Content uploaded by Madiha Hashmi
Author content
All content in this area was uploaded by Madiha Hashmi on Feb 28, 2020
Content may be subject to copyright.
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Critical Care Medicine www.ccmjournal.org 1
DOI: 10.1097/CCM.0000000000004222
1Fast and Chronic Programmes, Alexandra Hospital, National University
Health System, Singapore.
2Division of Respiratory and Critical Care Medicine, University Medicine
Cluster, National University Hospital, National University Health System,
Singapore.
3General Intensive Care Unit and Emergency Department, United Hos-
pital Ltd, Dhaka, Bangladesh.
4Division of Critical Care Medicine, Tata Memorial Hospital, Homi Bhabha
National Institute, Mumbai, India.
5Faculty Of Medicine, Department of Anesthesiology and Intensive Care,
Padjadjaran University, Dr. Hasan Sadikin National Referal Hospital
Bandung, Bandung, Indonesia.
6Adult Intensive Care Unit, Mahosot Hospital, Vientiane, Lao PDR.
7Anaesthesia and Critical Care Department, Mongolian National Univer-
sity of Health Science, Ulaanbaatar, Mongolia.
8Department of Anaesthesiology and ICU, Yangon General Hospital,
University of Medicine 1, Yangon, Myanmar.
9Department of Anesthesia and Intensive Care, Kathmandu Medical Col-
lege Teaching Hospital, Kathmandu, Nepal.
10Department of Anaesthesiology, Aga Khan University, Karachi,
Pakistan.
11Acute and Critical Care Institute, The Medical City, Pasig City, Philip-
pines.
12Network for Improving Critical Care Systems and Training, Colombo, Sri
Lanka.
Critical Care Bed Capacity in Asian Countries
and Regions
Jason Phua, MBBS, MRCP(UK)1,2; Mohammad Omar Faruq, MBBS, FACP, FACEP, FCPS, FCCM3;
Atul P. Kulkarni, MBBS, MD (Anaesthesiology), FICCM, PGDHHM4;
Ike Sri Redjeki, SpAn, KIC, KMN, MKes5; Khamsay Detleuxay, MD6;
Naranpurev Mendsaikhan, MD, PhD7;
Kyi Kyi Sann, MBBS, DA, MMed Sc (Anaesthesiology), Dr Med Sc (Anaesthesiology)8;
Babu Raja Shrestha, MBBS, MD9; Madiha Hashmi, FFARCSI10;
Jose Emmanuel M. Palo, MD, FCCM11; Rashan Haniffa, MD, PhD12; Chunting Wang, MD13;
Seyed Mohammad Reza Hashemian, MD, FCCM14; Aidos Konkayev, MD, PhD15,16;
Mohd Basri Mat Nor, MBBCh, MMed Anaesthesiology, EDIC17;
Boonsong Patjanasoontorn, MD, FCCP, FCCM18; Khalid Mahmood Khan Nafees, MBBS, BSc, MMed19;
Lowell Ling, MBBS, MPhil, MRCP(UK)20; Masaji Nishimura, MD, PhD21;
Maher Jaffer Al Bahrani, MBChB22; Yaseen M. Arabi, MD, FCCP, FCCM, ATSF23;
Chae-Man Lim, MD, PhD, FCCM24; Wen-Feng Fang, MD25,26; for the Asian Analysis of Bed Capacity in
Critical Care (ABC) Study Investigators, and the Asian Critical Care Clinical Trials Group
Copyright © 2020 by the Society of Critical Care Medicine and Wolters
Kluwer Health, Inc. All Rights Reserved.
13Department of Critical Care Medicine, Shandong Provincial Hospital,
Jinan, China.
14Chronic Respiratory Diseases Research Center, National Research Insti-
tute of Tuberculosis and Lung Diseases, Shahid Beheshti University of
Medical Sciences, Tehran, Iran.
15Anaesthesiology and Intensive Care Department, Astana Medical Uni-
versity, Nur-Sultan, Kazakhstan.
16Anaesthesia and ICU Department, Institution of Traumatology and Or-
thopedics, Nur-Sultan, Kazakhstan.
17International Islamic University Malaysia Medical Centre, Kuantan, Malaysia.
18Division of Respiratory and Critical Care Medicine, Department of Internal
Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand.
19RIPAS Hospital, Bandar Seri Begawan, Brunei, Darussalam.
20Department of Anaesthesia and Intensive Care, The Chinese University
of Hong Kong, Prince of Wales Hospital, Hong Kong, China.
21Tokushima Prefectural Central Hospital, Tokushima, Japan.
22Department of Anesthesia and Critical Care, Royal Hospital, Muscat,
Oman.
23King Saud Bin Abdulaziz University for Health Sciences, King Abdullah
International Medical Research Center, King Abdulaziz Medical City, Ri-
yadh, Kingdom of Saudi Arabia.
24Department of Pulmonary and Critical Care Medicine, Asan Medical
Center, Seoul, South Korea.
