Original Article Singapore Med J 2006; 47(6) : 491

Dietary intake, physical activity

and energy expenditure of
Malaysian adolescents
Zalilah M S, Khor G L, Mirnalini K, Norimah A K, Ang M

ABSTRACT had greater than 30 percent and less than

Introduction: Paediatric obesity is a public 55 percent of energy intake from fat and
health concern worldwide as it can track carbohydrate, respectively.
into adulthood and increase the risk of adult
Conclusion: The data suggested that a
morbidity and mortality. While the aetiology
combination of low energy expenditure
of obesity is multi-factorial, the roles of
adjusted for body weight and high dietary fat
diet and physical activity are controversial.
intake may be associated with overweight
Thus, the purpose of this study was to
and obesity among adolescents. To prevent
report on the differences in energy intake,
overweight and obesity among children and
diet composition, time spent doing physical
adolescents, strategies that address eating
activity and energy expenditure among
behaviours and physical activity are required.
underweight (UW), normal weight (NW)
Various segments of the society must be
and at-risk of overweight (OW) Malaysian
involved in efforts to promote healthful
adolescents (317 females and 301 males) aged
dietary intakes and active lifestyle in children
11-15 years. and adolescents.
Methods: This was a cross-sectional study
Keywords: adolescents, dietary intake, energy
with 6,555 adolescents measured for weights
expenditure, energy intake, obesity, physical
and heights for body mass index (BMI)
activity
categorisation. A total of 618 subjects were Department of Nutrition
randomly selected from each BMI category Singapore Med J 2006; 47(6):491-498 and Dietetics
Faculty of Medicine and
according to gender. The subjects’ dietary Health Sciences
Universiti Putra
intake and physical activity were assessed INTRODUCTION Malaysia
using self-reported three-day food and Paediatric obesity has become an increasing clinical Serdang 43400
Malaysia
activity records, respectively. Dietary intake and public health concern in both developed and
Zalilah M S, PhD
components included total energy and developing nations. Children of all ages are fatter Senior Lecturer
macronutrient intakes. Energy expenditure now than they were before(1-3). In many countries,
Khor G L, PhD
was calculated as a sum of energy expended rapid economic development has brought about Professor
for basal metabolic rate and physical activity. changes in the populationsʼ dietary intakes and Mirnalini K, PhD
Time spent (in minutes) in low, medium and lifestyle. Consequently, these changes contribute Associate Professor
high intensity activities was also calculated. significantly to the increasing prevalence of obesity Ang M, MSc
Graduate Student
Results: The OW adolescents had the highest and diet-related chronic diseases, both in the adult and
paediatric populations(4). Malaysia is no exception in Department of Nutrition
crude energy intake and energy expenditure. and Dietetics
However, after adjusting for body weight, the that the shift towards a Westernised dietary pattern Faculty of Allied Health
Sciences
OW subjects had the lowest energy intake and sedentary lifestyle has been implicated in the Universiti Kebangsaan
increasing rate of obesity and nutrition-related non- Malaysia
and energy expenditure (p-value is less Kuala Lumpur 50300
than 0.001). The study groups did not differ communicable diseases among its population(5). Malaysia

significantly in time spent for low, medium In Malaysia, the National Health and Morbidity Norimah A K, PhD
Survey 2 (1996) reported that the prevalence Associate Professor
and high intensity activities. Macronutrient
intakes differed significantly only among the of being overweight and obese among adults is Correspondence to:
Dr Zalilah Mohd Shariff
girls where the OW group had the highest 20.7% and 5.8%, respectively(6). The prevalence of Tel: (60) 3 8946 8551
intakes compared to UW and NW groups overweight adults (body mass index [BMI] >25.0) Fax: (60) 3 8942 6769
Email: zalilah@
(p-value is less than 0.05). All study groups in Malaysia appears to be higher than the levels medic.upm.edu.my
 Singapore Med J 2006; 47(6) : 492

