Ganle et al. BMC Public Health         (2019) 19:1561 
https://doi.org/10.1186/s12889-019-7898-3
RESEARCH ARTICLE Open Access
Childhood obesity in urban Ghana:
evidence from a cross-sectional survey of
in-school children aged 5–16 years
John Kuumuori Ganle1,2* , Priscilla Pokuaa Boakye1 and Leonard Baatiema3
Abstract
Background: Childhood obesity is a growing public health concern in many low-income urban settings; but its
determinants are not clear. The purpose of this study is to assess the prevalence of childhood obesity and
associated factors among in-school children aged 5–16 years in a Metropolitan district of Ghana.
Methods: A cross-sectional quantitative survey was conducted among a sample of 285 in-school children aged 5–
16 years. Pre-tested questionnaires and anthropometric data collection methods were used to collect data.
Descriptive, bivariate, binary and multivariate logistic regression statistical techniques were used to analyse data.
Results: Some 46.9% (42.2% for males and 51.7% for females) of the children were overweight. Of this, 21.2% were
obese (BMI falls above 95th percentile). Childhood obesity was higher in private school (26.8%) than public school
(21.4%), and among girls (27.2%) than boys (19%). Factors that increased obesity risks included being aged 11–16 as
against 5–10 years (aOR = 6.07; 95%CI = 1.17–31.45; p = 0.025), having a father whose highest education is
‘secondary’ (aOR =2.97; 95% CI = 1.09–8.08; p = 0.032), or ‘tertiary’ (aOR = 3.46; 95% CI = 1.27–9.42; p = 0.015), and
consumption of fizzy drinks most days of the week (aOR = 2.84; 95% CI = 1.24–6.52; p = 0.014). Factors that lowered
obesity risks included engaging in sport at least 3times per week (aOR = 0.56; 95% CI = 0.33–0.96; p = 0.034), and
sleeping for more than 8 h per day (aOR = 0.38; 95% CI = 0.19–0.79; p = 0.009).
Conclusion: Higher parental (father) educational attainment and frequent consumption of fizzy drinks per week
may increase obesity risks among in-school children aged 5–16 years in the Metropolitan district of Ghana.
However, regular exercise (playing sport at least 3 times per week) and having 8 or more hours of sleep per day
could lower obesity risks in the same population. Age and sex-appropriate community and school-based
interventions are needed to promote healthy diet selection and consumption, physical activity and healthy life
styles among in-school children.
Keywords: Childhood obesity, BMI, In-school children, Lifestyle, Dietary behaviour, Sedentary behaviour, Ghana
Background sedentary lifestyle, physical inactivity and poor eating
Childhood obesity is one of the most important child habits, including consumption of savoury foods with
and public health issues in many parts of the world hidden fats and sugars that impair metabolism [1, 2].
today [1]. In simple terms, obesity refers to abnormal or Other factors include biophysiological causes such as
excessive fat accumulation resulting from energy imbal- genetic causes, insulin resistance, hyperinsulinism,
ance between calories consumed and calories expended and disruption of the normal satiety feedback mech-
[1]. Several factors contribute to obesity. These include anisms [1, 2].
Globally, obesity prevalence has nearly tripled since
* Correspondence: jganle@ug.edu.gh 1970s [1]. For instance, about 1.9 billion adults (18+
1Department of Population,Family and Reproductive Health, School of Public years) were overweight in 2016, out of which 650 million
Health, University of Ghana, P.O. Box LG 13, Legon, Accra, Ghana
2Stellenbosch Institute for Advanced Study (STIAS), Wallenberg Research were obese [1]. In the same year, 41 million children
Centre at Stellenbosch University, Stellenbosch 7600, South Africa under age five were overweight or obese [1]. Among
Full list of author information is available at the end of the article
© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Ganle et al. BMC Public Health         (2019) 19:1561 Page 2 of 12
children and adolescents aged 5–19, over 340 million children and adolescents [3]. In addition, while studies
were overweight or obese in 2016 [1]. Evidence further have identified various factors that increase the risk of
suggests that low-income countries harbor majority of childhood obesity, including age [15–17], sex [14, 16,
obese people [1]. For instance, nearly half of overweight/ 18], educational level of parent/guardian [14, 19], as well
obese children under-five are in Asia, while one quarter as unhealthy diet and physical inactivity [20–24], it is
live in Africa [1]. In Africa in particular, the number of not entirely clear whether these factors are implicated in
overweight or obese children has nearly doubled from Ghana’s obesity epidemic. In order to prevent childhood
5.4 million to 10.6 million in in 1990 and 2016 respect- obesity, there is need for context-specific studies to esti-
ively [1]. mate prevalence and identify risk factors [25]. As Guh
In Ghana, nearly 43% of the adult population is either et al. have argued, one important step in preventing
overweight or obese [3]. Additionally, 45.6% of adult dia- childhood obesity with its associated problems lies in
betes patients in Ghana are either overweight or obese comprehensive studies of its prevalence and associated
[3]. Further, recent studies suggest higher overweight/ factors [26]. This study contributes to filling this import-
obesity prevalence among women compared to men [4, ant knowledge gap by examining prevalence of obesity
5]. For instance, Benkeser, Biritwum and Hill [4] found and associated factors among in-school children in a
that 64.9% of Ghanaian women aged 18+ years in the metropolitan district of Ghana.
Accra Metropolis were either overweight or obese. Im-
portantly, emerging research suggests high prevalence of Methods
childhood obesity in Ghana [6–8]. For example, one Study design
study among 5–14 year old children found 17.4% of A cross-sectional school-based quantitative survey was
them to be obese [6]. conducted among children aged 5–16 in the Tema
In current literature, there is recognition of the fact Metropolitan District of the Greater Accra region of
that many obese adults developed the condition in child- Ghana. The design, implementation and reporting of re-
hood and adolescence [1]. Childhood obesity also has sults followed the Strengthening the Reporting of Obser-
additional consequences, including higher risks of pre- vational Studies in Epidemiology (STROBE) checklist for
mature death and disability in adulthood, increased risk cross-sectional studies.
of fractures, increased future risks of breathing difficul-
ties, heart disease, hepatic impairment, diabetes, insulin Study context
resistance, vision problems, cancer, and psychological Empirical data collection was done in two of the largest
consequences such as low self-esteem [3, 9, 10]. The dif- basic schools – one public and the other private - in the
ficulty in treating childhood obesity and the social and Tema Metropolitan district, 30 km East of Accra, the
economic burden of managing the condition are other Capital City of Ghana. The metropolis is the second
consequences of childhood obesity [11]. In relation to most populous metropolis in the Greater Accra Region.
