Objectives: Expanding on cross-sectional studies, associations are examined between religious involvement and the 6-year course of depressive symptoms in older adults. Methods: Subjects are 1,840 community-dwelling older adults (aged 55 to 85) participating in three measurement cycles of the Longitudinal Aging Study, Amsterdam. Assessments include aspects of religious involvement, depressive symptoms, physical health, self-perceptions, social integration, urbanization, and alcohol use. Results: Church attendance is negatively associated with the course of depressive symptoms, also after adjustment for explanatory variables. Among respondents with functional limitations, lower depression scores are found for those who attend church on a regular basis. For respondents who are bereaved or nonmarried, however, slightly higher depression scores are found for those with high levels of orthodox beliefs. Discussion: There is a consistent negative association over time between church attendance and depressive symptoms in older Dutch citizens. Both stress-buffering as well as depression-evoking effects of religious involvement are found.