reported in other Asian countries, such as Thailand, Technology and Environment. For this study, multi-
China and Japan(7). However, this health problem is stage random sampling was utilised in the selection
not restricted only to the Malaysian adults. Among of 14 secondary schools. Other inclusion criteria
the adolescents, the prevalence of overweight and for the schools were: co-educational, multiracial in
obesity from independent studies ranges from 5% composition, non-religious and non-residential.
to 26%, depending on the methods used to define All Secondary 1 and 2 male and female students
overweight and obesity, age, ethnicity, gender and (n=6,555) in the age range of 11-15 years old were
locality(8-11). included in the screening process where they were
Reilly et al(12) showed that obesity in children measured for their weight and height. BMI was
and adolescents is associated with both short-term calculated for each respondent for the purpose of
and long-term consequences, although the latter categorisation into underweight, overweight and
may be of a greater health concern. Short-term normal weight groups. These respondents were
consequences of obesity in children include low self- then randomly sampled to obtain a total of 2,050
esteem, behavioural problems, clinical conditions respondents for the assessment of body image. From
such as asthma, systemic inflammation and type 1 this sample, another sub-sample of 618 respondents
diabetes mellitus, and cardiovascular risk factors. (317 female and 301 male adolescents) was randomly
Long-term consequences of obesity in childhood selected to complete additional measurements such
and adolescence may include adverse social and as three-day food and physical activity records.
economic outcomes (e.g. income, educational The study protocol and ethics were approved by the
achievement) and increased cardiovascular risks and Malaysian Ministry of Education. Permissions to
premature mortality in adulthood. conduct the study were also obtained from the state
The aetiology of obesity involves a complex education departments and participating schools. The
interaction of various factors, such as genetics, respondents were informed of the study and their
physical activity, diet, social, environment and participation was voluntary. Those who participated
health. The roles of diet and physical activity in were required to sign consent forms prior to the
body weight gain and body fat accumulation among inception of the study.
adolescents have remained equivocal. While several The weight and height of the respondents
studies have reported on the positive associations were measured with a TANITA digital scale and
between adiposity in children and adolescents a SECA body metre to the nearest 0.1 kg and 0.1
and energy intake, consumptions of high fat cm, respectively. Two measurements were taken for
foods and high sugar beverages and physical both weight and height, and the average of the two
inactivity(13-16), others have found no relationship(17-19). values were used in the analyses. For measurement
The relationship between physical fitness and accuracy, the respondents were asked to remove their
adiposity has also been researched in line with the shoes and empty their pockets, and the instruments
idea that overweight children had reduced physical were calibrated daily. BMI was calculated based
fitness due to increased physical inactivity(20-22). on the weight and height measurements and the
Presently, published research on the relationship categorisation of BMI levels were done according to
between adiposity with diet and physical activity the age and gender-specific BMI for adolescents(23).
among Malaysian adolescents is limited. Hence, the Thus, the BMI categories of UW, NW and OW
purpose of this study was to report on the energy corresponded to <5th percentile, 5th - 85th percentile
and macronutrient intakes, and energy expenditure, and ≥85th percentile, respectively.
among male and female adolescents according to For dietary intake, a self-administered three-
the various BMI levels, namely underweight (UW), day food record (consecutive two weekdays and
normal weight (NW) and at risk of overweight a weekend) were obtained from the respondents.
(OW). Respondents were asked to record their consumption
of all foods and beverages, including snacks, and
METHODS to provide detailed descriptions of the foods and
This cross-sectional study was conducted as part beverages, which included brand names and methods
of a multi-centre research project on “The study of of food preparation. Food portions were estimated by
the relationship between dietary habits, physical the respondents using standard household measuring
activity and nutritional status with body image cups and spoons, matchbox (for meat, fish and bean
among adolescents” with financial support from curd), ruler (for fruit slices) and counts (for eggs,
the Intensification of Research in Priority Areas biscuits, bread slices and fruits). Prior to that, the
(IRPA) grant of the Malaysian Ministry of Science, researchers explained the food record form and the
 Singapore Med J 2006; 47(6) : 493