the economic cost of childhood obesity for instance, one It has a population of 403,934, with nearly everyone liv-
study found a $14.1 billion annual cost for additional ing in urban localities [27]. The proportion of children
prescription drug, emergency room, and outpatient visit aged 5–16 is estimated to be 29.4% of the total popula-
healthcare costs annually [12]. Also, an obese 10 year old tion of the metropolis [27]. There are about 338 basic
child who maintains weight gain throughout adulthood schools (primary and junior high schools), of which 185
has a lifetime medical costs of $19,000 higher compared are private and 153 are public schools. The two schools
to a healthy-weight 10-year old who maintains a normal that were purposively chosen for this research were
weight throughout life [13]. When compared with a nor- some of the largest in terms of student numbers. The
mal weight child, a child who is obese for two consecu- number of students in the private and public schools
tive years has a $194 higher outpatient visit expenditure, that were studied were 320 and 610 respectively.
a $114 higher prescription drug expenditure, and a $12
higher emergency room expenditure [12]. Study population
While the evidence on childhood obesity and the po- We included children from the two selected basic
tential adverse effects in low-income settings is growing, schools who were aged 5-16 years.We however excluded
there is still a general paucity of data on prevalence of children within the 5-16 year age bracket who had some
childhood obesity in many African settings [14]. Specif- form of physical disability that could not allow accurate
ically, there is currently limited data on prevalence of determination of their true height.
obesity in children from 5 to 15 years in many low-
income settings [1]. In Ghana for example, a recent sys- Sample size and sampling
tematic view on overweight and obesity epidemic in One recent cross-sectional study among 218 in-school
Ghana highlighted the relative lack of focus on young children in northern Ghana reported childhood obesity
Ganle et al. BMC Public Health         (2019) 19:1561 Page 3 of 12
prevalence of 17.4% [6]. Based on this prevalence, and Data collection
assuming a 95% confidence level, alpha 0.05, 5% worst The second author (a specialist child health nurse) and
acceptable margin of error, and power of 80%, we es- two trained research assistants collected the data. Two
timated a minimum sample of 221 using Cochran’s kinds of data were collected. First, questionnaires were
sample size estimation formula for cross-sectional used to collect information on socio-demographic, diet-
studies [28]. We adjusted the minimum sample size ary, and physical activity characteristics of the children.
upward by 30% to account for non-response and as The questionnaire was developed specifically for the
well increase the power of the study. The final sample purposes of this study based on extensive literature re-
size was thus 287. view and expert consultation in Ghana (see Add-
We used a multistage sampling procedure to select itional file 1). It was however pretested on a total of 30
qualified children. In the first stage we obtained registers children aged 5–16 in two other schools not included in
of all students in each of the two schools. We screened the actual study. We tested the reliability of the instru-
each register to identify children who met our inclusion ment and found it to be reliable (Cronbach’s alpha coef-
criteria. In total, 737 children (189 private and 548 pub- ficient range = 0.80–0.92). The level of reliability
lic school) from the two basic schools met our inclusion observed for both individual items and the entire data
criteria. In stage two, we allocated our total sample of collection tool is considered in literature to be good
287 proportionate to the size of the population of chil- [29]. Details of the test-retest properties are provided in
dren aged 5–16 years in each school. This was to ensure Additional file 2.
that the sample for each school was commensurate with Second, anthropometric information such as the
the size of the population of eligible children. This weight and height of each child was collected to enable
yielded 191 children from the public school and 94 from calculation of body mass index (BMI) and obesity status.
the private school. In stage three, we entered the names The anthropometric measurement tools employed were
of all eligible children from each school into excel a Weighing Scale (Tanita WB-3000 Digital Doctor Scale,
spreadsheet and gave each a unique number (e.g. 001– Tanita Corporation of America, Inc., Illinois, USA) to
548 for the public school, and 001–189 for the private measure the weight of each child in kilograms as they
school). The lists of all the numbered children was then stood on it on a flat hard surface, and a height rod (HR-
exported into a google-based random number generator 200 Tanita Wall-Mounted Height Rod, Tanita Corpor-
software, and the required number of children (191 for ation of America, Inc., Illinois, USA) to measure the
public school and 94 for private school) for each school height of children in centimeters.
was randomly selected. We matched the randomly se- All data collection was done in the schools during
lected numbers to the corresponding names on our list break hours. Children were interviewed one at a time in
of eligible children. In stage four, the research team vis- a specially designated office where maximum confidenti-
ited the schools to meet all selected children. They were ality was assured. Interviews were done in English and
told about the purpose of the study, how they were se- three other local dialects (Ga/Dangme, Ewe and Twi) de-
lected, and what the study procedures involved. All ini- pending on each child’s preference.
tial questions were addressed in these meetings. As the
children were below the age of legal consent (i.e. 18 Measurement
years), we wrote personalised letters to their parents/ The outcome variable of the study is childhood obesity.
guardians. The letters explained the purpose of the Similar to previous studies, we used BMI count as a
study, how the children were selected and what the marker of obesity [11]. While adult BMI is easily calcu-
study procedures were. Information sheets giving further lated by dividing body weight in kilograms by height in
details about the study, including information about eth- meters squared, for children and teenagers, BMI requires
ical approval, rights of their child, informed consent, and tuning for age and sex [1]. We used the WHO Anthro-
contact details of the researchers were included with Plus software (version 1.0.4) for calculating BMI for chil-
each letter. Each child delivered the letter to their par- dren/adolescents aged 5–20 years. BMIs calculated using
ent/guardian. Each parent/guardian was given 1 week to the WHO AntroPlus were then compared with the CDC
decide on their child’s participation in the study. After curves. Any child with BMI >95th age-sex percentile was
this period, parents/guardians who consented to their considered obese [30]. Our final outcome was dichoto-
child’s participation were directed to sign or thumbprint mized into ‘obese’ and ‘not obese’.
the consent form and return it to their child. Only one A number of potential covariates were also measured
child did not receive parental consent to participate in using the questionnaires, including socio-demographic
the study and was subsequently dropped. All children characteristics like age, sex, educational level of parent/
who received parental consent were required to assent guardian, religion, occupation of parents and obesity his-
to their parents’ consent. All gave their assent. tory of child’s family; dietary/behavioural factors such as
Ganle et al. BMC Public Health         (2019) 19:1561 Page 4 of 12
consumption of processed foods and fizzy drinks; and intercept term (i.e. probability of a child being obese if
physical activity such as sports, means of transport to all the explanatory variables were equal to zero), ß1 to
school, sleeping hours and playing of computer games ßn = explanatory variables’ coefficients holding all other
other than outdoor games. In particular, sports activity variables constant, and ui = random error term.
