Abstract
Anaesthetic airway incidents persist as a significant concern in patient safety and, despite extensive investigations, continue to cause patient harm. Traditional safety investigations predominantly adhere to Safety-I principles, focusing on identifying and rectifying errors, often yielding limited new findings. In this analysis conducted within the webAIRS database, the focus shifted towards Safety-II principles. The aim of this study was to identify factors contributing to airway management safety by examining incidents that did not result in adverse patient outcomes. Incidents categorised as ‘difficult intubation’ or ‘failed intubation’ without causing harm to the patient and reported to webAIRS between 2016 and 2022, were included in the analysis.
An inductive qualitative content analysis of narrative data from 129 such incidents revealed that the majority of reported events depicted scenarios deviating from controlled and planned circumstances. During the analysis four themes were identified: patient factors, system factors, individual anaesthetist factors and airway management strategy. Within the first three themes, multiple factors were linked to airway management strategies. The findings of this qualitative analysis show that ‘Work as done’ often differs from ‘Work as imagined’.
This qualitative analysis highlighted the dynamic nature of human management, as individuals respond to unplanned or unexpected events, showcasing adaptability and positive contributions to incident performance. Expanding the understanding of patient safety to also include Safety-II principles, provides a deeper and wider understanding of airway management safety.
Introduction
Critical incident reports that describe difficult and failed intubation in anaesthetic airway management across Australia have been analysed for over 30 years. 1 Despite advances in equipment, monitoring and teaching, and the incorporation of human-factor principles and decision-making theories in airway management guidelines, these incidents continue. Opportunities to consider incident reports and patient safety through new theoretical and practical lenses exist. 2 This study has adopted a novel approach for exploring no-harm events in anaesthetic airway management to enhance understanding of system and patient factors that contribute to patient safety.
Safety principles
Airway incidents describing significant harm or death have been investigated by airway management societies globally.3,4 There are numerous, detailed publications exploring the findings of difficult and failed intubation incidents reported across Australia and New Zealand.1,5,6 However, to date, these reports on airway incidents in anaesthesia have mainly been viewed from the perspective of Safety-I principles—that is, identifying ‘what went wrong’.1,3,5,7 –9 Consequent recommendations and quality improvement endeavours have focused on the elimination of these failures. An alternative strategy is to consider incidents from the perspective of Safety-II principles. Advocates for Safety-II perspectives encourage data analysis that expands the understanding of safety to include the valuable information and insights gained from examining incidents that result in no harm to the patient,10 –13 thereby defining safety as successful work processes. 10
A concept from the Safety-II paradigm is to consider ‘work as imagined’ (WAI) versus the ‘work as done’ (WAD). 10 WAI describes expected work pathways and outcomes. WAD examines the execution of work in the real world, where situations may evolve dynamically and at times unexpectedly, requiring quick adaptations and improvisation by individuals identifying the disparity between planned procedures and practical task execution. In examining WAI versus WAD, the dynamic nature of complex systems can be better understood and provide insights into both the points of failure and the successful adaptations that lead to improved safety.2,10 –12,14,15
Describing WAI in anaesthetic airway management, Figure 1 outlines a typical scenario, where a proactive and thorough assessment of the patient’s airways is conducted electively, and an airway management strategy is developed. It is assumed that patients are in their optimal health condition, with information about their previous airway management available, and the formulated strategy is supported by favourable system factors. These factors include the immediate availability of suitable equipment, experienced assistance and a cohesive team in an optimal and supportive environment. This idealised viewpoint might differ significantly from how work has actually been done. Anaesthetists work in varied environments, with varied patient acuity and system complexities. It is unlikely that the WAI perspective can foresee all local contingencies and variables 12 and therefore, a focus on WAD in events of successful airway management when potential critical events occurred will likely provide useful insights and understanding of anaesthetic processes and airway management safety.
Work as imagined (WAI) airway management—simplified diagram.
