To Die For

The December 1997 SilkAir 185 crash in Indonesia and the October 1999 Egypt Air 990 crash in the U.S. have both focused attention on an aviation safety question that most of us would really rather not discuss: pilot suicide. Could psychological testing of pilots help prevent this sort of tragedy? Does cockpit crew size (three vs. two) make a difference? Does the FAA's policy of grounding pilots who take antidepressant medication help or hurt? AVweb's Ken Cubbin examines these and other facets of the problem.


Egypt Air 767OnOctober 31, 1999, 217 passengers and crew lost their lives when Egypt Air 990crashed into the sea off Nantucket Island, Mass., 40 minutes after having takenoff from New York’s JFK airport. With most of the wreckage now recovered fromthe ocean, investigators can find no evidence of mechanical failure that wouldwarrant reconstructing the airplane. Investigation into the cause of theaccident still continues. But, based on physical evidence, flight data and voicerecorder information, the predominant theory remains that pilot Gamil El-Batouty– who took the controls shortly after takeoff — deliberately caused the Boeing767 to dive into the sea. (The pilot’s family and Egypt Air officials vehementlydefend the pilot’s character and insist that investigators are incorrect intheir presumption.)

When aircraft manufacturers design airplanes, they incorporate failsafe andfail-operational features that make it most unlikely for any single mechanicalfailure to cause an accident. However, the one phenomenon that engineers find itimpossible to account for is when a pilot intentionally flies his aircraft intoterrain. This phenomenon is popularly referred to as “pilot suicide,”although in some cases it might be more accurate to call it “murder.”

Accidents like these raise troubling questions. How could such things happen,especially in the cockpit of a commercial airliner? How could a pilot whocommits such a crime have remained functional on a day-to-day basis before thetragedy? How could such an obviously troubled pilot have avoided detection byfamily and friends? Why couldn’t other crewmembers have stopped the pilot fromcarrying out his destructive act?

Could they have been stopped?

SilkAir 185 crash locationIn1997, a SilkAir Boeing 737-700 crashed into the Musi River in Indonesia, killing104 people. Investigators concluded that the most probable cause of thisaccident was deliberate, controlled flight into terrain by the captain who — inan effort to confuse future investigation into the accident — allegedly pulledcockpit voice recorder and flight recorder circuit breakers while returning froma visit to the rest room. If this scenario is true, then the captain’s actionsconstitute premeditated murder. Apparently, the captain had recently beendemoted from instructor and many other pilots who had flown with him hadinformally complained of his non-standard operation. Therefore, there was someevidence of aberrant behavior by this pilot prior to this accident that may havebeen a clue as to his state of mind.

In 1994, a Royal Air Maroc ATR-42-300 jet plunged into the ground ten minutesafter takeoff from Agadir killing all 44 people on board. Investigatorsconcluded that the captain had deliberately steered the plane into the ground.(The Moroccan Pilots Association disputes this claim.)

In 1982, a Japan Airlines (JAL) DC-8 crashed into Tokyo Bay while on approachto Haneda Airport in Japan, killing 24 and injuring 141. The captain allegedlypushed the nose down prematurely and pulled the inboard engines into reversewhile on approach to the airport. Despite attempts by the first officer andflight engineer to rectify the perilous flight path, the captain was successfulin his efforts to terminate the flight prematurely. After exhaustiveinvestigation of this accident, the captain was placed into a psychiatricinstitution. Sources at JAL who knew this captain personally informed accidentinvestigators that it was general knowledge the captain’s behavior had beenerratic before the accident. After the accident, there were accusations thatJAL’s management had been aware of the captain’s mental instability but hadfailed to relieve him of flight duty. However, Japanese authorities ultimatelyconcluded that both of these accusations were unfounded.

In several of the above examples, a pattern of aberrant behavior had beendemonstrated by the pilots who eventually took their own life and the lives ofothers.