food-measuring tools, and demonstrated the use of the dietary intake analysis. Dietary quality was
these instruments to the respondents. All dietary reported for macronutrient intakes (carbohydrate,
records were checked by researchers for accuracy fat and protein) as mean intakes in grammes and
and completeness. Further clarifications were sought percentages of energy.
from the respondents in cases where the food records As the data were parametric, one-way ANOVA
were incomplete or filled incorrectly. Calculation of and chi-square analyses were conducted to assess the
energy and macronutrient intakes was done for each relationship between the BMI categories according
day, and the average of the three daysʼ calculation to gender and demographical, socioeconomic and
was taken in the final statistical analyses. physical measurements. The interaction between
The respondents were also asked to record their ethnicity (Malay, Chinese and Indian) and household
physical activities simultaneously with their food income was tested with a two-way ANOVA; however,
records. This three-day activity record method no interaction effect was observed (F=0.65, p=0.63).
has been shown to be suitable for the estimation Finally, analysis of covariance (ANCOVA) was
of energy expenditure in children(24) and had been utilised to determine the mean differences in the
previously used among Malaysian adolescents(25). outcome variables among UW, NW and OW subjects
The activities including the body postures (sitting, with statistical control of household income and
standing, walking and squatting) were recorded ethnicity. These covariates were included individually
every five minutes. There were a total of 42 recorded and combined in the ANCOVA; however, the mean
activities and based on their characteristics and the scores for all outcome variables from ANCOVA with
circumstances of their performance, these activities individual covariate did not differ much from the
were then grouped into nine categories according mean scores when both covariates were included in
to energy cost or physical activity ratio (PAR) as ANCOVA. All statistical analyses were conducted
recommended by Torun(26). Thus, light, moderate and using Statistical Package for Social Sciences (SPSS)
high intensity activities were defined as activities version 11.5 (Chicago, IL, USA).
with PAR of 1-1.9, 2-2.9 and >3.0, respectively.
The sum of the duration (in minutes) of activities RESULTS
in each category was calculated and included in the Based on the screening sample (n=6,555), the
energy expenditure calculation(27). The calculation proportion of UW, NW and OW male and female
for the adolescentsʼ basal metabolic rate (BMR) in adolescents are presented in Table I. The prevalence
kcal/day, was based on a validated gender-specific of UW and OW was higher among male (14.8% and
BMR equation for Malaysian adolescents(25). Energy 19.7%) than female (7.9% and 16.7%) adolescents.
expended for physical activity was calculated as the The mean and range of BMI for male and female
difference between total energy expenditure and adolescents were 19.26 + 4.16 (10.58-45.71) and
energy expended for BMR. The reported values 19.72 + 4.12 (10.51-43.98), respectively. Table II
for energy expenditure and physical activity were describes the demographical characteristics and
calculated based on the average values of the three anthropometric measurements of the subjects by
days of activity record. gender and study group (n=618).
The dietary intakes of the respondents were Self-reported energy intake and expenditure and
analysed using Nutri-Cal Professional Version 1.01 energy balance of the subjects are shown in Table
which had the Malaysian food database(28). In cases III. The OW girls had significantly higher crude
where the foods or ingredients consumed were energy intake than the UW and NW groups (p<0.05).
not available in the database, other food databases Among the boys, the differences in mean crude
such as “Singapore Food Facts”(29) and “Thai Food energy intake were not statistically significant. When
Composition Tables”(30) were also utilised to assist body weight was considered, the OW subjects (girls

Table I. Prevalence of underweight, normal weight and overweight male and female adolescents (n=6,555)
Gender Total UW NW OW χ2 p-value
n n(%) n(%) n(%)
Male 3,353 498 (14.8) 2,195 (65.5) 660 (19.7) 101.6 < 0.001
Female 3,202 252 (7.9) 2,415 (75.4) 535 (16.7)
Total 6,555 750 (11.4) 4,610 (70.3) 1,195 (18.3)
BMI-for-age (WHO, 1995): <5th percentile (underweight: UW); 5th - 85th percentile (normal weight: NW);
≥ 85th percentile (at risk of overweight: OW).
 Singapore Med J 2006; 47(6) : 494

Table II. Characteristics of subjects by gender and body mass index (n=618).
Girls Boys
UW NW OW p-value UW NW OW p-value
(n=72) (n=123) (n=122) (n=82) (n=118) (n=101)
Age (year)a 13.0 (0.7) 13.1 (0.7) 13.2 (0.8) ns 13.1 (0.8) 13.2 (0.8) 13.1 (0.8) ns
Height (m)a 1.49 (0.1) 1.51 (0.08) 1.54 (0.05) <0.001 1.50 (0.10) 1.57 (0.09) 1.60 (0.08) <0.001
Weight (kg)a 32.8 (3.5) 42.6 (6.3) 64.2 (10.4) <0.001 33.5 (5.2) 45.7 (8.3) 65.5 (11.6) <0.001
BMI (kg/m2)a 14.7 (0.6) 18.7 (1.9) 27.0 (3.5) <0.001 14.8 (0.8) 18.5 (2.0) 25.5 (3.4) <0.001
Household incomea 1,328 (1,708) 1,082 (1,123) 1,124 (1,254) <0.05 958 (939) 1,002 (940) 1,193 (877) <0.01
(RM)
0-500 19 (26.4) 44 (35.8) 34 (27.9) 28 (34.1) 34 (28.8) 30 (29.7)
501-1000 31 (43.1) 50 (40.7) 58 (47.5) 35 (42.7) 60 (50.8) 39 (38.6)
>1000 22 (30.5) 29 (23.5) 30 (24.6) 19 (23.2) 24 (20.4) 32 (31.7)
Ethnicityb <0.05 <0.05
Malay 43 (59.7) 84 (68.3) 72 (59.0) 51 (62.2) 56 (47.5) 48 (47.5)
Chinese 24 (33.3) 33 (26.8) 36 (29.5) 16 (19.5) 46 (39.0) 43 (42.6)
Indian 5 (7.0) 6 (4.9) 14 (11.5) 15 (18.3) 16 (13.5) 10 (9.9)
UW: underweight; NW: normal weight; OW: at risk of overweight
1 USD = RM 3.8; a: Mean (SD); b: n (%)
ns: not significant

