If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Intensive Care Unit, Redcliffe Hospital, 4020, Redcliffe, Queensland, AustraliaFaculty of Medicine, University of Queensland, 4029, Brisbane, Queensland, Australia
Faculty of Medicine, University of Queensland, 4029, Brisbane, Queensland, AustraliaIntensive Care Units, Caboolture and Prince Charles Hospitals, Queensland, AustraliaThe George Institute for Global Health, Sydney, New South Wales, AustraliaUniversity of New South Wales, Sydney, New South Wales, Australia
Intensive Care Unit, Redcliffe Hospital, 4020, Redcliffe, Queensland, AustraliaFaculty of Medicine, University of Queensland, 4029, Brisbane, Queensland, Australia
ANZICS Centre for Outcome and Resource Evaluation, Camberwell, Victoria, AustraliaDepartment of Intensive Care, The Alfred Hospital, Prahran, Victoria, AustraliaThe Australian and New Zealand Intensive Care - Research Centre, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
ANZICS Centre for Outcome and Resource Evaluation, Camberwell, Victoria, AustraliaIntensive Care Unit, St John of God Hospital, Perth, Western Australia, AustraliaSchool of Medicine, University of Western Australia, Perth, Western Australia, Australia
The objective of this study was to describe family visitation policies, facilities, and support in Australia and New Zealand ( ANZ ) intensive care units (ICUs).
Methods
A survey was distributed to all Australian and New Zealand ICUs reporting to the Australian and New Zealand Intensive Care Society Centre for Outcomes and Resources Evaluation Critical Care Resources (CCR) Registry in 2018. Data were obtained from the survey and from data reported to the CCR Registry. For this study, open visiting (OV) was defined as allowing visitors for more than 14 h per day.
Setting and participants
This study included all Australian and New Zealand ICUs reporting to CCR in 2018.
Main outcome measures
The main outcome measures were family access to the ICU and visiting hours, characteristics of the ICU waiting area, and information provided to and collected from the relatives.
Findings
Fifty-six percent (95/170) of ICUs contributing to CCR responded, representing 44% of ANZ ICUs and a range of rural, metropolitan, tertiary, and private ICUs. Visiting hours ranged from 1.5 to 24 h per day, with 68 (72%) respondent ICUs reporting an OV policy, of which 64 (67%) ICUs were open to visitors 24 h a day. A waiting room was part of the ICU for 77 (81%) respondent ICUs, 74 (78%) reported a separate dedicated room for family meetings, and 83 (87%) reported available social worker services. Most ICUs reported facilities for sleeping within or near the hospital. An information booklet was provided by 64 (67%) ICUs. Only six (6%) ICUs required personal protective equipment for all visitors, and 76 (80%) required personal protective equipment for patients with airborne precautions.
Conclusions
In 2018, the majority of ANZ ICUs reported liberal visiting policies, with substantial facilities and family support.
Visiting policies including visiting hours, number of visitors, and support provision for relatives of patients admitted to the intensive care unit (ICU) are important components of family-centred critical care. Open visiting (OV), in which visiting hours have little to no restriction, may be of benefit to patients and their relatives by improving access, facilitating communication, and reducing stress.
Closed visiting, in which visiting hours are restricted, may be preferred by some ICU clinicians so that time can be focused on delivering patient care while relatives feel empowered to continue to attend to their own needs. While perception of an increased workload with OV has been reported,
The College of Intensive Care Medicine (CICM) of ANZ recommends a separate waiting area (with drinks dispenser, radio, television, and comfortable seating desirable), a separate interview room, and a separate area for distressed relatives, with consideration for overnight rooms for relatives.
In a statement on partnering with families in critical care, the Australian College of Critical Care Nurses recommends ICUs have facilities to provide family access to critical care areas at all times, including outside business hours.
They should have waiting areas close to the ICU, with a clear method to gain access, sufficient facilities such as a bathroom, telephone, comfortable sitting, and sustenance. They also recommend having nearby accommodation for families where possible.
The aim of this site-level survey was to describe visiting policies in ANZ ICUs and to compliment a survey of ICU staff members and their perceptions of visiting policies are published in the same issue of this journal.
