Structure, Organization, and Delivery of Critical Care

Structure, Organization, and Delivery of Critical Care in Asian ICUs Yaseen M. Arabi, MD1; Jason Phua, MBBS2; Younsuck Koh, MD3; Bin Du, MD4; Mohammad Omar Faruq, MD5; Masaji Nishimura, MD6; Wen-Feng Fang, MD7; Charles Gomersall, MD8; Hussain N. Al Rahma, MD9; Hani Tamim, PhD1; Hasan M. Al-Dorzi, MD1; Fahad M. Al-Hameed, MD10; Neill K. J. Adhikari, MD11; Musharaf Sadat, MBBS1; and the Asian Critical Care Clinical Trials Group Intensive Care Department, King Saud bin Abdulaziz University for Health Sciences, King Abdullah International Medical Research Center, Riyadh, Saudi Arabia. 2 Division of Respiratory and Critical Care Medicine, University Medicine Cluster, National University Hospital, National University Health System, Singapore. 3 Department of Pulmonary and Critical Care Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. 4 Department of Critical Care Medicine, Peking Union Medical College Hospital, Beijing, China. 5 Department of Critical Care Medicine, Birdem General Hospital, Ibrahim Medical College, Dhaka, Bangladesh. 6 Emergency and Critical Care Medicine, University of Tokushima Graduate School, Tokushima, Japan. 7 Division of Pulmonary and Critical Care Medicine, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of Medicine, and Chang Gung University of Science and Technology, Kaohsiung, Taiwan. 8 Department of Anaesthesia and Intensive Care, Chinese University of Hong Kong, Shatin, Hong Kong, China. 9 Intensive Care Department, Dubai Hospital, Dubai, United Arab Emirates. 10 Intensive Care Department, King Abdulaziz Medical City and King Saud bin Abdulaziz University for Health Sciences, Jeddah, Saudi Arabia. 11 Department of Critical Care Medicine, Sunnybrook Health Science Centre, Interdepartmental Division of Critical Care Medicine, University of Toronto, Toronto, Canada. Additional members of the Asian Critical Care Clinical Trials Group can be found in Supplemental Appendix 1 (Supplemental Digital Content 2, http://links.lww.com/CCM/B980). Supplemental digital content is available for this article. Direct URL citations 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). Supported by King Abdullah International Medical Research Center (KAIMRC), Riyadh, Saudi Arabia. Dr. Al-Dorzi received funding from Sanofi Aventis. The remaining authors have disclosed that they do not have any potential conflicts of interest. Address requests for reprints to: Yaseen M. Arabi, MD, FCCP, FCCM, Chairman, Intensive Care Department, MC-1425, Medical Director, Respiratory Services, Professor, College of Medicine, King Saud bin Abdulaziz University for Health Sciences, King Abdulaziz Medical City, P.O. Box 22490 Riyadh, 11426, Kingdom of Saudi Arabia. E-mail: arabi@ngha. med.sa Copyright © 2016 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved. DOI: 10.1097/CCM.0000000000001854 1

Critical Care Medicine

Objectives: Despite being the epicenter of recent pandemics, little is known about critical care in Asia. Our objective was to describe the structure, organization, and delivery in Asian ICUs. Design: A web-based survey with the following domains: hospital organizational characteristics, ICU organizational characteristics, staffing, procedures and therapies available in the ICU and written protocols and policies. Setting: ICUs from 20 Asian countries from April 2013 to January 2014. Countries were divided into low-, middle-, and high-income based on the 2011 World Bank Classification. Subjects: ICU directors or representatives. Measurements and Main Results: Of 672 representatives, 335 (50%) responded. The average number of hospital beds was 973 (se of the mean [sem], 271) with 9% (sem, 3%) being ICU beds. In the index ICUs, the average number of beds was 21 (sem, 3), of single rooms 8 (sem, 2), of negative-pressure rooms 3 (sem, 1), and of board-certified intensivists 7 (sem, 3). Most ICUs (65%) functioned as closed units. The nurse-to-patient ratio was 1:1 or 1:2 in most ICUs (84%). On multivariable analysis, single rooms were less likely in low-income countries (p = 0.01) and nonreferral hospitals (p = 0.01); negativepressure rooms were less likely in private hospitals (p = 0.03) and low-income countries (p = 0.005); 1:1 nurse-to-patient ratio was lower in private hospitals (p = 0.005); board-certified intensivists were less common in low-income countries (p < 0.0001) and closed ICUs were less likely in private (p = 0.02) and smaller hospitals (p < 0.001). Conclusions: This survey highlights considerable variation in critical care structure, organization, and delivery in Asia, which was related to hospital funding source and size, and country income. The lack of single and negative-pressure rooms in many Asian ICUs should be addressed before any future pandemic of severe respiratory illness. (Crit Care Med 2016; XX:00–00) Key Words: Asia; critical care; health services research; infection control; intensive care

A

sia is the largest and most populous continent on Earth. It comprises 30% of Earth’s land area and is the home for 60% of the human population (1). In www.ccmjournal.org