Table III. Energy intake, energy expenditure and energy expended for basal metabolic rate and physical
activity by gender and body mass index (n=618).
Girlsb Boysb
UW NW OW p-value UW NW OW p-value
(n=72) (n=123) (n=122) (n=82) (n=118) (n=101)
Energy intakea
Kcal/day 1,916 (73.7) 1,903 (56.5) 2,138 (56.7) <0.05 2,198 (92.4) 2,133 (77.1) 2,262 (84.1) ns
Kcal/kg body weight 59 (1.8) 46 (1.4) 34 (1.4) <0.001 66 (2.4) 49 (2.0) 36 (2.2) <0.001
Energy expenditurea
Kcal/day 1,581 (24.0) 1,778 (18.4) 2,189 (18.4) <0.001 1,917 (38.8) 2,195 (32.4) 2,871 (35.3) <0.001
Kcal/kg body weight 49 (0.5) 42 (0.4) 34 (0.4) <0.001 58 (0.7) 49 (0.6) 44 (0.6) <0.001
Basal metabolic ratea
Kcal/day 1,091 (11.9) 1,219 (9.2) 1,500 (9.2) <0.001 12,02 (18.8) 1,431 (15.7) 1,811 (17.1) <0.001
Physical activitya
Kcal/day 490 (17.7) 559 (13.6) 689 (13.6) <0.001 715 (28.7) 763 (23.9) 1,059 (26.1) <0.001
Energy balancea
Kcal/day 334 (74.6) 125 (57.2) -51 (57.4) <0.001 281.7 (94) -61 (79.1) -608 (86.3) <0.001
UW: underweight; NW: normal weight; OW: at risk of overweight
a
: Analysis of covariance controlling for ethnicity and household income
b
: Mean (standard error)
ns: not significant
BMR (kJ/day)
BMR (kcal/day) = 4.184 (kJ/kcal)

and boys) had the lowest energy intake compared the OW subjects (p<0.001). OW girls and boys had
to the UW and NW subjects (p<0.001). Crude significantly lower energy balance (negative energy
energy expenditure and energy expended for BMR balance) than NW and UW subjects (p<0.001).
and physical activity were significantly greater for The mean duration in minutes of low, moderate
OW boys and girls. However, our data also showed and high intensity physical activities performed by
that the total energy expenditure per kilogramme of the UW, NW and OW adolescents is presented in
body weight for boys and girls was lowest among Table IV. There was no significant difference in time
 Singapore Med J 2006; 47(6) : 495

Table IV. Mean duration (minutes) of low, moderate and high intensity physical activity by gender and body
mass index (n=618).
Girlsb (n=317) Boysb (n=301)
Physical activity UW NW OW p-value UW NW OW p-value
(n=72) (n=123) (n=122) (n=82) (n=118) (n=101)
Light intensitya 1,156 (9.4) 1,147 (7.2) 1,148 (7.2) ns 1,163 (9.4) 1,180 (7.8) 1,157 (8.6) ns
Moderate intensity a
245 (8.2) 252 (6.3) 251 (6.3) ns 207 (8.0) 205 (6.7) 218 (7.3) ns
High intensity a
40 (3.7) 41 (2.8) 41 (2.8) ns 69 (4.8) 55 (4.0) 65 (4.4) ns
UW: underweight; NW: normal weight; OW: at risk of overweight
a
: Analysis of covariance controlling for ethnicity and household income
b
: Mean (standard error)
Light: sleeping, lying down, sitting (resting, reading, writing, playing computer and video games, watching TV, eating), standing still
(watching TV, talking, queuing)
Moderate: walking, standing with movements (washing dishes, cleaning, dusting, gardening)
High: cycling, walking up and down the stairs, running, playing sports, exercise
ns: not significant

Table V. Macronutrient intakes by gender and body mass index (n=618).

