1. Nutrition 26 (2010) 75–81
www.nutritionjrnl.com
Applied nutritional investigation
Dietary intakes of essential nutrients among Arab and Berber ethnic
groups on rural Tunisian island
Thouraya Baroudi, Ph.D.a,*, Hedi Ben Maiz, Ph.D., M.D.b,
Hafaoua Kammoun Abid, Ph.D., M.D.b, Amel Benammar-Elgaaied, Ph.D.a,
and Leila Trabelsi Alouane, Ph.D.c
a
Laboratory of Genetics, Immunology and Human Pathologies, Faculty of Sciences of Tunis, El Manar II University, Tunis, Tunisia
b
Hospital Charles Nicolle, Internal Medicine A, Tunis, Tunisia
c
High School of the Sciences and Techniques of the Health of Tunis, University of Tunis II, Bab Suika BP 176, Tunis, Tunisia
Manuscript received February 16, 2008; accepted May 18, 2009.
Abstract Objective: The dietary intake was investigated and food sources were identified among Tunisian
ethnic groups from Jerba Island in the south of Tunisia.
Methods: Ninety-four subjects of moderate socioeconomic status (47 Berbers and 47 Arabs) aged 32
to 64 y completed a 1-mo qualitative food-frequency questionnaire and a single 24-h dietary recall, and
dietary intakes and demographic status were observed from 2006 to 2007.
Results: The prevalence of overweight and obesity was not significantly associated with Arab men
compared with Berber men. Therefore, obesity was significantly associated with Berber women
(P < 0.001). Height was significantly different between Arab and Berber women (P < 0.001). There
were no significant differences in energy intake between men and women. Protein intake was not
significantly different between ethnic groups. Milk and dairy products in the Berber group were
significantly different from the Arab group. Intakes of calcium, zinc, iron, and folate were below
recommended nutrient intakes in men and women in the two ethnic groups. Vitamin E intake was
greater in Berbers than in Arabs (P < 0.01).
Conclusion: Ethnicity was significantly associated with dietary intakes in the two ethnic groups of
Jerba Island. Ó 2010 Published by Elsevier Inc.
Keywords: Ethnicity; Dietary intake; Arabs; Berbers; Jerba Island
Introduction Although the two ethnic groups are Muslim, they differ by
their doctrines. Religious and cultural differences have repre-
Ethnicity is a complex construct of biology, culture, sented an obstacle to their intermixing and marriages usually
language, religion, and distinct health beliefs and behaviors occur between members from the same extended family in
encompassing a range of biological and environmental expo- the same ethnic group.
sures [1]. The Tunisian population is characterized by its The influence of lifestyle patterns on health status has
diversity. In Tunisia, there still exists, despite civilizations been evaluated in numerous ecologic and analytic studies.
mixing, areas where ethnic separation remains clear. Among Avoiding cigarette smoking [2], maintaining a normal body
these areas Jerba Island (in the southeast) shelters two differ- mass index (BMI) [3], and adding physical activity to daily
ent populations in their culture and their lifestyle. The first activities are well studied [4]. Lifestyle habits are strongly
consists of Berbers, original citizens of the island, whose associated with decreased rates of mortality. There is
history extends to the end of the Paleolithic period. The growing awareness that variations in dietary practices among
second community consists of Arabs installed in the island ethnic groups may help to explain interethnic differences
at the time of the Islamic conquest during the 7th century. in morbidity and mortality. These differences have been
identified among many ethnic groups [5–7].
*Corresponding author. Tel.: þ216-71-60-65-15; fax: þ216-71-82-17-0. To our knowledge, the assessment of ethnic differences in
E-mail address: Thourbar@yahoo.fr (T. Baroudi). food intake or physical activity patterns between Berbers and
0899-9007/10/$ – see front matter Ó 2010 Published by Elsevier Inc.
doi:10.1016/j.nut.2009.05.020

2. 76 T. Baroudi et al. / Nutrition 26 (2010) 75–81
Arabs has not yet been performed. The aims of this study bolic rate (BMR). To estimate the BMR we used predictive
were to describe the dietary patterns in two ethnic groups equations based on these factors [12]. Multiplying the PAL
on Jerba Island and to relate these patterns to anthropometric by the BMR produced the total energy expenditure.
measurements. The same investigator instructed the subjects in the use of
these physical activity records and inspected the completed
Methods forms.