Copyright © 2016 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

1

Arabi et al

addition to the large critical illness burden (2), Asia has been challenged by natural disasters, wars, and multiple epidemics, all adding strains on critical care services. According to the “Disasters in Asia and the Pacific: 2014 Year in review” report released by the United Nations Economic and Social Commission for Asia and the Pacific, more than half of the world’s 226 natural disasters of 2014 occurred in the Asia and Pacific region (3). Recent outbreaks of the Severe Acute Respiratory Syndrome (SARS), avian influenza, and Middle East Respiratory Syndrome Coronavirus (MERS-CoV) (4–6), which all emerged from Asia, have highlighted the importance of critical care infrastructure. It has been suggested that higher quality critical care incorporating best practices in structure, organization, and care delivery could reduce the high mortality in ICU patients in general (7), including mortality associated with these infections. Related factors such as architectural design to facilitate infection control (8), staffing with intensivists (9, 10), the closed-unit model (11–13), and nurse-to-patient ratio (9) have also been associated with improved patient outcomes. Our objectives of this study were to describe the structure, organization, and delivery of critical care in Asian ICUs, which previously have been examined in small or single-country studies. These data can be used for benchmarking and to identify urgent areas for improvement.

METHODS This was a multinational survey of Asian ICUs endorsed by the Asian Critical Care Trials Group. The group had established collaborative work to understand critical care practice in the Asia (14, 15). Investigators for each country obtained approval by institutional review boards as appropriate. Survey Development Items were first generated after a review of related literature and previously developed questionnaires (16, 17). Additional questions were designed to capture the recommendations formulated by the Society of Critical Care Medicine (18) and by the Leapfrog Group (19). The first draft of the questionnaire was discussed with an expert focus group in April 2011. After several rounds of review by the focus group, the final survey included the following domains: hospital organizational characteristics, ICU organizational characteristics (number of beds, single vs shared rooms, negative-pressure rooms, open vs closed ICU), staffing (medical, nurses, and others), procedures and therapies available in the ICU, and written protocols and policies. The survey questions are available in Table S1 (Supplemental Digital Content 1, http://links.lww.com/CCM/B870). Definitions We defined ICUs as units capable of providing invasive mechanical ventilation and organ support, such as using vasoactive medications and renal replacement therapy, and recognized to be ICUs by their hospitals. We defined a board-certified intensivist as a physician who had passed a certification examination in intensive care and a nonboard-certified intensivist as a physician who had completed training program in intensive care or who treated 2

www.ccmjournal.org

ICU patients and was recognized by his/her hospital as an intensivist but without certification by examination (20). ICU was labeled as closed if intensivists were responsible for admission, management, and discharge decisions for critically ill patients. Survey Administration The survey was constructed as a web-based survey. Using lists of ICUs from national critical care societies and networks, supplemented by regional and personal snowball sampling, we approached by E-mail ICU directors or representatives to participate in the survey between April 2013 and January 2014. Reminders were sent three times, 4 weeks apart. In Bangladesh, the survey was distributed in paper form; completed surveys were returned by mail and then entered into the database. Statistical Analysis SAS (version 19; SAS Institute, Cary, NC) was used to analyze the data. The countries were divided into low-, middle-, and highincome groups based on the World Bank Classification of Asian countries, which was based on the 2011 gross national income per capita (21) (Fig. S1 and Table S2, Supplemental Digital Content 1, http://links.lww.com/CCM/B870). The middle-income group combined the upper middle and lower middle categories (21). The study design involved clustering by country, which was accounted for in the statistical analysis by applying the various PROC SURVEY procedures with a “cluster” statement (SAS). Categorical variables were presented as frequencies and percents, estimated with the PROC SURVEYFREQ procedure. Rao-Scott modified chi-square test was used to evaluate the association between categorical variables. Continuous variables were presented as mean and its se (sem), calculated using PROC SURVEYMEANS. PROC SURVEYREG was used for comparison of means. We constructed multivariable linear or logistic regression analyses to evaluate the predictors for the percentage of ICU: hospital beds, percentage of single rooms, percentage of negative-pressure rooms, availability of 1:1 nursing staff, and closed ICU model. We used the following covariates: private (vs government) funding, nonuniversity (vs university) affiliation, nonreferral (vs referral) hospital status, number of hospital beds, and low-income (vs middle- and high-income) group. PROC SURVEYREG and PROC SURVEYLOGISTIC were used for linear and logistic regressions, respectively. The results were presented as either parameter estimates or odds ratios (ORs) with 95% CIs. A p value of less than 0.05 was considered significant.

