|Year : 2014 | Volume
| Issue : 3 | Page : 164-170
Lifestyle and genetic factors in the prevalence of adolescent underweight and obesity in an urban area in Nigeria
Akinola Oluwole Busayo, Medubi Leke Jacob, Tejumola Abimbola
Department of Anatomy, Faculty of Basic Medical Sciences, College of Health Sciences, University of Ilorin, Nigeria
|Date of Submission||04-Jun-2014|
|Date of Decision||16-Aug-2014|
|Date of Acceptance||18-Aug-2014|
|Date of Web Publication||19-Sep-2014|
Medubi Leke Jacob
Department of Anatomy, Faculty of Basic Medical Sciences, College of Health Sciences, University of Ilorin
Source of Support: None, Conflict of Interest: None
Background: The rise in the prevalence of overweight and obesity has continued to elicit genuine public health concerns across different disciplines. This underscores the need for continuous monitoring of anthropometric indices of obesity and predisposing factors across populations of different age groups. Unfortunately, few studies have documented the prevalence of obesity in Nigerian adolescents. Materials and Methods: A cross-section of adolescents attending high school in Ilorin, the Capital of Kwara State, Nigeria, was recruited into this study; following informed consent, they were requested to fill-in questionnaires in order to obtain their sociodemographic and lifestyle information. Subsequently, each participant's anthropometric measurements were taken which included, body weight, standing height, waist and hip circumferences. Results: Analyses of our data reveal gender disparity in the prevalence of obesity among participants. Based on body mass index, 6% of adolescent girls are obese, and 8.5% are overweight, while obesity and overweight among adolescent boys are 2% and 2.5%, respectively. Underweight was 7% and 15% among female and male participants, respectively. However, using waist-to-hip ratio (WHR), central obesity among male and female subjects is as high as 26% (WHR = 0.9) and 30% (WHR = 0.84), respectively. Lifestyle-factors analysis reveals that at least 31% of obese participants do not engage in any physical exercise; only 12.5% of obese subjects reported that they did not consume soft drink at all. Our data reveal familial tendencies of obesity -31.58% of overweight and obese subjected reported obesity in their families; only 18.78% of underweight and normal-weight subjects reported the same in their families. Conclusion: The prevalence of obesity is higher among female adolescents compared to male adolescents among the study population in Ilorin, Nigeria. While underweight is 2 times higher among male adolescents, each condition requires imminent attention because each poses a potential risk factor for ill-health.
Keywords: Adolescents, obesity, overweight, underweight
|How to cite this article:|
Busayo AO, Jacob ML, Abimbola T. Lifestyle and genetic factors in the prevalence of adolescent underweight and obesity in an urban area in Nigeria. J Obes Metab Res 2014;1:164-70
|How to cite this URL:|
Busayo AO, Jacob ML, Abimbola T. Lifestyle and genetic factors in the prevalence of adolescent underweight and obesity in an urban area in Nigeria. J Obes Metab Res [serial online] 2014 [cited 2020 Jul 7];1:164-70. Available from: http://www.jomrjournal.org/text.asp?2014/1/3/164/141146
| Introduction|| |
There is currently a justified global concern about the trend in overweight and obesity among different age groups of the human population. This is because overweight and obesity have been identified as risk factors for many conditions such as diabetes, cardiovascular diseases, asthma, and other respiratory diseases.  These ill-health consequences have made continuous studying and monitoring of anthropometric indices indispensable to the effective delivery of healthcare. It is well-documented that the obesity is burgeoning in both developed and developing countries and has cut across all socio-economic groups irrespective of ethnicity, gender and age.  For instance in the United States, between 1999 and 2002, the adult population was reported to be 65.1% overweight and 30.4% obese, while children aged 6-19 years were reported to be made up of 31% overweight and 16% obese individuals.  Among populations of developing countries, obesity is reported to be rising across gender and age brackets. 
Meanwhile, the most convenient diagnostic tool for identifying obesity is anthropometric measurements. It's simplicity has not restricted its use to resources-poor societies, rather anthropometric measurements is used the world over.
