Public Health Nutrition: 12(12), 2464–2472
Assessing individual dietary intake from common-plate meals:
a new tool for an enduring practice
Kathleen Abu-Saad1,*, Danit R Shahar1, Heiger Abu-Shareb2, Hillel Vardi1,
Natalya Bilenko1,3and Drora Fraser1
1S. Daniel Abraham International Center for Health and Nutrition, Faculty of Health Sciences, Ben-Gurion
University of the Negev, POB 653, Beer-Sheva 84105, Israel:2Department of Epidemiology and Health Services
Evaluation, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel:3Ministry of
Health, Southern District, Beer-Sheva, Israel
Submitted 4 September 2008: Accepted 8 February 2009: First published online 1 May 2009
Objective: The purposes of the present study were to estimate individual intake
from common-plate meals among Bedouin Arabs using a modified 24h recall
questionnaire, and to evaluate reported energy intake (EI) by comparison with
estimated energy requirement (EER).
Design: Weighed records were used to develop a method of quantifying intake
from common plates. Reported EI and nutrient intakes were obtained from
administration of the modified 24h recall. The relative standard error (RSE) was
used to evaluate the reliability of reported nutrient intakes. The FAO/WHO/
United Nations University and Oxford equations and reported physical activity
levels were used to compute ratios of reported EI to BMR and EER.
Setting: Population centres of traditionally semi-nomadic Bedouin Arabs under-
going sedentarization/urbanization in southern Israel.
Subjects: A convenience sample of 451 adults (aged 19–82 years).
Results: Mean (SE) energy intake was 9648 (276) kJ/d (2306 (66) kcal/d) for
men and 8230 (172) kJ/d (1967 (41) kcal/d) for women, of which carbohydrates
accounted for 63–64%. The nutrient intakes evaluated had RSE ratios of less
than 25%. EI:EER ratios ranged from 0?86 to 0?89, and from 0?87 to 0?93 among
non-dieters who ate the usual amount on the recall day.
Conclusions: The modified 24h recall produced plausible estimates of energy and
nutrient intakes, comparable to those obtained with the 24h recall in other
populations. The modified questionnaire makes an important contribution to
facilitating large-scale nutritional surveillance in the Bedouin population, and may
serve as a model for modifying dietary instruments to quantify individual intake in
other populations that practise common-plate eating.
The Negev Bedouin Arabs in southern Israel, a tradi-
tionally semi-nomadic population, historically ate their
meals from common plates. Despite the processes of
sedentarization, modernization and urbanization that
have occurred in this population over the past 50 years,
common-plate eating remains an enduring practice.
Dietary surveillance among Negev Bedouin Arabs has
been hampered by the lack of an appropriate dietary
assessment instrument for quantifying intake at the indi-
vidual level(1), thus current comprehensive nutrient
intake data for this population are lacking.
Common-plate eating occurs in many parts of the world,
but little work has been done to develop dietary assessment
methods for quantitatively measuring individual food intake
that are practical for use in epidemiological studies(2–4).
The 24h recall questionnaire is the main tool for dietary
surveillance and monitoring dietary intake over time(5,6). It is
based on identifying foods eaten and their portion sizes
served and eaten individually in a 24h period preceding the
interview. The traditional 24h recall is not designed to
measure food intake in societies in which common-plate
eating is practised.
During recent decades, chronic disease rates among
the Bedouin have been on the increase(7–10), as they have
in many of the populations that practise common-plate
eating(11,12). Thus, the development of appropriate diet-
ary assessment instruments that are practical for use in
large-scale dietary surveillance has become increasingly
important. To the best of our knowledge, no dietary
assessment tool has ever been adapted and/or used for
*Corresponding author: Email firstname.lastname@example.org
r The Authors 2009
measuring individual food intake in this population. In
the present paper, we describe the modification of the
24h recall for quantifying individual intake from com-
mon-plate meals, present the results from the adminis-
tration of the modified 24h recall among a sample of
Bedouin adults, and compare reported energy intakes
(EI) with estimated energy requirements (EER) derived
from the calculation of BMR and reported physical activity
Population and methods
The population and food culture
The history and characteristics of the Bedouin Arabs
living in the Negev desert in southern Israel have been
described elsewhere(9,13,14). When the Bedouin Nutri-
tional Study (BNS) was conducted (2001–3), the Negev
Bedouin population numbered 138000 of whom 60%
lived in government-planned towns. The remaining 40%
lived more traditionally in unrecognized villages not
connected to local and national planning and commu-
nications infrastructures. The Negev Bedouin have the
lowest socio-economic level of any population group in
Israel, reflected in educational, employment and income
Traditionally, Bedouin food culture was well adapted
to a semi-nomadic life which involved seasonal migration
and living in tents with limited household utensils that
were easy to transport. The most common traditional
meal patterns consisted of either one main dish served on
a large platter or of one or more dishes of cooked foods
or salads in common service dishes. Main dish meals were
usually eaten with the hand, and the rules of hygiene
required that a person eat only from the part of the plate
directly in front of him/her. In the second type of meal,
bite-sized portions of flat bread were used as utensils for
dipping in or scooping up food from common service
dishes. The bread was consumed along with the food it
held. Since many of the foods eaten from common plates
with bread were sauce-like or liquid, it was not possible
for a person eating with others to visualize how much he
or she consumed individually.
