Evaluation of the safety of inter-hospital transfers of critically ill patients led by advanced critical care practitioners

Secondary inter-hospital transfer of critically ill adults in the UK occurs frequently. The Intensive Care National Audit and Research Centre (ICNARC) case mix programme estimates that there are 10 000 critically ill adult transfers in the UK annually (Intensive Care Society (ICS) and Faculty of Intensive Care Medicine (FICM), 2019). Pre-transfer patients require high percentage oxygen, invasive mechanical ventilation and invasive monitoring. Due to the absence of an available intensive care bed or the need for definitive care at another hospital, patients often only receive a brief period of resuscitation before transfer. Patients are at high risk for deterioration in their neurological, haemodynamic and respiratory state, and even death.

The capacity of critical care services is a long-standing problem in the UK (FICM, 2018a). The centralisation of the NHS is driven by economy of scale, new technologies, corporate mergers, safety and standard concerns (Meadows et al, 2011). This increases the need for secondary transfer of critically ill patients. Critical care regional networks have responsibility for co-ordinating standards and processes for transfer (Association of Anaesthetists of Great Britain and Ireland (AAGBI), 2009; ICS and FICM, 2019). Despite developments, concerns over inter-hospital transfer safety continues. Studies highlight deficiencies in experience and training of staff, equipment, stabilisation before departure, and logistical difficulties (Droogh et al, 2015). In the realm of paediatric critical care, specialist transfer teams have been in place regionally for some years; in adults, this has only occurred in the pre-hospital arena. Adult critical care transfer continues to be delivered on a largely ad hoc basis in the UK, with transient staff, with varying levels of training and experience (Grier et al, 2020).

University Hospitals Birmingham NHS Foundation Trust (UHB) comprises four hospitals. Services across these hospitals vary, necessitating frequent secondary transfer of critically ill adults. Our critical care services evolved a dedicated transfer system, utilising staff trained in initial resuscitation and transfer of critically ill patients between three of these four hospitals. The team is consistent, adherent to national critical care transfer guidelines, non-transient, with a culture of governance and case series review to benchmark and revise standards. This team is a combination of advanced critical care practitioners (ACCPs) and a critical care outreach team (CCOT). The team consists of nurses and functions without direct supervision by doctors, the traditional model of transfer. This article will evaluate the safety and development of this model of inter-hospital transfer using case series data.

ACCPs in this service model do not operate a stand-alone transfer system, they are part of the multidisciplinary team (MDT) in our intensive care service, operating 24 hours a day, 7 days a week. ACCPs contribute to patient assessment and plans of care, ward rounds, invasive procedures, investigations and diagnosis, prescribing, communication with families, teaching and supervision. Lee et al (2018) discussed the ACCP role in more detail.

The Standards for Quality Improvement Reporting Excellence (SQUIRE 2.0) guideline is a structure for reporting new knowledge about how to improve health care. This will be used to frame the development of the ACCP transfer team and discuss its evaluation (Goodman et al, 2016).

Method

Context

Data collection pertaining to inter-hospital transfers of ventilated, critically ill patients was absent, both within the hospital group and within the critical care network. Critical care inter-hospital transfers were carried out by rotating, often locum, medical staff on an ad-hoc basis, with no ‘buy-in’ to the critical care service. This made quality improvement or obtaining baseline data challenging. Problems of note were a lack of training of the lead clinician and assistants, inappropriate equipment, and lack of preparation for transfer, leading to avoidable complications. This was the service's anecdotal experience, there were no data collection mechanisms in place to support these assertions.

Intervention

UHB comprises four hospitals; the Queen Elizabeth Hospital, a tertiary centre, will not be discussed here. Heartlands, Good Hope and Solihull hospitals (HGS) vary in size and on-site services/specialties. This necessitates frequent transfer of critical care patients between the Trust's hospitals to meet their specific needs.

ACCPs also conduct intra-hospital transfers at the Heartlands site in Birmingham, including patients in the emergency department, critical care, the wards and outlying areas. The majority of internal transfers of invasively ventilated patients are carried out by ACCPs at Heartlands. If a patient requires invasive mechanical ventilation at the Solihull site, the critical care MDT on that site delivers the initial intubation and resuscitation. The Heartlands ACCPs will travel to the Solihull site to assist with the stabilisation, and manage the transfer to the nearest available ICU bed (typically within the HGS system). The majority of inter-hospital transfers of ventilated patients from Solihull Hospital are carried out by ACCPs. The Good Hope Hospital ICU team has no consistent ACCP presence and usually manages its own inter-hospital transfer of mechanically ventilated patients. However, the ICU team is small, and workload can make inter-hospital transfer difficult. Workload permitting, the Heartlands ACCPs will assist the Good Hope team. ACCPs never conduct time-critical transfers for Good Hope Hospital. This system has been in place for 8 years; the Heartlands ACCPs have carried out a continuous data collection since the beginning of 2017.

