Depression didn’t crash the plane: the co-pilot did

BY EIKO FRIED. In response to the Germanwings crash, media and politicians have construed depression as cause for the tragedy, and depressed individuals as hidden danger we need to protect ourselves from. This stigmatization is inappropriate and harmful—what we really need is a better understanding of a severe and very common disease.

Written by Eiko Fried, a psychologist specializing in research on depression. He tweets via @EikoFried.

Eiko Fried

On March 24rd, the Germanwings flight 9525 crashed in the French Alps, and 150 people lost their lives. It has been 3 weeks since, and we have had some time to reflect upon what happened. The catastrophe received unparalleled attention, especially in France and Germany, and the media created a vast amount of hypotheses and pseudo-news to fill hundreds of hours of live coverage. Most of this coverage happened in the first week, based on very little factual information.

While the desire to understand what happened, and why it happened, is understandable, wild speculations about the possible mental state of the co-pilot, private details about his life and his family, and pictures of grieving relatives and friends of the victims did certainly not provide clarity.

Responses to the tragedy are based on misconceptions of depression

What we know now is that the co-pilot had been diagnosed with severe depression in 2009, and that investigators discovered a letter in his apartment signed by a doctor declaring him unfit to work. Depression has been construed as cause for the crash, and it has been suggested that individuals with mental disorders should be forced to disclose their current diagnoses and disease history to their employer. There have also been requests that people with a depression diagnosis should be banned from working. These calls are hasty and based on four depression myths I would like to correct.

First, very few suicides committed by individuals diagnosed with depression cause physical harm to other people, and it is a notable exception when another person is killed. In the rare cases this happens, the victims are usually close relatives whom the depressed person does not want to leave behind. The plane crash does not resemble this pattern at all, and should in turn be interpreted as a very rare exception to a very rare exception.

About 1 in 5 individuals living today will, at some point in their lives, be diagnosed with depression.

Second, depression is the most common mental disorder in the world, and about 1 in 5 individuals living today will, at some point in their lives, be diagnosed with it. It is not a very picky disorder that only befalls people of a specific disposition, skin color, religion, or profession.

That a pilot was diagnosed with depression in 2009 is not unlikely, because so many people suffer from depression––doctors, bus drivers, athletes, and hairdressers. About half of all people diagnosed with depression will have multiple episodes that can last from weeks to months, whereas the other half will only exhibit a single episode in their lives. Therefore, a diagnosis of depression six years ago does not necessarily tell us much about how the person is doing today.

Third, there are many different reasons why people become depressed: adverse life events, biological predispositions, chronic stress, and early traumatic experiences, to name but a few. In most cases, several risk factors come together. Depression also has many different faces: some people feel strong negative emotions such as sadness and anger, while others are better characterized by an absence of positive emotions. Some depressed patients sleep very little and lose weight, others sleep 14 or more hours a day and eat more than usual. It is not surprising that depression is acknowledged as one of the most diverse and complex diseases in psychology and psychiatry.

Many articles called out his depression as cause for the crash, but depression didn’t crash the plane: the co-pilot did.

This implies that knowing about the co-pilot’s previous diagnosis is not very informative in the face of the variability of the disorder; in fact, the diagnosis tells us incredibly little about the specific problems he may have suffered from, and his possible motives for crashing the plane. Many articles called out his depression as cause for the crash, but depression didn’t crash the plane: the co-pilot did. And he did so because of very specific problems, in combination with a very specific personality profile, very specific life circumstances, and a very specific personal life history.

Finally, depression is a real problem deserving of a lot more than a knee-jerk reaction by politicians and press. It’s much more than just being utterly sad––it causes as much disability as other chronic medical conditions and severely compromises the capacity for self-care and independent living. This means that people suffering from depression, who often also have other conditions like anxiety disorders, need our attention and support just as much as people with cancer, HIV, or Alzheimer’s disease. Just because there is the word mental in mental disorder doesn’t make depression any less real, or any less a disability. This problem needs real attention and real solutions—not reactive measures that will hurt in the long run.

Stigmatization and violation of privacy rights do more harm than good

In addition to the fact that many responses to the tragedy have been based on wrong assumptions, they also seem inappropriate.

