Objective: To assess prevalence and quality of end-of-life communication between persons with advanced AIDS and their clinicians and to identify patient and clinician characteristics associated with this communication.
Design: Prospective cohort study of 57 patients with AIDS and their primary care clinicians.
Setting: University-based and private clinics in Seattle, Washington.
Patients: Patients had a prior AIDS-defining illness and a CD4 cell count of less than 100 x 10(6) cells/l.
Main outcome measures: Quality of patient-clinician communication about end-of-life care, validated against patient satisfaction and patient-clinician concordance on advance directives and treatment preferences.
Results: Patients reported they had communication about end-of-life care with their clinician in 31 of 57 cases (54%) while clinicians reported they had this discussion in 36 of 57 cases (64%). Patients and clinicians gave concordant answers in 42 patient-clinician pairs. In 15 pairs (26%), patients and clinicians disagreed about whether end-of-life communication had occurred. African-American and Hispanic patients were less likely to report having communication than non-Hispanic white patients (chi-square analysis: chi2 = 4.67; P < 0.05); injection drug users and women with high-risk sexual partners were less likely to report communication than homosexual or bisexual men (chi2 = 4.67; P < 0.05). A four-item measure of patients' assessment of the quality of communication about end-of-life care had good internal consistency (Cronbach's alpha 0.81) and was significantly correlated with overall satisfaction with medical care (r2 = 0.76; P < 0.0001). Patients with lower income reported lower quality of communication (chi2 = 5.82; P = 0.05). If patients assessed quality of communication as high, their clinicians were more likely to know if the patient had a durable power of attorney for health care (chi2 = 4.95; P = 0.03) but were not more likely to predict patients' preferences for life-sustaining treatments.
Conclusions: Quality of patient-clinician communication about end-of-life care can be measured in a brief questionnaire; higher quality of this communication is associated with higher satisfaction with care and increased clinician knowledge of patients' advance directives. Since socioeconomic status and ethnicity are associated with both the occurrence and quality of this communication, future interventions in end-of-life care should assess the effect of these variables. Given the important and independent goals of improving patient-clinician communication about end-of-life care and improving the quality of care at the end of life, future studies should test interventions to improve the quality of communication and determine whether improving this communication improves the quality of care at the end of life.