Knowledge and attitudes toward end-of-life care in veterans with symptomatic metastatic cancer.

<bold>Objectives: </bold>The purposes of this study were to study symptomatic metastatic cancer patients' knowledge and attitudes toward end-of-life (EOL) care and to examine how patient-perceived health status affects attitudes toward EOL care and survival.<bold>Methods: </bold>From 1999 to 2002, 254 symptomatic metastatic cancer patients at the VA New Jersey Health Care System completed the Vermont Voices on Care of the Dying Questionnaire. Survival status and location of death were obtained. Descriptive statistics and the chi square method were used to assess the differences between African Americans (N=109) and Caucasians (N=135), and between different patient-perceived health status groups. A log-rank test was performed to assess for differences in median survival length between different patient-perceived health-status groups.<bold>Results: </bold>Veterans' responses to the Vermont questionnaire showed knowledge deficits regarding EOL care. There was wide variation in self-rankings of health status: 45.6% of patients rated their illness as serious and life threatening, 18.9% considered their health problem significant but not life threatening, 2.8% thought they were in good health, and one-third of patients were unsure about their health status. Most patients (86.2%) preferred physician frankness when communicating bad news and 61.8% preferred family involvement in EOL discussions. African American patients were less likely to have completed advance directives (p < 0.0001), to have knowledge about hospice programs (p < 0.00001), and to feel capable of assessing their health situation (p = 0.04). Patient-rated health status affected completion rates of advance directives and survival.<bold>Significance Of the Research: </bold>These findings demonstrate knowledge deficits and racial differences in attitudes and values toward EOL care in veterans with cancer. The Vermont questionnaire enables patients to state their EOL preferences but may not be detailed enough for clinical applications. Patient-rated health status may be an important explanatory variable for EOL preferences and length of survival. [ABSTRACT FROM AUTHOR]