Age and Ageing 2008; 37: 455–478
The Author 2008. Published by Oxford University Press on behalf of the British Geriatrics Society.
All rights reserved. For Permissions, please email: firstname.lastname@example.org
Trends in hospital inpatient episodes for
signs, symptoms and ill-defined conditions:
observational study of older people’s
hospital episodes in England, 1995–2003
to increase with age, and hospital admissions are rising more
rapidly for those aged over 65 years, going up by 3% per
year compared with 2% for all age groups [1–6]. Aside
from the impact on health services, hospitalisation leads to
of formal and informal care, loss of social networks,
infections [1, 7]. Despite this, in excess of 20% bed days
in the United Kingdom are thought to be avoidable if a
suitable alternative to acute admission were available [8–10].
In the United Kingdom and elsewhere there is, therefore,
increasing emphasis on reducing preventable admissions for
older people. However, access to intermediate or primary
Recent UK health policy has made a commitment to
services that promote health and independence and prevent
unnecessary hospitalisation, emphasising case management
of chronic disease in primary care [1, 6]. It is thought that
these services will be more effective where interventions
focus on specific conditions or patient groups at high risk of
avoidable admission [1, 11–15].
One group identified in health policy as potentially avoid-
able admissions are those patients discharged from hospital
with signs, symptoms and ill-defined conditions [1, 12, 14],
i.e. patients whose primary disease codes are categorised
under Chapter XVIII of the International Classification
of Diseases (ICD)-10 , ‘Signs, symptoms and abnormal
laboratory findings’ (usually abbreviated to ‘ill-defined con-
ditions’, a part of the chapter title in previous versions of the
ICD ). Ill-defined conditions include not only symptoms
such as chest pain and breathlessness but also common
geriatric syndromes such as delirium, falls, incontinence and
collapse [18–21]. These symptom-related diagnostic codes
are a common feature of older people’s emergency admis-
sions in the United Kingdom and elsewhere and, whilst
numbers are increasing for all age groups, they are rising
more rapidly for older people [4, 5, 22]. This type of admis-
sion is likely to be seen as problematic in the context of
tariff-based payment systems based on diagnoses, where
reimbursement may not cover the costs incurred by these
patients; costs per admission for ill-defined conditions are
been attributed to a decline in community services and to an
the oldest old [12, 24], but there is little evidence to support
this view.In order to beable to respondappropriatelytothis
trend, its underlying causes must be understood. This study
explored recent growth in emergency hospital admissions
for ill-defined conditions in people aged 65 years and over
in England, and examined the impact of population growth,
demographic change and changes to hospital activity on the
rate of increase in incidence of this type of admission.
in England was carried out using the national Hospital
Episodes Statistics (HES) database . Data on total
emergency finished consultant episodes (FCEs) and FCEs
for ill-defined conditions were obtained for people aged
65 years and above. This analysis included data from all
hospital departments but, pilot work in medicine and care
of the elderly showed very high rates in these settings,
supporting the relevance of this analysis for geriatric
to 2003, the complete years available using ICD version 10,
resulting in a total sample of 2,793,653 FCEs for ill-defined
numbers of episodes were tabulated. Population estimates
were obtained from the Office of National Statistics .
Data were analysed using Microsoft Excel 2003 and STATA
release 8. Codes for uncategorised data and unknown cause
of death were excluded. Standardised episode rates were
calculated to explore the growth over time adjusted for age
group composition, total population and total FCEs.
During the study period there was a steady increase in the
numbers of ill-defined condition episodes, from 210,908 to
402,325 (an increase of 91%), a trend apparent within each
age band (Figure 1). Numbers of these episodes tended to
increase with age, with the youngest age group experiencing
fewer such episodes than the oldest age groups (Figure 1)
and rates per 1,000 people increasing by age group in all
study years (Table 1). The fall in overall numbers of these
is associated with variations in population estimates within
the Journal website http://www.ageing.oupjournals.org/)
groups during this period (Table 1). Crude episode rates per
1,000 people aged 65 years and over rose from 27.4 to 50.6
in the study period, an increase of 84.7%. However, during
by guest on December 29, 2015
Table 1. Rates of hospital episodes for ill-defined conditions (per 1,000 people) adjusted for population, age group and
total FCEs in people aged ≥65 years in England from 1995 to 2003
Rate in age group
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Total crude rate
rate (95% CI)
199519961997 199819992000 2001 20022003
(4.21,4.24)(4.76,4.79) (5.01,5.04) (5.28,5.31)
Figure 1. Number of ill-defined conditions related FECs for people aged 65 years and over in England, 1995–2003.
(an increase of 52%), whilst the number of people aged
over 65 years rose from 7,696,700 to 7,947,000 (an increase
of 3%) (see Appendix 1 in the supplementary data on the
Journal website http://www.ageing.oupjournals.org/). Two
concurrent trends may therefore have influenced numbers
of admission episodes: the general increase in FCEs within
the National Health Service(NHS); and demographic trends
to an older population. The analysis was therefore adjusted
to account for changes to both population and FCEs. Direct
standardisation of episode rates to account for changes in
age distribution, population and total FCEs produces lower
rates. However, adjusted rates have risen year on year during
the study period (Table 1), from 3.0 in 1995 to 5.7 in 2003,
a change of 91%. The proportion of the older population
experiencing these episodes rose from 2.8 to 5.1% of people
aged over 65 years.