Putting a large group of people under general suspicion because of rare memorable situations like this plane crash is unjustified.

Demanding that patients should reveal previous diagnoses to employers is not only calling for the severe violation of privacy rights and medical confidentiality of millions of people—it is also highly unfitting, especially in the face of this one incident. People with mental disorders are much more often victims than perpetrators of crimes and deserve, if anything, special protection.

Putting a large group of people under general suspicion because of rare memorable situations like this plane crash is unjustified. Another major problem is that less than half of all people suffering from depression will seek help, often because they are afraid of disadvantages that could result from their diagnosis. The recent calls to force patients to disclose medical information to their employers foster the stigmatization of individuals with psychological problems, and will further decrease the willingness to look for help.

In sum, a terrible catastrophe happened, but we should carefully consider our options before we make rash decisions. People with mental disorders are as much individuals as people without, and there are large differences. Somebody with a specific phobia––for instance a fear of spiders or dogs––may be well suited to fly an airplane, whereas a patient suffering from sleep problems and severe fatigue after a traumatic experience may not.

And this variability exists not only across diagnoses, but also within: some individuals with depression may be able to do their jobs very well, depending on the particular symptoms, whereas others may not be able to do so. We need individualized solutions for individual people.



  1. If depression has such a wide range of symptoms, is it possible that it’s not just a single disorder, but actually a number of different, but related mental disorders (or “brain malfunction”; when people mistake mental disorders with something that’s not a bodily disease, we should help them: brain malfunctions are just as bodily as heart or stomach conditions) under one, somewhat misleading, single name?

    • Bernd: I’m going to try and answer this, even though I’m not a medical professional. If you look at diagnostic manuals which “define” things like depression, there’s usually a lengthy introduction (e.g. in the ICD-10, the standard diagnostic manual used in large parts of the world) that explains that these categories (depression etc.) are diagnostic tools: they group together similar issues so that it becomes easier to find appropriate classes of treatments. They’re “clusters”, to use a term from statistics.

      The goal of diagnosing isn’t to find exact boundaries — there aren’t really any definite ones, anyway –; it’s to reduce complexity, to find enough common ground between various cases to be able to draw on prior experience. Of course every single case is still unique — this is true for pretty much any psychological issue. Any decent professional helping with such problems will keep this in mind even when using a rather generic diagnosis.

      (Can there be value in subdividing the diagnosis? Absolutely, but only if someone finds treatments that work better for one sub-diagnosis than for others, for example; when that happens, the new diagnosis allows us to draw on a narrower set of prior experience so that we can hopefully find the right treatment more quickly. Additionally, there can be value in subdividing the diagnosis to open up more avenues for research, but in practice, from what I gather, it’s a huge challenge to actually do research on that kind of scale.)

      Also, on the topic of mental disorders: whether they are bodily or not is really very much debatable. It’s true that many pervasive psychological problems have observable effects on physiological processes, but that doesn’t mean the physiological processes are the “cause” or the “heart” of the issue. Crude analogy: suppose you have a modern mobile phone that gets hot all the time and loses battery charge really fast. These are physical, observable effects, but there are many different potential causes. One of them is “software issues”: maybe an app on your phone uses a lot of computing power even while the phone looks like it’s in standby. This problem isn’t best described as an electrical or electronics problem, even though it *causes* electrical problems (high power consumption). The most effective level on which to look at the issue, if you want to actually solve it, is that the *software* needs to be investigated. If you focus on the electrical side of things, you can only treat the symptoms (by installing a bigger battery, applying more cooling, …) but the underlying problem will still exist.

      If you consider the same kind of structure in psychiatric/psychological issues, it seems very practical to try and distinguish between issues that are *caused* by physiological changes, and issues that are only *accompanied* by physiological changes. This is probably very difficult in a significant number of cases; to the best of my knowledge, for many diagnoses in the diagnostic manuals there are no objective physiological tests, only more or less subjective “soft” criteria. Feel free to read the DSM-V or ICD-10 (section F) yourself to see what kind of criteria are commonly used for diagnosing…

      Of course, just like the article emphasizes, even if something doesn’t have a physiological cause, it can still be very serious, create very complicated problems in someone’s life, and there’s no still need to blame someone a la “what’s the big deal, it’s only in your mind”, so I agree with you in the sense that people who downplay the potential impact of psychological issues are, at best, misguided.