Psychological testing of pilots

According to American Psychological Association(APA) databases, a number of papers have been written on the subject of pilotsuicide in general aviation. For example, a 1994 paper by Timothy J. Ungs titled”Suicide by Use of Aircraft in the United States, 1979-1989″ statesthat the NTSB reported nine fatal accidents attributed to pilot suicide duringthe 1979-1989 period. NTSB investigations identified evidence of importantadverse psychological factors in most of the pilots, including depression ornegative life events.

In 1998, Anthony S. Cullen reported in his paper titled “AviationSuicide: A Review of General Aviation Accidents in the U.K., 1970-96” thatout of 415 general aviation accidents, three definite cases of pilot suicide andpossibly another seven occurred in the United Kingdom. Again, the pilots’previous psychiatric problems, familial instability and alcohol misuse wereidentified as causal factors.

In Germany, Bernhard Maeulen reported in his 1993 paper titled “AnAeronautical Suicide Attempt — Suicide and Self-Destructive Behavior inAviation” that:

“Approximately 2%-3% of all fatal general aviation accidents in Germany may be attributed to suicide, and in many other accidents in aviation there are grounds for inferring that self-destructive and suicidal behavior was involved.”

Investigators concluded that precursors to these accidents included pilotdepression, alcoholism and family problems.

Although these papers all refer to general aviation, their conclusions appearto concur that the offending pilots had a history of previous psychiatric ordomestic problems and/or alcohol abuse. It would be logical to assume thatsimilar problems exist in the lives of airline pilots. What can be done toensure that such tragic events do not occur again in commercial aviation?

Three vs. two

SilkAir 185 pilot Tsu Way MingAtfirst look, it would appear logical that three crewmembers in the cockpit standa better chance of ensuring such events will not occur. For example, had threecrewmembers been on the flight deck of the SilkAir B-737, the captain mighthave found it much more difficult to surreptitiously pull circuit breakers whenreturning to the cockpit from an in-flight visit to the rest room. If he had notbeen able to throw this curve ball to future investigators, he might haveaborted his plan to crash his airplane.

Conversely, the JAL DC-8 accident in 1982 would suggest that three personsare no more effective than two at preventing pilot suicide, since thefirst officer and flight engineer were unable to prevent the captain from hisdire actions. Such a conclusion might be premature, however, because untilapproximately 15 years ago, the patriarchal society that has existed in Japanfor centuries dictated that subordinate males revere and never question seniormales. Therefore, one can only wonder how vigorously the first officer andflight engineer fought with the captain for control of the aircraft. Of course,even if the first officer and flight engineer fought the captain with all meansat their disposal, the aircraft was at low altitude with two engines in reverse,so perhaps it was a case of “too little, too late.”

After Cockpit Resource Management (CRM) programs were introduced intoJapanese airlines in the mid-80s, the status quo of blind hierarchical obediencebegan to change. In addition, contemporary youths in Japan’s general populationare questioning the validity of “the old way” and are rebelling invarious ways against their elders. As a result, if the same circumstances wereto occur today, it is probable that a Japanese captain would be more vigorouslyrestrained by the other crewmembers. Still, as mentioned previously, if theaircraft were low and on approach, any efforts to recover the aircraft mightstill be unsuccessful.

During my flight engineer training experience over the last 20 years, I haveconducted episodes in the simulator where the captain was instructed to act asthough he had become incapacitated. The fastest method of restraining thecaptain under such conditions, I discovered, was for the flight engineer tograsp and pull back on the captain’s shoulder harness. This was entirelyadequate for subtle or non-violent incapacitation as it kept the captain frominterfering with the first officer’s control of the aircraft. However, if thecaptain were physically resisting restraint — as he might if he were intent onkilling himself and all on board — it might be necessary for the flightengineer to use force to ensure the first officer could recover the safeoperation of the airplane. How much force might be necessary?

According to one report of the SilkAir crash, the sound of what is thought tobe the first officer being struck by a heavy object is heard on the CVR shortlybefore the aircraft impacted the ground. If a flight engineer had been on board,would he or she have used a similar measure of force to restrain the captain intime for recovery? Can a rational crewmember come to the conclusion that thecaptain must be violently stopped from his criminal intentions in time? I thinkit highly unlikely.

