To study the relationship between social participation and depression trajectory and to explore the mediating effect of quality of life.

Based on the data of the four periods of China Health and Retirement Longitudinal Study (CHARLS) from 2011 to 2018, the heterogeneous trajectory of depressive symptoms in the elderly was identified by group-based trajectory model, the relationship between social participation and depression trajectory was studied by multi-class logistic regression, and the mediating effect of quality of life on social participation and depression trajectory of the elderly was analyzed by Karlson-Holm-Breen （KHB） method.

A total of 4 931 elderly people aged between 60 and 105 with at least two periods of depression were included, and four depression trajectory groups were identified: long-term non-depression group (61.69%), increased depression group (16.69%), depression remission group (13.83%), and long-term high depression group (7.79%). Compared with the subjects in the long-term non-depression group, the more social participation, the less likely the depression trajectory was attributed to the depression remission group, the increased depression group, and the long-term high depression group (RRR=0.899, 95%CI: 0.812-0.994; RRR=0.885, 95%CI: 0.786-0.995; RRR=0.747, 95%CI: 0.630-0.885). In addition, quality of life played a significant mediating role in the relationship between the two, and the mediating effect of self-rated health accounted for the largest contribution of 30.33% in the elevated depression group. The mediating effect of physical pain accounted for the largest contribution of 27.94% and 29.09, respectively in the depression remission group and the long-term high depression group. Similar results were obtained in the robustness test.

There is heterogeneity in the depression trajectory of the elderly, social participation has a significant impact on the attribution of depression trajectory, and quality of life is an important factor of social participation affecting depression trajectory.