End-of-life care at an academic medical center: are attending physicians, house staff, nurses, and bereaved family members equally satisfied? Implications for palliative care.

BACKGROUND: End-of-life care is deemed to be poor in the United States - particularly in large teaching hospitals. Via a brief survey, we examined satisfaction with end-of-life care for those patients who died in our academic medical center from provider and family perspectives. METHODS: To assess the correlation between overall satisfaction between providers (attending, housestaff, and nurses) as well as family members for decedents who died in our hospital, we conducted a satisfaction survey regarding care in the last three days of life. The nine item survey was administered within 1 week of the patient s death to care providers and approximately 8 to 12 weeks to next of kin. RESULTS: There were 166 deaths examined over the four month study period. Overall satisfaction with care was 3.02 out of 4.0, and differed by respondent group (p= 0.035). Correlation between respondents was very low (range 0.02 to 0.51). The least discordance was between residents and interns (0.5), who had the lowest level of satisfaction (2.72). Housestaff and attendings had the lowest overall correlation in mean satisfaction scores (0.05). Most providers knew their patients for 24 hours or less. CONCLUSIONS: Overall satisfaction was high, but there was discordance among different providers. Continuity of care was limited. Age and location of death alone did not significantly affect satisfaction with end-of-life care. Implications of this type of research for improving end of life care at academic centers are discussed.

Duke Scholars

Author Anthony Nicholas Galanos Medicine, Geriatrics

Author Carl F. Pieper Biostatistics & Bioinformatics, Division of Biostatistics

Author Karen E. Steinhauser Population Health Sciences