International trends in the provision and utilisation of hospital care

This article examines the pattern of change in hospital provision in established market economies, the countries of eastern Europe and the former Soviet Union, and lower income countries. Despite the great differences between these areas, hospitals in all systems have to deal with rising expectations and, more often than not, a need to contain the costs of health care. Outside the developing countries the generic response to this has been to reduce hospital stays and to improve the efficiency of the system, a strategy which seems to be at least partly successful. The experience of the health systems in the United States and the United Kingdom shows that cost pressures and changes in healthcare delivery mean that this strategy will lead to hospital mergers and closures in the longer term.

Summary points

• Increasing rates of admission are not just a British phenomenon

• Some countries have experienced reductions in admissions

• Almost all developed countries have made substantial reductions in beds, and in many places there is still further scope for this

• Eastern Europe and the former Soviet Union still have an overdeveloped hospital sector

• Hospitals are more of a growth industry in developing countries, but care is needed if the right investment is to be made

Trends in the established market economies

Medical practice, hospital management, and technology have undergone great changes over the past 20 years. This has led to substantial shifts in demand for secondary care facilities and notable changes in the types of facilities required. Hospital costs are the largest component of health expenditure in most countries in the Organisation for Economic Cooperation and Development (OECD), and they have therefore been a key focus in the drive for increased efficiency in the health sector.¹ While this financial pressure has played a part in developments in the provision of secondary care, the influence of technology and the ability to treat large proportions of patients on a day case basis have also shaped patterns in the provision and use of hospitals.

By using data compiled for selected OECD countries a number of trends can be identified. Table 1 provides data on selected indicators for all hospitals combined and acute hospitals only for 1986 and 1995.² Admission rates for both acute and all inpatient hospitals have increased slightly in most countries, the exception being the United Kingdom, which shows a substantial increase, particularly for acute hospitals. In tandem with this increase, both average length of stay and number of beds per 1000 population have shown a consistent reduction across nearly all countries, with particularly notable reductions in numbers of beds per 1000 population for inpatient hospitals in Scandinavia (Finland, Norway, and Sweden). Interestingly, occupancy rates have remained fairly static at 75-85% for most countries. These trends indicate large increases in the throughput of hospitals: many more patients are passing through hospital beds in much shorter periods of time. In the United Kingdom the improvement in efficiency has more than compensated for the reduction in beds, and it is probable that some of the recent problems experienced with emergency admissions are a result of trying to force a large volume of work through a much smaller stock of beds. This leaves the system with little flexibility and very vulnerable to small fluctuations in demand. This improved efficiency has been attributed to a combination of factors, with changes in the management of patients, improvements in clinical techniques, and financial incentives to reduce length of stay as the prime contributors.¹

Table 1.

Selected indicators for inpatient care hospitals and acute care hospitals in established market economies²

	Admissions (% of population)						Average length of stay (days)						No of beds per 1000 population
	Inpatient care			Acute care			Inpatient care			Acute care			Inpatient care			Acute care
	1986	1995		1986	1995		1986	1995		1986	1995		1986	1995		1986	1995
Australia	17.6*	13.8		17.2*	16.2†		17.0*	14.0†		7.4*	6.7		10.5*	8.9‡		5.2*	4.3‡
Austria	22.2	24.7		19.3	23.1		13.7	10.9		12.4	7.9		10.9	9.3		6.6	6.6
Belgium	17.3	19.8†		16.3	17.7‡		16.3	11.5		10.9	7.8		9.0	7.6†		5.9	4.8†
Canada	14.8	12.5‡		14.5	–		13.9	12.2		10.8	7.5		6.7	5.1		4.4	3.6‡
Denmark	20.3	20.4†		19.7	19.7†		10.2	7.5		7.5	6.0		6.9	4.9		5.0	4.0†
Finland	22.3	25.4		15.9	20.0		19.4	11.8		8.0	5.5		13.9	9.3		4.6	4.0
France	21.6	22.7		19.4	20.3		14.9	11.2		8.0	5.9		10.3	8.9		5.6	4.6
Germany	20.6	20.7		18.0	18.0		17.5	14.2		13.5	11.4		11.0	9.7		7.6	6.9
Greece	12.0	13.5‡		–	–		12.0	8.2		–	–		5.3	5†		3.9§	–
Iceland	24.6	28†		17.6¶	18.1		21.4	–		7.2¶	5.9		15.7	–		4.3‡	3.7
Ireland	17.0	15.5		16.0	14.8		8.0	7.2		7.4	6.7		8.0	5.0		4.8	3.4
Italy	16.7	16.0†		16.4	15.5†		12.1	10.5		–	8.8		8.1	6.4		6.3§	5.3†
Japan	7.3	8.9†		–	–		54	45.5		–	–		14.9	16.2†		–	–
Korea	3.5	6.3		3.4	6.1		11.0	13.0		11.0	11.0		2.6	4.4		1.7	2.8
Luxembourg	19.0	19.4†		18.1	18.4†		19.8	15.3		11.5	9.8		12.5	11.1†		7.5	6.7†
Netherlands	11.2	11.1		10.7	10.3		34.4	32.8		12.3	9.9		11.8	11.3		4.7	3.9
New Zealand	13.0	14.1		–	–		12.9	6.9		–	–		9.3	–		8.7	–
Norway	16.6	15.0		15.1	14.5		11.3	10		9.2	6.5		16.1	13.3		4.8	3.4
Portugal	8.7	11.3		8.5	11.1		13.5	9.8		10.4	7.9		4.7	4.1		3.7	3.4
Spain	9.2	10.0†		9.0	10†		13.1	11		9.9	8.8		4.5	4.0		3.5	3.2†
Sweden	19.7	18.5		17.0	16.2		20.8	7.8		7.3	5.2		14.2	6.3		4.5	3.1
Switzerland	13.3	15.0†		13.5	–		23.7	–		14.2	12.0		–	–		6.8	6.1‡
United Kingdom	15.7	23.0		12.9	21.2		15.2	9.9		7.8	4.8		7.2	4.7		2.6	2.0
United States	14.6	12.4		13.4	11.7		9.3	8.0		7.1	6.5		5.2	4.1		3.9	3.3