    • Bernd, you hit the nail on the head with your question. I agree with Jan’s very insightful response, but would like to elaborate on three points a bit more.

      First, many people (both lay man and medical professionals) believe that mental disorders are natural kinds––that is, things that exist in the universe objectively outside of human classification systems––similar to, for instance, atoms. The atom gold always has the same number of protons and neutrons, and anything with that number of protons and neutrons is the atom gold, no matter if we humans observe or discover or not. And this thinking about mental disorders has caused simplistic questions such as “what causes depression”, when depression is in fact a large number of different problems with different symptoms and different etiologies (pathways into the disease). The question “what causes depression”, or “which genes cause depression”, are not very good questions, because depression is not one thing. Acknowledging this variability, and that fact that depression is a broad category that encompasses a large number of very different problems, will really help to understand our current challenges better.

      Second, depression can indeed be seen as a number of related but different disorders (I would call it problems instead of disorders though). We recently examined 4000 people diagnosed with depression in a study, and found over 1000 unique symptom profiles. That means that, on average, about 4 people each had the same symptoms, and that these 4 people had different symptoms than every else in the large group of people. And keep in mind: they all had exactly the same diagnosis.

      Third, Jan rightfully hints at causation. There can be no question that all processes such as sadness or fatigue are *accompanied* by changes in your brain. It’s how the human body works. But when you see disorders framed as “brain disorders” in science and media, it is often often implied that the *cause* for the disorder lies in the brain (I’m not an expert here, but I think Alzheimer’s disease and Parkinson would be good examples for brain changes that then, downstream, afterwards affect other levels such as the symptoms). What we’ve been showing in our work however is that this causal model doesn’t work well for depression: depression symptoms like sadness, fatigue, insomnia and so forth are not caused by one brain failure that is the shared origin for all symptoms (I wish it were the case, because then we could fix that brain failure and all symptoms would disappear!), but by the fact that symptoms trigger each other in complex dynamic network systems (insomnia -> fatigue -> concentration problems -> psychomotor problems -> etc.).

      You can find a summary of these three points––and actually also most of the points Jan so elaborately made clear––in a very recent overview paper on the topic. The paper is freely accessible to anyone, and written in a way that also non-scientists can understand most of it.

  2. I am interested in understanding better ways to integrate mental health into our everyday lives, because right now at least in the US it seems to be very private and not understood in a wholistic way. For instance, how do we as a society support parents with depression? Children of parents with depression? How does BPD affect homeownership, and do we want to make sure people have support they need to own a home? To what extent is anxiety useful or not in particular workplaces and can we figure out ways of allowing people to keep jobs by working with their mental illness? If people are working, I also want to think more about ways that concretely it makes sense or doesn’t to have psych eval as part of taking on particular kinds of work — for everyone, not for particular people who have a “history” they must “disclose.” When/where is it important to know people’s psychological state outside of how they interact on a day-to-day basis with others? What would be considered inappropriate for which kinds of work? Would a wide-ranging and thorough psych eval actually help to prevent things like abuse of children by clergy or teachers? What could it tell us about our elected officials? Would it make us more vulnerable as workers? It seems like it would be good if it helped us to know how to work with each other better, but if it could be used to disqualify people, it could be a terribly effective avenue to formally class our society based on mental health, or to enact bias or personal dislike during the hiring process. Moreover, proper psych evals are supposed to by definition reveal things that we wouldn’t be able to/want to reveal ourselves if asked directly. Which means we’d be consenting, without exactly knowing what a test might reveal. It’s just a complicated, interesting idea. It makes me think a lot about how “free” we are and how separate our private selves ought to be from our public selves.

  3. Dit is op Maarten is benieuwd herblogden reageerde:
    Deze ‘KU Leuven blogt’ ontkracht enkele mythes die bestaan rond depressie en die sterk zijn uitgespeeld in de berichtgeving over de Germanwings-crash in de Franse Alpen. Wat mij betreft verplichte lectuur voor elk debat dat ver of nabij te maken heeft met depressie.

Reacties zijn gesloten.