionallow levels and littlediversified
patterns, e.g. the roots/plantains-based diets in several low-
income countries in sub-Saharan Africa.
Given the prospects for overall development (not very
promising for a number of developing countries*) and the
persistence of the other factors determining diet structures,
we can anticipate that countries will continue to have
widely differing diet profiles in the future, though
differences will be somewhat less pronounced than at
present. In some countries, progress will not be sufficient to
bring them even in 30 years time to levels of apparent per
capita consumption compatible with reduction of under-
nutrition to tolerable levels
8
. The nutritional divide will be
* The latest World Bank assessment of prospects (p. 9) foresees that
there will still be some 2 billion people in 2015 (compared with
some 2.5 billion in 2002) living on less than $2 a day in the
developing countries, the bulk of them in South Asia and sub-
Saharan Africa
11
.
N Alexandratos112

becoming increasingly evident within the group of the
developing countries themselves rather than along
the traditional dichotomy developed vs. developing. In
the process, countries moving out of ‘food poverty’ proper
will run the risk of falling into the trap of ‘health poverty’ as
they transit towards food consumption levels and patterns
associated with increases in the incidence of chronic NCDs,
even before they have resolved problems of undernutrition
of significant segments of their populations.
Promotion of the Mediterranean diet (MD) holds promise
of mitigating the undesirable effects that accompany the diet
transitions. We can gain some insights concerning the extent
to which this may be a realistic prospect by examining how
the MD has been faring in the Mediterranean region itself
under the socio-economic transformations and related diet
transitions of the last few decades.
Evolution of diets in some Mediterranean countries
To judge from the historical evolution of food
consumption levels and structures, as depicted in the
data of the national FBS, the experience does not augur
well. The traditional MD seems to be in a moribund state in
its very birthplace–allegedly Greece and within it Crete*.
With the exception of the high consumption of olive oil
and fresh fruits and vegetables, today’s food consumption
patterns in Greece (as national average) have moved away
from those that were close to the MD prototype and were
Table 1 Changes in the levels and commodity composition of apparent food consumption: world and major country groups
kg/person/year 1969/1971 1979/1981 1989/1991 1999/2001 2030
World
Cereals, food 149 160 171 165 165
Cereals, all uses 303 325 329 309 331
Roots and tubers (incl. plantains) 84 73 65 69 75
Sugar and sugar crops (raw sugar equivalent) 22 23 23 24 26
Pulses, dry 8 7 6 6 6
Vegetable oils, oilseeds and products (oil equivalent) 7 8 10 12 16
Meat (carcass weight) 26 30 33 37 47
Milk and dairy, excl. butter (fresh milk equivalent) 75 77 77 78 92
Other food (kcal/person/day) 216 224 241 289 320
Total food (kcal/person/day) 2411 2549 2704 2789 3030
Developing countries
Cereals, food 146 162 174 166 166
Cereals, all uses 192 219 239 238 268
Roots and tubers (incl. plantains) 79 70 60 67 75
Sugar and sugar crops (raw sugar equivalent) 15 18 19 21 25
Pulses, dry 9 8 7 7 7
Vegetable oils, oilseeds and products (oil equivalent) 5 7 9 10 14
Meat (carcass weight) 11 14 18 27 38
Milk and dairy, excl. butter (fresh milk equivalent) 29 34 38 45 67
Other food (kcal/person/day) 123 140 171 242 280
Total food (kcal/person/day) 2111 2308 2520 2654 2950
Industrial countries (Western Europe, North America, Japan, Australia, New Zealand)
Cereals, food 132 139 154 162 159
Cereals, all uses 531 542 544 592 641
Roots and tubers (incl. plantains) 74 67 69 67 61
Sugar and sugar crops (raw sugar equivalent) 41 37 33 33 33
Pulses, dry 3 3 3 4 4
Vegetable oils, oilseeds and products (oil equivalent) 13 16 19 22 24
Meat (carcass weight) 70 79 84 90 99
Milk and dairy, excl. butter (fresh milk equivalent) 189 201 211 214 223
Other food (kcal/person/day) 486 500 521 525 560
Total food (kcal/person/day) 3046 3133 3292 3446 3520
Transition countries (formerly Centrally Planned Economies, i.e. Eastern Europe and former USSR)
Cereals, food 201 189 179 169 164
Cereals, all uses 653 778 768 499 618
Roots and tubers (incl. plantains) 140 118 97 103 99
Sugar and sugar crops (raw sugar equivalent) 42 46 43 37 39
Pulses, dry 4 3 2 2 2
Vegetable oils, oilseeds and products (oil equivalent) 7 9 10 10 15
Meat (carcass weight) 50 63 71 44 59
Milk and dairy, excl. butter (fresh milk equivalent) 186 181 177 160 179
Other food (kcal/person/day) 331 372 333 317 365
Total food (kcal/person/day) 3323 3389 3280 2900 3145
Note: Cereals food consumption includes the grain equivalent of beer consumption and of corn sweeteners.