Regardless of criticism of the use of anthropometric measurement for derivation of indices such as body mass index (BMI) and waist-to-hip ratio (WHR), evidence abound that at certain cut-off marks based on these indices, the risk of certain diseases is significantly raised. For instance, overweight has been shown to be related to cardiovascular risk factors in adolescents and school children.  Furthermore, in adolescents as well as children, Katzmarzyk et al.,  used BMI to predict cardiovascular risk factors. High-risk of metabolic diseases has been shown to demonstrate a strong association with an enlarged waist.  In a study by Suwaidi et al.,  it was discovered that obesity was independently associated with coronary endothelial dysfunction in patients with normal or mildly diseased coronary arteries.
It is well known that obesity has intrinsic connections with body mass, body surface area, and body composition. The challenge of diagnosing the condition has not been a simple one, since all proxies so far developed has attracted both acceptance and criticism to nearly equal degree such that their continued use is believed to less scientific and more of convenience. For instance, body mass is easily determined by its weight on the earth surface (that is near the center of earth's gravity). Determining the surface area of the human body is not a measurement that can be done with ease and with no risk. Equally, quantifying the various component of the body such as fat or lean body component is not a cheap activity as it currently involves dual energy X-ray absorptiometry, computerized axial tomography, and magnetic resonance imaging - techniques that are regarded as not "feasible and too expensive for everyday use."  These factors combined makes the use of anthropometric measurement common, convenient, and currently enduring, but with certain flaws.
The conceptual flaws and continued use of BMI has continued to draw scientifically strong criticisms.  Although BMI is currently the most popular measure of fatness, it neither tell how much fat deposition a person has nor does it delineate the distribution of body fat. These two missing points are fundamental to scientific understanding of the impact of fat on an individual health, and very importantly, evidence abound for ethnic differences in fat content and distribution at a given BMI.  However, the continued reliance on BMI for classifying people into different categories such as underweight, normal-weight, overweight, and obese is possibly connected to its ease of derivation rather than the depth of its scientific meaning. So, it appears that anthropometric indices such as BMI, WHR will remain a choice diagnostic tool for the foreseeable future.
As in many areas of medical and biomedical research, paucity of data remains a daunting challenge in understanding the true extends of overweight and obesity among Nigerian adolescents. This investigation was, therefore, undertaken to fill part of the lacuna of knowledge that currently exist in estimating the prevalence of overweight and obesity with its associated health conditions.
| Materials and methods|| |
This is an anthropometric study involving 200 adolescent boys and 200 adolescent girls (13-17 years) attending high schools in the city of Ilorin, North-Central Nigeria. The study was done in 2013. Following informed written consents, the participants were requested to fill-in questionnaires on individual demographics and family history of obesity, hypertension, and diabetes mellitus.
Subsequently, certain anthropometric parameters were taken in duplicate from each subject, including body weight, standing height, hip circumference, and waist circumference (WC). Standing height was taken with the aid of a stadiometer, with the head in the Frankfurt (orbitomeatal) plane. Body weight was recorded using a Hana mechanical personal scale (China). Waist circumference was measured with a tape placed at a point mid-way between the costal margin and the iliac crest. Hip circumference was taken using an inelastic tape placed snugly at the widest portion of the buttocks.
From the data obtained, the following calculations were made:
The classification of participants as underweight, normal-weight, overweight and obese was done according CDC procedure (CDC, 2000). Statistical analyses were performed with GraphPad prism 5.0 for Windows (GraphPad Software, San Diego California USA).
| Results|| |
The percentage distribution of BMI classification of adolescent boys based on sex and age-specific percentile is shown in [Figure 1]. The percentage of those that are underweight is very high (15%) compared to overweight and obese combined (4.5%). [Figure 2] shows the percentage distribution of BMI classification of adolescent girls based on sex and the age-specific percentile - while 6% of female participants are obese, and 8.5% were overweight, only 7% were underweight.
|Figure 1. Percentage distribution of body mass index classification of the adolescent boys based on sex and age-specific percentile. The percentage of those that are underweight is very high (15%) compared to overweight and obese combined (4.5%)|
Click here to view
The percentage of male subjects with WHR of 0.90 (that is the 75 th percentile) is 26% while 30% of female subjects at the same percentile had WHR of 0.84 [Table 1]. This figure points to the high-prevalence of central obesity among the adolescents. This is a gender-specific percentiles distribution of WHR. Our data was not large enough to construct sex-age-specific percentiles.
