This study aimed to examine the stability of depression subtypes over time or the multiple factors and mechanisms that influence these subtypes. Data from two waves of the China Health and Retirement Longitudinal Study (CHARLS) for 2013 (T1) and 2018 (T2), separated by 5 years were included in latent class analysis (LCA) and latent transition analysis (LTA). Specifically, scores on the 10-item Chinese version of the Centre for Epidemiological Studies Depression Scale (CES-D-10) of 2,828 Chinese individuals aged 60 years and older were analyzed. Older Chinese adults were categorized into three classes based on depressive symptoms: the severe depressive symptoms, the mild depressive symptoms, and transition groups. At T2, 70.4% of participants reported severe depressive symptoms, 62.1% reported mild depressive symptoms, and 49.2% were placed in the transition group. Patients in the severe depressive symptoms and mild depressive symptoms tended to switch to the transition group over time, with probabilities of 21.1% and 28.2%, respectively. Further analysis showed that sex, place of residence, marital status, ability to perform activities of daily living (ADLs), self-rated health status, and life satisfaction significantly predicted this transition. This study demonstrated a transition pattern of depressive symptoms in Chinese older adults. We should pay more attention to key elderly groups such as unmarried women, alcohol consumption, and poor self-rated health, rationally plan mental health resources, and establish a dynamic tracking service system to prevent the depression and related adverse consequences.