Introduction Liver disease is the 3^rdcommonest cause of premature death in the UK. The end of life care strategy (Department of Health, 2008) noted many patients do not die in a place of their choosing, and that difficulties exist amongst physicians in identifying the dying process. Data from the Royal Free demonstrated that only 19% of patients assessed unsuitable for liver transplantation were referred to palliative care services, a median of 4 days before death. Across the United Kingdom between 2007–2011 81% of patients with alcoholic liver disease died in hospital, compared to 48% with cancer (Office of National Statistics). Opportunities for advanced care planning in this population are missed.

We aimed to design and validate a tool to identify inpatients with liver disease who stand to benefit from palliative care assessment and advanced care planning, and to create a model of care for such patients.

Method The department of health document ‘Getting it Right: improving end of life care for people living with liver disease’ and the NHS North East document ‘Framework for supportive care in advanced liver disease’ identify evidence based factors which are predictive of death in liver disease. Five of these factors (Childs Pugh C, >2 admissions within last 6 months, continued use of alcohol, unsuitable for liver transplantation, pre-admission WHO performance status >2) were assimilated into a screening tool.

Results The tool was retrospectively applied to all patients admitted to the Bristol Royal Infirmary with a diagnosis of cirrhosis over 90 consecutive days from 1^stJuly 2013 (n = 47). Based on the ‘Gold Standard Care’ framework question (“would you be surprised if your patient was not alive in one year”) mortality one year post admission was calculated. Sensitivity and specificity for predicting one year mortality when 2, 3 or 4 poor prognostic criteria were positive were analysed (see Table 1). On this basis, an admission score of 3 or more criteria was considered a “positive” poor prognosis screen.

Conclusion The tool has been successfully trialled and audited locally over the past year. Patients who screen positive are highlighted for discussion at a weekly hepatology MDT. Assuming MDT agreement this triggers a consultant led poor prognosis discussion with the patient, a poor prognosis letter to the GP, and involvement of the palliative medicine team to assist with symptom control and advance care planning. Communication skills training has been delivered to consultants and junior staff by the palliative medicine team.

Disclosure of interest None Declared.