Anaesthesia has a long history of pioneering patient safety initiatives, incident reporting and analysis of critical events. Professor Sir Robert Macintosh first published the idea of anaesthetic incident reporting in 1949, recommending and discussing the benefits of reporting adverse events. 16 Since then, incident reporting systems have evolved. In 1993, the first analyses of the Australian Incident Monitoring Study (AIMS) provided groundbreaking information concerning incidents during anaesthesia and continued to report their findings and for several years.1,8,9,17 AIMS was phased out in 2008 and ‘WebAIRS’, the first bi-national web-based anaesthetic incident reporting system was introduced. WebAIRS has collected data about incidents and adverse events during anaesthetic care across Australia and New Zealand since 2009.18,19 Registered anaesthetists across both countries can voluntarily and anonymously report no-harm events, near misses, significant adverse events or incidents leading to significant clinical harm or death. One-third of the incidents reported to webAIRS report no immediate harm to the patient at the time of the event, and two-thirds of the reporters classify ‘patient not affected’ as the final outcome of the incident. 19 These reports are classified as no-harm events.
This study is the first to explore the possibility of inductive qualitative no-harm event analysis in airway management using incidents reported to webAIRS, and we aim to identify factors that have contributed to airway management safety. Incidents reported to webAIRS from 2016 onwards and coded as ‘difficult intubation’ or ‘failed intubation’ have been analysed. In contrast to previous reports, this analysis focuses only on events that did not lead to adverse patient outcomes at the time of the incident.
Methods
In this qualitative study, incident narratives from webAIRS were analysed to address the following research question: ‘What safety features and elements are identified by incident reporters of no-harm events?’ Incidents categorised by the anaesthetist reporters as ‘difficult intubation’ or ‘failed intubation’ and classified via tick-box as no-harm events, were extracted into an Excel spreadsheet. Narratives of incidents where the reporter did not classify an episode outcome were reviewed, and if sufficient information was provided, then the missing classification was added by the analysts. Incidents where patient harm was reported either as a final outcome by the reporters or when identified via narrative analysis, were excluded from the analysis. Incidents with limited information preventing clear analysis were also excluded. Relevant incident narratives reported to webAIRS from July 2016 to the start of the analysis for this project in September 2022 were examined. This time frame was selected as incidents from the database from before July 2016 have previously been analysed 5 and the trial of a new method for analysing webAIRS data required a timebound data set.
Data entry
Incident data entry to webAIRS includes non-mandatory data entry fields (narratives and tick-boxes), to collect patient, anaesthetic and surgical demographic information. The reporters can categorise the immediate outcome of the incident, ranging from ‘no effect’ to ‘minor effect’, ‘case cancellation’, ‘intensive care unit (ICU)/high-dependency unit (HDU) admission’, ‘prolonged length of stay’ and ‘death’. The final patient outcome can be categorised as ‘not affected by the incident’, ‘temporary disability’, ‘permanent disability’ or ‘death’. Separate narrative entries include text boxes inviting the reporting anaesthetist to describe the incident in their own words, to reflect on contributing factors leading to the incident, to review alleviating factors and to add any additional information that they believe may be helpful to more fully appreciate the circumstances concerning the event (Figure 2).
Narrative boxes that (a) describe the incidents and (b) invite the reporter to reflect on the incident, with focus on contributing and alleviating factors, and further suggestions.
Inductive content analysis
Narrative data from incidents where ‘no effects’ were reported for an episode outcome were extracted and imported to a qualitative data analysis programme (QSR NVivo 12) and analysed. Inductive content analysis (ICA), as described by Vears and Gillam, guided this analysis. 20 The analysed narratives are straightforward in nature—short and concise—and therefore ideally suited for ICA.
Coding
Codes to categorise the narrative data were developed iteratively throughout the analysis process, a commonly described technique in inductive qualitative analysis.21,22 Each event narrative was coded line by line. For example, the narrative sentence ‘There was no other significant medical history other than resolving respiratory tract infection’ was coded as ‘Generally well patient’ and ‘Patient condition improving’. When the use of a videolaryngoscope of any type was mentioned within the sentence, this was coded as ‘Videolaryngoscope’ and the sentence ‘Assistance called for—anaesthetist from next theatre attended’ was coded as ‘Call for help’ and ‘Additional help available’. During the process, a range of codes addressing safety factors, such as patient and system factors, airway management strategies and team factors, were developed. Codes were initially divided into two main subgroups—factors that facilitated safety following Safety-II principles, and factors that threatened safety, with a focus on identifying mechanisms that restored safety and prevented a bad outcome.
