Description

On 24 March 2015, an Airbus A320 (D-AIPX) being operated by Lufthansa Group subsidiary Germanwings on a scheduled passenger flight (4U9525) from Barcelona to Dusseldorf under R/T call sign GWI18G was in the cruise at FL380 in day Visual Meteorological Conditions (VMC) when ATC lost contact with it. The wreckage of the aircraft was subsequently found scattered over a wide area in mountainous terrain in the Alpes-de-Haute-Provence region of south eastern France. The aircraft was completely destroyed and all 150 occupants were killed.

Investigation

A Safety Investigation was carried out by the French Civil Aviation Accident Investigation Agency, the BEA. Wreckage at the accident site, which was located in mountainous terrain at an altitude of 5085 feet (1550 metres) amsl, was highly fragmented - the largest parts of the aircraft were about 3 to 4 metres long - and was spread over an area of 4 hectares in a sloping, rocky ravine - see the illustration.

The Cockpit Voice Recorder (CVR) and Flight Data Recorder (FDR) were both recovered and their data successfully downloaded. The Quick Access Recorder was also recovered but found to have been subjected to too much damage for its data - the same as that on the FDR - to be read. A GPWS/TAWS Warning "Terrain, Terrain, Pull Up, Pull Up" was found to have been activated 25 seconds prior to impact and to have remained active until impact occurred.

The synchronised recorded data was used to construct the descent profile and track and to establish that there was no evidence of any relevant aircraft fault, that the aircraft had responded to intentional flight management inputs with the AP engaged and that these inputs had been made by the First Officer whilst the Captain was being intentionally prevented from returning to the flight deck after leaving it shortly after the aircraft had reached its FL380 cruising level. Once alone in the flight deck, it was noted that the First Officer had not responded to any ATC or crew intercom calls. Collision with terrain occurred less than 9 minutes after the Captain had left the flight deck and the descent had been continuous.

A Preliminary Report detailing the initial progress of the Investigation was released on 6 May 2015. This included the conclusion that "During the cruise phase, the co-pilot was alone in the cockpit. He then intentionally modified the autopilot instructions to order the aeroplane to descend until it collided with the terrain. He did not open the cockpit door during the descent, despite requests for access made via the keypad, the cabin interphone and knocks on the door."

FDR and CVR data (see the illustration below) also showed that during the outbound flight, whilst the Captain was absent from the flight deck for a period of just over 4 minutes, an ATC clearance to begin descent from the cruising level FL370 had been received, initially to FL 350 and then soon afterwards to FL 210. Initially, the selected altitude of FL 350 had been set and then it was "decreased to 100 ft for three seconds and then increased to the maximum value of 49,000 ft" before being returned to FL350. Almost immediately, the clearance to FL 210 was given and initially set before being again reduced to 100 ft for most of the next two minutes until just before the buzzer to request access to the flight deck by the Captain was recorded, at which point it was set to FL 250. Before reaching this level, a further clearance to FL170 was received and set and the rest of the flight was completed normally. It was noted that "due to the engaged autopilot modes, the changes in selected altitudes described above did not influence the aircraft descent flight path" and the Investigation concluded that these actions "may be interpreted as a rehearsal for suicide" and that all the actions of the First Officer could only be interpreted as determined by an intention to commit suicide.

The Investigation established that the flight deck door locking mechanism was one commonly fitted to A320 series aircraft for the purposes of preventing access to the flight deck except when permitted by its occupant(s). It was noted that inward opening door had, as would usually be the case, an outward opening escape hatch in its lower section and that the usual, in this case three, cameras were installed to allow the occupant(s) of the flight deck to clearly see the area of the forward cabin just outside the door.

The Investigation noted that both pilots were German nationals. The 34 year old Captain had, after initial pilot training with Lufthansa, served initially as a First Officer on the A320 at Condor Berlin and then, for just over 4 years as a First Officer on the A330 Family/A340 Family fleet at Lufthansa. Then, he had been transferred to the Lufthansa Group's wholly-owned subsidiary Germanwings where he had received command training on the A320 and completed his qualifications as a Captain on type just over 6 months prior to the investigated event. By this time, he had accumulated 259 hours in command and had accumulated a total of 6763 flying hours.

