Abstract Background Living will (LW) documents have been suggested to act in patients’ best interests in end-of-life care. It is unclear, however, whether the LWs influence medical decisions when death is impending. Aim To explore, whether a LW among the community-dwelling older people has an influence on intensity of end-of-life treatment, place of death, or length of the dying process. Design A 10-year follow-up; data concerning place and cause of death, event history, intensity of care, decision-making process, and length of the dying process were collected from the death certificates. Setting/Participants In a community-based longitudinal cardiovascular prevention trial (DEBATE) in Helsinki, home-dwelling older people with an atherosclerotic disease (n=378, mean age 80.2years) were questioned about the pre-existence of a written LW at baseline (n=44) in 2000. In 2010, all death certificates (n=207) were analysed, comparing people with a LW (n=30) at baseline with those without (n=177). Results Of the deceased, 77% died in hospital. Intensity of end-of-life treatment or cause of death did not differ between the groups. However, dying at home was more common among persons with a LW at baseline (16.7% vs. 5.6%, P=0.031), and length of the dying process was shorter in this group (<1week in 50.0% vs. 27.1%, P=0.013). Conclusions Length of the dying process was often shorter and home death more common among those with a LW at the beginning of the follow-up. More insight into the dying process is required to ensure that the decision-making process better accommodates the preferences of older people.