Treatment Decisions in Patients with Potentially Non-Survivable Burn Injury in Australia and New Zealand: A Registry-based Study.

Whilst burn-related mortality is rare in high-income countries, there are unique features related to prognostication that make examination of decision-making practices important to explore. Compared to other kinds of trauma, burn patients (even those with non-survivable injuries) may be relatively stable after injury initially. Complications or patient comorbidity may make it clear later in the clinical trajectory that ongoing treatment is futile. Burn care clinicians are therefore required to make decisions regarding the withholding or withdrawal of treatment in patients with potentially non-survivable burn injury. There is yet to be a comprehensive investigation of treatment decision practices following burn injury in Australia and New Zealand. Data for patients admitted to specialist burn services between July 2009 and June 2020 were obtained from the Burns Registry of Australia and New Zealand. Patients were grouped according to treatment decision: palliative management, active treatment withdrawn, and active treatment until death. Predictors of treatment initiation and withholding or withdrawing treatment within 24 hours were assessed using multilevel mixed-effects logistic regression. Descriptive comparisons between treatment groups were made. Of the 32,186 patients meeting study inclusion criteria, 327 (1.0%) died prior to discharge. Fifty-six patients were treated initially with palliative intent and 227 patients had active treatment initiated and later withdrawn. Increasing age and burn size reduced the odds of having active treatment initiated. We demonstrate differences in demographic and injury severity characteristics as well as end of life decision-making timing between different treatment pathways pursued for patients who die in-hospital. Our next step into the decision-making process is to gain a greater understanding of the clinician's perspective (e.g., through surveys and/or interviews).