^*1987. ^†1994. ^‡1993. ^§1988. ^¶1989. 

Advances in technology have enabled more and more procedures and treatments to be provided on an outpatient or day case basis. This trend towards day case care has been particularly marked in the 1990s, aided by the influence of financial reimbursement mechanisms in several countries. For example, in America insurers now generally reimburse patients for the cost of day surgery only.³ In the United Kingdom the day case rate has grown dramatically in recent years. In 1985, 17.7% of all admissions in England were treated as day cases, increasing to 38.9% in 1996-7.⁴ Whether this shift to outpatient care is a substitute for inpatient stay or represents additional work made possible by technological advances remains a matter of debate, but recent evidence in the United Kingdom suggests that a large proportion of this represents additional work facilitated by new technology and the availability of day case facilities rather than a substitute for inpatient procedures.

The current trends of increased efficiency, substitution between inpatient and outpatient care, and changes in clinical management are likely to continue. There is substantial potential to increase the proportion of outpatient care undertaken, and experience in the United States shows that the actual level of day surgery in many countries falls far short of the technical potential. Hospital planners will have to continue to keep pace with rapid changes in the requirements for secondary care; unfortunately, they will need to do this without a well developed body of research literature to support them.

Trends in the former socialist economies

The past decade has marked a turning point for healthcare and hospital systems in the former socialist economies of eastern Europe and the former Soviet Union. The Soviet healthcare system and those of the communist countries of central and eastern Europe emphasised the development of a substantial hospital infrastructure, resulting in comparatively high admission rates. Soviet and, to a lesser extent, east European medicine were also heavily controlled by medical care “normatives”—state guidelines prescribing aspects of treatment in detail, particularly minimum (but notmaximum) lengths of stay for different conditions.

Although they were never especially well funded, health systems in the former socialist economies underwent a grave financing shock after the collapse of communism. While the economies of the central and east European countries have stabilised rapidly, many of the former Soviet republics have yet to reverse the effects of collapsing household incomes and tax revenue, with inevitable consequences for healthcare spending. Many central and east European countries moved rapidly to reform their healthcare systems, often with the intention of reducing what was perceived to be an excessive emphasis on hospital care. Table 2 summarises key indicators for hospital systems in the former Soviet Union and the central and east European countries between 1986 and 1995 (the Baltic states are regarded here as central and east European countries).⁵ Overall, admission rates in the former Soviet Union have dropped substantially—primarily as a direct result of collapsing health sector finances. While trends in admission rates are mixed among central and east European countries, there has clearly not been the precipitate drop in admissions to hospital seen in the neighbouring former Soviet republics. Bed provision has dropped in both groups, but again more in the former Soviet Union, where dwindling finances have been met by successive waves of bed closures. The most telling indicator is perhaps length of stay in hospital. This has fallen in every central and east European country over the past decade, and the average duration of stay in hospital was substantially reduced—very much in line with trends in the established market economies. Yet length of stay actually rose slightly in most republics of the former Soviet Union between 1986 and 1995. There are two possible explanations for this phenomenon. Firstly, medical care norms continue to be more or less adhered to, with the effect that clinical practice has changed little for many years. Secondly, increasing shortages of drugs and supplies are leading patients to remain in hospital for longer periods, so that as a result of the economic “transition” it has become more difficult to provide effective care. Certainly, there seems reason to believe that the central and east European countries are beginning to converge on west European healthcare trends, whereas much of the former Soviet Union seems to be stuck in a trap that combines obstinate adherence to obsolete practice with continuously collapsing finances.

