Epidemiology of intensive care medicine:

supply versus demand

J F Bion* and D Bennett*

*Department of Intensive Care Medicine, University of Birmingham and Queen Elizabeth Hospital,
Birmingham, UK and ^Department of Intensive Care Medicine, St George's Hospital, London, UK

Developments in hospital medicine combined with social and demographic

changes are likely to increase the need for intensive care services at a time when
cost containment and cost-efficacy are the main items on the political agenda.
This will exaggerate the supply-demand outcome mismatch unless the problem is
approached in a constructive manner by clinicians, managers and politicians. More
resources will be required for intensive care, but these must be better targeted
and more efficiently employed. Opportunities for prevention should be explored,
with intensive care being given a pro-active rather than a re-active role. Intensive
care clinicians should understand that this expanded role cannot be achieved if
they are willing only to accept responsibility for patient care after the patient has
been admitted to the ICU. Clinicians and managers should develop methods for
linking the various disciplines which contribute to emergency care, to form an
acute care framework within the hospital. Research into the factors which
determine risk of critical illness should be combined with enhanced medical and
nursing training in intensive care, accompanied by an expansion in resources for
intermediate and high dependency care in countries like the UK where there is
clear evidence of rationing.

Intensive care originated from the polio epidemics of the 1950s. The
epidemic which afflicted Copenhagen was described by the epidemiologist
Professor Lassen1. Those polio victims who developed respiratory failure
were initially managed using 'iron lung' ventilators in different areas of
the hospital. Within a few weeks about 100 patients had been received.
The mortality rate of 90% combined with the lack of adequate numbers
of these ventilators suggested the need for a new approach, and the advice
of the anaesthetist Professor Ibsen was obtained. He showed, using the
Correspondence to:
relatively new technique of arterial blood gas analysis, that the patients
DrJF Bion, Department were dying from hypoxaemic ventilatory failure. Three important changes
of Intensive Care were then introduced: patient care was centralised within the hospital;
Medicine, airway control was provided via a surgically formed tracheostomy; and
Queen Elizabeth
Hospital, Edgbaston, the patients' lungs were ventilated with oxygen enriched air using
Birmingham B15 2TH, UK manual positive pressure provided by teams of medical students. This

British Medical Bulletin 1999,55 (No. 1): 2-11 C The British Council 1999

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Epidemiology of ICM: supply vs demand

resulted in a reduction in mortality to 40%. Lassen noted the psycho-

logical benefits of having a dedicated attendant permanently by the
bedside, but also pointed out that this was more expensive. He also
showed that while the new system of management produced many more
survivors, it delayed death amongst patients destined not to survive. The
old (cuirass) method allowed more patients to die faster and, therefore, at
lower cost.

Changes and challenges in intensive care

Cost-efficacy

Almost 50 years on, what progress have we made? The more visible
aspects of intensive care practice have changed substantially. What started
as a method of providing respiratory support to patients with a single
organ system failure has evolved into a range of technical developments
that allow temporary functional replacement of most organ systems. The
problem is that this can be achieved without necessarily effecting a cure.
In fact, it is precisely because of intensive care support that the pheno-
menon of multiple organ failure has become apparent. The 40% mortality
rate reported by Lassen as a success has now become a somewhat resistant
challenge for intensive care, particularly in the context of sepsis.
Moreover, as it may cost twice as much to die in intensive care as to
survive2, a substantial proportion of the intensive care budget is con-
sequently expended on patients who cannot benefit3. Clinical uncertainty
about individual patient outcomes cannot be resolved by scoring systems,
despite the significant contribution made by intensive care to this area of
severity description. In a resource-constrained environment, this is both a
financial and an ethical problem. Politicians and medical managers,
charged with the responsibility for maximising benefit within the medical
commons, tend to concentrate on volume-cost relationships, whereas
healthcare providers identify more closely with quality of care at the level
of the individual patient4. This produces tensions in intensive care where
costs are closely related to staffing levels (already identified by Lassen as
being a quality issue), since a lower nurse:patient ratio is often presented
as the simple solution to intensive care resource limitations.