Girlsb (n=317) Boysb (n=301)
UW NW OW p value UW NW OW p-value
(n=72) (n=123) (n=122) (n=82) (n=118) (n=101)
Total intake (grammes)a
Total carbohydrate 247 (9.8) 252 (7.5) 272 (7.5) <0.05 281 (1.8) 271 (9.9) 285 (10.7) ns
Total fat 70 (3.6) 69 (2.7) 81 (37.4) <0.05 80 (4.0) 81 (3.4) 86 (3.7) ns
Total protein 72 (3.2) 70 (2.4) 80 (2.5) <0.05 89 (5.9) 80 (4.9) 86 (5.4) ns
Percent energy (%)a
From carbohydrate 52 (0.9) 53 (0.7) 52 (7.8) ns 52 (0.9) 51 (0.8) 51 (0.8) ns
From fat 33 (0.8) 32 (0.6) 33 (0.6) ns 32.4 (0.7) 34 (0.6) 34 (0.7) ns
From protein 15 (0.3) 15 (0.3) 15 (0.3) ns 16 (0.4) 15 (0.4) 15 (0.4) ns
UW: underweight; NW: normal weight; OW: at risk of overweight
a
: Analysis of covariance controlling for ethnicity and household income
b
: Mean (standard error)
ns: not significant

spent in each physical activity category among the energy from carbohydrates (<55%).
study groups. In general, OW adolescents did not
appear to spend more time doing low to moderate DISCUSSION
activities compared to the UW and NW groups. The high proportion of adolescents at-risk of
This, however, could also be due to misreporting overweight in this multi-ethnic sample was similar
of activities, especially among the overweight and to(9) or higher than other findings among Malaysian
obese subjects. Macronutrient intakes (carbohydrate, adolescents(8-11). The prevalence of at-risk overweight
protein and fat), expressed as intake in grammes and reported in this present study, however, may not be
as percentage of energy intake are shown in Table V. comparable to those reported in previous studies
When compared to the UW and NW groups, the OW due to different methods and cut-offs used to
girls had the highest total grammes of macronutrient define overweight and obesity, age groups, ethnic
intake (p<0.05). No significant differences were composition and locality (urban and rural) utilised
observed for macronutrient intakes (total grammes in each study. Nevertheless, overweight and obesity
and percent of energy) among the UW, NW and OW among children and adolescents is of a concern as
boys. All groups, however, had more than 30% of the problem may persist into adulthood and increase
energy from fat and relatively lower percentage of the risk of adult morbidity and mortality(31-33).
 Singapore Med J 2006; 47(6) : 496