Table 2.

Selected indicators for inpatient care and acute care hospitals in former socialist economies⁵

	Admissions (% of population)						Average length of stay (days)						No of beds per 1000 population
	Inpatient care			Acute care			Inpatient care			Acute care			Inpatient care			Acute care
	1986	1995		1986	1995		1986	1995		1986	1995		1986	1995		1986	1995
Former Soviet Union
Armenia	16.1	7.5		15.8	7.3		16.4	15.2		14.6	13.6		9.0	7.6		7.7	6.9
Azerbaijan	16.6	6.8		–	–		17.5	17.7		–	–		9.8	10.0		–	–
Belarus	27.1	24.9		–	–		15.5	15.2		–	–		13.3	11.7		–	–
Georgia	16.7	5.0		16.3	4.9		14.9	13.4		13.0	10.9		10.0	6.3		8.7	5.6
Kazakhstan	24.8	17.1		24.0	16.5		16.5	17.1		13.6	14.5		13.6	11.6		11.5	10.8
Kyrgyzstan	25.0	17.0		24.1	16.3		15.5	14.7		13.1	12.4		12	8.6		9.8	7.0
Moldova	25.7	20.6		24.3	19.8		15.7	17.5		13.3	15.5		12.4	12.2		10.1	10.3
Russia	–	21.2		–	–		–	16.8		–	13.6		13.0	11.8		–	–
Tajikistan	21.8	12.8		21.3	12.6		14.9	15.0		14.1	13.4		10.6	8.0		9.0	7.3
Turkmenistan	20.7	–		20.2	–		15.6	–		13.5	–		11.4	–		9.4	–
Ukraine	26.9	21.9		25.6	20.8		15.8	16.8		13.0	14.6		12.8	11.9		10.3	9.8
Uzbekistan	25.3	18.6		–	–		15.0	14.2		–	–		12.0	8.3		–	–
Average (% change)	24.4	19.9 (−18)		23.8	17.7 (−26)		16.3	16.4 (1)		13.6	13.8 (1)		13.1	11.2 (−15)		10.2	9.3 (−9)
Central and eastern Europe
Albania	–	9.0		–	–		14.3	8.2		–	–		4.0	3.2		–	–
Bosnia-Hercegovina	10.2	–		8.7	–		14.3	–		10.6	–		4.8	–		3.7	–
Bulgaria	20.3	17.7		–	15.7		14.1	13.6		–	11.6		9.4	10.4		–	–
Croatia	15.3	13.4		–	11.9		16.2	13.2		12.7	10.2		7.4	5.8		4.7	3.8
Czech Republic	19.2	20.1		17.9	18.7		16.4	12.8		13.4	10.2		10.1	9.2		8.6	7.2
Estonia	21.4	18		20.2	16.7		17.5	12.7		14.5	10.5		12.2	8.1		9.7	6.4
Hungary	21.0	23.3		18.3	20.5		13.3	10.8		10.4	8.6		9.8	9.0		6.8	6.3
Latvia	26.4	21.3		–	–		17.5	14.9		–	–		14.2	11.0		–	–
Lithuania	22.2	20		–	–		18.2	14.8		–	–		12.8	10.8		–	–
Poland	–	–		–	–		–	10.8		–	–		6.6	6.3		–	–
Romania	23.5	20.6		–	–		11.1	11.0		–	–		8.9	7.6		–	–
Slovakia	16.1	18.3		–	–		13.2	11.7		–	–		7.4	7.5		–	–
Slovenia	16	15.7		15.4	15.1		12.1	10.4		10.5	8.8		6.3	5.8		5.3	4.8
Former Yugoslav Republic Macedonia	10.1	9.9		9.5	9.2		14.7	14.3		11.1	10.3		5.5	5.5		3.9	3.8
Average (% change)	18.9	19.1 (1)		16.3	17.9 (10)		13.5	11.7 (13)		11.8	9.7 (−18)		7.7	7.3 (−5)		6.2	6.1 (−2)

Trends in developing countries

There are no comprehensive databases providing detailed comparative data on hospital provision in developing countries. Tables 3 and 4 provide data from the World Bank’s Social Indicators of Development series.⁶ They present the number of beds per 1000 population for 125 countries for which data were available, for a year between 1970 and 1975, and a year between 1989 and 1994. These measures are clearly far from perfect, but they do at least provide a sense of the direction in which hospital bed provision is moving—namely, downwards for most of the world’s population. Yet this downward trend stems from different causes. In the countries with high incomes, it is driven by a reduction in the absolute number of beds. Despite the rhetoric demanding that health investment be focused on primary care, the absolute number of hospital beds in developing countries has tended to increase since the early 1970s. It has, however, done so at a slower rate than rates of population growth, so that the bed:population ratio has fallen.