Due to the absence of 40% of the population from
census data, the use of standard methods of drawing
a random sample (e.g. street addresses from voter
registration lists, telephone directories, etc.) would have
systematically excluded all those who live in settings
that lack official mapping/street addresses and land-line
telephones. To obtain a sample that included Bedouin
from both recognized and unrecognized localities, we
drew the study population from healthy Bedouin adults
visiting patients at Soroka University Medical Center, the
only regional hospital serving the southern (Negev)
region of Israel, and adults (primarily women) attending
Maternal and Child Health Care clinics in Bedouin towns.
The hospital and clinics serve the Bedouin from both the
government-planned towns and the surrounding unrec-
ognized villages, and thus provided us with access to a
broad geographical cross-section of the population.
We enrolled adults aged 19 years and above who
consented to provide dietary intake information. Data
were collected on all days of the week, including week-
Updating the food and nutrient database
The S. Daniel Abraham International Center for Health and
Nutrition food composition tables, which are currently
based on the US Department of Agriculture (USDA) Nutri-
ent Database for Standard Reference Release 19, were
modified to include Israeli-manufactured foods and com-
mon recipes(17), but did not include Bedouin foods. We
recorded recipes in homes of Bedouin women who
cooked a dish, while the research team member weighed
and measured the ingredients using digital scales and
standardized measuring tools. Cooking times were recor-
ded, and foods were weighed before and after cooking. We
thus added 145 Bedouin recipes/foods/beverages to the
database prior to beginning and throughout the dietary
assessment process. Recipes were calculated using another
computer system we developed based on the concept of
the American Food Information Analysis System (FIAS)
program(18). Trained nutritional data entry coders did the
Quantifying food intake
Field tests were conducted in eleven households in three
different Bedouin communities to determine what quanti-
fiable information Bedouin could give us on their food
intake at the individual level. Foods and beverages eaten in
individual servings (e.g. pieces of fruits, sandwiches, meat
not in stews) were easily quantified using the USDA
method(19). Respondents were also able to estimate the
quantity of bread they ate, since generally people take
portions of bread (e.g. a half or whole pita), which they
consume individually. Since bread is used as the utensil for
eating from common plates, it also serves as the vehicle
determining the amount of the common-plate foods con-
sumed, and could thus be used as the means for quanti-
fying the intake of these foods. To determine the ‘carrying
capacity’ of bread for common-plate foods of different
consistencies, the BNS Bedouin study staff were trained to
conduct weight tests under natural eating conditions
among a pool of volunteers recruited from their nuclear
and extended families by weighing the bread and com-
mon-plate food before and after a meal in which a single
common-plate dish was served. Weighings were conducted
among adults in five field locations in both urban and rural
settings. Multiple weighings of twenty-eight common-plate
foods of differing consistencies were carried out in a total of
seventeen households. Analysis of these data showed that
most common-plate foods fell into two main categories: A,
Assessing individual intake from common-plate meals2465
solid/semi-solid (e.g. egg dishes, humus salad, semi-solid
dips, sauces containing chunks of meat and/or vegetables,
thick cracked-wheat or lentil sauces); and B, liquid (e.g.
buttermilk sauces, thin vegetable sauces made with
potherbs not containing solid chunks). Aggregated avera-
ges of the weight data were used to establish the ‘carrying
capacity’ of bread, expressed as food:bread ratios, for these
two categories of foods (category A, 1?3g food:1?0g bread;
category B, 1?0g food:1?0g bread). During the pilot and
throughout the BNS data collection, the Bedouin BNS study
coordinator and a registered dietitian reviewed the results
produced by the food:bread ratios. For foods with differ-
ent/unique consistencies (e.g. dried thyme mixed with
olive oil) or for which the standard food:bread ratios
produced implausible results, additional weighings were
conducted under natural eating conditions and specific
food:bread ratios set.
Modification of the US Department of Agriculture
24h recall questionnaire
The USDA 24h recall questionnaire was modified to
allow for recording of the three eating practices of the
population: (i) eating an item as an individual portion;
(ii) eating from a common-plate with bread; or (iii) eating
directly from a large platter. We used the multi-pass
method(19)to administer the modified USDA 24h recall
questionnaire. At the appropriate stage, the interviewers
determined which foods were eaten together in one sit-
ting and whether the foods had been consumed with
bread as the utensil, individually or with others eating
from the same dishes. All items consumed with bread
from common plates in a single meal were enclosed by
brackets. The total amount of bread the respondent
consumed during that meal was obtained.
