The variability of critical care bed numbers in Europe
To quantify the numbers of critical care beds in Europe and to understand the differences in these numbers between countries when corrected for population size and gross domestic product.
Prospective data collection of critical care bed numbers for each country in Europe from July 2010 to July 2011. Sources were identified in each country that could provide data on numbers of critical care beds (intensive care and intermediate care). These data were then cross-referenced with data from international databases describing population size and age, gross domestic product (GDP), expenditure on healthcare and numbers of acute care beds.
We identified 2,068,892 acute care beds and 73,585 (2.8 %) critical care beds. Due to the heterogeneous descriptions of these beds in the individual countries it was not possible to discriminate between intensive care and intermediate care in most cases. On average there were 11.5 critical care beds per 100,000 head of population, with marked differences between countries (Germany 29.2, Portugal 4.2). The numbers of critical care beds per country corrected for population size were positively correlated with GDP (r2 = 0.16, p = 0.05), numbers of acute care beds corrected for population (r2 = 0.12, p = 0.05) and the percentage of acute care beds designated as critical care (r2 = 0.59, p < 0.0001). They were not correlated with the proportion of GDP expended on healthcare.
Critical care bed numbers vary considerably between countries in Europe. Better understanding of these numbers should facilitate improved planning for critical care capacity and utilization in the future.
KeywordsIntensive care Bed numbers Demand Supply
The need for critical care capacity worldwide is increasing . This has been described in the USA, where it is recognized that future provision of critical care is unlikely to be able to meet the estimated demands . This potential shortfall has also been described in other countries, such as Norway, as a result of changes in population demographics . Similar patterns are being described in many other countries, although most have been unable to accurately quantify the problem. The future increase in demand is due to a number of factors that include significant changes in the size and age of the population, together with increasing prevalence of relevant comorbidities and changing perceptions as to what critical care can offer [4, 5].
The identification of mechanisms to prevent this mismatch developing needs to take place with some urgency. Several factors have to be taken into account, all of which interact with each other at a variety of levels. Unless admission and referral practices change, the increased future demand can only be met by an increase in total capacity [2, 6]. Without an increase in capacity there will need to be rationing or triaging of available resource to ensure that patients who are most likely to benefit can receive the care they need . Although part of this change may be met by increased provision of outreach and intermediate care , there will also need to be an increase in the number of critical care beds and hence also an increase in the numbers of appropriately skilled healthcare professionals to care for the increased number of patients.
To plan for these changes there is a need to better understand the current situation of critical care bed availability [4, 9, 10, 11]. Although several countries publish the numbers of beds provided, little is known about how this varies between countries even within a confined geographical region such as Europe. This study therefore aims to identify the total numbers of critical care beds for each country in Europe and to adjust the bed numbers to the population in order to illustrate the differences in resource provided for this group of patients.
Materials and methods
This was an observational study assessing the numbers of adult critical care beds in each country in Europe between July 2010 and July 2011. For the purposes of this study critical care includes intensive care (ICU) and intermediate care beds (IMCU). To be included in this study, the bed had to be open, staffed and fulfil any relevant national criteria, where available. The following were excluded from the data collection: private healthcare providers, neonatal and paediatric intensive care beds, coronary care, stroke and pure renal units. The numbers of beds in each country were obtained by assessing data from reliable governmental sources (websites and contacts), national societies with a declared interest in intensive care medicine, national training boards, faculties or colleges and national registries where appropriate. Data obtained were then cross-referenced with the national council representative for the European Society of Intensive Care Medicine (ESICM) and other personal contacts with knowledge of their country, in order to provide face validity for the numbers obtained. In countries where, following this approach, data were still not forthcoming, personal contacts were used and numbers were estimated according to a local sample assessment.
Data describing the total population of each country were identified from a series of publicly available databases. These included the European Commission database (Eurostat) (ec.europa.eu/Eurostat), the World Health Organization (WHO) regional office for Europe, the Central Intelligence Agency (CIA) World Factbook (https://www.cia.gov/library/publications/the-world-factbook/) and the Organisation for Economic Co-operation and Development (OECD). Data were analysed using Graphpad Prism (version 5.1a) and are presented as numbers with a percentage. Linear regression analysis was performed in order to assess likely associations. A p value of less than 0.05 was taken to be significant.