Projections from Bruinsma, 2003
8
, Table 2.7, with revisions from ongoing work by the author.
* The term ‘traditional Mediterranean diet’ has a specific meaning. It
reflects food patterns typical of some Mediterranean regions in the
early 1960s, such as Crete, parts of the rest of Greece, and southern
Italy
12
.
The Mediterranean diet in a world context 113

prevalent as recently as the mid 1960s. Shifts that
accompanied rapid urbanisation, growing incomes,
technical change in the food industry, the rapid diffusion
of fast-food outlets, and globalisation have led to greatly
increased consumption of animal products, lipids other
than olive oil (added lipids as well as those embodied in
other foods) and sugar.
In the mid 1960s, Greece had a national average apparent
food consumption yielding 2900 kcal/person/day, with
29% coming from fats (13% from olive oil) and 43% from
cereals. Meat consumption was a mere 33 kg/person/year
and that of sugar 18 kg (raw sugar equivalent). Three and a
half decades later, the kcal/person/day had risen to 3700
and fats from 92 to 152 g/day, accounting for 36% of total
calories (12% from olive oil). Meat consumption had risen
to 88 kg/person/year and that of sugar to 32 kg. No wonder
that the incidence of obesity in Greece is among the highest
in Europe and the world, with Crete itself holding place of
pride in this transformation
13,14
. The evolution of diets in
the other Mediterranean countries on the European side
tells a similar story. Figure 1 shows these changes for
Greece, Italy and Spain.
In the obesity data (percentage of adult population with
body mass index . 30 kg/m
2
) of both the International
Obesity Task Force
13
and the OECD
15
, Italy is at the bottom
in the European obesity league, while Spain is in the middle
position and Greece has the highest estimates. How can
countries with fairly similar (though far from identical) diet
levels/structures have so widely differing obesity rates?
These divergences certainly raise the issue of reliability,
comparability and mutual compatibility of the data on both
apparent food consumption levels and obesity. On the side
of data on obesity, it is to be noted that Italy’s come from
self-reporting, a method which is generally considered to
underestimate the true incidence of the phenomenon
15
.On
the side of the data on apparent food consumption, those of
the FBS are certainly overestimates of actual intakes
because they are inclusive of post-retail waste, which
can be considerable
4
. They also contain the errors made
in recording production, imports, exports, estimates of
Table 2 Apparent consumption of fats and cholesterol in excess of recommended levels
Limits 1961/1963 1969/1971 1979/1981 1989/1991 1999/2001
No. of countries 158 158 158 158 178
Total fat . 30% of total kcal 28 31 43 54 61
Saturated fatty acids . 10% 47 49 50 55 62
Cholesterol . 300 mg/day 26 30 41 44 64
Source: Reproduced from Schmidhuber and Shetty, 2005
10
based on the FAO FBS. The larger number of
countries in 1999/2001 than in earlier years reflects the new countries formed after the transformations of the
former USSR and Eastern Europe.
Recommended levels are from Data Food Networking
16
.
0
20
40
60
80
100
120
140
160
180
1964/1966 1964/1966 1964/19661999/2001 1999/2001 1999/2001
Greece Spain Italy
Fats (grams/person/day)
0
10
20
30
40
50
60
70
80
90
Calories from fats OR cereals % of total calories
Fats from olive oil Fats from other vegetable oils/products.