25Division of Pulmonary and Critical Care Medicine, Department of Internal
Medicine, Kaohsiung Chang Gung Memorial Hospital, Chang Gung Uni-
versity College of Medicine, Kaohsiung, Taiwan.
26Department of Respiratory Care, Chang Gung University of Science and
Technology, Chiayi, Taiwan.
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Phua et al
2 www.ccmjournal.org XXX 2020 • Volume XX • Number XXX
Objective: To assess the number of adult critical care beds in
Asian countries and regions in relation to population size.
Design: Cross-sectional observational study.
Setting: Twenty-three Asian countries and regions, covering
92.1% of the continent’s population.
Participants: Ten low-income and lower-middle–income econo-
mies, five upper-middle–income economies, and eight high-
income economies according to the World Bank classification.
Interventions: Data closest to 2017 on critical care beds, in-
cluding ICU and intermediate care unit beds, were obtained
through multiple means, including government sources, national
critical care societies, colleges, or registries, personal contacts,
and extrapolation of data.
Measurements and Main Results: Cumulatively, there were 3.6
critical care beds per 100,000 population. The median number
of critical care beds per 100,000 population per country and
region was significantly lower in low- and lower-middle–income
economies (2.3; interquartile range, 1.4–2.7) than in upper-mid-
dle–income economies (4.6; interquartile range, 3.5–15.9) and
high-income economies (12.3; interquartile range, 8.1–20.8)
(p = 0.001), with a large variation even across countries and re-
gions of the same World Bank income classification. This number
was independently predicted by the World Bank income clas-
sification on multivariable analysis, and significantly correlated
with the number of acute hospital beds per 100,000 population
(r2 = 0.19; p = 0.047), the universal health coverage service cov-
erage index (r2 = 0.35; p = 0.003), and the Human Development
Index (r2 = 0.40; p = 0.001) on univariable analysis.
Conclusions: Critical care bed capacity varies widely across Asia
and is significantly lower in low- and lower-middle–income than
in upper-middle–income and high-income countries and regions.
(Crit Care Med 2019; XX:00–00)
Key Words: critical care; intensive care units; bed capacity; Asia;
population
Critical illness knows no boundaries and affects all
human beings at some time in their lives (1). Although
demand for critical care is expected to increase as pop-
ulations age and expectations rise, its reach is patchy and its
impact inconsistent. Indeed, there exists a huge variability in
the availability of critical care across continents and countries
(2). Careful analysis of this diversity allows a better under-
standing of the effect of different healthcare systems on patient
outcomes, and steps in this direction have been made in the
developed world. Most of these studies have revolved around
North America, Europe, Australia, and New Zealand (1–8).
Asia is Earth’s largest continent. It is estimated—albeit
based mainly on extrapolations of data from the West—that
Asia accounts for at least half of the cases of sepsis, acute res-
piratory distress syndrome, and mechanical ventilation in the
world (1). Given the heterogeneous distribution of wealth
across Asia’s mix of low-, middle-, and high-income countries
(9), it is likely to see a larger variability of critical care beds
than that in continents like Europe (4). Studies on critical
care bed capacity, both in resource-limited and resource-rich
parts of Asia, are however lacking (1, 10–15). The aim of the
current study is therefore to assess the number of adult crit-
ical care beds in Asian countries and regions in relation to
population size.
MATERIALS AND METHODS
Countries and Regions
We conducted this Asian Critical Care Clinical Trials Group-
endorsed cross-sectional study (the Asian Analysis of Bed Ca-
pacity in critical care or ABC study) between September 2017 and
September 2019. We adapted methods by Rhodes et al (4), who
examined critical care bed capacity in 31 European countries, to
facilitate comparisons between Asia and Europe. Although var-
ious definitions of Asian countries exist, we referred mainly to
the United Nations Asia-Pacific Regional Group (eTable 1, Sup-
plemental Digital Content 1, http://links.lww.com/CCM/F287)
(16). We sought to invite a national coordinator from each of
these countries and regions who is prominent in the field of
critical care and/or the national critical care society, and/or pas-
sionate in critical care research. Approval by institutional review
boards was obtained where appropriate, according to local regu-
lations in each participating country and region.
Critical Care Beds
Each national coordinator provided information on critical care
beds through a questionnaire (eAppendix, Supplemental Dig-
ital Content 1, http://links.lww.com/CCM/F287). Critical care
beds primarily not only refer to beds in an ICU which are staffed
but also includes beds in an intermediate care unit (IMCU) that
meet selected criteria. The definition of critical care beds varies
across countries and regions and even within countries (17).
