Article Text


PTH-062 Emergency Admissions For Alcohol Related Conditions: Making Sense Of Routine Data
  1. P Lekharaju1,
  2. E Thompson2,
  3. M Shawihdi3,
  4. M Pearson2,
  5. S Hood1,
  6. K Bodger3
  1. 1Dept of Gastroenterology, Aintree University Hospital, University of Liverpool, Liverpool, UK
  2. 2Aintree Health Outcome Partnership, University of Liverpool, Liverpool, UK
  3. 3Department of Gastroenterology, Institute of Translational Medicine, University of Liverpool, Liverpool, UK


Introduction Alcohol-related emergency admissions (ARA) are a major and rising hospital burden, resulting from conditions that range from short-term toxicity to end-stage organ damage, notably liver disease. We report a project to develop analyses of routine coding data, with a particular focus on metrics related to ‘frequent flyers’ (FFs) as targets for new service interventions.

Methods Sources of data: Hospital Episode Statistics for all English hospitals (2006–2008); Inpatient coding data and AED attendances for our Trust (2006–2013).

Analysis: Screening of all non-emergency episodes for alcohol-related codes at any position to flag ARAs; linkage of individual cases to extract all admissions and order chronologically; allocation of each admission to a category based on primary and lower order diagnoses (flagging LIVER and NON-LIVER admissions); identification of frequent flyers (FFs) based on various definitions of admission count; linkage of ARAs to all-cause AED attendances in local data; funnel-plot analyses of patterns across English Trusts; longitudinal trends in local data.

Results Nationally: 219,158 ARAs in 139,077 patients (2006–2008), mean age (sd): 49.5 yrs (16); males: 99,271 (71%); Deprivation : Quintile 1 (most deprived) 36.4%, Quintile 5 (least): 9.4%; Co-morbidity (Charlson), 0.44 (0.68); LoS: 7 (14) days; Inpatient death: 6,656 (4.8%). No. admissions ranged from 1–60 per individual. Frequent Flyers: In two years, a cut-off of 5+ admissions identified 5,404 FFs (4% of patients; accounting for 18% of ARAs nationally) whereas 10+ identified 909 FFs (1% of patients; 6% of ARAs). Mean ARAs per Trust was 927 (range: 235–3930) with 6-fold variation in% of FFs (1.3% to 7.7%) and 4-fold variation in% with liver disease across English hospitals (range: 7.6–30.2%). As expected, FFs coded with liver disease had significantly higher LoS and mortality risk consistent with end-stage organ damage and “unavoidable” admissions.

Locally: 21,308 ARAs in 16,305 patients (2006–2012), with annual number of cases rising from 1,615 to 4,603. Defining FFs as 5+ ARAs per year, there were 320 FFs (2% of patients; 10% of ARAs). There was a year-on-year rise in ARAs (2,454–5,510) and AED attendances without admission (2,499–5,979). However, FFs (5+ admissions) declined from 64 to 47 between 2006 and 2012 and non-liver FFs from 25 to 12, suggesting a positive impact of new local services on multiple attenders, especially those lacking established liver disease.

Conclusion We have developed a set of candidate metrics focussed on FFs, short stays and selected baskets of conditions to provide data to support front-line acute alcohol services.

Disclosure of Interest None Declared.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.