This was a quasi-experimental study designed to determine whether, in patients with advanced cancer, a palliative home care team (PHCT) modified hospital utilization in the last six months before death. Of 2503 cancer deaths in the municipality of Genoa, Italy, in 1991, 189 (7.5%) received care from a PHCT. Three hundred and seventy-eight controls matched for primary tumour were selected. The groups were similar in terms of age, gender and most other demographic variables, except that educational level was lower, and times to death, from first diagnosis and from diagnosis of advanced or metastatic cancer, were longer among PHCT patients compared with the controls. Before referral to a PHCT, or a matched time in controls, both groups spent about 15% of days in hospital. After admission to a PHCT, the percentages of days in hospital increased in both groups as death approached, but it was much higher in the control group (30.3%; 95% confidence interval (CI): 26-34) than in the PHCT group (19.0%; 95% CI: 15-23). The difference between groups was most marked in the last month of life, and disappeared among those patients who were in care for more than 120 days (throughout the course of their illness). We conclude that a PHCT appears to reduce days in hospital and allows patients to spend more time at home. Differences in time in care between groups requires further investigations.