Abstract title
Relationship of diet and physical activity with metabolic syndrome among urban and rural adolescents of north India
Co-authors	
M. Chopra1, A. Siddhu1, D. Bharadwaj2, N. Tandon3.
1Lady Irwin College, Food and Nutrition, New Delhi, India.
2Jawaharlal Nehru University, Systems Genomics Laboratory- School of Biotechnology, New Delhi, India.
3All India Institute of Medical Sciences, Endocrinology and Metabolism, New Delhi, India.
Abstract body	
Background
Increasing prevalence of metabolic syndrome and unhealthy lifestyle behaviour among adolescents residing in developing countries requires early identification and modification of risk factors as preventive measures to reduce the growing epidemic of adult metabolic diseases.

Aim
To determine the prevalence of metabolic syndrome (further categorized by location, gender and weight status) and assess its relationship with diet and physical activity (PA) among urban and rural adolescents of north India.

Method
In this cross-sectional study, schools (n=12) from urban (n=9; Delhi) and rural (n=3; Harsaru village, District Gurgaon, Haryana and Duhai village, District Ghaziabad, Uttar Pradesh) areas of north India were selected to enrol adolescents in the age group of 11-17 years. A total of 2484 adolescents were recruited from urban (n=1790) and rural areas (n=694), by setting up health camps in their respective schools. Measurement of anthropometric indices including height, weight (computation of BMI), waist circumference; sitting blood pressure (BP); and fasting blood samples for analyses of glucose and lipid profile were done in the screening phase. Information on diet was obtained using two 24hr recalls (1 school day and 1 weekend day), on non-consecutive days, and on PA using Global PA Questionnaire, version 2 (total PA and its domains) from the recruited participants. Metabolic syndrome was defined using the IDF criteria, and BMI for age guidelines by the International Obesity Task Force were used to classify adolescents according to weight status. Logistic regression analyses were conducted to assess relationship of metabolic syndrome with diet and PA variables. 

Results
Prevalence of metabolic syndrome for total sample was 3.5% (urban, 4.5% vs. rural, 1.0%, p<0.001; male, 3.7% vs. female, 3.4%, p>0.05; overweight/obese (OW/OB), 19.5% vs. normal weight, 0.5%, p<0.001). Adolescents with metabolic syndrome had significantly higher mean values of BMI, waist circumference, systolic BP, diastolic BP, total cholesterol, LDL-c, triglycerides; dietary energy, carbohydrate, fat, calcium, sodium, vitamin A, folic acid (free); sedentary behaviour, and lower values of HDL-c; dietary protein; total PA, moderate work, travel, vigorous and moderate leisure time PA (p-values ranging from <0.001-0.048) as compared to adolescents without metabolic syndrome. Logistic regression analyses with and without covariate adjustments are provided in Table 1. Dietary energy, carbohydrate and fat intakes significantly increased the odds of developing metabolic syndrome. Furthermore, among PA variables, sedentary behaviour significantly increased and moderate leisure time PA and total PA significantly decreased the likelihood of developing metabolic syndrome among adolescents, after adjustment for age, gender, location and BMI were done. 

Table 1: Association of metabolic syndrome with diet and PA variables using logistic regression analysis

 	Unadjusted	Unadjusted	Adjusted#	Adjusted#
 	OR (95% CI)	p-value	OR (95% CI)	p-value
Dietary variables (Average of 2 days)a
Energy (kcal)	1.20 (1.10-1.30)	0.016	1.10 (1.00-1.20)	0.031
Protein (g)	0.99 (0.97-1.01)	0.676	0.99 (0.96-1.02)	0.660
Carbohydrate (g)	1.16 (1.14-1.18)	0.024	1.06 (1.01-1.11)	0.047
Fat (g)	1.02 (1.00-1.03)	0.032	1.01 (1.00-1.02)	0.045
Calcium (mg)	1.00 (1.00-1.00)	0.998	1.00 (1.00-1.00)	0.239
Magnesium (mg)	1.00 (1.00-1.00)	0.002	1.00 (1.00-1.00)	0.013
Zinc (mg)	0.98 (0.85-1.14)	0.980	1.02 (0.85-1.21)	0.863
Vitamin C (mg)	1.00 (1.00-1.01)	0.479	1.00 (1.00-1.01)	0.503
Total dietary fibre (g)	0.84 (0.78-0.90)	0.045	0.99 (0.94-1.04)	0.687
PA variablesb
Total PA (MET/week)	0.97 (0.96-0.98)	<0.001	0.99 (0.98-1.00)	<0.001
Moderate work (MET/week)	1.00 (1.00-1.00)	0.980	1.00 (1.00-1.01)	0.574
Travel (MET/week)	1.01 (1.00-1.01)	<0.001	1.00 (1.00-1.01)	<0.001
Vigorous leisure time PA (MET/week)	0.99 (0.99-1.00)	0.025	1.00 (1.00-1.00)	0.042
Moderate leisure time PA (MET/week)	0.95 (0.90-1.00)	<0.001	0.98 (0.97-0.99)	0.001
Sedentary behaviour (minute/day)	1.20 (1.10-1.30)	<0.001	1.05 (1.01-1.09)	0.037
a,bIndividual analysis for diet and PA variables with metabolic syndrome. 
Significant variables from univariate analysis carried forward for multiple logistic regression analyses. #Adjusted for age, gender, location and BMI Significant values are highlighted. 
OR, odds ratio; CI, class interval; MET, metabolic equivalent of task. 


Discussion
Prevalence of metabolic syndrome was higher among urban and OW/OB as compared to rural and normal weight adolescents respectively, with elevated levels of anthropometry, clinical and biochemical variables. Furthermore, unhealthy diet and PA pattern was a characteristic feature among them with dietary energy, carbohydrate, fat and sedentary behaviour being positively associated with metabolic syndrome. Hence, a formal evaluation of lifestyle intervention for reducing cardio-metabolic risk is merited.

Conflict of Interest
I have no potential conflict of interest to disclose