Critical care transfer in an English critical care network: Analysis of 1124 transfers delivered by an ad-hoc system

Abstract

Background

Critical care transfers between hospitals are time critical high-risk episodes for unstable patients who often require urgent lifesaving intervention. This study aimed to establish the scale, nature and safety of current transfer practice in the South West Critical Care Network (SWCCN) in England.

Methods

The SWCCN database contains prospectively collected data in accordance with national guidelines. It was interrogated for all adult (>15 years of age) patients from January 2012 to November 2017.

Results

A total of 1124 inter-hospital transfers were recorded, with the majority (935, 83.2%) made for specialist treatment. The transferring team included a doctor in 998 (88.8%) and nurse in 935 (93.7%) transfers. In 204 (18.1%) transfers, delays occurred, with the commonest cause being availability of transport. Critical incidents occurred in 77 (6.9%).

Conclusions

This is the first published data on the transfer activity of a UK adult critical care network. It demonstrates that current ad-hoc provision is not meeting the longstanding expectations of national guidelines in terms of training, clinical experience and timeliness. The authors hope that this study may inform national conversation regarding the development of National Health Service commissioned inter-hospital transfer services for adult patients in England.

Keywords

Critical care transportation of patients patient transfer retrieval inter-hospital transfer

Introduction

Critical care transfers between hospitals are frequently time critical, high-risk episodes of care for unstable patients who require urgent lifesaving intervention.¹ Concerns regarding the safety of these transfers have been expressed for many years^2–6 and variability in service provision exists despite the publication of national recommendations.^1,7–11 There is no recent published data that provide an accurate estimate of the scale of critical care transfers in England per annum, but historical data estimated 10,000 in Great Britain in 1988¹² and over 11,000 in 1994.¹³ Increasing centralisation of specialist services¹⁴ in the National Health Service (NHS) and the establishment of major trauma networks¹⁵ in the United Kingdom (UK) are likely to have accelerated demand for both emergency transfers to access specialist resources and subsequent repatriation to the referring hospital for ongoing care.^1,15

In the UK, the approach to inter-hospital transfer services varies across the devolved nations. Dedicated transport teams undertake adult critical care transfers in Scotland and Wales,^16,17 and paediatric transfers across the UK.¹⁸ However, adult critical care transfers in England are performed on an ad-hoc basis utilising a range of in- and pre-hospital resources and personnel.^1,3,12,13 Multiple studies over several decades have highlighted operational and training inadequacies with this current ad hoc system.^1,12,13,19 Serious event incidence rates during critical care transfer are reported between 12.5% and 62% of transfers.^2–6 Technical failure, such as monitor, gas supply or vehicle problems are reported in 12.5–45%.^2–6,20 Aiming to address these issues, the Intensive Care Society (ICS) published Guidelines for the Transport of the Critically Ill Adult (3rd Edition 2011).¹ These guidelines provide clear recommendations on the necessary training, equipment, documentation, audit and governance to undertake critical care transfers¹ in accordance with international guidance.^7–9 In England, the responsibility for implementing these standards is devolved to regional Critical Care Networks and the ICS guidelines recommend that ‘critical care networks should consider whether the development and use of dedicated transport teams is appropriate to best meet the transport needs of their patient population’.¹

This study aimed to utilise critical care transfer documentation across an English Critical Care Network to establish the scale, nature and safety of current critical care transfer practice and evaluate the need for the provision of a dedicated specialist transport service.

Methods

The South West Critical Care Network (SWCCN) covers a large mixed urban and rural geographical area of England, within which a resident population of 5.5 million²¹ is served by 14 acute hospitals, including a tertiary centre for medicine and surgery and two major trauma centres, along with a number of smaller NHS and private institutions.^22,23

The SWCCN has prospectively collected a standardised dataset on all network inter-hospital transfers since 2012. Information collected is in accordance with the recommendations of national guidelines¹ and includes demographics, clinical condition, reasons for transfer, condition during transfer and critical incidents encountered. Data are entered into a single database for audit and quality improvement purposes.