The relationship between BMI and WC in adolescent boys and adolescent girls are shown in [Figure 3] and [Figure 4], respectively - regression analysis shows a positive significant correlation between BMI in adolescent boys (P < 0.0001), although the correlation is weak (R2 = 0.15); among adolescent girls, regression analysis show a positive significant correlation between BMI (P < 0.0001) and the correlation is a bit strong (R2 = 0.24). Among adolescent girls, WC increases with increasing BMI, but not nearly as fast a BMI.
|Figure 2. Percentage distribution of body mass index classification of the adolescent girls based on sex and age-specific percentile. While 6% of female participants are obese, and 8.5% were overweight, only 7% were underweight|
Click here to view
|Figure 3. Relationship between body mass index (BMI) and waist circumference in adolescent boys Regression analysis show a positive significant correlation between BMI in adolescent boys (P < 0.0001) although the correlation is weak (R2 = 0.15)|
Click here to view
|Figure 4. Relationship between body mass index (BMI) and waist circumference (WC) in adolescent girls. Regression analysis show a positive significant correlation between BMI in adolescent girls (P < 0.0001) and the correlation is a bit strong (R2 = 0.24). WC increases with increasing BMI but not nearly as fast a BMI|
Click here to view
Male and female sleep hour results according to the four BMI classifications are shown in [Figure 5] and [Figure 6], respectively. One-way analysis of variance followed by the Turkey-Kramer's multiple comparison test shows that the average hours of sleep did not differ significantly among the four BMI-based categories of subjects (P = 0.77) in males, while the same in females the average hours of sleep differs only significantly between normal-weight and obese subjects (95% confidence interval = 0.07-2.31).
|Figure 5. Male sleep hours according to the four body mass index (BMI) classifications. One way analysis of variance followed by the Turkey-Kramer's multiple comparison test shows that the average hours of sleep did not differ significantly among the four BMI-based categories of subjects, the differences (P = 0.77)|
Click here to view
|Figure 6. Female sleep hours according to the four body mass index classifications. One-way analysis of variance followed by the Turkey-Kramer's multiple comparison test shows that the average hours of sleep differs only significantly between normal-weight and obese subjects, (95% confidence interval = 0.07-2.31)|
Click here to view
Results in [Table 2] show that at least 31% of obese participants do not engage in any physical exercise. This is more than triple the percentage of normal-weight subjects that do not exercise at all. Other than this, the frequency of physical exercise per week appears similar.
|Table 2: Weekly frequency of physical exercise in relation to adolescent BMI |
Click here to view
The percentage (9.1%) of underweight and overweight participants that do not consume sugar-sweetened beverages was the same [Table 3]. However, 12.5% of obese subjects reported that they do not consume soft drinks at all. This is more than double the 5% of normal subjects that do not consume sugar-sweetened beverages. No overweight and obese participants report drinking up to eight bottles of sugar-sweetened beverages per week.
|Table 3: Weekly sugar-sweetened beverage consumption in relation to adolescent BMI |
Click here to view
Familial tendencies of obesity revealed a higher percentage of overweight and obese participant reporting obesity in their family members [Table 4]. Among the overweight and obese participants, 31.58% reported that at least a member of their family was obese while only 18.78% of the underweight and normal-weight participants reported the existence of obesity in their families.
|Table 4: Percentage of BMI-classified participants reporting obesity in their families |
Click here to view
Participants volunteered information about the incidence of obesity, diabetes mellitus and hypertension in their families. In some of the families, the health conditions co-occur [Figure 7]. While obesity was the highest reported condition, co-occurrence of diabetes mellitus and hypertension was highest.
|Figure 7. Co-occurrence of obesity, diabetes and hypertension in the family of participants. Participants volunteered information about the incidence of obesity, diabetes mellitus and hypertension in their families. In some of the families, the health conditions co-occur. While obesity was the highest reported condition, co-occurrence of diabetes and hypertension was highest. OB = obesity, DM = diabetes mellitus, HYPT = hypertension|
Click here to view
| Discussion|| |
Our study aimed at understanding the twin problem of obesity and underweight in adolescents in an urban setting in North-central Nigeria. Beyond obtaining anthropometric indices, we have attempted to elucidate two important factors affecting body weight - lifestyle and genetics - by requesting participants to volunteer information about how often they engage in physical exercise per week, their average sleep hours per night and incidence of obesity and some of its associated health conditions in their families.