Table 3.

Provision of hospital beds in 125 countries, by region

Region	No of countries	No of beds per 1000 population
		1970-5	1989-94	% change
East Asia	11	2.36	2.85	17
Former Soviet Union and central and eastern Europe	21	9.42	9.66	2
High income countries	24	9.37	7.46	−26
Latin America and Caribbean	24	3.73	2.84	−31
Middle East and North Africa	11	1.82	1.91	5
South Asia	6	0.82	0.80	−2
Sub-Saharan Africa	25	1.58	1.21	−31

Table 4.

Provision of beds in 125 countries, by income group

Income group	No of countries	No of beds per 1000 population
		1970-75	1989-94	% change
High	24	9.37	7.46	−26
Upper middle	15	4.34	4.31	−1
Lower middle	40	5.02	4.69	−7
Low	46	2.67	2.49	−7

We could find no data to illustrate trends over time in admission rates, length of stay, or other dimensions of hospital activity in developing countries, although a summary of one off data from the late 1980s from certain countries has been published. ⁷ The box provides a flavour by presenting 1996 data from Côte d’Ivoire. Problems facing hospital systems in developing countries frequently include excessive dominance of tertiary hospitals in large cities to the detriment of basic primary and secondary care services; lack of drugs and supplies, especially at first level referral hospitals, compromising the quality and effectiveness of care; extreme difficulties in recruiting and retaining staff in remote rural areas; and, in many countries, an ever increasing burden on the hospital system from adults with AIDS.

The hospital sector in Côte d’Ivoire

Côte d’Ivoire is a French speaking country of some 14.6 million people in West Africa. Although classified by the World Bank as a low income country, with a gross national product per head of $610 (£381) in 1995, it is not particularly poor by sub-Saharan African standards.⁶

In 1996, the population of Côte d’Ivoire was served by 6727 beds in the public network of 68 hospitals, with a further 524 beds provided by 25 private clinics—an overall bed:population ratio of 0.5 beds per 1000 population (see table 5).

Table 5 General, regional, and university hospitals in Côte d’Ivoire

Public hospitals (n=68)	Beds (n=6727)	Admissions (n=223 299)	Surgical operations, 1995 (n=24 809)	x Ray films, 1995 (n=117 988)
Hôpitaux Généraux (n=56)	3 490	114 074	8 302	4 300
Centres Hospitaliers Regionaux (n=8)	1 252	60 268	8 465	14 437
Centres Hospitaliers Universitaires (n=4)	1 985	48 957	8 042	99 251

Nearly 30% of all beds are located in the capital, Abidjan, which has 17% of the population, in three of the country’s four tertiary hospitals. The eight regional hospital centres are distributed in provincial cities across the country, but the small general hospitals provide the bulk of inpatient care to the mass of the population. They do so virtually without access to x ray facilities—a diagnostic technology regarded as an essential part of basic secondary care in wealthy countries, but to which most Ivoireans who live outside Abidjan have little or no practical access.⁹

Conclusions

The global trend of increasing admission rates and falling bed numbers may continue for many years, but clearly the extent to which it can be sustained is limited. The contradiction between falling bed numbers and the pressure on the hospital system from increasing admissions is probably a factor contributing to the institution of a review of bed numbers by the UK secretary of state for health. If there is, however, one lesson to be drawn from the variation in the extent of provision between similar countries and the often rapid year on year decline it is that there is probably no such thing as the “right” extent of bed provision, and if there were it would be incorrect within a year of being set. The questions that need to be answered are how to create flexibility to manage uncertainty and how to be able to have the capacity to cope with surges in demand without creating the potential for further increased admissions through the operation of supplier induced demand.

International comparisons—handle with care

Cross national comparisons of key indicators must be treated cautiously, keeping in mind the limitations of the data. There is no international consensus on concepts, definitions, and methods of calculation for the compilation of health statistics, and health concepts are not uniform across OECD countries.⁸ In other words, what constitutes a “bed” or an “admission” (and whether or not either is “acute” or “long stay”) may vary considerably. For example, in some countries these statistics may count beds simply by counting metal frames with mattresses, whether or not they are in use; in others, a “bed” is counted only if it is staffed and operational. However, comparisons can shed light on important international trends in the way care is delivered, and how hospital systems are changing and evolving over time.

Figure.

MARY EVANS PICTURE LIBRARY

St Luke’s Hospital, London