1 SCHOOL OF PUBLIC HEALTH COLLEGE OF HEALTH SCIENCES UNIVERSITY OF GHANA RISK FACTORS FOR OBESITY AMONG ADOLESCENTS IN SENIOR HIGH SCHOOLS IN THE TEMA METROPOLIS BY ABDULAI MOHAMMED SALIFU (10443330) THIS DISSERTATION IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MASTER OF PUBLIC HEALTH DEGREE JULY, 2014 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I declare that this thesis is the result of my own research. References from other people’s work have been duly acknowledged. Abdulai Mohammed Salifu Dr Samuel O. Sackey (Student) (Supervisor) Signature: ………………….. Signature: ………………… Date : ………………….. Date : ………………… University of Ghana http://ugspace.ug.edu.gh ii DEDICATION I would like to dedicate this dissertation to my mother, Ms Mariama Yayah, for her immense contribution to my life. She has been my constant source of inspiration. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENT My first gratitude goes to the Almighty for His favour upon my life and giving me the strength, wisdom and knowledge to complete this work. My sincere gratitude goes to my supervisor, Dr Samuel O. Sackey of the Epidemiology and Disease Control Department of the School of Public Health, who guided me with patience to the completion and success of this work. Special thanks to Zanno Emmanuella Billey of the Ledzokuku-Krowor Municipal Health Directorate for her support and encouragement. I want to thank the headmasters of Methodists and Datus SHS for their help and understanding. I also appreciate the support of the teachers of Methodist Day and Datus SHS. I am grateful to all the students, parents and guardians who made this research a success. I also want to thank Dr Ernest Adjepong-Tandoh of the Accident Center of Korle Bu Teaching Hospital and Vera Acheampong for their support and understanding. Finally, thank you to everyone who helped me in one way or the other to ensure success of this work. University of Ghana http://ugspace.ug.edu.gh iv ABSTRACT Introduction: Obesity in adolescents has reached alarming levels and its prevalence is continuing to rise to the extent that WHO has labeled it as an epidemic. Obesity in adolescence is of importance because the adolescents are known to carry the obesity into adulthood and develop Non Communicable Diseases (NCDs) in their early years of adulthood. Objective: The aim of the study was to determine the prevalence and risk factors for obesity among adolescents in senior high schools in the Tema Metropolis. Methods: A cross-sectional school based study was conducted on 423 adolescents in Methodist Day SHS and Datus SHS in the Tema Metroplolis. A multistage stratified random sampling was used to select the participants for the study. The students were asked to complete questionnaires that included risk factors of obesity such as dietary habits, physical inactivity, familial or genetic factors and socioeconomic status. Their weight and height were measured and used to compute their BMI. They were then categorized as obese, overweight, normal weight or underweight based on their BMI for age taking into account their sex using the CDC BMI for age growth chart. Results: Out of the 423 participants aged between 13 – 19 years, 159(37.6%) were males and 264(62.4%) were females. The overall prevalence of obesity was 15.4%. The prevalence of obesity in males and females was 6.3 % and 20.8% respectively. Seventy (16.5 %) were overweight while 3(0.7%) were underweight. The risk factors that showed significant association with obesity after adjusting for other variables were sex (AOR= 21.8, p=0.000), socioeconomic status (AOR=2.56, p=0.036), snacks consumption per week (AOR=1.88, p=0.046), not having a fat father (AOR=0.06, p=0.000), not having a fat mother (AOR=0.21, p=0.000), not having a fat University of Ghana http://ugspace.ug.edu.gh v relative (AOR=0.29, p=0.007 ), not being fat before age 10years (AOR=0.17, p=0.001) and duration of watching TV of > 4hours (AOR= 2.31, p=0.006) Conclusion: Sex was the most significant risk factor associated with obesity, with females being 21.8 times more likely to be obese compared to males. There is the need for increased awareness for obesity. Health education programs should be conducted for obese children but should target overweight children too. These health education programs must include diet, exercise and family based behavior approach. Further research is needed on obesity. Key words: Obesity, adolescents, senior high schools, Tema metropolis University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENTS DECLARATION ............................................................................................................................. i DEDICATION .................................................................................................................................ii ACKNOWLEDGEMENT .............................................................................................................. iii ABSTRACT .................................................................................................................................... iv TABLE OF CONTENTS ................................................................................................................ vi LIST OF FIGURES ........................................................................................................................ ix LIST OF TABLES ........................................................................................................................... x LIST OF ACRONYMS .................................................................................................................. xi CHAPTER ONE ............................................................................................................................. 1 INTRODUCTION .......................................................................................................................... 1 1.1 Background ........................................................................................................................... 1 1.2 Statement of problem ............................................................................................................ 4 1.3 Conceptual frame work ......................................................................................................... 6 1.4 Justification ........................................................................................................................... 7 1.5 OBJECTIVES ....................................................................................................................... 7 1.5.1 General Objective ........................................................................................................... 7 1.5.2 Specific Objectives ......................................................................................................... 8 1.5.3 Research Questions......................................................................................................... 8 CHAPTER TWO ............................................................................................................................ 9 LITERATURE REVIEW ............................................................................................................... 9 2.1 Prevalence of obesity ............................................................................................................ 9 2.2 Causes of obesity ................................................................................................................. 11 2.3 Measurement of obesity ...................................................................................................... 11 2.4 Risk factors for obesity ....................................................................................................... 13 2.4.1 Dietary habits ................................................................................................................ 13 University of Ghana http://ugspace.ug.edu.gh vii 2.4.2 Familial or Genetic factors ........................................................................................... 15 2.4.3 Metabolic or Physiological factors ............................................................................... 16 2.4.4 Physical inactivity......................................................................................................... 17 2.4.5 Socioeconomic status ................................................................................................... 19 CHAPTER THREE ...................................................................................................................... 21 METHODOLOGY ....................................................................................................................... 21 3.1 Study design ........................................................................................................................ 21 3.2 Study location or area .......................................................................................................... 21 3.3 Variables.............................................................................................................................. 22 3.3.1 Independent variables ................................................................................................... 22 3.3.2 Dependent variable ....................................................................................................... 22 3.4 Study population ................................................................................................................. 25 3.4.1 Inclusion criteria ........................................................................................................... 25 3.4.2 Exclusion criteria .......................................................................................................... 25 3.5 Sampling.............................................................................................................................. 26 3.5.1 Sampling method .......................................................................................................... 26 3.5.2 Sample size ................................................................................................................... 27 3.6 Data collection techniques or methods and tools ................................................................ 29 3.7 Quality control..................................................................................................................... 30 3.8 Data analysis and Statistical methods ................................................................................. 31 3.9 Ethical considerations ......................................................................................................... 32 3.10 Pretest or pilot study .......................................................................................................... 33 3.11 Limitations of the study..................................................................................................... 34 3.12 Assumptions ...................................................................................................................... 34 CHAPTER FOUR ......................................................................................................................... 35 4 RESULTS .................................................................................................................................. 35 University of Ghana http://ugspace.ug.edu.gh viii 4.1 Prevalence of Obesity and Characteristics of the Study Population ................................... 35 4.2. Risk factor for obesity ........................................................................................................ 52 4.2.1 Significant risk factors for obesity................................................................................ 56 CHAPTER FIVE .......................................................................................................................... 59 5 DISCUSSION ............................................................................................................................ 59 5.1 Prevalence of Obesity.......................................................................................................... 59 5.2 Risk factors for obesity ....................................................................................................... 60 5.2.1 Dietary habits ................................................................................................................ 60 5.2.2 Familial or Genetic factors ........................................................................................... 61 5.2.3 Metabolic or Physiological factors ............................................................................... 62 5.2.4 Physical inactivity......................................................................................................... 63 5.2.5 Socioeconomic status ................................................................................................... 65 CHAPTER SIX ............................................................................................................................. 66 CONCLUSIONS AND RECOMMENDATION ......................................................................... 66 6.1 CONCLUSION ................................................................................................................... 66 6.2 RECOMMENDATION ...................................................................................................... 67 6.2.1 Schools.......................................................................................................................... 67 6.2.2 Parents or Guardians ..................................................................................................... 67 6.2.3 Policy Makers and Health Institutions.......................................................................... 68 6.2.4 Research........................................................................................................................ 68 REFERENCES ............................................................................................................................. 69 APPENDIX ................................................................................................................................... 76 APPENDIX 1A: CONSENT FORM FOR STUDY PARTICIPANTS .................................... 76 APPENDIX 1B: CONSENT FORM FOR PARENT OR GUARDIAN .................................. 78 APPENDIX 1C: QUESTIONNAIRE FOR RISK FACTORS FOR OBESITY AMONG SENIOR HIGH SCHOOLS IN THE TEMA METROPOLIS .................................................. 80 University of Ghana http://ugspace.ug.edu.gh ix LIST OF FIGURES Figure 1: Conceptual framework……………………………………………………………….8 University of Ghana http://ugspace.ug.edu.gh x LIST OF TABLES Table 3.1: Study Variables….……………………………………….…………………………24 Table 4.1: Distribution of BMI status by background characteristics…………… ……………36 Table 4.2: Distribution of Obesity by background characteristics……………………………..38 Table 4.3: Distribution of Obesity by Dietary factors on breakfast, lunch and supper ………..41 Table 4.4a: Distribution of Obesity by Dietary factors on snacks, fruits and vegetable consumption…………………………………………………………………………………….43 Table 4.4b Distribution of obesity by Dietary Habits on types and frequency of food fast and sweetened beverage consumed…………………………………………………………………44 Table 4.5: Distribution of Obesity by Familial or Genetic factors and Physical Activity.…......45 Table 4.6: Bivariate logistic regression on socio-demographic factors and Dietary Habits for obesity.