Objective: To describe the demographic profile of a cohort of elderly patients with a 'do-not-resuscitate' (DNR) order at death and to study the specific supportive measures instituted or withdrawn during the DNR period and those in force at the time of death.
Methods: The case notes of patients who died between October 1996 and March 1997 in the Department of Geriatrics, Alexandra Hospital were studied retrospectively by a single observer.
Results: Only 95 out of an eligible 102 patients' case notes could be retrieved. Seventy-two (75.8%) patients had a DNR status at death. The racial distribution was as follows: 90.3% Chinese, 5.6% Indians, 2.8% Malays and 1.4% Others while their pre-admission domicile were: own home 79.2%, nursing home 19.4% others 1.4%. Those bedbound constituted 48.6% of the cohort while 29.2% had dementia and 43.1% were totally dependent for their activities of daily living. The commonest cause of death was pneumonia while the average duration patients were on the DNR status was 5.1 days before death. The commonest measures instituted during DNR period were as follows: oxygen therapy (38.9%), nasogastric tube insertion and feeding (30.6% and 33.3% respectively), intravenous fluid administration (33.3%), blood investigations (33.3%), opioid use (33.3%) and antibiotic use (29.2%). Measures withdrawn were intravenous fluid administration (36.1%), hourly monitoring of parameters (22.2%), antibiotics (13.9%), high dependency care (12.5%) and nasogastric tube feeding (6.9%).
Conclusion: The DNR status is decided late in the course of a patient's illness when he may have been too ill to partake in the decision making process. Even if a DNR status was ordered, a patient might still be subjected to CPR at death.