Study population Measurement of diet
The data for this study were collected from July 2006 to Usual dietary intake was assessed with the use of
July 2007 in two Arab villages, Midoun and Houmet Essouk, a 168-item semiquantitative food-frequency questionnaire.
and two Berber villages, Guellala and Sedwikch. These All questionnaires were administered by trained dieticians.
ethnic groups live far from each other on Jerba Island. The The food-frequency questionnaire consisted of a list of foods
sample for our analyses included 47 Berber and 47 Arab with a standard serving size. Participants were asked to report
subjects aged 32–64 y. The recruitment of the population their frequency of consumption of each food item during the
was done among the patients who came to medical centers previous month on a daily (e.g., bread) or weekly (e.g., rice or
on the island and those who accompanied them. Each partic- meat or fish) basis. Portion sizes of consumed foods were con-
ipant agreed to take part in our investigation by signing an verted from household measurements to grams. Each food
informed consent before entering the study; the appropriate and beverage was then coded according to the prescribed pro-
local ethical committee approved this research and it was tocol and analyzed for content of energy and other nutrients
performed in accordance with the Declaration of Helsinki. using Bilnut software (SCDA Nutrisoft, Cerelles, France),
Each individual was examined by medical personnel and which was designed for Tunisian foods. We used a 24-h
surveyed with a comprehensive questionnaire that they dietary recall to obtain detailed information about foods
completed at home (including information on their pheno- including brand names, preparation methods, and ingredients
typic measurements about age, smoking habits, medical used in the preparation. Three-dimensional food models,
history and current use of medications, lifestyle, diet, and measurement aids, and food-specific units were used to
socioeconomic status). estimate amounts consumed. We determined total calories,
The inclusion criteria in this study were to be an adult and total and percentage of calories from protein, carbohydrate
a life-long full-time resident of the island who never left or and fat, and the nutrients cholesterol, fiber, folates, vitamin
lived elsewhere; the spouse had to originate from the island C, vitamin E, calcium, zinc, iron, and magnesium.
and have the same ethnicity. These criteria seemed important We calculated the nutrient density (amount of nutrient per
to us to avoid any modification of the food culture. 100 kcal) for fiber, sucrose, and iron.
Because it is recognized that social inequalities play an An appropriate intake of energy with energy distributions
important role in health and disease outcomes in many of carbohydrate, protein, and fat was determined within the
regions of the world [8–10], we recruited in the study only recommended nutrient intakes (RNIs) established by the
subjects of moderate socioeconomic status. Participants U.S. National Academy of Sciences. No subject had lactose
were also invited to provide information on their education intolerance during his/her lifetime.
level (as number of years spent in the education system).
Anthropometric determination
Physical activity assessment
Measurements of height by using a stadiometer and
For physical activity assessment, subjects completed weight on a digital scale while wearing lightweight clothing
a physical activity–recall questionnaire. They were asked to without shoes was used to calculate BMI; BMI was defined
record whether they were sleeping, sitting, standing, or as weight in kilograms divided by height squared in meters.
watching television during each hour of the day. Subjects Non-obese subjects had a BMI <25 kg/m2, overweight
noted the time of day they started each new activity, body po- subjects had a BMI 25–30 kg/m2, and obese subjects had
sition during the activity (reclining, sitting, standing, or walk- a BMI >30 kg/m2. Obesity was defined as a BMI
ing), and effort (light, moderate, or vigorous). They were !30 kg/m2 according to recommendations of the World
asked about leisure and occupational physical activities per- Health Organization [13].Waist and hip circumferences
formed during the previous week. For the previous week, the were measured at the level of the umbilicus and the widest
average number of hours per week of each activity was cal- area around the buttocks, respectively, and the waist-to-hip
culated. The total hours of each activity was multiplied by ratio was also calculated [14].
the estimated metabolic cost of the activity and then the phys- To avoid the influence of height, the ratio of waist circum-
ical activity level (PAL) was computed as the total energy ference (WC) to body height (BH) was computed. The WC/
expended over 24 h divided by 24 [11]. Gender, age, height, BH ratio has been reported to have closer values between
and body weight are the main determinants of basal meta- men and women than BMI or WC; therefore, the same

3. T. Baroudi et al. / Nutrition 26 (2010) 75–81 77
boundary value may be applied to men and women [15]. a significantly lower BMI than Berbers. Arab men showed no
Meanwhile, a WC/BH ratio equal to 0.5 may be a simple difference in BMI from Berbers. BMI did not show a signifi-
and effective index to identify not only almost overweight cant difference between ethnic groups (P ¼ 0.66). WC was
people but also normal-weight people with higher metabolic larger in women than in men and there was no difference
risks [15,16]. between ethnic groups. Berber women showed a significantly
To avoid data variability from measurement techniques, shorter height than Arab women (P < 0.001). WC/BH ratio
all measurements were undertaken by the same individual. measurements were significantly different between sexes
(P < 0.001) and ethnicities (P < 0.01).
