Since 2015, a research team from Trinity College Dublin has been attempting to understand how sources of work-related stress are affecting pilot wellbeing and, ultimately, impacting on performance and flight safety By Captain Paul Cullen, IFALPA-accredited Accident Investigator A READ MORE About the author viation is considered ultra-safe. Fatal accidents are exceptionally rare but what if we are using the wrongmetric to measure safety? Should we be aimingnot only to avoid fatalities, but also to do no harm in the process? SThe traditional Safety-I model focuses on the relatively rare events when something goes wrong, and things end badly. The Safety-II model, or the new view of safety, examines the relatively higher proportion of events when things end well. Despite things not going exactly as planned or imagined, somehow because of resilience in the system a safe end result is still reached. We cannot manage what we cannot measure, and safety is no exception. Within aviation, safety performance is monitored using key performance indicators (KPIs), and safety is managed using a combination of preventative and reactive measures. Very often, however, the most dangerous hazards and threats are those of which we are not aware or those of which we are aware but, naively, assume do not apply to us. In terms of managing risks associated with pilot wellbeing, we are neither operating within the Safety-I nor Safety-II model, but, instead, within Safety-ZERO. We dont know why things go wrong for some pilots, or why things go right for others. In fact, we know little or nothing about the relationship between pilot wellbeing and flight safety. Unless a question is asked, an answer will never be found, and no evidence of a problem will continue to be taken as evidence of no problem. Theres ample evidence to indicate that pilots are suffering with mental health issues. A 2012 study of Brazilian pilots reported an incidence rate of 23.7% for common mental disorders. In 2016, Harvard University reported that over a twoweek period 12.6% of pilots met the threshold for clinical depression or a major depressive disorder. Over the same period, 4.1% of pilots reported having suicidal thoughts. Eindhoven University of Technology studied British pilots and, in 2018, reported that 40% of the participants experienced high burnout. There are approximately 130,000 commercial pilots flying today. Ifthe results of the Harvard study reflect what is going on throughout the world, there could be more than 16,000 pilots who meet the threshold for clinical depression and more than 5,000 who have had suicidal thoughts in the past fortnight. Equally, the Eindhoven study suggests there could be more than 50,000 pilots flying while experiencing burnout. In addition, if Brazilian pilots are representative of their colleagues across the globe, almost 31,000 pilots could be dealing with a common mental disorder right now. NOTAMS To begin with, the flight crew were unaware or could not recall information of a runway closure at SFO. The NTSB report indicates that this pre-flight information was presented via NOTAM page eight of 27, and was the fourth line on the in-flight ACARS information. Pilots reading this will know the reality of pre-flight reviews of NOTAMs. They are not easy to digest, and there is often insufficient time to read and disentangle poorly described bits of information. This can place an intolerably heavy burden on crews to find the needle in the haystack that is relevant to their upcoming flight, particularly when priority items may be buried under pages of other irrelevant items. In this incident, the first officer did not recall a specific NOTAM about 28L closure, while the captains actions in lining up on the taxiway suggest that he did not recall that information. Hence, for this approach, both flight crew had the expectation of illuminated normally configured parallel runways 28L/28R. Expectation bias and conflicting cues The captain and first officer were experienced and familiar with the SFO runway configuration. The captain had not experienced a previous occasion in which either runway had been dark, so both flight crew expected to see two illuminated parallel runways on their descent. While there would have been light cues coming from the taxiway, such as green (not white) centreline lights and taxiway in-pavement flashing guard lights, which differ to the surface of a runway, there were also a number of visually supporting clues that fed into taxiway Cs resemblance to runway 28R. Specifically, the navigation lights on the wingtips of the aeroplanes lined up on the taxiway partially resembled the runway edge lighting, while their flashing red beacon lights were consistent with features associated with approach lighting. Another supporting cue was the presence of runway and approach lights on the actual 28R, which would have appeared to the left of the pilots primary field of vision, and hence appeared to confirm they were correctly situated with respect to the normally adjacent parallel runway 28L. This expectation bias should not be understated runway lighting aids are normally powerfully conspicuous features, but humans are vastly better at recognising salient features that pop out than detecting missing features or slightly off-hue features, particularly under night-time conditions.. The NTSB analysis also notes that while a runway closure marker a flashing white X was placed at the start of runway 28L, it would not have been in the flight crews primary field of view, and the flash rate (2.5 seconds on and off) may have been too slow to capture their attention unless they were looking in that direction. The flight crew from the preceding DA 521 flight into SFO, who were aware of the runway closure, also reported that the taxiway lights gave the impression that the surface could have been a runway, especially as no aeroplane shapes could be seen. So, even though their expectations were different and they knew that runway 28L was closed they, too, felt the taxiway could be perceived as a runway. Setting these visual features aside, normal cues that could have been provided from the backup lateral guidance (via the localizer) were missing for AC 759. When the first officer (pilot monitoring) had set up the approach, he did not manually tune the ILS frequency, and the captain either did not notice, or did not address this. TODAYS PILOTS WORK IN AN ENVIRONMENT THAT IS RADICALLY DIFFERENT FROM THAT EXPERIENCED BY PREVIOUS GENERATIONS, THE HAZARDS OF WHICH ARE NOT FULLY IDENTIFIED OR UNDERSTOOD The role of pilot mental health in accidents When pilot mental health is mentioned, many people think of Germanwings 9525. In the aftermath of that tragedy, and to prevent a similar catastrophe, EASA went to great lengths to: 1.Detect and remove pilots with mental health issues 2.Give support to pilots with mental health issues As of summer 2020, Europes airlines will be required to introduce psychological screening, alcohol/drug testing and peer-support programmes. This response by EASA is seriously lacking a preventative element; it focuses on the symptoms of the problem while doing little to understand or