RESULTS Of 672 ICU representatives approached, 335 responded (50%) from 20 countries. The highest responses were from China (85/103, 83%), Japan (61/69, 88%), Bangladesh (51/53, 96%), and Saudi Arabia (38/147, 26%). The participating countries, number of approached ICUs, and the country response rate are shown in Figure S1. Hospital Organizational Characteristics Most hospitals (61.8%) were government-funded, 58.8% were affiliated with universities, and 81.2% were referral centers XXX 2016 • Volume XX • Number XXX


Online Clinical Investigation

Table 1. Hospital Organizational Characteristics According to the 2011 World Bank Classification of the Country Income

Variables

Total n = 335

Low Income n = 83

Middle Income n = 99

High Income n = 153

p

Hospital funding, n (%) Government

207 (61.8)

16 (19.3)

96 (97)

95 (62.1)

Private

128 (38.2)

67 (80.7)

3 (3)

58 (37.9)

University-affiliated

197 (58.8)

27 (32.5)

73 (73.7)

97 (63.4)

Nonuniversity-affiliated

138 (41.2)

56 (67.5)

26 (26.3)

56 (36.6)

272 (81.2)

56 (67.5)

80 (80.8)

136 (88.9)

63 (18.8)

27 (32.5)

19 (19.2)

17 (11.1)

0.102

Academic affiliation, n (%) 0.23

Referral status, n (%) Referral hospitals Nonreferral hospitals

0.31

No. of beds (mean, sem) No. of hospital beds No. of ICU beds in the hospital ICU: hospital bed percentage

973.2 (271.0)

333.9 (87.4)

1676.9 (175.5)

864.6 (137.5)

0.002

52.5 (11.9)

30.0 (15.1)

62.7 (3.3)

58.1 (18.6)

0.16

8.8 (2.5)

16.2 (1.0)

4.4 (1.1)

7.6 (2.5)

0.002

97 (98)

147 (96.1)

0.36

Other hospital characteristics, n (%) Infection control department or committee

285 (85.1)

41 (49.4)

68 (20.3)

6 (7.2)

18 (18.2)

44 (28.8)

0.38

Cardiac arrest team

185 (55.2)

36 (43.4)

41 (41.4)

108 (70.6)

0.42

Basic life support training required for physicians

237 (70.7)

35 (42.2)

86 (86.9)

116 (75.8)

0.27

Basic life support training required for nurses

233 (69.6)

37 (44.6)

82 (82.8)

114 (74.5)

0.38

Chronic ventilation unit

Analysis was Adjusted for Clustering.

(Table 1). The average number of hospital beds was 973 (sem, 271). The number of all ICU beds including adult, pediatric, and neonatal in the study hospitals was 52.5 (sem, 11.9) and the mean percentage of ICU beds: hospital beds was 8.8% (sem, 2.5%). Respondents reported that 282 hospitals (85.1%) had infection control departments or committees, 185 (55.2%) had cardiac arrest teams, and 68 (20.3%) had chronic ventilation units. Basic life support training was required in 237 (70.7%) of hospitals for physicians and in 232 (69.6%) for nurses. Hospital characteristics varied considerably according to the country-income group (Table 1) and among the countries with the highest responses (Fig. 1). ICU Organizational Characteristics The majority of these units admitted a wide patient mix: medical (91.6%), surgical (87.2%), trauma (71.9%), respiratory (83.9%), and neurologic (77.6%) patients. Burn patients were admitted in 37.9% of these units and cardiac surgery patients in 39.7% (Table 2). The average number of beds in the index ICUs was 21.0 (sem, 3.4) with a mean 36.9% of ICU rooms as single rooms Critical Care Medicine

(sem, 5.8%). A total of 45 ICUs (13.4%) had no single rooms. The percentage of single rooms out of all ICU rooms varied with an average of 21.6% (sem, 4.9%) in low-income, 31.7% (sem, 14.3%) in middle-income, and 48.4% (sem, 5.6%) in high-income countries (p = 0.04) (Table 2 and Fig. 1). Multivariable analysis confirmed that referral hospital status (nonreferral vs referral hospitals, –11.12 single rooms; 95% CI, –18.29 to –3.95; p = 0.01) and country income (lowincome vs higher income country, –21.24 single rooms; 95% CI, –34.73 to –7.76; p = 0.01) were independently associated with lower percentage of single rooms. Negative-pressure rooms constituted 12.1% (sem, 2.2%) of ICU beds: 2.9% (sem, 0.9%) of ICU beds in low-income countries, 14.3% (sem, 1.2%) in middle-income, and 15.6% (sem, 0.9%) in highincome countries (p = 0.002). Among the study ICUs, 123 ICUs (36.7%) had no negative-pressure rooms. Multivariable analysis identified low-income country group versus higher income group (–9.85%; 95% CI, –14.8 to –4.90%) and private versus government-funded hospitals (–3.89%; 95% CI, –7.27 to –0.51%) as independent predictors of lower percentage of www.ccmjournal.org


3

Arabi et al

Figure 1. Key hospital and ICU organizational characteristics among countries with the highest responses.