To estimate the portion size for each of the foods eaten
with bread as the utensil in multi-dish meals eaten with
others, the interviewer asked if the respondent ate a
relatively smaller, medium or larger portion (or equal
portions) from each dish. The total amount of bread was
then divided by the relative portions given, using 1 for
small, 2 for medium and 3 for large. The quantity of food
eaten from each dish was calculated on the basis of the
food:bread ratio for the consistency category of the food
(as described above), and multiplied by the amount of
bread (in grams) eaten with each dish.
For meals eaten from a single large common platter,
pictures with different relative portions removed from the
platter were used as reporting aids. Portion sizes for items
consumed individually, rather than from common plates,
were reported in the standard way.
The modified 24h recall questionnaire was piloted
among forty Bedouin adults. The results confirmed that
individuals were able to estimate the amount of bread
they ate at a meal and were able to estimate the relative
proportion they consumed from each dish of a common-
plate meal with bread.
Administration of the modified 24h recall
Trained interviewers conducted the interviews in Arabic
using a USDA food book and a food models/portion-size
booklet modified to include common Israeli foods and
Bedouin foods, utensils and portion sizes. Upon com-
pletion of the 24h recall, a number of questions on
general health status were asked, a physical activity
questionnaire was administered, and the respondents’
weight (in light street clothing) and height were measured
using a portable digital scale and collapsible measuring
stick. Two additional 24h recalls were completed on non-
consecutive days among a subsample of forty respon-
dents who agreed to be interviewed at home.
Quality control was applied at four stages. First, each
interview was checked for missing data within 1–3d of the
interview. Second, after data entry, the BNS study coordi-
nator, who had extensive knowledge of Bedouin foods
and the process for quantifying intake from common
plates, edited each questionnaire for accurate data entry
and appropriate application of the common-plate quanti-
fication method. Requests for corrections were then
returned to the coders and re-checked. At the third stage,
registered dietitians edited each questionnaire, returned
corrections to the coders and re-checked corrected data
entries; and, at the final stage, the BNS study coordinator
and registered dietitians made cross-interview checks to
detect unusual nutrient or food model values.
Physical activity levels
We used a physical activity questionnaire based upon a
synthesis of previously validated international ques-
tionnaires(20–22), modified for use in Israel(23). It included
questions about the time spent in and intensity of work,
recreational, leisure-time and household activities, further
modified to capture specific activities associated with
Bedouin lifestyle (e.g. herding, washing clothes by hand).
The definitions of the joint FAO/WHO/United Nations
University (UNU) report on human energy requirements(24)
were used to classify physical activities as sedentary/light
(non-strenuous occupations, no regular leisure-time or life-
style-required physical exercise); active/moderately active
(moderate occupational exertion or moderate/vigorous lei-
sure activity $1h/d); or vigorously active (very strenuous
occupational or leisure activities several hours daily).
Descriptive statistics were used to provide a profile of the
sample’s demographic characteristics, eating patterns and
nutrient intakes. We tested for differences in eating pat-
terns by categorical demographic variables using the x2
statistic and by continuous demographic variables using
Student’s t test. Following the method of the National
Health and Nutrition Examination Survey (NHANES)
1999–2000, we used a relative standard error (RSE; ratio of
2466K Abu-Saad et al.
the standard error of the mean to the mean, multiplied by
100) of greater than 25% as the statistical criterion to define
unreliable nutrient intake estimates(25). To evaluate the
plausibility of the reported EI obtained from the modified
24h recall questionnaire, we calculated the EER using the
FAO/WHO/UNU equations(24), which are based on the
Schofield equations(26), to estimate the BMR, and then
multiplied it by the appropriate PAL factor based on
respondents’ reported levels of physical activity. The PAL
factors were 1?53, 1?76 and 2?25 for sedentary/light activity,
active/moderately active or vigorously active lifestyles,
respectively, as defined by the FAO/WHO/UNU report on
human energy requirements(24). The respondents’ reported
EI was then divided by their EER to obtain the EI:EER ratio.
Owing to concerns that the Schofield equations over-
estimate BMR in non-European populations, the Oxford
equations(27)were developed based upon a database of
measured BMR data that included a better representation of
Asian and other non-European populations. Thus, we also
used the Oxford equations to calculate the BMR and EI:EER
ratios for our sample. Women who indicated that they were
pregnant and breast-feeding on the questionnaire were
excluded from this analysis. Student’s t tests and ANOVA
were used to determine whether or not mean EI:EER
ratios differed by sex, BMI (above or below median BMI),
dieters and non-dieters, and reported eating of ‘usual
amount’ on the day covered by the 24h recall. Among the
forty respondents who completed three repeat 24h recalls,
we computed the within-person CV for selected nutrients to
assess the level of day-to-day variability in nutrient intakes
using Generalized Estimating Equations (GEE) in the
STATA statistical software package version 9?2 (StataCorp
LP, College Station, TX, USA). All other statistical analyses
were conducted using the SPSS statistical software pack-
age version 14?0 (SPSS Inc., Chicago, IL, USA), using
P,0?05 to indicate significance.