We conducted a web-based survey to investigate current practices with regards to visiting policies in ANZ ICUs. Ethical approval was granted by the Ethics committee of the Alfred Hospital in Melbourne, Australia, reference 638/18. We report this study following good practice in the conduct and reporting of survey research from the Enhancing the Quality and Transparency of Health Research (EQUATOR) network.
The Welcome-CCR survey was designed to focus on three specific areas: (i) family access to the ICU and visiting hours, (ii) the characteristics of the ICU waiting area and the possibility to sleep within or near the hospital, and (iii) information provided and collected to and from the relatives.
To reduce the data burden on respondents and ensure high-quality information on resourcing, the survey was designed to be linked to ICU resource data reported to the Australian and New Zealand Intensive Care Society Centre for Outcomes and Resources Evaluation (ANZICS CORE) Critical Care Resources (CCR) Registry. The CCR invites annual, aggregate, ICU-level data submission from all Australian and New Zealand ICUs each financial year.
Participation is not mandatory; all ICUs, general and specialised ICUs of all subspecialities of intensive care medicine inclusive of adult, mixed, and paediatric ICUs, in ANZ are invited to participate. CCR annual reports are available on the ANZICS CORE webpage and provide data on resources, activity, and processes of care for participating ICUs, including the number of ICU beds, admissions and readmissions, occupancy, hours of ventilation, workforce data, and information related to ICU safety and quality indicators.
Our target was to design a survey that would require a maximum of 5 min to complete. The data collection tool was drafted to answer the three domains of focus of the study by the primary investigator (A.T.) on a text editing software. It was reviewed by another author (E.L.) to ensure the questions would match and not overlap the data collected through CCR. The draft data collection tool was circulated via email to all the investigators and commented during one video conference call and three rounds of email until the tool was finalised. The survey contained a total of 15 questions. Five questions focussed on waiting room and available services, including facilities to sleep on site. Six questions were about visiting, including visiting hours and the possibility to extend them, who is allowed as a visitor, and the use of personal protective equipment (PPE). Three questions regarding communication, including the availability of an information booklet, the evaluation of family satisfaction, and the use of ICU patient diaries, were included. The text document of the data collection tool can be downloaded with the electronic supplement. This document was used to design an online formulary (SurveyMonkey, San Mateo, CA, USA) to be disseminated for data collection. Pilot testing of the online survey was completed by all the investigators before dissemination. No data were recorded during the pilot testing phase. Reliability and validity testing were not performed.
Completed surveys were linked by ANZICS staff to existing CCR data via a unit identifier that was removed before data analysis.
2.2 Participants
All Australian ICUs participating in the ANZICS CORE CCR registry were invited to participate, and all ICUs that completed the CCR survey in 2018 were included in the study after data matching.
2.3 Survey administration
The primary contact for each ICU in the CCR registry, usually the ICU director or nurse unit manager, was contacted via email and invited to participate. This was separate to the annual CCR survey and was launched after its completion. A total of three email reminders were sent between February and May 2019.
2.4 Definitions
We defined ICU size by quartiles of available, funded, and equipped ICU beds as small (less than 6), moderate (7–10), large (10–18), and very large (more than 18 ICU beds). An OV policy was defined as allowing visitors for more than 14 h per day, matching the definition used in the Welcome-ICU staff survey.
Survey data were downloaded from SurveyMonkey and collated into a single encrypted database by ANZICS CORE. After linking this information to existing information in the CCR registry and removing identifiers from the database, it was securely transferred to the principal investigator and imported into SAS Software, version 9.4 (SAS Institute Inc., Cary, NC, USA) for data management and analysis.
Staffing characteristics were indexed on the number of available beds by dividing the full-time equivalent for each staff characteristic by the number of available beds. Turnover was defined by indexing the number of yearly admissions to the number of available beds.
Continuous data were summarised as median and interquartile range, and categorical data as proportions. Differences in categorical variables were tested using a Pearson chi-squared test or Fisher's exact test when appropriate. Continuous variables were analysed using the Student's T test or the Mann–Whitney U test depending on parametricity. Statistical significance was defined as p < 0.05.