In many countries, readily accessible data with regards to the provision of critical care bed numbers were not available. In some countries, for instance the UK, there were governmental census data. In other countries, data were available through national societies (for instance, Germany). In others no data were found, and local clinicians had to count the beds themselves (Portugal). There were marked differences in how critical care services were set up between countries, with some having separate intermediate and intensive care (Table ESM 1), whilst others manage both flexibly within single services. Some countries also included higher levels of care in acute general wards, for instance the Czech Republic.
Descriptors of population size, economic strength and health expenditure in European countries
Total size of population
Gross domestic product (GDP) ($millions)b
Gross domestic product (GDP) per inhabitant ($)b
Total expenditure on health as a % of GDPc
% of population over 65 years of agea
Data describing numbers of adult acute care, intermediate care and intensive care beds per European country
Acute care bedsa
Acute care beds/100,000 population
Intermediate care (IMCU) beds
Intensive care (ICU) beds
Critical care beds
ICU and IMCU beds/100,000 population
ICU beds as % of acute care beds
GDP ($million)/ICU beds
In this work we found marked heterogeneity in the numbers of critical care beds between European countries, even when corrected for population size and age distribution, gross domestic product, expenditure on healthcare and numbers of total acute care beds. The differences in provision can be exemplified by the fact the Germany has 6.9 times the number of intensive care beds compared with Portugal per head of population.
The artificial split of critical care beds into either intermediate or intensive care varies widely across Europe despite the move to standardize descriptions across the region [12, 13]. This lack of a consistent definition reduces our ability to compare clinical practice and organizational models across borders and therefore will not help individual countries to build the case for additional capacity in future years. We believe that it would be beneficial to have a European standard definition of exactly what an intensive care bed is, which could then be implemented within the different countries. This definition could include factors related to the unit’s ability to address organ dysfunction/failure, availability of beds throughout the day and week, patient/nurse and patient/doctor ratios, severity of illness and the operative rather than the planned mean level of care of the ICU .
The overall number of critical care beds for Europe was 11.5/100,000 head of population. This is in marked contrast to the number for the USA, which Carr found to be 28/100,000 in 2010 . The heterogeneity of the data between European countries is consistent with the findings reported by other groups. Wunsch and colleagues  presented similar data although only on a very limited number (six) of European countries in addition to a number of other non-European countries. The provision of intensive care beds that they found within the European region was very similar to the numbers presented in this study, despite their collection of data being from 2005, 5 years earlier.
An interesting question that arises from examination of this data is how the different countries cope with the widely differing levels of critical care capacity. Presumably, in a grossly homogeneous geographical and developed region such as Europe, one would expect that comparable numbers of patients would develop acute critical illness in the different countries. One would hence expect that the different levels of provision should have a major impact on practice and hence presumably outcomes [6, 16, 17, 18, 19]. Again comparing Portugal and Germany, it is impossible that Portugal is able to admit the same amount of patients to critical care as in Germany. The implication must therefore be that either patients in Portugal with need for critical care are unable to get it or that Germany overprovides intensive care for its population. Only further analysis of data that describe provision and practice of critical care in detail across countries will enable us to answer these questions. As a start, comparison of data from European countries with established national registries of intensive care could give more insight into such details. At present such registries are operative in England, Scotland, Norway, Finland and Sweden, The Netherlands, and Austria. With the exception of Austria, these are all countries with a relatively low number of ICU beds per population unit, as can be seen in Fig. 1.
We have been able to demonstrate that there are still major differences within Europe regarding provision of critical care services. These differences are too large to be explained purely by differences in the characteristics of the populations and are inadequately explained by the economic strength of the country. In that respect, the three wealthiest countries in Europe [measured in GDP ($)/inhabitant]: Luxembourg (107,358), Norway (83,936), and Switzerland (67,110) have respectively 21, 8, and 11 ICU beds/100,000 population. It seems likely that the healthcare models present in each country have a major impact on the development and prioritization of this resource. This is likely to reflect a variety of factors that range from specialty status, bed and patient models and bed utilization (admission and discharge criteria) protocols. In addition, the staffing of other hospital wards may also play a major role. We restricted this study to the provision of beds through the public healthcare systems, excluding private providers. There may, therefore, be an underestimation of numbers in some countries due to the missing private sector.
More research is urgently needed to understand how the differing numbers of critical care beds impact on practice and ultimately on patient outcomes. If the need for these beds continues to grow, then the most effective and cost-efficient use of this level of care must be developed in order for most countries to be able to afford this level of provision of healthcare.