The total BNS sample included 519 participants. We
excluded forty-five respondents interviewed during the
Islamic month of Ramadan who were fasting from sunrise
until sunset. Of the remaining participants, 451 (95%)
completed at least one reliable 24h recall and were
included in the analysis.
The demographic characteristics of the sample are
presented in Table 1. The respondents’ age distribution
ranged from 19 to 82 years with a mean of 34 years. Only
3% of the sample had BMI below 19?0kg/m2, while over
40% had BMI above 26?0kg/m2. More than 75% of par-
ticipants reported eating the usual amount on the day of
dietary recall, and approximately 8% reported currently
dieting. The proportion who reported taking a vitamin
supplement was low, and was concentrated among
women who were pregnant or breast-feeding. In the
remainder of the sample (n 286), only 6?3% reported
using vitamin supplements.
With regard to eating patterns, 88% reported eating at
least one common-plate meal, and this proportion was
higher among those living in rural (94%) than in urban
(85%) communities (x257?82, df51, P50?005). Those
who reported eating common-plate meals did not differ
significantly by sex, age or years of education from those
who did not report eating common-plate meals.
Table 2 contains the means, standard errors and medians
for the dietary intakes of selected nutrients for men and
women. All intake estimates have RSE ratios of less than
25%. Carbohydrate intake accounted for ,63% of total
Table 3 presents the mean EI for men and women
(excluding pregnant and breast-feeding women) and EER
calculated using the FAO/Schofield and the Oxford equa-
tions and respondents’ reported PAL. Low proportions
Table 1 Selected characteristics of the Bedouin Nutrition Study participants (n 451)
Demographic characteristicn or Mean% or SD
Amount of food eaten on recall day*
Less than usual
More than usual
Pregnant or breast-feeding*
Taking a vitamin supplement*
Number of eating occasions per questionnaire-
Total number of items per questionnaire/d-
Questionnaires with one or more common-plate eating occasions*
*Data presented are n and %.
-Data presented are mean and SD.
--Data available for 108 men and 119 women.
Assessing individual intake from common-plate meals 2467
reported engaging in physically demanding occupations
(11%) or household chores (e.g. hand-washing clothes,
6%) daily or in walking/exercise/sports for $1h/d (3%);
thus, 84% were classified as having sedentary/light activity,
16% as having moderately active lifestyles and none as
having vigorously active lifestyles. We report both the
EI:BMR and EI:EER ratios to facilitate comparisons with the
results of other studies. The reported EI of the BNS
respondents was closer to the EER calculated using the
Oxford equations than to that using the FAO/Schofield
equations, but with both equations was closer to the EER
among respondents who ate the ‘usual’ amount on the
recall day and were not dieting. The FAO/Schofield and
Oxford EI:EER ratios were significantly higher among those
who reported eating the ‘usual’ amount (0?88 and 0?91,
respectively) than among those who reported eating ‘less
than usual’ (0?70 and 0?72, respectively) on the recall day
(P50.001), and the trend for dieters and non-dieters was
similar (data not shown). The mean EI:EEROXFratios of
those with BMI$26?0kg/m2(EI:EEROXF50?83) also
Table 2 Dietary intake of selected nutrients among Bedouin men and women in the Bedouin Nutrition Study
Men (n 149) Women (n 302)
Total SFA (g/d)
Total MUFA (g/d)
Total PUFA (g/d)
Dietary fibre (g/d)
Vitamin C (mg/d)
Vitamin E (mg/d)
% of total energy
Table 3 Reported energy intake (EI), estimated energy requirement (EER) and ratios of reported EI to estimated BMR and EER among
men and women in the Bedouin Nutrition Study
All (n 149)Usual intake/non-dieters (n 111)All (n 138) Usual intake/non-dieters (n 101)
Total reported EI (kcal)-
FAO estimates and ratios
Oxford estimates and ratios
*Pregnant and breast-feeding women were excluded from this analysis.
-To convert to kJ, multiply kcal by 4?184.
--Calculated using the FAO/WHO/United Nations University equations(24)to obtain BMR, multiplied by physical activity level factor.
yCalculated using the Oxford equations(27)to obtain BMR, multiplied by physical activity level factor.
2468K Abu-Saad et al.
differed significantly from those with BMI,26?0kg/m2
The mean intakes of selected nutrients and the within-
person CV for three repeat 24h recalls administered to a
subsample of forty respondents are presented in Table 4.