negative-pressure rooms. The number of sinks and alcohol hand gel dispensers for hand hygiene also varied considerably according to income group, and was lowest in the low-income counties (Table 2). ICUs in low-income countries had the lowest number of hand washing sinks (mean, 5.2; sem, 2.6) and alcohol hand gel dispensers (mean, 13.2; sem, 8.7). Among the responding ICUs, 64.9% operated as closed units. Multivariable analysis showed that the closed-unit model was more likely to be present in government hospitals 4

www.ccmjournal.org

(private vs government hospital: OR, 0.28; 95% CI, 0.10–0.83; p = 0.02) and in larger hospital (OR, 1.09/100 hospital beds; 95% CI, 1.05–1.15; p < 0.001). Staffing The responding ICUs reported having an average of 6.6 (sem, 2.9) board-certified intensivists and 5.9 (sem, 1.5) nonboardcertified intensivists, with 53 ICUs (15.9%) having no boardcertified intensivists (Table 3). Having a board-certified XXX 2016 • Volume XX • Number XXX



Table 2. The Index ICU Organizational Characteristics According to the 2011 World Bank Classification of the Country Income

Variables

Total n = 335

Low Income n = 83


High Income n = 153

p

Type of ICU patients admitted, n (%) Medical

307 (91.6)

78 (94.0)

89 (89.9)

140 (91.5)

0.77

Surgical

292 (87.2)

76 (91.6)

86 (86.9)

130 (85.0)

0.79

Trauma

241 (71.9)

71 (85.5)

78 (78.8)

92 (60.1)

0.21

Respiratory

281 (83.9)

72 (86.7)

84 (84.8)

125 (81.7)

0.87

Neurologic

260 (77.6)

74 (89.2)

73 (73.7)

113 (73.9)

0.25

Burn

127 (37.9)

41 (49.4)

31 (31.3)

55 (35.9)

0.43

Cardiac

220 (65.7)

54 (65.1)

69 (69.7)

97 (63.4)

0.88

Cardiac surgery

133 (39.7)

23 (27.7)

38 (38.4)

72 (47.1)

0.68

No. of ICU beds No. of beds in the ICU (mean, sem)

21.0 (3.4)

16.5 (6.9)

22.8 (1.0)

22.2 (5.5)

0.72

8.1 (2.1)

4.4 (2.4)

7.2 (0.1)

10.6 (3.9)

0.45

36.9 (5.8)

21.6 (4.9)

31.7 (14.3)

48.4 (5.6)

0.04

4 (4.0)

7 (4.6)

0.21

Single rooms n (mean, sem) Percentage of ICU beds (mean, sem)

45 (13.4)

34 (41.0)

2.5 (0.7)

0.6 (0.4)

3.2 (0.3)

3.2 (1.0)

0.29

Percentage of ICU beds (mean, sem)

12.1 (2.2)

2.9 (0.9)

14.3 (1.2)

15.6 (0.9)

0.002

ICUs without negative-pressure rooms, n (%)

123 (36.7)

64 (77.1)

No. of sinks per ICU (mean, sem)

12.6 (2.9)

5.2 (2.6)

14.6 (0.1)

15.3 (5.1)

0.06

No. of alcohol hand gel dispenser per ICU (mean, sem)

24.3 (4.2)

13.2 (8.7)

27.3 (0.7)

28.5 (4.2)

0.37

ICUs without single rooms, n (%) Negative-pressure rooms n (mean, sem)

31 (31.3)

28 (18.3)

0.12

Operation model Closed ICU

216 (64.9)

33 (40.7)

87 (87.9)

96 (62.7)

Open ICU

117 (35.1)

48 (59.3)

12 (12.1)

57 (37.3)

Open visiting hour policy

70 (21.0)

20 (24.1)

12 (12.1)

38 (24.8)

Restricted visiting hours

264 (79.0)

62 (75.6)

87 (87.9)

115 (75.2)

2.7 (0.2)

1.2 (0.5)

0.099

Visiting hours, n (%)

If restricted, no. of visiting hours (mean, sem)

2.7 (0.9)

3.8 (1.1)

0.56 0.02

Analysis was adjusted for clustering.

intensivist was much less common in low-income countries (OR, 0.03; 95% CI, 0.01–0.09; p < 0.001) (Table 4). Most ICUs (84%) had a nurse-to-patient ratio as 1:1 or 1:2. It was 1:1 in 33%, 1:2 in 51%, and 1:3 or more in 16%. Multivariable analysis showed that the presence of 1:1 nursing staff was much lower in private hospitals (OR, 0.25; 95% CI, 0.10–0.66; p < 0.005). The following staff categories were also present: physiotherapists in 174 ICUs (52%), clinical pharmacists in 165 (49%), respiratory therapists in 136 (41%), infection control practitioners in 201 (60%), and dieticians in 160 (48%). Critical Care Medicine

Procedures, Therapies, Protocols, and Policies in ICU Procedures and therapies in the ICU (Table S3, Supplemental Digital Content 1, http://links.lww.com/CCM/B870) and protocols and policies (Table S4, Supplemental Digital Content 1, http://links.lww.com/CCM/B870) were generally less available in low-income countries. For example, in low-income countries, central oxygen supply was available in 92%, invasive arterial monitoring in 60%, continuous renal replacement therapy in 34%, intermittent hemodialysis in 71%, and enteral nutrition formulae in 70% of responding ICUs. There were also significant variation in the presence of written protocols www.ccmjournal.org


5

Arabi et al

Table 3.

