Article

Nutrition and Health

2018, Vol. 24(4) 231–240
Dietary practices of school children ª The Author(s) 2018
Article reuse guidelines:
in Sindh, Pakistan sagepub.com/journals-permissions
DOI: 10.1177/0260106018791859
journals.sagepub.com/home/nah

Ayesha Aziz , Mehak Pervaiz, Ayesha Khalid,

Ayesha Zahid Khan and Ghazala Rafique

Abstract
Background: Malnutrition in Pakistan has remained an unresolved issue for decades. Data on nutrient intake and variety
in children’s diet is lacking. Aim: This paper describes the dietary practices of school children across eight districts of
Sindh. Methods: A school-based survey of dietary practices among 1109 students from classes 2–5 in 36 schools was
conducted using a pre-tested interviewer-completed questionnaire. Results: Descriptive and comparative analysis of data
showed that more than 75% school children ate breakfast, lunch and dinner regularly. However, 10% children skipped
breakfast and nearly one-third of them skipped mid-morning and evening snacks. Skipping breakfast and snacks was related
to low socio-economic status and rural residence. Children’s intake of protein-rich food items and fruits also illustrated
that urban residence and better socio-economic status have a positive effect on the frequency and variety of daily protein
and fruit intake. Though affordability and geographic location were key determinants of children’s dietary practices,
significantly higher percentages of children studying in class 2, 3 and 4 consumed different kinds of protein-rich food items,
fruits and vegetables more frequently as opposed to children studying in class 5. Junk food consumption was more
common in urban areas and better socio-economic strata. Conclusion: Development of school-based nutrition edu-
cation programmes and behavior change strategies emphasizing the regular intake of meals and nutrient-rich snacks, and
encouraging consumption of easily available and affordable food items such as dried seeds, fruits, vegetables, milk and its
derivatives is essential for addressing nutrient deficiencies.

Keywords
Malnutrition, meal frequency, junk food, food choices, socio-economic status

Introduction (NIPS and ICF International, 2013). Data on minimum

acceptable diet and dietary diversity of school-aged chil-
Most studies on school-aged children in Pakistan report a
dren in Pakistan is lacking. However, evidence on dietary
high prevalence of stunting, wasting, underweight, anemia,
pattern depicts an increase in consumption of junk food
iodine deficiency and vitamin A deficiency (Government
(Consumers International, 2004; Hakeem et al., 1999).
of Pakistan, 2011; Khuwaja et al., 2005; Mian et al., 2002;
Long-term imbalanced dietary intake poses a threat to
NIPS and ICF International, 2013). Dietary patterns of
children’s physical, mental, social and intellectual devel-
school-aged children across different provinces in Pakistan
opment, and predisposes them to communicable and
show a low intake of protein-rich foods and a compensatory
chronic diseases (Cheah et al., 2010; Chirwa and Ngalawa,
high consumption of affordable carbohydrates (Aziz and
2006; Kapil and Bhavna, 2002; Liu et al., 2015). School
Hosain, 2014). Differences in dietary intake of rural and
age is a time for rapid growth and development when
urban areas depict a lower consumption of meat and fruits
individual dietary habits are formed (Best et al., 2010). It
in rural children and a lower intake of vegetables in urban
creates a window of opportunity for overcoming some of
settings (Perveen et al., 2016).
Appropriate nutrition to meet the caloric and nutrient
requirements of growing age warrants attention on the Human Development Programme, Aga Khan University Karachi, Karachi,
frequency of feeding and diversity of food in daily diet Pakistan
(World Health Organization, 2008). A national survey of
Corresponding author:
children aged 6–23 months in Pakistan found that only 15% Ayesha Aziz, Human Development Programme, Aga Khan University,
met the standards of minimum acceptable diet, while 22% IED-PDC, 1-5/B, Block 7, F B Area, Karimabad, Karachi-75950, Pakistan.
fulfilled the requirements of minimum dietary diversity Email: ayesha.azi@gmail.com
232
the nutrition deficiencies set out in early years of life
(Badruddin, et al., 2008; Government of Pakistan and
UNESCO, 2010). Understanding the existing dietary
practices of school-aged children and factors influencing
their food choices is necessary for addressing and pre-
venting nutritional deficiencies.
Dietary habits of school-aged children are influenced by
various socioeconomic factors, as well as home and school
environments (Choi et al., 2008). Studies done in selected
areas of Punjab have shown a significant relationship
between malnutrition among school-aged children and low
socio-economic status, rural residence, low parental edu-
cation and large family size (Babar et al., 2010; Khan et al.,
2015; Mian et al., 2002; Mushtaq et al., 2011; Perveen
et al., 2016; Siddique et al., 2013). In Southern Sindh,
stunting was found to be more common in girls as opposed
to boys, children older than 7 years of age as opposed to
children aged 5–7 years, and children of farmers, shop-
keepers and government employees as opposed to children
of businessmen and private employees (Khuwaja et al.,
2005). With respect to indicators of children’s nutrition
status, Sindh is one of the worst affected provinces in
Pakistan, with 28% households living below the food
poverty line, but a comprehensive study of dietary practices
of school children and their correlates in Sindh is lacking
(CRPRID, 2002; Government of Pakistan, 2011). Afford-
ability, availability and access have been identified as the
key determinants of children’s diet (Aziz and Hosain, 2014;
Government of Pakistan, 2011; Perveen et al., 2016).
Strategies to address malnutrition in Pakistan must
attend to issues of availability and cost of fulfilling the
caloric and nutrient requirements of school-aged children.
Knowledge of factors influencing school children’s diet-
ary practices will help design effective strategies for
cultivating appropriate nutrition practices in school-aged
children and focus interventions on the most vulnerable
population groups. This paper aims to describe the dietary
practices of school children across eight districts of Sindh,
within the ambit of the Sindh Community Mobilization
Program (CMP).
Methodology
A school-based survey was conducted during October–
December 2016, to explore the dietary practices of school
children in eight districts of Sindh (District Khairpur,
Sukkur, Larkana, Dadu, Kambar Shahdadkot, Jacobabad,
Kashmore, and Karachi). The Sindh CMP works in colla-
boration with the Government of Sindh in 409 government
schools of the eight districts. The programme aims to
increase community involvement in improving children’s
enrolment and nutrition. The school-based dietary practices
survey was part of a pre-intervention study of school
children’s dietary practices. Assuming that CMP’s inter-
vention will contribute to a 10% change in children’s
dietary pattern, a sample size of 800 was calculated, which
was inflated to 1000 after adding the design effect and
Nutrition and Health 24(4)
adjusting for a response rate of 80%. Considering the
financial and logistic limitations of the programme, the
sample of school children was selected from 36 CMP target
schools (8 schools in Karachi and 4 schools in each of the
other seven districts). Schools in each district were selected
from a list provided to CMP by the Government of Sindh,
in consultation with the district teams and school admin-
istration. The number of schools selected from Karachi
was increased to eight as the CMP programme was being
implemented in four towns of Karachi; therefore, two
schools in each town were included in the study.
Depending on the total enrollment of each selected school,
15–45 children from grades/classes 2–5 of each school
were included in the study.1 A total of 1205 school chil-
dren and their households were approached for consent.
Twenty-four children were not allowed to participate in
the study as their households did not consent to partici-
pation in the survey. An additional 72 children were
absent on the days of data collection in the selected
schools; therefore, data was gathered from the remaining
1109 school children and their households.
The survey was based on an interviewer-completed
questionnaire consisting of 40 items that gathered data
about school children’s socio-demographic characteristics
from the parents/relatives in their households and school
children’s dietary practices from children themselves.
Questions related to socio-demographic characteristics
were adapted from the Multiple Indictor Cluster Survey of
Sindh (Sindh Bureau of Statistics and UNICEF, 2015).
Information about the weekly frequency for intake of
meals, snacks and food items was collected through the
food frequency questionnaire. The questionnaire was
translated into local languages (Urdu and Sindhi) and pre-
tested with households and children living in settings
similar to the study population.
Ethical approval for the study was acquired from the
Ethics Review Committee of the Aga Khan University,
Karachi (4415-HDP-ERC-16). Institutional consent from
schools and parental consent from households of children
included in the study sample was obtained prior to data
collection. Children’s assent was taken on the day of
data collection.
Data from the survey were analyzed using SPSS 20.0. A
descriptive analysis of children’s socio-demographic
characteristics and dietary practices was performed. Sub-
sequently, a comparative analysis was done using the
chi-square test to explore significant socio-demographic
differences in the proportion of school children skipping
meals and snacks, and eating various food items.
Results
The results provide a detailed description of socio-
demographic characteristics such as the age, gender and
rural/urban residence of school children included in the
survey. The frequency of children’s daily meals, snacks and
different food items intake was graded as 0–7 days/week, at
Aziz et al. 233