Units known as ICUs in resource-limited settings may not be
considered ICUs in the developed world (10, 11). To ensure ap-
plicability across countries and regions, we adapted the defini-
tion of an ICU from that of a task force convened by the World
Federation of Societies of Intensive and Critical Care Medicine
(WFSICCM): an ICU is based in a defined geographic area of
a hospital and an organized system for the provision of care
to critically ill patients that provides intensive and specialized
medical and nursing care, an enhanced capacity for monitor-
ing, and multiple modalities of physiologic organ support to
sustain life during a period of acute organ system insufficiency
(18). Because some countries and regions differentiate IMCUs
(various names include high-dependency units, intermediate
care areas/units, step-up units, and step-down units) and ICUs
(4, 19, 20), we considered beds in such units in the definitions of
critical care beds if the units have at least the capability of level 1
ICUs as defined by the WFSICCM task force, that is, a dedicated
space within the hospital with a higher nurse-to-patient ratio
than a regular ward, equipped to monitor vital signs and oxygen
saturation intensively and the electrocardiogram continuously,
and to provide noninvasive ventilation, short-term invasive me-
chanical ventilation, or simple mechanical ventilation for stable
chronically ventilated patients. To allow comparisons with
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Clinical Investigation
Critical Care Medicine www.ccmjournal.org 3
similar studies, we excluded beds from coronary care, stroke,
and pure renal units, as well as PICUs and neonatal ICUs which
usually have a very different setup and staffing model from
adult units (4). We recorded whether the ICUs and IMCUs were
from the public or private sector, and whether they were located
in urban or rural areas as generally understood in the respective
countries and regions.
To ensure as complete a review as possible, a mix of sources
was used to obtain the data, with the following suggested order
of preference: official government census or sources, through
websites or contacts; national critical care societies, colleges,
or registries; counting through personal contacts with links to
ICUs and IMCUs; estimation through extrapolation of data
and relevant ratios (12); and other means appropriate for
each country and region. Where appropriate, data from sev-
eral sources were added to compute the total number of beds.
We sought data that are closest to the year 2017 and recorded
methodologic details, together with qualifiers, uncertainties,
and notes on unavailable data.
Other Variables
We collected other data to assess associations with critical care
bed capacity. Each national coordinator provided information
on acute hospital beds using methods similar to the counting
of critical care beds. Acute hospitals beds and acute hospitals
refer to beds and hospitals which are operational and staffed for
acute care. We excluded beds and hospitals which are primarily
for rehabilitative or long-term care. In countries and regions
where available census provides the total number of hospital
beds regardless of whether they are for acute, rehabilitative, or
long-term care, we used this number, albeit with qualifiers.
We categorized the countries and regions into low and
lower-middle income, upper-middle income, and high-
income, using to the World Bank classification which is based
on gross national income (GNI) per capita (9). We combined
low-income and lower-middle income countries because Nepal
was the only low-income country which participated (the oth-
ers in Asia are Afghanistan, North Korea, Syria, and Yemen).
We recorded each country and region’s population (number
of people living within, including both citizen and non-citizen
residents, and the percentage of the population who are ≥ 65
yr old) using the Central Intelligence Agency World Factbook
(21); gross domestic product (GDP) per capita using data from
the World Bank (22); health expenditure using data from the
World Health Organization (23); the universal health coverage
service coverage index (a scale of 100 which measures coverage
of essential services across four areas: reproductive, maternal,
neonatal, and child health; infectious disease; noncommuni-
cable diseases; and service capacity and access) from the World
Health Organization and the World Bank (24); and the Human
Development Index (a scale of 1.000 which takes into account
life expectancy at birth, years of schooling, and GNI per capita)
according to the United Nations (25).
Statistical Analysis
We present data as numbers, percentages, medians, and in-
terquartile range (IQR). We used the Mann-Whitney U test
and the Kruskal-Wallis test to compare nonparametric data
between countries and regions in different World Bank in-
come groups. We used both univariable and multivariable
linear regression analyses, including evaluating for multi-
collinearity, to assess associations of the collected variables
with the number of critical care beds per 100,000 popula-
tion in countries and regions. We consider a p value of less
than 0.05 as statistically significant. We used IBM SPSS Sta-
tistics for Windows (version 20.0; IBM Corp, Armonk, NY).
RESULTS
Enrolled Countries and Regions
Out of 44 Asian countries and regions, five were not contacted
because of an absence of contacts within, 12 were contacted
but did not respond, and four replied but did not provide data
eventually (eTable 2, Supplemental Digital Content 1, http://
links.lww.com/CCM/F287). Thus, 23 countries and regions
which make up 92.1% of the population of Asia (3,952,975,069
out of 4,292,142,884) were enrolled: 10 low-income and lower-
middle–income ones, five upper-middle–income ones, and
eight high-income ones (21). Table 1 provides details of each
country and region.
Critical Care Beds
Although most countries and regions have both ICUs and
IMCUs, Brunei, China, Kazakhstan, Mongolia, South Korea,
and Sri Lanka do not have IMCUs (eTable 3, Supplemental
Digital Content 1, http://links.lww.com/CCM/F287). All coun-
tries and regions have official definitions of an ICU except
Mongolia, Myanmar, Pakistan, and the Philippines (eTable 4,
Supplemental Digital Content 1, http://links.lww.com/CCM/
F287). Only Laos (where ICUs in main cities and provinces are
very different) and the Philippines (where critical care areas in-
clude cardiac catheterization laboratories, operating theaters,
and delivery rooms) have a definition/understanding of an
ICU which is not in keeping with the WFSICCM definition.
