• Kirsten Wentlandt, Philippe Toupin, Natalia Novosedlik, Lisa W Le, Camilla Zimmermann, and Ebru Kaya.
• Department of Supportive Care, University Health Network, Toronto, Ontario, Canada; Department of Family and Community Medicine, University of Toronto, Toronto, Ontario, Canada. Electronic address: Kirsten.Wentlandt@uhn.ca.
• J Pain Symptom Manage. 2018 Sep 1; 56 (3): 337-343.

ContextClinicians often rely on documentation to relay information, and this remains the mainstay of interprofessional communication regarding patient care. However, there has been scant research focused on clinicians' documentation of dying in hospital and how this is communicated to other team members in patient charting.ObjectivesTo understand the language used to describe the deterioration and death of patients in an acute academic tertiary care center and to identify whether patient diagnoses or palliative care (PC) involvement was associated with clearer descriptions of this process.MethodsWe conducted a retrospective chart review of the final admission of 150 patients who died on an inpatient internal medicine unit. Conventional and summative content analysis was performed of the language used to describe, either directly or indirectly, that the patient's death was imminent.ResultsOf the 150 deaths, the median age was 79.5 (range 22-101), 58% were males, and 69% spoke English. A total of 45% of deaths were from cancer, and 66% occurred with prior PC team involvement. There was no documentation of the dying process in 18 (12%) charts. In the remainder, clinicians' documentation of imminent death fell into three categories: 1) identification of the current state using specific labels; for example, dying (24.7%) or end of life (15.3%), or less specific language, unwell or doing poorly (6.0%); 2) predicting the future state using specific or more vague predictions; for example, hours to days (7.3%) or poor or guarded prognosis (26.0%); and 3) using care provided to the patient to imply patient status; for example, PC (49.3%) or comfort care (28.7%). PC involvement, but not a malignant diagnosis, was associated with more frequent use of specific language to describe the current state (P = 0.004) or future state (P = 0.02).ConclusionDeath and dying in hospital is inadequately documented and often described using unclear and vague language. PC involvement is associated with clearer language to describe this process.Copyright © 2018 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

25 keywords...

hide…