During the study period, total admission episodes for ill-
defined conditions increased. The literature suggests that
hospital episodes for ill-defined conditions are associated
with increased age, and therefore such episodes can be
expected to increase as the population ages . It is known
that the population of England has grown during the study
period, particularly the proportion of the population aged
over 65 years. However, another possible driver of rising
reported during the study period. Episode rates for ill-
defined conditions were therefore adjusted for the number
of older people in the population, the age distribution of
the older population and the overall number of FCEs. The
episodes amongst older people exceeded what was expected
on the basis of these factors alone. However, whilst the
adjusted standardised rates almost doubled during this time,
the adjusted incidence rate is much lower than what would
be expected from the crude episode data. This suggests that
much of the recent increases in this type of admission is
likely to be due to a more general rise in hospital episodes,
which have increased dramatically during the study period,
and to some extent may reflect changesto reporting patterns
or increased hospital activity. Nevertheless, a significant
increase in incidence rate for these episodes is demonstrated
by the adjusted analysis, although the underlying reasons for
of ill-defined condition episodes are due to organisational
changes that have occurred during this time period. One
explanation, favoured by current UK health policy, might
by guest on December 29, 2015
be that these admissions reflect a lack of access to primary
and community care services that would otherwise support
chronically ill older people and avoid the need for hospital
admission. However, other organisational factors, such as
new systems of out-of-hours care, reduced waiting times
in Accident and Emergency departments generally increased
and funding in primary care might all contribute to the
observed trend. During the study period, the mean length
of stay for hospital episodes has decreased more for ill-
defined conditions than other types of episodes, possibly
reflecting more internal transfers or shorter admissions and
coding accuracy and readmissions is not known. The oldest
and most frail individuals may be more susceptible to the
fragmentation of care that occurs with increasingly rapid
transit through the acute care system, making them more
likely to have overlooked or incorrect diagnoses. Older
people are known to be more likely to present with atypical
a challenge [20, 21]. However, accuracy of diagnostic coding
has been shown to be high when considered at this broad
chapter level, and there are indications that coding accuracy
is improving [26, 27].
Growing numbers of hospital episodes for ill-defined
conditions represent a challenge for health services. An
ageing population and rising hospital episodes will both
contribute to increased hospital episodes for ill-defined
conditions. However, these factors do not fully explain
the recentrisesin incidencerate ofadmissions for ill-defined
conditions, and it is likely therefore, that organisational
factors are also implicated. We also need to explore the
possibility of changes in accuracy of diagnosis or coding.
Further research is required into the relation between age,
health needs and different approaches to the organisation of
care in order to determine the most appropriate approach to
management of admissions for ill-defined conditions.
• Inpatient emergency episodes for ill-defined conditions
in older people are rising in the United Kingdom.
• These admissions rise with age and are likely to increase
as the population ages.
• Increases in these admissions are not fully explained by
either population growth or episode growth.
• Further research on the interaction between ageing and
organisational factors is required.
Conflicts of interest
Approval for this study was obtained from Southampton
and South West Hampshire Research Ethics Committee.
The corresponding author was funded to carry out this
work by a Department of Health National Co-ordinating
Centre for Research Capacity Development Post-Doctoral
Supplementary data for this article are available online at
BRONAGH WALSH1∗, HELEN C. ROBERTS2, PETER G. NICHOLLS1,
VALERIE A LATTIMER1
1School of Nursing and Midwifery, University of
Southampton, Highfield, Southampton, SO17 1BJ, UK
2Geriatric Medicine, University of Southampton, Southampton
General Hospital, Tremona Road, Southampton, SO17 6YD, UK
∗To whom correspondence should be addressed
1. DH. National Service Framework for Older People. London:
2. ONS. Healthy Life Expectancy—Living Longer, More Years
in Poor Health. London: TSO, 2004.
3. Brayne C, Matthews F, McGee MA et al. Health and ill-health
in the older population in England and Wales. Age Ageing
2001; 30: 53–62.
4. DH. Shaping the Future NHS: Long-Term Planning for
Hospitals and Related Services: Consultation Document on
the Findings of the National BedsInquiry. London: DH,2000.
5. DH. The NHS Plan. London: DH, 2000.
6. Commission for Healthcare Audit and Inspection. Living Well
7. Hanlon P, Walsh D, Whyte BW et al. Hospital use by an ageing
behavioural and social risk markers and subsequent hospital
utilisation. J Public Health Med 1998; 20: 467–76.
8. McDonagh MS, Smith DH, Goddard M. Measuring
and results. Health Policy 2000; 53: 157–84.