…………………………………………………………………………………………..48 Table 4.7: Bivariate logistic regression on Familial or genetic factors and Physical activity for obesity…………………………………………………………………………………………..50 Table 4.8: Significant risk factors for obesity…………………………………………………..54 University of Ghana http://ugspace.ug.edu.gh xi LIST OF ACRONYMS AOR ADJUSTED ODDS RATIO BMI BODY MASS INDEX CDC CENTERS FOR DISEASE CONTROL CI CONFIDENCE INTERVAL NCD NONCOMMUNICABLE DISEASE OR ODDS RATIO WC WAIST CIRCUMFERENCE WHO WORLD HEALTH ORGANISATION WHR WAIST TO HIP RATIO University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTION 1.1 Background Obesity is a disease and a known risk factor associated with many conditions such as High blood pressure, Heart disease, Stroke, Diabetes, Joint problems such as osteoarthritis, Sleep apnea and respiratory problems, gallstones, kidney stones, infertility (Bogers et al., 2007); Cancers of the esophagus, pancreas, breast, colon and rectum, gallbladder, ovary, endometrium and prostate (Wiseman, 2008); and Psychosocial effects like depression from bias and discrimination (Pan et al., 2011). Obesity is defined as a condition in which there is an abnormal or accumulation or excessive body fat that impairs or presents a risk to health (WHO, 2010). The causes of obesity are complex and multiple leading to an imbalance between energy intake and output often termed as energy imbalance between calories consumed and calories expended. WHO defines obesity as the BMI as equal to or greater than 30 kg/m2 or Body Mass Index (BMI) > 2 standard deviations above the WHO growth standard median for 5 to 19 year olds (WHO, 2006) while the CDC uses the BMI for age growth charts which takes into account the sex and categorizes obese as >95th percentile (Kuczmarski et al., 2000). There are different ways of measuring obesity but it is difficult to develop one simple index for the measurement of obesity in children and adolescents because their bodies undergo a number of physiological changes as they grow. It is however often measured as excessive weight for a given height which is the Body Mass Index (BMI). It is defined as a person's weight in kilograms divided by the square of the height in meters (kg/m2) (Swamy, 2011; WHO, 2006). University of Ghana http://ugspace.ug.edu.gh 2 Data available shows that the prevalence of obesity is increasing rapidly in developing countries (De Onis & Blössner, 2000) and this is placing significant financial burdens on health system to the extent that WHO has labeled it as an epidemic. At least 2.8 million people each year die as a result of being overweight or obese. According to WHO, obesity is the leading cause of preventable illnesses and death worldwide (WHO, 2010). Studying obesity in adolescence is of importance because the adolescents are known to carry the obesity into adulthood and hence develop Non Communicable Diseases (NCDs) early in their early years of adulthood. For most obesity related NCDs, the risks are associated with the age of onset and on the duration of obesity. It is important to note that obese adolescents tend to suffer from both short-term and long-term health consequences of obesity (WHO, 2008). Unfortunately, children and adolescents do not usually make the decisions on the environment in which they live or the food they eat. The long-term consequence of their behavior is probably not immediately clear to them hence their irresponsible dieting and physical inactivity. Westernization has led to new eating patterns, which affect dietary habits and even pattern of consumption (Lobstein, Baur, & Uauy, 2004). Changes in diets and lifestyles that has occurred with industrialization, urbanization and the world food economy has had a significant impact on the health and nutritional status of populations especially in developing countries. This has reflected in the shifting dietary patterns of increased consumption of energy dense diets that are high in fat, particularly polysaturated fatty acids and high in refined carbohydrates. This has led to protection from some nutritional deficiencies but not nutritional imbalance (Mendoza, Drewnowski, & Christakis, 2007; Mozaffarian, Hao, Rimm, Willett, & Hu, 2011; WHO, 2006). University of Ghana http://ugspace.ug.edu.gh 3 These changing dietary patterns combined with a decline in energy expenditure are associated with motorized transport, labor saving devices in the home, the decline in physical demanding manual tasks in the workplace, and the leisure time that is devoted to physically undemanding activities leads to obesity (Chatterjee & DeVol, 2012; S. Kautiainen, Koivusilta, Lintonen, Virtanen, & Rimpelä, 2005; WHO, 2010). The developing economies or countries are undergoing epidemiological transition (Ezzati et al., 2005), and are now facing a double or dual burden of disease (Boutayeb, 2006). Even though they continue to deal with the problems of infectious diseases and under-nutrition, they are experiencing a sudden increase in Non Communicable Diseases risk factors such as obesity, especially in urban settings (Yach & Beaglehole, 2004). Adolescents are exposed to foods that are high in fat, high in sugar, high in salt, energy dense, and poor in micronutrients. These foods are lower in cost and have little nutrient quality. This pattern of diet together with increased levels of physical inactivity from the increasingly sedentary nature of many forms of leisure or recreational time, changing modes of transportation, and increasing urbanization, results in sharp increases in adolescent obesity (Chatterjee & DeVol, 2012; Mendoza et al., 2007). Until recently obesity was thought to be rare in developing or low and middle income countries. It is now being reported in Africa. In a study to determine the distribution and trends of obesity in West Africa populations, urban residents and women were particularly at higher risk of obesity (Abubakari & Bhopal, 2008). University of Ghana http://ugspace.ug.edu.gh 4 Some African cultures consider weight gain and fat storage as signs of health and prosperity. Some Africans therefore make frantic efforts to put on weight (Arojo & Osungbade, 2013).. In Ghana, studies conducted on obesity in 2005 have indicated a prevalence of 5.5% among the general population, with prevalence higher in females than in males. The prevalence was 7.4% among females and 2.8% among males (Biritwum, Gyapong, & Mensah, 2005). The Ghana Demographic and Health Survey report for 2008 showed that 1.6% of female adolescents between the ages of 15 to 19 were obese with 14% of them in the urban areas and just 4.6% in the rural areas. The Greater Accra region alone had a prevalence of 19.4% for obese females, the least was 2.4% and that was recorded in both the Northern and Upper West Regions (Ghana Demographic and Health Survey Report, 2008). 1.2 Statement of problem Obesity is the leading cause of preventable illnesses and death worldwide. It is believed that obesity is associated with more deaths worldwide than underweight and 65% of the world's population are at risk (WHO, 2010). Obesity in adolescents is increasing and placing significant financial burdens on health system to the extent that WHO has labeled it as an epidemic (Reddy et al., 2012; WHO, 2010). About half of the diabetes burden, a quarter of the heart disease burden and between 7% and 41% of certain cancer burdens are attributable to obesity and overweight (WHO, 2008). Obesity in adolescence is of importance because the adolescents are known to carry the obesity into adulthood and develop non communicable diseases (NCDs) early in their early years of adulthood (WHO, 2010). The sad aspect for these obese adolescents is that they tend to suffer University of Ghana http://ugspace.ug.edu.gh 5 from both short-term and long-term health consequences of obesity (WHO, 2008). Childhood obesity is associated with a higher chance of obesity, premature death and disability in adulthood (Kalra, De Sousa, Sonavane, & Shah, 2012). Most students especially those in the urban areas like Tema Metropolis do very little or no physical activity. Many of these students vehicles as transport to school due to either long distance or their inability to go school by foot or cycling due to the unsafe nature of our roads and time constraints. As a result of technology, labor saving devices and equipment are widely used in homes for cooking, washing, cleaning and even controlling the air condition, radio and television. Children and adolescents are now using video and computer games that require sitting or lying down to participate. These limit the expenditure of energy in adolescents which results in excessive weight gain leading to obesity (Chatterjee & DeVol, 2012; Mendoza et al., 2007; Siervo, Grey, Nyan, & Prentice, 2005; WHO, 2006). Westernization has led to new eating patterns, which affect dietary habits and even pattern of consumption. Adolescents are exposed to foods that are high in fat, high in sugar, high in salt, energy dense and poor in micronutrients (Chatterjee & DeVol, 2012; Susanna Kautiainen, Koivusilta, Lintonen, Virtanen, & Rimpelä, 2005; Mendoza et al., 2007). In the face of increasing trends in obesity and overweight reliable data baseline on obesity in adolescents would be useful. There is available data on obesity as a whole but data on adolescent obesity is scanty and limited. The outcome of this research is expected to provide data to fill in the data gap and to reinforce existing studies on obesity in adolescents. University of Ghana http://ugspace.ug.edu.gh 6 1.3 Conceptual frame work Figure 1 Conceptual framework The conceptual framework shows the risk factors that lead to obesity. The main risk factors which are dietary habits, physical inactivity, physiological or metabolic, socioeconomic factors and family history or genetics are responsible for obesity. Any one of these risk factors can lead to obesity but it does not mean that someone with all these will be obese. Dietary Habits Physiological or Metabolic OBESITY Physical Inactivity Family History or Genetics Socioeconomic factors University of Ghana http://ugspace.ug.edu.gh 7 1.4 Justification Obesity is on the increase among adolescents and these adolescents tend to carry the obesity into adulthood. Obesity is associated with many conditions such as High blood pressure, Heart disease, Stroke and Diabetes (WHO, 2010). Even though studies have been done on obesity, most of these studies on obesity have been carried out on adults making data available on children and adolescents scanty and scarce. There is the need to focus on prevention of obesity especially among adolescents because they tend carry their obesity into adulthood. Obesity is also the leading cause of preventable illness and death globally (WHO, 2010). This study aims to determine the prevalence and the assess risk factors for obesity among adolescents in senior high school in the Tema metropolis to add to the pool of data currently available. The data will go a long way to help in the prevention of obesity through planning and education. The research data will also serve as a baseline data upon which further research could be done. 1.5 OBJECTIVES 1.5.1 General Objective To determine the prevalence and risk factors for obesity among adolescents in senior high school in the Tema Metropolis. University of Ghana http://ugspace.ug.edu.gh 8 1.5.2 Specific Objectives 1. To determine the prevalence of obesity among adolescents in Senior High School of Tema Metropolis. 2. To assess risk factors for obesity among adolescents in Senior High School of Tema Metropolis. 1.5.3 Research Questions 1. What is the prevalence of obesity in adolescents in Senior High Schools in the Tema Metropolis? 2. What are the risk factors for obesity in adolescents in Senior High Schools in the Tema Metropolis? University of Ghana http://ugspace.ug.edu.gh 9 CHAPTER TWO LITERATURE REVIEW 2.1 Prevalence of obesity The prevalence of obesity is rising at a fast rate worldwide especially in developing countries (Martorell, Stein, & Schroeder, 2001). Once considered a problem of only high income countries, overweight and obesity are now rising rapidly in low and middle income countries, particularly in urban settings (Abubakari & Bhopal, 2008; Amoah, 2003). While the world is concentrated on infectious diseases like malaria and tuberculosis, overweight and obesity have become global pandemics leading to dual or double burden of disease. The frightening issue with the obesity pandemic is that, it affects both children and adults(Boutayeb, 2006; WHO, 2010). A cross-sectional study using a multistage stratified random sampling technique was conducted in Kuwaiti among 5402 intermediate school adolescents aged 10 to 14 years to find out the prevalence of obesity and overweight. The overall prevalence of overweight and obesity from their study was 30.7% and 14.6%, respectively. The prevalence of obesity among females and males was 14.2% and 14.9% respectively (El-Bayoumy, Shady, & Lotfy, 2009). In another cross-sectional study conducted among children 6 to 17 years to determine prevalence of obesity in children and adolescents. The prevalence of obesity was 10.3% and 9.1% in males and females respectively, using the NHANES I definition (Savva et al., 2002). In a study of weight-for-height reference and the prevalence of obesity for school children and adolescents in Taiwan and Fuchien areas, the prevalence of childhood and adolescent obesity was 18.5% in boys and 15.0% in girls. By comparing data, they concluded that all ages, the University of Ghana http://ugspace.ug.edu.gh 10 prevalence of childhood and adolescent obesity for both males and females was higher in the 2002 compared to 1997. This evidence strongly indicates an increasing prevalence of obesity during childhood and adolescence in recent years, irrespective of sex (Huang, Wu, & Yang, 2003). Obesity has also increased throughout developed and developing countries (WHO, 2010). A study in 1989 to describe the prevalence of overweight and obesity among Brazilian adolescents was done through a nationwide home based survey involving 13,715 adolescents from 10 to 19 years of age. The prevalence of obesity and overweight was 7.7%, with 10.6% in the females and 4.8% in the males (Neutzling, Taddei, & Gigante, 2003). In a study on secular trends in overweight and obesity among Finish adolescents from 1977 to 1999. They had a sample size of 64, 147 with a response rate of 78.9%. Their study concluded that obesity and overweight increased linearly in all sex and age groups from 1977 to 1999 (S. Kautiainen, Rimpelä, Vikat, & Virtanen, 2002). A study conducted in 2005 to determine the prevalence of overweight and obesity and associated factors among students aged 11 to 17 years in Tehran, resulted in a prevalence of 17.9% and 7.1% respectively (Moayeri et al., 2006). In Bahraini, a study was conducted to determine the prevalence of overweight and obesity among of 506 students (249 males and 257 females) between the ages of 12 and 17 years. The overall prevalence of obesity among Bahraini boys and girls was higher than was previously reported. Obesity was higher in females, 35% than in males 21% (Al-Sendi, Shetty, & Musaiger, 2003). University of Ghana http://ugspace.ug.edu.gh 11 In general, research and existing data show a global increase in childhood and adolescent obesity (WHO, 2010). 2.2 Causes of obesity The causes of obesity are complex and multiple leading to an imbalance between energy intake and output often termed as energy imbalance between calories consumed and calories expended. The thrifty gene hypothesis postulates that, there are certain ethnic groups which have a higher chance of being obese when exposed to the same environment as others. This could be reason why it is the assumed that who evolved in a desert ecosystem like the Pima Indians tend to develop some of the highest rates of obesity when they are exposed to a Western lifestyles (Schulz et al., 2006). Some rare genetic conditions such as Prader-Willi syndrome, Bardet-Biedl syndrome, MOMO syndrome and Congenital leptin deficiency have obesity as the main feature (Ahmad, Ahmad, & Ahmad, 2010; Elena, Bruna, Benedetta, Stefania, & Giuseppe, 2012). 2.3 Measurement of obesity There are several ways in which obesity or body fat can be measured. The ideal method of checking body fat should accurately calculate the fat mass, be independent of other covariates of body mass such as height, be acceptable and reproducible, be inexpensive and have appropriate values of normality (Power, Lake, & Cole, 1997). It is however impossible to measure directly the body fat in vivo (RAO, 2011). The direct methods and indirect methods have therefore been developed. The direct methods consists of Bioelectric Impedance (BIA), Underwater Weighing (Densitometry), Air-Displacement Plethysmography, Dilution Method (Hydrometry), Dual Energy X-ray Absorptiometry (DEXA), University of Ghana http://ugspace.ug.edu.gh 12 Computerized Tomography (CT) and Magnetic Resonance Imaging (MRI) while the indirect methods consist of Body Mass Index (BMI), Waist Circumference, Waist-to-Hip Ratio and Skinfold Thickness (Hu & Hu, 2008). The direct methods are expensive, sometimes invasive and often require special equipments and skills. Although anthropometric measures are less accurate for measuring the excess body fat, they are less expensive, non-invasive and easy to use. The commonly used measurements are skin-fold thickness, weight and height. Anthropometry allows measurement of variations in physical dimensions and gross composition of the human body. At different ages, anthropometric measurements serve as an indicator of nutritional status (WHO, 1986). Body Mass Index is frequently used to categorize adolescents into underweight, normal weight, overweight and obese. However, this ratio provides only a crude measure of body fatness and does not distinguish between weight associated with muscles and weight associated with fat (WHO, 2006). BMI is also affected by physiological fluctuations in weight and height during growth at different age points in childhood and adolescence (Rolland-Cachera, 1995). In 1995, the WHO recommended the use of percentile BMI-for-age for adolescents 10-19 years to differentiate underweight (<5th percentile), normal (>=5th to <85th percentile), overweight (>85th-<=95th percentile) and obese (>95th percentile) (WHO, 1995). An expert committee declared BMI to be the principal measure of obesity in childhood because it is reproducible, valid and an easy measure of body fat (Bellizzi & Dietz, 1999). However, most studies have shown that BMI does not readily distinguish between body fat and muscle because University of Ghana http://ugspace.ug.edu.gh 13 it fails to take into account fat distribution making it a poor indicator of body fat (Rothman, 2008; Wickramasinghe et al., 2005). A cross-sectional analysis was conducted using 474 healthy 17year old adolescents to evaluate the diagnostic accuracy of body mass index (BMI, kg/m2), waist-circumference (WC) and waist- hip-ratio (WHR) as diagnostic tests for detecting fatness in adolescents using the measurements performed by air-displacement plethysmography as the reference. From the results, BMI and WC showed strong positive correlation in both sexes than WHR leading to a conclusion that BMI and WC performed well as diagnostic tests for fatness than WHR (Neovius, Linne, & Rossner, 2004). There are also various proposed reference values for obesity for BMI but there are not enough references for WC and WHR (Cole, Bellizzi, Flegal, & Dietz, 2000). 