So although it might seem logical that having three flight crewmembers aboardall flight decks might prevent such occurrences, this is by no means certain.Three crewmembers might still be duped by the covert action of one — and twocrewmembers might be unable, or unwilling, to use sufficient force to restrainthe third in time to recover safe operation of the airplane.

Who’s at risk?

Scientific AmericanLetme say up front that I have no psychological or clinical psychiatric expertise.But an article titled “The Neurobiology of Depression” published inthe June1998 issue of Scientific American presents some opinions that I findquite interesting. At the time the article was published, its author Charles B.Nemeroff was professor and chairman of the Department of Psychiatry andBehavioral Sciences at the Emory University School of Medicine.

In his article, Professor Nemeroff describes the symptoms of depression asbeing quite different from “the blues” that everyone feels at one timeor another, including grief from bereavement. He states that depression caninclude a sense of overwhelming sadness, guilt, and a sense ofself-worthlessness. A person suffering from depression may lose appetite andhave trouble sleeping — or conversely, want to eat and sleep constantly. Suchpeople can be preoccupied with suicide and have difficulty thinking clearly,remembering, or taking pleasure in anything.

How can someone suffer from such debilitating effects and yet remainfunctional? Eva Winer, a spokesperson for the APA, explained that in her careeras a testing officer in a psychiatric hospital, she had seen “manydeep-seated, practically asymptomatic cases of ‘smiling’ or ‘larvae’ depressionthat didn’t impair daily functioning and easily went undetected for a longtime.” Therefore, presumably, a person can be severelydepressed, yet hide it from his or her peers.

Professor Nemeroff suggests “that 5 to 12 percent of men and 10 to 20percent of women in the U.S. will suffer from a major depressive episode at sometime in their life. Roughly half of these individuals will become depressed morethan once, and up to 10 percent (about 1.0 to 1.5 percent of Americans) willexperience manic phases in addition to depressive ones, a condition known asmanic-depressive illness or bipolar disorder… As many as 15 percent of thosewho suffer from depression or bipolar disorder commit suicide each year.”

In what may be a very disturbing statistic in relation to pilot suicide,Professor Nemeroff contends that “many people who kill themselves do soin a way that allows another diagnosis to be listed on the death certificate, sothat families can receive insurance benefits or avoid embarrassment.”

The FAA as “Doctor No”

FAA logoA deficiencyin serotonin in the brain stem can result in the affected person sufferingailments such as severe and chronic depression. This subgroup of depressionsufferers will find no relief unless their condition is treated withmedication. Antidepressants such as Prozac have been very successful in treatingdepression.

However, antidepressant medication is currently not approved by the FAA, anda pilot who is prescribed such medication will have his or her medical revoked.For transitory depressions, the pilot who chooses to take antidepressants canhave his or her medical restored after the he or she has discontinued medicationfor 60 to 90 days and the prescribing doctor confirms in writing that thepilot’s original condition is no longer evident.

The makers of Prozac, Ely Lilly and Company, state that the drug’s benefitsmay not become apparent until one to four weeks after beginning the medication.It also recommends that antidepressant medication be taken for six to 12 monthsto monitor its success. Approximately 17 million Americans have been treatedwith Prozac and the drug has been proven safe and effective over the last tenyears. Side effects can include nausea, insomnia, drowsiness, anxiety,nervousness, weakness, loss of appetite, tremors, dry mouth, sweating, decreasedlibido, impotence or yawning. These side effects tend to go away after severalweeks of medication.

Presumably, since usage of Prozac and other antidepressants is so widespread,other professionals in highly demanding and responsible positions continue tofunction while taking the drug. Doctors, school bus drivers and police officersare just a few job classifications that come to mind. If a cardiac surgeon istaking Prozac, does he or she suffer the same scrutiny by his or her regulatingauthorities? I don’t recall reading any reports of doctors who have had theirlicense revoked simply because they were taking antidepressants.