Our analyses reveal that the overall prevalence of overweight among male and female participants differs significantly. As shown in Figure 1] and [Figure 2], our analyses reveal that 8.5% of girls are overweight (BMI ≥ 85 th and ≤ 95 th percentile) compared to 2.5% of adolescent males. Similarly, obesity among female subjects stood at 6% while it was 2% among boys (BMI ≥ 95 percentile). Although slightly higher percentage of boys (80%) have normal-weight (BMI ≥ 5 and ≥ 85 percentile) compared to females (78.5%), the problem of underweight was more pronounced in boys compared to girls. Among the adolescent male population, 15% were underweight compared to 7% among female subjects (BMI ≥ 5).
These differences as shown by our analyses clearly contrast in two major ways with the reports of similar studies among western populations. One: The problem of overweight and obesity is not as widespread in our study population in comparison with what is obtainable in some western society as reported in some studies. ,, Two: Obesity is more prevalent among female adolescents as opposed to the situation in some Western and Asian societies where adolescent boys are more likely to be obese than adolescent girls. 
The prevalence of underweight among our subjects, particularly the male subjects, is relatively higher than what has been reported in other countries. For example, studying trends of underweight in four countries (Brazil, China, Russia, and United States of America), Wang et al.  observed differences in the prevalence of underweight in adolescents aged 10-18 years in these countries. According to their study, China had the highest (11.5% in girls, 14.4% in boys), followed by Brazilian and Russian adolescents (6.5% in girls, 10.6% in boys and 8.6% in girls, 7.7% in boys, respectively), while the United States of America had the lowest underweight prevalence (3% in girls, 3.6% in boys). Obviously, as indicated by Wang et al., these differences appear to follow national socio-economic development. However, it is impossible to fully explain the gender difference in the prevalence of underweight based on socio-economic differences. In our study, the disparity in the prevalence of underweight between male and female is seemingly worrisome as it stands at 15% for adolescent boys - a prevalence which is more than double that of their female counterparts of 7%. Interestingly, however, we noticed that our data is comparable to those of Reddy et al.,  that revealed that among the South African black population, prevalence of underweight stood at 3.9% for the adolescent female population a figure that was more than quadrupled (17%) among South African adolescent male. According to the authors of the works, malnutrition was a suspected culprit for the large disparity as the primary cause of underweight, in the absence of illness. While we do not have enough data to determine the level of malnutrition in the study population, we cannot dispense with the fact that this high level of underweight is possibly connected with the well-documented malnutrition problem that is common in sub-Saharan Africa.  While the world is fixated on the high prevalence of overweight and obesity, our opinion is that it is necessary to bear in mind that some populations still suffer from underweight occasioned by poor nutrition, and that the impact of underweight on health is overshadowed by that of obesity. Each condition merits adequate attention as a matter of optimal health. Steps should be taken at reducing both the level of underweight and obesity in any given society and among all age groups and both genders.