Statistical analysis
Physical activity
Analysis was performed with SPSS 10.0 (SPSS Inc.,
Chicago, IL, USA). Subgroups were analyzed by gender, The evaluation of physical activity of the two groups did
age, and ethnic group. All results were presented as relative not show a significant difference according to ethnicity. The
and absolute data means. Dietary intakes were calculated only difference was found according to sex; the men were
from the food-frequency questionnaires using Bilnut 4.0, more active than the women (P < 0.00). The men were on
which was adapted for Tunisian foods. Group means were their feet for most of the day as part of their job but did not
compared by least significant difference at P < 0.05. take regular physical exercise. The two ethnic groups can
be considered as sedentary because the PAL was <1.70 [11].
Results
Energy intakes
Anthropometric characteristics of studied samples
Table 2 lists mean 6 SD daily dietary intake by ethnic
Table 1 lists means 6 standard deviations (SDs) for age and group and by sex. Mean 6 SD energy intakes in Arab men
anthropometric measurements (BMI, WC, and WC/BH) for were 1846 6 262 kcal. Energy intakes in Berber men were
men and women by ethnic group. The studied population con- 2039 6 316 kcal. Energy intakes were 1833 6 254 kcal in
sisted of 38 men (22 Arabs and 16 Berbers) and 56 women (25 Arab women and 1858 6 239 kcal in Berber women.
Arabs and 31 Berbers). Among the Arabs 12 men (54.5%) were Mean 6 SD energy intakes were 1839 6 256 kcal in Arab
overweight and four (18.2%) were obese. Among the Berbers subjects and 1927 6 283 kcal in the Berber population.
eight men (50%) were overweight and two (12.5%) were There were no significant differences in energy intake in
obese. Among the Arabs nine women (36%) were overweight men and women by ethnicity.
and eight (32%) were obese. Among the Berbers five women We compared energy intake with calculated expenditure
(16%) were overweight and 24 (77%) were obese. energy (PAL multiplied by BMR produced total energy
The prevalence of overweight and obesity was not expenditure). Over the entire studied population, men’s needs
significantly greater for Arab men compared with Berber were covered (81 6 16) was significantly less than those of
men. However, the prevalence of obesity was significantly women (102 6 16, P < 0.001). Among men, 14 Arabs and
greater for Berber women (P < 0.001). Arab women showed 5 Berbers covered <80% of their needs. Among women,
Table 1
Anthropometric characteristics in ethnic groups*
Arab group Berber group
Men Women Total Men Women Total
(n ¼ 22) (n ¼ 25) (n ¼ 47) (n ¼ 16) (n ¼ 31) (n ¼ 47)
Age (y) 49 6 7 50 6 10 50 6 9 54 6 10 49 6 11 50 6 11
Education levely (%)
Illiterate 4.5 40 23.4 31.3 74.2 59.6
Low 36.4 32 34.0 37.5 25.8 29.8
Moderate 31.8 16 23.1 25 0 8.5
High 27.3 12 19.1 6.3 0 2.1
Height (m) 1.68 6 0.08 1.57 6 0.06 1.62 6 0.09 1.66 6 0.09 1.52 6 0.08 1.57 6 0.1
Weight (kg) 78 6 13 71 6 15 74 6 14 68 6 12 78 6 14 74 6 14
BMI (kg/m2)z 27 6 5 29 6 6 28 6 6 25 6 5 33 6 5 90 6 6
WC (m)z 83.1 6 9.7 85.9 6 11.3 84.6 6 10.5 78.1 6 8.9 94.4 6 8.8 88.9 6 11.7
WC/BHyz 0.50 6 0.07 0.55 6 0.08 0.52 6 0.08 0.47 6 0.07 0.62 6 0.06 0.57 6 0.09
BH, body height; BMI, body mass index; WC, waist circumference
* Values are means 6 SDs or percentages of subjects.
y
P < 0.001 between Arab and Berber groups.
z
P < 0.001 between men and women.

