ObjectiveTo assess the influence of private attending physician status on the withdrawal of life-sustaining interventions among patients dying within a medical intensive care unit (ICU).DesignRetrospective cohort analysis.SettingAn academic tertiary care center.PatientsOne hundred fifty-nine consecutive patient deaths occurring in the medical ICU during a 12-month period.InterventionsNone.Measurements and Main ResultsWithdrawal of life-sustaining interventions (i.e., mechanical ventilation, dialysis, and/or vasopressors), duration of mechanical ventilation, length of intensive care unit stay, medical care costs, and patient charges were recorded.Life-sustaining interventions were actively withdrawn from 69 (43.4%) patients prior to death. Patients without a private attending physician were significantly more likely to undergo the withdrawal of life-sustaining interventions compared with patients having a private attending physician (odds ratio equals 2.5; 95% confidence interval equals 1.8, 3.6, respectively; p equals .005). A correlation was found between the possession of private health insurance and private attending physician status (r2equals .39, p less than .001). Multiple logistic regression analysis was subsequently used to control for demographic factors and severity of illness. Three independent predictors for the withdrawal of life-sustaining interventions were identified in this patient cohort (p less than .05): having a planned therapeutic trial of life-sustaining interventions outlined in the medical record (adjusted odds ratio equals 9.4; 95% confidence interval equals 5.6 to 15.6; p less than .001); lack of a private attending physician (adjusted odds ratio equals 4.4; 95% confidence interval equals 2.9 to 6.5; p less than .001); and the presence of clearly defined advance directives regarding patient preferences for medical care (adjusted odds ratio equals 3.6; 95% confidence interval equals 2.3 to 5.7; p equals .005). Patients with private attending physicians had significantly greater medical care costs and medical care charges compared with patients without a private attending physician.ConclusionAmong patients dying within a medical ICU, those patients without a private attending physician are more likely to undergo the active withdrawal of life-sustaining interventions.(Crit Care Med 1996; 24:968-975)