The SWCCN database was interrogated for all adult (>15 years of age) critical care transfers from the dates of 3 January 2012 and 24 November 2017. Data were anonymised and statistical analysis performed using Stata Version 14 (StataCorp). We used the Chi-square and Kruskal–Wallis tests to examine for statistical significance of categorical and non-normal continuous data, respectively. This study was registered with the University Hospitals Bristol NHS Foundation Trust Clinical Effectiveness Committee.

For the purposes of this study, time critical transfers were defined as those patients who required immediate specialist surgery or critical care at regional centres.

Results

Patient characteristics and indications for critical care inter-hospital transfer

During the study period of 3 January 2012 to 24 November 2017, 1124 inter-hospital transfers of critically ill adult patients were documented on the SWCCN database, representing a mean of 19 transfers per month. Patient age ranged from 16 to 92 years (median 57, IQR 27 years) and 656 (58.4%) were male. Table 1 provides an overview of the indications for critical care transfer.

Table 1.

Indications for critical care transfer.

Indication for critical care transfer	Number of patients (%)
Specialist treatment	935 (83.2%)
Specialist surgery	619 (66.2%)
Neurosurgery	323 (52.2%)
Trauma surgery	187 (30.2%)
Cardiothoracic surgery	58 (9.4%)
Vascular surgery	15 (2.4%)
Gastrointestinal/liver surgery	15 (2.4%)
Other surgery	21 (3.4%)
Specialist medicine	234 (25.0%)
Cardiology	75 (32.1%)
Neurology	45 (19.2%)
Gastrointestinal/liver	41 (17.5%)
Renal	27 (11.5%)
Respiratory	11 (4.7%)
Other medicine	35 (15.0%)
Burns	13 (1.4%)
Not specified / other	69 (7.4%)
Repatriation	118 (10.5%)
Critical care capacity	49 (4.4%)
Missing data	22 (2.0%)
All transfers	1124 (100%)

Most patient transfers originated in intensive care units (510, 45.4%) and emergency departments (487, 43.3%), with few originating on hospital wards (44, 3.9%), operating theatres (26, 2.3%) or from other areas (7, 0.6%); data were missing for 50 (4.5%) transfers.

Seven hundred forty-three (66.1%) patients were mechanically ventilated, with a greater proportion of those requiring specialist treatment being mechanically ventilated compared with the repatriation group (636, 68.0% vs. 60, 50.9%; p < 0.001). However, a larger proportion of patients being repatriated had tracheostomies compared with those requiring specialist treatment (34, 28.8% vs. 12, 1.3%; p < 0.001). Of those transferred for specialist treatment, 728 (77.9%) were time critical transfers requiring immediate specialist surgery or critical care at regional centres. A greater proportion of the time critical patients were mechanically ventilated compared with those who required non-time critical specialist treatment (520, 71.4% vs. 116, 56.0%; p < 0.001).

Timing of transfers

Transfers occurred throughout the 24-h period (Table 2 and Figure 1) with peaks in the middle and end of the working day. There was a wide range of time between hospital admission and transfer (0–108 days), although the median time was one day (IQR 0 to 3 days). These times varied depending upon the reason for transfer, with those requiring specialist treatment having a median time of 0 days (range 0–95), repatriation patients having a median of 13 days (range 0–108 days) and patients transferred for capacity reasons 0 days (range 0–21).

Figure 1.

Types of transfer by time of day.

Table 2.

Timings of critical care transfers.