4. 78 T. Baroudi et al. / Nutrition 26 (2010) 75–81
Table 2
Daily dietary intakes by ethnic group*
Arab group Berber group
Men Women Total Men Women Total
(n ¼ 22) (n ¼ 25) (n ¼ 47) (n ¼ 16) (n ¼ 31) (n ¼ 47)
Energy expenditure (cal/day) 2293 6 230 1825 6 154 2044 6 304 2404 6 574 1832 6 188 2027 6 454
Total energy intake (cal/day) 1966 6 185 1922 6 215 1943 6 201 2069 6 232 1951 6 152 1991 6 189
Average of covertures (%) 86.1 6 6.5 106.1 6 15.4 96.7 6 15.6 90.2 6 23.1 107.4 6 11.7 101 6 18.2
Protein (g/day) 56 6 10 55 6 11 55 6 11 65 6 14x 57 6 11 60 6 13
Protein (% TEI) 12.1 6 1.5 12.1 6 1.8 12.1 6 1.6 12.8 6 1.7 12.3 6 1.5 12.5 6 1.5
AP/VP 0.81 6 0.30 0.80 6 0.33 0.81 6 0.32y 0.71 6 0.44 0.65 6 0.27 0.67 6 0.30
Fat (g/day) 68 6 18 70 6 14 69 6 16y 83 6 19x 70 6 14 78 6 17
Fat (% TEI) 33.5 6 6.0 34.5 6 3.8 34.0 6 4.9y 36.7 6 4.7 36.4 6 4.4 36.5 6 4.5
SFA (g/day) 17.4 6 5.8 18.2 6 4.9 17.7 6 5.3 18.8 6 4.8 17.9 6 5.6 18.2 6 5.3
SFA (% TEI) 7.8 6 2.3 8.5 6 2.0 8.2 6 2.2 8.2 6 1.5 8.2 6 2.3 8.2 6 2.0
MUFA (g/day) 43.6 6 12.1 43.3 6 9.7 43.4 6 10.7y 53.5 6 13.4x 48.4 6 10.5 50.1 6 11.7
MUFA (% TEI) 19.8 6 4.7 20.1 6 3.1 20.0 6 3.9y 23 6 3.5x 22.2 6 3.7z 22.5 6 3.6
PUFA (g/day) 7.5 6 1.9 8.2 6 2.8 7.9 6 2.4y 10.7 6 5.6x 9.2 6 3.1 9.7 6 4.1
PUFA (% TEI) 3.4 6 0.8 3.8 6 1.1 3.6 6 1.0y 4.1 6 1.1 4.2 6 1.4 4.2 6 1.3
PUFA/SFA (g/day) 0.47 6 0.13 0.46 6 0.13 0.47 6 0.13y 0.51 6 0.14 0.55 6 0.20z 0.54 6 0.18
Cholesterol (mg/day) 127.4 6 100.1 101.4 6 76.1 113.6 6 88.1 94.3 6 84.1 109.8 6 98.9 104.6 6 93.5
Carbohydrates (g/day) 251 6 43 244 6 44 248 6 42 253 6 35 236 6 28 242 6 31
Carbohydrates (% TEI) 54.9 6 6.7 53.6 6 4.1 54.2 6 5.5y 50.5 6 4.9x 51.4 6 4.9 51.1 6 4.9
Sucrose (g/day) 34 6 22 33 6 22 33 6 22y 18 6 12x 25 6 23 23 6 20
Sucrose (% TEI) 764 7 6 4z 7 6 4y 4 6 2x 565 564
Sucrose density (g/day) 1.7 6 1.1 1.7 6 1.0 1.7 6 1.1y 0.9 6 0.6x 1.3 6 1.2 1.2 6 1.1
Fiber (g/day) 25 6 5 24 6 7 24 6 6 29 6 7x 25 6 5 27 6 7
Fiber density (g/day) 1.26 6 0.26 1.25 6 0.27 1.25 6 0.26 1.41 6 0.28 1.28 6 0.30 1.30 6 0.30
AP/VP, animal protein/vegetal protein ratio; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA, saturated fatty acid; TEI, total energy
intake
* Values are means 6 SDs.
y
P < 0.05 between Arab and Berber groups.
z
P < 0.05 between Arab and Berber women.
x
P < 0.05 between Arab and Berber men.
only one did not have 80% coverage, whereas seven women, Ethnicity comparisons indicated a higher fat intake by
four Arabs and three Berber, exceed the threshold of 120%. Berber subjects than by Arab subjects (P < 0.001). Fat intake
also was categorized as amounts of saturated fatty acid
Macronutrient intakes (SFA), monounsaturated fatty acid (MUFA), and polyunsat-
urated fatty acid (PUFA). The percentage of dietary fat from
Average protein intakes were, 56, 65, 55, and 57 g for SFA was similar across groups. The Arab group had the
Arab men, Berber men, Arab women, and Berber women, lowest percentage of MUFA (P < 0.0003) and PUFA
respectively (Table 3). There was no significant difference (P < 0.008) in men. The ratio of PUFA to SFA was signifi-
by ethnicity in women but the difference was significant cantly greater in the Berber group than in the Arab group
(P < 0.02) between Arab and Berber men. Nevertheless, (P < 0.02). There was an inverse relation between the
the ratio of animal protein to vegetal protein was significantly percentage of fatty acid intake and the percentage of carbohy-
greater in the Arab group than in the Berber group (P < 0.03). drates (r ¼ À0.54, À0.80, and À0.47 for SFA, MUFA, and
Average carbohydrate consumption showed no statistical PUFA, respectively, P < 0.01). Intakes of PUFA and
significance across groups when compared by ethnicity. How- MUFA decreased when the consumption of sucrose
ever, the Arab group consumed more sucrose than the Berber increased (r ¼ 0.31 and À0.52, P < 0.001, data are not
group (P < 0.02) and the Berber group consumed more fiber shown). There was no statistical difference for cholesterol
than the Arab group (P < 0.02). There was a positive correla- intake between the two ethnic groups.
