Preferences for resuscitation and intubation among patients with do-not-resuscitate/do-not-intubate orders

To determine the accuracy of do-not-resuscitate/do-not-intubate (DNR/DNI) orders in representing patient preferences regarding cardiopulmonary resuscitation (CPR) and intubation.

We conducted a prospective survey study of patients with documented DNR/DNI code status at an urban academic tertiary care center that serves approximately 250,000 patients per year. From October 1, 2010, to October 1, 2011, research staff enrolled a convenience sample of patients from the inpatient medical service, providing them with a series of emergency scenarios for which they related their treatment preference. We used the Kendall τ rank correlation coefficient to examine correlation between degree of illness reversibility and willingness to be resuscitated. Using bivariate statistical analysis and multivariate logistic regression analysis, we examined predictors of discrepancies between code status and patient preferences. Our main outcome measure was the percentage of patients with DNR/DNI orders wanting CPR and/or intubation in each scenario. We hypothesized that patients with DNR/DNI orders would frequently want CPR and/or intubation.

We enrolled 100 patients (mean ± SD age, 78 ± 13.7 years). A total of 58% (95% CI, 48%-67%) wanted intubation for angioedema, 28% (95% CI, 20%-3.07%) wanted intubation for severe pneumonia, and 20% (95% CI, 13%-29%) wanted a trial resuscitation for cardiac arrest. The desire for intubation decreased as potential reversibility of the acute disease process decreased (Kendall τ correlation coefficient, 0.45; P<.0002).

Most patients with DNR/DNI orders want CPR and/or intubation in hypothetical clinical scenarios, directly conflicting with their documented DNR/DNI status. Further research is needed to better understand the discrepancy and limitations of DNR/DNI orders.