9. Kalant N, Berlinguet M, Dragatakis L et al. How valid are
utilisation review tools in assessing appropriate use of acute
care beds? Can Med Assoc J 2000; 162: 1809–13.
10. Victor C, Khakoo A. Is hospital the right place? A survey of
‘inappropriate’ admissions to an inner London NHS Trust.
J Public Health Med 1994; 16: 286–90.
11. Gravelle H, Dusheiko M, Sheaff R et al. Impact of case
management (Evercare) on frail elderly patients: controlled
before and after analysis of quantitative outcome data. BMJ
2007; 334: 31–4.
12. Wanless D. Wanless Social Care Review. Securing Good Care
for Older People. Taking a Long-term View. London: Kings
13. Scott I. Optimising care of the hospitalised elderly. A literature
review and suggestions for future research. Aust N Z J Med
1999; 29: 254–64.
by guest on December 29, 2015
Research letters Download full-text
Programme in England, 2003–2004. Final Report, 2005;
15. Dr Foster Intelligence. Keeping People Out of Hospital. The
challenge of reducing emergency admission, 2006; http://www.
EmergencyAdmissions February 2006.
16. WHO. International Classification of Diseases and Related
Health Problems, 10thRevision. Geneva: WHO.
17. WHO. International Classification of Diseases and Related
Health Problems. 9thRevision. Geneva: WHO, 1978.
18. Walsh B, Roberts HC, Hopkinson J. Emergency hospital
admissions for ill-defined conditions amongst older people:
a review of the literature. Int J Older People Nurs 2007; 2:
19. Walsh B, Roberts H. Older people’s use of Accident and
Emergency services. Age Ageing 2005; 34: 535.
20. Fried LP, Store DJ, King DE et al. Diagnosis of illness
21. Berman P, Hogan DB, Fox RA. The atypical presentation of
infection in old age. Age Ageing 1987; 16: 201–7.
22. HES Online. Hospital Episodes Statistics http://www.
23. DH. NHS Reference Costs 2005, 2006; http://www.dh.gov.
frail elderly admitted to the Emergency Department for lack
of community support. Aging Clin Exp Res 1999; 11: 56–60.
25. OfficeofNational Statistics
26. Aylin P, Bottle A, Majeed A. 2007 Use of administrative data
or clinical databases as predictors of risk of death in hospital:
comparison of models. BMJ 2007; 334: 1044.
27. Mohammed MA, Stevens A. The value of administrative
databases. BMJ 2007; 334: 1014–5.
Published electronically 16 May 2008
Glucose control levels, ischaemic brain
lesions, and hyperinsulinaemia were
associated with cognitive dysfunction in
SIR—Type 2 diabetes mellitus (DM) in the elderly is
associated with impaired cognitive functioning and an
increased risk of dementia [1–4]. The pathogenesis of the
impairment, however, remains unclear.
Our previous study indicated that the performance of
a cognitive functional test was positively correlated with
glycohaemoglobin, HbA1c, which is an index of glucose
control . Recently, several studies suggested that use
of anti-diabetic medication was associated with improved
cognitive function , or prevented a decline in cognitive
Several reports have indicated that hyperinsulinaemia is
associated with cognitive dysfunction and dementia in the
general population [7, 8]. A small preliminary study reported
that insulin sensitivity measured by the euglycemic insulin
clamp method was inversely correlated with the cognitive
functional test score .
Many studies have reported that the diabetic elderly
have ischaemic brain lesions such as lacunae infarctions,
white matter lesions and paraventricular lesions even
without neurological symptoms . Several reports have
also suggested that these ischaemic lesions were associated
with cognitive dysfunction in the diabetic elderly [11, 12].
In this study, we analysed the association of HbA1c,
hyperinsulinaemia and ischaemic brain changes to DM-
related cognitive dysfunction by performing an assessment
of subjects’ profiles including a brain imaging assessment
Subjects and methods
For the present study, we recruited consecutively 77 patients
with type 2 DM from the Chubu Rosai Hospital’s Diabetic
criteria were as follows: malignancy, inflammatory disease
(such as collagen disease, thyroid disease and viral hepatitis),
severe cardiovascular disease (such as myocardial infarction
and unstable angina). None of the subjects had audio-visual
deficiencies that would prevent them from participating in
the cognitive functional assessment.
An ethical committee approved the study and all patients
After giving informed consent, the cognitive functional tests
were administered individually to each subject. On the day
of the assessment, the subjects had breakfast as usual and
the assessment was performed in the morning. Doctors
and nephropathy were diagnosed as below: neuropathy, ele-
vated vibratory perception thresholds or symptomatic neu-
nephropathy, microalbuminuria and more advanced.
Cognitive function was assessed by structured performance
(i) Mental status: the Mini-Mental State Examination
(MMSE) . (ii) Verbal memory (Word Recall): the Word
List (a subtest of the Alzheimer’s Disease Assessment Scale
(ADAS) ; with a score range of 0–10. (iii) Complex
psychomotor skill: the Digit Symbol Substitution (DSS)
by guest on December 29, 2015