Fats from animal products Calories from fats % of total calories (right scale)
Calories from cereals % of total calories (right scale)
Note. Fats include added fats as well as those embodied in other food products
Fig. 1 Greece, Spain, Italy: diet transition to increased fat consumption
N Alexandratos114

non-food uses and pre-retail losses of food commodities,
though such errors can result in either over- or under-
estimation–see discussion in Bruinsma, 2003
8
(pp. 380–1).
We have little option but to continue using these FBS
data, notwithstanding their considerable shortcomings,
because no other source of food consumption data
provides as comprehensive coverage (countries, time
periods, commodities) of what people have available to
eat. In addition, the need to interface food consumption
with agricultural production and trade in any policy
analysis requires an internally consistent set of data for all
variables concerned. Indirectly, some degree of validation
of the FBS data comes from the observed accordance of
the trends in per capita food consumption resulting from
these data with the trends in obesity measurements for
countries which have data on the latter for more than one
time period. Some examples are shown in Fig.2.
It is seen that obesity is shown as having increased in
Spain from 7% in 1987 to 13% in 2001, a trend that parallels
that of the apparent national average food consumption
revealed by the FBS. In contrast, data from Spain’s
Household Budget Surveys of 1980 –1981 and 1998– 1999
indicate that there has been a fairly generalised decline in
per capita food consumption (or availability)
16
, a trend
clearly at variance with the data showing growing obesity
and growing national food supply from production and
imports (net of exports). It would seem that these data
consistently underestimate consumption, mainly because
they more often than not capture only the part of food
purchased for home consumption. Yet, the share of total
food consumed away from home (restaurants, workplace
cantinas, fast-food outlets, etc.) has been growing by leaps
and bounds
17
.Byreductio ad absurdum we may note that
if per capita consumption of meat had not risen, it would be
difficult to explain how Spain’s production of meat (mostly
pigmeat) grew so fast (by some 80%) over that period, given
that increased net exports of pigmeat and products took
only a small part of the total increase in production.
Issues in diet transitions and thinking about policy
responses
A growing number of developing countries are embarking
on nutrition transitions. While such progress will improve
the welfare of significant parts of the world’s population
currently in poverty and suffering from inadequate access
to food, experience shows that the associated risks of diet-
related NCDs will also be rising. WHO’s proposed Global
Strategy on Diet, Physical Activity and Health
18
, if adopted
and acted upon, holds promise of maximising benefits of
such transition and minimising adverse effects. The
recommendations concerning increased intakes of fruits
and vegetables and substituting monounsaturated fatty
acids (plentiful in olive oil, but also in canola oil) for other
fats would favour the further adoption of the MD.
Countries exporting Mediterranean products would gain
some competitive advantage in the world market. Perhaps
it is no coincidence that in recent years a good part of the
expansion of world consumption of olive oil was
accounted for by increases in countries with no or little
tradition in olive oil production and consumption, e.g.
North America, non-Mediterranean Europe, Japan, Aus-
tralia and Brazil
19
.
The growing awareness of the high costs imposed on
society by the spread of diet-related diseases, both in
terms of the health welfare of individuals and the
pressures on the national health systems, provide a strong
case for policies to promote healthy diets. It also provides
2800
2900
3000
3100
3200
3300
3400
3500
3600
3700
3800
1976–1980 1988–1994 1999–2000
1980
1991
2001
1987
2001
USA USA USA UK UK UK Spain Spain
kcal/person/day
0
5
10
15
20
25
30
35
Obesity (%)
kcal/person/day (FBS) Obesity, percentage of adult population
Obesity data from
15
, Chart 4.11
Fig. 2 Changes over time: kcal/person/day and obesity rates
The Mediterranean diet in a world context 115

an additional argument in favour of action now to reduce
the incidence of hunger and undernutrition in the many
low-income countries facing that problem, given that,
according to some, children born to undernourished
mothers and/or in populations with long histories of food
deprivation face higher than average risks of developing
diet-related diseases in adulthood in environments of
easier access to food supplies
10,20
. There is less consensus
on what are appropriate policy responses–see discussion
in Schmidhuber, 2005
21
. At the one extreme are views
emphasising nutrition education and provision of infor-
mation on the pros and cons of alternative lifestyles and
the attributes of the different food products (e.g. by means
of appropriate labelling), but not otherwise interfering
with eating choices of individuals. In short, according to
this view, it is the responsibility of the informed individual
to adopt lifestyles that would reduce the risk of diet-related
diseases.