the time of data collection. For analysis and interpretation, Table 1. Household characteristics.
categorization in other studies was used as a reference and
frequency of children’s daily meals, snacks and different n (%)
food items intake was classified into categories of 0–2 Construction of housea (n ¼ 1107)
days per week, 3–5 days per week and 6–7 days per week Pakka 488 (44)
(Woodruff et al., 2010). A descriptive analysis of socio- Kacha pakka 308 (27.8)
demographic characteristics was performed for 1109 Kacha 311 (28)
children, while the descriptive and comparative analysis Ownership of house (n ¼ 1105)
for frequency of meals, snacks and food items was per- Self-owned 938 (84.9)
formed for 1017 children after exclusion of incomplete Rented 128 (11.6)
questionnaires. Landlord/relative’s houseb 39 (3.5)
Ownership of land for cultivation (n ¼ 1107)
Yes 215 (19.4)
Socio-demographic and household characteristics No 892 (80.6)
Ownership of livestock (n ¼ 1109)
Analysis of data gathered from 1109 school children and Yes 523 (47.2)
their households showed that the mean age of school No 586 (52.8)
children included in the survey was 9 years (SD +1.2). The Drinking water supply (n ¼ 1077)
study sample comprised 53.2% boys and 46.8% girls, of Ground water 533 (49.4)
whom 80% lived in rural areas and 20% resided in urban Piped/tap water 461 (42.8)
areas.2 With respect to educational level (grade/class), Purchased water 81 (7.5)
9.6% children studied in class 2, 45.8% came from class 3, Pond/river/rain water 2 (0.2)
34.1% from class 4 and 10.6% children studied in class 5. Drainage system (n ¼ 1076)
Open 629 (58.4)
Data from households was gathered from mothers,
Closed 447 (41.5)
fathers and relatives living in the same house; 78% of Toilet facility (n ¼ 1100)
mothers had received no formal or primary education, Flush toilet 666 (60.5)
while 22% reported having secondary education. Contra- Bush/no facility 233 (21.2)
rily, 62% fathers had received secondary or higher Pit latrine 201 (18.2)
education, while 38% had no formal or primary education. Family size (n ¼ 1109)
Table 1 shows the characteristics of households included 5 Members 152 (13.7)
in the survey. The total number of households for each 6–8 Members 487 (43.9)
characteristic varies according to the number of valid 9–11 Members 272 (24.5)
responses. As shown in Table 1, 71% households had a 12 Members 198 (17.9)
monthly income of Pakistani Rupees (PKR) 20,000 or less. Household monthly income (PKR) (n ¼ 952)
1000–10,000 325 (34.1)
The average monthly household expenditure on food was
11,000–20,000 353 (37)
found to be PKR 9000, while an average of PKR 1500 per 21,000–30,000 147 (15.4)
month was spent on children’s education and PKR 2000 per 31,000 127 (13.3)
month was spent on healthcare. Of the households that
a
reported their source of income, 54% identified agriculture Pakka (made of concrete and cement), kacha pakka (made of mud, hay
and/or daily wage labor as their main source of income, and mortar), kacha (made of mats or straw).
b
Landlord’s/relative’s house is different from rented, as tenants do not pay
while 7% households had multiple sources of income, cash for living there. Tenants living in a landlord’s house take care of the
which included small businesses, government or private landlord’s livestock and agricultural land, while those living in relative’s
service, domestic work and pension. house live as part of an extended family system.