2.4 Risk factors for obesity The main risk factors for obesity are dietary habits, physical inactivity, physiological or metabolic, socioeconomic factors and family history or genetics. 2.4.1 Dietary habits You are what you eat. What we eat and how we eat influences energy balance and therefore influence our weight. The macronutrient content of a diet determines the extent to which excess is stored and about 80% of the excess energy may be stored after carbohydrate overfeeding. The capacity for storage of carbohydrates is smaller compared to the unlimited capacity for storage of fat. However, when excess carbohydrate in the body is not used it is converted to body fat (Joosen & Westerterp, 2006). Dietary fat tends to induce appetite and therefore encourages one to eat more than is needed (Drewnowski & Almiron-Roig, 2010). University of Ghana http://ugspace.ug.edu.gh 14 From the WHO/FAO Expert Consultation Report, there is convincing evidence that the factors that increase the risk of obesity include sedentary lifestyle, high intake of energy-dense and micronutrient – poor foods, increased number of fast-food outlets and marketing of energy-dense foods, high intake of sugars-sweetened carbonated drinks and fruit juices as well as low consumption of fruits and vegetables (Nishida, Uauy, Kumanyika, & Shetty, 2004). In a randomised controlled trial of a primary school based intervention to reduce risk factors for obesity in Leeds, it was noticed that consumption of vegetable by 24 hour recall was lower in children in the control group than the intervention group. In the same study, consumption of fruit was lower in obese children in the intervention group compared to those in the control group. Similarly in that study, a three day recall diary showed a low consumption of highly sugary foods among overweight children in the control group than the intervention group. The study involved 634 primary school children aged 7 to 11 years (Sahota et al., 2001). There appears to be a basic shift in dietary patterns, induced mainly by shifts in income, prices and food availability, and also by the modern food industry and the mass media. There are marked differences between urban and rural eating patterns, particularly regarding the consumption of food prepared away from home (Popkin, 2006). Breakfast and lunch habits are associated with both overweight and obesity, eating breakfast was positively associated with not being overweight and obese (Grøholt, Stigum, & Nordhagen, 2008). Skipping and or infrequent intake of breakfast at home, frequent consumption of fast foods, low servings of fruits and vegetables per day, and frequent consumption of sweets, candy and carbonated drinks were all predictors of obesity and overweight among the school children in Saudi Arabia (Amin, Al-Sultan, & Ali, 2008). University of Ghana http://ugspace.ug.edu.gh 15 The shift from traditional staples to processed foods in these developing (especially urban areas) is strongly enhanced by borrowed western culture (Neutzling et al., 2003). Eating western food has become widely accepted in developing countries. Adolescents form part of this group of patrons. And as long as parents continue to leave the decision concerning what to eat to their wards, their wards are bound to patronize foods that may be unbalanced and unhealthy (WHO, 2010). 2.4.2 Familial or Genetic factors Some conducted studies have shown a strong predisposition that obesity can be inherited. Researchers have identified several genes that appear to be associated with obesity but most believe that one gene alone is not responsible for the entire obesity epidemic. From the results of study conducted on obesity in children and adolescents in Cyprus, the most significant associated factor for obesity was parental obesity status. The prevalence of obesity in the study in males was 10.3% and in females 9.1% (Savva et al., 2002). Similarly, a study conducted earlier in Italy showed that parental obesity was the main risk factor for obesity in children (Maffeis et al., 2006). The risk of childhood and adolescent obesity has been found to be associated with high parental body mass in another study conducted in Sao Paulo, Brazil. The study results showed that the risk of childhood obesity was associated with having one or both parents obese, the risk increased when both parents were found to be obese (da Costa Ribeiro, Taddei, Colugnatti, & others, 2003). A study has also shown an association between BMI of friends, siblings and spouses (Christakis & Fowler, 2007). University of Ghana http://ugspace.ug.edu.gh 16 In a case control study to determine the risk factors associated with overweight and obesity among adolescents enrolled in private high schools in the city of Pelotas, southern Brazil, the results showed that overweight and obesity were positively associated with mother and father’s BMI being greater than 30. Parental nutritional status was directly and significantly associated with overweight and obesity in adolescents. Adolescents whose parents’ BMI were greater than 30 were two times more likely to be overweight or obese than those whose parents had BMI less than 30 (Neutzling et al., 2003). 2.4.3 Metabolic or Physiological factors Overweight and obesity tends to be associated with the sex of individuals, being more prevalent in girls than in boys, a study done in Kuwait showed that obesity was prevalent in 14.9% of females and 14.2% of males. The BMI in females was also higher than in males. This denotes that the level of fatness among adolescent girls was found to be higher than among boys (El- Bayoumy et al., 2009). During adolescence, and especially during puberty, both girls and boys undergo fat storage, and because the spurt of height in boys is higher than in girls, the BMI of girls appears to be higher than boys (Poskit, 1988). In addition, boys lose their body fat during puberty faster than girls, and the fat deposition among girls still continues through puberty (Meredith & Dwyer, 1991). The difference in obesity between males and females can be further explained by the differences in the level of physical activity between them, that is, boys tend to be more physically active than girls (Sallis, 1993). University of Ghana http://ugspace.ug.edu.gh 17 However, some studies have showed higher prevalence of obesity in males than females, a study conducted showed the prevalence of obesity was 3.8% for boys and 3.3% for girls (Georgiadis & Nassis, 2007). In a case control study to determine the risk factors associated with overweight and obesity among adolescents enrolled in private high schools in the city of Pelotas, southern Brazil. Childhood overweight showed a direct relationship to overweight and obesity during adolescence. Adolescents who reported being overweight before they were 10 years old were 2.5 times more likely to be overweight or obese during their adolescence (Neutzling et al., 2003). 2.4.4 Physical inactivity Physical activity is defined as any bodily movement produced by skeletal muscles that requires energy expenditure (WHO, 2010). Literature available have reported that children and adolescents are more active than adults (Pratt, Macera, & Blanton, 1999). In a study conducted from 2000 to 2004, physical inactivity was positively associated with being overweight and obese. There were more overweight and obese adolescents who were physically inactive compared to those who were physically active (Grøholt et al., 2008). In a randomised controlled trial of 634 primary school children aged 7 to 11 years in Leeds to reduce risk factors for obesity, it was noticed sedentary behaviour was higher in overweight children in the intervention group than the control group (Sahota et al., 2001). The association between overweight and obesity and factors related to physical activity showed that adolescents using bus or car transport to school were about 30% more likely to be obese compared with adolescents who cycled or walked. Furthermore, whereas overweight increased University of Ghana http://ugspace.ug.edu.gh 18 moderately with declining physical activity outside school, the association with obesity was much stronger, with a doubling from the high to the low activity group. There was also an independent association between overweight and obesity and TV/PC use. Adolescents watching television/video 3 hours or more per day were about 60% more likely to be obese compared with adolescents watching television/video 0 to 2 hours per day (Savva et al., 2002) . The American Academy of Peadiatrics states that children and adolescents need 1-2hours a day on screens but it has however estimated that the average child spends over 7 hours a day on internet usage, TV viewing and video games (American Acamedy of Pediatrics, 2013). Sedentary lifestyle is positively associated with obesity, the amount of time spent in front of the television during adolescence has been found to be significantly associated with adiposity even after correction for obesity history (S. Kautiainen et al., 2005). Studies were reviewed to assess the role of television as tool for childhood obesity prevention, results showed that television watching or viewing replaced more vigorous activities; there was an association between time spent watching or viewing television and being overweight or obese; the prevalence of obesity and the number of hours that TV networks dedicate to children have also increase increased; finally for the last 30 years, the rate at which children watch television for more than 4 hours per day has also increased (Caroli, Argentieri, Cardone, & Masi, 2004). Physical activity and self-perceived physical fitness assessed in adolescents aged 16 to 18 years of age were used to predict the development of obesity in a study. The results showed that physical inactivity in adolescence strongly predicted the risk for obesity. The study also showed University of Ghana http://ugspace.ug.edu.gh 19 that poor physical fitness in adolescence also increased the risk for overall obesity. They concluded that physical inactivity in adolescence strongly and independently predicts total (and especially) abdominal obesity in young adulthood and therefore physical activity should be seriously recommended for obesity prevention in the young (Pietiläinen et al., 2008). According to the Ghana Health Service in 2007, physical inactivity contributed to the increasing prevalence of overweight and obesity and its associated diseases in Ghana (Ghana Health Service, 2007). In a cross sectional study conducted to compare the physical activity pattern of children, their results showed that obese children exhibited significantly lower daily accumulations of total counts, significantly lower levels of physical activity self-efficacy and were also involved in significantly fewer community organizations promoting physical activity (Trost, Sirard, Dowda, Pfeiffer, & Pate, 2003). 2.4.5 Socioeconomic status The relationship between social class and BMI varies globally. In the developed world pediatric obesity is generally more common in children and adolescents from families of lower socioeconomic status whereas in the developing world, higher socioeconomic status has usually been associated with higher risk of paediatric obesity (WHO, 2010). In a study carried out in Kuwait, 89.2% of the obese children belonged to families with high socioeconomic status (earning ≥1000 Kuwaiti Dinars), while 6.7% of them belonged to middle social class families (families earning 500 to <1000 Kuwaiti Dinars), and 4.1% of them belonged to families with low socioeconomic status (earning <500 Kuwaiti Dinars) (El-Bayoumy et al., 2009). University of Ghana http://ugspace.ug.edu.gh 20 In the developed countries women of a high social class are less likely to be obese while in the developing world, women, men, and children from high social classes have higher rates of obesity (McLaren, 2007). There appears to be a link between obesity and income levels of the families. Some studies have shown that obesity is prevalent among low socioeconomic class in developed countries but more prevalent among high socioeconomic class in developing countries (Burns, 2004). In the study to assess the magnitude of overweight and obesity and its associated socio- demographic factors in adolescents attending junior high schools in Xi’an, Shanxi province in China, 1804 adolescents were examined. Overweight and obesity were more prevalent in younger boys from richer families living in urban districts and those whose parents were either overweight or obese (Li, Dibley, & Yan, 2011). In a study conducted among 15 and 16 year olds in Norway , those with lower educational plans and poor family socioeconomic status were both significantly associated with overweight and obesity. In the same study, when compared with adolescents living in families with high socioeconomic status, the odds of one becoming overweight and obese was about 1.5 times higher among adolescents living in families with low socioeconomic status (Grøholt et al., 2008). A direct relation was established between the socioeconomic level and overweight and obesity. For adolescents in the most industrialized region of the country they had a higher risk for overweight and obesity than those in the least industrialized region. They also found that male adolescents who lived in the urban areas were at a higher risk of being overweight and obese than those in the rural areas (Neutzling et al., 2003). University of Ghana http://ugspace.ug.edu.gh 21 CHAPTER THREE METHODOLOGY 3.1 Study design A cross sectional study was conducted among adolescent students (aged 10 to 19 years) from Methodist Day and DATUS Senior High Schools in the Tema Metropolis using quantitative and qualitative data. The study took place from June to July 2014. 3.2 Study location or area Tema is the administrative capital of the Tema Metropolitan District in the Greater Accra Region of Ghana with a population of about 402,637 people. It is a coastal city situated 25 kilometres east of the national capital, Accra. The Greenwich Meridian (00 Longitude) passes through the city of Tema. It has a deep seaport that was developed in 1952 and became a Metropolitan in 1990. The Metropolitan shares boundaries with Ashaiman Municipal, Adentan Municipal, and Ledzokuku-Krowor Municipal to the west, to the east with Kpone Katamanso District, to the North with Dangme West District and to the South with the Gulf of Guinea. Towns in This Metropolis: Tema Community One to Eleven, Tema Newtown, Lashibi and Bethlehem. Tema has 13 Senior High Schools, Six (6) are public and Seven (7) are private. University of Ghana http://ugspace.ug.edu.gh 22 3.3 Variables 3.3.1 Independent variables Age Sex Ethnicity Level of education (class) Residential area Residential status in school Socioeconomic status of Parents or Guardians Dietary intake and habits Family history Physical activity level Weight Height. 3.3.2 Dependent variable BMI(Body Mass Index)- Obesity University of Ghana http://ugspace.ug.edu.gh 23 Table 3.1: Study Variables Study Variable Operational Definition Measurement Scale Data Collection Tool/ Technique Age Age at last birthday Continuous Questionnaire Sex Sex of student  Male or Female Nominal 1. Male 2. Female Questionnaire Ethnicity Ethnic group of association by birth Nominal 1. Ga 2. Fante 3. Ewe 4. Dagomba 5. Others Questionnaire Level of education (class) Current class of student  SHS1, SHS 2 or SHS 3 Nominal 1. SHS1 2. SHS 2 3. SHS 3 Questionnaire Residential area Residential area of the student when not in school Nominal 1. Tema 2. Ashaiman 3. Kpone 4. Others Questionnaire Residential status in school Whether student resides in school or comes to school from home daily  Day or Boarding Nominal 1. Day 2. Boarding Questionnaire Socioeconomic status of Parents or Guardians Highest level of education, occupation and monthly income of the father and/or mother or guardian; Type of residence; Ownership and number of some named items.  