Although only Hong Kong, India, Indonesia, Japan, Oman,
Taiwan, and Thailand have official definitions of an IMCU, all
countries and regions with IMCUs are able to describe their
usual understanding of an IMCU (eTable 5, Supplemental
Digital Content 1, http://links.lww.com/CCM/F287). Data on
the number of IMCU beds in Indonesia, India, Laos, Malaysia,
Pakistan, Saudi Arabia, and Thailand are unavailable (eTable 3,
Supplemental Digital Content 1, http://links.lww.com/CCM/
F287). Although a mix of methods was used to obtain data on
critical care bed numbers, the commonest approaches were
through official government census and sources (17 coun-
tries and regions) and counting through personal contacts (13
countries and regions) (eTables 6 and 7, Supplemental Dig-
ital Content 1, http://links.lww.com/CCM/F287). The majority
(17 countries and regions) obtained data on acute hospital
bed numbers through official government census and sources
(eTables 8 and 9, Supplemental Digital Content 1, http://links.
lww.com/CCM/F287).
Cumulatively, there were 141,034 critical care beds in
the studied countries and regions (Table 2). With a total
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Phua et al
4 www.ccmjournal.org XXX 2020 • Volume XX • Number XXX
population of 3,952,975,069, this works out to 3.6 critical care
beds per 100,000 population. The median number of critical
care beds per 100,000 population per country and region was
significantly lower in low and lower-middle–income econo-
mies (2.3; IQR, 1.4–2.7) than in upper-middle–income econ-
omies (4.6; IQR, 3.5–15.9) and high-income economies (12.3;
IQR, 8.1–20.8) (p = 0.001). This number ranged from 0.7 for
Bangladesh to 28.5 for Taiwan (Table 2 and Fig. 1). Where
data were available, there were more critical care beds in the
public than in the private sector (except in Bangladesh, Japan,
Nepal, the Philippines, and Taiwan) (eTable 10, Supplemental
Digital Content 1, http://links.lww.com/CCM/F287). The vast
majority were in urban areas (eTable 11, Supplemental Digital
Content 1, http://links.lww.com/CCM/F287). There was
TABLE 1. Characteristics of Countries and Regions
Country and
Region
Total Size of
Population
(2017)
Percentage of
Population of ≥
65 yr Old (2017) GDP per Capita
(USD) (2017)
Current Health
Expenditure per
Capita (USD)
(2016)
Current Health
Expenditure as
Percentage of
GDP (2016)
Universal Health
Coverage
Service
Coverage
Index (2015)
Human
Development
Index (2017)
Low- and lower-middle–income countries and regions
Bangladesh 157,826,578 6.23 1,563.99 34.2 2.4 46 0.608
India 1,281,935,911 6.24 1,981.50 62.1 3.6 56 0.640
Indonesia 260,580,739 7.01 3,836.91 111.6 3.1 49 0.694
Laos 7,126,706 3.89 2,423.85 55.2 2.4 48 0.601
Mongolia 3,068,243 4.29 3,671.95 140.7 3.8 63 0.741
Myanmar 55,123,814 5.53 1,249.83 62.1 5.1 60 0.578
Nepal 29,384,297 5.17 900.57 45.5 6.3 46 0.574
Pakistan 204,924,861 4.48 1,466.84 39.6 2.8 40 0.562
Philippines 104,256,076 4.49 2,981.93 129.4 4.4 58 0.699
Sri Lanka 22,409,381 9.67 4,104.63 153.1 3.9 62 0.770
Upper-middle–income countries and regions
China 1,379,302,771 10.81 9,770.85 398.3 5.0 76 0.752
Iran 82,021,564 5.32 5,627.75 415.4 8.1 65 0.798
Kazakhstan 18,556,698 7.65 9,030.32 262.0 3.5 71 0.800
Malaysia 31,381,992 6.10 10,117.57 361.5 3.8 70 0.802
Thailand 68,414,135 10.58 6,578.19 221.9 3.7 75 0.755
High-income countries and regions
Brunei 443,593 4.84 28,572.11 630.6 2.3 ≥ 80a0.853
Hong Kong 7,191,503 16.53 46,220.50 2,462.2 5.9b≥ 80a,c 0.933
Japan 126,451,398 27.87 38,331.98 4,233.0 10.9 ≥ 80a0.909
Oman 3,424,386 3.49 15,170.35 648.3 4.3 72 0.821
Saudi Arabia 28,571,770 3.44 20,803.74 1,147.3 5.7 68 0.853
Singapore 5,888,926 9.63 60,297.79 2,462.4 4.5 ≥ 80a0.932
South Korea 51,181,299 14.12 29,742.84 2,043.9 7.3 ≥ 80a0.903
Taiwan 23,508,428 13.72 22,683.7d1,346.7e6.3e≥ 80a,f 0.907g
GDP = gross domestic product, USD = U.S. dollars.
a
Values of 80 and over reported as ≥ 80 as the index lacks fine resolution at high values.
b
Data from Hong Kong’s Food and Health Bureau (35).
c
Official data from the World Health Organization and the World Bank not available; value imputed from knowledge of Hong Kong’s healthcare system.
d
Data from the International Monetary Fund (36).
e
Data from Taiwan’s Ministry of Health and Welfare (37).
f
Data from Taiwan’s Ministry of Health and Welfare (38).
g
Data from Taiwan’s Directorate General of Budget, Accounting and Statistics (39).
