The objectives of this study were to identify and describe the demographics and social conditions as well as characteristics of the places of death in a randomly selected sample that died in a county in Sweden during 2001. The present study reports part of the findings from a larger study undertaken during 2003 using a survey design of retrospective reviews of death certificates, medical records, and nursing records. Among several noteworthy findings are: first, an examination of these individuals’ living arrangements reveal not only a large and vulnerable group that lives alone at the end-of-life but also a group living separated from their partners near end-of-life; second, individuals differ with respect to residence prior to death, that is private homes versus residential care facilities, imparting highly different contexts of care at the end-oflife, and third, a high prevalence of institutionalized deaths

demonstrates that places of death other than hospitals and residential care facilities are uncommon, if not rare. It is concluded that increased attention to the social circumstances of the time period surrounding individuals’ at the end-of-life is essential.

There is need to develop more integrated models of care for dying people. Contemporary services available are not designed to meet a wide range of peoples’ needs at the endof- life. Home-based care, residential care and hospital care must be adapted to the changing patterns of dying. The services available should be organized to the benefit of the users rather than around the providers. This is the first study of its kind of a Swedish sample and while the data

are limited to one county in Sweden its findings may contribute to a deeper understanding of demographic and social patterns at the end-of-life in general.

To improve the understanding of specific clinical problemsat the end of life, including the role of advancing age, physicalfunction and cognitive function. Methods: The study is partof an explorative survey of data relevant to end-of-life healthcareservices during the last 3 months of life of a randomly selectedsample of the population of a Swedish county. Data were selectedthrough retrospective reviews of death certificates and medicalrecords, and comprise information from 12 municipalities and229 individuals. Results: A range of prevalent concerns wasfound. Overall deterioration, urinary incontinence, constipation,impaired skin integrity, anxiety and sleep disturbances weresignificantly associated with dependency on others for activitiesof daily living; pulmonary rattles and swallowing disturbanceswere associated with cognitive disorientation; excepting cough,advancing age did not have significant impacts on these prevalentclinical concerns. Conclusions: A range of distressing conditionsconstitute a common pathway for many individuals at or nearthe end of life. The incorporation of health promotion as aprinciple of palliative care will probably benefit individualsat the end of life, and includes a proactive focus and emphasison enhanced well-being at the time of diagnosis of a life-threateningillness. For individuals with physical and cognitive limitationsimparting a state of dependency, it is reasonable to provideassurance of care for individuals' specific needs by professionalswith both training for and competence in this special and sometimesunique clinical environment.

End-of-life care poses a growing clinical and policy concern since most people who are dying utilize health-care services during this period of life. Hence, end-of-life care is a common and integral part of the care provided by health-care systems. There is a growing call for the implementation of a palliative approach as an integral part of all end-of-life care. The purpose of this study was thus to provide policy-makers, health-care providers and professional caregivers with increased knowledge about mainstream patterns of health-care utilization during end-of-life. The patterns of use of health-care services in a Swedish population who accessed the health-care system during their last 3 months of life were in this study examined through a retrospective examinations of medical and nursing records (n = 229). We found high prevalences of use of both hospital care, primary care and care provided in people's homes and nearly three quarters of the persons included in the study used between two and three health-care services. However, the probability of using different health-care services was found to be strongly depending on demographic, social, functional and disease related characteristics. The study reveals a considerable use of different health-care services during end-of-life. It is hence essential to, on one hand delineate how such health-care services best can support people at the end-of-life, and on the other hand develop policies which facilitate the process of dying, both in hospitals as well as in peoples’ homes. Implications for policy are discussed.

Palliative care is increasingly organized within the setting of formal health care systems but the demarcation has become unclear between, on the one hand, care directed at cure and rehabilitation and palliative care aimed at relief of suffering on the other. With the purpose to increase the understanding about the turning point reflecting identification of dying and reorientation of care, this study explores this phenomenon as determined from health care records of a representative sample (n = 229). A turning point was identified in 160 records. Presence of circulatory diseases, sporadic confinement to bed, and deterioration of condition had a significant impact upon the incidence of such turning point. The time interval between the turning point and actual death ranged between one and 210 days. Thirty percent of these turning points were documented within the last day of life, 33% during the last 2–7 days, 19.5% during the last 8–30 days, 13% during the last 31–90 days, and 4.5% during the last 91–210 days of life. The time interval between the turning point and actual death was significantly longer among individuals with neoplasm(s) and significantly shorter among individuals suffering from musculoskeletal diseases. Perhaps this reflects a discrepancy between the ideals of palliative care, and a misinterpretation of the meaning of palliative care in everyday clinical practice. The findings underscore that improvement in timing of clinical identification of dying and reorientation of care will likely favour a shift from life-extending care to palliative care.