was defined as engaging in any of the following: playing
football, basketball, tennis, volley ball and ampe (a sim- Data processing and statistical analysis
ple jumping game played by school-aged children - Following completion of data collection, all question-
mostly girls - in Ghana and neighbouring countries and naires were first manually examined to check for com-
usually involving two or more players and requires no pleteness. Questionnaires were then hand-coded and
equipment), as well as running and cycling. Fizzy drinks entered separately into Epi info version 7 by two re-
were defined as non-alcoholic soft drinks that contain search assistants. The two data entries were compared,
carbonated water, a sweetener (sweetener may be sugar, and all data entry errors or inconsistencies were dis-
high-fructose corn syrup, fruit juice, a sugar substitute, cussed and resolved with the two research assistants
or some combination of these), and a natural or artificial who performed the data entry. Following from this, a
flavouring. single database was created, agreed upon, and imported
into STATA software for analysis.
Model specification Both descriptive and inferential statistical analyses were
In this study, childhood obesity (Y) is the response vari- done. For the descriptive analysis, frequency distribu-
able with two binary outcomes: Y = 1, when a child is tions and proportions were used to summarise categor-
obese (BMI falls above 95th percentile), and Y = 0, when ical variables. Mean and range were computed to
a child is not obese (BMI falls below 95th percentile). summarise continuous variables.
The independent variables for the response variable, Y, For the inferential statistical analysis, both Chi-square
are the socio-demographic, dietary and behavioural fac- test of independence and fisher’s exact test (for observa-
tors. If Y is the response variable, the probability that a tions with less than 5 cell counts) were performed to
child will be obese is (Yi) and the alternative other out- first to assess association between childhood obesity and
come is (1-Yi). Therefore, the odds ratio in favour of a independent variables. This was followed by binary and
child being obese is

  multivariable logistic regression analysis to estimate odd
Y ratios for factors that showed statistical association at
OR ¼ i ¼ β0 þ βiXi ð1Þ1−Y the bivariate level. A 95% confidence level and statisticali
significance of p < 0.05 were assumed in the regression
Taking the natural log of the odds ratio gives the logit analysis.
model:

 
Y Results
ln i ¼ β0 þ βiXi þ ui ð2Þ1−Y Characteristics of respondentsi
A total of 286 children were surveyed; one questionnaire
Where, ‘ln’ is the natural logarithm, Yi is the probability was missing, hence 285 were used for the analysis.
that a child will be obese, (1 − Y i) is the otherwise prob- Table 1 shows the background characteristics of respon-
ability, β0 denotes the intercept parameter, Xi denotes dents. Mean age was 11.27(SD = + 4.73), and a little over
the explanatory variables, βi denotes the coefficients to half (50.5%) were female. Majority (57.5%) were aged
be estimated, and ui is the error term. From Eq. (2), the 11–16 years. The majority of children reported that their
model specification for estimating factors that predict fathers’ (33.0%) and mothers’ (38.7%) highest educational
obesity status could be represented as follows: level was basic education. Table 2 also shows essential
anthropometric characteristics of respondents. Some
Yi 46.9% (42.2 for males and 51.7% for females) of the chil-
In ¼ β þ β age þ β religion dren were overweight for their age.
1−Yi 0 1 child 2 child
þ β3sexchild þ β4ethnicityChild
þ β5fathereducationlevel child Prevalence of childhood obesity
þ β6mothereducation levelchildβ7numbero f siblings Table 3 shows prevalence of obesity by background ofchild
þ β fatheroccupation children. Of 46.9% of respondents that were overweight,8 child
þ β motherccupation þ…β þ ui 21.2% were obese. Some 26.8% of children from the pri-9 child n
ð3Þ vate school were obese compared to 21.4% from the
public school. Childhood obesity was also higher among
Where In = natural log; Yi/(l-Yi) = odds ratio; ‘ß0’ = girls (27.2%) than boys (19%).
Ganle et al. BMC Public Health         (2019) 19:1561 Page 5 of 12
Table 1 Background characteristics of respondents (N = 285)
Characteristic Private school n(%) Public School n(%) Total n (%)
Sex of child
Male 55 (58.51) 86 (45.0) 141 (49.5)
Female 39 (41.5) 105 (55.0) 144 (50.5)
Age of child (years)
5–10 51 (54.3) 70 (36.7) 121 (42.5)
11–16 43 (45.7) 121 (56.3) 164 (57.5)
Religion of child
Christianity 90 (95.7) 177 (92.7) 267 (93.7)
Islam 4 (4.3) 12 (6.3) 16 (5.6)
Traditionalist 0 (0.0) 2 (1.0) 2 (0.7)
Number of siblings
None 0 (0.0) 1 (0.5) 1 (0.4)
1–4 73 (77.7) 118 (61.8) 191 (67.0)
5–10 18 (19.2) 69 (36.1) 87 (30.5)
11+ 3 (3.1) 3 (1.6) 6 (2.1)
Mother’s education
No education 5 (5.4) 34 (17.8) 39 (13.7)
Basic education 4 (4.3) 106 (55.5) 110 (38.7)
Secondary education 20 (21.5) 35 (18.3) 55 (19.4)
Tertiary education 64 (68.8) 16 (8.4) 80 (28.2)
Father’s education
No education 5 (5.3) 29 (15.2) 34 (11.9)
Basic education 2 (2.1) 92 (48.2) 94 (33.0)
Secondary education 11 (11.7) 58 (30.4) 69 (24.2)
Tertiary education 76 (80.9) 12 (6.2) 88 (30.9)
Mother’s occupation
Self-employed 54 (57.5) 176 (92.2) 230 (80.7)
Public sector employee 18 (19.2) 10 (5.2) 28 (9.8)
Private sector employee 22 (23.4) 5 (2.6) 27 (9.5)
Father’s occupation