nut, tobacco, slaked lime, etc.), as well as food items like

Children’s dietary practices
School children were questioned about their eating habits
by asking the frequency of their daily meal and snack
intake during the previous week. Children’s weekly intake
of different food items was explored through the food
frequency questionnaire. After excluding questionnaires
with missing data, a descriptive analysis was performed for
1017 children.
Table 2 shows school children’s weekly intake of meals,
snacks and various food items, categorized as 0–2 days per
week, 3–5 days per week and 6–7 days per week. Junk food
included non-nutritious food items such as betelnut/pan/
gutka (gutka is a locally made chewable mixture of betel
packed chips and toffees/chocolates, which have a high salt
and sugar content and produce satiety.
More than 75% children ate breakfast, lunch and dinner
regularly (6–7 days per week). However, a considerable
percentage of school children reported never or rarely
taking breakfast (0–2 days per week ¼ 10.3%), mid-
morning (0–2 days per week ¼ 28.9%) and evening
snacks (0–2 days per week ¼ 33.4%). Three-quarters of the
children reported eating potatoes on 3 days a week or more,
and more than 60% children acknowledged similar con-
sumption of milk and derivatives. However, intake of
animal and plant protein items and fruits was quite infre-
quent as more than half of the children reported rare intake
234 Nutrition and Health 24(4)

Table 2. School children’s weekly intake of meals, snacks and food items (n ¼ 1017).

0–2 days/week 3–5 days/week 6–7 days/week

Meals n (%) n (%) n (%)

Breakfast 105 (10.3) 117 (11.5) 795 (78.2)

Lunch 34 (3.3) 45 (4.4) 938 (92.2)
Dinner 21 (2.1) 56 (5.5) 940 (92.4)

Snacks 0–2 days/week n (%) 3–5 days/week n (%) 6–7 days/week n (%)

Mid-morning 294 (28.9) 173 (17) 550 (54.1)

Evening 340 (33.4) 231 (22.7) 446 (43.9)
Bed-time 481 (47.3) 90 (8.8) 446 (43.9)

Food items 0–2 days/week n (%) 3–5 days/week n (%) 6–7 days/week n (%)

Potatoes 214 (23.7) 457 (44.9) 319 (31.4)

Meat, chicken and fish 563 (55.4) 369 (36.3) 85 (8.4)
Egg 600 (59) 258 (25.4) 159 (15.6)
Milk and alternatives 376 (37) 213 (20.9) 428 (42.1)
Lentils/pulses 523 (51.4) 418 (41.1) 76 (7.5)
Nuts and seeds 658 (64.7) 210 (20.6) 149 (14.7)
Fruits 514 (50.5) 333 (32.7) 170 (16.7)
Vegetables 220 (21.6) 531 (52.2) 266 (26.2)
Junk fooda 94 (9.2) 265 (26.1) 658 (64.7)
Tea 59 (5.8) 39 (3.8) 919 (90.4)
a
Junk food included non-nutritious food items such as betelnut/gutka/pan, as well as the food items like packed chips and toffees/chocolates, which have a
high salt and sugar content and produce satiety.

(0–2 days per week) of meats, eggs, lentils, nuts and seeds,
and fruits. Vegetables were frequently a part of children’s
diet, with almost 80% reported consuming vegetables on 3
or more days a week. Junk food and tea were also fre-
quently included in children’s diet as more than 90%
children acknowledged eating junk food items and drinking
tea on 3 or more days a week.
Table 3 shows a comparative analysis of weekly
breakfast and snack intake in school children with respect
to gender, rural/urban residence, house construction and
educational level of children. A significantly higher pro-
portion of children living in rural areas and in kacha (made
of mats or straw) or kacha pakka (made of mud, hay and
mortar) houses were skipping breakfast, mid-morning and
evening snacks, as opposed to children living in urban areas
and pakka houses. A significantly higher percentage of
boys and children studying in class 5 were found to skip
mid-morning and evening snacks, as opposed to girls and
children studying in classes 2, 3 and 4. Family income and
educational status of mother and father were also included
as socio-demographic variables in the comparative analy-
sis, but they did not show a significant variation in pro-
portion of children skipping breakfast, mid-morning and
evening snacks.
School children’s weekly intake of different food items
and junk food was compared with respect to socio-
demographic variables such as gender, rural/urban resi-
dence, construction of house, educational level of children,
family income and educational status of mother and
father. Table 4 shows the comparative analysis of school
children’s animal protein and milk intake with respect to
significant socio-demographic variables. A significantly
higher proportion of girls, children living in urban areas
and pakka houses and children studying in class 2, 3 and 4
consumed animal protein food such as meat/chicken/fish
and eggs on 3 or more days per week, as opposed to their
comparative counterparts. A significantly higher percent-
age of boys and children living in rural areas reported more
frequent intake (3 or more days per week) of milk protein
items, as opposed to girls and children living in urban areas.
Table 5 shows the comparative analysis of school chil-
dren’s plant protein intake with respect to significant socio-
demographic variables. A significantly higher proportion
of children studying in class 2, 3 and 4 consumed lentils
and nuts and seeds on 3 or more days per week, as opposed
to children studying in class 5. Likewise, a significantly
higher percentage of children living in urban areas reported
eating nuts and seeds on 3 or more days per week, as
opposed to children living in rural areas.
Table 6 shows the comparative analysis of school chil-
dren’s fruit, vegetables and junk food intake with respect to
significant socio-demographic variables. A more frequent
intake of fruits was found to be common among children
living in urban areas and pakka houses, as well as in chil-
dren studying in class 2, 3 and 4, as opposed to children
living in rural areas, kacha pakka and kacha houses and
children studying in class 5. Children studying in class 2, 3
and 4 consumed vegetables more frequently as opposed to
those studying in class 5.
 Table 3. Comparative analysis of school children’s breakfast and snack intake (n ¼ 1017).