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Clinical Investigation
Critical Care Medicine www.ccmjournal.org 5
no significant difference in the proportion of acute hospital
beds used for critical care in low- and lower-middle–income
incomes (2.5%; IQR, 1.3–3.7%), upper-middle–income econ-
omies (2.9%; IQR, 1.3–4.5), and high-income economies
(5.7%; 2.2–9.0) (p = 0.15). This proportion ranged from 0.6%
for Sri Lanka to 9.2% for Saudi Arabia (Table 2; and eTable 12,
Supplemental Digital Content 1, http://links.lww.com/CCM/
F287).
On univariable linear regression analyses, the number of
critical care beds per 100,000 population in each country and
region was significantly correlated with the number of acute
hospital beds per 100,000 population (Fig. 2), the universal
health coverage service coverage index (Fig. 3), and the
Human Development Index (eFig. 1, Supplemental Digital
Content 1, http://links.lww.com/CCM/F287); but not corre-
lated with health expenditure as a percentage of GDP (eFig. 2,
Supplemental Digital Content 1, http://links.lww.com/CCM/
F287), and percentage of the population who are 65 years old
and older (eFig. 3, Supplemental Digital Content 1, http://
links.lww.com/CCM/F287). The r2 between the number of
critical care beds per 100,000 population and GDP per capita
was 0.17 (p = 0.054) (eFig. 4, Supplemental Digital Content
1, http://links.lww.com/CCM/F287). In a multivariable linear
regression analysis model including these same variables (the
TABLE 2. Number of Critical Care Beds
Country and Region ICUs ICU Beds IMCUs IMCU Beds Critical Care
Bedsa
Critical Care
Beds/100,000
Population
Critical Care Beds as
Percentage of Acute
Hospital Beds
Low- and lower-middle–income countries and regions
Bangladesh 84 878 41 296 1,174 0.7 3.6
India b29,997 b b 29,997 2.3 3.7
Indonesia 1,910 7,094 b b 7,094 2.7 2.3
Laos 22 152 b b 152 2.1 b
Mongolia 43 271 0 0 271 8.8 2.5
Myanmar 68 331 38 255 586 1.1 1.0
Nepal 67 578 22 237 815 2.8 b
Pakistan 114 3,142 b b 3,142 1.5 2.5
Philippines 450 2,315 2 20 2,335 2.2 3.7
Sri Lanka 100 519 0 0 519 2.3 0.6
Upper-middle–income countries and regions
China 3,569 49,453 0 0 49,453 3.6 0.9
Iran 426 3,769 2 21 3,790 4.6 2.9
Kazakhstan 582 3,948 0 0 3,948 21.3 3.9
Malaysia 105 1,060 b b 1,060 3.4 1.7
Thailand 1,221 7,100 b b 7,100 10.4 5.1
High-income countries and regions
Brunei 5 58 0 0 58 13.1 4.7
Hong Kong 24 287 41 224 511 7. 1 2.7
Japan 590 5,973 401 3,268 9,241 7. 3 1.3
Oman 26 196 61 303 499 14.6 8.7
Saudi Arabia 600c6,515 b b 6,515 22.8 9.2
Singapore 28 335 30 336 671 11.4 6.6
South Korea 581 5,402 0 0 5,402 10.6 2.0
Taiwan 344 5,758 65 943 6,701 28.5 9.1
IMCU = intermediate care unit.
a
Critical care beds refer to the sum of IMCU beds and ICU beds.
b
Data not available.
c
Number of ICUs estimated from the number of ICU beds.
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Phua et al
6 www.ccmjournal.org XXX 2020 • Volume XX • Number XXX
universal health coverage service coverage index, Human
Development Index, and GDP per capita were removed due to
multicollinearity) and the World Bank income classification,
the latter was the only predictor of critical care bed capacity
in each country and region: high-income economies had an
average of 12.2 more critical care beds per 100,000 population
than low- and lower-middle–income economies (95% CIs,
4.8–19.6; p = 0.003).
DISCUSSION
This is the largest ever survey of critical care bed capacity in
Asia, covering an area housing more than 90% of the conti-
nent’s population. We found the number of critical care beds
corrected for population size varied widely across 23 countries
and regions, and was significantly lower in low and lower-
middle–income economies than in upper-middle–income
economies and high-income economies. Indeed, this number
was independently predicted by the World Bank income classi-
fication on multivariable analysis, and significantly correlated
with acute hospital bed capacity, the universal health coverage
service coverage index, and the Human Development Index on
univariable analysis.