Self-employed 164 (85.9) 20 (21.3) 184 (64.6)
Public sector employee 15 (7.8) 23 (24.5) 38 (13.3)
Private sector employee 12 (6.3) 51 (54.3) 63 (22.1)
Table 2 Anthropometric characteristics by sex (N = 285)
Characteristic Male n(%) Female n(%) Factors associated with obesity
BMI Percentile To identify factors that significantly predict childhood
obesity, chi-square and fisher’s exact tests were first per-
< 5th 7 (4.9) 5 (3.8)
formed between a total of 16 theoretically relevant inde-
5th – 85th 75 (52.9) 65 (44.5) pendent socio-demographic and dietary variables and
85th – 95th 33 (23.2) 35 (24.5) the outcome variable. The results are shown in Tables 4
> 95th 26 (19.0) 39 (27.2) and 5. From the bivariate analysis, two socio-
Total Weight (kg) 141 (100) 144 (100) demographic factors (age of child [p = 0.004], and fa-
Mean (SD) 34.6 + 10.6 47.8 + 25.0 ther’s educational level [p = 0.002]), and three dietary/be-
havioural factors (sports activity per week [p = 0.029],
Height (cm)
regularity of fizzy drinks intake [p = 0.004], and sleep
Mean (SD) 135.1 + 19.8 138.3 + 20.7 hours per day [p = 0.001]) were statistically associated
Ganle et al. BMC Public Health         (2019) 19:1561 Page 6 of 12
Table 3 Prevalence of obesity by selected socio-demographic children aged 5–10 years (cOR = 7.60; 95% CI = 1.29–
characteristics (N = 285) 44.9; p = 0.032). After adjusting for other variables iden-
Characteristic Not Obese n(%) Obese n(%) tified as significant predictors of obesity, this difference
Type of School was still statistically significant (aOR = 6.07; 95% CI =
Private 69 (73.2) 25 (26.8) 1.17–31.45; p = 0.025). The risks of childhood obesity
generally increased as paternal education increases. Spe-
Public 150 (78.6) 41 (21.4)
cifically, children whose fathers obtained basic (cOR =
Sex of child 1.93; 95% CI = 0.73–5.09; p = 0.185), secondary (cOR =
Male 114 (81.0) 26 (19.0) 3.86; 95% CI = 1.52–9.78; p = 0.004), and tertiary (cOR =
Female 105 (72.8) 39 (27.2) 4.20; 95% CI = 1.67–10.59; p = 0.002) education were, re-
Age of child (years) spectively, 1.93 times, 3.86 times and 4.20 times more
5–10 105 (86.9) 16 (13.1) likely to be obese compared to children whose fathers
had no education. After adjusting for other variables
11–16 114 (69.4) 50 (30.6)
identified as significant predictors of obesity, the differ-
Religion ence between ‘no education’ and ‘secondary education’
Christianity 202 (75.8) 65 (24.2) (aOR = 2.97; 95% CI = 1.09–8.08; p = 0.032), and ‘no edu-
Islam 14 (85.0) 2 (15.0) cation’ and tertiary education (aOR = 3.46; 95% CI =
Traditionalist 1 (50.0) 1 (50.0) 1.27–9.42; p = 0.015) were still statistically significant.
Number of siblings Children who were involved in sporting activities for
at least 3 days per week had a 42% significant reduction
None 0 (0.0) 1 (100.0)
in their odds of being obese as compared to children
1–4 146 (76.5) 45 (23.5) who participated in sporting activities for less than 3
5–10 67 (77.5) 20 (22.5) days per week (cOR = 0.58; 95% CI = 0.36–0.95; p =
11+ 5 (86.7) 1 (13.3) 0.030). After adjusting for other variables identified as
Mother’s education significant predictors of obesity, this association was still
No education 36 (91.8) 3 (8.2) statistically significant (aOR = 0.56; 95% CI = 0.33–0.96;
p = 0.034). Further, children who consume fizzy drinks
Basic education 80 (72.7) 30 (27.3)
on some days (cOR = 2.39; 95% CI = 1.06–5.38; p =
Secondary education 46 (83.6) 9 (16.4) 0.035) and most days (cOR = 3.36; 95% CI = 1.60–7.06;
Tertiary education 56 (70.0) 24 (30.0) p = 0.001), had respectively 2.39 and 3.36 times the odds
Father’s education of being obese as compared to children who hardly or
No education 33 (98.4) 1 (1.6) never consume fizzy drinks. These differences were sta-
Basic education 64 (67.9) 30 (32.1) tistically significant. After adjusting for other variables
identified as significant predictors of obesity, children
Secondary education 60 (86.7) 9 (13.3)
who consumed fizzy drinks on most days were still 2.84
Tertiary education 62 (70.0) 26 (30.0) times more likely to be obese compared to children who
Mother’s occupation hardly or never consumed fizzy drinks (aOR = 2.84; 95%
Self-employed 178 (77.4) 52 (22.6) CI = 1.24–6.52; p = 0.014).
Public sector employee 21 (73.6) 7 (26.4) Finally, children who slept for more than 8 h per day
Private sector employee 20 (75.6) 7 (24.4) had 68% reduction in their odds of being obese as com-
pared to children who slept for less than 5 h (cOR =
Father’s occupation
0.32; 95% CI = 0.16–0.63; p = 0.001). This association
Self-employed 145 (78.7) 39 (21.3) was still significant after adjusting for other variables
Public sector employee 30 (77.9) 8 (22.1) (aOR = 0.38; 95% CI = 0.19–0.79; p = 0.009). Further-
Private sector employee 45 (70.8) 18 (29.2) more, children who slept between 5 and 8 h, had 34% re-
Overall 225 (78.8) 60 (21.2) duction in their odds of being obese as compared to
children who slept for less than 5 h (cOR = 0.66; 95%
CI = 0.31–1.41; p = 0.286). This association was however
with childhood obesity. These factors were pulled into not significant.
binary and multivariable logistic regression models and
odd ratios were estimated. The results are shown in Discussion
Table 6. This study aimed to estimate childhood obesity preva-
The results show that children aged 11–16 years had lence and identify key factors among in- school children
significantly higher odds of being obese as compared to (5–16 years) in a Metropolitan district of Ghana.
