TY - JOUR T1 - Preferred place of death in paediatric, teenage and young adult haemato-oncology patients: a retrospective review JF - BMJ Supportive & Palliative Care JO - BMJ Support Palliat Care DO - 10.1136/bmjspcare-2019-002005 SP - bmjspcare-2019-002005 AU - Philippa Stilwell AU - Ankit Bhatt AU - Keval Mehta AU - Ben Carter AU - Maggie Bisset AU - Louise Soanes AU - Ananth Shankar Y1 - 2020/03/24 UR - http://spcare.bmj.com/content/early/2020/04/21/bmjspcare-2019-002005.abstract N2 - Objectives Identifying the preferred place of death for children/young people with cancer and determining whether this is achieved is pertinent to inform palliative care service provision. The aims of this retrospective case series review were to determine where children/young people with cancer want to die and whether their preferred place of death was achieved.Methods Clinical/demographic details, including preferred and actual places of death, were recorded for 121 patients who died between 2012 and 2016 at a tertiary haematology–oncology centre. A logistic regression model was used to determine the odds of achieving the preferred place of death in patient subgroups.Results 74 (61%) patients had a documented discussion regarding place of death preference. Where a preferred location was identified, 72% achieved it. All patients who wanted to die in the hospital (n=17) or a hospice (n=9) did, but only 58% of patients who wanted to die at home (n=40) achieved this. Of the 42% (n=17) who wanted to die at home but did not, 59% of these were due to rapid deterioration in clinical status shortly after the discussion. Having supportive treatment in the last month of life was associated with increased odds of achieving the preferred place of death versus those who were undergoing chemotherapy/radiotherapy (OR 3.19, 95% CI 1.04 to 9.80, p value=0.04).Conclusion Where hospice/hospital was chosen as the preferred place of death, this was always achieved. Achieving home as the preferred place of death was more challenging and frequently prevented by rapid clinical deterioration. Clinicians should be encouraged to address end-of-life preferences at an early stage, with information provided adequately. Further research should explore implications of these findings on both end-of-life experience and overall service provision. ER -