Start of critical care transfer	Specialist treatment		Repatriation (n = 118)	Capacity (n = 49)
Start of critical care transfer	Time critical (n = 728)	Non-time critical (n = 207)	Repatriation (n = 118)	Capacity (n = 49)
In hours* (08:00–17:59)	256 (35.2%)	111 (53.6%)	87 (73.7%)	9 (18.4%)
Out of hours* (18:00–07:59)	335 (46.0%)	55 (26.6%)	15 (12.7%)	34 (69.4%)
Missing data	137 (18.8%)	41 (19.8%)	16 (13.6%)	6 (12.2%)

p < 0.001 for differences between transfer indication groups.

Critical care escorts

The transfer team for 998 (88.8%) patients included a doctor and for 710 (63.2%) a nurse, in addition to the ambulance crew (Table 3); 99.8% (996/998) of escorting doctors and 71.8% (510/710) of escorting nurses had received training in the transfer of critically ill patients; 93.7% (935/998) of escorting doctors had an anaesthesia or intensive care medicine background and the majority of nurses were of intensive care background (450, 63.4%). Table 3 gives an overview of escorting doctors’ and nurses’ training, grouped by transfer indication.

Table 3.

Transfer team composition and training.

	Specialist treatment		Repatriation	Capacity	p
	Time critical	Non-time critical	Repatriation	Capacity	p
Doctor escort n (%)	678 (93.1%)	161 (77.8%)	100 (84.8%)	47 (95.9%)	<0.001
Transfer trained	676 (99.7%)	161 (100%)	100 (100%)	47 (100%)	0.92
Doctors’ career levels
Foundation programme	22 (3.2%)	6 (3.7%)	0 (0.0%)	2 (4.3%)	<0.001
Core trainee	140 (20.7%)	50 (31.1%)	14 (14.0%)	14 (29.8%)
Speciality doctor	79 (11.7%)	19 (11.8%)	2 (2.0%)	5 (10.6%)
Registrar or advanced trainee	335 (49.4%)	68 (42.2%)	76 (76.0%)	23 (48.9%)
Consultant	42 (6.2%)	11 (6.8%)	3 (3.0%)	2 (4.3%)
Missing data	60 (8.9%)	7 (4.4%)	5 (5.0%)	1 (2.1%)
Nurse escort	410 (56.3%)	166 (80.2%)	106 (89.8%)	19 (38.8%)	<0.001
Transfer trained	302 (73.7%)	123 (74.1%)	63 (59.4%)	15 (79.0%)	0.07

Duration of critical care transfer and delays

Time points to calculate transfer intervals were documented relatively infrequently. Table 4 shows time intervals from referral to arrival at receiving hospital as well as duration of the ambulance journey, for those transfers where relevant data were available. In 204 (18.1%) transfers, the transferring team perceived a delay and recorded a reason for this. The majority (75, 36.8%) were due to delay or unavailability of transport followed by clinical problems with the patient (62, 30.1%) and lack of bed availability in the receiving hospital (25, 12.3%). Transfers were also delayed by availability of personnel, administrative reasons, lack of equipment and severe weather; see Table 4 for an overview of transfer delays.

Table 4.

Time from transfer request to arrival at destination, duration of ambulance journey and frequency and reason for delays of transfers.

Time intervals (where documented)	Specialist treatment		Repatriation	Capacity	p
Time intervals (where documented)	Time critical	Non-time critical	Repatriation	Capacity	p
Transfer request to leaving hospital (median, IQR)	105 min (72–155) n = 234	140 min (81–240) n = 75	270 min (122–390) n = 31	95 min (45–138) n = 23	<0.001
Ambulance journey (median, IQR)	60 min (45–75) n = 343	65 min (50–105) n = 103	81 min (60–100) n = 51	55 min (35–105) n = 31	<0.001
Transfer request to arrival in destination hospital (median, IQR)	159 min (120–220) n = 238	215 min (145–329) n = 76	355 min (245–460) n = 31	120 min (79–220) n = 23	<0.001
Documented reasons for delayed transfers	139/728 (19.1%)	33/207 (15.9%)	21/118 (17.8%)	11/49 (22.5%)	0.35
Transport availability	45 (32.4%)	5 (15.2%)	16 (76.2%)	9 (81.8%)	0.06
Patient preparation	44 (31.7%)	16 (48.1%)	1 (4.8%)	1 (9.1%)
Lack of bed at receiving unit	18 (13.0%)	5 (15.2)	2 (9.6)	0 (0.0%)
Transfer personnel or equipment	10 (7.2%)	0 (0.0%)	1 (4.8%)	0 (0.0%)
Unspecified/other	22 (15.8%)	7 (21.2%)	1 (4.8%)	1 (9.1%)