Resource allocation

The majority of developed countries have been examining methods for

controlling their health care budgets, and high-cost, low-throughput
specialities are an obvious target. Even in the USA which allocates around

British Medical Bulletin 1999;55 (No. 1)

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Intensive care medicine

US$60 billion annually5 to intensive care (20-30% of acute hospital costs,

or 1% of GDP6) there is a perceived need for cost-containment, as
exemplified by the development of managed care organisations. This
presents a particular problem for countries like the UK which has one of
the lowest national expenditures on health care generally (6.2% of GDP)
and on intensive care in particular (around 2% of the acute hospital
budget)7. Rationing intensive care is a well documented phenomenon8,
and probably explains why the UK has one of the most severely ill
intensive care patient populations9: the less ill patients continue to receive
care on ordinary wards, until their condition has deteriorated to the point
at which intensive care support can no longer be delayed. The inevitable
consequence of this approach is that patients are admitted to intensive
care late, at a stage where the chances of reversing or preventing organ-
system dysfunction are remote. This re-inforces cost-ineffective practice,
by concentrating scarce resources on patients who may no longer be able
to benefit, and whose death is then deferred at great emotional and
financial cost to the medical commons.
Given this background, at a national level one might have expected
governments to want to define fairly carefully the volume of intensive care
services that a particular population might require. In fact, the reverse
appears to be true, and the evidence suggests that intensive care resource
allocation bears no relationship whatsoever to population need, whether
expressed per capita or according to indices of population health10"12.
There are several reasons for this. Amongst the most important is the
multiple subspeciality status in which intensive care exists in many
countries, although this is at last changing to allow intensive care to
develop as a multidisciplinary speciality13. The effect of this is that
intensive care is often incorporated (inaccurately, if it is identified at all)
as an add-on to major speciality budgets, and hospitals, therefore,
perceive intensive care as an undesirable drain on resources, instead of as
a useful source of income from external contracts. The second reason is
that there has been virtually no research investment to determine
population needs in relation to public health and existing hospital
services14. Finally, despite the proliferation of scoring systems in intensive
care, there is no universally agreed common terminology to describe
intensive care activities, standards or case mix; this has made it difficult to
establish international comparisons and recommendations.

Demographic shifts

Many new technologies have been introduced into medical practice with
inadequate systematic evaluation of cost-efficacy, and this is also true of
intensive care. Part of the problem with evaluation is that the effects of

Bntish Medical Bulletin 1999;55 (No. 1)

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Epidemiology of ICM: supply vs demand

new technologies are complex when analysed at the level of whole

populations instead of the group of patients to whom the interventions
have specifically been applied. For example, improved peri-operative
care has made it possible to perform major surgery in patient groups
that only a few years ago would have been denied these operations. The
hospital and community populations have also changed, with earlier
hospital discharge of the less dependent patients, and a higher
proportion of patients with chronic disease and severe health limitation.
At the same time, the knowledge gained from intensive care practice,
when applied pro-actively in the field of resuscitation, has probably
reduced the need for intensive care for some of the less severely ill
patients (for example, gastrointestinal haemorrhage, asthma, diabetic
coma). The net result is that there has been a progressive severity shift
in intensive care case mix15, with an increase in the proportion of
severely ill patients with limited physiological reserve whose disease
processes are less susceptible to curative treatment.

The research minefield

Much of the research in intensive care during the last 20 years has focused
on established critical illness associated with sepsis. The proliferation of
biotechnology companies has tended to encourage a 'magic bullet'
approach to research, based more on the development of a marketable
intervention rather than basic scientific knowledge. Indeed, the field of
sepsis research is littered with studies which have been terminated early,
either because the active treatment arm showed no difference, or actually
demonstrated harm. In one instance, the active treatment was associated
with a significant increase in early, rather than late, deaths, presenting the
ethical dilemma of a potentially harmful drug which could regarded as
cost-effective by reducing treatment times in the non-survivors. There is
now a growing appreciation that more basic science research (and more
funding) is required in this area, including an examination of those factors
that determine susceptibility to critical illness.

Current developments and planning for the future

Intensive care needs to become better targeted, more pro-active, and
more cost-effective if it is to make more than just a supportive
contribution to hospital medicine. A collaborative effort is required to
integrate research in basic science, epidemiology, and management: the
Cochrane Collaboration is a good example of the sort of structure that
facilitates this type of interdisciplinary work. Areas for investigation

British Medical Bulletin 1999;55 (No. 1) 5

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Intensive care medicine

should include the pathogenesis of organ system dysfunction, the factors

that determine susceptibility to critical illness and methods for its pre-
vention, techniques for enhancing tissue repair, more accurate descriptors
of quality of care and outcomes, and explicit criteria for funding and con-
tracting related to population need. We consider some of these factors
below.

