This paper analyses the impact of spousal care-giving on survivors' depressive symptoms six months into widowhood, and examines the applicability of a ‘relief model’ of spousal adjustment during bereavement. We examine several aspects of the care-giving situation, including care-giver stress, care-giving demands, and type and duration of care and how these affect survivors' depressive symptomatology. The sample is drawn from two waves of the Changing Lives of Older Couples (CLOC) survey, which was conducted in the United States in the Detroit Metropolitan Area, Michigan. The first wave of data was collected from couples and the second from the surviving spouse six months after the death of the partner. We use multiple regression analysis to examine the effects of key variables on depressive symptoms six months into widowhood, controlling for various demographic characteristics and personal circumstances. The results demonstrate that the duration of care-giving is the most influential predictor of survivors' depressive symptoms six months after the death. Indeed, long-term care-givers experience greater relief than both non-caregivers and short-term care-givers, as the predicted probabilities indicate. The results lead us to emphasise that care-giving and spousal bereavement should be studied as related processes rather than distinct phenomena. Indeed, relief from a chronically stressful care situation may actually ameliorate the negative effect of spousal loss for survivors.