Critical incidents

Seventy-seven critical incidents (6.9% of all transfers) were recorded during the 1124 transfers. Incidents were classified into three domains – patient (including cardiac arrest, loss of intravenous or intra-arterial access, hypo- and hypertension, airway and ventilation problems and pupil dilatation), technical (including failure of equipment, gas supply and transfer trolley) and logistical (including vehicle problems and difficulty with access to the receiving hospital or unit). While the majority of incidents were patient-specific (46, 59.7%), around a third (26, 33.7%) concerned technical problems and five (6.5%) were logistical. The commonest patient-related incidents were hypo- or hypertension (20, 26.0%) and pupil dilatation in the context of neurological event or injury (12, 15.6%), while the commonest technical incidents concerned monitor (10, 13.0%) and syringe pump (8, 10.4%) failure. All logistical problems occurred during the transfer of time critical patients and these included vehicle breakdown and inability to access the receiving hospital on arrival. No patients died during transfer. Table 5 displays critical incidents stratified by transfer indication.

Table 5.

Critical incidents recorded during transfer.

	Specialist treatment		Repatriation (n = 118)	Capacity (n = 49)	p
	Time critical (n = 728)	Non-time critical (n = 207)	Repatriation (n = 118)	Capacity (n = 49)	p
Patient-specific incidents	39 (5.4%)	7 (3.4%)	0 (0.0%)	0 (0.0%)	0.13
Technical incidents	18 (2.5%)	4 (1.9%)	4 (3.4%)	0 (0.0%)
Logistical incidents	5 (0.7%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
All critical incidents	62 (8.5%)	11 (5.3%)	4 (3.4%)	0 (0.0%)	0.05

Discussion

This study represents the first published data on the critical care transfer activity of a UK adult critical care network. It demonstrates that inter-hospital critical care transfers of adult patients in a region of 5.5 million people occur at least every other day. Despite published national guidelines,¹ a considerable proportion of transfers are still undertaken by untrained and/or inexperienced doctors and nurses. We describe delays from transfer request to leaving the referring hospital of 105 min (median) in time critical patients, and critical incidents occur in at least 1 in 14 transfers. The majority of critical care transfers are time critical and occur soon after admission to hospital. As such, this demand represents a frequent, unpredictable and high-risk workload that in the Critical Care Network studied is not currently being met by a timely, experienced and safe response consistent with national guidelines.

In the United Kingdom, the Association of Anaesthetists of Great Britain and Ireland (AAGBI)^10,11 and the ICS¹ have published clear guidelines on critical care transfer. All of these guidelines recommend that hospital departments and critical care networks establish clinical governance procedures, that transfers are documented, critical incidents recorded and investigated and that standards of care should be the same as those delivered on the ICU. A postal survey of intensive care units examining inter-hospital transfers in 2003 and found that 25% of patients were escorted by senior house officers (maximum five years’ postgraduate training),²⁴ comparable to a similar study in 1997 (20%)¹⁹ and our finding of 22.3% of transferring doctors with less than four years postgraduate experience. This represents a group of relatively inexperienced staff transferring critically ill patients despite their acuity and potential for deterioration and unfortunately often reflects availability of staff within the hospital and also the difficult balance between maintaining the services of the transferring hospital while also ensuring safety of the transferred patient. Interestingly, despite multiple national guidelines that all agree on the importance of specific transfer training, no intensive care society or similar organisation has instigated a national training scheme for inter-hospital transport.³ Previous studies have demonstrated that 52–91% of transfer critical incidents (technical and patient related) are preventable with adequate preparation and training, such as undertaken by specialist retrieval teams.^3,4