Breakfast Mid-morning snack Evening snack

0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/
Characteristics n (%) n (%) n (%) P-value n (%) n (%) n (%) P-value n (%) n (%) week n (%) P-value

Gender
Male (n ¼ 543) 52 (9.6) 57 (10.5) 434 (79.9) 0.347 170 (31.3) 102 (18.8) 271 (49.9) 0.016* 199 (36.6) 121 (22.3) 223 (41.1) 0.056
Female (n ¼ 474) 53 (11.2) 60 (12.7) 361 (76.2) 124 (26.2) 71 (15) 279 (58.9) 141 (29.7) 110 (23.2) 223 (47)
Residence
Rural (n ¼ 809) 93 (11.5) 80 (9.9) 636 (78.6) 0.001* 265 (32.8) 140 (17.3) 404 (49.9) 0.000* 306 (37.8) 180 (22.2) 323 (39.9) 0.000*
Urban (n ¼ 208) 12 (5.8) 37 (17.8) 159 (76.4) 29 (13.9) 33 (15.9) 146 (70.2) 34 (16.3) 51 (24.5) 123 (59.1)
Construction of house
Pakka (n ¼ 456) 32 (7) 63 (13.8) 361 (79.2) 0.003* 105 (23) 74 (16.2) 277 (60.7) 0.001* 126 (27.6) 99 (21.7) 231 (50.7) 0.001*
Kacha pakka 38 (13.3) 34 (11.9) 213 (74.7) 88 (30.9) 53 (18.6) 144 (50.5) 100 (35.1) 69 (24.2) 116 (40.7)
(n ¼ 285)
Kacha (n ¼ 276) 35 (12.7) 20 (7.2) 221 (80.1) 101 (36.6) 46 (16.7) 129 (46.7) 114 (41.3) 63 (22.8) 99 (35.9)
Education
Class 2 (n ¼ 97) 11 (11.3) 14 (14.4) 72 (74.2) 0.649 23 (23.7) 11 (11.3) 63 (64.9) 0.000* 29 (29.9) 15 (15.5) 53 (54.6) 0.000*
Class 3 (n ¼ 463) 45 (9.7) 59 (12.7) 359 (77.5) 110 (23.8) 82 (17.7) 271 (58.5) 152 (32.8) 112 (24.2) 199 (43)
Class 4 (n ¼ 348) 37 (10.6) 31 (8.9) 280 (80.5) 107 (30.7) 58 (16.7) 183 (52.6) 101 (29) 79 (22.7) 168 (48.3)
Class 5 (n ¼ 109) 12 (11) 13 (11.9) 84 (77.1) 54 (49.5) 22 (20.2) 33 (30.3) 58 (53.2) 25 (22.9) 26 (23.9)

Table 4. Comparative analysis of school children’s animal protein and milk intake (n ¼ 1017).

Meat/chicken/fish Eggs Milk and alternatives

0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/
Characteristics n (%) n (%) n (%) P-value n (%) n (%) n (%) P-value n (%) n (%) week n (%) P-value

Gender
Male (n ¼ 543) 320 (58.9) 184 (33.9) 39 (7.2) 0.04* 340 (62.6) 130 (23.9) 73 (13.4) 0.029* 189 (34.8) 102 (18.8) 252 (46.4) 0.01*
Female (n ¼ 474) 243 (51.3) 185 (39) 46 (9.7) 260 (54.9) 128 (27) 86 (18.1) 187 (39.5) 111 (23.4) 176 (37.1)
Residence
Rural (n ¼ 809) 469 (58) 277 (34.2) 63 (7.8) 0.004* 510 (63) 182 (22.5) 117 (14.5) 0.000* 288 (35.6) 160 (19.8) 361 (44.6) 0.005*
Urban (n ¼ 208) 94 (45.2) 92 (44.2) 22 (10.6) 90 (43.3) 76 (36.5) 42 (20.2) 88 (42.3) 53 (25.5) 67 (32.2)
Construction of house
Pakka (n ¼ 456) 221 (48.5) 191 (41.9) 44 (9.6) 0.000* 238 (52.2) 134 (29.4) 84 (18.4) 0.000* 168 (36.8) 104 (22.8) 184 (40.4) 0.42
Kacha pakka 158 (55.4) 100 (35.1) 27 (9.5) 169 (59.3) 73 (25.6) 43 (15.1) 112 (39.3) 57 (20) 116 (40.7)
(n ¼ 285)
Kacha (n ¼ 276) 184 (66.7) 78 (28.3) 14 (5.1) 193 (69.9) 51 (18.5) 32 (11.6) 96 (34.8) 52 (18.8) 128 (46.4)
Education
Class 2 (n ¼ 97) 49 (50.5) 38 (39.2) 10 (10.3) 0.000* 52 (53.6) 30 (30.9) 15 (15.5) 0.003* 33 (34) 26 (26.8) 38 (39.2) 0.319
Class 3 (n ¼ 463) 258 (55.7) 168 (36.3) 37 (8) 262 (56.6) 133 (28.7) 68 (14.7) 181 (39.1) 100 (21.6) 182 (39.3)
Class 4 (n ¼ 348) 170 (48.9) 142 (40.8) 36 (10.3) 204 (58.6) 78 (22.4) 66 (19) 123 (35.3) 70 (20.1) 155 (44.5)

235
Class 5 (n ¼ 109) 86 (78.9) 21 (19.3) 2 (1.8) 82 (75.2) 17 (15.6) 10 (9.2) 39 (35.8) 17 (15.6) 53 (48.6)
 Table 5. Comparative analysis of school children’s plant protein intake (n ¼ 1017).

