Do-not-resuscitate practices in the chronically critically ill

To determine the frequency of do-not-resuscitate (DNR) orders in the chronically critically ill; to identify the differences in clinical and demographic characteristics of chronically critically ill patients who have DNR orders and those who do not; to identify the differences in the cost of care between patients with and without DNR orders; and to identify the differences in DNR practices between an experimental special care unit and the traditional intensive care unit (ICU).

Randomized, prospective design with a block randomization scheme.

Two hundred twenty patients who met the following eligibility criteria for enrollment in a parent study of the special care unit: an ICU stay of at least 5 days, an absence of pulmonary artery monitoring, an absence of frequent titration of intravenous vasopressors, an Acute Physiology and Chronic Health Evaluation II score of less than 18, and a Therapeutic Intervention Scoring System score of less than 39.

A large, urban academic medical center.

Clinical and demographic variables describing the study populations, mental status, and timing of DNR orders, mortality rates, and cost of hospitalization.

There was no difference in the frequency of DNR orders between the special care unit versus the intensive care unit--although patients in the special care unit had a longer interval between hospital admission and initiation of the DNR order. DNR patients differed from non-DNR in that they were older, less likely to be married, and had a higher Acute Physiology and Chronic Health Evaluation II score on admission to the study. The mortality rate in the DNR group was 71% versus 6% in the non-DNR group. There was no difference in total costs. DNR patients were also more likely to have an impaired mental status on admission, and more likely to have deterioration in mental status by the time of discharge than the non-DNR patients.