The CV for energy and macronutrients ranged from 24?8%
to 49?1%, those for micronutrients ranged from 31?3% to
We present a modified USDA 24h recall questionnaire that
quantifies individual intake from common-plate meals
among the Negev Bedouin Arab population in southern
Israel. Since 88% of the respondents reported eating
common-plate meals on the recall day, we can confirm
that common-plate eating is a widespread and enduring
practice among Negev Bedouin adults and that an appro-
priate dietary assessment instrument for common-plate
eating is needed. In addition, this population has under-
gone a major transition from semi-nomadic pastoralists and
agriculturalists to sedentarized/urbanized wage labourers,
and the transition has been accompanied by dramatically
rising rates of chronic diseases(7–10,13,14). It is reasonable to
assume that life changes, including dietary changes, have
contributed to increases in chronic disease rates. Thus, an
appropriate tool for nutritional surveillance, such as the
24h recall questionnaire, is crucial for assessing nutrient
intake quantity/quality and identifying dietary trends. It
will also enable health-care professionals to evaluate the
efficacy of interventions targeted at disease prevention and
The dietary assessment tools developed in European-
origin cultures are not suitable for measuring individual
dietary intake in contexts in which common-plate eating
is practised. Alternative methodologies have been devel-
oped, ranging from distribution algorithms(3)to observing
subjects while eating(2,4). Generally, these methodologies
are quite costly and thus infeasible for large epidemio-
logical studies or ongoing nutritional surveillance, parti-
cularly in developing countries where common-plate
eating is more likely to be practised.
The modified 24h recall used among Bedouin adults
produced estimates of individual nutrient intakes that
were comparable in terms of plausibility of reported EI
and day-to-day within-person variation to those of the
nutrient intake estimates obtained with the original USDA
24h recall in other populations. The fact that respondents
tend to underestimate their EI with the 24h recall has
been well documented, in both developed(5,28–30)and
developing countries/populations(31–33). We evaluated
the plausibility of the reported EI from the modified 24h
recall by comparing it with EER computed using the FAO/
WHO/UNU(24)and the Oxford(27)BMR equations. Several
of the older equations for calculating BMR (e.g. Harris–
Benedict(34)), including the FAO/WHO/UNU equations
based on Schofield’s BMR data collected in the 1930s(26),
have been shown to overestimate energy requirements
in modern populations, and particularly those of non-
European origin(26,27,35–38). The Oxford equations were
developed more recently using a data set of 10552 BMR
measurements that included a much larger number of
non-European subjects(27). In the present study, the BMR
calculated according to the Oxford equations was closer
to the EI of the BNS respondents than the BMR calculated
Table 4 Mean and within-person CV of three repeat 24h recalls for energy and selected nutrients among a subsample of Bedouin Nutrition
Men (n 15) Women (n 25)
Total SFA (g/d)
Total MUFA (g/d)
Total PUFA (g/d)
Dietary fibre (g/d)
Vitamin A (IU)
Vitamin C (mg/d)
Vitamin E (mg/d)
Assessing individual intake from common-plate meals 2469
according to other equations, as well as more sensitive to
differences in EI:EER ratios by BMI. The EI:EER and
EI:BMR ratios we obtained using both the FAO (Scho-
field) and Oxford BMR equations were well within the
range of EI:EER and EI:BMR ratios reported in the litera-
ture for 24h recall questionnaires(39), particularly when
our analysis was limited to non-dieters who ate the usual
amount (EI:EERFAO50?87, 0?91; and EI:EEROXF50?90,
0?93; men and women, respectively). A review of studies
validating reported energy intake produced a mean
EI:EER of 0?87 in studies comparing the reported EI from
24h recalls with EER measured by doubly labelled
water(40). Among the studies comparing EI from 24h
recalls with estimated BMR, EI:BMR ratios ranged from
1?37 to 1?51 for men and from 1?09 to 1?39 for women.
Comparison of reported EI from one 24h recall in
NHANES III with BMR estimated using Schofield’s equa-
tions resulted in mean EI:BMR ratios of 1?47 and 1?26 for
men and women, respectively(5). In our study, the mean
EI:BMR ratios calculated using the FAO/WHO/UNU
equations based on Schofield’s equations were 1?32 for
both men and women. Similar to our findings, Briefel
et al. reported higher EI:BMR ratios among those who
were not dieting, who reported eating the usual amount
on the recall day, and who were not overweight(5).
Another characteristic of the 24h recall and other 1d
dietary records is that they give poor estimates of an
individual’s habitual diet(41,42), and therefore repeat 24h
recalls from a subsample of respondents have been used
to estimate within-person variance in day-to-day dietary
intake(43–46). Based upon a subsample of forty BNS
respondents for whom we had three repeat 24h recalls
on non-consecutive days, we computed the within-per-
son variation in day-to-day nutrient intakes. Our results
were within the range of the within-person levels of
variation from 1d intake recalls and records found in the
published literature(41,43–45,47,48). Day-to-day variation in
intakes are on the level of 25% or higher for energy and
macronutrients(41,43,44,47). The CV are higher for many
micronutrients, particularly those that are found in high
amounts in a relatively small number of foods(41,48).