Defining the scope of intensive care

Few countries have explicit criteria for admission to intensive care, or

define what is meant by an intensive or high-dependency care unit. In the
UK, a multidisciplinary working party has produced national recommen-
dations based on the requirement for organ system support16, which
distinguish different levels of care in terms of standards and resources
(such as the nurse:patient ratio) and the degree of physiological support.
Intensive care is defined in this way as being appropriate for patients
requiring advanced respiratory support alone or combinations of two or
more other acute organ system failures (OSF); such patients require a 1:1
nurse:patient ratio. High-dependency care provides a lower nurse:patient
ratio (1:2) for patients needing support of a single acute OSF, excluding
advanced respiratory support. Admission/discharge criteria from the
Society of Critical Care Medicine in the USA17 and others18 are less
explicit. The problem with these definitions is that they have not been
developed a priori and, therefore, need independent evaluation. Most
importantly, such an evaluation should take into account denominator
data14 - that is, all hospitalised patients, in order to determine the
frequency with which the criteria might appear in the population 'at risk'.
Criteria based on interventions make no judgement about their
appropriateness and cannot, therefore, provide the basis for objective
assessment of efficacy of the interventions or of the service as a whole.
However, they do at least provide a standard for comparison.

Extending the role

One of the few areas in which intensive care knowledge and technology
have had a clear and beneficial effect on outcomes is in the management
of high risk surgical patients. In this group, relatively simple inter-
ventions, such as maximisation of circulating volume and oxygen
supply, have consistently led to reduced morbidity and improved
survival19"21. The fact that similar interventions applied later in the
course of established sepsis and multiple organ failure22*23 appear to be
of no benefit strengthens the argument for a pro-active approach to

British Medical Bulletin 1999;55 (No. 1)

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Epidemiology of ICM: supply vs demand

critical illness. This is relatively easy to apply (given staff, space, and time)
in surgical patients in whom the physiological insult is predictable and
semi-quantifiable, but less so in the general hospital population. However,
since a significant proportion of emergency referrals to intensive care24 or
hospital deaths25 are associated with substandard care in the ordinary
wards, it should be possible to develop systems such as emergency medical
teams26 which may facilitate earlier identification and intervention in
patients at risk of critical illness.
These developments should be accompanied by guidelines for the
discussion of treatment limitation decisions ('do not resuscitate orders')
and treatment preferences with patients and their families, in order to
reduce the number of inappropriate intensive care referrals and clarify
treatment goals before the patient becomes incompetent to do so27. Early
involvement of intensive care trained staff is an essential part of this
process as accurate information about outcomes substantially influences
patient preferences28. There seems to be an increasing awareness
amongst medical staff on both sides of the Atlantic that intensive care
has limitations, and that admission to intensive care may be subject to
societal constraints29. The problem is that, once intensive care has
started, it becomes increasingly difficult to withdraw - the cycle of com-
mitment. The SUPPORT study30 showed that providing physicians with
risk estimates of patients' mortality after admission to intensive care did
not increase the frequency of treatment limitation decisions; perhaps the
time to provide such estimates (were this possible) would be before ICU
admission, and in conjunction with the patient. Once again, this requires
the development of objective measures of prior risk, an integral
component of which is physiological reserve or susceptibility to critical
illness. How could this be achieved?

Prior risk and susceptibility to critical illness

The mainstay of mortality risk measurement in intensive care has been

the development of (predominantly physiology-based) scoring systems.
There are several problems with the concept of applying scoring systems
to patients before ICU admission. All scoring systems have been
calibrated in groups of patients who have been admitted to intensive
care, not those receiving care in non-ICU areas or those denied (or
refusing) intensive care. Scoring systems provide estimates of risk for
groups of patients, they do not provide predictions of outcome for
individuals. The other factor is that physiologically based scoring
systems measure the response to an acute illness, which is the product of
two unmeasured variables - the severity of the acute disease process and
the patient's physiological reserve. Physiological reserve includes factors

British Medial Bulletin 1999;55 (No. 1) 7

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Intensive care medicine

like cardiorespiratory fitness31"32, specific elements of nutrition such as

glutamine33>34, and genetic susceptibility or acquired resistance to disease
such as differences in cytokine responses to inflammation and infec-
tions35"38. Taking a broader view, since a proportion of mortality risk can
be attributed to lifestyle39 and place of residence40'41, it may not be too
fanciful to imagine a risk index in the future based at least in part on a
combination of genetic profiling and postcode.

