Abstract
OBJECTIVE: Though most patients wish to discuss end-of-life (EOL) issues, doctors are reluctant to conduct end-of-life conversations. Little is known about the barriers doctors face in conducting effective EOL conversations with diverse patients. This mixed methods study was undertaken to empirically identify barriers faced by doctors (if any) in conducting effective EOL conversations with diverse patients and to determine if the doctors' age, gender, ethnicity and medical sub-specialty influenced the barriers reported. DESIGN: Mixed-methods study of multi-specialty doctors caring for diverse, seriously ill patients in two large academic medical centers at the end of the training; data were collected from 2010 to 2012. OUTCOMES: Doctor-reported barriers to EOL conversations with diverse patients. RESULTS: 1040 of 1234 potential subjects (84.3%) participated. 29 participants were designated as the development cohort for coding and grounded theory analyses to identify primary barriers. The codes were validated by analyses of responses from 50 randomly drawn subjects from the validation cohort (n= 996 doctors). Qualitative responses from the validation cohort were coded and analyzed using quantitative methods. Only 0.01% doctors reported no barriers to conducting EOL conversations with patients. 99.99% doctors reported barriers with 85.7% finding it very challenging to conduct EOL conversations with all patients and especially so with patients whose ethnicity was different than their own. Asian-American doctors reported the most struggles (91.3%), followed by African Americans (85.3%), Caucasians (83.5%) and Hispanic Americans (79.3%) in conducting EOL conversations with their patients. The biggest doctor-reported barriers to effective EOL conversations are (i) language and medical interpretation issues, (ii) patient/family religio-spiritual beliefs about death and dying, (iii) doctors' ignorance of patients' cultural beliefs, values and practices, (iv) patient/family's cultural differences in truth handling and decision making, (v) patients' limited health literacy and (vi) patients' mistrust of doctors and the health care system. The doctors' ethnicity (Chi-Square = 12.77, DF = 4, p = 0.0125) and medical subspecialty (Chi-Square = 19.33, DF = 10, p =0.036) influenced their reported barriers. Friedman's test used to examine participants relative ranking of the barriers across sub-groups identified significant differences by age group (F statistic = 303.5, DF = 5, p < 0.0001) and medical sub-specialty (F statistic = 163.7, DF = 5, p < 0.0001). CONCLUSIONS AND RELEVANCE: Doctors report struggles with conducting effective EOL conversations with all patients and especially with those whose ethnicity is different from their own. It is vital to identify strategies to mitigate barriers doctors encounter in conducting effective EOL conversations with seriously ill patients and their families.