Content uploaded by Bénédicte MJ Merle
Author content
All content in this area was uploaded by Bénédicte MJ Merle
Content may be subject to copyright.
ORIGINAL PAPER
Adherence to a Mediterranean diet and energy,
macro-, and micronutrient intakes in older persons
Catherine Feart &Benjamin Alles &
Bénédicte Merle &Cécilia Samieri &
Pascale Barberger-Gateau
Received: 10 February 2012 /Accepted: 6 June 2012
#University of Navarra 2012
Abstract The Mediterranean-type diet combines sev-
eral foods and nutrients already individually proposed as
potential protective factors against adverse health out-
comes, such as cardiovascular diseases. The aim of the
present study was to describe the association between
adherence to a Mediterranean diet (MeDi) and intake of
energy, macronutrients, and micronutrients. The study
sample consisted of 1,595 individuals from Bordeaux,
France, included in 2001–2002 in the Three-City Study.
Adherence to a MeDi (scored as 0 to 9) was computed
from a food frequency questionnaire (FFQ). Total ener-
gy intake (EI) and nutrient intake were evaluated on a
24 h recall. Statistical analyses were stratified by gender.
Both in men and women, greater MeDi adherence was
associated with higher total vegetal protein, polyunsat-
urated fatty acids (PUFA), notably n−6 PUFA, and
lower total saturated fat intakes, as expressed in percent-
age of EI. Higher total monounsaturated fat and oleic
acid intakes (% EI) were observed with greater MeDi
adherence in men. Women with the highest MeDi ad-
herence exhibited a higher mean carbohydrate, polysac-
charide, and total n−3 PUFA intakes (% EI). The
consumption of fibers; vitamins B6, C, and E; folate;
magnesium; potassium; and iron increased with greater
MeDi adherence, both in men and women. However,
consumption of calcium significantly decreased with
greater MeDi adherence in women, while the ratio of
n−6/n−3 PUFA precursors increased. This cross-
sectional study provides the nutrient-related basis of
the Mediterranean-type diet of French elderly commu-
nity dwellers, which might participate to its well-
documented beneficial effects on health.
Keywords Mediterranean diet .Energy intake .
Nutrient intake
Introduction
In developed countries, the increasingly higher life
expectancy is mostly attributed to a decline of mortal-
ity at older ages. Better living conditions and notably
healthy dietary practices may explain this greater lon-
gevity in addition to considerable improvement in
health care of older persons [37]. Among the general
lifestyle recommendations, the traditional Mediterra-
nean diet (MeDi) seems an optimal dietary strategy for
J Physiol Biochem
DOI 10.1007/s13105-012-0190-y
C. Feart (*)
Equipe Epidémiologie de la nutrition et des
comportements alimentaires, INSERM, U897,
Université Bordeaux Ségalen,
ISPED Case 11, 146 rue Léo-Saignat,
33076 Bordeaux Cedex, France
e-mail: catherine.feart@isped.u-bordeaux2.fr
B. Alles :B. Merle :C. Samieri :P. Barberger-Gateau
INSERM, U897,
Bordeaux 33076, France
B. Alles :B. Merle :C. Samieri :P. Barberger-Gateau
Université Bordeaux Segalen,
Bordeaux 33076, France
health, since greater MeDi adherence has been associ-
ated with longer survival, reduced risk of cardiovas-
cular or cancer mortality, and reduced risk of
neurodegenerative diseases [22,29,30]. The tradition-
al MeDi is characterized by high consumption of plant
foods (vegetables, fruits, legumes, and cereals), high
intake of olive oil as the principal source of monoun-
saturated fat but low intake of saturated fat, moderate
intake of fish, low to moderate intake of dairy prod-
ucts, low consumption of meat and poultry, and wine
consumed in low to moderate amounts, normally with
meals [36,39]. Therefore, the MeDi combines several
foods and nutrients already proposed as potential pro-
tective factors against several conditions that may
contribute to the development and progression of
age-related diseases [6,12,13,22,23,26,29,30,
34]. However, due to its computation method [36], a
similar score on the MeDi does not mean a similar
intake of nutrients across populations. Only few stud-
ies have yet conducted precise descriptions of its nu-
tritional value [8]. Thus, a better knowledge of the
dietary characteristics of individuals adhering more
closely to a MeDi is necessary to support the biolog-
ical plausibility of its protective effects before promot-
ing this dietary pattern as a universal healthy diet. The
purpose of the current study was to describe the intake
of macronutrients, micronutrients, minerals, and trace
elements according to MeDi adherence among French
older community dwellers, in order to explain the
potential contribution of these nutrients to the health
benefits of the MeDi.
Methods
Participants
The data come from the Three-City (3C) study, a pro-
spective cohort study of vascular risk factors of demen-
tia which methodology has been described elsewhere
[33]. The 3C study protocol was approved by the Con-
sultative Committee for the Protection of Persons par-
ticipating in Biomedical Research at Kremlin-Bicêtre
University Hospital (Paris). A sample of 9,294 commu-
nity dwellers aged 65 and over was selected in 1999–
2000 from the electoral rolls of three French cities
(Bordeaux, Dijon, and Montpellier). All participants
gave written informed consent. At baseline, data collec-
tion included sociodemographic information, lifestyle,
symptoms and medical complaints, medical history,
blood pressure, smoking status, drug use, anthropomet-
ric data, neuropsychological testing, and blood sam-
pling. These data were completed by a comprehensive
dietary survey only in the Bordeaux center in 2001–
2002. The study sample comprises 1,595 participants
from Bordeaux, without missing nutritional data.
Dietary assessment
Participants were visited at home by a specifically
trained dietician who administered a food frequency
questionnaire (FFQ) and a 24-h dietary recall [11,25].
