Objective: To describe the cross sectional and longitudinal associations of dietary patterns and depressive symptoms among college students, so as to provide a reference for improving college students mental and physical health. Methods: From April to May 2019, 1 110 college students were randomly sampled in Hefei City, Anhui Province and Shangrao City, Jiangxi Province, and a follow up survey was conducted from September to October 2019. The depression subscale of Depression Anxiety Stress Scale-21 (DASS-21) was used to investigate depressive symptoms in college students. The Semi Quantitative Food Frequency Questionnaire (SQFFQ) was used to investigate the eating behaviors of college students. Diet patterns were analyzed by principal component analysis and exploratory factor analysis, and the scores were classified T1, T2 and T3. Multivariate Logistic regression models were used to analyze the association of baseline dietary patterns and depressive symptoms at baseline and follow up. Results: The detection rates of mild, moderate and above depressive symptoms among baseline college students were 7.03% and 14.23% , respectively. The results of principal component analysis and exploratory factor analysis showed that the dietary patterns of college students were divided into four patterns: sugary drinks, meat, fast food and healthy food. At baseline and follow up surveys, the detection rate of depressive symptoms in sugary drinks, meat and fast food was the highest ( χ 2=21.51, 32.25, 22.21； 23.54, 13.91, 19.98, P ＜0.05). The results of multivariate Logistic regression model showed that the meat and fast food T3 showed positive association with baseline and moderate follow up, and the fast food T3 showed positive association with mild depressive symptoms at follow up ( P ＜0.05). Conclusion Meat and fast food diet patterns can increase the risk of depressive symptoms, suggesting that improving their eating patterns has a positive effect on promoting the mental health of college students.