Introduction: Alcohol-related liver disease (ARLD) is a preventable cause of mortality. Historical epidemiological studies on ARLD often lack a detailed linked assessment of health-related contacts prior to death which limits understanding of opportunities for intervention. We aimed to analyse retrospective population-based data of all adult residents of Nottinghamshire dying from ARLD to determine the factors associated with delayed diagnosis of ARLD and the potential missed opportunities for interventions.

Methods: We linked the Office for National Statistics and Hospital Episode Statistics databases to identify adult (≥18 years) residents of Nottinghamshire, who died of ARLD over the 5-year period (1 January 2012 to 31 December 2017). Death was used as the primary outcome, and logistic regression analysis was conducted to test the association between key variables and mortality due to ARLD.

Results: Over 5 years, 799 ARLD deaths were identified. More than half had no diagnosis or a diagnosis of ARLD less than 6 months before death. Emergency presentation at first ARLD diagnosis and White ethnicity were significantly associated with a delay in diagnosis. Overall, the cohort had a median of five hospital admissions, four accident and emergency attendances and 16 outpatient appointments in the 5 years before death. Treatment was provided by a range of specialities, with general medicine the most common. Alcohol was associated with most admissions.

Discussion and conclusions: This study identified deficiencies in ARLD secondary care and provides us with a powerful methodology that can be used to evaluate and improve how alcohol issues are managed and where action can be best targeted.