Diagnosing delirium in very elderly intensive care patients.

OBJECTIVE: To determine the incidence of delirium in elderly intensive care patients and to compare incidence using two retrospective chart-based diagnostic methods and a hospital reporting measure (ICD-10). DESIGN: Retrospective study. SETTING: An ICU in a large metropolitan private hospital in Melbourne, Australia. PATIENTS: English-speaking participants (n=348) 80+ years, admitted to ICU for >24 hours. MEASUREMENTS AND MAIN RESULTS: Medical files of ICU patients admitted October 2009-October 2012 were retrospectively assessed for delirium using the Inouye chart review method, DSM-IV diagnostic criteria and ICD-10 coding data. General patient characteristics, first onset of delirium symptoms, source of delirium information, administration of delirium medication, hospital and ICU length of stay, 90 day mortality were documented. Delirium was found in 11-29% of patients, the highest incidence identified by chart review. Patients diagnosed with delirium had higher 90 day mortality, and those meeting criteria for all three methods had longer hospital and ICU length of stay. CONCLUSIONS: ICU delirium in the elderly is often under-reported and strategies are needed to improve staff education and diagnosis.

Doctors' attitudes regarding not for resuscitation orders.

Objectives The aims of the present study were to investigate doctors' attitudes regarding the discussion and writing of not for resuscitation (NFR) orders and to identify potential barriers to the completion of these orders. Methods A questionnaire-based convenience study was undertaken at a tertiary hospital. Likert scales and open-ended questions were directed to issues surrounding the discussion, timing, understanding and writing of NFR orders, including legal and personal considerations. Results Doctors thought the presence of an NFR order both should and does alter care delivered by nursing staff, particularly delivery of pain relief, nursing observations and contacting the medical emergency team. Eighty-five per cent of doctors believed they needed somebody else's consent to write an NFR order (seeking of consent is not a requirement in most Australian jurisdictions). Conclusion There are complex barriers to the writing and implementation of NFR orders, including doctors' knowledge around the need for consent when cardiopulmonary resuscitation is likely to be futile or excessively burdensome. Doctors also believed that NFR orders result in changes to goals-of-care, suggesting a confounding of NFR orders with palliative care. Furthermore, doctors are willing to write NFR orders where there is clear medical indication and the patient is imminently dying, but are otherwise reliant on patients and family to initiate discussion. What is known about the topic? Hospitalised elderly patients, in the absence of an NFR order, are known to have poor survival and outcomes following resuscitation. Further, Australian data on the prevalence of NFR forms show that only a minority of older in-patients have a written NFR order in their history. In Australian hospitals, NFR orders are completed by doctors. What does this paper add? To our knowledge, the present study is the first in Australia to qualitatively analyse doctors' reasons to writing NFR orders. The open-text nature of this questioning has been important in eliciting doctors' responses without hypothesis guessing bias. Further, we add to the literature on the breadth of considerations doctors may encounter with regard to NFR orders. What are the implications for practitioners? The findings indicate the issues impeding decision making around cardiopulmonary resuscitation relate to poor knowledge of the law, particularly around the issue of consent and confounding NFR orders with provision of palliative care. Such barriers to the completion of NFR orders expose elderly in-patients to futile and burdensome resuscitation events. The findings suggest consideration be given to education and training materials to inform doctors about jurisdictional law regarding resuscitation documentation, support decision making around cardiopulmonary resuscitation and promote goals-of-care discussions on admission.