Training, education and practice

Inexperienced staff fail to refer patients to intensive care in a timely

manner either because they have not recognised that the patient is deter-
iorating, or because they are reluctant to ask for help before the patient is
obviously moribund. Medical and nursing trainees in rotational training
programmes that include intensive care have more knowledge and con-
fidence to identify, stabilise and refer critically ill patients earlier, thereby
enhancing the principles of prevention. Part of this process involves
developing intensive care as a multidisciplinary speciality, with input into
undergraduate as well as postgraduate education. This is the pattern that
is gradually being adopted in Europe13. Sub-speciality ICUs (cardio-
thoracic, neurosurgical, transplant, etc.) may find that they loose oppor-
tunities for enhanced patient care and medical staffing if the adopt
isolationist policies in respect of training and collaborative medical
practice 42 . Outside the ICU, the trend toward increasing medical
specialisation in the USA has resulted in proposals for a new breed of
hospital general practitioner called a 'hospitalist'43. A better approach,
and one which would suit a European framework, would be for intensive
care to form organisational links with all disciplines involved in emer-
gency medicine, including accident and emergency, trauma, acute internal
medicine, and anaesthesia and peri-operative care. This would improve
communication and clinical management across the full range of
emergency services, and would help to foster better interdisciplinary
working relationships and training opportunities.

Organisational and managerial issues

Existing methods of funding intensive care vary from block contracts that
scarcely recognise the existence of the service, to very detailed fee-for-
service-and-item approaches. The precise method probably matters less
than whether those responsible for patient care have some involvement in,
and control over, fiscal management and resource allocation. Intensive
care services will not meet the demands made by developments in hospital

8 British Medical Bulletin 1999,55 (No. 1)

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Epidemiology of ICM: supply vs demand

medicine unless they are empowered to do so. This involves being given
authority and responsibility for delivering patient care, managing budgets,
and establishing intensive care contracts with purchasers. To achieve this
degree of independence involves support at a national as well as a local
level. Intensive care also needs to explore the possibility of collaborative
relationships with those groups most closely involved in emergency
medicine, as this will enhance opportunities for research, training, and
high quality clinical practice. The EURICUS study44, in particular, has
demonstrated substantial variability in organisational structures and
resource utilisation within European intensive care and, while some of this
may be a reflection of definitional or case mix differences, it also probably
reflects the 'Cinderella' status of intensive care in some regions.
Greater efficiency and efficacy could in theory be achieved by central-
isation of intensive care units and conversion of existing small ICUs into
intermediate care areas or 'holding' units. However, there are a number
of problems with this concept. First, large ICUs are not necessarily
cheaper or more efficient than small units. Second, there is little good
evidence that they have better outcomes. Third, the most valuable
resource in intensive care, the nursing staff, may not want to work a
long way from home. Finally, such an approach involves establishing
transport and retrieval teams, which are not inexpensive items. Rather
than centralising resources, there should be a progressive move toward
regionalisation, collaboration and resource-sharing
The training, retention, responsibilities and remuneration of ICU
nurses are of central importance to the development of the intensive care
service. Most high-quality ICUs practice collaborative care between
medical and nursing staff, but are unable to reward appropriately the
substantial responsibilities borne by highly-trained nurses. Governments
are beginning to recognise this problem, but inevitably would prefer to
see solutions achieved from within the existing healthcare budget.
Failure to invest in automated methods of bedside data capture means
that a substantial proportion of nursing time is consumed by clerical
jobs; this does nothing to maximise efficient resource use. Risk manage-
ment and quality issues are an integral part of this problem, which for
many ICUs would be diminished by the installation of any one of a
number of currently available clinical information systems.