The nutrient intake estimations of the BNS respondents
met the criterion for reliable estimates used for nutritional
surveillance with the 24h recall in the NHANES studies. EI
in the BNS sample was similar to EI reports for adults
based on one 24h recall in the NHANES data(25,49);
however, the macronutrient distribution differed. BNS
respondents obtained a higher proportion of their energy
from carbohydrates and lower proportions from protein
and fats (63–64%, 13% and 25–26%, respectively) than
NHANES respondents (48–50%, 15–16% and 34%,
respectively)(50). When we compared the macronutrient
distribution of a subsample of BNS respondents (aged
$35 years) with that of Jewish Israelis of the same age
group, the same trends were found, and the differences
were statistically significant(9). BNS macronutrient intakes
showed distributions closer to those reported in middle-
income/developing countries (e.g. Iran(51), Korea(52),
Chile(53), western Mali(54)) than to those in high-income/
The study has a number of limitations. The BNS sample
was not drawn randomly, given the practical, logistical
difficulties of randomizing sample selection in this
population, so the generalizability of the results is limited.
Nevertheless, the sample included a broad geographical
cross-section of Negev Bedouin adults from both rural
and urban settings among whom the modified 24h recall
was successfully administered, demonstrating the feasi-
bility of using this tool to estimate individual intake from
common-plate meals in the Negev Bedouin population.
As with the original 24h recall, the EI:EER ratios below
1?0 obtained with the modified 24h recall suggest that
respondents tend to underestimate EI; and one 24h recall
does not necessarily reflect habitual intake. These factors
must always be taken into account when analysing and
interpreting short-term dietary intake data. At the same time,
the levels of EI underestimation and day-to-day within-
person variation in food intake we found with the modified
24h recall were well within the ranges reported for the
original 24h recall in a variety of other populations.
The modified 24h recall instrument, which quantifies
individual dietary intake from common-plate meals, has
the potential to make an important contribution to facil-
itating large-scale nutritional surveillance in the Negev
Bedouin Arab population. Additional dietary studies
among the Bedouin are needed in order to further refine
and validate the methods of quantifying individual intake
from common plates, and future nutritional surveillance
studies should include a larger number of repeat 24h
recalls, which would facilitate the estimation of usual
nutrient intakes. It may also be possible to adapt the
modified 24h recall to other populations where bread is
used for eating from common plates by developing local
food:bread ratios. In addition, the model may be useful
for developing dietary assessment methods in other
common-plate eating contexts by identifying some food/
meal component that can be quantified individually and
can also be utilized as a vehicle for quantifying the foods
eaten from common plates.
This research was supported by funds from the S. Daniel
Abraham International Center for Health and Nutrition,
Department of Epidemiology and Health Services Eva-
luation, Faculty of Health Sciences, Ben-Gurion Uni-
versity of the Negev, Israel. None of the authors have any
financial conflicts of interest. K.A.-S. contributed to the
study conception and design, the analysis and inter-
pretation of data, and the drafting of the paper. D.R.S.
contributed to the interpretation of the data and the
2470 K Abu-Saad et al.
drafting of the paper. H.A.-S. contributed to the study
conception and design and interpretation of the data.
H.V. contributed to the conception and design of the
study and the analysis and interpretation of the data. N.B.
contributed to the conception of the study and the
interpretation of the data. D.F. contributed to the con-
ception and design of the study, interpretation of the data,
and the drafting of the paper. We are grateful to Dr Arkadi
Bolotin for his assistance with data analysis in STATA.
Manuscript preparation was supported by a post-doctoral
grant from Mr Robert Arnow.
1. Groen JJ, Balough M, Levy M & Yaron E (1964) Nutrition
of the Bedouins in the Negev Desert. Am J Clin Nutr 14,
Harrison GG (1998) Experiences with dietary assessment in
the Middle East. Public Health Rev 26, 55–63.
Hudson GJ (1995) Food intake in a West African village.
Estimation of food intake from a shared bowl. Br J Nutr 73,
Shankar AV, Gittelsohn J, Stallings R, West KP Jr, Gnywali T,
Dhungel C & Dahal B (2001) Comparison of visual
estimates of children’s portion sizes under both shared-
plate and individual-plate conditions. J Am Diet Assoc 101,
Briefel RR, Sempos CT, McDowell MA, Chien S & Alaimo K
(1997) Dietary methods research in the third National
Health and Nutrition Examination Survey: underreporting
of energy intake. Am J Clin Nutr 65, 4 Suppl., 1203S–1209S.
Wright JD, Borrud LG, McDowell MA, Wang CY, Radimer K
& Johnson CL (2007) Nutrition assessment in the National
Health and Nutrition Examination Survey 1999–2002. J Am
Diet Assoc 107, 822–829.
Abou-Rbiah Y & Weitzman S (2002) Diabetes among
Bedouins in the Negev: the transition from a rare to a
highly prevalent condition. Isr Med Assoc J 4, 687–689.
Cohen AD, Gefen K, Ozer A, Bagola N, Milrad V, Cohen L,
Abu-Hammad T, Abu-Rabia Y, Hazanov I & Vardy DA
(2005) Diabetes control in the Bedouin population in
southern Israel. Med Sci Monit 11, CR376–CR380.