ACCP training and supervision processes have been discussed in previous publications (Denton et al, 2019; Williams et al, 2019). The HGS ACCP team has incorporated transfer within its training structure since inception. This encompasses a master's level module on the subject, which is run by the University of Warwick (University of Warwick, 2020). The module integrates theoretical knowledge, portfolio development, direct observed practice assessment and objective structured clinical examination and simulation. Following final sign-off, and completion of the clinical components of the ACCP programme, ACCPs conduct transfers of mechanically ventilated patients (internal and inter-hospital) without direct medical supervision.

Study of the intervention

Collection of case series data was chosen to assess the impact of the ACCP-led critical care transfer service. It was not possible to take a ‘before and after’ quality improvement approach because no prior data were available. The team had evolved over several years; there was little specific funding for the ACCP development programme when it began. The comparator for evaluating the service was that of the published UK literature on the transfer of critically ill adults, combined with compliance to national guidance (AAGBI, 2009; ICS and FICM, 2019). There was no reason to assume HGS outcomes, complications and guideline compliance were different from the national climate at the time.

Measures

Anonymised data were collected using Google Forms. The form collected information on the date, time and indication for transfer. Patient information, acuity, transfer destination and any complications were captured (Denton, 2021).

There are no widely recognised data sets for audit of critical care transfer. National guidelines have evolved since the development of the data tool and changes were made as this guidance altered. Adaptations were made to the data set where incidents occurred.

Data for each event were submitted shortly after a completed transfer via the individual ACCP's phone. Monthly, the first author reviewed submitted data forms, sought clarification from individuals and resubmitted forms where inconsistencies were found.

Analysis

Data are presented using simple descriptive statistics. Categorical data are summarised by the number in each category and percentage of the group as a whole.

Ethical considerations

The development of the HGS ACCP workforce followed a Department of Health (DH) report (DH, 2008) and the FICM's curriculum (2018b) and did not require ethical approval. Data collection was approved by the Trust audit committee (reference 4461). The first author (GD) developed the online data tool and led the data collection under the supervision of the critical care audit lead (NA).

Results

Indications for transfer

Between January 2017 and September 2020, ACCPs conducted 934 transfers of mechanically ventilated patients without direct doctor supervision. A total of 286 of these were inter-hospital transfers, which are the focus of this article.

In the majority of inter-hospital transfers, lack of critical care beds in the original hospital (52.8%, 151) was the indication. The second most frequent indication for transfer was the need for specialist care in a tertiary centre, largely for neurosurgical intervention (29%, 83). Only 2.4% (7) required transfer outside of the HGS hospitals for non-clinical reasons.

Acuity

All patients in this cohort were in receipt of invasive mechanical ventilation (IMV). Half of these patients required more than 50% oxygen and 14% (40) necessitated a positive end expiratory pressure of at least 10 cmH₂O. Four (1.3%) patients had refractory hypoxia, meriting airway pressure release ventilation to maintain acceptable oxygen saturation.

In 18.9% of patients (54), IMV was the only form of organ support. Most patients required some form of cardiovascular support from a vasopressor or inotrope (67.1%, 192). In 3.5% (10) of cases, more than one vasopressor was required, and in two cases (0.7%) intra-aortic balloon pump support was necessary.

Transfer assistance and checklist utilisation

In this cohort, 3.1% (9) of transfers did not have an assistant other than a paramedic crew. CCOT nurses attended 56.2% (161) of transfers. ICU nurses provided the majority of transfer assistance where a CCOT nurse was not available. The emergency department (ED) was the origin of 42.3% (121) of patients. ED nurses (due to staffing issues) infrequently accompanied transfers (3.1%, 9). EDs in the HGS system saw increasing demand; release of staff in the resuscitation bay for the transfer was a growing problem. This led to transfer assistants not accompanying ACCP transfers. Recognising the clinical risk, from lack of an assistant, our CCOT team supported transfer of ED patients. The benefit of this was that CCOT nurses have transfer training and experience, which ED staff typically do not. Exceptions are made when awaiting a transfer assistant would lead to a delay in a time-critical transfer.