The time elapsed between transfer request and leaving the referring hospital amounted to a median of 105 min even when patients required time critical transfer. This delay was multifactorial and included transport vehicle availability, patient preparation and bed availability at the receiving hospital. System improvements that have been instituted in other regions to minimise transport delay include dedicated transport platforms^16,17 and network collaboration to streamline the transfer pathway with resultant 98-min reductions in total transfer time.²⁵

Interestingly the 6.9% critical incident rate seen in this study is relatively low in comparison with previous published studies which report patient-related incidents, such as cardiovascular instability, respiratory complications or neurological deterioration, in 6–31% of transfers^2–4 with around one in four resulting in patient harm.⁶ Consistent with previous studies, we consider the documented rate of critical incidents in this study to represent an underestimate of actual rates of critical incidents.²⁶ The extensively described reasons for the under-reporting of medical critical incidents, including fear of adverse consequences, process and reporting systems, are particularly applicable to the unsupervised and impromptu nature of current critical transfer practice.²⁷ An additional and related consideration in analysing this data is that, although completion of transfer documentation is regarded as a standard of care in the Critical Care Network, it is likely that factors including emergency clinical priorities, difficulties in locating transfer documentation and failure to submit completed paperwork may make the database an underestimate of actual transfer activity in the region.

This study demonstrates that current ad-hoc provision of critical care transfer is not currently meeting the longstanding expectations of national guidelines in terms of training, clinical experience and timeliness. The data from an English Critical Care Network presented in this study may inform national conversation regarding the development of National Health Service (NHS) commissioned inter-hospital transfer services for adult patients in England. Critically ill patients are transferred by specialist retrieval teams in countries such as the Netherlands, Australia and New Zealand³ and the UK precedent has already been set by the establishment of similar services in Scotland and Wales.^16,17 Moreover, all neonatal and the overwhelming majority of paediatric transfers in England are performed by dedicated transport teams which are NHS commissioned, adequately resourced, led by senior consultant and nursing staff and working to high standards within strict clinical governance procedures. These data suggest that this standard should be provided for any patient in the United Kingdom, regardless of age, geographical location, or time of day.

Footnotes

Acknowledgements

We thank Sarah Miller and Ian Scammell from the South West Critical Care Network for assisting with database interrogation and thank the South West Critical Care Network for allowing us to access their database.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Whiteley S, Macartney I, Mark J, et al. Guidelines for the transport of the critically ill adult (3^rd Edition 2011). Intensive Care Society 2011, www.ics.ac.uk/AsiCommon/Controls/BSA/Downloader.aspx?iDocumentStorageKey=79bdff30-7676-46dd-856d-b47e4d667568&iFileTypeCode=PDF&iFileName=Transport%20of%20the%20Critically%20Ill%20Adult (accessed 13 November 2018).

Lovell

Mudaliar

Klineberg

. Intrahospital transport of critically ill patients: complications and difficulties. Anaesth Intensive Care 2001; 29: 400–405.

Droog

Smit

Absalom

, et al.

Transferring the critically ill patient: are we there yet?

Crit Care 2015; 19: 62–69.

Ligtenberg

JJM

Arnold

Stienstra

, et al. Quality of interhospital transport of critically ill patients: a prospective audit. Crit Care 2005; 9: R446–R451.

Droogh

Smit

Hut

, et al. Inter-hospital transport of critically ill patients: expect surprise. Crit Care 2012; 16: R26.

Beckmann

Gillies

Berenholtz

, et al. Incidents relating to the intra-hospital transfer of critically ill patients. Intensive Care Med 2004; 30: 1579–1585.