ESICM: Claudia Arena and Giulia Evolvi, Andorra: Antoni Margarit, Austria: Philipp Metnitz, Austrian Center of Documentation and Quality Assurance in Intensive Care Medicine (ASDI), Belgium: Federal Government. Patrick Ferdinande and Eric Hoste, Bulgaria: National Center of Health Informatics-Public Health Statistics, Croatia: Ino Husedzinovich, president of the Croatian Society of Intensive Care Medicine, Czech Republic: Ivan Novak, scientific secretary of the Czech Society of Anesthesia and Intensive Care Medicine, Cyprus: Department of Medical and Public Health Service-Maria Psalti and Theodorus Kyprianou, Denmark: National Board of Health, Estonia: Indrek Rätsep, from the database of the Estonian Society of Anaesthesiologists, Finland: Ville Pettila-Finnish Society of Intensive Care Medicine, France: Agence technique de l’information hospitalière (ATIH), Germany: German Association of Anaesthesiology and Intensive Care, Greece: Apostolos Armaganidis, president of the Greek Society of Intensive Care Medicine, Hungary: Akos Csomos for the Hungary national health fund database, Iceland: Kari Hreinsson, president of the Icelandic Society of Anesthesiology and Intensive Care Medicine, Ireland: Patrick Seigne, The Prospectus Report (A Review of Adult Critical Care Services in Ireland, 2009), Italy: Ministry of Health, Latvia: Indulis Vagas (president of the National Society of Anesthesiology and Reanimation of Latvia), Lithuania: Jurate Sypilaite-Lithuanian Society of Intensive Care Medicine, Luxembourg: Max Martin, Malta: Carmel Abela, The Netherlands: Jozef Kesecioglu and Hans van der Spoel, Norway: Hans Flaatten, Poland: Krystof Kusza-Polish Ministry of Health, Portugal: Rui Moreno; Colégio de Medicina Intensiva da Ordem dos Médicos; Sociedade Portuguesa de Medicina Intensiva, Romania: National Center for Statistic of the Ministry of Health-Ioana Grigoras and Natalia Hagau, Slovakia: Jozef Firment Slovakian Society of Intensive Care Medicine, Slovenia: Slovenian Health Statistic Yearbook, Spain: Cristobal Leon Gil and the SEMICYUC and Cesar Aldecoa, Sweden: Goran Karlstrom, Switzerland: Hans Rothen, National coordinator, data from the Swiss Society of Intensive Care Medicine, UK: Department of Health (2010), Scottish Intensive Care Society, Linda Mulholland, Critical Care Networks of Wales and Northern Ireland.
Conflicts of interest
- 2.Angus DC, Kelley MA, Schmitz RJ, White A, Popovich J, for the Committee on Manpower for Pulmonary and Critical Care Societies (COMPACCS) (2000) Current and projected workforce requirements for care of the critically ill and patients with pulmonary disease: can we meet the requirements of an aging population? JAMA 284:2762–2770Google Scholar
- 5.Piers RD, Azoulay E, Ricou B, Dekeyser Ganz F, Decruyenaere J, Max A, Michalsen A, Maia PA, Owczuk R, Rubulotta F, Depuydt P, Meert AP, Reyners AK, Aquilina A, Bekaert M, Van den Noortgate NJ, Schrauwen WJ, Benoit DD (2011) Perceptions of appropriateness of care among European and Israeli intensive care unit nurses and physicians. JAMA 306:2694–2703PubMedCrossRefGoogle Scholar
- 8.Hillman K, Chen J, Cretikos M, Bellomo R, Brown D, Doig G, Finfer S, Flabouris A, Merit Study Investigators (2005) Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. Lancet 365:2091–2097 Google Scholar
- 9.Wunsch H, Angus DC, Harrison DA, Collange O, Fowler R, Hoste EAJ, de Keizer NF, Kersten A, Linde-Zwirble WT, Sandiumenge A, Rowan KM (2008) Variation in critical care services across North America and Western Europe. Crit Care Med 36(2787–2793):e2781–e2789Google Scholar
- 13.Valentin A, Ferdinande P, ESICM Working Group on Quality Improvement (2011) Recommendations on basic requirements for intensive care units: structural and organizational aspects. Intensive Care Med 37:1575–1587Google Scholar