FAA Civil Aeromedical Institute logoHowever,since the FAA will not allow a pilot to fly while on antidepressant medicationand for a period of up to three months after cessation, if a pilot elects to betreated in this manner, he or she could be out of work for well over a year. Inadopting this attitude toward a treatment that has been proven safe over thelast ten years, the FAA tacitly forces a professional pilot with severedepression to make an agonizing choice: give up his career, or continue to flywithout treatment.

FAR Part 67.107 states that a pilot can have no established medical historyor clinical diagnosis of psychosis (delusions, hallucinations, etc.),personality disorder, neurosis, bipolar disorder or other mental condition. Thisis explicit and precludes a pilot any hope of getting treatment for depressionwithout having his or her medical revoked. A group of online AMEs, VirtualFlight Surgeons (,point out that the primary medical responsibility of the FAA is to gauge whethera pilot’s medical condition is compatible with safe flight. Medication and itseffect on treatment is secondary, and the FAA may also ground pilots who sufferfrom side effects of medication.

Perception vs. reality

Senate ChamberUnfortunately,as an agency that depends on Congress for its budget, the FAA has to beconcerned not only with actual safety issues, but also with the perception ofthe travelling public. In this regard, the powers-that-be at FAA Headquartersseem to believe that the travelling public would never accept the idea of havingtheir flight piloted by a person taking antidepressant medication.

The truth, however, is that by disqualifying pilots who are prescribedantidepressant medication, the FAA has created a situation that is potentiallymuch more serious: The traveling public may have their aircraft piloted by aperson who is severely depressed but masking his or her symptoms — or aperson who is taking medication on the sly — in order to keep his or her job.Under such circumstances, which flight will prove to be the last straw for thepilot who can see no light at the end of a tunnel?

Despite becoming more sophisticated in terms of mental disorders and acceptedtreatment, there is still a stigma assigned to those people who receive suchtreatment by the general population. Any number of stand-up comics and comediansroutinely refer to those who take Prozac and other antidepressants in aderogatory manner. A person who has a serotonin deficiency has a similar needfor medication in order to remain functional as a diabetic who needs insulin.However, one never hears a comedy routine belittling those who regularlytake insulin shots. Apparently, one chemical deficiency is socially acceptable,and the other is the butt of jokes. Ironically, the FAA will now considerreissuing a third-class pilot’s license on a case-to-case basis for those peoplewho require insulin to control diabetes. Several other medications used to treatother potentially debilitating ailments, such as migraine headaches andParkinson’s Disease, also have FAA approval.

As a result of the present situation, even if a pilot is aware thatantidepressants might alleviate his or her depression, he or she might elect to”tough it out” in order to remain employed and provide an income forhis or her family. To make matters worse, a pilot may be severely depressed butnot recognize the insidious nature of his or her condition; in this case,seeking medical attention might not even occur to him or her.

Between a rock and a hard place

It is the very nature of flying single-pilot aircraft that no other personcan assist while the aircraft is in the air. The pilot is the person who mustmake the decision, good or bad. Therefore, because a pilot spends the formativeyears of his or her flying career as a sole pilot-in-command, he or she developsa strong sense of independence and self-reliance. This independent streak mightprevent a pilot from seeking help if he or she has a problem that seemsinsurmountable. Alternatively, he or she may refuse to acknowledge that aproblem really exists.

For an airline pilot, flying changes from an enjoyable hobby to a career.Professional pilots are expected to conduct themselves with propriety anddiligence. Every year, a physical examination is conducted to assure authoritiesthat each pilot is physically and mentally capable of continuing to fly.

Most airline pilots marry, have children, and assume all the financialburdens typical of modern adults. As a consequence, the pressure of needing tocontinue to work to support a family might restrict a professional pilot fromtaking time off to address his or her problems.

Another factor that may cause airline pilots to hide or ignore their problemsand avoid medical help is the complex and sometimes adversarial relationshipthat exists between pilots, the FAA and airline management. Airline managers askpilots to confide their innermost problems, while simultaneously threatening totake away their livelihood. The FAA is even more intolerant, with the criteriafor losing a pilot’s medical mostly spelled out in the black-and-white clausesof FAR Part 67.