Ganle et al. BMC Public Health         (2019) 19:1561 Page 7 of 12
Table 4 Socio-demographic factors associated with child obesity (bivariate) (N = 285)
Variable Not Obese n(%) Obese n(%) X2 p-value
Type of School 0.762 0.383
Private 69 (73.2) 25 (26.8)
Public 150 (78.6) 41 (21.4)
Sex of child 1.959 0.162
Male 114 (81.0) 26 (19.0)
Female 105 (72.8) 39 (27.2)
Age of child (years) 10.783 0.004*
5–10 105 (86.9) 16 (13.1)
11–16 114 (69.4) 50 (30.6)
Religion¥ 2.306 0.270
Christianity 202 (75.8) 65 (24.2)
Islam 14 (85.0) 2 (15.0)
Traditionalist 1 (50.0) 1 (50.0)
Number of siblings¥ 1.574 0.783
None 0 (0.0) 1 (100.0)
1–4 146 (76.5) 45 (23.5)
5–10 67 (77.5) 20 (22.5)
11+ 5 (86.7) 1 (13.3)
Mother’s education 6.867 0.076
No education 36 (91.8) 3 (8.2)
Basic education 80 (72.7) 30 (27.3)
Secondary education 46 (83.6) 9 (16.4)
Tertiary education 56 (70.0) 24 (30.0)
Father’s education¥ 14.537 0.002*
No education 33 (98.4) 1 (1.6)
Basic education 64 (67.9) 30 (32.1)
Secondary education 60 (86.7) 9 (13.3)
Tertiary education 62 (70.0) 26 (30.0)
Mother’s occupation 0.168 0.919
Self-employed 178 (77.4) 52 (22.6)
Public sector employee 21 (73.6) 7 (26.4)
Private sector employee 20 (75.6) 7 (24.4)
Father’s occupation 1.214 0.545
Self-employed 145 (78.7) 39 (21.3)
Public sector employee 30 (77.9) 8 (22.1)
Private sector employee 45 (70.8) 18 (29.2)
¥Fisher’s exact test; *p < 0.05
Findings highlight childhood obesity as an important prevalence in our study is lower than the 43% recently
public and child health issue that needs attention. To reported for the general adult population in Ghana [6].
start with, 21.2% of the children in our study were obese, This notwithstanding, the relatively high obesity preva-
and more children from the private school (26.8%) than lence in our study suggests that children are equally vul-
the public school (21.4) were obese. Childhood obesity nerable to obesity in cities in Ghana. This could be
prevalence in our study was higher than the 17.4% related to the fact that children share the same or similar
prevalence reported in a similar previous study in the obesogenic environments with adults, and are therefore
northern part of Ghana [6]. However, childhood obesity increasingly exposed to risk factors such as sedentary life
Ganle et al. BMC Public Health         (2019) 19:1561 Page 8 of 12
Table 5 Dietary and behavioural factors associated with child as sedentary life styles. On diet, children aged 5–10 are
obesity (bivariate) (N = 285) more likely to have their dietary choices better regulated
Variable Not Obese Obese X2 p-value than those aged 11–16. For instance, children aged 5–10
Sports activity per week 4.7463 0.029* are likely to carry home-made meals to school than chil-
< 3 days 52 (68.6) 55 (31.4) dren aged 11–16 who may have access to money and
may therefore purchase their own meals especially in
> 3 days 110 (81.8) 68 (18.2)
school. This may expose children aged 11–16 to un-
Where breakfast is obtained 1.352 0.509 healthy diet compared to those aged 5–10, which may
Home 95 (74.3) 80 (25.7) affect weight gain and subsequently the BMI of children
Outside of home 52 (71.9) 32 (18.1) aged 11–16. As a number of studies have shown, eating
No breakfast 15 (77.7) 11 (22.3) outside of home, especially in fast food eateries, is corre-
Regularity of eating fruits¥ 3.139 0.208 lated positively with overweight and obesity in children
[31]. In terms of sedentary lifestyle, children aged 5–10
Everyday 42 (78.3) 30 (21.7)
may again be more regulated in terms of their sleep time
Sometimes 107 (74.6) 89 (25.4) as well as time spend on watching TV and playing com-
Hardly or never 13 (76.5) 4 (23.5) puter games within the home environment. Within the
Regularity of eating junk foods¥ 3.042 0.219 school, 5-10 year old children may also be more involved
Never 8 (92.7) 3 (7.3) in outdoor games and playground activities than those
Monthly 73 (80.8) 47 (19.2) aged 11–16 who may spend more time doing classroom
work. When this is combined with the possibility that
Weekly 81 (72.6) 73 (27.4)
children aged 11–16 may be more exposed to unhealthy
Regularity of fizzy drinks intake 11.150 0.004* diet, the likelihood of weight gain and higher BMI could
Hardly or never 37 (97.1) 11 (2.9) be more in this age group. In line with the recent
Some days 80 (70.0) 80 (30.0) WHO’s global action plan on physical activity 2018–
Most days 45 (58.4) 32 (41.6) 2030 [2], we recommend targeted interventions such as
Television exposure 5.009 0.082 sport and other physical activities as well as dietary edu-
cation and nutrition counselling among children aged
Never 10 (86.7) 5 (13.3)
11–16 to ensure that they maintain healthy weight.
Some days 53 (68.6) 56 (31.4) Further, the risks of childhood obesity appeared to
Everyday 99 (61.5) 62 (38.5) increase with increase in paternal education. This re-
Sleep hours per day 14.369 0.001* sult is very counter-intuitive precisely because better
Less than 5 h 17 (57.8) 28 (42.2) educated parents/guardians generally have greater op-
5–8 h 34 (67.9) 37 (32.1) portunities for obtaining information related to
healthy dietary practices as well as healthy behaviour
8+ h 111 (85.7) 58 (14.3)
change information. Therefore, one would expect bet-
¥Fisher’s exact test; *p < 0.05
ter outcomes and lifestyle indicators for children
whose parents have higher education. This however
style in urban settings. Also, there is the perception in appears not to be the case in our study. Though
Ghana about body weight: people are perceived to be liv- counter-intuitive, this result is nevertheless not sur-
ing good when they look fat [4]. Consequently, many prising. This is because in many low-income settings,
parents may take steps to ensure that their children con- higher education is often linked to higher socio-
form to this expectation in order to be praised for good economic status, including higher purchasing power.