Paper-based analysis
When codes were finalised, content categories and subcategories were further analysed using a paper-based method. The main analyst (YE) transferred the developed codes onto separate pieces of paper. Each code was reviewed and inductively grouped into themes. When aggregating codes into themes, elements of alignment with Hollnagel’s framework addressing WAI against WAD were identified. 10 This process was repeated until themes pertinent to WAD in anaesthetic airway management and their relationships with anaesthetic airway management principles and guidelines were identified. The results, including group allocations and connections, were presented, discussed with co-authors and reviewed until all analysts agreed on the themes. The concept of WAI vs WAD was employed to support the results section and guide the ensuing discussion.
Presentation of results
Identifying factors present in the narratives—such as, for example, precise patient age, biological sex or detailed descriptions of the surgical procedure—are deliberately omitted from the subsequent results section. There is an existing protocol that mandates the maintenance of de-identifiability for all incidents reported to webAIRS. This redaction process aims to uphold the confidentiality and privacy of reporters and patients, while allowing aggregated findings and overarching patterns to be presented without compromising sensitive information. Each incident received a randomly generated identification number (ID) ranging from ID 1 to ID 129.
Ethical considerations and permissions
WebAIRS was created, and is funded by, the Australian Society of Anaesthetists, the Australian and New Zealand College of Anaesthetists and the New Zealand Society of Anaesthetists.18,19 WebAIRS data collection complies with current ethics requirements for collecting de-identified quality assurance data in Australia. 19 Additional ethics approval to ensure that the data collection meets National Health and Medical Research Council requirements has been obtained from the Royal Brisbane and Women’s Hospital Human Research Ethics Committee (HREC/11/QRBW/311) and the Nepean Blue Mountains local health district (HREC/12/NEPEAN/18). Ethics approval in New Zealand was obtained from the Health and Disability Ethics Committee (MEC/09/17/EXP).
Results
Of the 271 reports coded as difficult or failed intubation, 129 no-harm events were identified and included in this analysis (Figure 3).
Identification of eligible incidents for analysis. ICU, intensive care unit; HDU, high-dependency unit; NVivo, data analysis software.
Themes
The detailed thematic analysis in this qualitative study identified four themes: (1) variable patient factors; (2) variable system factors; (3) the role of the treating anaesthetist; and (4) the airway management strategy. Within the first three themes, multiple factors were linked to the airway management strategies.
Theme 1, ‘Variable patient factors’, included airway, surgical and medical factors. Theme 2, ‘Variable system factors’, included the theatre team, equipment and the work environment. Variables described within these themes shaped the initial airway management strategy developed by the treating anaesthetist and were either favourable or contributory to the development, escalation and resolution of the reported no-harm incidents. During the course of the no-harm events, initial airway management was either immediately successful or, if unsuccessful, progressed to successful rescue airway management. Theme 3, ‘The treating anaesthetist’, identifies factors related specifically to the reporting anaesthetists. From the perspective of the analysts, the contribution of the anaesthetist to no-harm incidents as an individual was often explicitly noted. Theme 4, ‘The airway management strategy’, includes data specific to the initial management or successful rescue management of patient airways.
Most of the reported events differed from WAI and were not describing ideal controlled and planned situations. Instead, the treating anaesthetist had to take variable patient and system factors into account to shape their airway management strategy. When securing the airway was initially unsuccessful, the treating anaesthetist was faced with dynamic and changing patient and system conditions until airway management was successfully performed. Figure 4 illustrates the variables described in this WAD, as identified from the qualitative analysis of the critical incident reports included in this study.
Work as done (WAD): Factors affecting successful airway management.
Theme 1: Variable patient factors
Patient factors identified by reporters included the complexity of the patient’s airway, underlying medical condition(s) and surgical requirements. Very few of the recounted incidents described WAI; that is, patients, environments and systems with an ideal confluence of safety features. Instead, a discernible variability prevailed across patient characteristics, conditions and medical exigencies. Within these, the WAD included anaesthetists managing patients with significant challenges in airway management, navigating intricate medical complexities and being confronted by diverse surgical demands.
Airway factors
Several challenging airway scenarios were described, wherein patients presented with difficult airways that were identified during the preoperative airway assessments: “[D]ifficult airway previously diagnosed as lymphangioma of floor of mouth and grade 4 larynx at the time now for repeat drainage of abscess by ENT” (ID 42). The early identification of patients’ difficult airways, and therefore an ‘anticipated difficult airway’ was described as a Safety-II factor that alleviated the risk to patient safety by some reporters and was an important factor in many no-harm events: “This would have been a very difficult case if it were unanticipated” (ID 52). Potential difficult airways were identified by commonly used airway assessment methods, including the Mallampati score, mouth opening and thyromental distance. Documentation of previous airway management was available in some instances, “Previous documentation of Grade 3 airway” (ID 24) but noted as unavailable in other events, “The hospital at which the patient’s airway incident had occurred was contacted but the records from the time of the incident were not available” (ID 124).