During the 24-h recall, the dietician registered all the
meals and beverages consumed in a period of exactly
24 h before the subject awoke on the day of the
interview. No weekend day was recorded. Quantities
were assessed according to a book of photographs [32]
edited for the SUVIMAX study [16]. A table gave the
correspondence between the portion size and the
weight of the food item. Photographs of dishes and
glasses with the corresponding volume were also
available. The same dietician then entered the data of
the 24-h recall in the Bilnut® software to obtain an
estimation of the daily nutrient intake of each partici-
pant. Food composition tables were already included
in the Bilnut® software and complemented as earlier
described [11]. Among the whole nutrient database,
key macronutrients (proteins, carbohydrates, fats, and
alcohol) were used to compute total energy intake
(EI). The different classes of lipids (saturated (SFA),
monounsaturated (MUFA), and polyunsaturated
(PUFA) fatty acids) were recorded, as the consump-
tion of carotenes; fibers; retinol; vitamins B1, B2, B5,
B6, B12, C, D, and E; folate; and of minerals and trace
elements (calcium, phosphorus, magnesium, potassi-
um, iron, and zinc).
Computation of the MeDi score
Based on the FFQ, frequency of consumption of 40
categories of foods and beverages for each of the three
main meals and three between-meal snacks was
recorded in 11 classes. The number of usual weekly
servings, but not portion size, of each of the 148 food
items and alcohol, ranging as a result from 0 to 42 was
recorded (with a maximum of six meals a day). The food
items were aggregated into 20 food and beverage groups
[25] and we identified those considered to be part of the
C. Feart et al.
MeDi: vegetables, fruits, legumes, cereals (including
bread, pasta, and rice, without distinction between
whole and refined grains), fish, meat, and dairy prod-
ucts. The number of servings per week for each food
group was determined and the MeDi score was comput-
ed as follows: a value of 0 or 1 was assigned to each
food group using sex-specific medians of the population
as cutoffs [35]. Briefly, subjects received one point if
their intake was higher than the sex-specific median for
a presumed protective component (vegetables, fruits,
legumes, cereals, fish, and MUFA to SFA ratio) and
lower than the sex-specific median for a presumed del-
eterious component (meat and dairy products). For al-
cohol, one point was given to mild to moderate
consumers. Cutoffs, chosen to be close to the second
quartile of distribution of total alcohol consumption,
were defined in men and women separately [13]. One
point was attributed for men if consumption was within
six to 14 glasses per week (8.5 to 20 g/day) and for
women if consumption was within one to four glasses
per week (1.4 to 5.7 g/day). Data from the 24-h recall
were used to compute the MUFA to SFA ratio (ratio of
the nutrient intake in gram per day). The MeDi score
was generated by adding the scores for each food cate-
gory for each participant. Thus, the MeDi score could
range from 0 to 9, with higher scores indicating greater
MeDi adherence [13].
Covariates
Sociodemographic information included age, gender,
education, income, and marital status. Vascular risk
factors included measured Body Mass Index (BMI in
kilogram per square meter), diabetes, hypercholester-
olemia (total cholesterol ≥6.2 mmol/L), and hyperten-
sion (measured blood pressure ≥140/90 mmHg or
treated).
Statistical analyses
All statistical analyses were performed with SAS Sta-
tistical package (Version 9.1 SAS Institute). Partici-
pants were classified according to categories of MeDi
adherence (Low MeDi adherence, score 0–3; Middle
MeDi adherence, score 4–5; or High MeDi adherence,
score 6–9) defined to be nutritionally relevant, as
previously described and used in the literature [13,
26,35]. Demographic and clinical characteristics were
compared between the three categories of the MeDi
score. We then explored associations between the
MeDi adherence at baseline and EI and macronutrient
intakes, as expressed in proportion of total energy
intake (% EI) with alcohol. We also compared micro-
nutrient intakes, expressed in gram per day, across
categories of MeDi adherence. Separate analyses were
performed for each gender, since the MeDi score was
computed according to the sex-specific medians of
consumption of food groups.
Results
The study sample consisted of 1,595 individuals (607
men and 988 women), aged 76.1 years on average
(range 67.7–94.9) in 2001–2002. The MeDi score
ranged from 0 to 9 with 1.1 % of the total population
in the extreme values (0 or 9); 43.5 % participants had
a MeDi score of 4 or 5. The mean (SD) MeDi score
was 4.36 (1.67) and was slightly but significantly
higher in men (4.54 (1.72) vs. 4.25 (1.63) in women,
P00.0006). High MeDi adherence (score 6–9) was
significantly more frequent in men (30.1 % vs.
23.0 % in women, P00.003). As expected, greater
MeDi adherence was characterized by higher intake
of vegetables, fruits, legumes, cereals, and fish and
lower intake of meat and dairy products (Table 1).
Greater MeDi adherence was associated with male
sex and being married (Table 2). Individuals in the
middle or high MeDi categories had a lower mean
BMI than those in the lowest category of MeDi score.
Greater MeDi adherence was not associated with hyper-
tension, hypercholesterolemia, or diabetes (Table 2).
In men, mean EI did not significantly differ across
categories of MeDi adherence (Table 3). However,
men in the middle MeDi category (MeDi score 4–5)
had a significant higher mean EI computed without
alcohol than the others. Regarding total protein intake,
there was no significant difference among the three
MeDi categories, although the mean protein intake
from vegetal origin was higher with greater MeDi
adherence. The mean carbohydrate and total fat
intakes did not vary with MeDi adherence. MeDi
adherence was inversely associated with total SFA
intake and consumption of myristic, palmitic, and
stearic acids in men. Total MUFA intake was higher
in men in the middle and highest MeDi categories, and
oleic acid intake was only higher in the middle MeDi
category compared with the others. Total PUFA,
Mediterranean diet adherence and nutrient intakes
notably total n−6 PUFA and linoleic acid intakes,
were higher in the middle MeDi category. Moreover,
there was no association between n−3 PUFA intake
and MeDi adherence in men. The consumption of
alcohol was lower in men with the highest MeDi
adherence (Table 3).
Regarding the consumption of macronutrients in
gram per day, similar trends were observed, since the
MeDi adherence was positively associated with vegetal
proteins, total MUFA, oleic acid, total PUFA, n−6
PUFA, and linoleic acid intakes in men. Moreover, an
inverse association between total SFA and myristic acid
intakes (gram per day) and MeDi adherence was ob-
served (data not shown). In men, the mean linoleic/α-
inolenic acid ratio was positively associated with MeDi
adherence, ranging from 8.78 (SD 5.51) among low
MeDi adherents to 9.46 (SD 6.09) and 10.42 (SD
6.00) among middle and high MeDi adherents, respec-
tively (P00.034).