tion between calorie and fiber intakes (r ¼ 0.55, P < 0.001)
and a negative correlation between fiber and sucrose intakes Mean micronutrient intakes
(r ¼ À0.33, P < 0.001). To avoid data variability from energy
intake, we calculated the nutrient densities for fiber and The minimum mean calcium requirement of adults is the
sucrose and the negative correlation between fiber and sucrose mean intake at which ingestion and excretion are equal;
remained (r ¼ À0.29, P < 0.005; Fig. 1). this occurs at an intake of about 520 mg/d [17].

5. T. Baroudi et al. / Nutrition 26 (2010) 75–81 79
Table 3
Daily micronutrient intakes by ethnic groups*
Arab group Berber group
Men Women Total Men Women Total
(n ¼ 22) (n ¼ 25) (n ¼ 47) (n ¼ 16) (n ¼ 31) (n ¼ 47)
Calcium (mg) 513.2 6 235.2 455.4 6 144.7z 482.4 6 192.4y 395.4 6 190.8 345.2 6 117.7 362.3 6 146.6
Zinc (mg) 6.7 6 1.4 6.3 6 1.5 6.5 6 1.5 6.6 6 2.4 5.9 6 1.5 6.1 6 1.9
Magnesium (mg) 266.2 6 87.3 272.1 6 119.3 269.3 6 104.5 312.7 6 64.9 261.1 6 66.1 278.6 6 69.5
Phosphorus (mg) 954.9 6 196.2 895.3 6 185.6 923.2 6 190.9 1054.5 6 222.0 905.3 6 165.6 956.1 6 197.6
Iron (mg) 9.9 6 2.5 9.7 6 3.1 9.8 6 2.8y 13.0 6 3.1x 11.1 6 2.9 11.7 6 3.1
Iron density (mg) 0.51 6 0.13 0.50 6 0.13 0.50 6 0.13y 0.62 6 0.10x 0.57 6 0.14 0.59 6 0.13
Vitamin E (mg) 7.9 6 2.5 7.8 6 2.9 7.9 6 2.7y 10.2 6 2.5x 8.8 6 2.2 9.3 6 2.4
Vitamin E/PUFA (mg) 1.07 6 0.21 0.98 6 0.25 1.02 6 0.23 1.04 6 0.24 1.00 6 0.23 1.02 6 0.23
Vitamin C (mg) 91.5 6 25.1 92.6 6 39.6 92.1 6 33.2 98.2 6 34.8 81.4 6 28.9 87.1 6 31.7
Folates (mg) 185.8 6 42.9 181.7 6 47.9 183.6 6 45.0 185.7 6 67.2 162.7 6 40.2 170.6 6 51.5
PUFA, polyunsaturated fatty acid
* Values are means 6 SDs.
y
P < 0.05 between Arab and Berber groups.
z
P < 0.05 between Arab and Berber women.
x
P < 0.05 between Arab and Berber men.
Mean calcium intakes were below the RNI in Berber men the RNI [20] at 59.6% in the Berber group and 55.3% in
and were 76% and 66% of the minimal requirement in Berber the Arab group.
women. In contrast, mean zinc intakes in both sexes and Vitamin E intake was greater in Berbers than in Arabs
ethnic groups were <63.9% of the RNI (59.6% and 68.1% (P < 0.01), and this was related to consumption of olive oil
in the Arab and Berber groups, respectively) [18]. Overall, and nuts, which were more important in Berber cuisine.
dietary iron intakes were <75% of the RNI [19] (60.6% for The vitamin E/PUFA ratio was approximately 1.02, which
the total population, 72.3% in the Arab group, and 48.9% is well above the 0.4 ratio that would be considered adequate
in the Berber group). Iron density positively correlated to [21]. The olive oil was not rich in PUFAs and rich in vitamin
fiber density (r ¼ 0.8, P < 0.001) and negatively correlated E, explaining the high vitamin E/PUFA ratio [22].
to the animal protein/vegetal protein ratio (Fig. 2). This result Although folates are found in a wide variety of foods, they
can be explained by the origin of iron, which was principally were present in a relatively low density except in the liver
from vegetal food and particularly from seeds, grain, and [23].
nuts. These vegetal products were rich in magnesium; hence, The diet contained much wheat and a small amount of
the mean intake of magnesium was 278 mg/d and exceeded fresh green vegetables and animal products; this could
explain the low intake rate for folates (178 mg/d). In addition,
6,00 R-Deux = 0.08
2,5
2,0
4,00
sucrosed
1,5
2,00 1,0
,5 AP/VP ratio
iron density
0,00 fibre density
1,00 1,50 2,00 0,0 iron density
fiberden ,3 ,4 ,5 ,6 ,7 ,8
Fig. 1. Relation between nutrient density for fibers and sucrose intake. Fig. 2. Relation among iron density, fiber density, and AP/VP. AP/VP, ratio
fiberden, fiber density; R-Deux, R2; sucrosed, sucrose density. of animal protein to vegetal protein.