Low, Medium or High Ordinal 1. Low 2. Medium 3. High Questionnaire and Modified Kuppuswamy scale Dietary Intake and Habits What one ate for breakfast, lunch, supper and snacks How many times one eats breakfast, lunch, snacks, supper, fruits and vegetables in a week  <3 days/week or >3days/week Nominal 1. Beverages(e. g. koko, tea) 2. Grains(e.g. rice, waakye) 3. Solid starchy foods(e.g. fufu, banku) 4. OTHER Binary 1. <3 days/week 2. >3days/week Questionnaire University of Ghana http://ugspace.ug.edu.gh 24 How many times one eats in a day  3 or others Patronizing fast foods and sugar sweetened drinks  Yes or No Binary 1. 3 2. Others Binary 1. Yes 2. No Family History Perception of a fat father or mother or guardian or relative or being fat yourself before age 10 years?  Yes or No Binary 1. Yes 2. No Questionnaire Physical Activity Level Engaging consciously in physical activity  Yes or No How long one engages in physical activity  < 30minutes or > 30minutes Number of hours spent watching TV OR Movies in a day  < 4HOURS or > 4 HOURS Physical Education as part of the school curriculum  Yes or No Binary 1. Yes 2. No Binary 1. < 30minutes 2. > 30minutes Binary 3. < 4HOURS 4. > 4 HOURS Binary 1. Yes 2. No Questionnaire Weight The weight to the nearest 0.1kg with students standing with heels, buttocks and upper back in a straight line in a complete upright position against the height rod of the machine on the digital scale without shoes, with feet together and arms at the sides and in light clothing with emptied pockets Continuous Digital column weighing scales with height rod (with BMI calculation) Height The height to the nearest 0.1cm with students standing with heels, buttocks and upper back in a straight line in a complete upright position against the height rod of the machine on the digital scale without shoes, with feet together and arms at the sides and in light clothing with emptied pockets Continuous Digital column weighing scales with height rod (with BMI calculation). BMI BMI=Weight(kg)/height(m) 2 Obese: BMI for age > =95th percentile Overweight: BMI for age >= 85th and < 95th percentile Normal: BMI for age >= 5th and <85th percentile Underweight: BMI for age <5th percentile  Obese or Not Obese Binary 1. Obese 2. Not Obese Digital column weighing scales with height rod (with BMI calculation) and CDC BMI for age growth chart. University of Ghana http://ugspace.ug.edu.gh 25 3.4 Study population The study population were adolescent students (aged 10 to 19 years) from Methodist Day and DATUS Senior High Schools in the Tema Metropolis. 3.4.1 Inclusion criteria: The following were included in the study; Senior High School students aged 10 to 19 years from Methodist Day and DATUS Senior High Schools in the Tema Metropolis. All students aged 18years and above who consented or all students aged 10 to 17 years whose parents consented The age of 10 to 19 years old represent the age range for adolescents. 3.4.2 Exclusion criteria: The following were excluded from the study; All those below 10years All those above 19years were excluded Any student who does not attend Methodist Day or DATUS SHS in Tema Metropolis. Any student aged 18years who declined to participate in the study or any students aged 10 to 17 years whose parents declined consent. Any student who was seriously ill and could not take part in the study. University of Ghana http://ugspace.ug.edu.gh 26 3.5 Sampling 3.5.1 Sampling method Multistage stratified random sampling technique was used. There are 13 Senior High Schools in the Metropolis made up of six (6) public and seven (7) private Senior High Schools. The Senior High Schools were divided into two groups - Private and Public schools. Two Senior High Schools were selected by random sampling from each of the two groups: One public and one private SHS were picked by the balloting technique. This idea was to enrich the data with students with different socioeconomic status backgrounds. Systematic sampling was used for the selection of the students in each school. For each of the schools a master list was obtained and numbered. The first student for the study was chosen at random by balloting. After the first student had been chosen, the next students were chosen by a sampling interval. The interval was calculated by dividing the sample size by the number of classrooms. The value obtained was then be divided by the number of students in each class to obtain the sampling interval. Any student who refused was replaced with the next student on the list. This same pattern was used for any other student who refused to participate in the study. University of Ghana http://ugspace.ug.edu.gh 27 3.5.2 Sample size The sample size can be calculated using the Cochran’s formula(Cochran, 1977) : N= deff z2pq d2 N is the Sample size deff is the design effect (the ratio between the variance of the cluster design to the variance that would be obtained from a simple random sampling) z is the Confidence limits p is the assumed prevalence of the dependent variable q is given by 1-p d is the acceptable deviation from the true value For this study deff= 2 z= 1.96 for CI at 95% p= 15% =0.15 (an assumed prevalence of obesity among adolescents in Ghana of 15%) q=1-0.15= 0.85 d= 5%= 0.05 Sample size, N= 2 x 1.962 (0.15) x (0.85) 0.052 N= 392 University of Ghana http://ugspace.ug.edu.gh 28 A non response rate of 10% was then factored into the sample size: 392 = 435.5 (1-0.1) This was approximated to 436 (Four hundred and thirty six) but a total sample size of 440 (Four hundred and forty) was used. Study participants per selected school Proportionate sampling was used to determine the number of participants per each senior high school taking into account their respective total population. This was because the two schools had different populations. Methodist Day SHS has a total population of 736 students while DATUS SHS has a total population of 95 students making the total student population for the two schools 832. Sample size for Methodist Day SHS; with a student population of 736 students N for Methodist Day SHS = 736 x 440 = 389. 7 832 This was approximated to 390. Sample size for DATUS SHS; with a student population of 95 students N for DATUS SHS= 95 x 440 = 50.3 832 This was approximated to 50. 390 students from Methodist Day SHS and 50 students from DATUS SHS were recruited for the study. University of Ghana http://ugspace.ug.edu.gh 29 Out of the 440 questionnaires, 423 questionnaires were fully completed and analysed. The overall response rate was 96.1 % (423/440). 3.6 Data collection techniques or methods and tools Interviews using self-administered questionnaires were conducted to obtain background information on students and to elicit the risk factors such as dietary habits, physical activity and family history. Questionnaires were distributed to the students and the questions explained to them to make sure they did not leave questions unanswered. The students were also supervised to make sure that they answered the questionnaires independently. However, the question on the monthly income was added to the consent form for the parents or guardians to answer. The socioeconomic status of the parents were assessed using both the modified Kuppuswamy scale (Kuppuswamy, 1981) and the modified version of the socioeconomic status questionnaire used by Balogun et al (1990). Based on the summative score, the participants were categorized into into three levels- Low, Middle and High. The scoring of the questionnaire items was based on the head of the family’s highest level of education, occupation and monthly income. However, the following were also taken into consideration while assessing the socioeconomic status of the participants-Highest level of education, occupation and monthly income of the father and/or mother or guardian (If participant is not living with you parents); Location of residence; Type of residence(owned by parents, owned by guardian, family house or rented); Total number of rooms in your residence; Ownership and number of the following items (Television, DVD player, refrigerator, microwave, generator, car, personal computer or laptop, electric or gas stove, motorbike; Subscription to any TV NETWORK ( DSTV, Multichoice) University of Ghana http://ugspace.ug.edu.gh 30 Height, weight and BMI were measured with digital column weighing scales with height rod (with BMI calculation). The students were made to stand on the digital scale without shoes, with feet together and arms at the sides and in light clothing with emptied pockets. The students stood with heels, buttocks and upper back in a straight line in a complete upright position against the height rod of the machine. The height and weight were then measured to the nearest 0.1cm and 0.1kg respectively. The BMI was then calculated with the same machine and then recorded. The height and weight were done twice by two different pairs of research assistants and the average was used. Before each reading the digital scale was checked to make sure it was on zero kg. 3.7 Quality control The whole process of data collection was standardized to obtain uniform and of high quality data. The research assistants were trained to make sure they knew the objectives and methodology of the study. They were trained to use consistent and correct techniques through demonstration and role playing. They were trained with WHO recommended measurement protocols. The researcher supervised the data collection process and ensured there were no protocol deviations. The digital column weighing scales with height rod (with BMI calculation) were calibrated daily and before each measurement and placed on a flat floor before taking weights and height. University of Ghana http://ugspace.ug.edu.gh 31 The height and weight for each student was measured two times by two different pairs of research assistants. The mean weight and height was then used to minimize intra-observer and inter-observer biases. The student were supervised and told to answer the questionnaires independently. Questionnaires were checked for completeness before they were accepted. Questionnaires were numbered during data entry to ensure that the questionnaires were not entered twice. Data entry clerks were closely monitored; data was entered twice by two different data entry clerks to make sure data was correctly entered. 3.8 Data analysis and Statistical methods Data entry, cleaning and analysis was done using EpiData 3.1, Microsoft Excel 2010 and Stata 11. The data were inspected and sorted. There were no missing values. The data was described using descriptive statistics. BMI was determined as weight in kg divided by the height in metre-squared, this was calculated directly by the digital machine used for the measurements. Subjects were classified based on their BMI for age taking into account their sex, as obese, overweight, normal or underweight but with emphasis on the obese status. Those with BMI for age > =95th percentile were considered obese, those with BMI for age >= 85th and < 95th percentile were considered overweight, those with BMI for age >= 5th and <85th percentile were considered normal and those with BMI for age <5th percentile were considered underweight. The classification was done using the CDC BMI for age growth chart. University of Ghana http://ugspace.ug.edu.gh 32 Univariate analysis was done using a Chi-square test to identify the association between the dependent and independent variables. A p-value of less than 0.05 was considered statistically significant throughout the data analysis. Bivariate analysis was carried out on variables that showed association in the univariate analysis. The variables that showed statistical significance in the bivariate analysis were inco-operated into a multivariate logistic regression model for further analysis to determine the most significant risk factors for obesity. The adjusted and unadjusted odds ratios at 95% confidence interval was determined using Stata 11. 3.9 Ethical considerations The Ethical Review Committee of the Ghana Health Service approved the study. Permission was obtained from the Ghana Education Service, Tema Metropolitan Health Management Team and the School authorities before the research was started. The students gave their consent before they were recruited into the research. No student was forced or coerced to take part in the study. They were made to know that participation was voluntary and there was no penalty for refusing to participate. Any student who refused to participate in the research was respected. Parents or Guardians were also asked to give consent before the data was taken from the adolescents or students. An informed written consent was used for both parents or guardians and the study participants. All those who gave consent were assured of confidentiality and anonymity. University of Ghana http://ugspace.ug.edu.gh 33 Participants were made to know that there were no incentives and they could withdraw at any time during the study. The procedures that were used did not cause any physical, emotional or mental harm. Participants were told their BMI status and those who were obese were educated on the health implications of obesity and how to reduce their weight through regular physical activity and healthy eating habits. Obese participants were also advised to see dieticians and experts in physical activity to help them maintain a healthy weight. The data obtained was analyzed solely for the objective of the study and utmost discretion was exercised in the handling of the personal information provided. The questionnaires were secured in a locked cupboard. Data from the study was password- protected and stored on an external storage device. 3.10 Pretest or pilot study Pre testing of the questionnaires was carried out among 20 adolescents in Achimota Senior High School. This school is not in the area of study but has similar characteristics as the schools in the study area because it is also in a metropolis. This process enabled me to clarify the adequacy of the questions, reaction of the respondents to the research questions, estimate the approximate time for each measurement and help make the necessary corrections or adjustments for the questionnaire for the actual study. Pretesting was also done to enable the data collectors practice the data collection technique. University of Ghana http://ugspace.ug.edu.gh 34 3.11 Limitations of the study SHS 3 students were on vacation so they did not participate in the study. Only SHS1 and SHS2 students participated in the study. The consent forms which were given out to the parents or guardians resulted in some delayed response because some parents or guardians were not at home. Some parents or guardians prevented their adolescents from taking part in the research even though those adolescents really wanted to take part. The BMI of the parents or guardians were not checked. 