Advance Care Planning in hospitalised frail older patients: a missed opportunity? A systematic review and narrative synthesis

06 Jun 2018

Background: Frail older people have multiple hospital admissions in the last year of life and account for 40% of UK deaths. Advance care planning (ACP) has been shown to improve outcomes for patients and families. Most frail older people have not participated in ACP but would like to do so. We sought to understand whether an acute hospital admission represents an opportunity for frail older patients to engage in and benefit from ACP through a systematic review and narrative synthesis. Method: Electronic literature search of MEDLINE, CINAHL, ASSIA, PsycINFO, and Embase databases from 1 January 1990 to 1 September 2017. Population: Patients aged 60 years or older with multiple co-morbidities or frailty or need for help with personal care. Setting: Acute hospital wards in any healthcare system. Review questions: Does ACP occur and is it of benefit? What are the views of patients and clinicians? What are the barriers and facilitators? Results: The search returned 8076 unique articles, 16 of which met inclusion criteria. One further study was identified from reference searching. Rates of ACP with hospitalised frail older patients are very low (0-3%) but most patients would welcome the opportunity to participate in ACP, which was found to improve outcomes, including respecting patient wishes and reduced relatives’ depression after death. Patients and families perceive the most important facilitator is access to doctors willing to discuss ACP, whereas clinicians believe the biggest barrier is patient and family members’ difficultly at accepting a poor prognosis. ACP uptake is increased by trained facilitators and patient ACP information packs. Conclusions: Hospital admission is a missed trigger opportunity for ACP with frail older patients, with a wide gap between evidence and clinical practice. Review findings are limited by variable study quality and scarcity of research in some areas.