School-based health programs implemented by the Malaysian Ministry of Education (MOE) through the National Physical Fitness Standard (SEGAK) assessments provided an important platform in health status monitoring among schoolchildren. However, to date, there is still no reliability study conducted on this method. Therefore, this study aimed to determine the reliability of the anthropometric data collected by physical education (PE) teachers in the SEGAK assessments. Anthropometry measurements of standard six school adolescents involved in the Health of Adolescents in Terengganu study were taken by trained researchers using a standardised protocol. The anthropometrics data were then compared with PE teachers’ measurements from the SEGAK assessments obtained from the specifically developed Health Monitoring System database. Reliability of the anthropometric measurements were analysed using Pearson’s correlation test, Intraclass Correlation Coefficients (ICC), Bland-Altman plot and Cohen’s Kappa statistics. Intraclass correlation coefficient between teacher-measured and researcher-measured values shows good correlation in weight (ICC = 0.93), height (ICC = 0.98) and BMI (ICC = 0.91). The Bland-Altman plot showed a relatively small difference in mean of weight, height, and BMI between teacher-measured and researcher-measured value. The mean difference between teacher-measured and researcher-measured value of weight, height, and BMI were 1.8kg, 0.1cm, and 0.8kg/m2 respectively. Overall, Cohen’s Kappa statistics showed substantial agreement (κ = 0.642) in BMI categorisation between the two measurements. Findings from reliability analysis conducted affirmed that anthropometrics assessments conducted by PE teachers in SEGAK assessments are reliable to be used for identification of body weight status among school children and adolescents particularly in Terengganu, Malaysia.

Worldwide prevalence of obesity among children and adolescents increased substantially over decades. Addressing potential risk factors of obesity among adolescents is very important for a successful intervention program in this population. The present study aimed to identify the sociodemographic determinants of obesity among school adolescents in Terengganu. A cross-sectional survey involving 3,798 school adolescents age 12 years old from 136 government primary schools in Kuala Terengganu and Besut districts were carried out from November 2014 to June 2015. For the purpose of this paper, the data for 2,842 school adolescents classified as either normal BMI (< +1SD) (n= 2,305) or obese (+2SD) (n= 537) based on WHO cut-off points were taken for analysis. Sociodemographic information on subjects and their parents were obtained from self-reported questionnaire. Anthropometric measurements were conducted by Physical Education teachers and uploaded into a specific developed database. The prevalence of obesity was relatively high in both genders in this study. Binary logistic regression analysis found gender, parental BMI, household income, household size and maternal working status were independently associated with obesity among school adolescents in this population. In the final model, being male, having working mother, and having obese parents were identified to be potential risk factors for obesity whilst having large household size lower the risk of obesity among these adolescents. Prevention programs are needed to increase awareness about the risk factors of obesity in adolescent and interventions should now focus on family member as well mainly the parents.
Sociodemographic ; obesity ; school adolescents ; Terengganu ; Malaysia

Worldwide prevalence of obesity among children and adolescents increased substantially over decades. Addressing potential risk factors of obesity among adolescents is very important for a successful intervention program in this population. The present study aimed to identify the sociodemographic determinants of obesity among school adolescents in Terengganu. A cross-sectional survey involving 3,798 school adolescents age 12 years old from 136 government primary schools in Kuala Terengganu and Besut districts were carried out from November 2014 to June 2015. For the purpose of this paper, the data for 2,842 school adolescents classified as either normal BMI (< +1SD) (n= 2,305) or obese (+2SD) (n= 537) based on WHO cut-off points were taken for analysis. Sociodemographic information on subjects and their parents were obtained from self-reported questionnaire. Anthropometric measurements were conducted by Physical Education teachers and uploaded into a specific developed database. The prevalence of obesity was relatively high in both genders in this study. Binary logistic regression analysis found gender, parental BMI, household income, household size and maternal working status were independently associated with obesity among school adolescents in this population. In the final model, being male, having working mother, and having obese parents were identified to be potential risk factors for obesity whilst having large household size lower the risk of obesity among these adolescents. Prevention programs are needed to increase awareness about the risk factors of obesity in adolescent and interventions should now focus on family member as well mainly the parents.