As the scientific basis for the use of BMI continues to be debated, other anthropometric indices have gained attention. It is, therefore, only necessary to have more than one index to be able to properly discriminate against different types of obesity. This underscores our inclusion of WHR to identify the prevalence of central obesity. It has been reported that if the use of BMI for determination of obesity were supplanted by the use of WHR, the percentage of obese persons worldwide will triple.  And more importantly, some investigations have shown that the use of WHR more accurately predict the risk of obesity-associated health outcomes such as cardiovascular diseases.  From our analysis, it is clear that if WHR were adopted to screen for obesity, a higher percentage of our study population must be classified as obese as shown in [Table 1]. We report that the WHR gender-specific 75 th percentile for male subjects is 0.9, but we suspect this to be a WHR challenge. Our suspicion hangs on the fact that many studies' recommendations for adult male WHR hover around (or are much lower than) this figure (please refer to the works of Berber et al.,  and Lin et al.,  ). This forms the basis of our inference that if WHR were adopted to discriminate against obesity, 26% of male adolescents in this study will classified as obese while 30% of female adolescent will be similarly classified. However, since no national WHR cut-off marks currently exist for Nigerian adults and adolescents - it is hard to conclude the 75 th percentile which gives a suitable cut-off mark. We believe that this requires further studies to clarify whether the 75 th percentile ought to represent the cut-off for Nigerian adolescents or that the cut-off should be raised to reflect ethnic and environmental uniqueness of the Nigerian society. Although Sanya et al.,  had attempted this fit, their work appears to have suffered from critical methodological flaws leading to inferential naivetι. Their decision to recruit both adolescent and aged subjects without discriminating their WHRs based on age and gender undermined the mean WHR they came up with. To us, any future work that will adequately address this must be rigorous and clinically-oriented to delineate at what cut-off central obesity significantly raises the risk of obesity-associated health conditions among Nigerian adolescent population.
Although we do not consider our sample size large enough to construct gender-age-specific percentiles for WHR, our analysis shows that at any given percentile, female participants exhibited lower WHR compared to male participants. This is consistent with well-documented evidence that adult and adolescent females characteristically exhibit lower WHR than males. Unfortunately, we do not find convincing relevant studies on Nigerian population with which we can compare our data since there are no recommended WHR cut-off marks for Nigerian adolescents.
Importantly, our analyses demonstrate the existence of differences between some aspects of lifestyles of underweight, normal-weight and obese subjects. First, as shown in Figure 4, it seems that the average hour of night sleep does not affect male adolescent bodyweight as classified by BMI. However, Figure 5, indicates that the average duration of nighttime sleep in obese female subjects was significantly lower compared to females with normal-weight. Although the possibility of nighttime sleep duration exerting an independent effect on BMI is still being debated, , Pileggi et al.,  have demonstrated a relationship between chronic short-sleeping-time and obesity in Southern Italian schoolchildren aged 10 years. Furthermore, our data is similar to those of Storfer-Isser et al.,  who reported an association between short-sleep time and obesity in male subjects in a prospective study that had 8.2 years of follow-up. However, unlike the data from Storfer-Isser et al.,  our data showed sleep-obesity association in female rather male participants. We have no direct explanation for these differences (at this juncture), except the fact that this may represent a peculiarity of the gender that exhibit a higher prevalence of obesity. As noted earlier, female adolescents in our study have a higher prevalence of obesity as opposed to Western male adolescents who had been shown to exhibit a higher prevalence of obesity compared to their female counterparts. ,,
Furthermore, we observed that high-proportion (90%) of our subjects engaged in different physical exercises. However, as shown in [Table 2], this did not cut evenly across the four BMI classes. Specifically, 31.3% of obese participants reported not engaging in physical exercise. This is more than triple the percentage of normal-weight subjects that do not exercise at all. Whereas more than 50% of underweight, normal-weight and overweight subjects reported their weekly frequency of exercise to be around 1-3, just a little above one-third of obese subjects had a similar frequency. Frequency of physical exercise of at least 4 times/week was similar in all BMI classes. The relationship between physical exercise and BMI is controversial as the intensity, duration and type of exercise have direct bearing on body weight, possibly more than the frequency since physical exercise may be used to build or shed weight.  We, therefore, cannot conclusively determine the impact of exercise on our participant's BMI since there were no data on other aspects of physical exercise. However, it is generally accepted that physical exercise can be used to control weight in obese persons, especially when combined with the change in dietary habits. 
On the consumption of carbonated drinks (sugar-sweetened beverages), obese subjects had the highest percentage of abstinence (12.5%). This was more than double of the 5% of normal-weight subjects who reported not consuming sugar-sweetened beverages. Consumption of an average of 1-3 bottles of carbonated drinks per week was similar among all the BMI classes. However, we notice that no overweight or obese subject report consuming more than eight bottles per week. This might represent some attempt at modifying lifestyle behavior that is perceived a risk for weight gain. While some studies have identified an association between carbonated drinks and obesity  others have failed to identify such connections. 