In women, the total mean EI computed with or
without alcohol was positively associated with MeDi
adherence (Table 3). As expressed in percent of EI
computed with alcohol (Table 3), MeDi adherence in
women was positively associated with intake of vegetal
protein, total carbohydrate, polysaccharides, total
PUFA, total n−6 PUFA, linoleic acid, total n−3PUFA,
and alpha-linolenic acid, and inversely associated with
total SFA and myristic, palmitic, and stearic acid intake
(Table 3). Regarding the macronutrient intake as
expressed in gram per day, similar trends between MeDi
adherence and higher mean consumptions of total pro-
tein, total MUFA, and oleic acid were observed (data
not shown). In addition, the mean (SD) linoleic/α-lino-
lenic acid ratios were statistically different among wom-
en with low, middle, and high MeDi adherence (i.e.,
8.46 (5.51), 10.71 (7.93), and 10.61 (7.69), respectively,
P<0.0001).
Regarding micronutrient, mineral and trace element
intake (Table 4), there was a positive significant asso-
ciation between MeDi adherence and the mean con-
sumption of fibers; vitamins B6, C, and E; folate;
magnesium; potassium; and iron both in men and
women. In men, there was also a significant associa-
tion between carotene intake and MeDi adherence,
which was not observed in women. Conversely, vita-
minB1intakewaspositivelyandsignificantlyassociated
Table 1 Mean number of servings per week for individual food
categories, proportion of mild to moderate alcohol consumers
and mean ratio of daily intake of MUFA to SFA by categories of
Mediterranean diet score among older persons living in Bor-
deaux, the Three-City study (2001–2002) (N01,595)
Low category
(MeDi score 0–3) n0492
Middle category
(MeDi score 4–5) n0693
High category
(MeDi score 6–9) n0410
Food categories
*
Mean serving/week (SD)
Dairy products 20.65 (8.02) 18.08 (7.69)
**
16.14 (6.71)
**,***
Meat 5.36 (2.59) 4.80 (2.41)
**
4.18 (2.22)
**,***
Vegetables 15.56 (6.41) 19.37 (7.11)
**
23.44 (6.61)
**,***
Fruits 10.66 (6.79) 13.84 (6.55)
**
16.45 (6.50)
**,***
Legumes 0.42 (0.68) 0.62 (0.62)
**
0.82 (0.65)
**,***
Cereals 19.78 (6.52) 22.16 (5.76)
**
24.54 (5.48)
**,***
Fish 2.02 (1.49) 2.91 (1.69)
**
3.76 (1.73)
**,***
Mild to moderate alcohol (%) 14.63 25.40
**
46.10
**,***
MUFA to SFA ratio, mean (SD) 0.77 (0.26) 0.88 (0.32)
**
1.00 (0.31)
**,***
MeDi Mediterranean diet, MUFA to SFA monounsaturated fatty acid to saturated fatty acid ratio (intake g/d)
*
All Pvalues <.0001 for the analysis of variance (ANOVA) or for the Chi-square test (proportion of mild to moderate alcohol
consumers) among categories of Mediterranean diet score
**
2×2 significant comparisons (P< 0.05) for mean number of servings per week for individual food categories, proportion of mild to
moderate alcohol consumers, and mean ratio of daily intake of MUFA to SFA between categories of Mediterranean diet score taking the
lowest category as reference group
***
2×2 significant comparisons (P< 0.05) for mean number of servings per week for individual food categories, proportion of mild to
moderate alcohol consumers, and mean ratio of daily intake of MUFA to SFA between individuals in middle and high categories of
Mediterranean diet score
C. Feart et al.
with MeDi adherence in women only. Finally, there was
an inverse association between MeDi adherence and
calcium intake which was the lowest in the middle and
highest MeDi categories in women, while this inverse
association was of borderline significance in men
(Table 4).
Discussion
This large population-based cross-sectional study in
older persons living in South-western France shows a
high coherence between dietary data collected by FFQ
and used to assess a diet modeled on the traditional
MeDi, and the nutrient intake estimated by 24-h recall.
As expected, greater MeDi adherence was associated
with total mean higher vegetal protein intake (% EI),
lower total mean SFA intake (% EI), higher total mean
PUFA intake (% EI), but also notably with total n−6
PUFA and linoleic acid intakes, both in men and
women. In men, higher total mean MUFA and oleic
acid intakes but less alcohol consumption (% EI) were
observed with greater MeDi adherence. Total carbohy-
drate and polysaccharide intakes (% EI) and total mean
n−3 PUFA and alpha-linolenic acid intakes (% EI) were
positively associated with MeDi adherence in women
but not in men. The consumption of several micronu-
trients, minerals, and trace elements (i.e., fibers; vita-
mins B6, C, and E; folate; magnesium; potassium; and
iron) increased with higher MeDi adherence, in both
men and women. Only the consumption of calcium
decreased with greater MeDi adherence in women and
with borderline significance in men.