3. Results
We contacted 213 of the 217 ICUs in ANZ. Of those, 170 had participated in the ANZICS CORE CCR registry for the same calendar year (Fig. 1). We obtained data from 110 ICUs, of which 95 had complete survey data and matching registry data. This analysed sample represents 56% (95/170) of the ICUs contributing to CCR and 44% (95/217) of ICUs operating in ANZ. As shown in Table 1, they were variably distributed across ANZ, with no differences in visiting policies across jurisdictions (p = 0.8).
Fig. 1Flowchart of included ICUs. CCR, Critical Care Resources; ICU, intensive care unit.
States with 1 response have been merged with another state to protect anonymity. Hospital beds are available to provide overnight accommodation. ICU beds denote the number of ICU beds with advanced life support capability that is fully staffed and funded. ICU admissions denote the number of ICU admissions for the 2017–18 financial year. Categorical data are expressed as n (%), and continuous data as mean ± SD for normally distributed data. Level 3/PICU category includes all Level 3 and Paediatric ICUs.
CICM, College of Intensive Care Medicine; FTE, full-time equivalent; ICU, intensive care unit; PICU, paediatric intensive care unit; SD, standard deviation.
As shown in Table 1, we obtained responses from a range of rural, metropolitan, tertiary, and private ICUs. The distribution of ICU size was broad, with 27% ICUs reporting less than seven available beds, 26% seven to 10 available beds, 23% 11 to 18 beds, and 25% more than 18 beds.
3.2 Visiting hours and family access
Visiting hours varied from 1.5 to 24 h per day, with 64 (67%) ICUs open to visitors 24 h a day and 68 (71%) with an OV policy (Fig. 2, Fig. 3). The median visiting hours was 24 with an interquartile range of 12–24 h per day. As shown in Table 1, OV was more frequent in CICM Level 3 ICUs (large units) (p = 0.02). The difference was not statistically significant when comparing ICUs by hospital or ICU size, crude mortality, hospital classification, or the number of admissions. Case-mix of the ICUs, as evidenced by admission of trauma, spinal, neurosurgical, burns, cardiothoracic, extra-corporeal membrane oxygenation (ECMO), cancer services or transplantation patients was not associated with increased probability of an ICU operating an OV policy (data not shown).
Fig. 2Distribution of ICUs open to visitors according to time of the day. ICU, intensive care unit.
Because there was only one factor associated with OV in the univariate analysis, we did not conduct a logistic regression to further investigate possible associations.
Most (80, 84%) allowed a maximum of two, a few allowed three (6, 6%), or four (3, 3%) or an unlimited (6, 6%) number of simultaneous visitors for a patient. Circumstances where visiting hours or the number of visitors could be increased are highlighted in Table 3.
3.3 Waiting room and facilities
Most (77, 81%) had a waiting room as part of the ICU; for 13 (14%), it was elsewhere in the hospital, and five (5%) did not have a waiting area for ICU patients’ relatives (Table 2). There was no difference in the availability or location of the waiting room between small, medium, large, or very large ICUs (p = 0.3).
Table 2Waiting room and facilities.
Waiting room, n = 95
Part of the ICU
77 (81%)
Shared with relatives of patients admitted elsewhere
11 (12%)
In a different part of the hospital than the ICU
2 (2%)
No waiting area available
5 (5%)
Facilities available, n = 77∗
A vending machine
18 (23%)
Coffee or tea making
55 (71%)
A microwave or other cooking facilities
26 (34%)
A bathroom
35 (45%)
A shower
8 (10%)
How does a family member inform the ICU of their presence, n = 95
A ring bell
28 (29%)
Intercom
47 (49%)
Videophone
35 (37%)
Receptionist
30 (32%)
Is there a separate dedicated meeting room, n = 95
74 (78%)
Where are families permitted to sleep? n = 95
Dedicated room within the hospital
16 (17%)
Hospital provided accommodation
21 (22%)
Hospital subsidised accommodation
24 (25%)
Charitable organisation sponsored accommodation
18 (19%)
Waiting room, no bed provided
49 (52%)
Waiting room, bed provided
10 (11%)
Families are never permitted to sleep anywhere in the hospital
6 (6%)
Data are presented as n (%) for all values.
∗Available facilities for those who report having a waiting room as part of the ICU.
Most (74, 78%) had a separate room dedicated to meetings with the families. This was more common for those who had a dedicated waiting room (86% vs 44%, p = 0.0005) and in CICM Level 3 (large tertiary units) compared with Level 2 or 1 ICUs (93% vs 67% vs 61%, p = 0.007).