Fraser D, Bilenko N, Vardy H, Abu-Saad K, Shai I, Abu-
Shareb H & Shahar DR (2008) Differences in food intake
and disparity in obesity rates between adult Jews and
Bedouins in southern Israel. Ethn Dis 18, 13–18.
Tamir O, Peleg R, Dreiher J et al. (2007) Cardiovascular risk
factors in the Bedouin population: management and
compliance. Isr Med Assoc J 9, 652–655.
Choi BC, McQueen DV, Puska P et al. (2008) Enhancing
global capacity in the surveillance, prevention, and control
of chronic diseases: seven themes to consider and build
upon. J Epidemiol Community Health 62, 391–397.
Musaiger AO (2002) Diet and the prevention of coronary
heart disease in the Arab Middle East countries. Med
Principles Pract 11, Suppl. 2, 9–16.
Abu-Saad K, Weitzman S, Abu-Rabiah Y, Abu-Shareb H &
Fraser D (2001) Rapid lifestyle, diet and health changes
among urban Bedouin Arabs of southern Israel. FNA 28,
Fraser D, Abu-Saad K & Abu-Shareb H (2001) The relative
importance of traditional and ‘modern’ foods for Israeli
Negev Bedouins – a population in transition. Nutr Metab
Cardiovasc Dis 11, Suppl. 4, 66–69.
The Center for Bedouin Studies and Development (2004)
Statistical Yearbook of the Negev Bedouin, No. 2. Beer
Sheva: The Center for Bedouin Studies and Development
and the Negev Center for Regional Development, Ben-
Gurion University of the Negev.
Swirski S & Hasson Y (2006) Invisible Citizens: Israel
Government Policy Toward the Negev Bedouin. Beer-Sheva:
Shai I, Vardi H, Shahar DR, Azrad AB & Fraser D (2003)
Adaptation of international nutrition databases and data-
entry system tools to a specific population. Public Health
Nutr 6, 401–406.
US Department of Agriculture, Human Nutrition Informa-
tion Service (1996) Food Intake Analysis System, Version 3.
Houston, TX: University of Texas Health Science Center at
Houston School of Public Health.
Conway JM, Ingwersen LA, Vinyard BT & Moshfegh AJ
(2003) Effectiveness of the US Department of Agriculture
5-step multiple-pass method in assessing food intake in
obese and nonobese women. Am J Clin Nutr 77,
Pols MA, Peeters PH, Ocke ´ MC, Bueno-de-Mesquita HB,
Slimani N, Kemper HC & Collette HJ (1997) Relative validity
and repeatability of a new questionnaire on physical
activity. Prev Med 26, 37–43.
Sesso HD, Paffenbarger RS & Lee IM (2000) Physical
activity and coronary heart disease in men: the Harvard
Alumni Health Study. Circulation 102, 975–980.
Wareham NJ, Jakes RW, Rennie KL, Mitchell J, Hennings S &
Day NE (2002) Validity and repeatability of the EPIC-
Norfolk Physical Activity Questionnaire. Int J Epidemiol 31,
Azrad A (2002) Association between physical activity,
energy balance, dietary habits and health-related parameters
in a working population. Masters Thesis, Ben-Gurion
University of the Negev.
Food and Agriculture Organization of the United Nations/
World Health Organization/United Nations University
(2004) Human Energy Requirements: Report of a Joint
FAO/WHO/UNU Expert Consultation. FAO Food and Nutri-
tion Technical Report Series no. 1. Rome: FAO.
Wright JD, Wang CY, Kennedy-Stephenson J & Ervin RB
(2003) Dietary intake of ten key nutrients for public health,
United States: 1999–2000. Adv Data 334, 17 Apr, 1–4.
Schofield WN, Schofield C & James WPT (1985) Basal
metabolic rate – review and prediction, together with an
annotated bibliography of source material. Hum Nutr Clin
Nutr 39, Suppl. 1, 5–41.
Henry CJK (2005) Basal metabolic rate studies in humans:
measurement and development of new equations. Public
Health Nutr 8, 1133–1152.
Novotny JA, Rumpler WV, Riddick H & Herbert JR (2003)
Personality characteristics as predictors of underreporting
of energy intake on 24-hour dietary recall interviews. J Am
Diet Assoc 103, 1146–1151.
Tooze JA, Subar AF, Thompson FE, Troiano R, Schatzkin A
& Kipnis V (2004) Psychosocial predictors of energy
underreporting in a large doubly labeled water study. Am
J Clin Nutr 79, 795–804.
Ferrari P, Slimani N, Ciampi A et al. (2002) Evaluation of
under- and overreporting of energy intake in the 24-hour
diet recalls in the European Prospective Investigation into
Cancer and Nutrition (EPIC). Public Health Nutr 5,
Azizi F, Esmaillzadeh A & Mirmiran P (2005) Correlates of
under- and over-reporting of energy intake in Tehranians:
body mass index and lifestyle-related factors. Asia Pac J
Clin Nutr 14, 54–59.