parenting. Together with previous studies highlighting Having higher purchasing power means the ability to
growing prevalence of childhood obesity, our findings afford, for example, personal means of transport such
here suggest a need for promotive health interventions as a car, which may be used to transport children to
(e.g. healthy eating and physical activity) targeted not school compared to children with lowly educated par-
only at adults but also children in our study context as ents who may lack such purchasing power and may
well as in other African settings such as Nigeria [15], walk or ride bicycles. Also, better purchasing power
Uganda [18], and Ethiopia [14] where similarly high may increase the chances of consumption of more proc-
levels of childhood obesity have been reported. essed foods, unhealthy snacking, consumption of high fat
Children who were aged 11–16 were 6 times more diets within and away from home. This link has been sug-
likely to be obese compared to those aged 5–10 years. It gested in a number of low-income settings [18]. In
is not entirely clear why this difference exists. However, addition, better purchasing power may also encourage
we believe it could be related to dietary practices as well more sedentary behaviour through access to TV and
Ganle et al. BMC Public Health         (2019) 19:1561 Page 9 of 12
Table 6 Predictors of childhood obesity (multivariate) (N = 285)
Variable cOR (95% C.I.) p-value aOR (95% C.I.) p-value
Age of child (years)
5–10 (ref) 1 1
11–16 7.60 (1.29–24.94) 0.032** 6.07 (1.17–21.45) 0.025**
Father’s education
No education (ref) 1 1
Basic 1.93 (0.73–5.09) 0.185 1.46 (0.51–4.15) 0.483
Secondary 3.86 (1.52–9.78) 0.004** 2.97 (1.09–8.08) 0.032**
Tertiary 4.20 (1.67–10.59) 0.002** 3.46 (1.27–9.42) 0.015**
Sports activity per weeka
< 3 days (ref) 1 1
> 3 days 0.58 (0.36–0.95) 0.030** 0.56 (0.33–0.96) 0.034**
Regularity of fizzy drinks intakeb
Hardly or never (ref) 1 1
Some days 2.39 (1.06–5.38) 0.035** 2.09 (0.84–5.16) 0.112
Most days 3.36 (1.60–7.06) 0.001** 2.84 (1.24–6.52) 0.014**
Sleep hours per day
Less than 5 h (ref) 1 1
5–8 h 0.66 (0.31–1.41) 0.286 0.70 (0.31–1.57) 0.391
8+ h 0.32 (0.16–0.63) 0.001** 0.38 (0.19–0.79) 0.009**
aSports activity was defined as engaging in any of the following: playing football, basketball, tennis, volley ball and ampe as well as running and cycling)
Ampe is a simple jumping game played by school-age children (mostly girls), in Ghana and neighbouring countries and usually involving two or more players and
requires no equipment
bFizzy drinks were defined as non-alcoholic soft drinks that contain carbonated water, a sweetener (sweetener may be sugar, high-fructose corn syrup, fruit juice,
a sugar substitute, or some combination of these), and a natural or artificial flavouring
**p < 0.05; cOR Crude odds ratio, aOR Adjusted odds ratio - aOR for each observation was derived a model which included all variables in the table; CI Confidence
interval, ref Reference category
computer games and related indoor activities that reduce of the energy balance equation, which influences whether
physical activity. Another way that higher education of energy would be expended leading to a healthy weight or
parents could expose children to obesity risks is work. accumulated leading to child obesity. Regular physical ac-
Parents with higher education are likely to be engaged in tivity could potentially lead to reduction in the odds of a
paid employment. Tight work schedules may make it diffi- child being obese. Therefore, parents and teachers who are
cult for parents to develop healthy meal plans for their the primary caregivers of children should endeavour to in-
households and this could increase consumption of con- crease children’s physical activity. Physical education in
venient foods which may be unhealthy. Indeed, our results schools should be placed on learning time tables for more
here support growing research evidence that suggests that than three times per week to increase children’s physical ac-
whereas higher socio-economic status is inversely related tivity. Parents should also encourage children to engage in
to obesity in high-income settings [10], it is positively as- outdoor games. This will however require local government
sociated with obesity in low-income settings [32]. authorities to create save and child-friendly spaces and
Not surprisingly, engaging in sport activities for at playgrounds within urban communities – something cur-
least 3 days per week reduced the odds of being obese by rently lacking in many urban settings in Ghana - to encour-
42% compared to children who participated in sport ac- age more outdoor physical activities among children.
tivities for less than 3 days per week. Our results here Children who consumed fizzy drinks on most days
highlight a need for age- and sex-appropriate sports and were 2.84 times more likely to be obese compared with
physical activity-based interventions such as football, those who hardly or never consumed fizzy drinks. This
basketball, tennis, volley ball, ampe as well as running is not surprising given that fizzy drinks typically are
and cycling among school children. First, while the im- sugar-sweetened beverages, and usually have high con-
portance of diet/physical activity in relation to obesity is tent of fructose [33]. Also, most sweetened foods are
not new insight, physical activity does represent one side typically calorie dense, which when combined with less
Ganle et al. BMC Public Health         (2019) 19:1561 Page 10 of 12
physical activity, could easily result in energy imbalance. bias. This is because respondents were asked about events
Indeed, previous studies have found that children who that occurred several weeks before the interview. Some re-
consumed sweetened foods are more likely to be over- spondents may also have given socially desirable dietary
weight or obese compared to those who do not [3, 8, 19, and behavioural responses such as exercise in order to
22]. In this regard, strategies such as developing healthy present themselves as leading active lives when in fact this
meal plans for the household could help regulate chil- may not be the case. Third, the study was conducted in
dren’s consumption of unhealthy foods. Teachers and only two schools and involved only 285 children in one
school authorities could be involved to promote the sale metropolitan district. Therefore, the limitations of
and consumption of more healthy foods especially generalizability due to the non-representativeness of the
within the vicinity of educational facilities. In Ghana, sample are acknowledged. Fourth, our statistical analysis
most private schools provide lunch at a fee or regulate approach was largely data-driven. That is, co-variates were
eating times. For private schools that provide food, they selected based on bivariate significance rather than a pre-
need to develop healthy food timetables for children. In specified theoretical view of potential causal structure.
place of fizzy drinks, naturally prepared fruit drinks Consequently, some variables that may be important after
should be served. Also, parents as much as possible conditioning on other variables may have been missed.
should avoid giving children extra money which they This is a limitation of our study. Finally, other factors that
could easily use to purchase fizzy drinks. Parents at were not measured in the study such as parental nutrition
home should discourage fizzy drinks consumption by knowledge, may nevertheless have had an influence on re-
not consuming them. Parents must particularly be exam- spondents dietary and behavioural behaviours. These limi-
ples to their children by consuming healthy and nutri- tations notwithstanding, the results provide useful
tious foods they expect their children to consume as this evidence that could inform large-scale research as well as
may drive home the message of healthy eating and living school-based interventions to reduce the risk of childhood
among children. The Government could also help by obesity in Ghana.