The 27 year old First Officer was found to have been selected for ab initio pilot training by Lufthansa in 2008 and been issued with a Class 1 Medical Certificate by an Aeromedical Examiner (AME) employed by the Lufthansa Aeromedical Centre (AeMC) in April 2008. He had then commenced pilot training in September 2008, but after only two months, this had been "suspended for medical reasons" for nine months before restarting due to a period of depression which included hospitalisation. His Class 1 Medical was not renewed in April 2009 "due to depression and the taking of medication to treat it" and his application to the Lufthansa AeMC for renewal three months later was rejected and the LBA (German CAA) were informed. Then, two weeks later, following an assessment by a psychiatrist at the same AeMC which concluded that "the depressive episode was over", the AeMC issued him with a restricted Class 1 Medical Certificate "with a waiver stating that it would become invalid if there was a relapse into depression". No further referral to the LBA was made at this time; the effect of the waiver was that any AME about to conduct an assessment for the extension or renewal of the Certificate must contact the licence issuing authority (the LBA) before proceeding with such a medical evaluation in order to appraised of information on the applicants medical history of depression. The same waiver was included in each subsequently-issued Class 1 Medical Certificate including the one still current at the time of the crash, issued on 28 July 2014 and valid until 14 August 2015. There was no interruption to continuity of successive annual Medical Certificates because the non-specialist AMEs involved in each extension/renewal did not detect any symptoms causing them concern in respect of mental health. The Investigation noted that "the waiver issued neither included the requirement for regular specific assessments by a psychiatrist nor reduced the time in-between two assessments".

Following the issue of the Class 1 Medical Certificate in July 2009, pilot training had been re-commenced the following month and continued during 2010, 2011 2012 and 2013. From 15 June 2011 until 31 December 2013, he was "under contract as a flight attendant for Lufthansa while continuing his training". Pilot training eventually culminated with the initial issue of a Multi-crew Pilot Licence - MPL(A) - endorsed with an A320 Type Rating on 11 February 2014, by which time he had been transferred to Germanwings to be prepared for line operations. This transfer was in line with the system for initial pilot training which meant that at the end of their training by the Lufthansa Flight Training School, "the management board decides where to allocate the personnel, between the Group’s operators: Lufthansa, SWISS, Austrian Airlines, or Germanwings." After training at Germanwings, he had qualified as a First Officer there in June 2014 and by the time of the investigated event, he had accumulated 540 hours on the A320 out of a total of 919 flying hours.

Then, in December 2014, the now-qualified First Officer "started to exhibit symptoms, possibly associated to a psychotic depressive episode" and "consulted several doctors, including the psychiatrist treating him on at least two occasions, who prescribed anti-depressant medication". The Investigation found that during the month prior to the crash, "a private physician diagnosed a psychosomatic disorder and an anxiety disorder and referred the co-pilot to a psychotherapist and psychiatrist" and, just two weeks before it, "diagnosed a possible psychosis and recommended psychiatric hospital treatment". On both occasions, sick leave certificates were issued as were several others by different physicians during February and March 2015. It was found that "not all of these sick leave certificates were forwarded to Germanwings" and that there had been "times when the co-pilot flew during some of these periods of sick leave, in particular on the day of the accident". It was also found that during this period, a psychiatrist had prescribed "anti-depressant and sleeping aid medication". The Investigation noted that in the three months prior to the crash, he "had had six documented periods of sickness...and had flown on 35 days over that same period". The Investigation concluded that on the day of the crash, "the pilot was still suffering from a psychiatric disorder, which was possibly a psychotic depressive episode, and was taking psychotropic medication" and was quite clearly unfit to fly. However, it was concluded that "no action could have been taken by authorities and/or his employer to prevent him from flying that day" because the pilot had failed to inform them or his AME of his 'relapse' (which was expressly contrary to the obligations placed on holders of all Pilot Medical Certificates) and no other party - the physicians treating him who knew he was ill and colleagues or family members who might have observed corresponding indications - had done so. All available information on the First Officer's medical history, including documentation related to the consultations with private physicians which he had sought during the four months prior to the crash, was collected by the German Air Accident Investigation Agency, the BFU, and shared with the BEA. This was then "analysed in detail by a German expert in aviation medicine and a German psychiatrist" and this analysis was then reviewed by "a team of experts, formed by the BEA, and composed of British aeromedical and psychiatric experts as well as French psychiatrists". This team concluded that "the limited medical and personal data available to the safety Investigation did not make it possible for an unambiguous psychiatric diagnosis to be made". However, it was noted that the majority of the team "agreed that the limited medical information available may be consistent with the (First Officer) having suffered from a psychotic depressive episode that started in December 2014, which lasted until the day of the accident" although "other forms of mental ill-health cannot be excluded and a personality disorder is also a possibility". It was observed that the assessment had been hindered by the refusal of both the relatives of the First Officer and his private physicians to be interviewed by the BEA and/or the BFU and the absence of any means to compel this. Toxicological examinations of his remains found at the accident site were noted to have found traces of two antidepressant medications and one sleeping-aid medication.

