Aim of study - The aim of the present work was to compare patients with advancedcancer receiving specialized palliative home care with those receiving conventional carewith regard to time spent at home and place of death (Paper I), factors associated with thepatients’ choice of palliative care (Paper II), and to describe the health-related quality oflife (HRQL) and the occurrence and changes in symptoms over the last three months ofthe patients’ lives (Paper III). In addition, sociodemographic and clinical differencesbetween participants and dropouts were explored in order to establish whether themissing data could be regarded as missing at random (Paper IV).Background - A hospital-based advanced home care programme was initiated in Malmö,Sweden in December1995, in order to enable patients to remain at home for a longerperiod and to choose to die at home. The allocation to advanced home care (AHC) orconventional care (CC) was performed according to the patients’ preferences. Thisintervention offered an opportunity to investigate the effect of AHC and to describe theHRQL of patients with a progressive, life-threatening disease.Methods - A prospective longitudinal comparative study was performed including 297patients consecutively recruited over 2½ years. The two groups of patients (AHC andCC) were compared and the main outcome variables were place of death, time spent ininstitution or at home, and the patients’ HRQL. The latter was assessed monthly by usingself-reported questionnaires, including the European Organisation for Research andTreatment of Cancer Core Quality of Life Questionnaire (EORTC QLQ-C30), the Impactof Event Scale (IES), five questions about social support (MA) and two items concerninggeneral well-being (NT). At the same time, in Trondheim, Norway, a similar palliativecare intervention programme was started. The design of the study was slightly different,but patients were evaluated using the same sociodemographic and clinical variables, andquestionnaire data were collected in the same way as in the Swedish study.Results - The AHC patients spent more time outside the hospital and more patients diedat home compared with patients in the CC group (Paper I). However, there was adifference between patients in the two interventions. Compared with the CC patients,patients who chose the AHC programme had lived longer with their cancer diagnosis,had a shorter survival period after study enrolment, and had poorer performance status. Inaddition, although all patients had reduced HRQL at inclusion, the AHC patients reportedpoorer functioning and more symptoms than the CC patients (Paper II). Results from theHRQL questionnaires returned during the last three months of the patients’ lives wereanalysed in order to describe the symptom burden of dying patients. Due to considerableattrition over the study period, data from the Norwegian study were included in order toincrease the study sample to 116 patients. Generally, a marked, but gradual deteriorationwas seen during the patients’ last three months of life. A more rapid increase in problemswas observed between two and one month before death regarding Physical, Cognitiveand Social function, as well as Fatigue. The mean scores for Pain, Nausea/Vomiting,Sleeping problems, Diarrhoea and Financial impact were stable over time (Paper III).Analyses of the characteristics of patients who dropped out of the study after filling in thefirst questionnaire after inclusion confirmed that they had a less favourable prognosis,with regard to both clinical variables and HRQL data. This pattern was not found whenanalysing data from patients dropping out of the study during the last two months beforedeath (Paper IV).Conclusions -The results obtained in this work add knowledge about patients’preferences regarding care at the end of life, and changes in their HRQL during the lastfew months of life. The palliative intervention with AHC proved successful, and enabledmore patients to stay at home for a longer period, and to die at home. Dying at home isstrongly associated with the patients’ preference to do so. Due to the design of the study,the impact of such preferences cannot be separated from the impact of the AHC. Theresults indicate that patients are reluctant to accept home care until necessary due to theseverity of functioning impairments and symptom burden. It is important to be aware ofthe high level of symptoms in the patients’ last three months of life. Symptom assessmentusing standardised questionnaires may enhance the focus on symptom management.Results from HRQL studies must be interpreted with care due to non-random attrition.This is less evident close to death.