Various studies have shown that overweight or respectively. However, The National Coordinating
obese children did not consume more energy; they Committee on Food and Nutrition of Malaysia
consumed less energy and had higher physical (2005), recommends that carbohydrates should
activity than their normal weight peers, yet contribute 55-75%, fat 20-30% and proteins 10-15%
maintained their adiposity(14,17-18). However, other of total daily energy intake.
researchers have reported that energy intake was In comparison to the NW and UW adolescents
positively related to adiposity(13,15-16). Our findings in this study, the OW adolescents expended more
showed that while crude energy intake did not energy per day for physical activity and BMR.
differ significantly among UW, NW and OW male The higher BMR among the overweight subjects
adolescents, OW female adolescents had significantly was expected because the increase in body weight
higher crude energy intake than the other two groups. will result in an increase in BMR(39). However, the
A mean difference in the range of 113-173 kcal/day higher energy expended for physical activity by the
between the NW and OW male and female subjects OW subjects remained questionable. Several studies
in this study (with the OW subjects consuming the have shown that obese children are physically less
extra energy), could as well contribute to a weight active than non-obese children, however, due to
gain of several pounds per year(34). higher body weight, the energy requirement of
We reported that after body weight was the same activity in obese children is much higher
considered, the overweight subjects actually had the than in the non-obese(40-41). Thus, the obese children
lowest energy intake compared to the subjects in the expended more energy for physical activity than did
other two groups. Our results are similar to those of non-obese children, yet were not necessarily more
Gazzaniga and Burns(35) in that the researchers also active. It is also possible that the OW adolescents had
found that the 9- to 11-year-old obese children had misreported the duration and intensity of physical
higher crude daily energy intakes but lower energy activities which then caused an overestimation of
intake per kg of body weight than did the non- physical activity- associated energy expenditure(42).
obese children. These findings may actually reflect The most significant finding from this study is
systematic underreporting among the OW subjects that when body weight is considered, energy intake
which consequently leads to the difficulty to identify and energy expenditure are lowest among the OW
excess energy intake as the main contributor to subjects. Gazzaniga and Burns(35) have also found
energy imbalance. that among boys and girls aged 9-11 years, energy
In children, percent of energy from dietary intake, total energy expenditure, resting energy
fat is positively related with increasing expenditure and energy expended for physical
adiposity(16,19,36). Percentage of body fat has also activity were substantially higher among obese
been shown to correlate positively with intakes of than non-obese children; however, the total energy
total, saturated and monounsaturated fatty acids expenditure and energy intake per kg body weight
and negatively with carbohydrate intake, even were lower. Perhaps, among the overweight and
after adjusting for energy intake, resting energy obese individuals, the combination of a genetically-
expenditure and physical activity among pre- determined low fat oxidation capacity when the
adolescent children(35). Similarly, Eck et al(36) have diet is high in fat, and a sedentary lifestyle with a
shown that children at high risk of obesity consumed low level of energy expended on physical activities
higher percentage of fat energy but lower percentage (contributing to a low total energy expenditure) may
of carbohydrate energy than children at low risk of put the individuals at risk of further weight gain(43),
obesity. In contrast, Grant et al(38) reported that obese independent of energy intake.
children did not consume significantly more energy It is always important to highlight the limitations
from macronutrients than non-obese children. of the study that may influence the study findings and
In the present study, while mean total gramme warrant caution in the interpretation of the findings.
intake of macronutrients (carbohydrate, fat, protein) Firstly, the design of the study is cross-sectional,
was significantly higher (p<0.05) in OW than NW and which provided information only on the overweight
UW girls, no significant differences were observed status and not on the development of the overweight
among the OW, NW and UW boys. Unlike previous status. The study did not obtain important information
studies, we did not find any significant difference related to weight changes i.e. gain, loss or stable.
in percentage of energy from macronutrients by Thus, the data could only report on the behaviours
gender and study groups. The adolescents in this that were associated with the current weight status
study consumed 32-34%, 51-53% and 14-16% of and not behaviours that contribute to weight gain.
energy intake from fat, carbohydrates and proteins, For example, excess energy intake and low physical
 Singapore Med J 2006; 47(6) : 497