236
Lentils Nuts and seeds

0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/week
Characteristics n (%) n (%) n (%) P-value n (%) n (%) n (%) P-value

Gender
Male (n ¼ 543) 285 (52.5) 217 (40) 41 (7.6) 0.729 355 (65.4) 121 (22.3) 67 (12.3) 0.054
Female (n ¼ 474) 238 (50.2) 201 (42.4) 35 (7.4) 303 (63.9) 89 (18.8) 82 (17.3)
Residence
Rural (n ¼ 809) 420 (51.9) 328 (40.5) 61 (7.5) 0.776 553 (68.4) 155 (19.2) 101 (12.5) 0.000*
Urban (n ¼ 208) 103 (49.5) 90 (43.3) 15 (7.2) 105 (50.5) 55 (26.4) 48 (23.1)
Construction of house
Pakka (n ¼ 456) 232 (50.9) 182 (39.9) 42 (9.2) 0.062 278 (61) 97 (21.3) 81 (17.8) 0.057
Kacha pakka (n ¼ 285) 135 (47.4) 133 (46.7) 17 (6.0) 186 (65.3) 60 (21.1) 39 (13.7)
Kacha (n ¼ 276) 156 (56.5) 103 (37.3) 17 (6.2) 194 (70.3) 53 (19.2) 29 (10.5)
Education
Class 2 (n ¼ 97) 49 (50.5) 34 (35.1) 14 (14.4) 0.000* 50 (51.5) 30 (30.9) 17 (17.5) 0.000*
Class 3 (n ¼ 463) 248 (53.6) 180 (38.9) 35 (7.6) 322 (69.5) 85 (18.4) 56 (12.1)
Class 4 (n ¼ 348) 152 (43.7) 170 (48.9) 26 (7.5) 200 (57.5) 81 (23.3) 67 (19.3)
Class 5 (n ¼ 109) 74 (67.9) 34 (31.2) 1 (0.9) 86 (78.9) 14 (12.8) 9 (8.3)

Table 6. Comparative analysis of school children’s fruits, vegetables and junk food Intake (n ¼ 1017).

Fruits Vegetables Junk fooda

0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/week 0–2 days/week 3–5 days/week 6–7 days/week
Characteristics n (%) n (%) n (%) P-value n (%) n (%) n (%) P-value n (%) n (%) n (%) P-value

