Re: Variations in hospice use among cancer patients.

Numerous studies have been done to analyze hospice use among terminally ill pa tients. The recent study by Keating et al. ( 1 ) analyzed variations in hospice use among cancer patients. I applaud their efforts but wish to highlight other factors that could have been considered to make the results more comprehensive. The authors reported that the rate of hospice enrollment varied by the race of the prin cipal physician. However, it is unclear whether they also analyzed the enrollment rates among patients who were treated by a physician of the same race as the patient, i.e., did the enrollment rates still differ when the patient and the physician shared a common ethnic background. The literature suggests that patients and physicians communicate more effectively when they belong to the same racial group ( 2 ), and this sharing of ethnic background could have an impact on patient decision-making process. Given the cultural and language barriers that exist in the US health care system, some patients may refuse to take medical assistance from a health care provider who does not share the same ethnic background as the patient. Also, some patients may not divulge vital information, such as drug allergies and medication adherence, to a health care provider of different ethnicity and thus may appear uncooperative during counseling sessions. It has been reported ( 3 , 4 ) that patients belonging to minority ethnic groups are not completely aware of the available hospice facilities because of the existing cultural and language barriers. Keating et al. did not report the initial level of awareness of the hospice eligibility criteria among the patients who were enrolled in their study. Patients with a previous knowledge of hospice eligibility criteria may show higher enrollment rates than patients with no previous knowledge of hospice enrollment criteria, thus creating a potential bias. Also, the authors did not clarify whether other health care providers, such as pharmacists and nurses, counseled the patients about hospice enrollment during the study period. Such counseling could have created a potential bias. Future studies that examine hospice use by cancer patients on different health coverage plans (including patients on federal plans such as Medicare and Medicaid) and that attempt to identify possible barriers to hospice enrollment in patients suffering from different types of cancers will be important.