References

Lassen HCA. A preliminary report on the 1952 epidemic of poliomyelitis in Copenhagen with
special reference to the treatment of acute respiratory insufficiency. Lancet 1953; 1: 37-41
Sage WM, Rosenthal MH, Silverman JF. Is intensive care worth it? An assessment of input and
outcome for the critically ill. Crit Care Med 1986; 14: 777-82
Atkinson S, Bihari D, Smithies M, Daly K, Mason R, McColl I. Identification of futility in
intensive care. Lancet 1994; 344: 1203-6

British Medical Bulletin 1999;55 (No 1)

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Intensive care medicine

4 Blumenthal D. Quality of health care, Part 4. The origins of the quality of care debate. N Engl
JMed 1996; 335: 1146-9
5 Halpern NA, Bettes L, Greenstein R. Federal and nationwide intensive care units and healthcare
costs: 1986-1992. Crit Care Med 1994; 22: 2001-7
6 Sirio CA, Angus DC, Rosenthal GE. Cleveland Health Quality Choice (CHQC) - an ongoing
collaborative, community-based outcomes assessment program. New Honz 1994; 2: 321-5
7 Bion J. Cost containment: Europe. The United Kingdom. New Honz 1994; 2: 341-4
8 Metcalfe MA, Sloggett A, McPerson K. Mortality among appropriately referred patient refused
admission to intensive care units. Lancet 1997; 350: 7-11
9 Le Gall JR., Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based
on a European/North American multicenter study. JAMA 1993; 270: 2957-63
10 Rapoport J, Teres D, Barnett R et at. A comparison of intensive care unit utilization in Alberta
and western Massachusetts. Cnt Care Med 1995; 23: 1336-46
11 Angus DC, Linde-Zwirble W 1, Sirio CA et at. Rural-urban differences in ICU use in Florida.
Chest 1995; 108: 105S
12 Reis Miranda D, Ryan DW, Schaufeli WB, Fidler V (Eds). Organisation and management of
intensive care. A prospective study in 12 European countries. Berlin: Springer, 1997
13 Bion JF, Ramsay G, Roussos C, Burchardi H. Intensive care training and speciality status in
Europe: international comparisons. Intensive Care Med 1998; 24: 372-7
14 Angus DC, Sirio CA, Clermont G, Bion J. International comparisons of critical care outcome
and resource consumption. In: Kvetan V, Vincent J-L, Dobb GJ (Eds). Critical Care Clinics,
April 1997. Philadelphia: Saunders, 1997
15 Jakob SM, Rothen HU. Intensive care 1980-1995: change in patient characteristics, nursing
workload and outcome. Intensive Care Med 1997; 23: 1165-70
16 Department of Health. Guidelines on admission to and discharge from intensive care and high
dependency units. London: NHS Executive, 1996
17 Society of Critical Care Medicine. Guidelines for ICU admission and discharge criteria. In:
Guidelines and Practice Parameters. Anaheim: SCCM
18 Bone RC, McElwee NE, Eubanks DH, Gluck EH. Analysis of indications for intensive care unit
admission. Clinical efficacy assessment project: American College of Physicians. Chest 1993;
104: 1806-11
19 Shoemaker WC, Appel PL, Kram HB et at. Prospective trial of supranormal values of survivors
as therapeutic goals in high-risk surgical patients. Chest 1988; 94: 1176-86
20 Bishop MH, Shoemaker WC, Appel PL et at. Prospective randomised trial of survivor values of
cardiac index, oxygen delivery, and oxygen consumption as resuscitation endpoints in severe
trauma. / Trauma 1995; 38: 780-7
21 Boyd O, Grounds RM, Bennett ED. A randomised clinical trial of the effect of deliberate
perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA
1993; 270: 2699-707
22 Hayes MA, Timmins AC, Yau EH, Palazzo M, Hinds CJ, Watson D. Elevation of systemic
oxygen delivery in the treatment of critically ill patients. N Engl J Med 1994; 330: 1717-22
23 Gattinoni L, Brazzi L, Pelosi P et at. A trial of goal-oriented hemodynamic therapy in critically
ill patients. SvO2 Collaborative Group. N Engl J Med 1995; 333: 1025-32
24 McQuillan P, Pilkington S, Allan A et at. Confidential inquiry mto quality of care before
admission to intensive care. BM] 1998; 316: 1853-8
25 Dubois RW, Brook RH. Preventable deaths: who, how often, and why? Ann Intern Med 1988;
109: 582-9
26 Lee A, Bishop G, Hillman KM, Daffurn K. The medical emergency team. Anaesth Intensive
Care 1995; 23: 183-6
27 Aarons EJ, Beeching NJ. Survey of ~do not resuscitate orders' in a district general hospital. BM]
1991; 303: 1504-6
28 Murphy DJ, Burrows D, Santilli S et at. The influence of the probability of survival on patients'
preferences regarding cardiopulmonary resuscitation. N Engl J Med 1994; 330: 545-9
29 Osborne M, Evans 1 W. Allocation of resources in intensive care: a transatlantic perspective.
Lancet 1994; 343: 778-80
30 Knaus WA, Harrell Jr FE, Lynn J et at. The SUPPORT prognostic model. Objective estimates
of survival for seriously ill hospitalized adults. Study to understand prognoses and preferences