Larger ICUs were more likely to have separate meeting rooms (65% vs 62% vs 86% vs 100%, p = 0.005) for up to six vs 10 vs 18 or more than 18 available ICU beds.
As shown in Table 2, most reported the possibility to provide, sponsor, or subsidise accommodation for relatives. Most (83, 87%) reported the availability of social work services in the ICU, with a wide range of full-time equivalent.
3.4 Information, satisfaction, and intensive care diary
An information booklet was provided by 64 (67%). This was more frequent for CICM Level 3 (large tertiary units) (Level 1: 46%, Level 2:57%, Level 3: 83%, p = 0.01), but similar when comparing ICUs by size (the number of available beds) (p = 0.3) or their visitation policies (p = 0.9).
Family satisfaction was always evaluated for 21 (22%), most of the time for 25 (26%), sometimes for 45 (47%), and never for four (4%). This was similar for open and closed ICUs (p = 0.4)
An intensive care diary was offered for all patients by five (5%) and for selected patients by 22 (23%). This was not different according to the visiting policies (p = 0.7) or ICU level status (p = 0.09).
3.5 Personal protective equipment
PPE was required as a standard for visitors of all patients in six (6%) of the ICUs; 61 (64%) of the ICUs required PPE for visitors of patients with drug-resistant pathogens, 74 (78%) for visitors of patients with droplet precautions, 76 (80%) for visitors of patients with airborne precautions and 64 (67%) for visitors of patients with clostridium difficile.
4. Discussion
The main purpose of this study was to describe visiting policies in ANZ ICUs, and the results indicated a liberal approach to patient visitation, with most ICUs offering OV to the friends and relatives of their patients.
OV has been variably defined, from unrestricted visiting to an increase in the previous restrictions on visiting policies, with wide variations between different reports.
We arbitrarily defined OV as allowing visits during 14 or more hours of the day. Of the 95 units who responded to the survey, 72% had an OV status and 67% permitted patient visits at any time of the day or night (24 h). As described in Table 4, restrictions to visitation were present in all countries where it has been investigated, with many allowing only brief and infrequent visits to ICU patients. Before making any comparisons, it is important to acknowledge the high risk of bias as there are no worldwide consolidated data on ICU visiting policies. Available reports used various methodologies and definitions, with most being surveys and spanned over a period of 14 years, only describing an incomplete and possibly biased picture of nine countries.
Table 4Visiting hours as reported in other settings.
Summary of visiting policies as reported in the medical literature. Percentages of unrestricted visiting hours represent the reported % of ICUs offering unrestricted visiting hours to the relatives of ICU patients as reported in the publication. Year denotes the year the data were collected. If not available, the year of publications is reported.
suggested OV was safe with no differences in mortality or ICU-acquired infections in the pooled results of the available studies. Their results suggested the frequency of delirium and anxiety may be reduced in patients exposed to the OV. One study suggested a significant risk of staff burnout as it increased from 34.5% to 43.6% after the implementation of OV in eight Italian ICUs.
This was not confirmed in a cluster randomised trial of 36 ICUs, 1685 patients, 1295 relatives, and 826 clinicians conducted by the same group in Brazil.
They successfully delivered the intervention as the mean duration of visits was 3.4 h higher in the OV than in the CV group (4.8 vs 1.4 h per day). The incidence of delirium was similar between both groups. They confirmed safety outcomes as there were no differences in mortality, ICU-acquired infections, or staff burnout. They demonstrated decreased anxiety and depression scores in family members exposed to OV, suggesting substantial benefit to this strategy.
There are limited data to explain the differences between visiting policies, within or between countries. In this setting and with society guidelines recommending OV, we may speculate that maintaining a CV is the result of a combination of societal attitudes, cultural factors, and workplace behaviours, with healthcare resourcing and staffing. Acknowledging the risk of error in such comparisons, where data are available, we note a temporal increase in reported visiting hours in Italy, France, and the USA.
This also indicates a baseline of restrictive policies. It is possible that ICUs have been closed or very restrictive to visitors when they were first built, and some may have maintained these policies because “it has always been like that”. Cultural factors were described by Speroni et al.
