Abstract: Objective To explore whether the increase of television watching time can increase the risk of obesity in Nanjing residents.Methods Population based corss sectional study was conducted among local residents older than 35 years in three urban districts and one rural county in July 2004.The subjects who were non obese in the base line survey were selected to participate the follow-up survey in July 2007.Multiple linear regression andmultivaria telogistic regression analysis were used to estmiate the incidence risk of obesity.Results Of 3 936 subjects without obesity in base line survey,2 837 participated the follow-up survey,and the follow-up rate was 72.1%.Three year cumulative incidence rate of obesity was 2.7%(2.1% in men and 2.3% in women).After adujsted for age,the partic ipants watching TV per week 7.14h and>14 had a significant higher risk of developing obesity compared to those watching TV≤7h/week(Ad.jRR=2 353,95%CI=1.186-4.669;4.061,95%CI=2.105-7.833,reapectively).Adjusted for age and other relevant covariates,the RRs of obesity still increased gradually with the increase of the time watching TV(adj.RR=2.357,95%CI=1.126-4.934; 3.639,95%CI=1.761-7.517 for 7-14 and>14h/week,respectively).After additional adjusmtent for base line BMI,WC and HC,the positive associations were attenuated but rema ined statistically significant.In multivariate logistic regression analysis,each 1 hr/w kincrese in TV watching time was associated with 2.3% and 4.9% increase in like lihood of obesity for the men and women in the study,respectively.Conclusion Sedentary lifestyle indicated by prolonged TV watching time is related to significantly elevated risk of obesity.Our findings suggest the miportance of reducing sedentary behavior in the prevention of obesity.