本研究旨在透過橫斷性研究與長期追蹤研究探討臺灣青少年身體活動、坐式行為、飲食習慣與肥胖之關係。2008年9月進行橫斷性研究，樣本3,221人來自臺灣11所國中，另外，對其中1所國中進行持續追蹤研究（樣本221人）。資料搜集包含身體活動、看電視、使用電腦、寫功課等時間，以及吃早餐與喝飲料習慣。持續追蹤研究共蒐集三次（2008年9月、2009年9月、2011年6月）。本研究採用多變量邏輯迴歸檢驗身體活動量、飲食習慣與肥胖之橫斷性關係；另外，透過長期追蹤的三次資料蒐集，以概化估計方程式進一步探討該關係。研究結果顯示：橫斷性研究發現，習慣喝含糖飲料與每天看電視2小時或以上之青少年，比習慣喝白開水與每天看電視少於2小時之青少年，肥胖機率較高。在長期追蹤研究中，廣義估計方程式結果顯示，只有一個變項達到顯著水準，亦即習慣喝含糖飲料之青少年，其肥胖機率比習慣喝白開水之青少年高。本研究結論：本研究與既有之研究證據均顯示，習慣喝含糖飲料與肥胖有顯著相關；然而，肥胖與身體活動、坐式行為及吃早餐等行為之關係仍然不明確。

This study used cross-sectional and longitudinal analyses to examine physical activity, sedentary behaviors and eating habits and their associations with occurrence of obesity among Taiwanese adolescents. A cross-sectional survey was conducted in September 2008 with a total number of 3,221 seventh graders from 11 junior high schools in Taiwan. Among them, one school was selected for the follow-up study (n = 221). The parameters, including time spent on physical activity, watching TV, computer use and academic-related activities, and data on breakfast consumption and drinking habits were collected. The data has been collected in three sessions (2008 September, 2009 September and 2011 June) with the follow-up group. A multiple logistic regression was performed to evaluate the cross-sectional association between activity levels and eating habits with obesity risk in 2008. A Generalized Estimating Equations (GEE) was employed to further examine this association with the three-wave data from the follow-up group. Cross-sectional results showed that adolescents, who intake sugar-sweetened beverages and watching TV two or more hours per day, were more likely to be overweight or obese than the reference groups. Longitudinally, significant association from GEE suggested that adolescents, who consumed sugar-sweetened beverages were more likely to be subsequently overweight or obese compared to those with water intake. This study added additional knowledge to the existing evidence showing that consuming sugar-sweetened beverages is associated with occurrence of obesity risk. The relationships between obesity and physical activity, sedentary behavior, and breakfast consumption remain equivocal.