activity may be contributory factors to overweight To promote a healthy lifestyle in children and
and obesity in adolescents, but reduced energy adolescents, which eventually can prevent overweight
intake and high or increased physical activity may and obesity, health promotional strategies that address
be the outcome behaviours of overweight and obese a broad range of factors, especially environmental
adolescents in their attempts to reduce weight. The factors that relate to eating behaviours and physical
present study, however, was not able to determine activity, may be required. Efforts to promote healthy
the direction of these dietary and physical activity eating habits and regular physical activities among
variables. Perhaps, future research with a longitudinal children and adolescents must engage all segments
study design may overcome these limitations. of the society, such as the government and non-
Secondly, the use of three-day food and physical government agencies, health professionals, food
activity records among the adolescents depended industries, media, school authorities and parents, to
very much on the initiative, cooperation and honesty help create an environment in which the children live
of the respondents to record their daily food intakes and form their behaviours.
and activities. The act of recording food intake and
physical activity over a period of time can also lead ACKNOWLEDGEMENT
the respondents to simplify the recording processes The project was funded by the Intensification of
i.e. reduce the number, types and frequency of Research in Priority Areas (IRPA) grant (project
foods and snacks eaten or types and duration of number: 06-02-05-9005) of the Malaysian Ministry
activities performed(44). Consequently, the records of Science, Technology and Environment.
may significantly under-report or over-report
energy and nutrient intakes and energy expenditure. REFERENCES
Thirdly, the physical activity and energy expenditure 1. Sawaya AL, Dallal G, Solymos G, et al. Obesity and malnutrition in a
shanty town population in the city of Sao Paulo, Brazil. Obesity Res
information obtained were of crude measurements 1995; 3:107S-15S.
compared to assessments using doubly-labelled 2. Popkin BM, Udry JR. Adolescent obesity increases significantly in
second and third generation US immigrants: the National Longitudinal
water method, indirect calorimetry, heart rate
Study of Adolescent Health. J Nutr 1998; 128:701-6.
monitoring or motion sensors. In addition, the use of 3. OʼLoughlin J, Paradis G, Meshefedjian, G, Gray-Donald K. A five-
equations to calculate BMR and the assumption that year trend of increasing obesity among elementary schoolchildren
in multiethnic, low-income, inner-city neighbourhoods in Montreal,
the energy cost for all activities in each category is Canada. Int J Obes Relat Metab Disord 2000; 24:1176-82.
the same may introduce error in the determination 4. World Health Organization. Obesity. Preventing and managing the
of total energy expenditure. global epidemic. Geneva: WHO, 1998.
5. Tee ES. Nutrition of Malaysians: where are we heading? Mal J Nutr
Finally, the aetiology of overweight and 1999; 5:87-109.
obesity is multi-factorial. At present, the role of 6. Lim TO, Ding LM, Zaki M, et al. Distribution of body weight, height
individual factors (e.g. dietary intake, genetic and and body mass index in a national sample of Malaysian adults. Med
J Malaysia 2000; 55:108-28.
physical activity) and their interactions are still 7. Khor GL. Nutrition and cardiovascular disease: an Asia Pacific
inconclusive. Our study, however, did not consider perspective. Asia Pac J Clin Nutr 1997; 6:122-42.
8. Bong ASL, Safurah J. Obesity among years 1 and 6 primary school
genetic factors which may influence the roles of
children in Selangor Darul Ehsan. Mal J Nutr 1996; 2:21-7.
diet composition, energy intake and expenditure 9. Judson JP, Jeyalingam K. Small yet large - obesity profile in
in the maintenance of overweight and obesity Malaysian school children. In: Ismail MN, ed. Proceedings of the 2nd
Scientific Meeting on Obesity. Kuala Lumpur: MASO, 1998: 59-68.
among the adolescents. Despite the limitations
10. Kasmini K, Idris MN, Fatimah A, et al. Prevalence of overweight
in study design and methodology, our findings and obese school children aged between 7 and 16 years amongst the
indicated that the overweight adolescents were not major 3 ethnic groups in Kuala Lumpur, Malaysia. Asia Pac J Clin
Nutr 1997; 6:172-4.
consuming more energy than their non-overweight 11. Aminah A, Ain A, Suriah AR. Prevalence of obesity among Malay
counterparts, after body weight was adjusted. primary school children. In: Ismail MN, ed. Proceedings of the 3nd
However, compared to non-overweight adolescents, Scientific Meeting on Obesity. Kuala Lumpur: MASO, 1999: 3-9.
12. Reilly JJ, Methven E, McDowell ZC, et al. Health consequences of
the overweight adolescents had significantly lower obesity. Arch Dis Child 2003; 88:748-52.
energy expenditure per kilogramme of body weight. 13. Tucker LA, Seljaas GT, Hager RL. Body fat percentage of children
varies according to their diet composition. J Am Diet Assoc 1997;
Despite no significant difference in total energy
97:981-6.
contribution from macronutrients, all study groups 14. Troiano RP, Briefel RR, Carrol M, Bialostosky K. Energy and fat
had higher percentage of energy from fat (>30%) intakes of children and adolescents in the United States: data from
the National Health and Nutrition Examination Surveys. Am J Clin
but lower percentage of energy from carbohydrate
Nutr 2000; 72(Suppl):1343S-53S.
(<55%). Our data suggested that a combination of a 15. Andersen RE, Crespo CJ, Bartlett SJ, Cheskin LJ, Pratt M.
diet high in fat (>30% of total energy intake) and low Relationship of physical activity and television watching with body
weight and level of fatness among children. JAMA 1998; 279:938-
energy expenditure may contribute to overweight 42. Comment in: JAMA 1998; 279:959-60, JAMA 1998; 280:
and obesity among the adolescents(36,42). 1230-2.
 Singapore Med J 2006; 47(6) : 498