Mendez MA, Wynter S, Wilks R & Forrester T (2004) Under-
and overreporting of energy is related to obesity, lifestyle
factors and food group intakes in Jamaican adults. Public
Health Nutr 7, 9–19.
Assessing individual intake from common-plate meals2471
33.Scagliusi FB, Ferriollo E & Lancha AH Jr (2006) Under- Download full-text
reporting of energy intake in developing nations. Nutr Rev
Harris JA & Benedict FG (1919) A Biometric Study of Basal
Metabolism in Man. Publication no. 279. Washington, DC:
Carnegie Institute of Washington.
Case KO, Brahler CJ & Heiss C (1997) Resting energy
expenditures in Asian women measured by indirect
calorimetry are lower than expenditures calculated from
prediction equations. J Am Diet Assoc 97, 1288–1292.
Douglas CC, Lawrence JC, Bush NC, Oster RA, Gower BA &
Darnell BE (2007) Ability of the Harris–Benedict formula to
predict energy requirements differs with weight history and
ethnicity. Nutr Res 27, 194–199.
Leung R, Woo J, Chan D & Tang N (2000) Validation of
prediction equations for basal metabolic rate in Chinese
subjects. Eur J Clin Nutr 54, 551–554.
Owen OE, Holup JL, D’Alessio DA et al. (1987) A
reappraisal of the caloric requirements of men. Am J Clin
Nutr 46, 875–885.
Jonnalagadda SS, Mitchell DC, Smiciklas-Wright H, Meaker
KB, Van Heel N, Karmally W, Ershow AG & Kris-Etherton
PM (2000) Accuracy of energy intake data estimated by a
multiple-pass, 24-hr dietary recall technique. J Am Diet
Assoc 100, 303–311.
Livingstone MBE & Black AE (2003) Markers of the validity
of reported energy intake. J Nutr 133, Suppl. 3, 895S–920S.
Hegsted DM (1982) The classic approach – the USDA
nationwide food consumption survey. Am J Clin Nutr 35,
Todd KS, Hudes M & Calloway DH (1983) Food intake
measurement: problems and approaches. Am J Clin Nutr
Ortega MI & Valencia ME (2002) Measuring the intakes of
foods and nutrients of marginal populations in north-west
Mexico. Public Health Nutr 5, 907–910.
Beaton GH, Milner J, Corey P et al. (1979) Sources of
variance in 24-hour dietary recall data: implications for
nutrition study design and interpretation. Am J Clin Nutr
Beaton GH, Milner J, McGuire V, Feather TE & Little JA
(1983) Source of variance in 24-hour dietary recall data:
implications for nutrition study design and interpretation.
Carbohydrate sources, vitamins and minerals. Am J Clin
Nutr 37, 986–995.
Moshfegh A, Goldman J & Cleveland L (2005) What We Eat
in America, NHANES 2001–2002: Usual Nutrient Intakes
from Food Compared to Dietary Reference Intakes.
Washington, DC: US Department of Agriculture, Agricul-
tural Research Service.
Block G & Hartman AM (1989) Issues in reproducibility and
validity of dietary studies. Am J Clin Nutr 53, 1133–1138.
Nelson M, Black AE, Morris JA & Cole TJ (1989) Between-
and within-subject variation in nutrient intake from infancy
to old age: estimating the number of days required to rank
dietary intakes with desired precision. Am J Clin Nutr 50,
US Department of Agriculture, Agricultural Research
Service (2003–2004) What we eat in America, NHANES
2003–2004: nutrient intakes Table 1 – mean amounts
consumed per individual, one day, 2003–2004. http://
0304/Table_1_NIF.pdf (accessed September 2008).
US Department of Agriculture, Agricultural Research
Service (2003–2004) What we eat in America, NHANES
2003–2004: nutrient intakes Table 2 – mean amounts and
percentages of calories from protein, carbohydrate, fat, and
alcohol, one day, 2003–2004. http://www.ars.usda.gov/
(accessed September 2008).
Mirmiran P, Esmaillzadeh A & Azizi F (2006) Under-
reporting of energy intake affects estimates of nutrient
intakes. Asia Pac J Clin Nutr 15, 459–464.
Kim J, Kim Y, Ahn Y, Paik H, Ahn Y, Tokudome Y,
Hamajima N, Inoue M & Tajima K (2003) Development of a
food frequency questionnaire in Koreans. Asia Pac J Clin
Nutr 12, 243–250.
Albala C, Vio F, Kain J & Uauy R (2002) Nutrition transition
in Chile: determinants and consequences. Public Health
Nutr 5, 123–128.
Torheim LV, Barikmo I, Hatløy A, Diakite ´ M, Solvoll K,
Diarra MM & Oshaug A (2001) Validation of a quantitative
food-frequency questionnaire for use in Western Mali.
Public Health Nutr 4, 1267–1277.
2472 K Abu-Saad et al.