Reporters described scenarios where assessment of the patient airway indicated an advanced airway management strategy should be implemented, but patient preferences required a departure from the ideal approach: “The patient stated that * weeks earlier the other anaesthetist did an awake intubation. * recalled a lot of what went on and * had found it unpleasant. [The patient] asked if there was a real need for [them] to go through the same again” (ID 12). While advanced asleep intubation techniques initially failed, the ability to facemask ventilate and oxygenate this patient, including excellent communication with the surgical team and lateral thinking of the treating anaesthetist, prevented patient harm.
Unwell patients in extremis, requiring urgent airway management were described: “Seen patient urgently with severe dyspn[o]ea/swollen tougue (sic) MP3, Mouth opening 2–3 cm and unable to lie flat with hoarse voice” (ID 49). In these instances, there were clear risks to patient safety. Described Safety-II factors included excellent planning “Everything prepared with plan written on white board everyone allocated roles”, successful facemask ventilation, “2 handed 2 person technique with high APL pressures able to slow bag up to 100% with guedel airway insitu” and the use of a hyperangulated videolaryngoscope, which led to a successful outcome despite an initially failed first intubation attempt (ID 49).
Surgical factors and medical factors
Surgical and medical factors considered contributing to an increased patient safety risk in the included no-harm incident reports were variable. WAD differed significantly from WAI, as described in a number of reports, including surgical urgencies requiring the rapid care of patients with complex medical issues: “Emergency * laminectomy + evacuation of epidural abscess …PMHx: Morbid obesity … Severe OSA … Untreated. Poorly controlled T2DM HTN Dyslipidaemia” (ID 129), the time of day when urgent cases presented to the operating theatre, with significantly unwell patients requiring immediate surgical management in unfavourable hours and not allowing time for detailed preparation and optimisation of the patient and environment: “patient in their 50’s, morbidly obese … was admitted for an emergency [surgery], and presented to ED just before midnight in extremis” (ID 108). Surgical factors affecting airway management also included the need for rapid changes in the initial airway management plan due to changing surgical requirements and were described commonly in incidents involving obstetric patients: “Just about to change to a long spinal needle when the obstetricians declared urgency of LSCS now immediate as foetal bradycardia” (ID 17).
Described Safety-II principles within the narratives, included the early recognition of difficult airways, coupled with consideration of the patient’s complex surgical and medical requirements and preparation involving rapid decision-making in evolving unexpected conditions.
Theme 2: Variable system factors
Reporters identified a number of system factors that affected patient safety, including the theatre team and their team dynamics, availability of appropriate equipment and the physical environment during airway management procedures. The analyses of these components highlighted significant variability within health systems, with some factors contributing to the development of near-miss events and some reducing risk and supporting positive patient outcomes.
Theatre team
The theatre team comprises surgical and anaesthetic professionals complemented by specialised nursing and maintenance staff. Team dynamics in this analysis was viewed as the interplay of interpersonal relationships, communication patterns, roles and collaborative behaviours among the theatre team working together to secure the patient’s airway. At both anaesthetic and surgical team levels, Safety-II was reinforced by these non-technical skills. The analysis included scenarios illustrating ideal team compositions and interactions within an ideal professional environment: “Opted for awake fibreoptic intubation in conjunction with ENT surgeon who was scrubbed and available for emergency tracheostomy.” (ID 50), but also, situations marked by understaffing and non-collegial team interactions. Communication issues, “… although there was good communication between the tech and myself there was not great communication with the rest of the team” (ID 47) and tense team dynamics were described, “Solo anaesthetist … Consultant surgeon had been agitating all day” (ID 1). Some reporters indicated to emotional distress consequently: “… by this stage I was feeling a little flustered” (ID 1). The ability to successfully oxygenate the patient via facemask between failed intubation attempts and the support of a different anaesthetist arriving in theatre and using advanced airway equipment to intubate this patient were owing to Safety-II factors preventing a harmful outcome in these events.
