Evaluating Variability in Outcomes after Stroke
Among Hospitals

About one in four stroke patients have “Do-Not-Resuscitate” orders written during hospitalization.5 Stroke patients with DNR orders account for over 80% of deaths by 30 days. These patients have significantly higher mortality rates than anticipated based upon their initial severity of illness. It is possible that the increased mortality seen in patients with DNR orders is due to incomplete adjustment for severity of illness. Alternatively, the care of patients made DNR may be suboptimal or centered around less aggressive interventions, both of which may be inappropriate. The paucity of understanding of DNR utilization and care of DNR patients with stroke is underscored by the few published studies on DNR use in stroke. In a cohort study of 140 stroke patients made DNR at one academic hospital, 53% continued to receive normal nutrition. Despite the fact that almost half of these patients were not given nourishment, the authors interpreted their results as reassuring evidence that DNR orders did not mean failure to provide care. In contrast, an accompanying editorial suggested that the percentage of patients receiving nutrition was too high, stating “the real question is why 53% of stroke patients received nutrition, despite having DNR orders”. These disparate conclusions make clear the need for a better physician understanding of appropriate care of strokes patients with DNR orders.

In addition, the hospital rate of use of Do-Not-Resuscitate orders has been recently correlated with higher in-hospital mortality rates from intracranial hemorrahge, a finding which extends even to hemorrhage patients that do NOT have DNR orders. The frequency of DNR use by hospital has been suggested to be a marker for some aspect of overall care by that hospital. We are evaluating the relationship between hospital use of DNR orders within one day of admission and in-hospital mortality in patients hospitalized for stroke in Northeastern Ohio.