Gender
Male (n ¼ 543) 285 (52.5) 176 (32.4) 82 (15.1) 0.256 112 (20.6) 278 (51.2) 153 (28.2) 0.273 65 (12) 160 (29.5) 318 (58.6) 0.000*
Female (n ¼ 474) 229 (48.3) 157 (33.1) 88 (18.6) 108 (22.8) 253 (53.4) 113 (23.8) 29 (6.1) 105 (22.2) 340 (71.7)
Residence
Rural (n ¼ 809) 440 (54.4) 247 (30.5) 122 (15.1) 0.000* 174 (21.5) 424 (52.4) 211 (26.1) 0.967 89 (11) 221 (27.3) 499 (61.7) 0.000*
Urban (n ¼ 208) 74 (35.6) 86 (41.3) 48 (23.1) 46 (22.1) 107 (51.4) 55 (26.4) 5 (2.4) 44 (21.2) 159 (76.4)
Construction of house
Pakka (n ¼ 456) 204 (44.7) 157 (34.4) 95 (20.8) 0.001* 97 (21.3) 246 (53.9) 113 (24.8) 0.814 25 (5.5) 123 (27) 308 (67.5) 0.000*
Kacha Pakka 145 (50.9) 94 (33) 46 (16.1) 65 (22.8) 141 (49.5) 79 (27.7) 26 (9.1) 74 (26) 185 (64.9)
(n ¼ 285)
Kacha (n ¼ 276) 165 (59.8) 82 (29.7) 29 (10.5) 58 (21) 144 (52.2) 74 (26.8) 43 (15.6) 68 (24.6) 165 (59.8)
Education
Class 2 (n ¼ 97) 42 (43.3) 38 (39.2) 17 (17.5) 0.001* 19 (19.6) 48 (49.5) 30 (30.9) 0.009* 12 (12.4) 25 (25.8) 60 (61.9) 0.017*
Class 3 (n ¼ 463) 238 (51.4) 155 (33.5) 70 (15.1) 109 (23.5) 216 (46.7) 138 (29.8) 36 (7.8) 109 (23.5) 318 (68.7)
Class 4 (n ¼ 348) 163 (46.8) 108 (31) 77 (22.1) 65 (18.7) 204 (58.6) 79 (22.7) 28 (8) 97 (27.9) 223 (64.1)
Class 5 (n ¼ 109) 71 (65.1) 32 (29.4) 6 (5.5) 27 (24.8) 63 (57.8) 19 (17.4) 18 (16.5) 34 (31.2) 57 (52.3)
a
Junk food included non-nutritious food items such as betelnut/gutka/pan, as well as the food items like packed chips and toffees/chocolates, which have a high salt and sugar content and produce satiety.
Aziz et al.
Significantly higher percentages of girls, children living
in urban areas and pakka houses, and children studying in
class 2, 3 and 4 reported consuming junk food items such as
candy, packed chips and betel nuts/betel leaves/gutka on 3
or more days per week, as opposed to boys, children living
in rural areas, kacha pakka and kacha houses, and children
studying in class 5. Among those who frequently consumed
one or more junk food items, more girls acknowledged
frequent consumption of candy and packed chips, whereas
a significantly higher percentage of boys ate betel nuts/
betel leaves/gutka.
Discussion
The findings of this study highlight the dietary practices of
school children in eight districts of Sindh, with particular
reference to weekly frequency of meals, snacks and dif-
ferent food items. We found that more than 75% of chil-
dren ate breakfast, lunch and dinner regularly. However,
10% of children skipped breakfast and nearly one-third
skipped mid-morning and evening snacks. Similar pro-
portions of school-aged children skipping breakfast and
mid-morning snacks have been reported in other studies in
Sindh (Paracha et al., 2016; USAID, 2014). Reflecting on
the reasons for skipping breakfast, mid-morning and
evening snacks is important, as intake of three meals and
two to three snacks per day is recommended to meet the
caloric and nutrient requirements of growing age children
(Kleinman and Greer, 2014).
Our study found that skipping breakfast and snacks was
related to low socio-economic status and rural residence
as a significantly higher percentage of children living in
rural areas and kacha or kacha pakka houses skipped
breakfast and mid-morning and evening snacks. Previous
studies on school-aged children in Punjab and Sindh have
also highlighted the association of children’s under-
nutrition with rural residence and low socio-economic
status (Khan et al., 2015; Khuwaja et al., 2005; Mushtaq
et al., 2011; Parveen et al., 2016). A deeper investigation
of the processes and factors affecting regular intake of
meals and snacks is necessary as school timings, distance
between home and school, mode of transportation, avail-
ability of snacks and recess time in schools could mediate
the effect of rural residence and socio-economic status on
intake of meals and snacks.
The quantity and quality of food items consumed in
meals and snacks determines the caloric and nutrient
intake in children’s diet and is known to impact upon their
satiety, eating patterns and nutrition status (Fayet-Moore
et al., 2017; Mahan et al., 2012). Our analysis showed
that 3.8% of the overall sample (37% of the 105 children
who skipped breakfast) skipped both breakfast and
mid-morning snack, while 28% of those who consumed
breakfast on 3 or more days a week skipped mid-morning
snack. These findings call for further investigation of the
socio-economic circumstances of children who skip both
breakfast and mid-morning snack and the adequacy of
237
their caloric and nutrient intake. In addition, they imply
that children consuming regular breakfast may be eating
food items that produce satiety and cause them to skip
mid-morning snacks.
Studies of snacking trends in developed countries show
that snack intake is not related to hunger and accounts for
more than one-quarter of children’s daily energy intake
(Hess et al., 2016). Our findings revealed that snacking
trends among children from low socio-economic back-
ground and rural areas of Sindh are poor. As our study
focused on weekly frequency of snacks we were unable to
discern the contribution of snacks to children’s daily energy
intake, but regular or frequent skipping of snacks signifies
lack of food diversity which may result in micronutrient
deficiencies. The value of consuming nutrient-dense foods
as snacks cannot be overemphasized, not only to meet the
energy requirements of growing age children, but to main-
tain a consistent supply of vitamins and minerals in their
bodies (Fayet-Moore et al., 2017). School-aged children in
Pakistan are known to have macro and micronutrient defi-
ciencies (Government of Pakistan, 2011; Khuwaja et al.,
2005; Mian et al., 2002; NIPS and ICF International, 2013).
The evidence of meal and snack intake from our study
highlights the need to develop school-based nutrition edu-
cation programmes and to design behavior change commu-
nication strategies focusing on regular intake of meals, as
well as nutrient-rich snacks to prevent and address their
vitamin and mineral deficiencies.
Our analysis of children’s intake of different protein-
rich food items and fruits depicted that urban residence
and better socio-economic status have a positive effect on
the frequency and variety of daily protein and fruit intake
as significantly higher percentages of children living in
urban areas and pakka houses (made of concrete and
cement) ate meats, eggs, nuts and seeds and fruits on 3 or
more days per week, as opposed to children living in rural
areas and kacha or kacha pakka houses. These findings are
similar to the feeding patterns of young children across
Pakistan as wealth and urban residence were found to have
a positive impact on the frequency and diversity of their
daily diet (NIPS and ICF International, 2013). This
highlights the need for developing nutrition-sensitive and