deal with the causes. It is normal practice, as per ICAO Annex 13, to investigate accidents without apportioning blame, to prevent re-occurrence. Every last bit of wreckage is thoroughly examined and every piece of evidence scrutinised. The investigators dig until they can dig no deeper. Key causal factors are identified, and a solution is identified and implemented. Since 1982, worldwide, at least eight fatal accidents involving commercial airliners have been attributed to pilots deliberate actions. In investigating these accidents, is it possible that pilot mental health has been listed incorrectly as a causal factor? That it was a contributory factor in most of these cases is beyond dispute, but was it a causal factor? By listing mental health as a causal factor, perhaps the investigators stopped digging too soon. Since 1982, more than 1,400 fatal accidents involving commercial aircraft have not been attributed to the deliberate actions of the pilot. Over the past 37 years, fewer than 1% of fatal accidents have had deliberate actions of the pilot listed as a causal factor but pilot mental health could have played a role in at least some of the 99% for which it was not listed. Approximately 80% of aircraft accidents have human/pilot error as a causal factor. Is it possible that the pilots state of mind triggered the error in some of these? Not every pilot with a mental health issue is hellbent on destroying their aircraft and all those on board. Despite struggling with mental health issues, many are doing their best to perform their duties to the best of their ability. Being human, however, they have their limits as do pilots without mental health issues. The role played by pilot mental health needs to be considered, without demonising it, in the investigation of all accidents and incidents. To truly do our best to prevent future reoccurrence, we need to dig further and attempt to understand what causes mental health issues in the first place, and consider what mitigations can be implemented to improve the resilience of pilots. Work-related stress Recent research indicates that aspects of a pilots job undermine wellbeing, and affect their physical, mental and social health. Pilot performance is affected by a variety of performance-shaping factors (PSF). These can be positive or negative, enhancing or degrading our performance on any given day. PSFs can be external such as the weather or organisational culture or internal, suchas alertness, experience, confidence or mental state. NegativePSFs are considered to increase the likelihood of human error substantially. Work-related stress (WRS) is defined as the response people may have when presented with work demands and pressures that are not matched to their knowledge and abilities, and which challenge their ability to cope. Things outside the workplace, such as family problems or debt, can be responsible for stress (personal stressors). A person experiencing stressful life events may find that they are less able tocope with the demands of work, even though work is not the cause and may not have been aproblem before. Pillars of wellbeing Since 2015, a research team from Trinity College Dublin comprising Captain Paul Cullen and Dr Joan Cahill and Dr Keith Gaynor from University College Dublin, has been trying to understand how sources of WRS affect pilotwellbeing and, ultimately, impact on performance andflightsafety. The biopsychosocial model of wellbeing allows us to think of our health as a threelegged stool, with each leg representing theinterdependent pillars of our wellbeing: physical, mental and social. Ifwe weaken one leg, the stool is weakened overall. The same applies to the pillars of our wellbeing. Our physical health affects ourmental health, and so forth. Preliminary field research involved informal interviews with morethan 100 pilots, through which sources of WRS were identified. Thisdata was used in conjunction with the biopsychosocial model of wellbeing, an extensive review of the scientific literature and inputfrom numerous medical experts toconstruct the LivedExperience of a Pilot model. This was furtherrefined and validated by a series of pilot workshops, held inearly 2018. The research examining the impact on flight safety is beyond the scope of this short article. Where do we go from here? To effectively address the safety implications of pilot mental health, we need to not only address the symptoms of the problem, but also the causal factors. In no other aspect of aviation do we rely solely on reactive defences. A preventative strategy needs to be adopted in parallel. Todays pilots work in an environment that is radically different from that experienced by previous generations, the hazards of which are not fully identified or understood. They are unknowingly participating in one of the largest, ethically unapproved, biopsychosocial experiments ever conducted. The parameters are not defined and the ultimate outcome is unknown; pilots are blindly heading into the abyss. Airlines are investing heavily in evidence-based training and upset prevention and recovery training. Evidence indicates, however, that the mental health of pilots is being upset by aspects of their profession, yet little training appears to be in place to prevent or recover from such upsets. It is not enough to focus solely on preventing aircraft from crashing. The complex interrelationships between sources of workrelated stress, the wellbeing of pilots and the impact on flight safety need to be understood and addressed. Further research is required to understand not only the frequency and severity of pilots mental health issues, but also to understand why this is happening and why, in many cases, it is not happening. Although some pilots are struggling, others are thriving within the same work environment. So what is it that makes some pilots resilient and others susceptible to mental health issues? Ifunderstood, this would greatly assist in identifying possible threats and mitigations. The what and the why are equally necessary if effective change is to be implemented. To this end, and as a starting point from which to advance, the research team has constructed a detailed general health questionnaire (GHQ). This will mirror existing studies, by containing standard instruments for the measurement of common mental disorders, and will look in detail at the biological, psychological andsocial aspects of the participants daily lives, inside and outsidethe cockpit. The GHQ can be accessed at www.surveymonkey.com/r/PilotWellbeingStudy If the pathogenesis of pilot mental health issues is not understood, we will be unable to reduce the likelihood of suchissues developing and will continue merely firefighting thisepidemic. If you would like to stay up to date with Paul and his teams work, visit www.tcd.ie/cihs/projects/pilot-lived.php READ MORE Captain Paul Cullen is an Airbus pilot, and previously held the role of Director ofSafety and Technical with the Irish Airline Pilots Association (IALPA). He is an IFALPAaccredited accident investigator and a research associate with the School ofPsychology, Trinity College Dublin. BURNOUT TIME FOR A NEW SAFETY YARDSTICK?