placing high taxes on fizzy drinks to discourage their
consumption. Conclusions
Also, more sleep hours appeared to reduce the risk of This study has provided further insights into the preva-
childhood obesity. Children who slept for more than 8 h lence of, and risk factors, for childhood obesity which have
per day had 62% reduction in their odds of being obese implications for interventions to reduce obesity in Ghana
compared to children who slept for less than 5 h. This re- and similar contexts elsewhere. Though predictive socio-
sult support one study among school-aged children in demographic factors such as age of child and fathers’ edu-
northern Ghana which linked shorter sleep hours to in- cational attainment may not be amenable to intervention
creased obesity risks [6]. In contrast, a study conducted in to reduce childhood obesity risks, dietary/behavioural fac-
China revealed that short sleep duration was not associ- tors observed to influence childhood obesity could be tar-
ated with obesity [34]. It is not entirely clear why this dis- geted to encourage the maintenance of healthy weight and
crepancy exists and further research is required in lifestyle among children. Therefore, age and sex-
different contexts to better explore this issue. However, appropriate community and school-based interventions
our results do suggest a need for parents to encourage are needed to promote healthy diet selection and con-
their children to have sufficient sleep. As a consensus sumption, physical activity and healthy life styles among
statement of the American Academy of Sleep Medicine in-school children. Finally, there is need for large-scale
recommends, children aged 6–12 need between 9 and 12 community-based population studies in Ghana to estimate
h of sleep per 24 h on a regular basis to promote optimal obesity prevalence in different contexts and identify risk
health whereas those aged 13–18 should sleep 8 to 10 h factors in different populations. Such information could
per 24 h on a regular basis to promote optimal health [35]. be vital for national policy and intervention development.
Our study has some limitations. First, although the an-
thropometric measurement instrument (e.g. weighing
scale) were continuously calibrated and monitored, it is Supplementary information
possible that extended period of use could have affected Supplementary information accompanies this paper at https://doi.org/10.1186/s12889-019-7898-3.
the accuracy of some of the measurement. Related to the
above, the use of only BMI as screening tool for obesity Additional file 1: Reliability testing of data collection instrument
has limitations. For instance, concerns have been raised Additional file 2: Study questionnaire
regarding the reliability of weight and height measurement
in research contexts [36–38]. Nevertheless, BMI is a
Abbreviations
valuable population-level indicator used widely in epide- BMI: Body mass index; CDC: Centre for Disease control and Prevention;
miologic research. Second, there could have been recall TV: Television; WHO: World Health Organisation
Ganle et al. BMC Public Health         (2019) 19:1561 Page 11 of 12
Acknowledgements 8. Mohammed H, Vuvor F. Prevalence of childhood overweight /obesity in
This manuscript was drafted when the first author (JKG) was a Fellow at the basic school in Accra. Ghana Med J. 2012;46(3):124–7.
Stellenbosch Institute for Advanced Study (STIAS), Wallenberg Research 9. Franks PW, Hanson RL, Knowler WC, Sievers ML, Bennett PH, Looker HC.
Centre at Stellenbosch University, Stellenbosch 7600, South Africa. Writing Childhood obesity, other cardiovascular risk factors, and premature death. N
space for the manuscript was graciously provided by STIAS. We are grateful Engl J Med. 2010;362(6):485–93.
for this support. 10. Wang YC, McPherson K, Marsh T, Gortmaker S, Brown M. Health and
economic burden of the projected obesity trends in the USA and the UK.
Authors’ contributions Lancet. 2011;378(9793):815–25.
PPB and JKG conceived the study and jointly contributed to the design. PPB 11. Cunningham SA, Kramer MR, Narayan V. Incidence of childhood obesity in
and LB performed literature search and review. PPB led the field work. PPB the United States. N Engl J Med. 2014;370:403–11.
and JKG performed statistical analysis. JKG and LB interpreted the data. JKG 12. Trasande L, Liu Y, Fryer G, Weitzman M. Effects of childhood obesity on
and LB jointly drafted the manuscript. JKG and PPB revised the first draft. All hospital care and costs, 1999-2005. Health Aff. 2009;28(4):751–60. https://doi.
authors read and approved the manuscript for submission for publication. org/10.1377/hlthaff.28.4.w751.
13. Finkelstein EA, Graham WC, Malhotra R. Lifetime direct medical costs of
Funding childhood obesity. Pediatrics. 2014;133:854–62.
This research did not receive any specific grant from funding agencies in the 14. Desalew A, Mandesh A, Semahegn A. Childhood overweight, obesity and
public, commercial, or not-for-profit sectors. However, this manuscript was associated factors among primary school children in Dire Dawa, eastern
drafted when the first author (JKG) was a Fellow at the Stellenbosch Institute Ethiopia; a cross-sectional study. BMC Obesity. 2017;4:20. https://doi.org/10.
for Advanced Study, South Africa. Writing space for the manuscript was gra- 1186/s40608-017-0156-2.
ciously provided by the Stellenbosch Institute for Advanced Study. We are 15. Musa DI, Toriola AL, Monyeki MA, Lawal B. Prevalence of childhood and
grateful for this support. adolescent overweight and obesity in Benue state, Nigeria. Trop Med Int
Health. 2012;17(11):1369–75. https://doi.org/10.1111/j.1365-3156.2012.03083.x.
Availability of data and materials 16. Sigmund E, El Ansari W, Sigmundová D. Does school-based physical activity
The datasets used and/or analysed during the current study are available decrease overweight and obesity in children aged 6–9 years? A two year
from the corresponding author on reasonable request. non-randomized longitudinal intervention study in the Czech Republic.
BMC Public Health. 2012;12:570. https://doi.org/10.1186/1471-2458-12-570.
Ethics approval and consent to participate 17. Hills AP, Andersen LB, Byrne NM. Physical activity and obesity in children. Br
Ethical approval was obtained from the Ghana Health Service Ethical Review J Sports Med. 2011;45(11):866–70.
Committee (Protocol ID NO: GHSERC056/01/18). Informed written consent – 18. Peltzer K, Pengpid S, Alafia TS, Zcan NK, Mantilla C, Rahamefy OH, et al.
parental consent and child assent - were obtained (either by signing or Prevalence of overweight/obesity and its associated factors among
thumb printing) from all respondents. university students from 22 countries. Int J Environ Res Public Health. 2014;
11(7):7425–41. https://doi.org/10.3390/ijerph110707425.
Consent for publication 19. Odume BB, Ofoegbu OS, Aniwada EC, Okechukwu EF. The influence of
Not Applicable. family characteristics on glycaemic control among adult patients with type
2 diabetes mellitus attending the general outpatient clinic, National
Competing interests Hospital, Abuja, Nigeria. S Afr Fam Pract. 2015;57(6):347–53.
