W.
Stuart Hislop * and
Robert C. Heading #
SMJ 2004 49(2): 57-60
(On
behalf of the Caledonian Society of Gastroenterology)
*
Dept. of Gastroenterology, Royal
Alexandra Hospital, Paisley, PA2
9PN.
#
Centre for Liver and Digestive Disorders, Royal Infirmary, Edinburgh, EH3 9YW.
Correspondence
to; Dr WS Hislop, BSc, FRCP
Abstract
Background
Concern among Scots gastroenterologists about alcohol related illness
prompted this inpatient prevalence study during the winter of 2000-01.
Aims
To study gastroenterology inpatient
workload due to alcohol-related illness, to determine how much was specialty
specific, and if there were regional variations.
Methods
40 Consultant Gastroenterologists throughout Scotland collected data on
the prevalence of alcohol related conditions among inpatients under their care
on each of three specified days during the winter of 2000/2001. All inpatients
under the care of participating consultants on the designated study days were
included in the study. Overall return rate was 65%.
Patients
were categorised as follows; (a) general medical inpatients admitted for reasons
other than alcohol related illness (b) general medical inpatients with no
gastrointestinal or liver disease, but whose admission to hospital was primarily
related to alcohol misuse, (c)
gastrointestinal (including liver) inpatients admitted for reasons unrelated to
alcohol intake, and with no alcohol related disease, and
(d) gastrointestinal inpatients whose admission to hospital resulted from
alcohol related disease. Additionally, the numbers of patients with (e)
decompensated liver disease of all causes ,
(f)
decompensated alcoholic liver disease, and (g) the numbers “blocking” acute
beds after initial hospitalisation with an alcohol related illness were
collected.
Results Overall, 829
general medical and 538 gastroenterology inpatients were entered in the study;
total 1367 (705 male, 662 female). Of these, 25% (337/1367) were admitted
because of alcohol related illness:
15% (201/1367) had decompensated
alcoholic liver disease. Of 538
gastroenterology inpatients, 238(44%) had problems related to alcohol, and 201
of these (37% of all gastroenterology inpatients) had decompensated alcoholic
liver disease. Of 246 inpatients
with decompensated liver disease, 82% (201) had alcoholic liver disease.
Alcohol related illness was significantly more prevalent among male
inpatients in the West of Scotland. 10% of specialist gastroenterology beds were
occupied by patients whose discharge was delayed because of alcohol related
problems.
Conclusions Most Scottish gastroenterologists
contribute to general medical receiving but their specialist inpatient workload
is dominated by treatment of patients with alcohol related disease. (44% in
gastroenterology v 12% in general
medicine). Inpatients with decompensated alcoholic liver disease form 37% of
gastroenterology workload. Alcohol related disease contributes to delayed
discharge in acute medical units, especially in gastroenterology wards.
There are regional differences in prevalence of alcohol related disease,
which is greatest in male inpatients in the West of Scotland. Here, alcoholic
liver disease accounts for nearly all decompensated liver disease. The findings
point to a need to review the current patterns of acute service provision for
alcohol related illnesses, so as to assess and improve both the clinical
effectiveness and cost effectiveness of care, and to ensure that alternatives to
acute hospital admission are available when appropriate. This need should not be
neglected while efforts are simultaneously being made to improve the early
detection of alcohol abuse and prevent irreversible alcohol related disease.
Keywords
Alcohol
related illness. Gastroenterology
workload. Scotland.
Alcoholic Liver Disease
Introduction
During discussions in 1999 and 2000 about resources
needed to manage and treat an increasing number of hepatitis C patients, several
Scottish gastroenterologists expressed the opinion that they
were encountering an even bigger increase in the numbers of alcohol
related problems in their clinical practice, particularly among inpatients, such
that this workload was becoming their dominant concern.
A small one-month study undertaken in one hospital around this time had
found the prevalence of patients with alcohol related problems in
gastroenterology wards to be as much as 56% [1], but no other up to date
information quantifying this workload appeared to exist. Morbidity and mortality
data have shown an increasing problem of alcoholic liver disease in Scotland.
For example, between 1983, (ten years after the 1973 Clayson Reforms [2]
liberalising the alcohol licensing laws) and 1995, deaths from alcoholic
cirrhosis in Scottish hospitals rose by 160% [3].
More recent figures show a 280% rise in total annual inpatient discharges
due to alcoholic liver disease, from 1466 in 1996 to 3813 in 2000 [4]. These
statistics do not necessarily provide a direct indicator of clinical workload,
however, and it was therefore agreed that a co-ordinated study of alcohol
related hospital admissions should be attempted by all consultant
gastroenterologists in Scotland.