Van Lieshout

. Richtlijn voor het transport van Intensive Care patiënten. [Translation: guideline for the transport of intensive care unit patients]. NVICMon 2001; 5: 22–25.

Australian and New Zealand College of Anaesthetists (ANZCA). Guideline for transport of critically ill patients. ANZCA 2015, www.anzca.edu.au/documents/ps52-2015-guidelines-for-transport-of-critically-i (accessed 13 November 2018).

Warren

Fromm

Orr

, et al. Guidelines for the inter- and intrahospital transport of critically ill patients. Crit Care Med 2004; 31: 256–262.

10.

Farling PA, Andrews PJD, Cruickshank S, et al. Recommendations for the safe transfer of patients with brain injury. Association of Anaesthetists of Great Britain and Ireland 2006, www.aagbi.org/sites/default/files/braininjury.pdf (accessed 13 November 2018).

11.

Goldhill D, Gemmel L, Lutman D, et al. AAGBI Safety guideline: interhospital transfer. Association of anaesthetists of Great Britain and Ireland 2009, www.aagbi.org/sites/default/files/interhospital09.pdf (accessed 13 November 2018).

12.

Wright

McDonald

Rogers

, et al. Provision of facilities for secondary transport of seriously ill patients in the United Kingdom. BMJ 1988; 296: 543–545.

13.

Mackenzie

Smith

Wallace

PGM

. Transfer of adults between intensive care units in the United Kingdom: postal survey. BMJ 1997; 314: 1455–1456.

14.

NHS England, Public Health England, Health Education England, Monitor, Care Quality Commission, NHS Trust Development Authority. Five year forward view, www.england.nhs.uk/wp-content/uploads/2014/10/5yfv-web.pdf (accessed 13 November 2018).

15.

McCullough

Haycock

Forward

, et al. Major trauma networks in England. BJA 2014; 113: 202–206.

16.

Emergency Medical Retrieval Service (EMRS) Scotland, www.emrsscotland.org (accessed 13 November 2018).

17.

Emergency Medical Retrieval and Transfer Service (EMRTS) Cymru, www.emrts.cymru (accessed 13 November 2018).

18.

Paediatric Intensive Care Society Acute Transport Group, http://picsociety.uk/pics-subgroups/acute-transport-group (accessed 13 November 2018).

19.

Dunn

. Secondary insults during interhospital transfer of head-injured patients: an audit of transfers in the Mersey Region. Injury 1997; 28: 427–431.

20.

Wiegersma

Droogh

Zijlstra

, et al. Quality of interhospital transport of the critically ill: impact of a Mobile Intensive Care Unit with a specialized retrieval team. Crit Care 2011; 15: R75.

21.

Wikipedia authors. South West England. Wikipedia, 2017, https://en.wikipedia.org/wiki/South_West_England (accessed 13 November 2018).

22.

South West Critical Care Network. SWCCN Network Members. SWCCN 2017, http://www.uhbristol.nhs.uk/for-clinicians/swccn/network-members (accessed 13 November 2018).

23.

North Bristol NHS Trust. North Bristol NHS Trust becomes Major Trauma Centre for Severn region. North Bristol NHS Trust 2012, www.nbt.nhs.uk/news-media/latest-news/major-trauma-centre (accessed 13 November 2018).

24.

Hay

Smith

Hilditch

, et al. Transfer of adults between intensive care units in the United Kingdom: clinical practice. J Intensive Care Soc 2005; 6: 45–56.

25.

Dickinson

Eynon

. Improving the timeliness of time-critical transfers: removing ‘referral and acceptance’ from the transfer pathway. J Intensive Care Soc 2014; 15: 104–108.

26.

Mahajan

. Critical incident reporting and learning. Br J Anaesth 2010; 1: 69–75.

27.

Archer

Hull

Soukup

, et al. Development of a theoretical framework of factors affecting patient safety incident reporting: a theoretical review of the literature. BMJ Open 2017; 7: e017155.