The authors declare that they have no competing interests. 20. Badawi NE, Barakat AA, Awad S, Sherbini E. Prevalence of overweight and
obesity in primary school children in port said city. Egypt Paediatr Assoc
Author details Gazette. 2013;61(1):31–6.
1Department of Population,Family and Reproductive Health, School of Public 21. Nuru H, Mamang F. Association between snacking and obesity in children: a
Health, University of Ghana, P.O. Box LG 13, Legon, Accra, Ghana. review. Int J Commun Med Public Health. 2015;2(3):196–200.
2Stellenbosch Institute for Advanced Study (STIAS), Wallenberg Research 22. Gebremichael B, Chere A. Prevalence of childhood overweight and obesity
Centre at Stellenbosch University, Stellenbosch 7600, South Africa. 3Regional and its determinants among primary school children in Addis Ababa,
Institute for Population Studies, University of Ghana, Accra, Ghana. Ethiopia: a cross-sectional study. J Nutr Disord Therapy. 2015;S1:002. https://
doi.org/10.4172/2161-0509.S1-002.
Received: 5 April 2019 Accepted: 5 November 2019 23. Boyland EJ, Halford JC. Television advertising and branding. Effects on
eating behaviour and food preferences in children. Appetite. 2013;62:236–
41. https://doi.org/10.1016/j.appet.2012.01.032.
References 24. Berentzen NE, Smit HA, van Rossem L, Gehring U, Kerkhof VDM, Postma DS,
1. World Health Organization. Obesity and overweight - Fact sheet. Accessed on et al. Screen time, adiposity and cardiometabolic markers: mediation by
20 Feb 2018 at https://www.who.int/news-room/fact-sheets/detail/obesity- physical activity, not snacking, among 11-year-old children. Int J Obes
and-overweight. (Lond). 2014;38(10):1317–23. https://doi.org/10.1038/ijo.2014.110.
2. World Health Organization. Global action plan on physical activity 2018– 25. Atuahene M, Ganle JK, Adjuik M, Atuahene NF, Kampitib BG. Overweight
2030: more active people for a healthier world. Geneva: World Health and obesity prevalence among public servants in Nadowli district, Ghana,
Organization; 2018. and associated risk factors: a cross-sectional study. BMC Obesity. 2017.
3. Ofori-asenso R, Agyeman AA, Laar A, Boateng D. Overweight and obesity https://doi.org/10.1186/s40608-017-0153-5.
epidemic in Ghana — a systematic review and meta-analysis. BMC Public 26. Guh DP, Zhang W, Bansback N, Amarsi C, Birmingham L, Anis HA. The
Health. 2016;16:1239. https://doi.org/10.1186/s12889-016-3901-4. incidence of co-morbidities related to obesity and overweight: a systematic
4. Benkeser RM, Biritwum R, Hill AG. Prevalence of overweight and obesity and review and meta-analysis. BMC Public Health. 2009;25(9):88. https://doi.org/
perception of healthy and desirable body size in urban, Ghanaian women. 10.1186/1471-2458-9-88.
Ghana Med J. 2012;46(2):66–75. 27. Ghana Statistical Service. Population and housing census 2010: district analytical
5. Addo PNO, Nyarko KM, Sackey SO, Akweongo P, Sarfo B. Prevalence of report, Tema metropolitan assembly. Accra: Ghana Statistical Service; 2014.
obesity and overweight and associated factors among financial institution 28. Cohrane WG. Sampling technique. 3rd ed. New York: Wiley; 1977.
workers in Accra Metropolis, Ghana : a cross sectional study. BMC Res Notes. 29. Tavakol M, Dennick R. Making sense of Cronbach’s alpha. Int J Med Educ.
2015;8:599. https://doi.org/10.1186/s13104-015-1590-1. 2011;2:53–5.
6. Mogre V, Gaa PK, Abubakari RNS. Overweight, obesity and thinness and 30. Hamid F, Islam R, Chandraray P. Childhood obesity – an emerging problem:
associated factors among school-aged children (5-14 years) in tamale, a review article. Bangladesh J Child Health. 2013;37(2):122–6. https://doi.org/
northern Ghana. Eur Sci J. 2013;9(20):1857–7881. 10.3329/bjch.v37i2.17269.
7. Amidu N, Owiredu WKBA, Saaka M, Quaye L, Wanwan M, Kumibea PD, et al. 31. Neto FA, Eto FN, Sabrina T, Pereira S, Carletti L, Bisi C. Active and sedentary
Determinants of childhood obesity among basic school children aged 6–12 behaviours in children aged 7–10 years old: the urban and rural contexts,
years in tamale Metropolis. J Med Biomed Sci. 2013;2(3):26–34. Brazil. BMC Public Health. 2014;14(1174):1-10.
Ganle et al. BMC Public Health         (2019) 19:1561 Page 12 of 12
32. Gupta N, Goel K, Shah P, Misra A. Childhood obesity in developing
countries: epidemiology, determinants, and prevention. Endocr Rev. 2012;
33(1):48–70. https://doi.org/10.1210/er.2010-0028.
33. Bray AG. Soft drink consumption and obesity: it is all about fructose. Curr
Opin Lipidol. 2010;21:51–7.
34. Cao M, Zhu Y, He B, Yang W, Chen Y, Ma J, et al. Association between sleep
duration and obesity is age and gender dependent in Chinese urban
children aged 6–18 years: a cross-sectional study. BMC Public Health. 2015;
15:1029. https://doi.org/10.1186/s12889-015-2359-0.
35. Paruthi S, Brooks LJ, D’Ambrosio C, Hall WA, Kotagal S, Lloyd RM, et al.
Recommended amount of sleep for pediatric populations: a consensus
statement of the American Academy of sleep medicine. J Clin Sleep Med.
2016;12(6):785–6.
36. Skinner AC, Skelton JA. Prevalence and trends in obesity and severe obesity
among children in the United States, 1999-2012. JAMA Pediatr. 2014.
https://doi.org/10.1001/jamapediatrics.2014.21.
37. Alemu E, Atnafu A. Prevalence of overweight and/or obesity and associated
factors among high school adolescents in Arada sub city, Addis Ababa,
Ethiopia. J Nutr Food Sci. 2004;4(2):10–4. https://doi.org/10.4172/2155-9600.
1000261.
38. Arora M, Nazar GP, Gupta VK, Perry CL, Reddy KS, Stigler MH. Association of
breakfast intake with obesity, dietary and physical activity behavior among
urban school-aged adolescents in Delhi, India: results of a cross-sectional
study. BMC Public Health. 2012;12(881):1–11.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.