The postponement of the initiation of airway management procedures was described as important Safety-II factors in reports where measures were taken to improve the team composition and ensure the presence and shared understanding among the entire team concerning the formulated airway strategies. “Anaesthetic registrar, fellow, consultant present for procedure on weekend late afternoon. Planned not to commence until ENT consultant in the room and all aware of airway plan” (ID 93).
Equipment
Several airway adjuncts and advanced types of airway equipment were described as being either used or accessible across all reported incidents. These included bougies, videolaryngoscopes featuring various laryngoscopy blades, flexible bronchoscopes, nasendoscopes, airway exchange catheters, high-flow nasal oxygen devices, rigid laryngobronchoscopes, ultrasound and X-ray. Availability and familiarity with these items of equipment guided the initial airway management strategy but also influenced rescue strategies: “… nasendoscopy revealed a swollen tongue and epiglottis, but it was possible to see the glottic opening. The patient was sedated and a combination of a fiberoptic bronchoscope with the tube loaded and a C-Mac video laryngoscope allowed the patient to be safely intubated with no episodes of hypoxia” (ID 78).
Numerous incident reports described events where the required equipment was not immediately available, where staff were unfamiliar with equipment, the equipment was not performing as intended or was broken. Reverting to facemask ventilation by the treating anaesthetist was commonly a successful technique to bridge the time until the required equipment arrived. “CMAC 4 with portable monitor, floppy epiglottis unable to lift. Changed to a D blade, however D blade brought was not compatible with the portable monitor. There was a delay getting a compatible CMAC monitor” (ID 126). In that case, rescue oxygenation using a laryngeal mask which consequently was used as a conduit for flexible bronchoscopy ensured a successful outcome without patient harm.
Environment
The operating room environment, defined in this study as the location where airway management was actively managed, included the operating theatre, as well as the ICU or emergency department (ED). Considerations relating to environment regulation, including infection control requirements, the anaesthetic workstation, monitoring equipment, ventilators and airway management devices were also examined in this theme. Most incident reports did not include mention of specific focus on the airway management environment unless it was suboptimal. A number of incidents described the optimisation of the environment by transporting critically ill patients urgently to the operating theatre: “Transferred to OT for immediate AFOI” (ID 81).
Examples where infection control requirements negatively affected the work environment and complicated airway management were described in scenarios requiring Covid precautions. The requirement led to critical situations of reduced equipment and reduced the number of staff in the operating theatre. “… During induction, it was only myself and the anaesthetic nurse present in the theatre (as being treated as a possible Covid)” (ID 14). Clear communication between the anaesthetist and the anaesthetic nurse, in addition to the difficult airway trolley brought into the theatre at the anaesthetist’s request, contributed to a safe patient outcome.
Theme 3: The treating anaesthetist
What became apparent in the analysis of incident reports was the impact of the treating anaesthetist and how they perceived their role in various aspects of the incidents. This included the initial airway management strategy but also initiating the rescue airway management and dealing with the dynamically changed patient and system factors after the failed initial airway attempt. The involved anaesthetists in this study included clinicians of various levels of expertise “Senior FANZCA anaesthetist involved with case …” (ID 7), working at variable times of the day, in a number of different team sittings. This included solo clinicians, clinicians educating trainees and teams where the complexity of the patients and system factors required more than one senior anaesthetist to be involved. “A second consultant anaesthetist was recruited to assist with induction and airway management as well as a second anaesthetic tech” (ID 124). “Having second anaesthetist present to assist with positioning was helpful to achieving intubation” (ID 25). The nature of the incidents required these anaesthetic clinicians to be particular in their planning but also adaptable and dynamic in their responses to unexpected events: “Airway plan already discussed with anaesthetic nurse at start of case in event of conversion” (ID 4). The anaesthetist’s integration of patient-specific and system-factor considerations in initial and rescue airway management planning appeared as factors that improved patient safety in numerous instances and contributed to favourable patient outcomes across all reported no-harm incidents.
Anaesthetists were also required in out-of-theatres airway management, often for significantly ill and critical patients in ICU or ED. These incidents added extra complexity, often required rapid actions in an unfamiliar remote environment without their usual anaesthetic team support network and equipment: “Insufficient advanced notice, planning and call for anaesthetic presence and assistance – ICU consultant not available via phone – CICO kit not immediately available in ICU airway trolley” (ID 20). The patient was ventilated with a supraglottic device until an experienced anaesthetist arrived, who then successfully intubated the patient before any harm occurred, using a hyperangulated videolaryngoscope and bougie.
