We retrospectively reviewed 110 consecutive admissions to continuing palliative care units, which were designed as part of a regionalized, comprehensive palliative care program in Edmonton, Canada. Ninety-six patient charts met the criteria for evaluation. Demographic characteristics, and, when available, symptom profiles, cognitive status, and risk for a history of substance abuse were described. The medications on admission were tabulated, and in those 93 patients who had consults done by a palliative care consultant, these are compared to recommended medications. This study showed an older cohort of patient (mean +/- SD = 75 +/- 11 years) than had previously been described in a tertiary unit in the same community. Median length of stay was 21 days (range, 0-> 200 days). Cognitive impairment was higher than would be anticipated on the basis of age alone, with 32/47 [64% (confidence interval (CI) 55%-81%)] of patients who had had cognitive testing done on the day of consult being found to be cognitively impaired. Symptoms, as measured by the Edmonton Symptom Assessment Scale, were similar to those found for patients admitted to the tertiary palliative care unit. In the 93 patients who had palliative care consults done on admission, there were a total of 179 recommendations for medication or hydration changes. Overall compliance with these recommendations was 84% (CI, 79%-89%). The highest compliance was observed for recommendations to start hydration clysis [27/27, 100% (CI, 100%)], and the lowest rate was observed for altering or decreasing hypnotic medications [11/22, 50% (CI, 29%-71%)]. We conclude that the patients were of higher acuity than anticipated.