To be fair to airline management and the FAA, they have to imposestrict control over pilots’ fitness. It would be irresponsible of them notto remove pilots with severe mental problems or alcohol abuse from the flightline. However, the knowledge that a pilot can be relieved of flight duty andpossibly dismissed undoubtedly prevents many pilots from seeking medicalassistance, and discourages crewmembers from reporting misdemeanors committed bytheir peers. Who wants to be responsible for a pilot’s family suffering becauseof a loss of income? What pilot would voluntarily confess to a condition thatwould mean imminent removal from flight duty unless forced to do so?

In my airline career, I have known of only two crewmembers who ultimatelylost their job because of alcohol abuse. Both of these pilots were commonlyknown to have severe drinking problems, but none of their peers complained tomanagement nor refused to fly with them — probably for the reason stated above.Ultimately, both of the pilots self-destructed: One embarrassed himself and theairline by his behavior while deadheading and the other was arrested forinsobriety while on a layover. Both were given an opportunity by the airlineinvolved to seek help for their problem — however, both of the pilots resigned.

Only one pilot, with whom I have been acquainted, was reported to airlinemanagement by his peers for irrational behavior. When his manager called theaccused pilot onto the mat, he was faced with the reports by his compatriots –as a result, he resigned. Whether the pilot resigned over a sense of guilt orembarrassment, or whether he just couldn’t imagine continuing to work with otherpilots who had complained of his behavior, I will never know. To this day, thereare many pilots and flight engineers — myself included — who believe thatpersonality clashes with a few of this pilot’s peers brought about complaints toairline management that were both unwarranted and exaggerated. Depending on yourpoint of view, either a mentally unstable pilot was rightfully removed fromflight duty, or an innocent-but-eccentric individual was vilified by a fewtreacherous fellow crewmembers.

Alcohol abuse is treatable and a full recovery to normal sobriety ispossible. However, for more insidious mental disorders such as depression due tochemical imbalances in the brain, a pilot may have no hope of maintaininghis or her employment if he or she seeks help. That person is stuck between arock and a hard place.

AVweb’s survey results

A recent informal survey conducted by AVwebasked and received responses for the following three questions:

  1. Have you ever had a medical condition for which you chose not to seek treatment for fear that disclosure might jeopardize your flying? (567 responses)
    • YES — 46%
    • NO — 54%
  2. Have you ever had a medical condition for which you sought medical treatment, but then failed to disclose it on your FAA medical application for fear that disclosure might jeopardize your flying? (563 responses)
    • YES — 32%
    • NO — 68%
  3. Do you take medication about which you have not told the FAA? (561 responses)
    • YES — 21%
    • NO — 79%

The responses to these questions clearly indicate that a significant numberof pilots would rather self-medicate or try and work through their problemswithout professional medical assistance, rather than risk losing their medical.

These statistics may actually understate the problem. Feedback to AVwebon the survey questions indicates that a number of pilots did not respond tothe survey for fear the FAA would somehow find out their identity.Therefore, the statistics above may actually underestimate thenumber of pilots who are reticent to seek medical treatment for fear of incitingthe wrath of the FAA.

Aggravating factors

JAL B-747Asa crewmember on long-haul international flights, it’s my opinion that duty cyclepatterns that cause a pilot to be isolated from his or her family and home lifefor extended periods of time can only exacerbate depression. We all know whathappens: The day after you walk out the door to go to work, your child has anasthma attack and has to be hospitalized or the plumbing in the attic leaks andfloods the house. Whatever the catastrophe, it’s sure to occur when you are awayfrom home. A depressed pilot may feel responsible for these family crises andblame him or herself.

Cumulative fatigue caused by chronic circadian rhythm disruption and heavyworkloads imposed by the airline duty roster can weigh heavily on anypilot’s shoulders. If a depressed pilot is already finding it difficult tosleep, then the weight added to his or her fatigue by flying the line may beenough to push him or her over the edge. To top it off, a depressed pilot’sfeeling of being unable to seek medical help can only serve to deepen his or hersense of hopelessness.