16. Robertson SM, Cullen KW, Baranowski J, et al. Factors related to 31. Must A, Jacques PF, Dallal GE, Bajema CJ, Dietz WH. Long-term
adiposity among children aged 3 to 7 years. J Am Diet Assoc 1999; morbidity and mortality of overweight adolescents. A follow-up of
99:938-43. the Harvard Growth Study of 1992 to 1935. New Eng J Med 1992;
17. Rolland-Cachera MF, Bellisle F. No correlation between adiposity 327:1350-5.
and food intake: why are working class children fatter? Am J Clin 32. Guo S, Roche AF, Chumlea WC, Gardner JD, Siervogel RM. The
Nutr 1986; 44:779-87. predictive value of childhood body mass index values for overweight
18. Calderon LL, Johnston PK, Lee JW, Haddad EH. Risk factors for at age 35 years. Am J Clin Nutr 1994; 58:810-9.
obesity in Mexican-American girls: Dietary factors, anthropometric 33. Must A. Morbidity and mortality associated with elevated body weight
factors and physical activity. J Am Diet Assoc 1996; 96:1177-9. in children and adolescents. Am J Clin Nutr 1996; 63 (3Suppl):445S-
19. Maffeis C, Zaffanello M, Schutz Y. Relationship between physical 447S.
inactivity and adiposity in prepubertal boys. J Pediatr 1997; 131:288- 34. Klesges RC, Klesges LM, Eck LH, Shelton ML. A longitudinal
92. Erratum in: J Pediatr 1998; 132:747. analysis of accelerated weight gain in preschool children. Pediatrics
20. Grund A, Dilba B, Forberger K, et al. Relationships between physical 1995; 95:126-30.
activity, physical fitness, muscle strength and nutritional state in 5- to 35. Gazzaniga JM, Burns TL. Relationship between diet composition
11-year-old children. Eur J Appl Physiol 2000; 82:425-38. and body fatness, with adjustment for resting energy expenditure and
21. Reybrouck T, Weymans M, Vinckx J, Stijns H, Vanderschueren- physical activity in preadolescent children. Am J Clin Nutr 1993;
Lodeweyckx M. Cardiorespiratory function during exercise in obese 58:21-8.
children. Acta Paediatr Scand 1987; 76:342-8. 36. Eck LH, Klesges RC, Hanson CL, Slawson D. Children at familial
22. Cooper DM, Poage J, Barstow TJ, Springer C. Are obese children risk for obesity: an examination of dietary intake, physical activity
truly unfit? Minimizing the confounding effect of body size on the and weight status. Int J Obesity 1992; 16:71-8.
exercise response. J Pediatr 1990; 116: 223-30. 37. Obarzanek E, Schreiber GB, Crawford PB, et al. Energy intake and
23. World Health Organization. Physical status: the use and interpretation physical activity in relation to indexes of body fat: the National
of anthropometry. Geneva: World Health Organization; 1995 Heart, Lung and Blood Institute Growth and Health Study. Am J Clin
Technical Series Report No. 854. Nutr 1994; 60:15-22.
24. Bouchard C, Tremblay A, Leblanc C, et al. A method to assess energy 38. Grant AM, Ferguson EL, Toafa V, Henry TE, Guthrie BE. Dietary
expenditure in children and adults. Am J Clin Nutr 1983; 37:461-7. factors are not associated with high levels of obesity in New Zealand
25. Poh BK, Ismail MN, Zawiah H, Henry CJK. Predictive equations for Pacific preschool children. J Nutr 2004; 134:2561-5.
the estimation of basal metabolic rate in Malaysian adolescents. Mal 39. Bandini LG, Schoeller DA, Dietz WH. Energy expenditure in obese
J Nutr 1999; 5:1-14. and nonobese adolescents. Pediatr Res 1990; 27:198-203.
26. Torun B. Energy cost of various physical activities in healthy children. 40. Johnson ML, Burke BS, Mayer J. Relative importance of inactivity
In: Schürch B, Scrimshaw NS, eds. Activity, Energy Expenditure and overeating in the energy balance of obese high school girls. Am
and Energy Requirements of Infants and Children. Cambridge, MA: J Clin Nutr 1956; 4:37-44.
International Dietary Energy Consultancy Group, 1989: 139-82. 41. Molnar D, Livingstone B. Physical activity in relation to overweight
27. World Health Organization. Energy and protein requirements. and obesity in children and adolescents. Eur J Pediatr 2000; 159
Geneva: World Health Organization; 1985 Technical Series Report Suppl 1:S45-55.
No. 724. 42. Buchowski MS, Townsend KM, Chen KY, Acra SA, Sun M. Energy
28. Tee ES, Ismail MN, Nasir MA, Khatijah I. Nutrient Composition expenditure determined by self-reported physical activity is related
of Malaysian Foods. 4th ed. Kuala Lumpur: Institute for Medical to body fatness. Obes Res 1999; 7:23-33.
Research, 1997. 43. Astrup A. Macronutrient balances and obesity: the role of diet and
29. Department of Nutrition Singapore. Singapore food facts. Singapore: physical activity. Public Health Nutr 1999; 2:341-7.
Institute of Health, 2000. 44. Rebro SM, Patterson RE, Kristal AR, Cheney CL. The effect of
30. Puwastien P, Raroengwichit M, Sungpuag P, Judprasong K. Thai keeping food records on eating patterns. J Am Diet Assoc 1998;
food composition tables. Thailand: Institute of Nutrition Mahidol 98:1163-5.
University, 1999.