An epidemiological study of varying emergency medical admission rates in Glasgow

Blatchford, Oliver (1999) An epidemiological study of varying emergency medical admission rates in Glasgow. PhD thesis, University of Glasgow.

Abstract

Background: Emergency medical admissions in the United Kmgdom have
been rising for many years. This rise has resulted in increasing pressures
on hospitals' resources, with consequent difficulties in coping with peaks of
admissions. This rise has not been intended or planned. The epidemiology
of emergency medical admissions is poorly understood.
Aims: To investigate the epidemiology of emergency medical admissions in
Glasgow in terms of time, person and place. To explore variations in
Glasgow's general practices' and hospitals' emergency medical admission
rates.
Literature review: Articles relating to variations in emergency medical
admission rates were identified by searching bibliographic databases, cross
referencing from known articles, consulting other researchers and hand
searching of journal indexes. Relevant articles were included in a
systematic review of the epidemiology of varying rates of hospitals'
emergency admissions. Articles that postulated causes of the rise of
emergency admissions or factors associated with varying admission rates
were also reviewed. A summary of mechanisms whereby hospitals might
cope with pressures of emergency admissions concluded the literature
review.
Published evidence of variations of rates of hospital emergency admission
was limited. Most articles were found to contain postulated associations
with variations in hospitals' emergency admissions. While many published
mechanisms for hospitals to adapt to pressures from emergency admissions
were identified, only a minority of these had been formally evaluated.
3
Setting: Greater Glasgow Health Board residents (813,029 adults at June
1997). Data obtained from the Health Board's Community Health Index
(CHI).
Subjects: 537,798 Greater Glasgow Health Board residents admitted to
Glasgow hospitals7 medical beds between 1980 and 1997 (43,236 patients
in 1997). Data obtained from Scottish Morbidity Record database one
(SMRl).
Methods: Anonymised CHI and SMRl datasets linked by patients' general
practitioners' codes, using a computer database package. Standardised
emergency medical admission rates were calculated by the database.
Computerised maps of standardised emergency medical admission ratios
were plotted for Glasgow7s postcode sectors to show geographical
variations. Correlation and logistic regression were used to explore
variations in standardised emergency medical admission ratios.
Outcome measures: Crude emergency medical admission rates.
Standardised emergency medical admission ratios adjusted for patients7
age, sex and Carstairs7 deprivation categories.
Results: The numbers of emergency medical admissions doubled between
1980 and 1997. Emergency medical admission rates increased steeply with
increasing age of patients, more than doubling for every two decades. Men
above 40 years had approximately 20% higher age specific emergency
medical admission rates than women. Emergency medical admission rates
were more than twice as high amongst patients from Glasgow's most
deprived areas, compared with the most affluent.
Cardiovascular disease (ICD10 chapter IX) discharge diagnoses were
commonest (27.2% in 1997)' followed by the non-specific diagnoses in
ICDlO chapter XVII (21.0%). The non-specific diagnoses mainly comprised
chest pain (9.6%) which was the commonest reason for admission. Between
1980 and 1997, non-specific diagnoses (ICD10 chapter XVII) increased at
twice the rate of all other ICDlO chapters of diagnoses.
There were two areas of Glasgow that had raised standardised emergency
medical admission ratios (adjusted for patients' ages, sex and deprivation).
These corresponded to the catchment areas of two acute hospitals, which
had substantially higher adjusted emergency medical admission ratios
than had the other three.
There was a 2.51 fold variation between the top and bottom deciles of
Glasgow's general Practices' crude emergency medical admission rates.
After adjustment for their patients' age, sex and deprivation
characteristics, this reduced to a 1.87 fold variation. Additional adjustment
for general practices' admitting hospitals (along with their patients' age,
sex and deprivation) accounted for a total of 84% of the inter-practice
variation in crude emergency medical admission rates.
Fundholding general practices had modestly raised emergency medical
admission rates (odds ratio 1.06.) There were no associations between
practices' rates of emergency medical admissions and any other measured
practices' characteristics (including numbers of partners, practices' sizes or
dispersions, immunisation and cytology rates).
Conclusions: This large study discovered epidemiological associations
between emergency medical admission rates and patients' sex and socioeconomic
deprivation that had not previously been shown. It showed that
emergency admission rate variations between general practices were only
partly accounted for by patient characteristics. Apart from fundholding
status, difference between practices were not related to variations in their
rates of emergency medical admissions. However, a substantial part of the
variation between general practices could be attributed to differences
between their admitting hospitals.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Health & Wellbeing > Public Health
• Supervisor's Name: Capewell, Dr Simon and McEwen, Professor James
• Date of Award: 1999
• Depositing User: Adam Swann
• Unique ID: glathesis:1999-8483
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 05 Oct 2017 14:57
• Last Modified: 05 Oct 2017 14:57
• URI: https://theses.gla.ac.uk/id/eprint/8483