6. 80 T. Baroudi et al. / Nutrition 26 (2010) 75–81
folate loss during harvesting, storage, distribution, and Berbers. There were no differences between genders but in
cooking can be considerable [24]. the Arab group these collations and nibblings provided
Dietary intakes of phosphorus and vitamin C met or more sucrose (sweet food) than in the Berber group in which
exceeded the RNI [24,25] at >80% for both sexes and ethnic these extra meals provided more fat (nuts, bread, and olive
groups. oil; Table 4).
Type of meals Discussion
Every day all subjects took three meals (breakfast, lunch, To our knowledge, this is the first study that compared
and dinner). Everyday 12 Arabs and 10 Berbers took one or anthropometric measurements and dietary intakes of Arabs
two collations. Nibbling was declared by 14 Arabs and 6 and Berbers. The inclusion criteria of this study decreased
Table 4
Macronutrient intake according to meal pattern*
Arabs Berbers
Men Women Total Men Women Total
Breakfast
Subjects 22 25 47 16 31 47
Energy (kcal) 23.8 6 4.7 25.1 6 5.5 24.5 6 5.1y 19.2 6 4.9x 21.4 6 6.8z 20.7 6 6.2
Protein (g) 2.7 6 1.3 2.7 6 0.9 2.7 6 1.1 2.3 6 0.5 2.5 6 0.5 2.5 6 0.5
Fat (g) 6.0 6 3.7 6.4 6 2.7 6.2 6 3.2 5.0 6 3.7 5.8 6 2.4 5.6 6 2.9
Carbohydrates (g) 14.0 6 4.8 16.0 6 4.2 15.1 6 4.6y 11.9 6 2.3 13.4 6 4.0z 12.9 6 3.6
Morning collation
Subjects 8 3 11 3 7 10
Energy (kcal) 1.9 6 3.8 0.7 6 2.6 1.3 6 3.3 0.1 6 0.4 2.0 6 5.1 1.4 6 4.2
Protein (g) 0.2 6 0.4 0.1 6 0.3 0.1 6 0.3 0.03 6 0.1 0.2 6 0.6 0.2 6 0.5
Fat (g) 0.8 6 1.7 0.3 6 0.9 0.5 6 1.4 0.1 6 0.3 0.6 6 1.5 0.4 6 1.3
Carbohydrates (g) 0.9 6 1.9 0.4 6 1.5 0.6 6 1.7 0.03 6 0.1 1.2 6 3.1 0.8 6 2.6
Lunch
Subjects 22 25 47 16 31 47
Energy (kcal) 35.6 6 11.2 36.6 6 10.6 36.1 6 10.8 40.4 6 10.6 39.3 6 10.3 39.7 6 10.3
Protein (g) 4.3 6 1.6 4.6 6 1.9 4.4 6 1.7 5.3 6 2.0 5.1 6 1.8 5.2 6 1.9
Fat (g) 13.2 6 5.2 13.9 6 4.7 13.6 6 4.9y 16.4 6 4.7 15.4 6 4.2 15.7 6 4.4
Carbohydrates (g) 18.1 6 5.4 18.1 6 5.4 18.1 6 5.3 18.7 6 5.5 18.8 6 4.9 18.7 6 5.1
Afternoon collation
Subjects 7 5 12 3 4 7
Energy (kcal) 1.9 6 4.5 1.4 6 4.2 1.7 6 4.3 1.4 6 3.9 1.1 6 3.9 1.2 6 3.9
Protein (g) 0.3 6 0.6 0.1 6 0.4 0.2 6 0.5 0.2 6 0.5 0.2 6 0.5 0.2 6 0.5
Fat (g) 0.4 6 1.2 0.5 6 2.0 0.5 6 1.7 0.4 6 1.1 0.3 6 1.1 0.3 6 1.0
Carbohydrates (g) 1.2 6 2.8 0.8 6 2.2 1.0 6 2.5 0.9 6 2.3 0.6 6 2.3 0.7 6 2.3
Dinner
Subjects 22 25 47 16 31 47
Energy (kcal) 28.1 6 8.2 27.7 6 8.4 27.9 6 8.2y 35.9 6 10.4x 31.0 6 7.6z 32.7 6 8.8
Protein (g) 3.6 6 1.5 3.7 6 1.6 3.6 6 1.5 4.6 6 1.9 3.6 6 1.4 4.0 6 1.7
Fat (g) 9.7 6 4.0 9.9 6 4. 9.8 6 4.4y 13.6 6 4.4x 12.6 6 3.9 13.0 6 4.0
Carbohydrates (g) 14.9 6 4.4 14.2 6 4.3 14.5 6 4.3 17.7 6 6.0 14.7 6 4.2 15.7 6 5.0
Evening collation
Subjects 0 0 0 1 0 1
Energy (kcal) 0.03 0.03
Protein (g) 0.01 0.01
Fat (g) 0.01 0.01
Carbohydrates (g) 0.01 0.01
Nibbling
Subjects 6 8 14 3 3 6
Energy (kcal) 1.9 6 3.6 3.6 6 7.8 2.8 6 6.2y 0.6 6 1.3 0.4 6 1.7 0.5 6 1.5
Protein (g) 0.1 6 0.3 0.2 6 0.5 0.2 6 0.4y 0.03 6 0.1 0.04 6 0.2 0.03 6 0.2
Fat (g) 0.9 6 2.0 1.6 6 3.4 1.3 6 2.8y 0.3 6 0.6 0.3 6 1.2 0.3 6 1.0
Carbohydrates (g) 0.9 6 1.8 1.7 6 4.0 1.3 6 3.1y 0.3 6 0.8 0.1 6 0.4 0.2 6 0.5
* Values are numbers of subjects or means 6 SDs.
y
P < 0.05 between Arab and Berber groups.
z
P < 0.05 between Arab and Berber women.
x
P < 0.05 between Arab and Berber men.

7. T. Baroudi et al. / Nutrition 26 (2010) 75–81 81