Income

ICU Staffing According to the 2011 World Bank Classification of the Country

Total n = 335

Low Income n = 83


High Income n = 153

p

No. of board-certified intensivists (mean, sem)

6.6 (2.9)

1.8 (0.7)

13.0 (2.4)

5.1 (1.7)

0.03

No. of nonboard-certified intensivists (mean, sem)

5.9 (1.5)

3.4 (0.8)

8.1 (1.6)

5.9 (1.7)

0.17

195 (58.5)

28 (33.7)

64 (64.6)

103 (67.3)

0.44

One or more nurses: one bed

112 (33.4)

19 (22.9)

50 (50.5)

43 (28.1)

0.374

One nurse: two beds

170 (50.7)

49 (59)

32 (32.3)

89 (58.2)

Variables

Physicians

Accredited intensive care fellowship program, n (%) Nurse-to-patient ratio, n (%)

One nurse: three beds

41 (12.2)

12 (14.5)

15 (15.2)

14 (9.2)

One nurse: four or more beds

12 (3.6)

3 (3.6)

2 (2.0)

7 (4.6)

Presence of nurse manager, n (%)

305 (91.0)

69 (83.1)

91 (91.9)

145 (94.8)

0.52

No. of registered nurses (mean, sem)

46.9 (8.3)

No. of nonregistered nurses (mean, sem)

32.6 (17.8)

43.3 (0.3)

57.0 (11.5)

0.50

29.5 (22.3)

8.2 (1.8)

14.4 (0.1)

50.9 (46.4)

0.06

Physiotherapists

174 (51.9)

58 (69.9)

26 (26.3)

90 (58.8)

0.19

Clinical pharmacist

165 (49.3)

19 (22.9)

42 (42.4)

Respiratory therapists

136 (40.6)

23 (27.7)

32 (32.3)

81 (52.9)

0.29

Social workers

120 (35.8)

18 (21.7)

17 (17.2)

85 (55.6)

0.14

Infection control practitioners

201 (60)

35 (42.2)

67 (67.7)

99 (64.7)

0.60

Dieticians

160 (47.8)

44 (53)

23 (23.2)

93 (60.8)

0.27

Other staff, n (%) 104 (68)

0.07

Analysis was adjusted for clustering.

such as for the management of severe sepsis and acute respiratory distress syndrome and weaning of mechanical ventilation (Table S4, Supplemental Digital Content 1, http://links.lww. com/CCM/B870).

DISCUSSION Our study demonstrates considerable variation in the organization and staffing among Asian ICUs. The variability was observed in the structure of the unit, the operation model, staffing levels as well as different care processes. The variations were related to several factors including hospital size and country income. An important finding is the deficiency in the availability of single and negative-pressure rooms, which has significant implication for infection control, particularly with the emergence of new infectious diseases. The survey highlights that Asian hospitals in general have a percentage of ICU beds (8.8%; sem, 2.5%) that is comparable to reported percentages from U.S. ICUs. In a 2004 study from United States, adult ICU beds accounted for 9.0% of acute care hospital beds (22). Interestingly, the percentage was lower in Canada (3.4%) (22) and Europe (the average was 2.8% in 2011 6

www.ccmjournal.org

and ranged between 1.3% in the Czech Republic and 5.1% in Germany) (23). The study also shows variation among Asian countries with China and Japan having lower ICU hospital beds compared with other studied countries. This is consistent with earlier single country studies showing that ICU beds accounted for 1.8% (interquartile range, 1.3–2.1%) of total hospital beds in China (24), 2% (range, 0.6–3.5%) in Japan (25), and 5–10% in India (26). It is important to point out that the higher ICU hospital beds does not necessarily reflect higher ICU provision to the general population, but simply higher ICU provision relative to hospital bed provision—both may be inadequate. The low percentage of single and negative-pressure rooms in the surveyed hospitals reflects an old design of ICUs in many hospitals as an open shared space, which is thought to facilitate care and reduce staffing needs. However, such structure has also implications for patient privacy and for infection control. This finding, along with variability in the presence of hand hygiene facilities and infection control departments or committees, highlights potential deficiencies in infection control. This is particularly true for low-income countries, where resource limitations may impede funding of basic infection control procedures. Asia has been the origin of outbreaks of XXX 2016 • Volume XX • Number XXX



Table 4. Predictors of Percentage of ICU: Hospital Beds, Percentage of Single Rooms, Percentage of Negative Pressure Rooms, Closed ICU Model, Availability of a BoardCertified Intensivist and Availability of 1:1 Nursing Staff Variables