Methods
The study was performed through the Caledonian
Society of Gastroenterology by asking all 53 consultant physician members
working in Scottish hospitals to count the inpatients under their care on three
separate days during the winter of 2000-01. These three dates (14/12/00, 18/1/01
& 22/2/01) provided a “snapshot” which was deemed to be representative
of the pattern of inpatients ordinarily managed by these consultants. (Each
individual consultant was asked to complete the return on only his own patients,
to avoid “double counting” in larger units with several consultants.)
Further categorisation of the patients was
sought on the basis of underlying diagnoses:
(a)
general medical (non-alcohol
related)
(b)
general medical (alcohol
related)
(c)
gastroenterology
(non-alcohol related)
(d)
gastroenterology (alcohol
related)
(e)
decompensated liver disease
(all causes)
(f)
decompensated alcoholic
liver disease
(g)
blocked beds due to chronic
alcohol problems
Inpatients were
classified as having alcohol related pathology if the condition leading
to admission was directly caused by alcohol (e.g. alcoholic liver disease,
alcoholic peripheral neuropathy, alcohol withdrawal states, Korsakov psychosis
etc.), or was one in which alcohol played a major contributory part (e.g.
post-binge upper gastrointestinal bleeding, starvation
hypoglycaemia, opportunistic infection with a background of chronic
alcoholism, Wernicke’s encephalopathy,
self-poisoning while intoxicated etc.)
In addition to identifying overall prevalence, the
possible existence of differences between the East of Scotland and the West of
Scotland were explored.
Where indicated, statistical analysis was on the
basis of the chi-square test applied
Results
Forty of the 53 consultants returned data for
at least one day, giving results totalling 82 “consultant/days”. The
potential data returns over the 3 days would appear to have been 159 consultant
days (53x3), but 32 days of consultant annual and other leave were reported,
thus reducing the maximum relevant return to 127 consultant days. The response
rate was therefore approximately 65% (82/127). Only 3 units sent back no returns
on any of the 3 study days. Eighty
four percent of returns came from consultants who participated in acute medical
receiving in their hospitals: the other 16% from physicians working in
gastroenterology only. About half
the returns were from consultants who, along with their gastroenterologist
colleagues, took responsibility for all gastroenterology patients admitted to
their hospitals. Consultants in district general hospitals generated 52% of the
returns: 48% came from university or teaching hospitals.
The
Overall Picture
Information
about 1,367 inpatients was obtained for the 3 study days (M/F 1.06/1) (Table
1). Around 61% (829 patients) were deemed to be general medical,
and 538 (39%) were gastroenterology inpatients. . Of the 829 general medical
patients, 99 (12%) were admitted on account of alcohol related problems. Among
the 538 gastroenterology patients, however, 238 (44%) were identified as being
admitted because of alcohol related illness (p<0.001).
The
12% of general medical inpatients admitted due to alcohol problems indicates
what a consultant physician of any sub-specialty in the participating hospitals
might expect in the course of acute general medical receiving. For the
gastroenterologists participating in general medical receiving, however, 25%
(337/1367) of their total inpatients are in hospital because of alcohol related
illness – 12% of the general medical patients and 44% of the gastroenterology
patients.
Among
the gastroenterology inpatients, the percentage of males with alcohol related
problems was approximately 52% (153/297)
whereas the corresponding figure for females was 35% (85/241; p<0.001). (Table
2).
Decompensated
liver disease represented 46% of all gastroenterology inpatients (246/538), and
in more than 82% of patients alcohol was the cause (201/246). There was a
significant difference (p<0.001) between
the men (89%; 128/144) and women (72%; 73/102) as might be expected.
Regional
Differences
Slightly
fewer inpatients (1,323) made up this part of the study because it was not
possible to identify conclusively the source hospital for 2 returns. The basis
of the geographic split is shown in Table
3.
Among
the general medical inpatients, 12% (98/810) were admitted primarily because of
alcohol related disease, but again geographic and gender differences were
apparent. In the West of Scotland
19% of male inpatients (44/236) had alcohol related problems, but the figure was
13% (21/164) in the East (p>0.05; n.s.). Only 8% of female general medical
inpatients had alcohol related illness – an identical proportion in the West
and the East.
Forty
four percent (225/513) of the gastroenterology inpatients were admitted because
of alcohol related problems (Table
5). In the West of Scotland, figures were 59% (77/130) among males and
35% (37/106) in females (p<0.001). The gender difference was less
marked in the East where the corresponding proportions were 43% (67/155) in
males and 36% (44/122) in females, (p>0.1; n.s.)
Thirty
seven percent (191/513) of all the gastroenterology patients in this part of the
study were admitted because of decompensated alcoholic liver disease.
In the West of Scotland, the figure was 50% of males (65/130), and 29%
(31/106) in females ( p<0.05), but in the
East no gender difference was evident ; 35% (55/156) in men and 33% (40/122) in
women.