The concept of MeDi has been first introduced in
the Seven-Country study by Keys et al. who reported
low all-cause and coronary heart disease death rates in
cohorts of the Mediterranean basin with olive oil as
the main fat [18]. Not a specific pattern but a collec-
tion of eating habits traditionally followed by the
populations of the Mediterranean basin first defined
the so-called “traditional”MeDi. However, there is no
single MeDi but several definitions because dietary
habits vary considerably across the Mediterranean
countries bordering the sea [3,28]. Compared with
Table 2 Demographic and clin-
ical characteristics by categories
of Mediterranean diet score
among older persons living in
Bordeaux, the Three-City study,
2001–2002 (N01,595)
Pvalue for the Chi-square test or
analysis of variance (ANOVA)
among categories of Mediterra-
nean diet score
MeDi Mediterranean diet, BMI
Body Mass Index
*
2× 2 significant comparisons (P
<0.05) for means of age and BMI
between categories of Mediterra-
nean diet score taking the lowest
category as reference group
Low category
(MeDi score 0–3)
(n0492)
Middle category
(MeDi score 4–5)
(n0693)
High category
(MeDi score 6–9)
(n0410)
P
Demographic characteristics
Age (n), mean (SD) (y) (492) 76.4 (5.1) (693) 76.2 (5.1) (410) 75.8 (4.7) 0.13
Wom en (n) (%) (325) 66.1 (436) 62.9 (227) 55.4 0.003
Education (n)(%)
No or primary school (182) 37.1 (221) 32.1 (125) 30.7 0.11
Secondary (136) 27.8 (192) 27.9 (106) 26.0
High school (102) 20.8 (143) 20.8 (90) 22.1
University (70) 14.3 (133) 19.3 (86) 21.1
Monthly income (€)(n)(%)
Refused to answer (40) 8.1 (55) 7.9 (26) 6.3 0.06
<750 (36) 7.3 (53) 7.7 (22) 5.4
750–1,500 (168) 34.1 (202) 29.1 (109) 26.6
1,500–2,250 (122) 24.8 (166) 24.0 (111) 27.1
>2,250 (126) 25.6 (217) 31.3 (142) 24.6
Marital status (n)(%)
Married (237) 48.2 (371) 53.5 (257) 62.7 0.0006
Divorced/separated (43) 8.7 (49) 7.1 (29) 7.1
Widowed (183) 37.2 (227) 32.8 (97) 23.7
Single (29) 5.9 (46) 6.6 (27) 6.6
Clinical characteristics
BMI (n), mean (SD) (480) 26.9 (4.8) (676) 25.2 (4.1)
*
(408) 26.1 (3.7)
*
0.01
Hypertension (n) (%) (371) 75.9 (529) 76.6 (322) 78.5 0.62
Hypercholesterolemia (n) (%) (257) 56.0 (372) 57.8 (237) 59.9 0.52
Diabetes (n) (%) (55) 11.2 (64) 9.2 (35) 8.6 0.36
Mediterranean diet adherence and nutrient intakes
usual single-food or nutrient methods, the dietary pat-
tern approach is appealing because analyses based on
single nutrients ignore important interactions (addi-
tive, synergistic, or antagonist effects) between com-
ponents of diet and because people did not eat isolated
nutrients [17]. The computation of the MeDi score
[35], as used in the present study, combines several
foods providing nutrients which have been identified
as potential protective factors against cardiovascular
diseases such as fruit and vegetables, fish, as the main
provider of long-chain n−3 PUFA, olive oil as main
source of MUFA, and a moderate consumption of
Table 3 Energy, macronutrient, and alcohol intakes as expressed in proportion of energy intake by categories of Mediterranean diet
score among older persons living in Bordeaux, the Three-City study, 2001–2002 (N01,595)
Men Women
Low MeDi
category
(score 0–3)
(n0167)
Middle MeDi
category
(score 4–5)
(n0257)
High MeDi
category
(score 6–9)
(n0183)
PLow MeDi
category
(score 0–3)
(n0325)
Middle MeDi
category
(score 4–5)
(n0436)
High MeDi
category
(score 6–9)
(n0227)
P
Total energy intake
(kcal) mean (SD)
1,967 (555) 2,058 (533) 2,004 (498) 0.19 1,453 (457) 1,540 (449)
*
1,585 (467)
*
0.002
Energy intake without alcohol
(kcal) mean (SD)
1,786 (516) 1,909 (510)
*
1,873 (494) 0.049 1,400 (441) 1,492 (435)
*
1,538 (451)
*
0.0007
Nutrient intake mean (SD)
a
Proteins 17.1 (4.6) 17.3 (4.2) 17.3 (4.2) 0.85 18.6 (5.1) 18.5 (5.0) 18.3 (4.7) 0.77
Animal proteins 12.3 (4.8) 12.2 (4.4) 12.1 (4.6) 0.89 13.9 (5.4) 13.3 (5.4) 12.8 (5.0) 0.07
Vegetal proteins 4.8 (1.5) 5.1 (1.3) 5.3 (1.5)
*
0.005 4.7 (1.6) 5.1 (1.6)
*
5.5 (1.7)
*
<0.0001
Carbohydrates 44.5 (10.0) 44.3 (10.0) 46.1 (9.3) 0.14 46.1 (10.0) 46.9 (9.9) 48.2 (9.6)
*
0.044
Mono/disaccharides 20.0 (10.3) 18.7 (8.0) 20.0 (6.7) 0.18 23.3 (9.8) 22.6 (8.1) 23.1 (8.0) 0.50
Polysaccharides 24.5 (8.4) 25.6 (7.3) 26.1 (7.4) 0.16 22.8 (8.7) 24.3 (8.4)
*
25.2 (8.0)
*
0.003
Total fat 29.7 (8.3) 31.2 (8.7) 30.0 (8.6) 0.13 31.9 (9.2) 31.6 (8.8) 30.6 (8.9) 0.26
Total SFA 13.6 (4.3) 13.4 (4.3) 11.9 (4.1)
*,**
<0.0001 14.8 (5.1) 13.4 (4.6)
*
11.9 (4.1)
*,**
<0.0001
Myristic acid 1.58 (0.75) 1.49 (0.70) 1.22 (0.64)
*,**
<0.0001 1.78 (0.89) 1.48 (0.76)
*
1.20 (0.62)
*,**
<0.0001
Palmitic acid 6.64 (2.18) 6.59 (2.22) 6.12 (2.26)
*,**
0.042 7.21 (2.60) 6.66 (2.52)
*
6.15 (2.36)
*,**
<0.0001
Stearic acid 2.97 (1.22) 2.87 (1.27) 2.58 (1.17)
*,**
0.0007 3.06 (1.29) 2.86 (1.37) 2.56 (1.19)
*,**
<0.0001
Total MUFA 10.2 (3.7) 11.3 (4.1)
*
11.2 (4.1)
*
0.01 10.9 (3.9) 11.2 (4.1) 11.5 (4.1) 0.23
Palmitoleic acid 0.72 (0.45) 0.78 (0.63) 0.78 (0.71) 0.51 0.80 (0.60) 0.83 (0.83) 0.74 (0.67) 0.32
Oleic acid 8.70 (3.44) 9.73 (3.95)