There is a striking heterogeneity in critical bed capacity
even among Asian countries and regions with the same World
Bank income classification.
As shown in Fig. 1, corrected
for population size, Taiwan
had four times the number
of beds compared with Japan,
Kazakhstan had six times that
of Malaysia, and Mongolia had
13 times that of Bangladesh.
The situation is similar in
Europe (2, 5, 10), in which
Rhodes et al (4) found a seven-
fold difference in capacity be-
tween Germany and Portugal.
Heterogeneity notwith-
standing, we found that low
and lower-middle–income
economies had signifi-
cantly fewer critical care beds
than upper-middle–income
economies and high-income
economies. All the low and
lower-middle–income coun-
tries in our study except for
Mongolia had fewer than three
critical care beds per 100,000
population, whereas all the
high-income ones had at least
seven. Specifically for low-
income countries, although we
found that Nepal has 2.8 crit-
ical care beds per 100,000 pop-
ulation, a systematic review found that Uganda was the only
other low-income country in the world with corresponding
data (0.1 adult ICU beds per 100,000 population) (10, 26).
Limiting the analysis to high-income countries and regions,
the median number of critical care beds per 100,000 popula-
tion in our study (12.3; IQR, 8.1–20.8) and the study of Rhodes
et al (4) (9.5; IQR, 6.5–13.5 for 28 high-income economies)
was similar (p = 0.13). It should be noted though that the lat-
ter review was conducted back in 2010 and 2011 and did not
include private healthcare providers. Corresponding numbers
in New Zealand (adult and pediatric beds in 2016), Australia
(adult and pediatric beds in 2016), Canada (adult and pedi-
atric beds in 2009), and the United States (adult beds in 2010)
were respectively 5.3, 9.0, 9.5, and 27.0 per 100,000 population
(7, 8, 27).
We found a moderate correlation between a country or
region’s critical care bed capacity and the universal health
coverage service coverage index and Human Development
index. The former index is a measure of the United Nations’
Sustainable Development Goal of universal coverage of es-
sential health services (24) but does not specifically evaluate
access to critical care. Regardless, there have been calls to con-
sider essential emergency and critical care as part of universal
health coverage, although not necessarily in state-of-the-art
ICUs (28). The latter index tracks not just health through life
Figure 1. Number of critical care beds per 100,000 population.
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Clinical Investigation
Critical Care Medicine www.ccmjournal.org 7
expectancy but also education
and income (25). As both in-
dices are known to be higher in
high-income economies, their
correlation with critical care
bed capacity is perhaps unsur-
prising. Similar to Rhodes et al
(4), we found no relationship
between critical care bed ca-
pacity and health expenditure
and ageing in the population.
It is likely that critical care bed
capacity is influenced by mul-
tiple other factors related to
healthcare systems and mod-
els of care, and availability of
equipment and expertise in
different parts of the hospital
including general wards (11).
Our study has several
implications. First, interna-
tional comparisons of critical
care bed capacity allow gov-
ernments and policy makers
to benchmark their resources
with others (1, 2). Although
an over-supply of critical care
beds may lead to unneces-
sary admissions, prolonged
death for the terminally ill, and
high costs, a lack may result
in delayed or refused admis-
sions to ICUs for appropriate
patients and increased pre-
ventable mortality (11, 29–31).
The discrepancy in bed num-
bers across countries and re-
gions suggests differences in
patterns of referrals, triage,
rationing, and admissions,
with many critically ill patients
being treated outside of ICUs
and IMCUs in resource-lim-
ited settings. This should trig-
ger debate on what degree of
investments in critical care
is appropriate in countries
which continue to struggle
with more basic healthcare
(11, 15, 31). Second, such het-
erogeneity points toward the
need for caution in extrapo-
lating findings from studies
conducted in high-income
(and often Western) countries
to low- and middle-income
Figure 2. Linear regression analysis of number of critical care beds per 100,000 population against number of
acute hospital beds per 100,000 population. Lines represent linear regression analysis with 95% CIs.
Figure 3. Linear regression analysis of number of critical care beds per 100,000 population against the
universal health coverage service coverage index. Lines represent linear regression analysis with 95% CIs.
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Phua et al
8 www.ccmjournal.org XXX 2020 • Volume XX • Number XXX
ones. For example, fluid resuscitation protocols, long consid-
ered standard of care in the West, has been associated with
increased mortality in African countries, presumably related
at least partly to the lack of ICUs for mechanical ventilation
in patients who developed hypervolemia and/or acute respi-
ratory distress syndrome (32, 33). To this end, clinical practice
guidelines which specifically target resource-limited settings
will prove valuable (31). Third, this exercise is an important
first step toward the creation of national and international reg-
istries of ICUs and IMCUs which are crucial for a better under-
standing of the epidemiology and burden of critical illness and
preparation for pandemics (1, 2).