Theme 4: The airway management strategy
Factors identified in Themes 1, 2 and 3 all relate to airway management strategies. Thorough planning and formulation of airway management strategies—including utilisation of additional support and equipment—were described in several incidents. A large number of the reports describe the necessity and importance of a thought through airway management strategy and the need for meticulous planning, especially in the context of potentially time-critical situations: “I also realise how difficult it is to think under pressure in this situation and hence the reason for clear planning” (ID 47).
In the context of elective procedures, for patients with anticipated difficult airways, the implementation of comprehensive airway management strategies successfully mitigated adverse events: “A patient … was scheduled for an elective [procedure] … had been assessed as potentially having a difficult airway … the patient disclosed that … prior at another hospital, they had difficulty putting [the patient] to sleep and [the patient] woke up later that day in the ICU with an emergency tracheostomy in situ and cancellation of the surgery. [The patient] was unable to expand further on what had happened … the records from the time of the incident were not available … A second consultant anaesthetist was recruited to assist with induction and airway management as well as a second anaesthetic tech. The patient was ramped … and prior to pre-oxygenation, the cricothyroid membrane was scanned and marked via ultrasound. Pre-oxygenation was completed with an ETO2 > 80% prior to induction. A glide scope with the S4 Lo-Pro blade was used for laryngoscopy and achieved a POGO of 100%. Intubation occurred with an #8.0 ETT loaded onto the glide scope intubating stylet and only a single attempt … Oxygen saturation remained at 100% throughout” (ID 124).
Airway management
Only a few incidents described WAI including successful initial airway management: “All appropriate equipment (C-MAC) and Duty Anaesthetist in attendance. Intubation uneventful” (ID 26), involving successful oxygenation via facemask after induction, followed by successful intubation. Successful laryngeal mask insertions as initial airway securement devices were not commonly described within this cohort, as this qualitative audit was analysing incidents reported to webAIRS categorised as difficult or failed intubation. The successful use of the laryngeal mask as a rescue device was described in a number of incidents.
A lot of the incidents described unsuccessful initial airway management, thereby requiring rescue airway management strategies, even when planning was meticulous beforehand: “The patient was assessed to be a likely difficult intubation – anterior larynx, but not assessed to be difficult ventilation. The plan was to use video laryngoscope—sent for—disposable CMAC brought/Induction of anaesthesia commenced before I realised – Grade 4 view achieved with disposable CMAC, grade 3b direct—changed to D—blade and easily intubated – prolonged BMV whilst waiting for equipment” (ID 105). Facemask ventilation was often described as a successful tool to bridge other definitive airway attempts but was also used as the final airway device in a few incidents: “Able to bag mask ventilate easily with one hand. Given that the surgery was concluding, the decision was made to abandon intubation attempts and patient woken up on bag mask ventilation. No desaturation was noted throughout the management” (ID 125). However, equally, there were other situations, where facemask ventilation was not successful, and the anaesthetist moved on to other rescue airway management techniques: “Unable to bag mask ventilate despite 2 hands/OPA; rescue LMA successful ventilation” (ID 54).
Rescue airway management was affected by variable patient and system factors and the individual anaesthetist’s response to these. Safety-II factors were identified in reports of healthy patients or pre-oxygenated patients: “The patient was fortunate enough to not desaturate during the period (SpO2 remained >92% for the duration of this event), however [the patient] was a young fit patient who had reserve” (ID 14), and in successful rescue oxygenation attempts using different airway devices, the absence of desaturation progression was commonly mentioned: “The patient was never given time to desaturate” (ID 10). Responses to call for additional assistance, “Help arrived – both anaesthetists and technicians” (ID 3) and advanced equipment was common practice, “… immediately requested the Morgan trolley be b[r]ought into the room, sux and propofol to be drawn up and administered – this occurred within ∼1 min” (ID 14), accompanied by team communications and planning to discuss the appropriateness and to enhance the efficacy of the rescue airway interventions: “Discussed whether to wake patient up or proceed with 3rd and final laryngoscopy attempt” (ID 28).