The local transfer checklist was employed in 83.2% (238) of transfers. Access to checklists varies, critical care network observation charts for transfer went through several iterations, which have omitted and then included checklists. This makes meaningful interpretation of checklist compliance in the course of the case series difficult.

Complications

Uneventful transfer occurred in 81.4% (233) of cases. Complications included logistical and equipment issues. Hypotension was the most common clinical deterioration (4.9%, 14). Four episodes of hypoxia were captured (1.4%, 4). There were no extubations or airway-related events during any transfers in either the inter-hospital or internal cohorts. Equipment failure seems to be a significant contributor to the complications that did occur. Half of complications included logistical issues such as repatriation of equipment.

Unexpected consequences/benefits and failures

Data collection is ongoing, adaptations have been made as a result of the data, and to changes in demand on ICU services. One such issue is the timeliness of neurosurgical transfers. This is a frequently highlighted topic in national literature (Leach et al, 2007; Dickinson and Eynon, 2014). Concerns centre on administration, communication, and timely decision making, rather than the process of transfer itself. To identify delay issues, the data forms were adjusted to capture timing in time-critical transfers. The change in the data set has yet to generate enough cases to make any assumptions about whether this aspect has improved.

Members of the ACCP team teach on a critical care transfer master's module. As the team's reputation has grown, members have received increasing requests from clinicians from external organisations to ‘shadow’ our ACCPs to gain further experience in transfer.

Responses to an ACCP-led transfer, as opposed to the traditional doctor-led one, varied. A receiving team could be accepting, had heard about our team, had a positive response to us and wanted to know more. On occasion, the team has met with some apprehension, which was dispelled by the ACCPs' conduct, the quality of handover and the condition of the patient on arrival. Sometimes the team has met with a degree of hostility. Over the ensuing years, the more negative responses have dissipated as the transfer system became normalised in the region.

Missing data

It was not possible to calibrate the data form against other administrative systems to identify missing data. The variability of location of patients means administrative systems differ between departments. Data from the ACCP transfer case series is also used as part of the yearly performance review, revalidation and competency portfolio, creating an incentive for individuals to submit data.

Summary

An ACCP-led transfer service has replaced an ad hoc process of transfer, with one of a consistent team combined with a process of training, governance and continual review. Data collection has allowed continual review of the team's compliance with evolving national guidelines for the transfer of critically ill adults. A developing ACCP team has the potential to apply these elements to their own service.

Research review

Non-medical critical care transfer in the UK

Leslie and Bose (1999) first described advanced neonatal nurse practitioner (ANNP)-led transportation of critically ill neonates. Leslie and Stephenson (2003) published data considering the safety and efficacy of UK ANNP-led transfers, showing parity with medical counterparts. Fenton and Leslie (2009) argued that there is no evidence that neonatal transfer outcomes differ between professional groups. ACCP-led transfer is a new concept in adult critical care. ACCP teams are in varying degrees of development, many providing intra-hospital transfer. Most ACCP teams do not perform inter-hospital transfers. Concerns include indemnity for inter-hospital transfers, risks of airway management, maturity of the team, not being seen as an early adopter, and service need. The logistics of our three critical care sites relies on the ACCP team delivering this element of care for the service model to function.

Compliance with national guidelines

The ICS and FICM have produced guidance for the transportation of critically ill patients (ICS and FICM, 2019). This guidance mandates minimum monitoring and equipment for critical care transfers. The team adheres to this, including dedicated transfer trolleys, ventilator equipment, kit bags and checklists. Guidance places emphasis on training of staff (AAGBI, 2009; ICS and FICM, 2019). The guidance states that staff should have supernumerary experience (ICS and FICM, 2019). Grier et al (2020) found that 25% of transfers were carried out by specialist trainees in their first 2 years of training. Grier et al (2020) argues that doctors at this point of training do not have the experience or the training to manage inter-hospital transfer. Cook and Allan (2008) surveyed acute care common stem (ACCS) doctors, finding that 30% of trainees felt they could not assess the suitability of a patient for transfer, 22% did not feel prepared for such transfers, and that 33% had attended a transfer course. Despite this, Cook and Allan (2008) highlighted that 88% of ACCS trainees had undertaken an inter-hospital transfer of a critically ill patient.