10 British Medical Bulletin 1999;55 (No. 1)

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017
 Epidemiology of ICM: supply vs demand

for outcomes and risks of treatments. Ann Intern Med 1995; 122: 191-203
31 Older P, Smith R, Courtney P, Hone R. Preoperative evaluation of cardiac failure and ischemia
in elderly patients by cardiopulmonary exercise. Chest 1993; 104: 7014
32 Montgomery HE, Marshall R, Hemingway H et al. Human gene for physical performance.
Nature 1998; 393: 221-2
33 Griffiths RD, Jones C, Palmer TE. Six-month outcome of critically ill patients given glutamine-
supplemented parenteral nutrition. Nutrition 1997; 13: 295-302
34 Houdijk APJ, Rjjnsburger ER, Jansen J et al. Randomised trial of glutamine-enriched enteral
nutrition on infectious morbidity in patients with multiple trauma. Lancet 1998; 352: 772-6
35 Westendorp RGJ, Langermans JAM, Huizinga TWJ et al. Genetic influence on cytokine
production and fatal meningococcal disease. Lancet 1997; 349: 170-3
36 Sorensen TTA, Nielsen GG, Andersen PK, Teasdale 1 W. Genetic and environmental influences
on premature death in adult adoptees. N Engl ] Med 1988; 318: 727-32
37 Bennett-Guerrero E, Ayuso L, Hamilton-Davies C et al. Relationship of preoperative
antiendotoxin core antibodies and adverse outcomes following cardiac surgery. JAMA 1997;
277: 646-50
38 Stuber F, Petersen M, Bokelmann F, Schade U. A genomic polymorphism within the tumor
necrosis factor locus influences plasma tumor necrosis factor-alpha concentrations and
outcome of patients with severe sepsis. Crit Care Med 1996; 24: 381-4
39 Erikssen G, Liestol K, Bjornholt J, Thaulow E, Sandvik L, Erikssen J. Changes in physical
fitness and changes in mortality. Lancet 1998; 352; 759-62
40 Watt GC, Ecob R. Mortality in Glasgow and Edinburgh: a paradigm of inequality in health. /
Epidemiol Community Health 1992; 46: 498-505
41 Kunst AE, Groenhof F, Mackenbach JP, Health EW. Occupational class and cause specific
mortality in middle aged men in 11 European countries: comparison of population based
studies. EU Working Group on Socioeconomic Inequalities in Health. BMJ 1998; 316: 163642
42 Marshall MF, Schwenzer KJ, Orsina M, Fletcher JC, Durbin CG. Influence of political power,
medical provincialism and economic incentives on the rationing of surgical intensive care unit
beds. Cnt Care Med 1992; 20: 387-94
43 Wachter RM, Goldman L. The emerging role of 'hospitahsts' in the American health care
system. N Engl J Med 1996; 335: 514-7
44 Miranda DR, Ryan DW, Schaufeli WB, Fidler V (Eds). Organisation and management of
intensive care. A prospective study in 12 European countries. Berlin: Springer, 1998

British Medical Bulletin 1999;55 (No. 1) 11

Downloaded from https://academic.oup.com/bmb/article-abstract/55/1/2/475354/Epidemiology-of-intensive-care-medicine-supply

by guest
on 23 September 2017