*
9.60 (3.82) 0.017 9.37 (3.67) 9.63 (3.85) 10.04 (3.95) 0.13
Total PUFA 3.8 (2.1) 4.3 (2.6) 4.7 (2.3)
*
0.004 3.9 (2.1) 4.6 (2.6)
*
5.0 (3.2)
*
<0.0001
Total n−6 PUFA 2.9 (1.9) 3.4 (2.3) 3.7 (2.1)
*
0.005 2.9 (2.0) 3.6 (2.4)
*
3.8 (2.8)
*
<0.0001
Linoleic acid 2.85 (1.87) 3.28 (2.31) 3.58 (2.12)
*
0.006 2.84 (2.01) 3.49 (2.40)
*
3.76 (2.80)
*
<0.0001
Arachidonic acid 0.07 (0.07) 0.08 (0.08) 0.09 (0.09) 0.13 0.09 (0.10) 0.09 (0.09) 0.09 (0.09) 0.81
Total n−3 PUFA 0.6 (0.5) 0.6 (0.6) 0.7 (0.6) 0.36 0.6 (0.7) 0.7 (0.8) 0.8 (0.8)
*
0.041
a-Linolenic acid 0.37 (0.25) 0.42 (0.39) 0.40 (0.28) 0.32 0.37 (0.19) 0.39 (0.32) 0.44 (0.44)
*
0.03
EPA 0.07 (0.17) 0.06 (0.15) 0.08 (0.18) 0.39 0.06 (0.19) 0.08 (0.21) 0.09 (0.18) 0.26
DHA 0.12 (0.29) 0.12 (0.31) 0.17 (0.40) 0.24 0.14 (0.41) 0.16 (0.40) 0.20 (0.42) 0.23
Alcohol 9.0 (7.7) 7.2 (6.1) 6.6 (5.9)
*
0.007 3.5 (4.8) 3.0 (4.3) 2.9 (4.2) 0.23
P-value for ANOVA comparing energy intake, macronutrient, and alcohol intakes in g/d among the three Mediterranean Diet Score
categories
MeDi Mediterranean diet, SFA saturated fatty acids, MUFA monounsaturated fatty acid, PUFA polyunsaturated fatty acid, EPA
eicosapentaenoic acid, DHA docosahexaenoic acid
a
Expressed as percentage of energy intake (computed with alcohol)
*
2×2 significant comparisons (P< 0.05) for means of nutrient intake between categories of Mediterranean diet score taking the lowest
category as reference group
**
2×2 significant comparisons (P<0.05) for means of nutrient intake between individuals in middle and high categories of Mediterranean
diet score
C. Feart et al.
alcohol. On the other hand, higher total fat and SFA
intake have been considered as harmful factors for car-
diovascular diseases and related mortality [7,20,24].
Our findings support the presumed beneficial effects
of the MeDi. Regarding macronutrients, higher adher-
ence to a MeDi was characterized by a high intake of
vegetal protein but not animal protein; complex carbo-
hydrates, hence a lower glycemic index; a low con-
sumption of SFA which detrimental effects are well
documented—although debated [1], and a high con-
sumption of MUFA, especially oleic acid, and PUFA.
Despite a higher intake of n−3 PUFA with great er MeDi
adherence, notably in women, the ratio of n−6/n−3
PUFA precursors increased with MeDi score. This
may be, in part, due to the high proportion of consumers
of oils from various vegetal origins, which are providers
of linoleic acid, the precursor of n−6 PUFA. Indeed,
60 % of the whole sample were regular consumers (i.e.,
preferred dietary fats used for dressing or cooking) of
olive oil and 44 % of sunflower oil, although 32 % were
simultaneously regular consumers of both oils (data not
shown). Hence, the ratio of n−6/n−3 PUFA precursors,
Table 4 Micronutrient intakes, as expressed in mg/d, by categories of Mediterranean diet score among older persons living in
Bordeaux, the Three-City study, 2001–2002 (N01,595)
Men Women
Low MeDi
category
(score 0–3)
(n0167)
Middle MeDi
category
(score 4–5)
(n0257)
High MeDi
category
(score 6–9)
(n0183)
PLow MeDi
category
(score 0–3) (n0325)
Middle MeDi
category
(score 4–5)
(n0436)
High MeDi
category
(score 6–9)
(n0227)
P
Nutrient intake, mean (SD)
a
Miconutrients
Carotenes 2.97 (4.59) 3.06 (4.07) 4.58 (6.67)
*,**
0.003 3.60 (5.65) 3.38 (5.11) 3.73 (4.72) 0.69
Fibres 16.82 (7.20) 20.28 (8.24)
*
21.47 (8.01)
*
<0.0001 13.88 (6.88) 16.41 (6.89)
*
18.20 (8.07)
*,**
<0.0001
Retinol 708.1 (1,851.6) 644.1 (1,545.2) 736.7 (2,021.0) 0.85 840.7 (4,652.4) 628.5 (2,621.7) 607.4 (1,828.3) 0.62
Vitamin B1 1.08 (0.41) 1.15 (0.52) 1.17 (0.44) 0.20 0.88 (0.39) 0.95 (0.40) 1.01 (0.41)
*
0.0008
Vitamin B2 1.69 (0.66) 1.72 (0.63) 1.68 (0.66) 0.85 1.50 (0.81) 1.46 (0.75) 1.50 (0.66) 0.76
Vitamin B5 4.31 (1.76) 4.56 (1.59) 4.60 (1.71) 0.20 3.76 (1.94) 3.85 (1.68) 4.05 (1.72) 0.17
Vitamin B6 1.52 (0.54) 1.69 (0.62)
*
1.68 (0.58)
*
0.008 1.22 (0.51) 1.34 (0.53)
*
1.42 (0.58)
*
0.0001
Folates 271.3 (119.9) 305.2 (128.7)
*
325.9 (144.5)
*
0.0005 230.8 (125.9) 257.9 (119.8)
*
294.0 (143.6)
*,**
<0.0001
Vitamin B12 6.17 (10.39) 5.91 (8.19) 6.00 (10.23) 0.96 5.45 (12.78) 5.44 (12.20) 5.23 (0.64) 0.97
Vitamin C 74.7 (63.6) 84.74 (60.52) 95.25 (67.72)
*
0.011 74.05 (57.53) 82.16 (57.01) 92.10 (66.30)
*
0.002
Vitamin D 1.68 (1.91) 2.13 (3.11) 1.78 (2.15) 0.16 1.55 (2.70) 1.64 (2.96) 1.71 (2.93) 0.79
Vitamin E 5.72 (4.99) 7.02 (4.17)
*
8.59 (5.58)
*,**
<0.0001 5.10 (3.47) 6.39 (4.63)
*
6.86 (4.30)
*
<0.0001
Minerals and trace elements
Calcium 947.7 (481.9) 953.7 (433.6) 862.1 (415.7) 0.07 892.9 (471.9) 810.2 (402.1)
*
773.5 (381.4)
*
0.002
Phosphorus 1.24 (0.42) 1.27 (0.39) 1.22 (0.35) 0.38 1.02 (0.39) 1.03 (0.36) 1.04 (0.36) 0.87
Magnesium 276.0 (78.5) 294.7 (83.3) 301.2 (80.0)
*
0.01 221.0 (71.5) 235.1 (74.8)
*
252.6 (83.7)
*,**
<0.0001
Potassium 2.90 (0.88) 3.04 (0.86) 3.13 (0.83)
*
0.04 2.37 (0.76) 2.52 (0.83) 2.72 (0.88)
*,**
<0.0001
Iron 12.53 (5.56) 13.01 (4.71) 14.44 (8.14)
*
0.01 8.79 (4.53) 9.73 (4.76)
*
10.88 (5.62)