specific programmes for lower socio-economic strata and
residents of rural areas.
Mother’s education has been related to better intake of
milk and dietary diversity in young children (Khan et al.,
2013). Our results showed that intake of milk was more
frequent among boys and children living in rural areas,
irrespective of socio-economic status. This implies that
availability and access have a significant effect on food
choices. Though affordability has been found to be a key
determinant of dietary practices in school aged children, it
may be argued on the basis of previous evidence that
increasing knowledge about healthy food choices that are
available in the local market and should be accessible to
both girls and boys will improve children’s dietary
238
practices and their nutritional status (Aziz and Hosain,
2014; Khan et al., 2013; NIPS and ICF International, 2013).
Most studies in Pakistan have focused on children under
5 years, while the food choices and nutritional needs of
school-aged children have been ignored (Best et al., 2010).
Meal frequency and dietary diversity questionnaires have
been used to monitor dietary trends and as proxy indicators
for appropriate caloric and nutrient intake in children aged
6–23 months (NIPS and ICF International, 2013; World
Health Organization, 2008). However, a standard guideline
for assessing meal frequency and dietary diversity in
school-aged children is missing. Dietary studies of school-
aged children in Pakistan depict low intake of protein-rich
foods, fruits and vegetables (Aziz and Hosain, 2014;
Parveen et al., 2016). Our study adds depth to this
knowledge by showing that milk and its derivatives are
the most frequently consumed source of protein, whereas
meats and lentils are the least regularly consumed sources
of protein. Eggs, nuts and seeds are also frequently skip-
ped in the diet of school children. Consumption of fruits is
more infrequent as opposed to vegetables. These findings
can help design specific behavior change messages for
school-aged children to encourage consumption of the
easily available, accessible and affordable protein- and
vitamin-rich foods such as lentils and dried seeds of
pumpkin, melons and sunflower. Consumption of mineral
-and vitamin-rich seasonal fruits and vegetables, as well
as milk and its derivatives can also be promoted by sup-
porting the agrarian economy and encouraging utilization
of local products to boost immunity and reduce chances of
chronic diseases.
The study results also depict that significantly higher
percentages of children studying in class 2, 3 and 4 con-
sumed different kinds of protein-rich food items, fruits and
vegetables more frequently as opposed to children studying
in class 5. On the other hand, significantly higher percen-
tages of children living in urban areas and pakka houses
acknowledged frequent consumption of junk food items
such as candy, packed chips and betel nuts/betel leaves/
gutka. More girls ate candy and packed chips, whereas a
significantly higher percentage of boys ate betel nuts/betel
leaves/gutka. It may be implied that increasing age has an
adverse effect on the variety of food in children’s diet,
while a better socio-economic status and multiple expo-
sures in the urban environment promote frequent con-
sumption of junk food. Increased consumption of junk food
has been related to urban environments, easy availability,
mass advertising, children’s preference for spices and lack
of parental control in Pakistan and other developing
countries (Alimi, 2016; Consumers International, 2004;
Hakeem et al., 1999). It is essential for parents and elders in
Pakistani families to pay attention to children’s nutritional
needs and dietary diversity, while guiding their food
choices during changes in age, socio-economic status and
rapidly urbanizing environments.
A major limitation of this study is that it provides
insight about the dietary practices and related socio-
Nutrition and Health 24(4)
demographic variables in 36 consultatively selected
schools of eight districts in Sindh. Large-scale represen-
tative studies based on standardized meal frequency and
dietary diversity guidelines for school-aged children
across Pakistan are needed to validate the relationship
between meal frequency, dietary diversity and children’s
nutrition status. Exploration of school children’s food
choices and the factors and processes affecting regular
intake of meals and snacks is also worth exploring through
qualitative models. This study lays a foundation for
research on dietary practices of school children in Paki-
stan and delineates the various key messages and strate-
gies that must be emphasized in nutrition intervention
programs in Sindh.
Acknowledgements
We would like to take this opportunity to express our pro-
found gratitude to our team members, namely Mr Rafique
Tunio, Mr Abid Khoso, Ms Ghazala Raheel, Ms Saina Ali
and Ms Gul Afshan for contributing to the data collection,
entry and cleaning process as part of the Community
Mobilization Programme at the Human Development Pro-
gramme, Aga Khan University, Karachi. We are also
thankful to Ms Kausar S. Khan from the Community
Health Sciences Department of Aga Khan University,
Karachi for providing guidance during the design of this
study. We would also like to extend our regards to our
collaborators, the International Relief and Development
(IRD), the Indus Resource Centre (IRC) and the Health
and Nutrition Development Society (HANDS). Data col-
lection for this study would not have been possible with-
out their assistance and cooperation. Last but not the least,
we are immensely grateful to the school children, their
families and the school teachers for the time and responses
provided for this research.
Author contributions
AA conceptualized and wrote the manuscript. MP contrib-
uted to the literature search, data analysis and introduction of
manuscript. AK conducted the analysis of data set and con-
tributed to interpretation of results. AZK and GR contributed
to the design of this study and reviewed the manuscript.
Availability of data and material
The survey questionnaire and data gathered for the survey
can be made available on request to the corresponding
author.
Consent for publication and ethical approval
Ethical approval for the study was acquired from the Ethics
Review Committee of the Aga Khan University, Karachi
(4415-HDP-ERC-16). Informed consent for data collection
and publication was obtained from parents and school
administration, and an assent form was signed by the school
children included in the study.
Aziz et al.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest
with respect to the research, authorship, and/or publication
of this article.
Funding
The author(s) disclosed receipt of the following financial
support for the research, authorship, and/or publication of
