Use this URL to cite or link to this record in EThOS:
Title: Risk factors for hospital admission in an ageing English cohort : an analysis linking prospectively collected data with Hospital Episode Statistics
Author: Simmonds, Shirley Jill
ISNI:       0000 0004 7656 2755
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The UK has an ageing population in which overall life expectancy is growing faster than healthy life expectancy. The result is that an increasing number of older people are in poor health; these people contribute to rising trends in hospital admission and are commonly associated with the financial crisis in the NHS. Reducing demand on hospitals is enshrined in Government policy, but requires improved understanding of the determinants of service use; these may differ according to the type of admission under consideration. Only when the determinants of admission are clear can effective preventive measures be put in place. This thesis describes the creation of a novel resource in which to investigate risk factors for hospital admission. This is achieved by combining data from two sources. The first is the Hertfordshire Cohort Study; a group of community-dwelling men and women whose physical, mental and social health were comprehensively characterised in 1998-2004 when they were aged 59-73 years. To these baseline data, routinely collected information about hospital admissions that participants experienced during the following decade has been added. The resulting resource has several advantages over hospital data alone: it includes variables that were prospectively collected; that go beyond those that are clinically relevant; and that cover non-admitted as well as admitted individuals. Three papers based on the combined dataset are presented. Each takes the same panel of 25 predictor variables - chosen to summarise baseline demography and anthropometry, lifestyle, social circumstances, physical function and morbidity - and compares it with one of three admission outcomes: 30-day readmission, emergency admission or elective admission. Across both sexes, the papers identify six risk factors for 30-day readmission, five for emergency admission and four for elective admission. Two markers of overall burden of disease: poor self-rated health (SRH) and increased number of body systems medicated (NSM); conferred risk for all three outcomes. Poor physical function and history of smoking were associated with risk of two outcomes each. SRH was associated with emergency admission in both sexes and NSM with elective admission in both sexes. The likelihood of each type of admission rose with the number of risk factors an individual had that were specific to it. The primary implication is that older people at highest risk of admission of any sort could be identified prospectively in a primary care setting by screening for poor SRH and a high number of systems medicated. This is suggested as an area for focus in revising the Quality and Outcomes Framework. Meanwhile, Public Health departments should seek to decrease risk profiles in younger generations and thus to increase healthy life expectancy. An important additional point relates to the linkage methodology. This PhD has shown that linked data have more value than either cohort or routinely collected data alone; this methodological contribution is arguably as important as the substantive findings. However, the work has been threatened from the outset by changing regulations on data sharing - a problem that urgently needs to be addressed. The Digital Economy Act that is currently progressing through Parliament presents an opportunity for Government to remedy the access problems that beset research of this nature.
Supervisor: Evandrou, Maria Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available