the influence from socioeconomic level and acculturation of National Health and Nutrition Examination Survey. Am J Clin Nutr
food choices. These conditions were often considered factors 2003;78:1120–7.
[9] Wang MC, Dixon LB. Socioeconomic influences on bone health in
of variability.
postmenopausal women: findings from NHANES III, 1988–1994.
Low education level is a factor that exacerbates the prob- Osteoporos Int 2006;17:91–8.
lem of obesity in Tunisia. Women without education do not [10] Cohen AD, Gefen K, Ozer A, Bagola N, Milrad V, Cohen L, et al.
recognize the risk factor and health consequences associated Diabetes control in the Bedouin population in southern Israel. Med
with overweight and obesity. It is more worrisome that these Sci Monit 2005;11:CR376–80.
[11] Food and Agriculture Organisation (FAO). Human energy require-
women consider fatness and obesity to be desirable, because
ments: energy requirements of adults. Report of a joint FAO/WHO/
these traits are associated with higher social status, fertility, UNU expert consultation. Food and nutrition technical report series.
and prosperity [25]. Rome: Food Agriculture Organisation; 2001, p. 35–52.
In our study low education level was markedly important [12] Black AE, Coward WA, Cole TJ, Prentice AM. Human energy
among women but it did not influence BMI or WC. In this expenditure in affluent societies: an analysis of 574 doubly-labelled
water measurements. Eur J Clin Nutr 1996;5:72–92.
study we investigated the effect of ethnicity on dietary
[13] World Health Organization. Use and interpretation of anthropometric
intakes in two ethnic groups, Arabs and Berbers. The most indicators of nutritional status. Report of WHO working group. World
striking observations were seen for dietary intakes of total Health Organ Tech Rep Ser 1986;64:929–94.
fat, MUFA, sucrose, calcium, zinc, folates, and vitamin E. [14] Wang J, Thornton JC, Bari S, Williamson B, Gallagher D, Heymsfield SB,
Dietary intakes of calcium were influenced primarily by et al. Comparisons of waist circumferences measured at 4 sites. Am J Clin
Nutr 2003;77:379–84.
ethnicity and to some extent by sex and education level. Ber-
[15] Hsieh SD, Muto T. The superiority of waist-to-height ratio as an
bers were found to some consume less calcium than Arabs. anthropometric index to evaluate clustering of coronary risk factors
Among Berbers and Arabs, dietary intakes of calcium in- among non-obese men and women. Prev Med 2005;40:216–20.
creased with increasing education level. The nutrient density [16] Hsieh SD, Yoshinaga H, Muto T. Waist-to-height ratio, a simple and
analyses indicated that Arabs were actually consuming more practical index for assessing central fat distribution and metabolic
risk in Japanese men and women. Int J Obes Relat Metab Disord
calcium-rich food than Berbers. The incidence of osteoporo-
2003;27:610–6.
sis in Berbers has not been studied, but the lower calcium in- [17] Food and Nutrition Board. Institute of Medicine, National Academy of
take in this group, the increased adiposity, and decreased Sciences, Report on dietary reference intakes for calcium, phosphorus,
exercise suggest that Berber women may be at increased risk. magnesium and vitamin D. Washington, DC: National Academy Press;
Iron intake was greater in Berbers than in Arabs; however, 1997, p. 7.1–7.30.