A survey of emergency medicine and anaesthesia trainees (registrar level and above) found 20% did not have training in the transfer, and only 9% had supervision from a more senior doctor during transfer (Paton et al, 2014).

The authors' ACCP group emphasises training for transfer. ACCPs develop an extensive portfolio of supervised transfers as part of their MSc Critical Care Transfer module and their FICM ACCP credentialing log (FICM, 2018b). The FICM ACCP credentialing document forms the basis for the ACCP training curriculum. Simulation forms part of the training process within the MSc model.

Patients should receive a period of stabilisation/resuscitation before transfer to minimise the likelihood of deterioration (ICS and FICM, 2019). ACCPs are part of this process from the outset of patient care, and where not, will assess if further interventions are required to optimise patients. Pre-departure checklists form part of our transfer preparations; our team were involved in the latest West Midlands Critical Care Network iteration of checklists and documentation for transfers.

Complications during transfer

Lack of uniform nomenclature for complications makes comparison between publications difficult. Complications associated with transfers are broadly categorised into: patient, system, equipment failure, and staff factors (Easby et al, 2002; Ligtenberg et al, 2005; Droogh et al, 2015). A Scottish review found that in 18% of transfers, equipment was secured inappropriately (Fried et al, 2010). All transfers in the authors' service are carried out using a transfer trolley that secures infusion pumps, monitors, gas supplies, suction and ventilator. Equipment failure featured in 9.8% of ventilated patients in a Scottish group (Fried et al, 2010), versus 2.7% (8) in the authors' ACCP group. Fried et al (2010) found 4% experienced cardiovascular and respiratory problems, contrasting with 6.3% (18) events in the ACCP cohort. Ligtenberg et al's (2005) transfer review found that 34% of patients had an adverse event. Droogh et al (2015) found cardiovascular complications of 6–24% and respiratory issues in 0–15%. Despite disparate definitions of adverse events, the overall rate of complications in this ACCP cohort is comparable to other published studies and in some cases produced a lower number of complications.

Grier et al (2020) published data on UK adult critical care transfer, finding 6.9% of transfers experience complications. Patient-related incidents were most common, with 26% of complications being related to blood pressure. Technical problems occurred in 34% of complications, 15% suffered neurological deterioration (Grier et al, 2020). Complications in the authors' ACCP cohort were 18.5% (53), hypotension being the most common issue (4.9%, 14). The ACCP cohort is smaller than the Grier et al (2020) study, a larger data set over a longer period is needed to make more valid comparisons.

Critical care transfer and airway management

Under supervision, ACCPs carry out drug-assisted oro-tracheal intubation, most in the team having performed over 200 intubations (Denton et al, 2019). Many ACCP teams do not carry out inter-hospital ventilated transfers independently based on concerns surrounding independent airway management. The authors challenge this rationale.

Inadvertent extubation is a risk of transfer. Great effort, planning and preparation goes into maintaining the integrity of the airway during transfer. Zero extubations occurred during this case series. That does not mean to say an extubation during transfer is not possible. Considering all transfers in the authors' case series (internal and inter-hospital) there have been no extubations or a need to change an airway in 934 transfers. Fried et al (2010) identified one extubation in 248 intubated transfers. In ventilated inter-hospital transfers, patients are invariably sedated and paralysed for the process. In the event of an inadvertent extubation of the airway, the patient does not require rapid sequence induction, they are already receiving general anaesthesia and neuromuscular blockade, leaving the provider with the option to re-intubate, or revert to basic airway measures such as a supraglottic device until the intended destination is reached.

The NAP4 study highlighted the risk of airway management in the critically ill outside of the theatre environment (Cook et al, 2011). Airway management is unpredictable in the critically ill. In the context of transfer, the nature of the patient's airway is known, including the ease of mask ventilation, the feasibility of a supraglottic airway, and difficulty of laryngoscopy. Many of the patients captured in this case series are likely to have been intubated by the ACCP performing the transfer. Intubation has been part of the ACCP role in the critical care team for some years. Critical care intubations case series data are collected by the ACCP team, but are unpublished and separate from this data set. This simplifies the planning of airway management in the event of difficulties during transfer.