In a survey of Swiss ICUs, they compared the length of permitted visiting in different linguistic regions which define different cultural groups within Switzerland. The median visiting time was 6.5 h in French-speaking areas, 8.0 h in the Italian-speaking region, and 8.8 h in the German-speaking region.
This difference was the only statistically significant predictor of longer versus shorter visiting times and was interpreted by the authors as a reflection of cultural differences between those regions. Similarly, Giannini et al
reported in a survey of Italian ICUs that regional areas of the country and higher volume ICUs were the only two factors independently associated with more restricted visiting. This was interpreted by the authors as the result of cultural differences and possibly the result of conservative and paternalistic attitudes of the staff in those areas.
While no direct evidence is available, it is likely that regional and cultural factors influence healthcare attitudes with regards to family presence within the critical care environment.
Implementation or maintenance of an OV may also be related to the difficulties in implementing any change in healthcare systems combined with the reluctance or negative attitudes of staff towards modifications of their current visiting policy.
While the risk of burnout was not confirmed in the only large-scale randomised controlled trial published to date, it has been reported both as a concern and a complication after policy change,
and is a likely barrier to the implementation of OV. ICU clinicians in Australia and internationally have reported that OV may cause workflow interruptions and increased workload.
Additional resources may be required to balance the workload caused by increased family presence. Research investigating the relative resource requirements of open and closed ICU visitation is lacking. However, in an analysis of the ORCHESTRA study of organisational characteristics of Brazilian ICUs, liberal visitation policies were more frequent in ICUs with lower standardised mortality ratio and high efficiency in resource use.
In our study, OV was more frequent in CICM Level 3 ICUs than in Level 1 or 2 units. In ANZ , Level 3 ICUs represent tertiary referral units, and beyond their role in managing complex multisystem life support, they need to demonstrate a commitment to academic education and research.
Although we may speculate that such institutions, with an increased commitment to education and research, possibly have higher adherence to guidelines, there is insufficient supporting evidence. Any such differences may be due to increased resourcing for Level 3 ICUs or other unmeasured factors. Other than ICU level, no variables were significantly associated with the open or restrictive visitation policy of the ICUs that participated in our study.
Widely reported availability of a waiting room within the ICU, facilities and sustenance for visitors, possibilities for accommodation, and separate waiting and meeting areas were in keeping with national recommendations.
Only five ICUs reported not having a waiting room for visitors of ICU patients anywhere within the hospital. Those were all in either private or rural/remote hospitals. It is unknown if not having a waiting room was due to architectural constraints and if they would be included in any possible future development plans.
Information to families, evaluation of their satisfaction, and the availability of social work services within the ICU were, in general, reported by most ICUs. Similar to OV, they are cornerstones of family-centred critical care.
Importantly, a large proportion (45%) of the smaller units (CICM Levels 1 and 2) did not provide relatives with an information booklet. It is unknown if this information was provided by any other means such as the hospital website, social media, or redirecting the relatives to other resources. Similarly, half of the ICUs reported evaluating family satisfaction only sometimes or never. Collecting this information is an important tool for targeting quality improvement initiatives.
Our survey does not detail if there were any other means of feedback for families, such as an evaluation at the hospital or health service level. We did not collect specific information on which tools were used for information or evaluation of satisfaction, their contents and if they were validated, the mode of distribution, or the reasons why they were not available. This could be investigated by a survey or mixed-methods study dedicated to family feedback in critical care.
Since this study was conducted, the COVID-19 pandemic has emerged and resulted in significant changes to various aspects of intensive care practice – including visitation. Visits to critically ill patients have been strictly limited or stopped altogether to minimise the risk of COVID-19 transmission and to conserve limited PPE supplies.
ICUs have been extended outside of their usual walls, waiting and meeting rooms have been repurposed in clinical or storage areas, and nonvital equipment such as coffee or vending machines have been removed as potential fomite and surface hazards. It is unknown how information and evaluation of family satisfaction has been pursued during this period of restrictions.
Before COVID-19, requiring visitors to wear PPE outside of defined infection control indications could have been described as a “ritual from the past” and was not supported by scientific evidence.