Predictors of percentage of ICU:hospital beds

Parameter Estimate/OR

p

CI

Parameter estimate

Private vs government funding

2.00

–5.11 to 9.10

0.48

Nonuniversity vs university affiliation

1.46

–3.66 to 6.57

0.47

Nonreferral vs referral hospital status

1.52

–5.62 to 8.66

0.59

–0.22

–0.76 to 0.31

0.31

6.00

–1.98 to 13.97

0.11

3.77

–3.51 to 11.05

0.22

–2.90

–18.71 to 12.91

0.64

–11.12

–18.29 to –3.95

0.012

–1.64 to 1.19

0.68

–34.73 to –7.76

0.012

No. of hospital beds (per hundred) Low-income group vs intermediate- and high-income group Predictors of percentage of single rooms Private vs government funding Nonuniversity vs university affiliation Nonreferral vs referral hospital status No. of hospital beds (per hundred) Low-income group vs intermediate- and high-income group

–0.22 –21.24

Predictors of percentage of negative-pressure rooms Private vs government funding

–3.89

–7.27 to –0.51

0.03


–4.73

–11.04 to 1.58

0.11

1.56

-3.41 to 6.53

0.43

No. of hospital beds (per hundred)

–0.13

–0.44 to 0.17

0.29

Low-income group vs intermediate- and high-income group

–9.85

–14.8 to –4.90

0.005

Predictors of closed ICU model

OR



0.28

0.10–0.83

0.02


1.27

0.52–3.12

0.60


1.08

0.47–2.51

0.85


1.09

1.05–1.15

< 0.001


0.86

0.46–1.59

0.62


1.53

0.89–2.63

0.12


0.44

0.28–0.67

0.0002


0.78

0.22–2.76

0.69


1.03

0.94–1.14

0.51


0.03

0.01–0.15

< 0.0001


0.25

0.10–0.66

0.005


1.60

0.85–3.01

0.15


0.64

0.30–1.38

0.26


1.02

0.99–1.04

0.13


1.17

0.85–1.61

0.34

Predictors of the presence of a board-certified intensivist

Predictors of the availability of 1:1 nursing staff

OR = odds ratio. Analysis was adjusted for clustering.


www.ccmjournal.org


7

Arabi et al

some emerging viruses, notably SARS and MERS-CoV (4–6), during which neighboring patients, healthcare staff, and even visitors became victims of healthcare-associated spread of these viruses (27, 28). Hence, we believe that most of, if not all, ICU rooms should be single-rooms and that ICUs should have sufficient numbers of negative-pressure rooms, although there is no clear guidelines about the required number of these rooms in the ICU. Conversion to single rooms may substantially reduce infections (29) and delirium (30), although some studies reported an increased rate of noninfectious adverse events with single rooms (31). We also believe that all ICUs should have adequate hand hygiene infrastructure and infection control and prevention programs in accordance with international guidelines (32). Our survey showed that 53 (15.9%) of the surveyed ICUs had no board-certified intensivists. There are substantial regional variations in the presence of intensivist coverage. Studies showed that 53% of ICUs in the United States had no intensivist coverage (33) whereas 71.7% of European ICUs were staffed 24 hours by intensivists (34). Additionally, the closed ICU model was reported in 64.9% of the surveyed ICUs in our study. Interestingly, this is relatively higher than what has been reported in a U.S. survey (26%) (33) and a European study (59.1%) (35), and is consistent with earlier single-country studies from Asia. A study from China reported that 52% of ICUs were closed, 36% semi-closed, and the others were open ICUs (24). These numbers may also reflect biased sampling, since the contact persons for the survey were mainly intensivists; therefore, ICUs without intensivists were less likely to be reached. Although we observed substantial variation in the presence, number, and qualifications of intensivists, the mean ratio of board-certified intensivists-toICU beds was 1:6, which increased proportionally with increasing country income. A multicenter observation study in the United States found that a closed ICU status was not associated with ICU mortality (36). However, a systematic review found that high-intensity staffing was associated with lower hospital mortality and significant reductions in hospital and ICU length of stay compared with low-intensity staffing (37). In our survey, the nurse-to-patient ratio was mostly 1:2 or less in 67% of Asian ICUs. A previous study from China showed that the nurse-to-patient ratio ranged from 1:1.4 to 1:2.0 (24). Higher ICU nurse-to-patient ratio has been associated with improved outcomes (38, 39). In a large U.S. study, the median nurse-to-patient ratio was 1:1.7 and higher ratio was associated with improved survival (36). Additionally, our study demonstrated that country income and hospital size were important determinants of staffing in Asian ICUs. Given the fact that low-income countries were less represented in our study compared with middle- and high-income countries, it is likely that Asian ICUs have actually a considerable shortage of intensivists and ICU nurses (40). Our study also found significant variation in the use of written policies and clinical protocols. A recent observational study in the United States also showed that the median number of clinical protocols per ICU was 19 (Q1, Q3: 15–21.5) with the lack of association number of protocols and mortality (41). 8