The Investigation noted that "depressive disorders and medications used to treat depression are usually medically disqualifying for pilots" and that "aviation authorities have not unified their approach on the consequences that the use of specific antidepressants and/or use of psychotherapy may have on pilot medical certification" with some regulatory authorities allowing pilots taking specific antidepressants such as Selective Serotonin Reuptake Inhibitors (SSRI), to fly subject to mandatory monitoring. It was also noted that a psychotic disorder or psychosis, which it appeared likely that the First Officer had been suffering from in the period leading up to the crash "is characterised by a loss of connectedness with reality, in the form of delusions, hallucinations, or disorganised thoughts" and that "it may be chronic, intermittent, or occur in a single episode". Expert medical opinion to the effect that "serious mental illness such as acute psychosis was relatively rare and its onset difficult to predict" was noted.

A key finding of the Investigation in respect of the First Officer's clear unfitness to fly for three months prior to the crash was that the private physicians and psychiatrists he had consulted had, despite likely awareness of his occupation, considered that medical confidentiality should not be breached in the wider public interest. This position was found to be underpinned by the requirement under German data protection regulations that meant that only a determination of "imminent danger" to the patient or others constituted an acceptable reason for a physician diagnosing an illness not compatible with pilot duties or with flight safety to be divulged without patient consent. No such determination was made by any of the private physicians who saw the First Officer.

It was noted that with effect from 8 April 2013, the EU “Aircrew Regulation” had come into force in Germany and that a standardisation inspection carried out at the LBA by EASA in July 2014 "had concluded that the LBA lacked powers to fulfil its responsibilities in the medical domain and that no process for internal audits or safety risk management was in place" and noted the existence of "possible conflicts between German data protection laws and the application of the Aircrew Regulation Part-MED" which require that the State Safety Regulator must have effective oversight of AMEs including those working at Company AeMCs. It was noted that EASA had advised the BEA in February 2016 that some related findings from this Inspection were still open.

The Investigation also noted that in the presence of a locked flight deck door, the mitigation of the risk of 'pilot incapacitation' was effectively valid only in the case of physical illness if the procedures advocated in ICAO Doc 8984 the "Manual of Civil Aviation Medicine" were applied. It was also noted that "specialists in aerospace medicine and psychiatrists......generally agree that serious mental illnesses involving sudden psychosis are relatively rare and their onset is impossible to predict" and that "for recurrent mental disorders that come in cycles, crises can leave no traces" so that when medical assessments take place in "the calm period of a cycle", any disorders can go undetected. In any case, "detection tools and methods can remain ineffective in cases where the patient is intentionally hiding any history of mental disorder and/or is faking being in good health".

The Investigation noted a small number of previous accidents and incidents in which the mental health of one of the pilots had been or may have been a precipitating factor including in particular a 2012 incident involving a JetBlue A320 on a domestic flight in the USA and the fatal accident to a Mozambique Airlines E190 when in Namibian airspace on an international flight the following year. However, no corresponding actions taken as a result were identified.

Safety Action The Investigation noted action taken by a number of bodies to review the risks posed by pilots experiencing mental health problems whilst the BEA Investigation was in progress as follows:

• The EASA, in response to a request from the European Commission, set up a Task Force which made six recommendations in its Final Report to the Commission on 17 July 2015 and subsequently followed up with an Action Plan on 7 October 2015.

• The German Federal Ministry of Transport and Digital Infrastructure (BMVI) set up a 'Task Force - Airline Safety' under the auspices of the German Aviation Association (BDL) to "determine what lessons could be leaned from the crash". This published a Final Report in November 2015.

• The UK Department for Transport (DfT) tasked the UK CAA in April 2015 "to review the assessment of mental health in pilots and to consider whether any changes to the current assessment system should be proposed". 'A Mental Health Working Group' was therefore established and made 11 Recommendations. Neither these nor any Report on how they were arrived at has yet (March 2016) been published but the text of all the Recommendations was included in the Official Report of this BEA Investigation (pps 79-80).

• The FAA in the light of both the Germanwings crash and the disappearance of the Malaysian Airlines Boeing 777 in 2014 established in May 2015 the Pilot Fitness ARC which developed as yet (March 2016) unpublished recommendations.

• The (US-based) Aerospace Medical Association which had already setup a 'Pilot Mental Health Expert Working Group' following the JetBlue Incident in 2012, reconvened following the Germanwings crash to review and update its Recommendations which were then published.

The formally-stated Conclusion of the Investigation in respect of Causes was as follows:

The collision with the ground was due to the deliberate and planned action of the co-pilot who decided to commit suicide while alone in the cockpit. The process for medical certification of pilots, in particular self-reporting in case of decrease in medical fitness between two periodic medical evaluations, did not succeed in preventing the co-pilot, who was experiencing mental disorder with psychotic symptoms, from exercising the privilege of his licence.

The following factors may have contributed to the failure of this principle:

• the co-pilot’s probable fear of losing his ability to fly as a professional pilot if he had reported his decrease in medical fitness to an AME;
• the potential financial consequences generated by the lack of specific insurance covering the risks of loss of income in case of unfitness to fly;
• the lack of clear guidelines in German regulations on when a threat to public safety outweighs the requirements of medical confidentiality.