In our survey, only 6% of ICUs required PPE for all visitors, while others adjusted requirements depending on the patients’ infectious status. However, currently, owing to COVID-19, anecdotal evidence points to a requirement for high-level PPE for all visitors, when they are allowed inside the hospital or the ICU.
Availability of PPE and the ability of patient relatives to safely don and doff PPE and prevent transmission will be significant factors to consider for future patient visitation policies. There will be a need to balance benefits to patients and relatives of open visitation, with the need to conserve PPE and maintain staff, patient, and family safety in a world with COVID-19. This survey provides a baseline from which changes necessitated by the COVID-19 pandemic will be measured as part of the ANZICS and the George Institute Point Prevalence Program in 2020.
Limitations of this work include the voluntary nature of the survey and a response rate of 44% of ICUs in ANZ, which may have caused sampling bias, potentially self-selecting respondents with an interest in OV and whose ICUs have implemented the strategy more frequently than nonrespondents. There may also be some desirability bias, where individual ICUs may have reported more permitted visiting hours than what is offered to visitors. The combination of those two biases may have led to this document reporting a higher frequency of OV than is actually the case in ANZ ICUs.
The lack of onsite monitoring left a risk of erroneous data. This was compounded by collecting surveys from the usual ANZICS CORE contact, usually a senior staff member with knowledge of policies and processes. The design of the survey did not include reliability or validity testing. This may have led to collecting erroneous data and a subsequent decrease in the internal validity of our results. This is balanced by the factual nature of this survey because all questions relate to organisational and policy items with minimal subjective elements.
Our findings may not be generalisable outside of ANZ where standards dictate a nurse-to-patient ratio of 1:1 for ventilated patients, 1:2 for nonventilated lower acuity patients, and at least one supernumerary team-leader nurse per shift.
These ratios are higher than those reported from other regions. This may decrease the need for additional resources to provide OV, and our findings may not be comparable with those from other countries, which can only be overcome with an international survey.
5. Conclusion
Most ANZ ICUs had an OV policy, with many open to visiting 24 h a day. OV was more likely in CICM Level 3 accredited ICUs but was not associated with any of the ICU characteristics or resourcing variables that were analysed. There was flexibility to liberalise policies in multiple clinical situations. Waiting rooms with services, separate meeting rooms, social work services, and information to and evaluation of family satisfaction were widely reported demonstrating the commitment to family-centred critical care in ANZ.
While the impact of COVID-19 on visiting policies in ANZ ICUs remains to be studied, this study shows how important family-centred critical care is in our culture. This highlights the urgency to devise policies that will allow safe visiting and OV as an integral component of critical care in ANZ ICUs.
CRediT authorship contribution statement
Alexis Tabah: Conceptualisation, Methodology, Validation, Formal analysis, Investigation, Data Curation, Writing – original draft, Writing – review & editing, Visualisation. Mahesh Ramanan: Conceptualisation, Methodology, Validation, Formal analysis, Investigation, Data Curation, Writing – original draft, Writing – review & editing, Visualisation. Rachel L Bailey: Conceptualisation, Methodology, Validation, Investigation, Writing – review & editing. Shaila Chavan: Software, Investigation, Resources, Data Curation, Writing – review & editing, Project administration. Stuart Baker: Writing – review & editing. Sue Huckson: Software, Investigation, Resources, Data Curation, Writing – review & editing, Project administration. David Pilcher: Conceptualisation, Methodology, Validation, Investigation, Resources, Writing – review & editing, Supervision. Edward Litton: Conceptualisation, Methodology, Validation, Investigation, Resources, Data Curation, Writing – review & editing, Visualisation, Supervision.
Funding acknowledgements
This project has been realised by the authors without specific funding. In Kind logistical and information technology support was provided by ANZICS CORE . Sue Huckson and Shaila Chavan are employed by ANZICS CORE.
Conflicts of interests
Dr Alexis Tabah has nothing to disclose, A/Prof Edward Litton has nothing to disclose, Dr M Ramanan has nothing to disclose, Prof David Pilcher has nothing to disclose, Sue Huckson has nothing to disclose, Shaila Chavan has nothing to disclose, Dr Stuart Baker has nothing to disclose.