www.ccmjournal.org

Critical care medicine is expensive, with annual costs approaching 82 billion USD in United States (42). An important question is how much should be invested in critical care, especially in low- and middle-income countries. It has been suggested that the high mortality in the recent Ebola outbreak in Africa could have been reduced substantially had the basic critical care been provided to affected patients (43). It has been also suggested that a short period of critical care is cost-effective for the treatment of the life-threatening conditions and a public health argument can be made for intensive care in lowincome settings (44). The fact that outbreaks of life-threatening infections in low-income countries can become epidemics affecting populations locally, regionally, and globally mandates more international aid targeted to the care of acute conditions, including illnesses due to infectious diseases, in these countries. The strengths of our study include the inclusion of ICUs from 20 countries, which improves study generalizability. However, the study has several limitations. Most respondents were from large referral government hospitals. This may be a reflection of the funding source being governmental in many Asian ICUs (24, 45). However, it may reflect a sampling issue, as representatives of such ICUs may have been more likely to be contacted and more likely to have responded. Representatives of ICUs from academic and referral hospitals and possible representatives of ICUs with more favorable characteristics were also more likely to be reached and to respond. Information about approached ICUs that did not respond was not available. Therefore, some of the findings may underestimate the true deficiencies in ICUs across Asia. We surveyed staffing quantitatively but we did not assess the quality of training and competencies. We also did not assess patient outcomes, cost, or resource utilization. This survey highlights considerable variation in the organization and staffing among Asian ICUs. This variation was related to the hospital funding source, hospital size, and country income. In particular, this survey reveals suboptimal infection control and prevention infrastructure, such as inadequate hand hygiene setup and lack of single and negative-pressure rooms in many Asian ICUs posing concerns on the adequate management of life-threatening infections if they become widespread.

REFERENCES

1. World Population Statistics. Population of Asia 2016. Available at: http://www.worldpopulation- statistics.com/population-of-asia-2014/. Accessed June 3, 2016 2. Adhikari NK, Fowler RA, Bhagwanjee S, et al: Critical care and the global burden of critical illness in adults. Lancet 2010; 376: 1339–1346 3. The United Nations Economic and Social Commission for Asia and the Pacific (ESCAP). Disasters in Asia and the Pacific: 2014 Year in Review. Available at: http://www.unescap.org/resources/disastersasia-and-pacific-2014-year-review-0. Accessed June 3, 2016 4. Millman AJ, Havers F, Iuliano AD, et al: Detecting spread of Avian Influenza A(H7N9) Virus beyond China. Emerg Infect Dis 2015; 21:741–749 5. Zhou G, Yan G: Severe acute respiratory syndrome epidemic in Asia. Emerg Infect Dis 2003; 9:1608–1610 6. Oboho IK, Tomczyk SM, Al-Asmari AM, et al: 2014 MERS-CoV outbreak in Jeddah–a link to health care facilities. N Engl J Med 2015; 372:846–854 XXX 2016 • Volume XX • Number XXX


Online Clinical Investigation 7. Vincent JL, Marshall JC, Namendys-Silva SA, et al; ICON investigators: Assessment of the worldwide burden of critical illness: The intensive care over nations (ICON) audit. Lancet Respir Med 2014; 2:380–386 8. Seto WH, Tsang D, Yung RW, et al; Advisors of Expert SARS group of Hospital Authority: Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS). Lancet 2003; 361:1519–1520 9. Pronovost PJ, Jenckes MW, Dorman T, et al: Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA 1999; 281:1310–1317 10. Li TC, Phillips MC, Shaw L, et al: On-site physician staffing in a community hospital intensive care unit. Impact on test and procedure use and on patient outcome. JAMA 1984; 252:2023–2027 11. Carson SS, Stocking C, Podsadecki T, et al: Effects of organizational change in the medical intensive care unit of a teaching hospital: A comparison of ‘open’ and ‘closed’ formats. JAMA 1996; 276:322–328 12. Multz AS, Chalfin DB, Samson IM, et al: A “closed” medical intensive care unit (MICU) improves resource utilization when compared with an “open” MICU. Am J Respir Crit Care Med 1998; 157:1468–1473 13. Hanson CW 3rd, Deutschman CS, Anderson HL 3rd, et al: Effects of an organized critical care service on outcomes and resource utilization: A cohort study. Crit Care Med 1999; 27:270–274 14. Phua J, Koh Y, Du B, et al; MOSAICS Study Group: Management of severe sepsis in patients admitted to Asian intensive care units: Prospective cohort study. BMJ 2011; 342:d3245 15. Phua J, Joynt GM, Nishimura M, et al; ACME Study Investigators and the Asian Critical Care Clinical Trials Group: Withholding and withdrawal of life-sustaining treatments in intensive care units in Asia. JAMA Intern Med 2015; 175:363–371 16. Groeger JS, Strosberg MA, Halpern NA, et al: Descriptive analysis of critical care units in the United States. Crit Care Med 1992; 20:846–863 17. Gay PC, Dellinger RP, Shelhamer JH, et al: The practice of critical care medicine. A national survey report. ACCP Council on Critical Care. Chest 1993; 104:271–278 18. Brilli RJ, Spevetz A, Branson RD, et al; American College of Critical Care Medicine Task Force on Models of Critical Care Delivery. The American College of Critical Care Medicine Guidelines for the Defintion of an Intensivist and the Practice of Critical Care Medicine: Critical care delivery in the intensive care unit: Defining clinical roles and the best practice model. Crit Care Med 2001; 29:2007–2019 19. The Leapfrog Group: ICU Physician Staffing. Available at: http:// www.leapfroggroup.org/sites/default/files/Files/IPS%20Fact%20 Sheet.pdf. Accessed June 3, 2016 20. Levy MM, Rapoport J, Lemeshow S, et al: Association between critical care physician management and patient mortality in the intensive care unit. Ann Intern Med 2008; 148:801–809 21. The World Bank. Changes in Country Classifications Classifications. Available at: http://data.worldbank.org/news/2010-GNI-incomeclassifications. Accessed June 3, 2016 22. 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–e9 23. Rhodes A, Ferdinande P, Flaatten H, et al: The variability of critical care bed numbers in Europe. Intensive Care Med 2012; 38: 1647–1653 24. 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 25. Sirio CA, Tajimi K, Tase C, et al: An initial comparison of intensive care in Japan and the United States. Crit Care Med 1992; 20:1207–1215 26. Yeolekar ME, Mehta S: ICU care in India–status and challenges. J Assoc Physicians India 2008; 56:221–222


