The identification of nature and natural history of alcohol related frequent attenders to hospital

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

In the UK, alcohol use costs the NHS £3.5 billion annually, with 80% spent on hospital based care. Alcohol related hospital admissions have more than doubled since 2003/04 to over a million admissions in England in 2015/16. A significant proportion of people who attend hospital for an alcohol related reason attend many times during a year and are known as "alcohol related frequent attenders" (ARFAs). With no universal way of identifying ARFA hospital admissions the true burden on the NHS is unknown including use of A&E services, costs of inpatient admissions and longXterm health and social care. In addition, little is known about predictors of alcoholXrelated frequent attending. My thesis presents the findings of studies investigating the nature and natural history of alcoholXrelated frequent attenders to hospital (ARFAs). ARFAs are defined as patients with a whollyXattributable alcoholXrelated diagnosis, and multiple hospital admissions during a 1 year period. The thesis includes a systematic review of the literature, presenting what is already known about the nature and characteristics of alcoholXrelated frequent attenders to hospital; along with new studies comprising a crossXsectional analysis of the characteristics of ARFAs, including inXdepth analysis of coXmorbid diagnoses; national longitudinal study of health service use by ARFAs and their outcomes; a longitudinal national study looking at the predictors of transition to ARFA; and estimates of the costs associated with ARFAs nationally to the NHS. All of the studies include samples of patients drawn from national Hospital Episodes Statistics (HES) 2011/12X2015/16. From pseudonymised inpatient hospital records, data on natural history of ARFAs including: incidence, comorbidities (ICD 10 code), mode of admission, length of stay, readmissions, age, gender and geography are compared to 3 other groups of patients from 2011/12X 2015/16 national HES: nonXalcoholXrelatedXfrequent attenders, nonX alcoholXnonXfrequent attenders and alcoholXrelatedXnonXfrequent attenders. Longitudinal analysis of the 2011/12 ARFA cohort followed through to 2015/16, 3 yields information on service use patterns once someone is identified as being an ARFA, existing and developing coXmorbidities, and outcomes such as readmissions and health status. In contrast, longitudinal analysis of a cohort of new ARFAs (no previous history of being an ARFA) yields information on service use patterns and other predictors of outcomes. Using findings from the longitudinal analyses, costs of health service usage by ARFAs is modelled using national tariffs and occupied beddays. These novel studies measuring the scale of alcoholXrelated frequent hospital admissions in England over a 5Xyear period and associated burden identify a typical ARFA as being male, aged approximately 55 years, living in a more deprived area, with multiple physical and mental health comorbidities including a chronic alcohol diagnosis. ARFAs are more complex than other patient groups and alcohol is a key factor in causing inpatient admissions (often the primary diagnosis) rather than being an incidental finding. ARFAs’ medical history is chronic in nature and 10% of ARFAs are frequently admitted to hospital every year for 5 years. ARFAs have poor health outcomes: with higher prevalence rates of alcohol related liver disease compared to other patient groups. ARFAs have a lower probability of survival at 5 years than other patient groups. Nationally, ARFAs are significant enough in number to place a substantial burden on the NHS: an average ARFA occupies 10 extra bed days per year compared to a nonXalcohol frequent attender and this is because ARFAs have longer lengths of stay compared to other groups.
Date of Award1 Jun 2019
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorColin Drummond (Supervisor) & Michael Lynskey (Supervisor)

Cite this

'