3.12 Assumptions It was assumed that participants gave accurate and honest information, and the non-availability of incentives from the principal investigator did not skew the responses to any direction. University of Ghana http://ugspace.ug.edu.gh 35 CHAPTER FOUR 4 RESULTS 4.1 Prevalence of Obesity and Characteristics of the Study Population Four hundred and twenty (423) three Senior High School students aged between 13 – 19years participated in the study. Three hundred and eighty one (90.1%) students were from Methodist Day Senior High School and 42(9.9%) were from Datus Senior High School. This was made up of 159(37.6%) males and 264(62.4%) females. For Methodist Day SHS, there were 135 males and 246 females while Datus SHS had 24 male and 18 female participants. Majority of the participants were in SHS 1(69.3%) and day students (92.7%). All the boarding students 31(7.3%) came from Datus SHS. The overall prevalence of obesity was 15.4%. The prevalence of obesity in Methodist Day Senior High School and Datus Senior High School was 15.0% and 19.1% respectively. The overall prevalence of obesity amongst males and females was 10(6.3%) and 55(20.8%) respectively. The mean age was 16.8 ±1.12 years for all the participants with majority 147(34.8%) being 17years of age. The mean age was 16.64± 1.12 years in males and 16.84±1.12years in females. The average height was 1.6±0.07 metres with males having an average height of 1.65± 0.08metres and females 1.59± 0.06metres. The average weight was 61.2±11.8 kilograms in both sexes but it was 61.24 ±10.6kilograms in males and 62.20 ±12.5 kilograms in females. The average BMI was 23.6± 4.53 kg/m2. The mean BMI for females and males was 24.26±4.90 kg/m2 and 22.39 ±3.56 kg/m2 respectively. University of Ghana http://ugspace.ug.edu.gh 36 Table 4.1 shows the distribution of BMI status by background characteristics. The BMI was categorized as obese, overweight, normal and underweight according to guidelines given by the CDC BMI for age growth chart. Majority of the participants 285(67.4%) were normal while 65(15.4%) were obese, 70(16.5 %) were overweight and 3(0.7%) were underweight. The normal category had the most number of participants with respect to the demographic characteristics. All the underweight participants, 3(100%) were females. There were more obese females 55 (20.8%) than overweight females 53(20.1%). There were 3(100.0%) participants who were 17years old. The underweight participants were all in Methodist Day SHS. University of Ghana http://ugspace.ug.edu.gh 37 Table 4.1 : Distribution of BMI status by background characteristics Variable Obese (n=65) Overweight(n=70) Normal(n=285) Underweight(n=3) No. (%) No. (%) No. (%) No. (%) Sex Male 10 (6.3) 17 (10.7) 132(83.0) 0(0.0) Female 55(20.8) 53 (20.1) 153(58.0) 3(100) Age 13 2 (100) 0(0.0) 0(0.0) 0(0.0) 14 0(0.0) 0(0.0) 6(100) 0(0.0) 15 5(12.2) 8(19.5) 28(68.3) 0(0.0) 16 26(21.1) 8(6.5) 89(72.3) 0(0.0) 17 17(11.6) 18(12.2) 109(74.2) 3(2.0) 18 9 (11.5) 24(30.8) 45(57.7) 0(0.0) 19 6(23.1) 12(46.2) 8(30.8) 0(0.0) School Methodist Day SHS 57(15) 60(15.8) 261(68.5) 3(0.8) DATUS SHS 8(19) 10(28.3) 24(57.1) 0(0.0) Educational level SHS 1 50(17.1) 40(13.7) 200(68.3) 3(1.0) SHS 2 15(11.5) 30(23.1) 85(65.4) 0(0.0) Ethnicity Ga 26 (22.41) 17(14.7) 73(62.9) 0(0.0) Ewe 22(20.4) 19(17.6) 67(62.0) 0(0.0) Akan 6(7.8) 12(15.6) 59(76.6) 0(0.0) Fante 11(15.3) 11(15.3) 47(65.3) 3(4.2) Others 0(0.0) 11(22.0) 39(78.0) 0(0.0) Residential Area Tema 40 (22.0) 34 (18.7) 108(59.3) 0(0.0) Ashaiman 19 (13.3) 18 (13.3) 103(72.0) 3(0.0) Others 6(6.1) 18(18.4) 74(75.5) 0(0.0) Residential status in school Day 57 (14.5) 64 (16.3) 268(68.4) 3(0.8) Boarding 8(25.8) 6(19.4) 17(54.8) 0(0.0) Socioeconomic Status (SES) Low SES 9 (15.8) 6 (10.5) 42(73.7) 0(0.0) Middle SES 21(10.3) 35(17.2) 144(71.0) 3(1.5) High SES 35 (21.5) 29 (17.8) 99(60.7) 0(0.0) N=423 University of Ghana http://ugspace.ug.edu.gh 38 Table 4.2 shows the distribution of obesity by background characteristics. There were 264(62.4%) females and 159(37.6%) males. There were 55(20.8%) obese females and 10(6.3%) males with a strong evidence of association (p= 0.000). Majority of the participants 147 (34.8%) were aged 17 years but majority of the obese participants 26(21.1%) were 16years old. None 0(0.0%) of the 14year olds were obese. Age showed a strong association with obesity (p= 0.000). Datus SHS had a high prevalence of obesity of 19.1 % compared to 15.0% for Methodist Day SHS. The school of a participant was insignificant (p= 0.399). There were 293(69.3%) SHS1 students and 190(30.7 %) SHS2 students. There were 50(17.1%) obese SHS1 students as against 15(11.5%) obese SHS2 students. The level of education showed no association with obesity (p= 0.055). There were 116(27.4%) Gas forming the majority and this same ethnic group had the most obese participants 26(22.41%). Ethnicity showed a strong association with obesity (p= 0.000). Majority of the participants 182(43.0%) were resident in Tema. Those resident in Tema had the most obese participants 40(22.0%), while those resident in Others had 6(9.52%) obese participants. Residential area showed a significant association with obesity (p= 0.008). Residential status in school was made of 392(92.7%) day and 31(7.3%) boarding participants. 57 of the day students and 8 of the boarding students were obese respectively, however the boarding students had a higher prevalence of obesity of 25.8% compared to the day students. The residential status in school was nonetheless insignificant with obesity (p= 0.314). University of Ghana http://ugspace.ug.edu.gh 39 Majority of the participants 203(48.0%) were of middle socioeconomic status, 163(38.5%) were of high socioeconomic status and 57(13.5%) were of low socioeconomic status. However, majority of the obese participants 35(21.5%) were of high socioeconomic status and the least number of obese participants 9(15.8%) were of low socioeconomic status. The socioeconomic status showed significance with obesity (p=0.032). University of Ghana http://ugspace.ug.edu.gh 40 Table 4.2 : Distribution of obesity by background characteristics Variable Participants Obese Pearson (ᵡ²) No. (%) No. (%) (p-value) Sex ᵡ²= 30.000 Male 159 (37.6) 10 (6.3) p= 0.000 Female 264(62.4) 55 (20.8) Age ᵡ²= 65.944 13 2 (0.5) 2(100.0) p= 0.000 14 6(1.4) 0(0.0) 15 41(9.7) 5(12.2) 16 123(29.1) 26(21.1) 17 147(34.8) 17(11.6) 18 78 (18.4) 9(11.5) 19 26(6.2) 6(23.1) School ᵡ²= 2.952 Methodist Day SHS 381(90.1) 57(15.0) p= 0.399 DATUS SHS 42(9.9) 8(19.1) Educational level ᵡ²= 8.065 SHS 1 293 (69.3) 50(17.1) p= 0.055 SHS 2 130 (30.7) 15(11.5) Ethnicity ᵡ²= 57.976 Ga 116(27.4) 26 (22.41) p= 0.000 Ewe 108(25.5) 22(20.4) Akan 77 (18.2) 6(7.8) Fante 72(17.0) 11(15.3) Others 50(11.9) 0(0.0) Residential Area ᵡ²= 27.116 Tema 182 (43.0) 40 (22.0) p= 0.008 Ashaiman 143 (33.8) 19 (13.3) Others 98(23.2) 6(9.52) Residential status in school ᵡ²= 3.550 Day 392 (92.7) 57 (14.5) p= 0.314 Boarding 31 (7.3) 8(25.8) Socioeconomic Status( SES) ᵡ²= 13.789 Low SES 57 (13.5) 9 (15.8) p= 0.032 Middle SES 203 (48.0) 21(10.3) High SES 163(38.5) 35 (21.5) University of Ghana http://ugspace.ug.edu.gh 41 Table 4.3 shows the distribution of obesity by dietary habits on number of meals per day, breakfast, lunch and supper. Most of the participants 343(81.1%) ate 3meals a day, 51(12.0%) ate 4 times in a day and 29(6.9%) ate 2 times in a day. No participant ate once a day or more than 4 times a day. Fifty one of the obese participants ate 3 times a day and the rest of the obese participants ate 4 times a day. No obese participant ate 2 times a day. The number of meals per day showed strong statistical significance with obesity (p=0.001). Majority of the participants 220(52%) had breakfast daily while 57(13.5%) had breakfast 1-3 days/week. Thirty four (15.5%) of those who ate breakfast daily were obese and 9(15.8%) of those who ate breakfast 1-3 days/week were obese. Eating breakfast was not associated with obesity (p=0.150). For food for breakfast, those who ate beverages (e.g. koko, tea) were 374(88.4%) and those who ate solid starchy foods (e.g. fufu, banku) were 3(0.7%). None of those who ate solid starchy foods for breakfast were obese. Fifty four of the participants who ate beverages for breakfast were obese. Food for breakfast per week showed a strong statistical significance with obesity (p=0.000). Lunch consumption per week had 219 (51.8%) participants eating lunch daily and 38(9.0%) participants eating lunch 1-3 days/week. Thirty nine (17.8%) of those who ate lunch daily were obese while 6(15.8%) obese participants ate lunch 1-3 days/week. Lunch consumption per week was not associated with obesity (p=0.200). For food for lunch, those who ate grains (e.g. rice, waakye) were 340(80.4%) and those who ate solid starchy foods (e.g. fufu, banku) were 45(10.6%). Fifty two of the participants who ate University of Ghana http://ugspace.ug.edu.gh 42 grains (e.g. rice, waakye) were obese. Six of those who ate solid starchy foods for breakfast were obese. Food for lunch showed association with obesity (p=0.001). For Supper consumption per week, majority 300(70.9%) of the participants ate supper daily. Those who ate supper 1-3 days/week were 14(3.3%). None of those who ate supper 1-3 days/week was obese while majority of the obese participants 43(14.3%) ate supper daily. Supper consumption per week was not associated with obesity (p=0.103). For food for supper, those who ate solid starchy foods (e.g. fufu, banku) were 304(71.9%) and no participants ate beverages (e.g. koko, tea). Fifty of the participants who ate solid starchy foods (e.g. fufu, banku) were obese and the rest of the 15 obese participants ate grains (e.g. rice, waakye). Food for supper showed no association with obesity (p=0.356). University of Ghana http://ugspace.ug.edu.gh 43 Table 4.3 Distribution of obesity by Dietary Habits on Number of meals per day, Breakfast, Lunch and Supper Variable Participants Obese Pearson (ᵡ²) No. (%) No. (%) (p-value) Number of meals per day ᵡ²= 21.499 2 times 29(6.9) 0(0.0) p= 0.001 3 times 343(81.1) 51(14.9) 4 times 51(12.0) 14(27.5) Breakfast consumption per week ᵡ²= 9.447 Daily 220(52.0) 34(15.5) p= 0.150 1-3 days/week 57(13.5) 9(15.8) 4-6 days/week 146(34.5) 22(15.1) None 0(0.0) 0(0.0) Lunch consumption per week ᵡ²= 5.243 Daily 219(51.8) 39(17.8) p= 0.200 1-3 days/week 38(9.0) 6(15.8) 4-6 days/week 166(39.2) 20(12.1) None 0(0.0) 0(0.0) Supper consumption per week ᵡ²= 10.552 Daily 300(70.9) 43(14.3) p= 0.103 1-3 days/week 14(3.3) 0(0.0) 4-6 days/week 109(25.8) 22(20.2) None 0(0.00 0(0.0) Food for breakfast ᵡ²=8.045 Beverages(e.g. koko, tea) 374 (88.4) 54 (14.4) p=0.060 Grains(e.g. rice, waakye) 38 (9.0) 11 (29.0) Solid starchy foods(e.g. fufu, banku) 3 (0.7) 0 (0.0) None 8(1.9) 0(0.0) Food for lunch ᵡ²= 33.823 Beverages(e.g. koko, tea) 19(4.5) 7(36.8) p= 0.001 Grains (e.g. rice, waakye) 340 (80.4) 52 (15.3) Solid starchy foods(e.g. fufu, banku) 45 (10.6) 6 (13.3) None 19(4.5) 0(0.0) Food for supper ᵡ²=6.638 Beverages(e.g. koko, tea) 0 (0.0) 0(0.0) p=0.356 Grains(e.g. rice, waakye) 116 (27.4) 15 (12.9) Solid starchy foods(e.g. fufu, banku) 304 (71.9) 50 (16.5) None 3(0.7) 0(0.0) University of Ghana http://ugspace.ug.edu.gh 44 Table 4.4a shows distribution of obesity by Dietary Habits on snacks, fruits and vegetables consumption. For snacks consumption per week, majority 175(41.4%) of the participants ate snacks 4-6days/week. Those who did not eat snacks were 17(4.0%). Most of the obese participants 27 ate snacks 4-6days/week while 8 of the obese participants did not eat snacks. Snacks consumption per week showed statistical significance with obesity (p=0.001) With fruits consumption per week, majority 233(55.1%) of the participants ate fruits 1-3 days/week. Those who did not eat fruits were 30(7.1%). Most of the obese participants (32) ate fruits 1-3days/week while 15 of the obese participants did not eat fruits. Fruits consumption per week showed some statistical significance with obesity (p=0.045) Vegetable consumption per week, most of the respondents 208(49.2%) ate vegetables 1- 3days/week. Those who did not eat vegetables were 21(5.0%). Most of the obese participants (38) ate vegetables 1-3days/week while 9 of the obese participants did not eat vegetables. Vegetable consumption per week showed a strong statistically significance with obesity (p=0.000). University of Ghana http://ugspace.ug.edu.gh 45 Table 4.4a Distribution of obesity by Dietary Habits on snacks, fruits and vegetables consumption. Variable Participants Obese Pearson (ᵡ²) No. (%) No. (%) (p-value) Snacks consumption per week ᵡ²= 28.150 Daily 57(13.5) 13(22.8) p= 0.001 1-3 days/week 174(41.1) 17(9.8) 4-6 days/week 175(41.4) 27(15.4) None 17(4.0) 8(47.1) Fruits consumption per week ᵡ²= 8.065 Daily 22(5.2) 0(0.0) p= 0.045 1-3 days/week 233(55.1) 32(13.7) 4-6 days/week 138(32.6) 18(13.0) None 30(7.1) 15(50.0) Vegetable consumption per week ᵡ²=71.912 Daily 55 (13.0) 0 (0.0) p=0.000 1-3 days/week 208 (49.2) 38(18.3) 4-6 days/week 139 (32.9) 18 (13.0) None 21 (5.0) 9 (42.9) University of Ghana http://ugspace.ug.edu.gh 46 Table 4.4b shows the distribution of obesity by Dietary Habits on types and frequency of fast food and sweetened beverages consumed. Majority of the respondents 400(94.6%) consumed fast food while 23(5.4%) did not. Sixty three of the obese participants consumed fast food and 2 did not consume fast food. Fast food consumption showed a strong association with obesity (p=0.000). The types of fast food listed were fried rice, fried chicken, pizza, potato chips and burgers. 356(84.2%) of the participants consumed fried rice. Out of the 65 obese participants, 58 consumed fried rice. Fried Chicken was consumed by 380(89.8%) participants and 60(15.8%) obese participants. Ninety eight (23.2%) participants consumed potato chips while 20(20.4%) obese participants consumed potato chips. Pizza was also consumed by 98(23.2%) participants with 33(33.7%) of the obese participants consuming pizza. The least consumed fast food was the burger, only 58(13.7%) participants consumed burgers. Nine (15.5%) obese participants consumed burgers. With the frequency of fast food consumption, majority of the participants 233(55.1%) ate fasts food 1-3days/week, 108(32.6%) ate fast food 4-6days/week and 26(5.2%) ate fast foods daily. Majority of the obese participants 22 ate fast food 1-3days/week. 18 of the obese participants ate fast food 4-6 days/week while 15 obese participants ate fast food daily. Sugar-sweetened beverages consumption had most of the participants 414(97.9%). Those who did not consume sugar-sweetened beverages were 9(2.1%). All the 65 obese participants were those who consumed sugar-sweetened beverages. Sugar-sweetened beverages consumption was statistically significant (p=0.000). University of Ghana http://ugspace.ug.edu.gh 47 The types of sweetened beverages listed were coke, fanta, sprite and other soft drinks or “minerals”. Majority of the participants 336(79.42%) patronized coke. Those who patronized fanta were 339(80.1%). Those who patronized sprite and other soft drinks were 306(72.3%) and 317(74.9%) respectively. Out of the 65 obese participants, 56 consumed both coke, fanta and other soft drinks while sprite was patronized by 50 obese participants. With the frequency of sweetened beverages consumption, majority of the participants 227(53.7%) took sweetened beverages 1-3days/week, 94(22.2%) took sweetened beverages 4- 6days/week and 63(14.9%) took sweetened beverages daily. Majority of the obese participants 22 took sweetened beverages 1-3days/week. Twenty of the obese participants took sweetened beverages 4-6 days/week while 17 obese participants took sweetened beverages daily. University of Ghana http://ugspace.ug.edu.gh 48 Table 4.4b Distribution of obesity by Dietary Habits on types and frequency of fast food and sweetened beverages consumed. Variable Participants Obese Pearson (ᵡ²) No. (%) No. (%) (p-value) Fast food consumption ᵡ²=49.423 Yes 400(94.6) 63(15.8) p=0.000 No 23(5.4) 2(8.7) Type of fast food Fried Rice 356(84.2) 58(16.3) Fried Chicken 380(89.8) 60(15.8) Potato chips 98(23.2) 20(20.4) Pizza 98(23.2) 33(33.7) Burger 58(13.7) 9(15.5) Fast food consumption Daily 26(5.2) 15(57.7) 1-3 days/week 233(55.1) 22(9.4) 4-6 days/week 108(32.6) 18(16.7) Other 33(7.8) 8(4.2) None 23(5.4) 2(8.7) Sugar-sweetened beverages consumption ᵡ²=46.498 Yes 414(97.9) 65(15.7) p=0.000 No 9 (2.1) 0 (0.0) Type of sweetened beverage Coke 336(79.4) 56(16.7) Fanta 339(80.1) 56(16.5) Sprite 306(72.3) 50(16.3) Others 317(74.9) 56(17.7) Sweetened beverages consumption Daily 63 (14.9) 17(27.0) 1-3 days/week 227(53.7) 22(9.7) 4-6 days/week 94(22.2) 20(20.3) Other 32(7.6) 6 (18.8) None 9(2.1) 0(0.0) University of Ghana http://ugspace.ug.edu.gh 49 Table 4.5 shows the distribution of obesity by familial or genetic factors and physical activity. Those who had a fat father were 54(14.0%) and those without a fat father were 364(86.0%). Thirty four of the obese respondents had a fat father. Having a fat father was strongly associated with obesity (p=0.000) Those who had a fat mother were 142(33.6%) and those without a fat mother were 281(66.4%). Forty four of the obese respondents had a fat mother. Having a fat mother was statistically significant with obesity (p=0.000) Those who had a fat relative other than a father or mother were 184(43.5%) and those without a fat relative were 239(56.5%). Forty eight of the obese respondents had a fat relative. Having a fat relative was strongly associated with obesity (p=0.000). Those who had a fat sibling were 69(16.3%) and those without a fat sibling were 354(83.7%). Thirty four of the obese respondents had a fat sibling. Having a fat sibling was statistically significant with obesity (p=0.000) Those who were fat before age 10years were 66(15.6%) and those who were not fat before 10years were 357(84.4%). Thirty two of the obese respondents were fat before age 10years. Being fat before age 10years was statistically significant with obesity (p=0.000). Majority of the participants 316(74.7%) engaged in conscious Physical Activity. Most of the obese participants (37) did not engage in conscious physical activity. Engaging in conscious physical activity was statistically significant (p=0.000) For duration of Physical Activity, most of the respondents 166(39.2%) had a duration of Physical Activity of <30 minutes. 