Rescue strategies were affected by variable patient and system factors as well as by the treating anaesthetist. Patients had undergone unsuccessful initial airway manipulations, potentially becoming unstable, while advanced rescue airway equipment was not available in all instances. The treating anaesthetist usually faced a higher cognitive load and potentially higher stress levels because of unforeseen events. The majority of reporters highlighted skilled assistance, a well-structured airway plan, favourable patient factors, a predicted difficult airway as well as advanced equipment as primary factors that preserved or enhanced safety. Analysis of the narratives identified the adaptivity of the treating anaesthetist, including their technical and non-technical skills as a Safety-II feature, that avoided the incident from progressing from a near miss to a harmful event. Returning to basic oxygenation techniques when initial intubation attempts failed, especially by using bag-mask ventilation and supraglottic airway devices, prevented the progression to a harmful event in a number of reports.
Discussion
In anaesthesia, effective airway management is essential and significantly impacts patient outcomes and safety. The complex interplay between patient factors, system factors and the anaesthetist’s expertise, including the dynamic nature of airway challenges, requires exploration beyond the examination of quantitative measures. Through a detailed inductive exploration of the narratives provided by anaesthetists to webAIRS, this qualitative investigation sheds light on strategies that shape airway management practices, contributing to a richer understanding of this pivotal aspect of anaesthesia care.
Analysing the narratives of reported airway events offers important insights into real-life work practices amid increasing procedural complexity and patient comorbidities. The incidence of difficult or failed airway management across Australia and New Zealand occurs in less than 1 airway incident per 1180 episodes of anaesthetic care. 7 It is remarkable how often anaesthetic airway management goes well 3 even when conditions are unfavourable. However, most research has mainly focused on adverse events, until now.
Analyses of adverse airway events reported to webAIRS
Focusing solely on adverse events offers limited new information. In the first 4000 incidents reported to webAIRS from July 2009 to June 2016, 4.25% were categorised as difficult or failed intubation, with a detailed analysis conducted on their contributing factors. 5 The analysis primarily focused on identifying factors leading to adverse events and their outcomes. Recommendations included improved attention to airway assessment, planning and equipment maintenance standards. 5
The same percentage of ‘difficulty with intubation’ incidents was published in the AIMS report in 1993. 1 The main focus was on contributing factors leading to these incidents providing similar recommendations. 1 A subsequent analysis of the first 4000 incidents reported to AIMS in 2005 reiterated the 4% rate of difficult and failed intubation, again, maintaining a primary focus on identifying contributing factors and similar conclusions. 6
The latest audit of airway incidents reported to webAIRS was a multicentre study across 12 tertiary centres in Australia and New Zealand in 2018, with comparable findings. 7 Reported patient outcomes varied from events with no harm to those resulting in patient injury or death. Identified contributing factors included, again, deficiencies in airway assessment and planning, equipment access and the necessity to enhance airway management strategy. Minimal emphasis on alleviating factors was described in this analysis.
Similar studies have been performed worldwide, with the focus on safety being the findings and recommendations of incident analyses. 13
The findings of Safety-II factors in this analysis complement the recommendations of these previous studies and confirmed that airway assessment, an airway management strategy and the availability of basic and advanced equipment are some of the important safety features to avoid patient harm. Additional important Safety-II factors preventing patient harm have been identified in this analysis (Table 1).
Safety-II features identified within the narratives.
MDT, multidisciplinary team.
Historically, the development of guidelines to manage difficult airways were developed from anaesthetic incident analysis findings and recommendations, thereby mainly focusing on Safety-I principles. 13 These guidelines describe planned and designed work procedures on how tasks should be performed,23 –26 similar to the concept of WAI. The findings in this analysis show that airway management is far more complex than described. In fact, the unpredictable and considerable variability of patient and system factors makes it impossible to devise guidelines to address every conceivable scenario. While Safety-I and WAI are valued and essential to ensure patient safety, it is noticeable that solely relying on Safety-I may not effectively address the multifarious variables involved in inconsistent and changing patient and system factors. Despite unpredicted complexities and complications, anaesthetic airway management still, more often than not, ends successfully. Adding Safety-II principles and reviewing the difference between WAI and WAD would make learning from events more thorough. Within the observed cohort of this paper all incidents were managed successfully, mostly due to adaptation to the changing situational demands and dynamic reactions of the anaesthetic frontline worker. While Safety-I views human factors and their inherent variability as a weakness, Safety-II recognises them as a source of resilience. This recognition is reassuring.