Each of these influencing factors can be alleviated to some degree if airlinemanagement and pilots’ unions work together. However, if the FAA were to changeits policy and allow pilots to resume normal functionality with the use of safeand effective drugs, then the potential of pilot suicide may be alleviated.

No easy answers

What a kettle of worms! Statistically many pilots will likely suffer fromdepression at some point in their career, but may not seek treatment out of fearof losing their job. Experts say that it is quite possible for a pilot to maskthe seriousness of his or her condition. Even if a pilot recognizes his or hercondition and desires to seek medical assistance, treatments such as Prozac maynot be available without throwing away his or her career. If the airline or theFAA discovers that a pilot has sought medical assistance for depression and/orwas prescribed medication, then there is a good possibility that the pilot willbe suspended from flight duty. The FAA is charged with maintaining aviationsafety, but by its lack of latitude in allowing a pilot to fly onantidepressants, it is arguably eroding safety rather than enhancing it. The FAAsays it’s looking at the problem, but in reality the agency is probably scaredto death of the political repercussions if it were to relax its absoluteprohibition on psychotropic medications

Airline management, having become more aware that such things as pilotsuicide exist in our modern world, have to tighten their scrutiny ofpilots’ mental fitness while maintaining an empathetic attitude towards theaffected pilots’ careers. Airline unions would likely resist any exhaustivepsychological testing of their members each year as such tests could beused to justify airline managers who selectively dismiss troublesome employees.However, a balance of what is good for the pilots and what is good for thesafety of all who fly with them must be met. Airline management can actto alleviate the fatigue factor caused by long duty-days and long patterns, andshould do so immediately.

One airline with which I am acquainted includes a psychological test as partof a crewmember’s annual physical. However, the test is hardly comprehensive –the questions asked by the medical examiner are general in nature and easilydeflected by the crewmember being examined. As far as this test isconcerned, it seems as if the airline is content with making a token effort atensuring each of its crewmembers is mentally fit. Having said that, at least itis making some effort to assess crewmembers’ mental alacrity andwell-being.

The sad truth is that it may be up to us — pilots and flight engineers — tobring other crewmembers’ aberrant behavior to the attention of airlinemanagement. As much as we all hate to tattle on our peers, such action might benecessary in the interest of safety. But what constitutes aberrant behavior?Should a pilot be reported because he or she is depressed over an impendingdivorce? How do you protect an eccentric individual from being wrongfullyaccused? Who is to set these criteria? I, for one, would be reluctant to treadthis path unless a pilot had blatantly put his or her crew and passengers inimminent danger … but by then, it may be too late.

Even though evidence exists that some pilots have functioned normally on aday-to-day basis while undergoing severe depression, it seems highly unlikely tome that someone, somewhere, was not privy to the affected person’smental state. In each of the airline pilot suicide cases mentioned at thebeginning of this article, the pilot’s family refused — out of a sense ofloyalty, or out of denial — to believe their loved one would commit such anact. However, if members of the pilot’s family were truly honest withthemselves, maybe they would recall evidence of their loved one’s illness. Thoseclosest to the depressed pilot are the most likely to identify the problem.However, family members and close friends are also the least likely to turn thepilot in (ex-wives and ex-husbands excepted).

There are no easy answers. At the very least, I would suggest that acooperative study be conducted by pilot unions, airlines, the FAA andappropriate medical authorities to determine what can be done to assist pilotswho suffer from serious depression without jeopardizing their employment.Anecdotal and clinical evidence exists that prove a person can return to normalfunction after being treated with antidepressants and waiting for any associatedside effects of the drugs to dissipate. In the hectic pace of modern life wheremore and more people are becoming disassociated from each other and there neverseems to be enough hours in the day to accomplish all that we need to do, theprevalence of depression is likely to increase. Somehow, we must all worktogether to ensure that those who need help for depression can get it withoutjeopardizing their career.

The wheels turn slowly at the FAA, but it is high time that pilots werereclassified as normal human beings who suffer from everyday, treatable ailmentsthat can be controlled effectively by the use of medication. The alternativecould be murder.