Security requirements led to cockpit doors designed to resist forcible intrusion by unauthorized persons. This made it impossible to enter the flight compartment before the aircraft impacted the terrain in the French Alps.

A total of 11 Safety Recommendations were made as a result of the Investigation covering:

• the medical evaluation of pilots with mental health issues;
• anti-depressant medication and pilot flying status;
• the balance between medical confidentiality and public safety;
• the routine analysis of in-flight pilot incapacitation;
• the mitigation of the consequences of the loss of a professional pilot licence;
• the promotion of pilot support programmes.

These were as follows:

• that the EASA require that when a class 1 medical certificate is issued to an applicant with a history of psychological/psychiatric trouble of any sort, conditions for the follow-up of his/her fitness to fly be defined. This may include restrictions on the duration of the certificate or other operational limitations and the need for a specific psychiatric evaluation for subsequent revalidations or renewals. [FRAN2016011]

• that the EASA include in the European Plan for Aviation Safety an action for the EU Member States to perform a routine analysis of in-flight incapacitation, with particular reference but not limited to psychological or psychiatric issues, to help with continuous re-evaluation of the medical assessment criteria, to improve the expression of risk of in-flight incapacitation in numerical terms and to encourage data collection to validate the effectiveness of these criteria. [FRAN-2016-012]

• that the EASA, in coordination with the Network of Analysts, perform routine analysis of in-flight incapacitation, with particular reference but not limited to psychological or psychiatric issues, to help with continuous re-evaluation of the medical assessment criteria, to improve the expression of risk of in-flight incapacitation in numerical terms and to encourage data collection to validate the effectiveness of these criteria. [FRAN-2016-013]

• that the EASA ensure that European operators include in their Management Systems measures to mitigate socio-economic risks related to a loss of licence by one of their pilots for medical reasons. [FRAN2016014]

• that the IATA encourage its Member Airlines to implement measures to mitigate the socio-economic risks related to pilots’ loss of licence for medical reasons. [FRAN-2016-015]

• that the EASA define the modalities under which EU regulations would allow pilots to be declared fit to fly while taking anti-depressant medication under medical supervision. [FRAN-2016-016]

• that the World Health Organisation (WHO) develop guidelines for its Member States in order to help them define clear rules to require health care providers to inform the appropriate authorities when a specific patient’s health is very likely to impact public safety, including when the patient refuses to consent, without legal risk to the health care provider, while still protecting patients’ private data from unnecessary disclosure. [FRAN-2016-017]

• that the European Commission, in coordination with EU Member States, define clear rules to require health care providers to inform the appropriate authorities when a specific patient’s health is very likely to impact public safety, including when the patient refuses to consent, without legal risk to the health care provider, while still protecting patients’ private data from unnecessary disclosure. These rules should take into account the specificities of pilots, for whom the risk of losing their medical certificate, being not only a financial matter but also a matter related to their passion for flying, may deter them from seeking appropriate health care. [FRAN-2016-018]

• that without waiting for action at EU level, the German Federal Ministry of Transport and Digital Infrastructure (BMVI) and the Bundesärztekammer (BÄK) (the German Medical Association) edit guidelines for all German health care providers to:
  • remind them of the possibility of breaching medical confidentiality and reporting to the LBA or another appropriate authority if the health of a commercial pilot presents a potential public safety risk.
  • define what can be considered as “imminent danger” and “threat to public safety” when dealing with pilots’ health issues.
  • limit the legal consequence for health care providers breaching medical confidentiality in good faith to lessen or prevent a threat to public safety.

[FRAN-2016-019 and FRAN-2016-020]

• that the EASA ensure that European operators promote the implementation of peer support groups to provide a process for pilots, their families and peers to report and discuss personal and mental health issues, with the assurance that information will be kept in-confidence in a justculture work environment, and that pilots will be supported as well as guided with the aim of providing them with help, ensuring flight safety and allowing them to return to flying duties, where applicable. [FRAN-2016-021]

The Final Report of the Investigation was published in English on 13 March 2016 and simultaneously also in French, German and Spanish.

Related Articles

• Unlawful Interference: Guidance for Controllers
• Flight Deck Security

Further Reading

• ICAO Annex 1
• ICAO Manual of Civil Aviation Medicine (Doc 8984)
• Final Report of the EASA 'Task Force on Measures Following the Accident of Germanwings Flight 9525',17 July 2015. - see source below.
• Action plan for the implementation of the Germanwings Task Force recommendations; EASA, 7 October 2015.
• The German Aviation Association (BDL) Report 'Task Force Airline Safety'
• AsMA Pilot Mental Health Working Group Recommendations as submitted to the FAA in September 2015.