*,**
<0.0001
Zinc 6.41 (4.98) 7.24 (5.27) 7.21 (5.76) 0.24 7.30 (7.51) 7.53 (7.19) 8.40 (7.42) 0.20
P-value for the ANOVA comparing energy, macronutrient, alcohol and micronutrients intakes among the three Mediterranean Diet
Score categories
MeDi Mediterranean diet
a
Expressed as mean intake in mg/d except for fibres, sodium, phosphorus, and potassium, all expressed in g/d, and for retinol, vitamin
D, and folates, all expressed in μg/d
*
2×2 significant comparisons (P< 0.05) for means of energy and nutrient intake between categories of Mediterranean diet score taking
the lowest category as reference group
**
2×2 significant comparisons (P< 0.05) for means of energy and nutrient intake between individuals in middle and high categories of
Mediterranean diet score
Mediterranean diet adherence and nutrient intakes
high in our sample and far above the recommended
value of 5 [19], might be a restriction to beneficial
effects of the MeDi. Higher adherence to the MeDi
was also associated with higher intake of various anti-
oxidants (vitamins C and E, carotenes in men, and also
probably polyphenols from plants although we could
not ascertain their amount) which contribute to protect
long-chain n−3 PUFA from lipid peroxidation. The
MeDi also provided B vitamins, especially folate. Nev-
ertheless, higher MeDi adherence is characterized by
lower intake of dairy products. Therefore, the lower
dietary intake of calcium associated with higher MeDi
adherence in women, may be a threat to their bone
health. This finding suggests that the MeDi as presently
defined may not be a universally protective diet. Refine-
ment in the computation of the score is needed to
disentangle the presumed detrimental effects of some
SFA from the beneficial effects of calcium found in
dairy products.
Overall, the results of the present analysis showed a
high coherence between the two methodologies used
in the 3C study to describe the anatomy of the
Mediterranean-type diet of older participants of the
3C cohort. However, some expected associations be-
tween MeDi adherence and various nutrient intakes
(for instance, total mean n−3 PUFA in men, carotene
in women, and vitamins D and B12) have not been
highlighted. This may be due to the use of a single 24-
h recall which cannot handle high intraindividual var-
iation [11]. Thus, a reported single day of intake is
unlikely to be representative of usual individual in-
take. However, a sufficiently large sample may allow
to determine average intake of defined subgroups [38].
Moreover, a single 24-h recall prevented us to capture
the intake of nutrients which are not consumed on a
daily basis, but less often. The FFQ used in the present
study assessed number of servings but not portion size.
The lack of consideration of amount of each food con-
sumed may lead to consider people with the same fre-
quency of each food group of the MeDi score, and the
same MeDi score, as comparable, although they may
have different quantitative food consumptions. We can-
not exclude that the quantity of presumed protective or
deleterious nutrients consumed by men and women with
a similar MeDi score may differ, which reduced our
chance to evidence associations between food and nutri-
ent intakes. Finally, alcohol was an important component
of total EI, notably in men [11], leading us to consider
this nutrient provider of energy in the computation of the
total EI in the present analysis. However, a low to
moderate alcohol consumption is also one of the nine
components of the MeDi score, which may have added
bias in the present study.
Analyses of single nutrients using 24-h dietary recalls
could help understand specific mechanisms involved in
physiopathological processes leading to diseases, by
establishing biological plausibility. Therefore, the bio-
logical basis for the well-documented health benefits of
the MeDi involves a decrease in oxidative stress and
inflammation which also participates in the dysregula-
tion of cellular function, the exacerbation of degenerative
diseases, such as cardiovascular diseases or atheroscle-
rosis [5,7,20,31]. Individuals with higher MeDi adher-
ence have been shown to have higher reduced to
oxidized glutathione ratio and lower insulin resistance,
oxidized low-density lipoprotein, C-reactive protein, and
interleukin-18 levels [4,9,15]. It has been hypothesized
that the n−3 PUFA and a balanced n−6/n−3PUFAratio
associated with a high intake of antioxidant in the MeDi
could be important determinants in influencing cardio-
vascular and cerebral diseases [14,21,27]. In the current
study, participants with lower MeDi adherence exhibited
also greater BMI. Surprisingly, there were no apparent
vascular benefits of a greater MeDi adherence in the
current sample where the prevalence of hypertension,
hypercholesterolemia, and diabetes did not differ across
MeDi categories. This has already been observed in
similar samples of elderly people involved in the 3C
study, who exhibited a global healthy lifestyle [13,33].