Several instances described returning to basic principles, such as reverting to oxygenation using a facemask, to be sufficient in creating room for cognitive space, team conversations, next-step planning, organising equipment and related essentials. Like the previous incident analyses focusing on Safety-I,1,5,7,8 this Safety-II analysis confirms the efficacy of oxygen provision via basic techniques. In reverting to these foundational principles and simplifying complexities, anaesthetists often successfully managed complex events where conventional guidelines might not have been sufficient.
While the events examined herein yielded positive patient outcomes, they were not without significant safety concerns and highlight the potential for divergent outcomes. A fundamental inquiry persists: Are there pre-emptive measures to avert adverse airway management events altogether? Previous analyses of incidents caused by difficult or failed intubation repeatedly recommend and refer to the importance of better airway assessment, planning and preparation, yet the percentage of incidents reported to webAIRS addressing failed and difficult intubation remains unchanged.5 –7
Using webAIRS purely as an incident reporting system is unlikely to answer these questions; per the nature of incident reporting, only data relating to adverse events or near misses are collected. Expanding on incident reporting and collecting data on all patients undergoing anaesthesia might provide deeper insights. One option is the development of specific audits that examine all airway management over a certain period, including routine events, without any incidents or near misses. While webAIRS has not been used in this way, the database might have the technical capacity to provide such a platform, as indicated in one of its previous airway audits. 7
Qualitative inductive content analysis
This analysis was an explorative study to examine whether inductive qualitative narrative analysis is a useful method for webAIRS incident analysis. Previous airway incident analyses by the first author used an established deductive narrative analysis approach, specifying details on what data to extract and review before the detailed narrative analysis. This approach is based on expert opinion and traditional webAIRS incident analysis methods and is useful for quantifying qualitative data, thereby providing a numeric overview, aiding the interpretation of the results.
In comparison, this current analysis used an evidence-based inductive method, which allowed themes to be identified during the detailed examination of narratives. Analysing and coding the text line by line led to a highly detailed breakdown of events, provided useful insight into how anaesthetists consider their situation and how they managed the different variables that emerged during adverse events. The benefit of this inductive approach is that the result provides a clearer representation of the reporters’ perspectives and experiences.
Considering future webAIRS incident analyses, the authors recommend using a mixed-method analysis combining qualitative inductive detailed content analysis with the analysis of the quantitative aspects of the data.
Limitations
The qualitative analysis of these data was performed by examining reports from anaesthetists. Therefore, the variables, factors and conceptual framework (Figure 4) that are identified and examined in this study may not reflect the perspectives of other health professionals nor of patients. As with other qualitative studies, the bias of the research team needs to be considered. In this study, the team consisted of one anaesthetist and two non-anaesthetic health experts. Interpretation of the data and themes may be different with a different team reviewing the same dataset. In this analysis, we focused on the narrative descriptions reported to webAIRS with demographic data of the reporters, and the involved patients were only included if specifically mentioned within the free-text entries. The entry fields in webAIRS are non-mandatory and voluntarily, and text entries might be incomplete. As the data and reporter details are de-identified it is not possible to collect additional information about the events.
Conclusion
The findings of this study emphasise the dynamic nature of health professionals’ adaptability. Adaptability served as a strength when patient safety was threatened and when WAI varied significantly from the work that needed to be done. Expanding the investigation of patient safety to include Safety-II principles provides a deeper understanding of airway management safety. Reviewing routine procedures without incident in addition to analyses of adverse events and near misses is valuable and should continue to be explored in future research. Detailed qualitative ICA of the webAIRS text-box narratives provided deep and meaningful exploration of the provided information and we recommend its use in future incident analyses to complement the traditionally performed quantitative analyses.
Footnotes
Acknowledgements
The authors would like to thank the large number of anaesthetists across Australia and New Zealand who report incidents to webAIRS in the interests of patient safety. The authors would like to acknowledge the contribution of all members of the Australian and New Zealand Tripartite Anaesthetic Data Committee (ANZTADC), past and present. Special thanks to Dr Chris Acott, Dr Anna Steer and to the ANZTADC Publication Committee for their contributions to the editing and review of this manuscript.
Author Contribution(s)
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article. Dr Yasmin Endlich is the current ANZTADC Medical Director and involved in quality improvement projects involving webAIRS and its database.