107(25.3%) of the respondents did not engage in Physical Activity. University of Ghana http://ugspace.ug.edu.gh 50 Majority of the obese respondents (37) did not engage in physical activity. No participant with duration of Physical Activity between 61-120 minutes was obese. Duration of Physical Activity was strongly associated with obesity (p=0.000). All the participants 423(100.0%) had Physical Education as part of school curriculum. All the 65 obese participants had Physical Education as part of school curriculum. For duration of watching TV, majority of the respondents 235(55.6%) had a duration of watching TV < 1 hour. Sixty nine (16.3%) of the participants had a duration of watching TV of > 4 hours. Most of the obese respondents (25) had a duration of watching TV of 1-4 hours while 19 of the obese participants had a duration of watching TV of > 4hours. Duration of watching TV was strongly associated with obesity (p=0.000). University of Ghana http://ugspace.ug.edu.gh 51 Table 4.5: Distribution of obesity by Familial or Genetic factors and Physical Activity Variable Participants Obese Pearson (ᵡ²) No. (%) No. (%) (p-value) Fat Father ᵡ²= 103.935 Yes 59(14.0) 34(57.6) p= 0.000 No 364(86.0) 31(8.5) Fat Mother ᵡ²= 80.033 Yes 142(33.6) 44(31.0) p= 0.000 No 281(66.4) 21(7.5) Fat Relative ᵡ²= 41.918 Yes 184(43.5) 48(26.1) p= 0.000 No 239(56.5) 17(7.1) Fat Sibling ᵡ²= 48.508 Yes 69(16.3) 34(15.5) p= 0.000 No 354(83.7) 31(15.8) Fat before age 10years ᵡ²= 66.388 Yes 66(15.6) 32(48.5) p= 0.000 No 357(84.4) 33(9.2) Conscious Physical Activity ᵡ²= 47.594 Yes 316(74.7) 28(8.9) p= 0.000 No 107(25.3) 37(34.6) Duration of Physical Activity ᵡ²= 90.899 <30 minutes 166(39.2) 14(8.4) p= 0.000 30-60 minutes 91(21.5) 11(12.1) 61-120 minutes 56(13.2) 0(0.0) >120 minutes 3(0.7) 3(100.0) None 107(25.3) 37(34.6) Physical Education as part of school curriculum Yes 423(100.0) 65(15.4) No 0(0.0) 0(0.0) Duration of watching TV ᵡ²= 62.150 < 1 hour 235(55.6) 21(8.9) p= 0.000 1-4 hours 119(28.1) 25(21.0) > 4 hours 69(16.3) 19(27.5) University of Ghana http://ugspace.ug.edu.gh 52 4.2. Risk factor for obesity The variables that showed statistical significance with obesity using the chi square test (p-value <0.05) in the univariate model were sex, age, ethnicity, residential area, socioeconomic status, number of meals per day, fruits consumption per week, snacks consumption per week, vegetable consumption frequency, food for lunch, fast food consumption, sugar-sweetened beverages consumption, fat father, fat mother, fat sibling, fat before age 10years, conscious physical activity, duration of physical activity and duration of watching TV. A bivariate logistic regression was then performed to see the strength of association between each of these risk factors and obesity. Table 4.6 shows a bivariate logistic regression socio demographic factors and dietary habits for obesity. For sex, females were 3.92 times likely to become obese compared to males (OR= 3.92, p=0.000). Socioeconomic status was still significantly associated with being obese, those from a high SES background were 1.56 times more likely to become obese compared to those from low SES background (OR=1.56, p=0.036). Those whose ate other than 3times a day are 2.73 times likely to become obese compared to those who ate 3 times in a day (OR= 2.73, p=0.001). For snacks consumption per week, those who consumed snacks >3days/week had a 63% reduced risk of becoming obese compared to those who consumed snacks <3days/week (OR=0.37, p=0.013). Eating grains as food for lunch reduces the risk of being obese by 69% compared to those who ate beverages as food for lunch (OR=0.31, 95% CI 0.12-0.82) University of Ghana http://ugspace.ug.edu.gh 53 Table 4.6: Bivariate logistic regression on socio demographic factors and dietary habits for obesity Variable Obese Crude OR( 95% CI) p-value Sex Male 1 p= 0.000 Female 3.92 (1.94-7.94) Age 0.90(0.71-1.14) p= 0.368 Ethnicity Ga 1 p= 0.710 Others 0.90(0.47-1.68) Residential Area Tema 1 p= 0.054 Others 0.37(0.15-1.93) Socioeconomic Status( SES) Low SES 1 p= 0.048 High SES 1.56(1.02-2.36) Number of meals per day 3 times 1 p= 0.001 Others 2.73(1.50-4.96) Fruits consumption per week >3 days/week 1 p= 0.089 <3 days/week 1.42(0.78-1.84) Snacks consumption per week >3 days/week 1 p= 0.013 <3 days/week 0.37(0.17-0.81) Vegetable consumption frequency >3 days/week 1 p=0.146 <3 days/week 1.29(0.90 -1.65) Food for lunch Beverages(e.g. koko, tea) 1 p= 0.019 Grains (e.g. rice, waakye) 0.31 (0.12-)0.82 Fast food consumption Yes 1 p=0.817 No 1.02(0.88-1.17) Sugar-sweetened beverages consumption Yes 1 p=0.719 No 1.04(0.85-1.28) University of Ghana http://ugspace.ug.edu.gh 54 Table 4.7 shows bivariate logistic regression on familial or genetic factors and physical activity for obesity. Not having a fat father reduces the risk of being obese by 93% compared to having a fat father (OR=0.07, p=0.000). For those without a fat mother, it reduces the risk of being obese by 82% compared to having a fat mother (OR=0.18, p=0.000). Not having a fat relative reduces the risk of being obese by 78% compared to having a fat relative (OR=0.22, p=0.000). Not being fat before age 10years reduces the risk of being obese by 89% compared to being fat before age 10years (OR=0.11, 95% CI 0.06-0.20). Those who did not engage in conscious physical activity were 5.44 times more likely to be obese compared to those who engage in conscious physical activity (OR= 5.44, p=0.000). Those who engage in physical activity for < 30 minutes were 5.74 times more likely to be obese compared to those who engage in physical activity for > 30 minutes (OR= 5.74, p=0.000). Those with duration of watching TV of > 4hours were 3.87 times more likely to be obese compared to those with duration of watching TV < 4hours (OR= 3.87, p=0.000) University of Ghana http://ugspace.ug.edu.gh 55 Table 4.7: Bivariate logistic regression on Familial or genetic factors and Physical Activity for obesity Variable Obese Crude OR( 95% CI) p-value Fat Father Yes 1 p= 0.000 No 0.07(0.04-0.13) Fat Mother Yes 1 p= 0.000 No 0.18(0.10-0.32) Fat Relative Yes 1 p= 0.000 No 0.22(0.12-0.39) Fat Sibling Yes 1 p= 0.200 No 1.64(0.77-3.49) Fat before age 10years Yes 1 p= 0.000 No 0.11(0.06-0.20) Conscious Physical Activity Yes 1 p= 0.000 No 5.44(3.12-9.48) Duration of Physical Activity <30 minutes 1 >30 minutes 5.74(2.92-11.29) p= 0.000 Duration of watching TV < 4 hours 1 > 4 hours 3.87(1.94-7.74) p= 0.000 University of Ghana http://ugspace.ug.edu.gh 56 4.2.1 Significant risk factors for obesity For the bivariate logistic regression, the variables sex, socioeconomic status, snacks consumption per week, food for lunch, having a fat father, having a fat mother, having a fat relative, fat before age 10years, conscious physical activity, duration of physical activity and duration of watching TV showed statistical significance with obesity. A multivariate logistic regression was carried out for the statistically significant in the bivariate logistic regression model. The variables sex, socioeconomic status, number of meals per day, snacks consumption per week, food for lunch, fat father, fat mother, fat relative, fat before age 10years and duration of watching TV were statistically significant with obesity in the multivariate logistic regression model. Table 4.8 shows the significant of risk factors for obesity. For sex, females were 21.8 times likely to become obese compared to males after adjusting for other variables (AOR= 21.8, p=0.000). Socioeconomic status was still significantly associated with being obese after adjusting for other variables those from a high SES background were 2.56 times more likely to become obese compared to those from low SES background (AOR=2.56, p=0.036). For snacks consumption per week, those who consumed snacks <3 days/week were 1.88 times more likely to be obese compared to those who consumed snacks >3 days/week after controlling for other variables (AOR=1.88, p=0.046). University of Ghana http://ugspace.ug.edu.gh 57 After adjusting for other variables, not having a fat father reduces the risk of being obese by 94% compared to having a fat father (AOR=0.06, p=0.000). For those without a fat mother, it reduces the risk of being obese by 79% compared to having a fat mother after controlling for other variables (AOR=0.21, p=0.000). After controlling for other variables, not having a fat relative reduces the risk of being obese by 71% compared to having a fat relative (AOR=0.29, 95%CI 0.12-0.71). Not being fat before age 10years reduces the risk of being obese by 83% compared to being fat before age 10years after adjusting for other variables (AOR=0.17, 95% CI 0.06-0.46). Those with duration of watching TV of > 4hours were 2.31 times more likely to be obese compared to those with duration of watching TV < 4hours after controlling for other parameters (AOR= 2.31, p=0.006) University of Ghana http://ugspace.ug.edu.gh 58 Table 4.8: Significant risk factors for obesity Variable Obese Adjusted OR(95% CI) p-value Sex Male 1 p= 0.000 Female 21.8 (5.57-85.60) Socioeconomic Status( SES) Low SES 1 p= 0.036 High SES 2.56(1.10-5.98) Number of meals per day 3 times 1 p= 0.069 Others 2.60(0.93-7.25) Snacks consumption per week >3 days/week 1 p= 0.016 <3 days/week 2.7(0.17-0.81) Food for lunch Grains (e.g. rice, waakye) 1 p= 0.040 Solid starchy foods(e.g. fufu, banku) 0.30 (0.89-0.94) Fat Father Yes 1 p= 0.000 No 0.04(0.01-0.11) Fat Mother Yes 1 p= 0.000 No 0.21(0.09-0.50) Fat Relative Yes 1 p= 0.007 No 0.29(0.12-0.71) Fat before age 10years Yes 1 p= 0.001 No 0.17(0.06-0.46) Conscious Physical Activity Yes 1 p= 0.135 No 0.02(0.00-3.53) Duration of Physical Activity >30 minutes 1 p= 0.072 <30 minutes 1.90(0.94-3.84) Duration of watching TV <4hours 1 p= 0.006 >4hours 2.31(1.27-4.20) University of Ghana http://ugspace.ug.edu.gh 59 CHAPTER FIVE 5 DISCUSSION 5.1 Prevalence of Obesity The overall prevalence of obesity was 15.4%. The prevalence of overweight was 16.5%. The combined prevalence of obesity and overweight was 31.9%. The prevalence of obesity in Methodist Day Senior High School and Datus Senior High School was 15.0% and 19.1% respectively. The overall prevalence of obesity amongst males and females was 6.3% and 20.8% respectively. This is similar to a study which showed that females had a higher prevalence of obesity of 14.9% compared to 14.2% for males (El-Bayoumy et al., 2009). This study is also similar to a study conducted in 1989 to describe the prevalence of overweight and obesity among Brazilian adolescents aged 10 to 19 years, in which the prevalence of obesity and overweight was of 7.7%, with 10.6% in the females and 4.8% in the males (Neutzling et al., 2003). However, our study is in contrast to a cross-sectional study conducted among children 6 to 17 years to determine prevalence of obesity in children and adolescents in which the prevalence of obesity was 10.3% and 9.1% in males and females respectively, using the NHANES I definition (Savva et al., 2002). Another study with a higher prevalence of obesity in males compared to females was a study of weight-for-height reference and the prevalence of obesity for school children and adolescents in Taiwan and Fuchien areas, the prevalence of childhood and adolescent obesity was 18.5% in boys and 15.0% in girls(Huang et al., 2003) University of Ghana http://ugspace.ug.edu.gh 60 5.2 Risk factors for obesity 5.2.1 Dietary habits In this study when it comes to snacks consumption per week, those who consumed snacks 1- 3days/week were 1.88 times more likely to be obese compared to those who consumed snacks daily after controlling for other variables (AOR=1.88, p=0.046). For this study 96.9% of the obese participants consumed fast food such as fried rice, fried chicken, potato chips and pizza. Fast food consumption showed a strong association with obesity in the univariate model (p=0.000). Sugar-sweetened beverages consumption was seen in 97.9% of the participants. All the obese participants (100.0%) consumed sugar-sweetened beverages. Sugar sweetened beverages consumption was statistically significant with obesity in the univariate model (p=0.000). This agrees with the WHO report in 2010, which indicates that eating western food has become widely accepted in developing countries with adolescents forming part of this group of patrons. And as long as parents continue to leave the decision concerning what to eat to their wards, their wards are bound to patronize foods that may be unbalanced and unhealthy (WHO, 2010). However, fast food consumption and sugar-sweetened beverages consumption did not show a significant association with obesity in the bivariate model (p=0.817 and p=0.719 respectively) In this study none of those who consumed fruits and vegetables daily were obese. This agrees with the study conducted in Saudi Arabia which showed that low servings of fruits and vegetables were all predictors of obesity and overweight among the school children (Amin et al., 2008). University of Ghana http://ugspace.ug.edu.gh 61 In this study, 52.3% obese participants ate breakfast daily. Eating breakfast was not associated with obesity (p=0.150) and this is in contrast to a study conducted in Norway that showed that eating breakfast was positively associated with not being overweight and obese (Grøholt et al., 2008). This study therefore coincides with another study that showed that factors that increase the risk of obesity include high intake of energy-dense and micronutrient – poor foods, increased number of fast-food outlets and marketing of energy-dense foods, high intake of sugars-sweetened carbonated drinks and fruit juices as well as low consumption of fruits and vegetables (Nishida et al., 2004). Another study that agrees with this study is a randomized controlled trial of primary school based intervention to reduce risk factors for obesity in Leeds, which showed that consumption of fruits was lower in obese children in the intervention group compared to those in the control group and a three day recall diary showed a low consumption of highly sugary foods among overweight children in the control group than the intervention group (Sahota et al., 2001) 5.2.2 Familial or Genetic factors Studies have shown a strong predisposition towards the fact that obesity can be inherited. In this study, after adjusting for other variables, not having a fat father reduces the risk of being obese by 94% compared to having a fat father (AOR=0.06, p=0.000). For those without a fat mother, it reduces the risk of being obese by 79% compared to having a fat mother after controlling for other variables (AOR=0.21, p=0.000). The risk of childhood and adolescent obesity has been found to be associated with high parental body mass index in another study in which the results University of Ghana http://ugspace.ug.edu.gh 62 showed that the risk of childhood obesity was associated with having one or both parents obese, the risk increased when both parents were found to be obese (da Costa Ribeiro et al., 2003). Similarly, a case control study conducted to determine the risk factors associated with overweight and obesity among adolescents enrolled in private high schools in the city of Pelotas, southern Brazil. Their results showed adolescents whose parents’ BMI were greater than 30 were 2 times more likely to be overweight or obese than those whose parents had BMI less than 30 (Neutzling et al., 2003). From the results of a study conducted on obesity in children and adolescents in Cyprus, the most significant associated factor for obesity was parental obesity status (Savva et al., 2002). Another study carried out in Italy showed that parental obesity was the main risk factor for obesity in children (Maffeis et al., 2006). 5.2.3 Metabolic or Physiological factors The males had an average height of 1.65± 0.08metres and females had an average height 1.59± 0.06metres. The males were taller than the females. The average weight was 61.24 ±10.6kilograms in males and 62.20 ±12.5 kilograms in females. The mean BMI for females and males was 24.26±4.90 kg/m2 and 22.39 ±3.56 kg/m2 respectively. This indicates that the females had a higher BMI and weight than the males. This corresponds to a study done in Kuwait in which females had higher BMI than males (El-Bayoumy et al., 2009). During adolescence, and especially during puberty, both girls and boys undergo fat storage, and because the spurt of height in boys is higher than in girls, the BMI of girls appears to be higher than boys (Poskit, 1988). University of Ghana http://ugspace.ug.edu.gh 63 In this study the overall prevalence of obesity amongst females and males was 20.8% and 6.3% respectively. The most significant risk factor after adjusting for other variables was sex (AOR=21.8, p=0.000). This is similar to a study which showed females had a higher prevalence of obesity of 14.9% compared to 14.2% for males (El-Bayoumy et al., 2009). In this study not being fat by the age of 10 years is protective, those who were not fat before age 10 years had a reduced risk of being obese by 89% compared to those who were fat before age 10 years after controlling for the effects of other variables (AOR=0.11, 95% CI 0.06-0.20). This agrees with a case control study conducted to determine the risk factors associated with overweight and obesity among adolescents enrolled in private high schools in the city of Pelotas, southern Brazil which demonstrated that childhood overweight showed a direct relationship to overweight and obesity during adolescence. Adolescents who reported being overweight before they were 10 years old were 2.5 times more likely to be overweight or obese during their adolescence (Neutzling et al., 2003). 