this article: This study was funded by the United States
Agency for International Development. Grant number
391-C-13-00006.
ORCID iD
Ayesha Aziz http://orcid.org/0000-0003-4384-9605
Notes
1. In Pakistan, schools have classes that are equivalent to grades
in schools of most other countries. Primary schools have chil-
dren studying in elementary grades that are commonly referred
to as class 1, 2, 3, 4 and 5.
2. The Government of Sindh has classified areas and schools
located therein as rural and urban. This classification was used
as a basis for categorization of rural/urban residence in this
study. Districts of Karachi, Sukkur and Jacobabad had both
rural and urban areas, while all other districts included in the
study were rural.
References
Alimi BA (2016) Risk factors in street food practices in develop-
ing countries: A review. Food Science and Human Wellness
5(3): 141–148.
Aziz S and Hosain K (2014) Carbohydrate (CHO), protein and fat
intake of healthy Pakistani school children in a 24 hour period.
Journal of Pakistan Medical Association 64(11): 1255–1259.
Babar NF, Muzaffar R, Khan MA, et al. (2010) Impact of socio-
economic factors on nutritional status in primary school chil-
dren. Journal of Ayub Medical College 22(4): 15–18.
Badruddin SH, Agha A, Peermohamed H, et al. (2008) Tawana
project-school nutrition program in Pakistan – its success, bot-
tlenecks and lessons learned. Asia Pacific Journal of Clinical
Nutrition 17(Suppl 1): 357–360.
Best C, Neufingerl N, van Geel L, et al. (2010) The nutritional
status of school-aged children: Why should we care? Food and
Nutrition Bulletin 31(3): 400–417.
Cheah WL, Muda WW and Zamh Z-H (2010) A structural equa-
tion model of the determinants of malnutrition among children
in rural Kelantan, Malaysia. Rural and Remote Health 10(1):
1248.
Chirwa EW and Ngalawa H (2006) Determinants of Child
Nutrition in Malawi. Working Paper No. 2006/01. Malawi:
Department of Economics, University of Malawi.
Choi E, Shin N, Jung E, et al. (2008). A study on nutrition knowl-
edge and dietary behavior of elementary school children in
Seoul. Nutrition Research and Practice 2(4): 308–316.
Consumers International (2004) The junk food generation: A
multi-country survey of the influence of television advertise-
ments on children. Kuala Lumpur, Malaysia: Consumers Inter-
national Asia Pacific Office.
239
CRPRID (2002) Pakistan Human Condition Report 2002. Islama-
bad: Center of Reduction on Poverty Reduction and Income
Distribution (CRPRID).
Fayet-Moore F, Peters V and McConnell A, et al. (2017) Week-
day snacking prevalence, frequency, and energy contribution
have increased while foods consumed during snacking have
shifted among Australian children and adolescents: 1995, 2007
and 2011-12 National Nutrition Surveys. Nutrition Journal 16:
65. Avaiable at: http://doi.org/10.1186/s12937-017-0288-8.
Government of Pakistan (2011) National Nutrition Survey 2011.
Islamabad, Pakistan: Planning Commission, Planning and
Development Division, Government of Pakistan.
Government of Pakistan and UNESCO (2010) School Health
Programme: A Strategic Approach for Improving Health &
Education in Pakistan. Islamabad, Pakistan: Ministry of Edu-
cation, Curriculum Wing, Government of Pakistan and United
Nations Educational, Scientific and Cultural Organization
(UNESCO).
Hakeem R, Thomas J and Badruddin SH (1999) Rural-urban dif-
ferences in food and nutrient intake of Pakistani children.
Journal of Pakistan Medical Association 49(12): 288–294.
Hess JM, Jonnalagadda SS and Slavin JL (2016) What is a snack,
why do we snack, and how can we choose better snacks? A
review of the definitions of snacking, motivations to snack,
contributions to dietary intake, and recommendations for
improvement. Advances in Nutrition 7: 466–475.
Kapil U and Bhavna A (2002) Adverse effects of poor micronu-
trient status during childhood and adolescence. Nutrition
Reviews 60(5): S84–S90.
Khan AZ, Rafique G, Qureshi H, et al. (2013) A nutrition educa-
tion intervention to combat undernutrition: Experience from a
developing country. International Scholarly Research Notices
Nutrition 2013: 210287.
Khan T, Khan RE and Raza M (2015) Gender analysis of
malnutrition: A case study of school-going children in
Bahawalpur. Asian Development Policy Review 3(2): 29–48.
Khuwaja S, Selwyn B and Shah S (2005) Prevalence and correlates
of stunting among primary school children in rural areas of
southern Pakistan. Journal of Tropical Pediatrics 51(2): 72–77.
Kleiman RE and Greer FR (eds) (2014) Pediatric Nutrition, 7th
Edition. Elk Grove Village, IL: American Academy of
Pediatrics.
Liu L, Oza S, Hogan D, et al. (2015) Global, Regional and
national causes of child mortality in 2000-13, with projections
to inform post-2015 priorities: An updated systematic analysis.
Lancet 9966: 430–440.
Mahan KL, Escott-Stump S and Raymond JL (eds) (2012)
Krause’s Food and the Nutrition Care Process, 13th Edition.
St. Louis, MO: Elsevier/Saunders.
Mian R, Ali M, Ferroni P, et al. (2002) The nutritional status of
school-aged children in an urban squatter settlement in Paki-
stan. Pakistan Journal of Nutrition 1(3): 121–123.
Mushtaq MU, Gull S, Khurshid U, et al. (2011). Prevalence and
socio-demographic correlates of stunting and thinness among
Pakistani primary school children. BioMed Central Public
Health, 11:790.
NIPS and ICF International (2013) Pakistan Demographic and
Health Survey 2012-13. Islamabad, Pakistan, and Calverton,
Maryland, USA: National Institute of Population Studies
(NIPS) [Pakistan] and ICF International.
Paracha PI, Bakht S, Paracha SI, et al. (2016) Nutritional status,
dietary practices and physical activities of adolescents in
240
public and private schools of Karachi, Pakistan. Obesity
Research 3(2): 30–39.
Parveen R, Ashiq Z, Naz S, et al. (2016) Evaluation and com-
parison of dietary habits of rural and urban areas school chil-
dren: A case study. Pakistan Journal of Food Sciences 26(1):
1–9.
Siddique S, Ayub M, Shore N, et al. (2013) Nutitional status of
primary school children in Abbottabad. Journal of Ayub Med-
ical College 25(1–2): 123–126.
Sindh Bureau of Statistics and UNICEF (2015) Sindh Multiple
Indicator Cluster Survey 2014, Final Report. Karachi, Paki-
stan: Sindh Bureau of Statistics and UNICEF.
Nutrition and Health 24(4)
USAID (2014) Sindh Nutrition and Health Practices Survey Plan-
ning Report. Sindh Community Mobilization Program (CMP)
and the United States Agency for International Development
(USAID).
Woodruff SJ, Hanning RM and McGoldrick K, et al. (2010).
Healthy eating index-C is positively associated with family
dinner frequency among students in grades 6–8 from Southern
Ontario, Canada. European Journal of Clinical Nutrition 64:
454–460.
World Health Organization (2008) Indicators for Assessing Infant
and Young Child Feeding Practices. Geneva, Switzerland:
World Health Organization.