Impact on people and systems, costs and strategic trade-offs

Transfer training is a key requirement in ACCS training for anaesthesia, intensive care medicine and emergency medicine. Despite meeting minimum requirements, some doctors still feel unprepared for critical care transfer (Cook et al, 2011; Paton et al, 2014). Loss of training opportunities to gain experience to achieve sign-off is a frequently cited concern in the training and utilisation of advanced practitioners. The HGS ACCP team is integral to the support of doctors in the service. The ACCPs provide supervised intra/inter-hospital transfer for junior doctors, fostering closer MDT working and forming working relationships with the registrars and consultants of the future. Support of ACCS trainee doctors by ACCPs nurtures a sense of team working, rather than competition and infringement on professional boundaries. The ACCP group has demonstrated contributions towards supervision of junior doctors in other domains (Denton et al, 2019).

Before the ACCP transfer system, out-of-hours transfer was a significant issue for the Trust's critical care consultants, necessitating demanding on-site presence during any out-of-hours mechanically ventilated patient transfers. Having a 24-hour, 7-days-a-week ACCP team reduces the consultants' out-of-hours workload. Typically, the ICU consultants are on site between 08:00 and 18:00, external transfers formerly meant that a consultant would have to come into the hospital from home while the ICU doctor carried out the transfer. ACCP external transfers occurred out of hours in 183 cases (64%), as ACCPs are an addition to the ICU medical team, in all likelihood, this prevented 183 instances of ICU consultants having to come in from home while on call.

The transfer service does not just move patients from one location to another. Focus is placed on communication between teams on different sites, between family and the MDT. ACCPs are part of, rather than external to, the hospital system, and have an intimate knowledge of the organisation. This includes ongoing resuscitation and stabilisation within the HGS policies and procedures, adding to continuity of care.

Service costs are limited to equipment, consisting of a critical care transfer trolley and the necessary monitoring, pumps and ventilator on each site. Since ACCPs are an embedded element of the MDT, the provision of this service does not incur additional costs. The trade-off for this model is that we are not a standalone service; we can be absent from the MDT on our primary site for some hours while delivering care to a patient requiring transfer. The team performs approximately 100 external transfers a year, a standalone service would be grossly inefficient and costly on this basis as it would be largely inactive. The benefit over a regional service is that the ACCP team members have an intimate knowledge of the HGS system, processes and relationships, adding continuity and consistency.

Limitations

As a single-centre study involving a well-established team operating over 6 years, external validity is debatable. ACCPs nationally are in various stages of training and maturity. The HGS ACCP training goes beyond the requirements of the FICM credentialing document, incorporating a comprehensive module on transfer. Extrapolation of this data nationally may be questionable.

The size of any cohort is a factor when considering safety and uncommon adverse events. There were no adverse airway events in this case series, and one could argue that the population was too small to detect such events. Rare events may be underestimated. Reporting is voluntary and retrospective, this gives the potential for recall bias and under-reporting of complications and adverse events.

Conclusion

This case series data provides an example of safe and effective advanced practice in an area that is traditionally a medically led domain. Data available surrounding transfer of critically ill adults is sparse and focuses on the problems of ad hoc systems of care. ACCPs can provide an alternative process of care for critically ill adults who require external transfer and a benchmark for audit and quality improvement.

This ACCP transfer service adds diversity to the ACCP role, but still uses the inherent skills and level of practice of ACCPs. Role diversity may help aid retention of ACCPs within a service.

The HGS ACCP transfer service is a prime mover in utilising the skills and level of practice of ACCPs in this arena. Publication may help other services to see advanced practitioners functioning at this level as less contentious, and a valid option for service development.

A study of external transfer by ACCPs on a national level may aid in confirming the external validity of this model of transfer. This would provide momentum to integrate external transfer more explicitly in the FICM ACCP competency document and wider acknowledgment in national guideline formation.

KEY POINTS

• Advanced clinical practitioners have been providing transfer of critically ill patients without medical supervision for 20 years
• An advanced critical care practitioner (ACCP) team in one trust has safely conducted transfer of critically ill adults for 8 years and has outcomes comparable to those of medical colleagues
• Transfer of critically ill adults can be integrated into the training of ACCPs and form part of the role that supports the overall critical care service
• ACCPs can provide adult critical care transfers to the standards laid out by the Intensive Care Society

CPD reflective questions

• What standards and guidelines apply to the transfer of critically ill adults?
• Think about the risks involved in the transfer of the critically ill adult
• What training and skills should nurses assisting in the transfer of the critically ill adult have?