5.2.4 Physical inactivity In this study 58.6% of the obese participants did not engage in conscious physical activity, engaging in conscious physical activity was statistically significant with obesity (p=0.000). This is similar to study from 2000 to 2004 in Norway, in which physical inactivity was positively associated with being overweight and obese. There were more overweight and obese adolescents who were physically inactive compared to those who were physically active (Grøholt et al., 2008). For this study those with duration of watching TV of > 4hours were 2.31 times more likely to be obese compared to those with duration of watching TV < 4hours after controlling for other University of Ghana http://ugspace.ug.edu.gh 64 variables (AOR= 2.31, p=0.006) and this agrees with a study that showed there is an independent association between overweight and obesity and TV/PC use. Adolescents watching television/video 3 hours or more per day were about 60% more likely to be obese compared with adolescents watching television/video 0 to 2 hours per day (Savva et al., 2002). This study agrees with a randomised controlled trial of 634 primary school children aged 7 to 11 years in Leeds to reduce risk factors for obesity, which showed that sedentary behaviour was higher in overweight children in the intervention group than the control group (Sahota et al., 2001). This study coincides with a study by Pietiläinen et al., (2008) in which physical activity and self-perceived physical fitness assessed in adolescents aged 16 to 18 years of age were used to predict the development of obesity and the results showed that physical inactivity in adolescence strongly predicted the risk for obesity. This study is similar to a cross sectional study conducted to compare the physical activity pattern of children, in which their results showed that obese children exhibited significantly lower daily accumulations of total counts, significantly lower levels of physical activity self-efficacy and were also involved in significantly fewer community organizations promoting physical activity (Trost et al., 2003). This study also agrees with a study conducted in which the amount of time spent in front of the television during adolescence has been found to be significantly associated with adiposity even after correction for obesity history (S. Kautiainen et al., 2005). University of Ghana http://ugspace.ug.edu.gh 65 5.2.5 Socioeconomic status The relationship between social class and BMI varies globally. In the developed world pediatric obesity is generally more common in children and adolescents from families of lower socioeconomic status whereas in the developing world, higher socioeconomic status has usually been associated with higher risk of pediatric obesity (WHO, 2010). In this study majority of the obese participants 35(55.6%) were of high socioeconomic status, 19(30.1%) were of middle socioeconomic status and 9(14.3%) of the obese participants were of low socioeconomic status. The socioeconomic status showed significance with obesity (p=0.032). This agrees with findings from a study carried out in Kuwait of which 89.2% of the obese children belonged to families with high socioeconomic status (earning ≥1000 Kuwaiti Dinars), while 6.7% of them belonged to middle social class families (families earning 500 to <1000 Kuwaiti Dinars), and 4.1% of them belonged to families with low socioeconomic status (earning <500 Kuwaiti Dinars) (El-Bayoumy et al., 2009). For this study the socioeconomic status was still significantly associated with being obese after adjusting for other variables (AOR=2.56, p=0.036) and it agrees with a study in which adolescents living in families with high socioeconomic status were 1.3–1.5 times more likely of becoming overweight and obese compared to adolescents living in families with low socioeconomic status (Grøholt et al., 2008). This study is also similar to study that has shown that obesity is more prevalent among high socioeconomic class in developing countries (Burns, 2004). University of Ghana http://ugspace.ug.edu.gh 66 CHAPTER SIX CONCLUSIONS AND RECOMMENDATION 6.1 CONCLUSION The overall prevalence of obesity was 15.4%. There were 70 (16.5%) overweight students and these must be targeted to prevent them from becoming overweight. The combined prevalence of obesity and overweight was 31.9%. This combined prevalence of obesity and overweight is very high. The incidental finding of underweight of 0.7% means that under nutrition is an issue in adolescents but to a lesser extent. Under nutrition must therefore not be overlooked. The study showed that there were associations between obesity and sex, socioeconomic status, snacks consumption per week, food for lunch, having a fat father, having a fat mother, having a fat relative, being fat before age 10 years and duration of watching TV. Sex was the most significantly associated with obesity, with females being 21.8 times more likely to be obese compared to males. Watching TV > 4hours increases the risk of being obese by 131%. Not having a fat parent or relative, consuming snacks daily and eating of grains for lunch are protective. This study provides useful baseline information for future studies on obesity in Senior High Schools. University of Ghana http://ugspace.ug.edu.gh 67 6.2 RECOMMENDATION Even though schools are the main institutions able to reach a large number of children and adolescents, the fight against obesity should be multidisciplinary. It should involve the schools, parents or guardians, policy makers, health professionals or institutions and everybody. 6.2.1 Schools  Educational programs aimed at adolescents should encourage physical activity and adequate dietary habits.  Basic nutrition should be introduced into the curricula of schools to educate students on healthy eating habits and its benefits.  School authorities need to regulate foods being sold by vendors in their premises. School authorities should also ensure that food served to boarding students are of good nutrients in addition to adequate vegetables and fruits.  Adolescents should be encouraged to engage in physical activity outside schools. They must also be made to participate fully in the Physical Education activities in school.  School health programs should be given all the needed support and must be mandated to provide school health services to ensure our future workforce are healthy. 6.2.2 Parents or Guardians  Parents or guardians should limit the duration of watching TV by their children at home.  Parents or guardians should also provide their children with adequate snacks and fruits for school or encourage them to buy healthy snacks or meals in school. University of Ghana http://ugspace.ug.edu.gh 68 6.2.3 Policy Makers and Health Institutions  Health policies should be directed at ensuring that the adolescent are healthy by developing effective public health education programs to increase the awareness and causes of obesity as well as its prevention.  More recreational facilities should be built to support physical activity. These facilities should be made affordable if not free to use.  There is the need for systematic surveillance and screening of growth indices to identify overweight and obese adolescents for prompt action. 6.2.4 Research  Further research into determinants of obesity in adolescents should be undertaken. The research can include the measuring of parental BMI. University of Ghana http://ugspace.ug.edu.gh 69 REFERENCES Abubakari, A. R., & Bhopal, R. S. (2008). Systematic review on the prevalence of diabetes, overweight/obesity and physical inactivity in Ghanaians and Nigerians. Public Health, 122(2), 173–182. Ahmad, Q. I., Ahmad, C. B., & Ahmad, S. M. (2010). Childhood Obesity. Indian Journal of Endocrinology and Metabolism, 14(1), 19–25. Al-Sendi, A. M., Shetty, P., & Musaiger, A. O. (2003). Prevalence of overweight and obesity among Bahraini adolescents: a comparison between three different sets of criteria. European Journal of Clinical Nutrition, 57(3), 471–474. American Acamedy of Pediatrics. (2013). Childhood Obesity & Technology – What’s the connection? Retrieved July 22, 2014, from https://www.floridahospital.com/blog/childhood-obesity-technology-whats-connection Amin, T. T., Al-Sultan, A. I., & Ali, A. (2008). Overweight and Obesity and their Association with Dietary Habits, and Sociodemographic Characteristics Among Male Primary School Children in Al-Hassa, Kingdom of Saudi Arabia. Indian Journal of Community Medicine : Official Publication of Indian Association of Preventive & Social Medicine, 33(3), 172–181. doi:10.4103/0970-0218.42058 Amoah, A. G. (2003). Sociodemographic variations in obesity among Ghanaian adults. Public Health Nutrition, 6(08), 751–757. Arojo, O. O., & Osungbade, K. O. (n.d.). Trends of Obesity Epidemic and its Socio-cultural Dimensions in Africa: Implications for Health Systems and Environmental Interventions. Retrieved from https://www.iconceptpress.com/download/paper/12060803212512.pdf Bellizzi, M. C., & Dietz, W. H. (1999). Workshop on childhood obesity: summary of the discussion. The American Journal of Clinical Nutrition, 70(1), 173s–175s. Biritwum, R. B., Gyapong, J., & Mensah, G. (2005). The epidemiology of obesity in Ghana. Ghana Medical Journal, 39(3), 82. Bogers, R. P., Bemelmans, W. J., Hoogenveen, R. T., Boshuizen, H. C., Woodward, M., Knekt, P., … others. (2007). Association of overweight with increased risk of coronary heart disease partly independent of blood pressure and cholesterol levels: a meta-analysis of 21 cohort studies including more than 300 000 persons. Archives of Internal Medicine, 167(16), 1720–1728. Boutayeb, A. (2006). The double burden of communicable and non-communicable diseases in developing countries. Transactions of the Royal Society of Tropical Medicine and Hygiene, 100(3), 191–199. University of Ghana http://ugspace.ug.edu.gh 70 Burns, C. (2004). A review of the literature describing the link between poverty, food insecurity and obesity with specific reference to Australia. Melbourne: Victorian Health Promotion Foundation. Retrieved from http://secure.secondbite.org/sites/default/files/A_review_of_the_literature_describing_the _link_between_poverty_food_insecurity_and_obesity_w.pdf Caroli, M., Argentieri, L., Cardone, M., & Masi, A. (2004). Role of television in childhood obesity prevention. International Journal of Obesity, 28(S3), S104–S108. doi:10.1038/sj.ijo.0802802 Chatterjee, A., & DeVol, R. C. (2012). Waistlines of the World. The Effect of Information and Communication Technology on Obesity. Milken Institute. Santa Monica. Retrieved from https://www.milken-inst.org/pdf/Waistlines-of-the-World.pdf Christakis, N. A., & Fowler, J. H. (2007). The spread of obesity in a large social network over 32 years. New England Journal of Medicine, 357(4), 370–379. Cochran, W. G. (1977). Sample techniques. New York: John Wills & Sons. Cole, T. J., Bellizzi, M. C., Flegal, K. M., & Dietz, W. H. (2000). Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ (Clinical Research Ed.), 320(7244), 1240–1243. Da Costa Ribeiro, I., Taddei, J. A., Colugnatti, F., & others. (2003). Obesity among children attending elementary public schools in Sao Paulo, Brazil: a case–control study. Public Health Nutr, 6(7), 659–663. De Onis, M., & Blössner, M. (2000). Prevalence and trends of overweight among preschool children in developing countries. The American Journal of Clinical Nutrition, 72(4), 1032–1039. De Onis, M., Blössner, M., & Borghi, E. (2010). Global prevalence and trends of overweight and obesity among preschool children. The American Journal of Clinical Nutrition, 92(5), 1257–1264. Drewnowski, A., & Almiron-Roig, E. (2010). Human Perceptions and Preferences for Fat-Rich Foods. In J.-P. Montmayeur & J. le Coutre (Eds.), Fat Detection: Taste, Texture, and Post Ingestive Effects. Boca Raton (FL): CRC Press. Retrieved from http://www.ncbi.nlm.nih.gov/books/NBK53528/ El-Bayoumy, I., Shady, I., & Lotfy, H. (2009). Prevalence of obesity among adolescents (10 to 14 years) in Kuwait. Asia-Pacific Journal of Public Health. Retrieved from http://aph.sagepub.com/content/early/2009/02/03/1010539509331786.short Elena, G., Bruna, C., Benedetta, M., Stefania, D. C., & Giuseppe, C. (2012). Prader-Willi Syndrome: Clinical Aspects. Journal of Obesity, 2012, 1–13. doi:10.1155/2012/473941 University of Ghana http://ugspace.ug.edu.gh 71 Ezzati, M., Vander Hoorn, S., Lawes, C. M., Leach, R., James, W. P. T., Lopez, A. D., … Murray, C. J. (2005). Rethinking the “diseases of affluence” paradigm: global patterns of nutritional risks in relation to economic development. PLoS Medicine, 2(5), e133. Georgiadis, G., & Nassis, G. P. (2007). Prevalence of overweight and obesity in a national representative sample of Greek children and adolescents. European Journal of Clinical Nutrition, 61(9), 1072–1074. doi:10.1038/sj.ejcn.1602619 Ghana Demographic and Health Survey Report. (2008). Ghana Demographic and Health Survey Report. Ghana Health Service. Ghana Health Service. (2007). Ghana Health Service Annual Report. Grøholt, E.-K., Stigum, H., & Nordhagen, R. (2008). Overweight and obesity among adolescents in Norway: cultural and socio-economic differences. Journal of Public Health, 30(3), 258–265. Hu, F. B., & Hu, F. B. (2008). Measurements of adiposity and body composition. Obesity Epidemiology, 53–83. Huang, Y.-C., Wu, J.-Y., & Yang, M. J. (2003). Weight-for-height reference and the prevalence of obesity for school children and adolescents in Taiwan and Fuchien Areas. J Chin Med Assoc, 66(10), 599–606. Joosen, A. M., & Westerterp, K. R. (2006). Energy expenditure during overfeeding. Nutr Metab (Lond), 3, 25. Kalra, G., De Sousa, A., Sonavane, S., & Shah, N. (2012). Psychological issues in pediatric obesity. Industrial Psychiatry Journal, 21(1), 11. Kautiainen, S., Koivusilta, L., Lintonen, T., Virtanen, S. M., & Rimpelä, A. (2005). Use of information and communication technology and prevalence of overweight and obesity among adolescents. International Journal of Obesity, 29(8), 925–933. doi:10.1038/sj.ijo.0802994 Kautiainen, S., Koivusilta, L., Lintonen, T., Virtanen, S. M., & Rimpelä, A. (2005). Use of information and communication technology and prevalence of overweight and obesity among adolescents. International Journal of Obesity, 29(8), 925–933. Kautiainen, S., Rimpelä, A., Vikat, A., & Virtanen, S. M. (2002). Secular trends in overweight and obesity among Finnish adolescents in 1977-1999. International Journal of Obesity and Related Metabolic Disorders: Journal of the International Association for the Study of Obesity, 26(4), 544–552. Kuczmarski, R. J., Ogden, C. L., Grummer-Strawn, L. M., Flegal, K. M., Guo, S. S., Wei, R., … Johnson, C. L. (2000). CDC growth charts: United States. Advance Data, (314), 1–27. University of Ghana http://ugspace.ug.edu.gh 72 Li, M., Dibley, M. J., & Yan, H. (2011). School environment factors were associated with BMI among adolescents in Xi’an City, China. BMC Public Health, 11(1), 1–6. doi:10.1186/1471-2458-11-792 Lobstein, T., Baur, L., & Uauy, R. (2004). Obesity in children and young people: a crisis in public health. Obesity Reviews, 5(s1), 4–85. Maffeis, C., Consolaro, A., Cavarzere, P., Chini, L., Banzato, C., Grezzani, A., … Tatò, L. (2006). Prevalence of Overweight and Obesity in 2- to 6-year-old Italian Children. Obesity, 14(5), 765–769. doi:10.1038/oby.2006.87 Martorell, R., Stein, A. D., & Schroeder, D. G. (2001). Early nutrition and later adiposity. The Journal of Nutrition, 131(3), 874S–880S. McLaren, L. (2007). Socioeconomic status and obesity. Epidemiologic Reviews, 29(1), 29–48. Mendoza, J. A., Drewnowski, A., & Christakis, D. A. (2007). Dietary Energy Density Is Associated With Obesity and the Metabolic Syndrome in U.S. Adults. Diabetes Care, 30(4), 974–979. doi:10.2337/dc06-2188 Meredith, C. N., & Dwyer, J. T. (1991). Nutrition and Exercise: Effects on Adolescent Health. Annual Review of Public Health, 12(1), 309–333. doi:10.1146/annurev.pu.12.050191.001521 Ministry of Health. (2007). MOH Annual Report. Moayeri, H., Bidad, K., Aghamohammadi, A., Rabbani, A., Anari, S., Nazemi, L., … Hatmi, Z. N. (2006). Overweight and obesity and their associated factors in adolescents in Tehran, Iran, 2004–2005. European Journal of Pediatrics, 165(7), 489–493. Mozaffarian, D., Hao, T., Rimm, E. B., Willett, W. C., & Hu, F. B. (2011). Changes in Diet and Lifestyle and Long-Term Weight Gain in Women and Men. The New England Journal of Medicine, 364(25), 2392–2404. doi:10.1056/NEJMoa1014296 Neovius, M., Linne, Y., & Rossner, S. (2004). BMI, waist-circumference and waist-hip-ratio as diagnostic tests for fatness in adolescents. International Journal of Obesity, 29(2), 163– 169. Neutzling, M. B., Taddei, J. A. A., & Gigante, D. P. (2003). Risk factors of obesity among Brazilian adolescents: a case–control study. Public Health Nutrition, 6(08), 743–749. Nishida, C., Uauy, R., Kumanyika, S., & Shetty, P. (2004). The joint WHO/FAO expert consultation on diet, nutrition and the prevention of chronic diseases: process, product and policy implications. Public Health Nutrition, 7(1a), 245–250. University of Ghana http://ugspace.ug.edu.gh 73 Organization, W. H., & others. (2008). School policy framework: Implementation of the WHO global strategy on diet, physical activity and health. Retrieved from http://apps.who.int/iris/handle/10665/43923 Pan, A., Sun, Q., Czernichow, S., Kivimaki, M., Okereke, O. I., Lucas, M., … Hu, F. B. (2011). Bidirectional association between depression and obesity in middle-aged and older women. International Journal of Obesity, 36(4), 595–602. Parsons, T. J., Power, C., Logan, S., & Summerbelt, C. D. (1999). Childhood predictors of adult obesity: a systematic review. International Journal of Obesity, 23. Retrieved from http://www.bewegenismedicijn.nl/files/downloads/Parsons%20et%20al.,%201999.pdf Pietiläinen, K. H., Kaprio, J., Borg, P., Plasqui, G., Yki-Järvinen, H., Kujala, U. M., … Rissanen, A. (2008). Physical inactivity and obesity: a vicious circle. Obesity, 16(2), 409–414. Popkin, B. M. (2006). Global nutrition dynamics: the world is shifting rapidly toward a diet linked with noncommunicable diseases. The American Journal of Clinical Nutrition, 84(2), 289–298. Poskit, E. M. E. (1988). Management of childhood obesity. The Indian Journal of Pediatrics, 55(4), 470–472. doi:10.1007/BF02868426 Power, C., Lake, J. K., & Cole, T. J. (1997). Measurement and long-term health risks of child and adolescent fatness. International Journal of Obesity and Related Metabolic Disorders: Journal of the International Association for the Study of Obesity, 21(7), 507– 526. Pratt, M., Macera, C. A., & Blanton, C. (1999). Levels of physical activity and inactivity in children and adults in the United States: current evidence and research issues: Medicine & Science in Sports & Exercise, 31(Supplement 1), S526. doi:10.1097/00005768- 199911001-00007 RAO, S. (2011). STUDY OF BODY COMPOSITION IN CHILDREN AGED SEVEN TO NINE YEARS USING BIO-ELECTRIC IMPEDANCE ANALYSIS & ANTHROPOMETRY. Rajiv Gandhi University of Health Sciences. Retrieved from http://14.139.159.4:8080/jspui/handle/123456789/7101 Reddy, S. P., Resnicow, K., James, S., Funani, I. N., Kambaran, N. S., Omardien, R. G., … Mbewu, A. (2012). Rapid Increases in Overweight and Obesity Among South African Adolescents: Comparison of Data From the South African National Youth Risk Behaviour Survey in 2002 and 2008. American Journal of Public Health, 102(2), 262– 268. doi:10.2105/AJPH.2011.300222 Rolland-Cachera, M.