At the other extreme, there are those who consider that
it is the responsibility of the state to adopt more
interventionist policies that would influence choices in
favour of healthier food consumption patterns and
lifestyles, e.g. policies akin to those used to discourage
smoking. Examples include the banning of advertising of
unhealthy foods, particularly to children, or outright
restricting access to them (e.g. in schools), or making them
more expensive by taxing them (e.g. a tax on fattening
foods). In the extreme case, the option is sometimes aired
of taxing fat people, e.g. through higher contributions to
the health insurance schemes. This is predicated on the
notion of the existence of externalities, i.e. that people
with ‘inappropriate’ eating behaviour leading to obesity
and associated disease should be made to bear a higher
proportion of the costs themselves rather than impose
them on others in collectively funded health systems–
something akin to the ‘polluter pays principle’.
The policy dilemmas are certainly of no easy resolution,
particularly in the light of the evidence that, at least in
industrial countries, spreading obesity is increasingly
encountered among the poor rather than in the well-to-do
classes. In such cases, making fat people (hence
predominantly poor) pay more would tend to subvert
the very notion of social justice underlying most national
health systems. This policy option may, however, have
some merit for the developing countries undergoing diet
transitions, where overweight and obesity are encoun-
tered, at least in the early stages, in the better– off
population groups rather than among the poor.
Taxing food items rich in fat may or may not be effective
in shifting diets of the population groups tending to
overconsume such foods. If they are rich, higher (after tax)
prices of these foods will probably have little effect as the
rich have low price elasticities of their demand for food. In
addition, it may penalise those who need high-energy
intakes because of the work they do and who are more
likely than not to be poor, e.g. rural workers. If the
overweight people are poor, most likely the case in the
high-income countries, the effectiveness of such policies
would depend, inter alia, on whether they will be able to
afford sufficient quantities of at least some of the more
expensive healthier foods (fruits, vegetables, fish) to
substitute for part of the energy-dense food items that are
often cheaper
22
. It may happen that, following the
imposition of the tax on ‘bad’ foods, the prices of
the more preferred foods would rise because of the shift of
the demand in their favour. In such situations, the poor
could easily be priced out of the market of such healthier
foods and revert to the consumption of the ‘bad’ ones.
The limiting case is that they may end up changing their
diets further in favour of the ‘bad’ foods. This could
happen if the general rise in the food price level following
a tax on fat foods were to reduce their overall purchasing
power, i.e. make them poorer and reinforce their spending
patterns in favour of cheap calories.
There is certainly something to be said for policies
that, rather than raise the price of ‘bad’ foods, actually
reduce that of the healthier ones, e.g. through measures
to reduce production and, particularly, distribution and
marketing costs, or that reduce barriers to imported
food. It is likely that aficionados of the MD would abhor
any thought of having fresh fruits and vegetables be
imported, sometimes from far-away places, just for the
sake of having them cheaper. After all, the very notion
of promoting the MD is predicated not only on its health
merits but also on the benefits it bestows to society at
large through the preservation of the culture (with local
production being an integral part of it) that gave rise to
the MD in the first place. However, we should be aware
that such arguments can be, and often are, abused to the
point of becoming thinly veiled positions for trade
protectionism.
References
1 Brown L. Tough Choices: Facing the Challenge of Food
Scarcity. New York: WW Norton, 1996.
2 Alexandratos N, ed. World Agriculture: Towards 2010, an
FAO Study. Chichester/Rome: John Wiley and Sons/FAO,
1995.