Of
the 234 patients with decompensated liver disease, 82% were due to alcohol (Table
6). Among men in the West of Scotland, 94% of decompensated liver disease
was due to alcohol (65/69) compared to only 72% (31/43) in women. (p<0.001) The corresponding figures for the East were 83% (55/66) and
71% (40/56): this gender differences was not significant (p>0.05).
Among the male inpatients with decompensated liver disease, alcohol was
significantly more likely to be the cause in the West than in the East.
(p<0.01)
Bed
Occupancy
A
further item of information requested of the participating consultants was
identification of any of their acute beds that were blocked by inpatients with
long-term alcohol problems (e.g. Korsakov psychosis/dementia, chronic peripheral
neuropathy or alcohol related brain injury) for whom more appropriate care was
awaited. It was reported that 54 of the 1,367 inpatients were blocking
acute beds. This constitutes only
4% of total inpatient beds to which gastroenterologists have access.
However, because of the lengthy nature of their stay in hospital, and the
way most Scottish hospitals now structure general medical admitting [6,7], these
patients had all been moved from acute general medical receiving beds to
specialist gastroenterology care, where their proportionate impact was
correspondingly greater at 10% (54/538). Moreover, these figures show that
Scottish gastroenterologists consider the acute care setting to be inappropriate
for 16% (54/337) of those inpatients admitted with alcohol related problems.
Discussion
This
study shows that the care of individuals with alcohol related illness is a major
part of Scottish gastroenterologists’ inpatient responsibilities. With
hindsight, of course, it is easy to see that additional information could also
have been sought in the study, and perhaps steps should have been taken to
increase the data return rate. Furthermore, no study undertaken at one point in
time could corroborate the opinion expressed at the outset that patients with
alcohol related illnesses are an increasing proportion of consultants’
workload. Nevertheless, the
prevalence of alcohol related illness now demonstrated requires consideration of
its implications, irrespective of how the trends of alcohol related illness go
in the future.
A
particular involvement of gastroenterologists in the management of patients
admitted to hospital with alcohol related illness may be expected, partly
because most consultant gastroenterologists in the United Kingdom participate in
acute general medical receiving, and partly because alcohol related end-organ
damage so frequently affects the gastrointestinal tract and liver. New triage
and specialty based methods of acute medical receiving [6,7], which have been
widely adopted in Scotland as elsewhere in the UK in recent years, have tended
to direct many inpatients with alcohol related illness to gastroenterology beds
whenever there is suspicion of liver disease. Consequently, gastroenterologists
deal with more alcohol related illness than colleagues in other medical
subspecialities.[1]
The
results also demonstrate that gastroenterology inpatient practice in Scotland is
now dominated by the treatment of decompensated alcoholic liver disease,
particularly among male inpatients in the West of Scotland. In contrast, there
are relatively few inpatients with decompensated liver disease due to causes
other than alcohol. The
West – East differences that have been shown presumably reflect the
well-recognised health problems related to poverty, unemployment and urban
deprivation and, additionally, cultural attitudes to excessive drinking endemic
in the West of Scotland.
Recognition
that alcohol related illness is now a major cause of morbidity and premature
death [3,4,8] and incurs major costs for the NHS has prompted consideration of
what should be done [9]. Not unreasonably, more effort to prevent alcohol abuse
(especially in young people), screening for dependence, observance of
detoxification protocols, and the provision of better support and specialised
alcohol psychiatry services in acute hospitals have been recommended. [9] In
Scotland, a comprehensive plan for action has been issued by the Scottish
Executive Health Department [10]. The thrust of most of this proposed activity
is the prevention of alcohol related illness. However the present study
demonstrates there is also a need to reassess the pattern of acute care now
being given.
In
view of the evident substantial usage of acute hospital beds, and the fact that
patients with chronic liver disease often require costly treatments and
interventions, it is regrettable that there is so little information available
about the effectiveness and cost effectiveness of the care given [11].
Is the time and resource commitment given to the patients with
decompensated liver disease by gastroenterologists and other specialists
appropriate, given the other calls on their time and on the available resources?
In 1997, one American report declared that it costs $110,000 to die from
end stage liver disease. In essence, this report showed that the treatments
given to these patients were expensive and achieved little benefit, except when
liver transplantation was undertaken in suitable patients
[12]. This may also be the current position in the United Kingdom. For example,
there is still controversy about when a TIPS shunt, (an
intervention calculated in Spain to cost 21,600 Euros per patient),
should be undertaken in patients with bleeding oesophageal varices, and whether
or not it prolongs life [13-15]. Although only a small minority of patients with
alcoholic liver disease are candidates for transplantation, alcoholic liver
disease is now the commonest indication for a liver transplant in Scotland [16].