Finally, the computation of the MeDi score was able to
assess associations between a diet modeled on the tradi-
tional MeDi, but which is specific to the sample studied.
Indeed, the use of sex-specific cutoff points does not
measure adherence to a universal traditional MeDi pat-
tern but rather to a specific pattern, which precludes
generalizing results of the present study to other popula-
tions who have, on the whole, different food intake and
MeDi adherence [2]. Moreover, including only nine food
groups, the MeDi score does not consider the overall
correlation between all foods [17], among which some
food groups could reflect specific health concerns and
behavior, such as the consumption of dietary supple-
ments. Despite these limitations, the strengths of the
present study are its sample size, the population-based
design, and the accuracy of food intake assessment
which validates the methodology employed.
In conclusion, this study conducted in French el-
derly community dwellers allowed to describe the
C. Feart et al.
basis of their Mediterranean-type diet and the actual
consumption of nutrients associated with MeDi adher-
ence. As residual confounding by other aspects of
lifestyle cannot be ruled out, these nutrients probably
do not fully explain the better health of persons who
adhere to this MeDi, but they likely contribute direct-
ly. At least regarding clinical parameters, old consum-
ers who conform with MeDi have only a lower BMI
and no adverse effects of an elevated n−6/n−3 PUFA
ratio. This dietary pattern may indirectly constitute an
indicator of a complex set of favorable social and
lifestyle factors that contribute to global better health.
Indeed, food choices and dietary habits which reflect
individual food preferences are highly sensitive to
culture, education, socioeconomic status, sociodemo-
graphic characteristics, environmental and lifestyle
determinants, as well as to age- or nutrition-related
diseases [10].
Funding/Support The Three-City Study is conducted under a
partnership agreement between the Institut National de la Santé
et de la Recherche Médicale (INSERM), the Institut de Santé
Publique et Développement of the Victor Segalen Bordeaux 2
University, and Sanofi-Aventis. The Fondation pour la Recher-
che Médicale funded the preparation and initiation of the study.
The 3C Study is also supported by the Caisse Nationale Maladie
des Travailleurs Salariés, Direction Générale de la Santé,
Mutuelle Générale de l’Education Nationale, Institut de la Lon-
gévité, Regional Governments of Aquitaine and Bourgogne,
Fondation de France, and Ministry of Research—INSERM
Programme “Cohortes et collections de données biologiques.”
Role of the sponsors Study sponsors played no role in the
design and conduct of the study; collection, management, anal-
ysis, and interpretation of the data; and preparation, review, or
approval of the manuscript.
Conflict of interest P. Barberger-Gateau received fees for
conferences from Danone, Lesieur, Bauch & Lomb, and Aprifel
and benefits from research grants from Danone and Lesieur. The
other authors declared no support from any institution for the
submitted work.
References
1. Astrup A, Dyerberg J, Elwood P, Hermansen K, Hu FB,
Jakobsen MU, Kok FJ, Krauss RM, Lecerf JM, LeGrand P,
Nestel P, Riserus U, Sanders T, Sinclair A, Stender S,
Tholstrup T, Willett WC (2011) The role of reducing intakes
of saturated fat in the prevention of cardiovascular disease:
where does the evidence stand in 2010? Am J Clin Nutr
93:684–688
2. Bach A, Serra-Majem L, Carrasco JL, Roman B, Ngo J,
Bertomeu I, Obrador B (2006) The use of indexes evaluat-
ing the adherence to the Mediterranean diet in epidemiolog-
ical studies: a review. Public Health Nutr 9:132–146
3. Bach-Faig A, Berry EM, Lairon D, Reguant J, Trichopoulou
A, Dernini S, Medina FX, Battino M, Belahsen R, Miranda G,
Serra-Majem L (2011) Mediterranean diet pyramid today. Sci-
ence and cultural updates. Public Health Nutr 14:2274–2284
4. Bach-Faig A, Geleva D, Carrasco JL, Ribas-Barba L, Serra-
Majem L (2006) Evaluating associations between Mediter-
ranean diet adherence indexes and biomarkers of diet and
disease. Public Health Nutr 9:1110–1117
5. Bartali B, Frongillo EA, Guralnik JM, Stipanuk MH, Allore
HG, Cherubini A, Bandinelli S, Ferrucci L, Gill TM (2008)
Serum micronutrient concentrations and decline in physical
function among older persons. JAMA 299:308–315
6. Berr C, Portet F, Carriere I, Akbaraly TN, Feart C, Gourlet
V, Combe N, Barberger-Gateau P, Ritchie K (2009) Olive
oil and cognition: results from the Three-City Study.
Dement Geriatr Cogn Disord 28:357–364
7. Bhupathiraju SN, Tucker KL (2011) Coronary heart disease
prevention: nutrients, foods, and dietary patterns. Clin Chim
Acta 412:1493–1514
8. Bountziouka V, Tzavelas G, Polychronopoulos E, Constantinidis
TC, Panagiotakos DB (2011) Validity of dietary patterns
derived in nutrition surveys using a priori and a poste-
riori multivariate statistical methods. Int J Food Sci
Nutr 62:617–627
9. Dai J, Jones DP, Goldberg J, Ziegler TR, Bostick RM,
Wilson PW, Manatunga AK, Shallenberger L, Jones L,
Vaccarino V (2008) Association between adherence to the
Mediterranean diet and oxidative stress. Am J Clin Nutr
88:1364–1370
10. Drewnowski A, Eichelsdoerfer P (2009) The Mediterranean
diet: does it have to cost more? Public Health Nutr 12:1621–
1628
11. Feart C, Jutand MA, Larrieu S, Letenneur L, Delcourt C,
Combe N, Barberger-Gateau P (2007) Energy, macronutri-
ent and fatty acid intake of French elderly community
dwellers and association with socio-demographic character-
istics: data from the Bordeaux sample of the Three-City
Study. Br J Nutr 98:1046–1057
12. Feart C, Samieri C, Barberger-Gateau P (2010) Mediterra-
nean diet and cognitive function in older adults. Curr Opin
Clin Nutr Metab Care 13:14–18
13. Feart C, Samieri C, Rondeau V, Amieva H, Portet F, Dartigues
JF, Scarmeas N, Barberger-Gateau P (2009) Adherence to a
Mediterranean diet, cognitive decline, and risk of dementia.