-F. (1995). Prediction of adult body composition from infant and child measurements. Body Composition Techniques in Health and Disease, 36, 100. University of Ghana http://ugspace.ug.edu.gh 74 Rothman, K. J. (2008). BMI-related errors in the measurement of obesity. International Journal of Obesity, 32(S3), S56–S59. doi:10.1038/ijo.2008.87 Sahota, P., Rudolf, M. C. J., Dixey, R., Hill, A. J., Barth, J. H., & Cade, J. (2001). Randomised controlled trial of primary school based intervention to reduce risk factors for obesity. BMJ : British Medical Journal, 323(7320), 1029. Sallis, J. F. (1993). Epidemiology of physical activity and fitness in children and adolescents. Critical Reviews in Food Science & Nutrition, 33(4-5), 403–408. Savva, S. C., Kourides, Y., Tornaritis, M., Epiphaniou-Savva, M., Chadjigeorgiou, C., & Kafatos, A. (2002). Obesity in children and adolescents in Cyprus. Prevalence and predisposing factors. International Journal of Obesity and Related Metabolic Disorders: Journal of the International Association for the Study of Obesity, 26(8), 1036–1045. Schulz, L. O., Bennett, P. H., Ravussin, E., Kidd, J. R., Kidd, K. K., Esparza, J., & Valencia, M. E. (2006). Effects of traditional and western environments on prevalence of type 2 diabetes in Pima Indians in Mexico and the US. Diabetes Care, 29(8), 1866–1871. Siervo, M., Grey, P., Nyan, O. A., & Prentice, A. M. (2005). Urbanization and obesity in The Gambia: a country in the early stages of the demographic transition. European Journal of Clinical Nutrition, 60(4), 455–463. doi:10.1038/sj.ejcn.1602337 Swamy, P. (2011, July 8). Simple Tips to Reduce Obesity Naturally: Childhood Obesity - Prevention Programs. Simple Tips to Reduce Obesity Naturally. Retrieved from http://simpletipstoreduceobesity.blogspot.com/2011/07/childhood-obesity-prevention- programs.html Trost, S. G., Sirard, J. R., Dowda, M., Pfeiffer, K. A., & Pate, R. R. (2003). Physical activity in overweight and nonoverweight preschool children. International Journal of Obesity, 27(7), 834–839. doi:10.1038/sj.ijo.0802311 WHO. (1986). Use and interpretation of anthropometric indicators of nutritional status. Bulletin of the World Health Organization, 64(6), 929–941. WHO. (1995). Physical status: The use of and interpretation of anthropometry, Report of a WHO Expert Committee. Retrieved from http://apps.who.int/iris/handle/10665/37003 WHO. (2006). WHO Report 2006. WHO. (2008). WHO Report 2008. WHO. (2010). WHO | World health report. WHO. Retrieved July 18, 2014, from http://www.who.int/healthsystems/topics/financing/healthreport/whr_background/en/ Wickramasinghe, V. P., Cleghorn, G. J., Edmiston, K. A., Murphy, A. J., Abbott, R. A., & Davies, P. S. W. (2005). Validity of BMI as a measure of obesity in Australian white University of Ghana http://ugspace.ug.edu.gh 75 Caucasian and Australian Sri Lankan children. Annals of Human Biology, 32(1), 60–71. doi:10.1080/03014460400027805 Wiseman, M. (2008). The second World Cancer Research Fund/American Institute for Cancer Research expert report. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Proceedings of the Nutrition Society, 67(03), 253–256. Yach, D., & Beaglehole, R. (2004). Globalization of risks for chronic diseases demands global solutions. Perspectives on Global Development and Technology, 3(1), 213–233. University of Ghana http://ugspace.ug.edu.gh 76 APPENDIX APPENDIX 1A: CONSENT FORM FOR STUDY PARTICIPANTS Project Title: Risk factors for obesity among adolescents in Senior High Schools in the Tema Metropolis Institution of affiliation: School of Public Health, University of Ghana, Legon. Background of interviewer: My name is …………………………………………………. from ……………………………………………. (I am a student who is here) or (I am helping a student) to collect data purely for academic work for a degree in Masters in Public Health Procedure: Information required from you for this study includes background characteristics and risk factors for obesity among adolescents. Data collection is through the administration of a structured questionnaire, taking of weight and height. Risks and benefits: There are minimum or no risks if you take part in this study. There are also no incentives but the information you provide will help you improve on your health and that of your loved ones. Right to refuse: Your consent to participate in this study is voluntary and you can withdraw from this study at any time. Anonymity and Confidentiality: You are assured of strict anonymity and confidentiality on any information you give. If you have any further information or questions about the study, you may contact the principal investigator, Abdulai Mohammed Salifu, on phone number: 020 8077 230 or email: doctorcalculus@yahoo.com University of Ghana http://ugspace.ug.edu.gh 77 Your rights as a Participant: This research has been reviewed and approved by the Ethical Review Committee of the Ghana Health Service. If you have any questions about your rights as a research participant you can contact the ERB Office between the hours of 9am – 4pm on Monday to Friday through 0507041223 (Miss Hannah Frimpong). I have read the information above, or it has been read to me. I consent voluntarily to be a participant in this study Name of Participant :…………………………………………………. Signature or Thumb print of Participant : ………………………………………………….. Date : ………………………………………………….. Thank you for agreeing to participate Name of witness : ………………………………………………….. Signature or Thumb print of witness :…………………………………………………… Date :…………………………………………………… I confirm that the individual has not been coerced into giving consent, and the consent has been given freely and voluntarily. Name of Researcher or Principal investigator :…………………………………………….. Signature of Researcher :…………………………………………….. Date :……………………………………………… University of Ghana http://ugspace.ug.edu.gh 78 APPENDIX 1B: CONSENT FORM FOR PARENT OR GUARDIAN Project Title: Risk factors for obesity among adolescents in Senior High Schools in the Tema Metropolis Institution of affiliation: School of Public Health, University of Ghana, Legon. Procedure: Information required from you for this study includes background characteristics and risk factors for obesity among adolescents. Data collection is through the administration of a structured questionnaire, taking of weight and height. Risks and benefits: There are minimum or no risks if you take part in this study. There are also no incentives but the information you provide will help you improve on the health of your child and loved ones. Right to refuse: You have a right to refuse your child to participate Anonymity and Confidentiality: You are assured of strict anonymity and confidentiality on any information that is given. If you have any further information or questions about the study, you may contact the principal investigator, Dr Abdulai Mohammed Salifu, on phone number: 020 8077 230 or email: doctorcalculus@yahoo.com Your rights as a Parent or Guardian: This research has been reviewed and approved by the Ethical Review Committee of the Ghana Health Service. If you have any questions about your rights as a research participant you can contact the ERB Office between the hours of 9am – 4pm on Monday to Friday through 0507041223 (Miss Hannah Frimpong). University of Ghana http://ugspace.ug.edu.gh 79 I have read the information above, or it has been read to me. I consent voluntarily for my child to be a participant in this study Name of Parent or Guardian :………………………………………………….. Signature or Thumb print of Participant : ………………………………………………….. Date : ………………………………………………….. Thank you for agreeing for your child to participate Name of Researcher or Principal investigator :…………………………………………….. Signature of Researcher :…………………………………………….. Date :……………………………………………… University of Ghana http://ugspace.ug.edu.gh 80 APPENDIX 1C: QUESTIONNAIRE FOR RISK FACTORS FOR OBESITY AMONG SENIOR HIGH SCHOOLS IN THE TEMA METROPOLIS INTRODUCTION Greetings, my name is…………………………………………... I am a member of a team from the University of Ghana conducting a research on the risk factors for obesity among Senior High Schools in the Tema Metropolis. If you agree to take part in this study, I will give you a questionnaire to fill and then measure you height and weight. The questions and measurements will take about 25 to 30 minutes. Your responses to all questions will be confidential and will not be shared with anyone other than members of our study team. No answer is wrong. Your participation in the study is voluntary and you are free to end the interview or measurement process at any time. However, I will be happy if you participate in the study to contribute to existing knowledge on obesity in adolescent. Questionnaire number :…………………………………………………………………... Name of interviewer :…………………………………………………………………... Date :…………………………………………………………………… University of Ghana http://ugspace.ug.edu.gh 81 SECTION A: SOCIODEMOGRAPHIC CHARACTERISTICS 1 Date of birth 2` Age 3 Sex MALE …………1 FEMALE …………2 4 Level of Education SHS 1 …………1 SHS 2 …………2 SHS 3 …………3 5 Ethnicity GA …………1 EWE ……..2 AKAN …….3 DAGOMBA …….4 FANTE ………5 OTHER ……….6 6 Residential area (where do you stay?) TEMA . . . . . . . . . 1 ASHAIMAN . . . . . . . . . 2 KPONE . …......3 OTHERS . . . . . . . . .4 7 Residential status in school DAY . . . . . . ..1 BOARDING . . . . . . . .2 University of Ghana http://ugspace.ug.edu.gh 82 SECTION B: SOCIOECONOMIC STATUS OF PARENTS OR GUARDIANS 8 Father’s highest level of education NONE …......1 PRIMARY ……..2 JHS/MIDDLE SCH ….……3 SHS ..…….4 TERTIARY …………5 9 Father’s occupation 10 Mother’s highest level of education NONE …......1 PRIMARY ……..2 JHS/MIDDLE SCH ….……3 SHS ..…….4 TERTIARY …………5 11 Mother’s occupation 12 Guardian’s highest level of education NONE …......1 PRIMARY ……..2 JHS/MIDDLE SCH ….……3 SHS ..…….4 TERTIARY …………5 13 Guardian’s occupation 14 Relationship to guardian AUNT …......1 UNCLE ……..2 SISTER ….……3 BROTHER ..…….4 University of Ghana http://ugspace.ug.edu.gh 83 15 Type of residence OWNED BY PARENTS ......1 OWNED BY GUARDIAN …..2 FAMILY HOUSE ….……3 RENTED ..…….4 16 Do you own any of the following? YES-1 NO-2 TELEVISION DVD PLAYER REFRIGERATOR MICROWAVE GENERATOR CAR PERSONAL COMPUTER ELEC OR GAS COOKER MOTORBIKE 17 Do you subscribe to any TV NETWORK? YES ……….1 NO ……….2 University of Ghana http://ugspace.ug.edu.gh 84 SECTION C: DIETARY HABITS 18 How many times do you eat in a day? ONE …......1 TWO ……..2 THREE ….……3 FOUR ….….4 OTHER …..……5 19 How many times do you eat breakfast in a week? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 NONE .…….4 OTHER ………5 20 How many times do you eat snacks in a day? ONE …......1 TWO ……..2 THREE ….……3 FOUR ….….4 OTHER …..……5 21 How many times do you eat snacks in a week DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 NONE .…….4 OTHER ………5 University of Ghana http://ugspace.ug.edu.gh 85 22 What do you eat for snacks? 23 How many times do you eat lunch in a week? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 NONE .…….4 OTHER ………5 24 How many times do you eat supper in a week? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 NONE .…….4 OTHER ………5 25 How many times do you eat vegetables in a week? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 NONE .…….4 OTHER ………5 26 How many times do you eat fruits in a week? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 NONE .…….4 OTHER ………5 University of Ghana http://ugspace.ug.edu.gh 86 27.1 What did you eat for breakfast yesterday? Beverages(e.g. koko, tea) …....1 Grains(e.g. rice, waakye) ……....2 Solid starchy foods(e.g. fufu, banku) ....…3 OTHER ……….4 27.2 What did you eat for lunch yesterday? Beverages(e.g. koko, tea) …....1 Grains(e.g. rice, waakye) ……....2 Solid starchy foods(e.g. fufu, banku) ....…3 OTHER ……….4 27.3 What did you eat for supper yesterday? Beverages(e.g. koko, tea) …....1 Grains(e.g. rice, waakye) ……....2 Solid starchy foods(e.g. fufu, banku) ....…3 OTHER ……….4 27.4 What else did you eat yesterday? 28.1 What did you eat for breakfast three days ago? Beverages(e.g. koko, tea) …....1 Grains(e.g. rice, waakye) ……....2 Solid starchy foods(e.g. fufu, banku) ....…3 OTHER ……….4 28.2 What did you eat for lunch three days ago? Beverages(e.g. koko, tea) …....1 Grains(e.g. rice, waakye) ……....2 Solid starchy foods(e.g. fufu, banku) ....…3 OTHER ……….4 University of Ghana http://ugspace.ug.edu.gh 87 28.3 What did you eat for supper three days ago? Beverages(e.g. koko, tea) …....1 Grains(e.g. rice, waakye) ……....2 Solid starchy foods(e.g. fufu, banku) ....…3 OTHER ……….4 28.4 What else did you eat three days ago? 29 Do you patronize any of the following foods? Fried Rice, Fried Chicken, Burger, Pizza, Potato chips YES ……….1 NO ……….2 30 If YES, How often? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 OTHER ………5 NA ……….. 9 31 Do you patronize any of the following drinks? Fanta, coke, sprite, other soft drinks and minerals YES ……….1 NO ……….2 32 If YES, How often? DAILY …......1 1-3 DAYS/WEEK .……..2 4-6 DAYS/WEEK ...……3 OTHER ………5 NA ……….. 9 University of Ghana http://ugspace.ug.edu.gh 88 SECTION D: FAMILIAL OR GENETIC HISTORY 33 Do you think your father is fat? YES ……….1 NO ……….2 34 Do you think your mother is fat? YES ……….1 NO ……….2 35 Do you think your guardian is fat? YES ……….1 NO ……….2 NA ………9 36 Do you know of any relative you think is fat? YES ……….1 NO ……….2 37 If YES to 36, what is the relationship? AUNT …......1 UNCLE ……..2 SISTER ...……3 BROTHER .…….4 OTHER ………5 NA ………9 38 Were you fat when you were a child under 10year old? YES ……….1 NO ……….2 University of Ghana http://ugspace.ug.edu.gh 89 SECTION E: PHYSICAL ACTIVITY 39 Do you consciously engage in physical activity? YES ……….1 NO ……….2 40 If YES to 39, how long do you engage in physical activity? <30 MINUTES ……….1 30-60 MINUTE ……….2 61-120 MINUTES ……….3 >120MINUTES ……….4 NA ……….9 41 If YES to 39, what type of physical activity do you engage in? ACTIVITIES ……….1 ………………………………………………. NA ………9 42 How many hours do you spend watching TV OR Movies in a day( when you are at home or vacation)?: < 1HOUR ……….1 1-4 HOURS ……….2 > 4 HOURS …………3 43 Is Physical Education part of your school curriculum? YES ……….1 NO ……….2 44 If YES to 43, what type of physical education activity do you engage in? ACTIVITIES ……….1 ………………………………………………. NA ………9 45 Was Physical Education part of your JHS curriculum? YES ……….1 NO ……….2 University of Ghana http://ugspace.ug.edu.gh 90 SECTION F: ANTHROPOMETRIC MEASUREMENTS MEASUREMENT FIRST MEASUREMENT SECOND MEASUREMENT AVERAGE HEIGHT (m) WEIGHT (kg) BMI (kg/ m2) BMI STATUS OBESE ……....1 OVERWEIGHT .……..2 NORMAL .......…3 UNDERWIGHT ……….4 University of Ghana http://ugspace.ug.edu.gh