In this paper we examine the effects of increasing as well as decreasing caregiving demands on depressive symptomatology. In addition, we focus on spousal caregivers' activity restriction as an explanatory mechanism for changes in depressive symptomatology in the caregiving context. Two databases are used to answer our research questions. An increase of caregiving demands is assessed in study 1, which includes prospective data on 127 spousal caregivers of stroke, hip fracture, congestive heart failure and myocardial infarction patients. A decrease of caregiving demands is examined in study 2, which includes prospective data on 110 spousal caregivers of bypass operation patients. The results generally support the hypothesis that an increase in caregiving demands results in increased depressive symptomatology, while a decrease in caregiving demands reduces depressive symptomatology. The results also support the notion of activity restriction as a critical mediator of changes in depressive symptoms. Cross-sectionally it mediates the association between caregiving and depressive symptomatology, and longitudinally it contributes to changes in depressive symptomatology in both samples.