JAMA 302:638–648
14. Feart C, Torres MJ, Samieri C, Jutand MA, Peuchant E,
Simopoulos AP, Barberger-Gateau P (2011) Adherence to a
Mediterranean diet and plasma fatty acids: data from the
Bordeaux sample of the Three-City Study. Br J Nutr 106
(1):149–58
15. Giugliano D, Esposito K (2008) Mediterranean diet and
metabolic diseases. Curr Opin Lipidol 19:63–68
16. Hercberg S, Preziosi P, Briancon S, Galan P, Triol I, Malvy
D, Roussel AM, Favier A (1998) A primary prevention trial
Mediterranean diet adherence and nutrient intakes
using nutritional doses of antioxidant vitamins and minerals
in cardiovascular diseases and cancers in a general popula-
tion: the SU.VI.MAX study—design, methods, and partic-
ipant characteristics. SUpplementation en VItamines et
Mineraux AntioXydants. Control Clin Trials 19:336–351
17. Kant AK (2004) Dietary patterns and health outcomes. J
Am Diet Assoc 104:615–635
18. Keys A, Menotti A, Karvonen MJ, Aravanis C, Blackburn
H, Buzina R, Djordjevic BS, Dontas AS, Fidanza F, Keys
MH et al (1986) The diet and 15-year death rate in the seven
countries study. Am J Epidemiol 124:903–915
19. Martin A (2001) The “apports nutritionnels conseilles (ANC)”
for the French population. Reprod Nutr Dev 41:119–128
20. Mente A, de Koning L, Shannon HS, Anand SS (2009) A
systematic review of the evidence supporting a causal link
between dietary factors and coronary heart disease. Arch
Intern Med 169:659–669
21. Panagiotakos DB, Kastorini CM, Pitsavos C, Stefanadis C
(2011) The current Greek diet and the omega-6/omega-3 bal-
ance: the Mediterranean diet score is inversely associated with
the omega-6/omega-3 ratio. World Rev Nutr Diet 102:53–56
22. Roman B, Carta L, Martinez-Gonzalez MA, Serra-Majem L
(2008) Effectiveness of the Mediterranean diet in the elder-
ly. Clin Interv Aging 3:97–109
23. Samieri C, Feart C, Letenneur L, Dartigues JF, Peres K,
Auriacombe S, Peuchant E, Delcourt C, Barberger-Gateau P
(2008) Low plasma eicosapentaenoic acid and depressive
symptomatology are independent predictors of dementia
risk. Am J Clin Nutr 88:714–721
24. Samieri C, Feart C, Proust-Lima C, Peuchant E, Tzourio C,
Stapf C, Berr C, Barberger-Gateau P (2011) Olive oil con-
sumption, plasma oleic acid, and stroke incidence: the
Three-City Study. Neurology 77:418–425
25. Samieri C, Jutand MA, Feart C, Capuron L, Letenneur L,
Barberger-Gateau P (2008) Dietary patterns derived by hy-
brid clustering method in older people: association with
cognition, mood, and self-rated health. J Am Diet Assoc
108:1461–1471
26. Scarmeas N, Luchsinger JA, Schupf N, Brickman AM,
Cosentino S, Tang MX, Stern Y (2009) Physical activity,
diet, and risk of Alzheimer disease. JAMA 302:627–637
27. Simopoulos AP (2001) The Mediterranean diets: what is so
special about the diet of Greece? The scientific evidence. J
Nutr 131:3065S–3073S
28. Sofi F (2009) The Mediterranean diet revisited: evidence of
its effectiveness grows. Curr Opin Cardiol 24:442–446
29. Sofi F, Abbate R, Gensini GF, Casini A (2010) Accruing
evidence on benefits of adherence to the Mediterranean diet
on health: an updated systematic review and meta-analysis.
Am J Clin Nutr 92:1189–1196
30. Sofi F, Cesari F, Abbate R, Gensini GF, Casini A (2008)
Adherence to Mediterranean diet and health status: meta-
analysis. BMJ 337:a1344
31. Solfrizzi V, D’Introno A, Colacicco AM, Capurso C,
Palasciano R, Capurso S, Torres F, Capurso A, Panza F
(2005) Unsaturated fatty acids intake and all-causes mortal-
ity: a 8.5-year follow-up of the Italian Longitudinal Study
on Aging. Exp Gerontol 40:335–343
32. SU.VI.MAX (2000) Portions alimentaires. Manuel photos
pour l’estimation des quantités
33. The 3C Study Group (2003) Vascular factors and risk of
dementia: design of the Three-City Study and baseline
characteristics of the study population. Neuroepidemiology
22:316–325
34. Trichopoulou A, Bamia C, Trichopoulos D (2009) Anatomy
of health effects of Mediterranean diet: Greek EPIC pro-
spective cohort study. BMJ 338:b2337
35. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D
(2003) Adherence to a Mediterranean diet and survival in
a Greek population. N Engl J Med 348:2599–2608
36. Trichopoulou A, Kouris-Blazos A, Wahlqvist ML, Gnardellis
C, Lagiou P, Polychronopoulos E, Vassilakou T, Lipworth L,
Trichopoulos D (1995) Diet and overall survival in elderly
people. BMJ 311:1457–1460
37. Trichopoulou A, Vasilopoulou E (2000) Mediterranean diet
and longevity. Br J Nutr 84:S205–S209
38. Willett W (1998) Nutritional epidemiology. Oxford Univer-
sity Press, NY, p 514
39. Willett WC, Sacks F, Trichopoulou A, Drescher G, Ferro-
Luzzi A, Helsing E, Trichopoulos D (1995) Mediterranean
diet pyramid: a cultural model for healthy eating. Am J Clin
Nutr 61:1402S–1406S
C. Feart et al.