Another interesting finding from our analyses is the fact that obesity tends to run in the family. Our analyses convincingly indicate the existence of strong familial tendencies of obesity (and associated health conditions). In this study, obesity is the single most common condition reported as existing in families of participants. As shown in [Table 4], more than 31% of overweight and obese participants reported having at least one obese family member. This is against 18% of the underweight and normal-weight participants who reported obesity in their family. Obesity is often a risk factor for other poor health conditions such as diabetes and hypertension, hence, it's reported association with these conditions. However, as shown in Figure 7, we noticed the presence of these other health conditions independent of obesity, an indication suggesting that the obesity is just one of the many risk factors and not the only risk factor.
Meanwhile, the observation that 18% of our participants who at the time of this study were either underweight or having normal-weight, but nonetheless reported incidence of obesity in their families, require careful consideration. Importantly, it must be noted that this does not in any way invalidate the familial tendencies of obesity. In fact, the question that comes to mind should be: Do these categories of participants stand the risk of future obesity? While it is certainly beyond the scope of this study to respond affirmatively to this question, the likelihood cannot be totally dispensed with. We are convinced that future studies will be necessary to understand the factors that are involved in the transition of underweight adolescents to obese adults in Nigeria. Hopefully, such investigations will succeed in discriminating between predisposing factors that are genetic and those that are lifestyle related.
A limitation of our study is principally connected to its sample size and the lack of national reference value of cut-off marks. Taken together, these point to the need for Nigeria to have nationally representative surveys that will provide baselines for future studies.
| Conclusion|| |
Our study has revealed a comparatively low level of obesity in the population of adolescents in an urban area in Nigeria. We also uncovered the fact that the prevalence of obesity among adolescent girls triples that of the boys while the prevalence of underweight among adolescent boys is more than double that among girls. These are important findings that require further investigations into factors responsible for these wide gender differences. More importantly, there is a need for nationally representative surveys that can provide reliable cut-offs for different anthropometric indices.
| Acknowledgments|| |
We acknowledge the cooperation of the Principal of schools that allowed us access to the students, and also the students who volunteered to participate in this study.
| References|| |
|1.||Daniels SR, Arnett DK, Eckel RH, Gidding SS, Hayman LL, Kumanyika S, et al. Overweight in children and adolescents: Pathophysiology, consequences, prevention, and treatment. Circulation 2005;111:1999-2012. |
|2.||Kosti RI, Panagiotakos DB. The epidemic of obesity in children and adolescents in the world. Cent Eur J Public Health 2006;14:151-9. |
|3.||Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999-2002. JAMA 2004;291:2847-50. |
|4.||Mushtaq MU, Gull S, Abdullah HM, Shahid U, Shad MA, Akram J. Prevalence and socioeconomic correlates of overweight and obesity among Pakistani primary school children. BMC Public Health 2011;11:724. |
|5.||Kelishadi R, Gheiratmand R, Ardalan G, Adeli K, Mehdi Gouya M, Mohammad Razaghi E, et al. Association of anthropometric indices with cardiovascular disease risk factors among children and adolescents: CASPIAN Study. Int J Cardiol 2007;117:340-8. |
|6.||Katzmarzyk PT, Srinivasan SR, Chen W, Malina RM, Bouchard C, Berenson GS. Body mass index, waist circumference, and clustering of cardiovascular disease risk factors in a biracial sample of children and adolescents. Pediatrics 2004;114:e198-205. |