3 Alexandratos N. World food and agriculture: outlook for the
medium and longer term. Proceedings of the National
Academy of Sciences of the USA 1999; 96: 5908–14.
4 Kantor L. A Dietary Assessment of the US Food Supply:
Comparing Per Capita Food Consumption with Food Guide
Pyramid Serving Recommendations. Agricultural Economics
Report No. 772. Washington, DC: US Department of
Agriculture, 1998.
5 Food and Agriculture Organization of the United Nations.
The State of Food Insecurity in the World. Rome: FAO, 2004.
6 Alexandratos N. Countries with rapid population growth and
resource constraints: issues of food, agriculture, and
development. Population and Development Review 2005;
31(2): 237–58.
7 World Health Organization. Diet, Nutrition and the
Prevention of Chronic Disease. Report of a Joint WHO/FAO
N Alexandratos116

Expert Consultation. WHO Technical Report Series 916.
Geneva: WHO, 2003.
8 Bruinsma J, ed. World Agriculture: Towards 2015/30, an
FAO Perspective. London/Rome: Earthscan/FAO, 2003.
9 Dhillon A. India has no beef with fast-food chains: the
country has adapted American fare to satisfy its own taste–
spicy and vegetarian. Financial Times, 23 March 2002.
10 Schmidhuber J, Shetty P. The nutrition transition to 2030,
why developing countries are likely to bear the major
burden. Plenary paper presented at the 97th Seminar of the
European Association of Agricultural Economists, University
of Reading, England, 21–22 April 2005 (http://www.fao.org/
es/esd/gstudies.htm).
11 World Bank. Global Economic Prospects 2006. Washington
DC: World Bank, 2005.
12 International Consensus Statement Dietary Fat, the
Mediterranean Diet, and Lifelong Good Health 2000,
International Conference on the Mediterranean Diet,
Royal College of Physicians, London, 13–14 January 2000
(http://europa.eu.int/comm/agriculture/prom/olive/
medinfo/uk_ie/consensus/index.htm).
13 International Obesity Task Force. EU Platform on Diet,
Physical Activity and Health–Briefing Paper. Prepared in
collaboration with the European Association for the Study of
Obesity, Brussels 15 March 2005 (http://www.iotf.org/
media/euobesity3.pdf).
14 Moschandreas J, Kafatos A. Food and nutrient intakes of
Greek (Cretan) adults: recent data for food-based dietary
guidelines in Greece. British Journal of Nutrition 1999;
81(Suppl. 2): S71–6.
15 Organization for Economic Co-operation and Development.
Health at a Glance, OECD Indicators 2003. Paris: OECD,
2003.
16 Data Food Networking–DAFNE (www.nut.uoa.gr).
17 Cutler D, Glaeser E, Shapiro J. Why have Americans Become
More Obese? Harvard Institute of Economic Research,
Discussion Paper No. 1994, 2003 (http://post.economics.
harvard.edu/hier/2003papers/2003list.html).
18 World Health Organization. Process for a WHO Global
Strategy on Diet, Physical Activity and Health. Geneva:
WHO, 2003.
19 Alexandratos N. Mediterranean countries and world
markets: basic foods and Mediterranean products. In:
Brauch HG, Liotta HP, Marquina A, Rogers P, Selim EM,
eds. Security and the Environment in the Mediterranean –
Conceptualising Security and Environmental Conflicts.
Heidelberg: Springer, 2003; 813–26.
20 Food and Agriculture Organization of the United Nations.
Fighting Hunger Today Could Help Prevent Obesity
Tomorrow. Rome: FAO, 2004 (http://www.fao.org/
newsroom/en/news/2004/36 847/index.html).
21 Schmidhuber J. The growing global obesity problem: some
policy options to address it. Paper presented at the 97th
Seminar of the European Association of Agricultural
Economists, University of Reading, England, 21– 22 April
2005 (http://www.fao.org/es/esd/gstudies.html).
22 Darmon N, Briend A, Drenowski A. Energy-dense diets are
associated with lower diet costs: a community study of
French adults. Public Health Nutrition 2004; 7(1): 21–7.
The Mediterranean diet in a world context 117