ANZICS CORE Acknowledgements
The authors and the ANZICS CORE management committee would like to thank clinicians, data collectors and researchers at the following contributing sites: Albury Wodonga Health ICU, Alfred Hospital ICU, Armadale Health Service ICU, Auckland City Hospital CV ICU, Auckland City Hospital DCCM, Austin Hospital ICU, Ballarat Health Services ICU, Bankstown-Lidcombe Hospital ICU, Bathurst Base Hospital ICU, Bendigo Health Care Group ICU, Bowral Hospital HDU, Box Hill Hospital ICU, Buderim Private Hospital ICU, Bundaberg Base Hospital ICU, Caboolture Hospital ICU, Calvary Mater Newcastle ICU, Campbelltown Hospital ICU, Canberra Hospital ICU, Canterbury Hospital ICU, Concord Hospital (Sydney) ICU, Dandenong Hospital ICU, Epworth Freemasons Hospital ICU, Epworth Geelong ICU, Epworth Hospital (Richmond) ICU, Fairfield Hospital ICU, Fiona Stanley Hospital ICU, Flinders Medical Centre ICU, Footscray Hospital ICU, Gold Coast University Hospital ICU, Gosford Hospital ICU, Goulburn Base Hospital ICU, Greenslopes Private Hospital ICU, Hawkes Bay Hospital ICU, Hervey Bay Hospital ICU, Holy Spirit Northside Hospital ICU, Hornsby Ku-ring-gai Hospital ICU, Ipswich Hospital ICU, John Hunter Hospital ICU, Joondalup Health Campus ICU, Kempsey District Hospital HDU, Knox Private Hospital ICU, Lake Macquarie Private Hospital ICU, Latrobe Regional Hospital ICU, Launceston General Hospital ICU, Lismore Base Hospital ICU, Liverpool Hospital ICU, Mackay Base Hospital ICU, Maitland Hospital HDU/CCU, Manning Rural Referral Hospital ICU, Maroondah Hospital ICU, Mater Adults Hospital (Brisbane) ICU, Mater Private Hospital (Brisbane) ICU, Mildura Base Hospital ICU, Monash Medical Centre-Clayton Campus ICU, Nelson Hospital ICU, Nepean Hospital ICU, Nepean Private Hospital ICU, Noosa Hospital ICU, North Shore Hospital ICU, Northeast Health Wangaratta ICU, Orange Base Hospital ICU, Palmerston North Hospital ICU, Peninsula Private Hospital ICU, Port Macquarie Base Hospital ICU, Prince of Wales Hospital (Sydney) ICU, Princess Alexandra Hospital ICU, Queen Elizabeth II Jubilee Hospital ICU, Redcliffe Hospital ICU, Rockhampton Hospital ICU, Rotorua Hospital ICU, Royal Brisbane and Women's Hospital ICU, Royal Children's Hospital (Melbourne) PICU, Royal Melbourne Hospital ICU, Shoalhaven Hospital ICU, St Andrew's Hospital Toowoomba ICU, St Andrew's War Memorial Hospital ICU, St George Hospital (Sydney) ICU, St John of God Hospital (Bendigo) ICU, St Vincent's Hospital (Melbourne) ICU, St Vincent's Hospital (Sydney) ICU, St Vincent's Private Hospital Fitzroy ICU, Sunnybank Hospital ICU, Sunshine Coast University Hospital ICU, Sydney Adventist Hospital ICU, Sydney Children's Hospital PICU, Tamworth Base Hospital ICU, Tauranga Hospital ICU, The Memorial Hospital (Adelaide) ICU, The Northern Hospital ICU, The Prince Charles Hospital ICU, The Townsville Hospital ICU, The Townsville Hospital ICU-paeds, The Wesley Hospital ICU, Timaru Hospital ICU, Toowoomba Hospital ICU, Tweed Heads District Hospital ICU, Wairarapa Hospital HDU, Wakefield Hospital (NZ) ICU, Warringal Private Hospital ICU, Western District Health Service (Hamilton) ICU, Westmead Hospital ICU, Whanganui Hospital ICU, Wollongong Hospital ICU, Wyong Hospital ICU.
Appendix A. Supplementary data
The following are the supplementary data to this article:
00103-X&id=gr1.jpg){kind=link}
00103-X&id=gr2.jpg){kind=link}
00103-X&id=gr3.jpg){kind=link}