[18] World Health Organization/Food and Agriculture Organization of the
the source of iron was mostly vegetal food in the Berber
United Nations. Human vitamin and mineral requirements. Zinc,
group versus animal food in the Arab group. Consequently, Report of a joint FAO/WHO expert consultation, Bangkok, Thailand.
the bioavailability of iron may be lowest in the Berber group. Rome: World Health Organization/Food and Agriculture Organization;
In conclusion, the comparisons of nutrient intakes of 2002, p. 257–70.
Berbers and Arabs clearly indicate a significant effect of [19] World Health Organization/Food and Agriculture Organization of the
United Nations. Human vitamin and mineral requirements. Iron, Report
ethnicity.
of a joint FAO/WHO expert consultation, Bangkok, Thailand. Rome: World
Health Organization/Food and Agriculture Organization; 2002, p. 195–221.
References [20] World Health Organization/Food and Agriculture Organization of the
United Nations. Human vitamin and mineral requirements. Magne-
[1] Chaturvedi N. Ethnicity as an epidemiological determinant—crudely sium, Report of a joint FAO/WHO expert consultation, Bangkok,
racist or crucially important? Int J Epidemiol 2001;30:925–7. Thailand. Rome: World Health Organization/Food and Agriculture
[2] Menotti A, Mulder I, Nissinen A, Feskens E, Giampaoli S, Organization; 2002, p. 223–33.
Tervahauta M, Kromhout D. Cardiovascular risk factors and 10-year [21] World Health Organization/Food and Agriculture Organization of the
all-cause mortality in elderly European male populations. The FINE United Nations. Human vitamin and mineral requirements. Vitamin
study. Eur Heart J 2001;22:573–9. E, Report of a joint FAO/WHO expert consultation, Bangkok,
[3] Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW. Body-mass Thailand. Rome: World Health Organization/Food and Agriculture
index and mortality in a prospective cohort of U.S. adults. N Engl J Med Organization; 2002, p. 121–31.
1999;341:1097–105. [22] Abaza L, Taamalli W, Temine S, Daoud D, Gutierrez F, Zarrouk M.
[4] Steffen BL, Nichaman MZ, Goff DC, Frankowski RF, Hanis CL, Natural antioxidant composition as correlated to stability of some
Ramsey DJ, Labarthe DR. Change in level of physical activity and Tunisian virgin olive oils. Riv Ital Sostanze Grasse 2005;82:12–8.
risk of all-cause mortality or reinfarction: the Corpus Christi Heart [23] World Health Organization/Food and Agriculture Organization of the
Project. Circulation 2000;102:2204–9. United Nations. Human vitamin and mineral requirements. Vitamin
[5] Zhu S, Heymsfield SB, Toyoshima H, Wang Z, Pietrobelli A, Heshka S. E, Report of a joint FAO/WHO expert consultation, Bangkok,
Race-ethnicity–specific waist circumference cutoffs for identifying Thailand. Rome: World Health Organization/Food and Agriculture
cardiovascular disease risk factors. Am J Clin Nutr 2005;81:409–15. Organization; 2002, p. 53–63.
[6] Daida Y, Novotny R, Grove JS, Acharya S, Vogt TM. Ethnicity and [24] World Health Organization/Food and Agriculture Organization of the
nutrition of adolescent girls in Hawaii. J Am Diet Assoc 2006; United Nations. Human Vitamin and Mineral Requirements. Vitamin
106:221–6. E, Report of a joint FAO/WHO expert consultation, Bangkok,
[7] Diaz VA, Mainous AG, Koopman RJ, Carek PJ, Geesey ME. Race and Thailand. Rome: World Health Organization/Food and Agriculture
diet in the overweight: association with cardiovascular risk in Organization; 2002, p. 73–86.
a nationally representative sample. Nutrition 2005;21:718–25. [25] Mokhtar N, Elati J, Chabir R, Bour A, Elkari K, Schlossman NP,
[8] Mazur RE, Marquis GS, Jensen HH. Diet and food insufficiency among et al. Diet culture and obesity in northern Africa. J Nutr 2001;131.
Hispanic youths: acculturation and socioeconomic factors in the Third 887S–92.