27. Lee N, Hui D, Wu A, et al: A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003; 348: 1986–1994 28. Memish ZA, Zumla AI, Assiri A: Middle East respiratory syndrome coronavirus infections in health care workers. N Engl J Med 2013; 369:884–886 29. Teltsch DY, Hanley J, Loo V, et al: Infection acquisition following intensive care unit room privatization. Arch Intern Med 2011; 171:32–38 30. Caruso P, Guardian L, Tiengo T, et al: ICU architectural design affects the delirium prevalence: A comparison between single-bed and multibed rooms. Crit Care Med 2014; 42:2204–2210 31. Zahar JR, Garrouste-Orgeas M, Vesin A, et al: Impact of contact isolation for multidrug-resistant organisms on the occurrence of medical errors and adverse events. Intensive Care Med 2013; 39: 2153–2160 32. World Health Organization: WHO Guidelines on Hand Hygiene in Health Care: First Global Patient Safety Challenge Clean Care Is Safer Care. Geneva, 2009. Available at: http://www.who.int/gpsc/5may/ tools/who_guidelineshandhygiene_summary.pdf. Accessed June 3, 2016 33. Angus DC, Shorr AF, White A, et al; Committee on Manpower for Pulmonary and Critical Care Societies (COMPACCS): Critical care delivery in the United States: Distribution of services and compliance with Leapfrog recommendations. Crit Care Med 2006; 34:1016–1024 34. Vincent JL, Suter P, Bihari D, et al: Organization of intensive care units in Europe: Lessons from the EPIC study. Intensive Care Med 1997; 23:1181–1184 35. Azoulay E, Timsit JF, Sprung CL, et al; Conflicus Study Investigators and for the Ethics Section of the European Society of Intensive Care Medicine: Prevalence and factors of intensive care unit conflicts: The conflicus study. Am J Respir Crit Care Med 2009; 180:853–860 36. Checkley W, Martin GS, Brown SM, et al; United States Critical Illness and Injury Trials Group Critical Illness Outcomes Study Investigators: Structure, process, and annual ICU mortality across 69 centers: United States Critical Illness and Injury Trials Group Critical Illness Outcomes Study. Crit Care Med 2014; 42:344–356 37. Wilcox ME, Chong CA, Niven DJ, et al: Do intensivist staffing patterns influence hospital mortality following ICU admission? A systematic review and meta-analyses. Crit Care Med 2013; 41:2253–2274 38. Needleman J, Buerhaus P, Pankratz VS, et al: Nurse staffing and inpatient hospital mortality. N Engl J Med 2011; 364:1037–1045 39. Shekelle PG: Nurse-patient ratios as a patient safety strategy: A systematic review. Ann Intern Med 2013; 158:404–409 40. Chan ZC, Tam WS, Lung MK, et al: A systematic literature review of nurse shortage and the intention to leave. J Nurs Manag 2013; 21:605–613 41. Sevransky JE, Checkley W, Herrera P, et al; United States Critical Illness and Injury Trials Group-Critical Illness Outcomes Study Investigators: Protocols and hospital mortality in critically ill patients: The United States Critical Illness and Injury Trials Group Critical Illness Outcomes Study. Crit Care Med 2015; 43:2076–2084 42. Halpern NA, Pastores SM: Critical care medicine in the United States 2000-2005: An analysis of bed numbers, occupancy rates, payer mix, and costs. Crit Care Med 2010; 38:65–71 43. Fowler RA, Fletcher T, Fischer WA 2nd, et al: Caring for critically ill patients with ebola virus disease. Perspectives from West Africa. Am J Respir Crit Care Med 2014; 190:733–737 44. Firth P, Ttendo S: Intensive care in low-income countries–a critical need. N Engl J Med 2012; 367:1974–1976 45. Arabi Y, Al Shimemeri A: Critical care medicine in Saudi Arabia. East Mediterr Health J 2006; 12:225–230

www.ccmjournal.org


9