The limitations of our study must be acknowledged. First,
although we reached out as widely as possible, 21 countries did
not participate. Second, exact definitions of—and the quality of
care provided in (34)—ICU and IMCU beds differ across Asia,
and there is often a lack of a clear distinction between IMCU
and general wards beds used for more intensive monitoring and
treatment in low- and middle-income countries. To deal with
this heterogeneity, we used the WFSICCM task force’s defini-
tion as a reference, and found that all countries but Laos and the
Philippines have a similar basic understanding of what ICUs
and IMCUs are (eTables 4 and 5, Supplemental Digital Content
1, http://links.lww.com/CCM/F287). Third, most countries and
regions did not have official up-to-date numbers of critical care
beds, and we had to use multiple methods to obtain as accu-
rate an estimate of true capacity as possible, including extrap-
olation of data for India, Laos, Malaysia, and the Philippines
(eTable 6, Supplemental Digital Content 1, http://links.lww.
com/CCM/F287). eTable 7 (Supplemental Digital Content 1,
http://links.lww.com/CCM/F287) details the limitations of
the information sources such as the absence of data on IMCU
beds in seven countries. Fourth, critical care capacity is not uni-
form throughout each country, with beds being concentrated
in urban areas.
CONCLUSIONS
In conclusion, critical care bed capacity varies widely across
Asia and is significantly lower in low- and lower-middle–
income economies than in upper-middle–income economies
and high-income economies.
ACKNOWLEDGMENTS
We thank all colleagues and partners who provided valuable
assistance during the data collection process, including Pro-
fessor Jigeeshu V. Divatia, Dr. Bambang Wahyuprajitno, Pro-
fessor Du Bin, Dr. Tan Cheng Cheng, Professor Charles D.
Gomersall, and Professor Younsuck Koh.
Supplemental digital content is available for this article. Direct URL cita-
tions appear in the printed text and are provided in the HTML and PDF
versions of this article on the journal’s website (http://journals.lww.com/
ccmjournal).
Dr. Detleuxay received support for article research from the National Insti-
tutes of Health and disclosed government work. Dr. Patjanasoontorn dis-
closed government work. Dr. Fang disclosed work for hire. The remaining
authors have disclosed that they do not have any potential conflicts of
interest.
For information regarding this article, E-mail: jason_phua@nuhs.edu.sg
REFERENCES
1. Adhikari NK, Fowler RA, Bhagwanjee S, et al: Critical care and the
global burden of critical illness in adults. Lancet 2010; 376:1339–1346
2. Murthy S, Wunsch H: Clinical review: International comparisons in
critical care - lessons learned. Crit Care 2012; 16:218
3. Wallace DJ, Angus DC, Seymour CW, et al: Critical care bed growth
in the United States. A comparison of regional and national trends.
Am J Respir Crit Care Med 2015; 191:410–416
4. Rhodes A, Ferdinande P, Flaatten H, et al: The variability of crit-
ical care bed numbers in Europe. Intensive Care Med 2012; 38:
1647–1653
5. Wunsch H, Angus DC, Harrison DA, et al: Variation in critical care
services across North America and Western Europe. Crit Care Med
2008; 36:2787–2793, e1
6. Martin JM, Hart GK, Hicks P: A unique snapshot of intensive care re-
sources in Australia and New Zealand. Anaesth Intensive Care 2010;
38:149–158
7. Fowler RA, Abdelmalik P, Wood G, et al; Canadian Critical Care Trials
Group; Canadian ICU Capacity Group: Critical care capacity in Canada:
Results of a national cross-sectional study. Crit Care 2015; 19:133
8. Halpern NA, Goldman DA, Tan KS, et al: Trends in critical care beds
and use among population groups and Medicare and Medicaid ben-
eficiaries in the United States: 2000-2010. Crit Care Med 2016;
44:1490–1499
9. The World Bank: World Bank Country and Lending Groups. Avail-
able at: https://datahelpdesk.worldbank.org/knowledgebase/arti-
cles/906519. Accessed June 1, 2019
10. Murthy S, Leligdowicz A, Adhikari NK: Intensive care unit capacity
in low-income countries: a systematic review. PLoS One 2015;
10:e0116949
11. Dondorp AM, Iyer SS, Schultz MJ: Critical care in resource-restricted
settings. JAMA 2016; 315:753–754
12. Du B, Xi X, Chen D, et al; China Critical Care Clinical Trial Group
(CCCCTG): Clinical review: Critical care medicine in mainland China.
Crit Care 2010; 14:206
13. Haniffa R, De Silva AP, Iddagoda S, et al: A cross-sectional survey
of critical care services in Sri Lanka: A lower middle-income country.
