Aims. To introduce, to assess the use of and the staff understanding o
f a do not resuscitate policy. Methods. A policy was developed with le
gal, medical, nursing and ethical input. Two distinct methods for a do
not resuscitate order were allowed. The first method is where the pat
ient requests a do not resuscitate order. The second is the medically
indicated do not resuscitate order. The policy was trialled on two flo
ors at Dunedin Hospital and an audit performed. Information was collec
ted on patients receiving do not resuscitate orders. The notes of all
patients dying on these floors were reviewed. A questionnaire was sent
to clinical staff working on the trial floors which included question
s on their understanding of issues related to do not resuscitate order
s. Results. 86% of deaths had a do not resuscitate order, 26% in accor
dance with the formal policy. Thirty percent of the time there was no
record of a discussion of the do not resuscitate order with the patien
t or their family. Problems identified by staff included difficulty wi
th raising and discussing these orders with patients and their familie
s. Staff misunderstanding of consent and responsibility for treatment
decisions for mentally competent and incompetent patients was common.
Conclusions. There is a need far further staff education and discussio
n of the issues surrounding do not resuscitate orders. Discussion with
patients and their families could be improved.