Cancer pain management in home hospice settings: a comparison of primary care and oncologic physicians.

We compare the effectiveness of terminal cancer pain management delivered by primary care physicians with that of specialists in home-based hospice settings. Each visit record for 223 outpatients in three hospice programs was abstracted for physician type, patient age, medication usage, level of pain reported, cancer type, and metastatic status. Thirteen percent of patients reported no pain at any visit and 19% reported pain at all visits; half of the patients reported pain at two thirds of their visits. No difference was found in the presence or absence of pain between primary care and oncologic patients. When available, the level of pain reported (0-10 scale) was statistically (p < 0.01) but not clinically different between physician groups; average pain rating for primary care patients was 3.7 while the mean pain rating for oncologic patients was 3.1. The reported pain level varied significantly among facilities, as did physician mix. Multivariate analysis revealed that program and an interaction term between program and physician type, but not physician type independently, explained a significant amount of variation in pain level. Overall, reported pain remained higher than optimal. Research elsewhere has shown that application of the World Health Organization (WHO) cancer pain management guidelines can control 70%-90% of cancer pain. Strategies for implementing pain guidelines that emphasize a systems approach may be effective.