Without
disputing the benefit that will come in time from successful prevention of
alcohol abuse and consequent alcohol related illness, there is an immediate need
to examine critically the care now being given in the acute medical and acute
gastroenterology settings to patients with alcohol related illness to ensure
that resources are expended to good effect and that more appropriate care
settings are made available when required. Education and training for healthcare
staff in the recognition of alcohol dependence, and on how to intervene
constructively has been urged by the Royal College of Physicians [9], and may
improve the care of patients with subtle manifestations of alcohol dependence.
However, the evidence in the present study that 16% of inpatients with alcohol
related illnesses are thought to be “blocking” acute beds, implies that
those same healthcare staff need help now to evaluate what they are presently
doing, and to identify what could be done more appropriately. Both in general
medical receiving units and in gastroenterology wards, the cost effectiveness of
current practices and possible alternatives to them should be examined,
especially in respect of the large number of patients with decompensated
alcoholic liver disease.
Acknowlegements
We
acknowledge the help and support of the following colleagues, all members of the
Caledonian Society of Gastroenterology, who contributed to the patient data
collection, and without whom this study would not have been possible.
Dr
JG Allan, Dr AD Beattie, Dr PM Bramley, Prof. KM Cochran, Dr RW Crofton, Dr GW
Curry,
Dr
BJ Danesh, Dr JB Dilawari, Dr JF Dillon, Dr OE Eade,
Dr AM El-Nujumi, Dr NDC Finlayson, Dr E Forrest, Dr. JAH Forrest, Dr CJR
Goddard, Prof. PC Hayes, Dr RJ Holden, Dr DA Johnson,
Dr
PR Mills, Dr AJ Morris, Dr NAG Mowat, Dr AJ McGilchrist, Dr JF Mackenzie, Dr AW
McKinlay, Dr JR McPeake, Dr KR
Palmer, Dr RHR Park, Dr ID Penman, the
late Prof.CR Pennington,
Dr
J Plevris, Dr AT Prach, Dr T
Reilly, Dr JDR Rose, Dr WSJ Ruddell, Dr AS Taha,
Dr
AN Shepherd, Dr K Simpson, Dr KC Trimble, and Dr AJK Williams.
References
Butler SR, Hislop WS, Fisher BM, et al. Consultants workload due to
alcohol related conditions in acute medical receiving, gastroenterology and
endocrinology. Scot. Med. J. 2001; 46: 104-105
Clayson C. Report of the Departmental Committee on Scottish Licensing
Law. Edinburgh: HMSO, 1973
Chick J. Evidence suggesting increasing health damage in Scotland
related to alcohol. Health Bull. 1997; 55 (3): 134-139
Scottish Health Statistics, ISD, Edinburgh, 1997-2001 (www.show.scot.nhs.uk/isd/)
Swinscow TDV. Statistics at Square One. London: BMA, 1976
Dorward AJ. Changes in Patterns of Acute Medical Receiving in
Scotland 1996 to 2001. Scot. Med. J. 2002; 47:105-108
Hanlon P, Beck S, Robertson G et al. Coping with the inexorable rise
in medical admissions: evaluating a radical reorganisation of acute medical
care in a Scottish district general hospital. Health Bull. 1997; 55 (3):
176-184
Annual Report of the Chief Medical Officer, 2001. Department of
Health, London.
Alcohol - can the NHS afford it?
Report of a Royal College of Physicians Working Party. Royal College
of Physicians, London, 2001
Plan for action on alcohol problems. Scottish Executive Health
Department. 2002.
Neuberger J, Bryan S. Clinical economics review: cost-effectiveness
in the therapy of liver disease. Aliment Pharmacol Ther 1997; 11: 61-68.
Wong LL, McFall P,
Livingston MF. The cost of dying of end-stage liver disease. Arch.
Internal Med.1997; 157: 1429-1432.
Escorsell A, Banares R, Garcia-Pagan J C, Gilabert R, Moitino E,
Piqueras B et al. TIPS versus drug therapy in preventing variceal rebleeding
in advanced cirrhosis: a randomised controlled trial. Hepatology 2002; 35:
385-392.
Jalan R, Bzeizi KI, Tripathi D, Lui HF, Redhead DN, Hayes PC. Impact
of transjugular portosystemic stent-shunt for secondary prophylaxis of
oesophageal variceal haemorrhage: a single centre study over an 11 year
period. Eur J Gastroenterol Hepatol 2002; 14: 615-626
Harrison P, Karani J. Secondary prophylaxis of oesophageal variceal
haemorrhage: a role for transjugular intrahepatic portosystemic shunt? Eur J
Gastroenterol Hepatol 2002 ; 14: 591-593.
Annual Report 2001-2002. Scottish Liver Transplant Unit.