|7.||Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: The Bogalusa Heart Study. Pediatrics 1999;103:1175-82. |
|8.||Al Suwaidi J, Higano ST, Holmes DR Jr, Lennon R, Lerman A. Obesity is independently associated with coronary endothelial dysfunction in patients with normal or mildly diseased coronary arteries. J Am Coll Cardiol 2001;37:1523-8. |
|9.||de Koning L, Merchant AT, Pogue J, Anand SS. Waist circumference and waist-to-hip ratio as predictors of cardiovascular events: Meta-regression analysis of prospective studies. Eur Heart J 2007;28:850-6. |
|10.||Deurenberg-Yap M, Chew SK, Deurenberg P. Elevated body fat percentage and cardiovascular risks at low body mass index levels among Singaporean Chinese, Malays and Indians. Obes Rev 2002;3:209-15. |
|11.||Dietz WH, Robinson TN. Clinical practice. Overweight children and adolescents. N Engl J Med 2005;352:2100-9. |
|12.||Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999-2000. JAMA 2002;288:1728-32. |
|13.||Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007-2008. JAMA 2010;303:242-9. |
|14.||Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr 2002;75:971-7. |
|15.||Reddy SP, Resnicow K, James S, Kambaran N, Omardien R, Mbewu AD. Underweight, overweight and obesity among South African adolescents: Results of the 2002 National youth risk behaviour survey. Public Health Nutr 2009;12:203-7. |
|16.||Lartey A. Maternal and child nutrition in Sub-Saharan Africa: Challenges and interventions. Proc Nutr Soc 2008;67:105-8. |
|17.||Yusuf S, Hawken S, Ounpuu S, Bautista L, Franzosi MG, Commerford P, et al. Obesity and the risk of myocardial infarction in 27,000 participants from 52 countries: A case-control study. Lancet 2005;366:1640-9. |
|18.||Berber A, Gómez-Santos R, Fanghänel G, Sánchez-Reyes L. Anthropometric indexes in the prediction of type 2 diabetes mellitus, hypertension and dyslipidaemia in a Mexican population. Int J Obes Relat Metab Disord 2001;25:1794-9. |
|19.||Lin WY, Lee LT, Chen CY, Lo H, Hsia HH, Liu IL, et al. Optimal cut-off values for obesity: Using simple anthropometric indices to predict cardiovascular risk factors in Taiwan. Int J Obes Relat Metab Disord 2002;26:1232-8. |
|20.||Sanya AO, Ogwumike OO, Ige AP, Ayanniyi OA. Relationship of waist-hip ratio and body mass index to blood pressure of individuals in Ibadan North local government. Afr J Physiother Rehabil Sci 2009;1:7-11. |
|21.||Cappuccio FP, Taggart FM, Kandala NB, Currie A, Peile E, Stranges S, et al. Meta-analysis of short sleep duration and obesity in children and adults. Sleep 2008;31:619-26. |
|22.||Patel SR, Hu FB. Short sleep duration and weight gain: A systematic review. Obesity (Silver Spring) 2008;16:643-53. |
|23.||Pileggi C, Lotito F, Bianco A, Nobile CG, Pavia M. Relationship between chronic Short sleep duration and childhood body mass index: A school-based cross-sectional study. PLoS One 2013;8:e66680. |
|24.||Storfer-Isser A, Patel SR, Babineau DC, Redline S. Relation between sleep duration and BMI varies by age and sex in youth age 8-19. Pediatr Obes 2012;7:53-64. |
|25.||Ogden CL, Kuczmarski RJ, Flegal KM, Mei Z, Guo S, Wei R, et al. Centers for disease control and prevention 2000 growth charts for the United States: Improvements to the 1977 National center for health statistics version. Pediatrics 2002;109:45-60. |
|26.||Fogelholm M, Kukkonen-Harjula K. Does physical activity prevent weight gain - a systematic review. Obes Rev 2000;1:95-111. |
|27.||Shaw K, Gennat H, O'Rourke P, Del Mar C. Exercise for overweight or obesity. Cochrane Database Syst Rev 2006;:CD003817. DOI: 10.1002/14651858; Accessed on April 24, 2014. URL: http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003817.pub3/pdf |
|28.||Basu S, McKee M, Galea G, Stuckler D. Relationship of soft drink consumption to global overweight, obesity, and diabetes: A cross-national analysis of 75 countries. Am J Public Health 2013;103:2071-7. |
|29.||Gómez-Martínez S, Martín A, Romeo J, Castillo M, Mesena M, Baraza JC, et al. Is soft drink consumption associated with body composition? A cross-sectional study in Spanish adolescents. Nutr Hosp 2009;24:97-102. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4]