J Crit Care 2014; 29:764–768
14. Mendsaikhan N, Begzjav T, Lundeg G, et al: A nationwide census
of ICU capacity and admissions in Mongolia. PLoS One 2016;
11:e0160921
15. Arabi YM, Phua J, Koh Y, et al; Asian Critical Care Clinical Trials
Group: Structure, organization, and delivery of critical care in Asian
ICUs. Crit Care Med 2016; 44:e940–e948
16. United Nations: Regional Groups of Member States. Available at:
http://www.un.org/depts/DGACM/RegionalGroups.shtml. Accessed
October 20, 2018
17. Halpern NA, Pastores SM: Critical care medicine beds, use, occu-
pancy, and costs in the United States: A methodological review. Crit
Care Med 2015; 43:2452–2459
18. Marshall JC, Bosco L, Adhikari NK, et al: What is an intensive care
unit? A report of the task force of the World Federation of Societies of
Intensive and Critical Care Medicine. J Crit Care 2017; 37:270–276
19. Vincent JL, Rubenfeld GD: Does intermediate care improve patient
outcomes or reduce costs? Crit Care 2015; 19:89
20. Prin M, Wunsch H: The role of stepdown beds in hospital care. Am J
Respir Crit Care Med 2014; 190:1210–1216
21. Central Intelligence Agency: The World Factbook. Available at: https://
www.cia.gov/library/publications/the-world-factbook/. Accessed
June 1, 2019
22. The World Bank: Gross Domestic Product per Capita: World Bank
National Accounts Data, and OECD National Accounts Data Files.
Available at: https://data.worldbank.org/indicator/NY.GDP.PCAP.
CD. Accessed September 9, 2019
Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Clinical Investigation
Critical Care Medicine www.ccmjournal.org 9
23. World Health Organization: Global Health Expenditure Database.
Available at: http://apps.who.int/nha/database. Accessed June 1,
2019
24. Hogan DR, Stevens GA, Hosseinpoor AR, et al: Monitoring universal
health coverage within the Sustainable Development Goals: Devel-
opment and baseline data for an index of essential health services.
Lancet Glob Health 2018; 6:e152–e168
25. United Nations Human Development Report Office: Human Devel-
opment Index. Available at: http://hdr.undp.org/en/content/human-
development-index-hdi. Accessed June 1, 2019
26. Kwizera A, Dünser M, Nakibuuka J: National intensive care unit bed
capacity and ICU patient characteristics in a low income country.
BMC Res Notes 2012; 5:475
27. Australian and New Zealand Intensive Care Society Centre for
Outcome and Resource Evaluation: ANZICS CORE Annual Re-
port 2017. Available at: https://www.anzics.com.au/wp-content/
uploads/2018/10/ANZICS-CORE-Annual-Report-2017.pdf.
Accessed October 20, 2018
28. Schell CO, Gerdin Wärnberg M, Hvarfner A, et al: The global need for
essential emergency and critical care. Crit Care 2018; 22:284
29. Wunsch H: Is there a Starling curve for intensive care? Chest 2012;
141:1393–1399
30. Vukoja M, Riviello ED, Schultz MJ: Critical care outcomes in resource-
limited settings. Curr Opin Crit Care 2018; 24:421–427
31. Schultz MJ, Dunser MW, Dondorp AM, et al; Global Intensive Care
Working Group of the European Society of Intensive Care Medicine:
Current challenges in the management of sepsis in ICUs in resource-
poor settings and suggestions for the future. Intensive Care Med
2017; 43:612–624
32. Andrews B, Semler MW, Muchemwa L, et al: Effect of an early re-
suscitation protocol on in-hospital mortality among adults with sepsis
and hypotension: A randomized clinical trial. JAMA 2017; 318:1233–
1240
33. Maitland K, Kiguli S, Opoka RO, et al; FEAST Trial Group: Mortality
after fluid bolus in African children with severe infection. N Engl J Med
2011; 364:2483–2495
34. Phua J, Koh Y, Du B, et al; MOSAICS Study Group: Management of
severe sepsis in patients admitted to Asian intensive care units: Pro-
spective cohort study. BMJ 2011; 342:d3245
35. Food and Health Bureau: Hong Kong’s Domestic Health Accounts.
Available at: https://www.fhb.gov.hk/statistics/download/dha/en/a_
estimate_1617.pdf. Accessed June 1, 2019
36. International Monetary Fund: Taiwan, Gross Domestic Product. Avail-
able at: http://www.imf.org/external/pubs/ft/weo/2016/02/weodata/
weorept.aspx?sy=2014&ey=2021&scsm=1&ssd=1&sort=country&
ds=.&br=1&pr1.x=57&pr1.y=8&c=528&s=NGDPD%2CNGDPDP
C&grp=0&a=. Accessed October 20, 2018
37. Ministry of Health and Welfare, Taiwan: National health expenditure.
Available at: https://www.mohw.gov.tw/np-130-2.html. Accessed
June 1, 2019
38. Ministry of Health and Welfare, Taiwan: UHC service coverage index.
Available at: https://www.hphconferences.org/fileadmin/user_up-
load/P3_Wang.PDF. Accessed September 9, 2019
39. Directorate General of Budget, Accounting and Statistics, Taiwan:
Human Development Index. Available at: http://eng.stat.gov.tw/
public/data/dgbas03/bs2/gender/International%20Gender/%E4%
BA%BA%E9%A1%9E%E7%99%BC%E5%B1%95%E6%8C%87
%E6%95%B8.xls. Accessed June 1, 2019
