Nothing to do with the content of your essay, but I was initially confused by the title of the magazine "Works in Progress". I was thinking that, like your recent "Peer Review Request", you were posting an actual work in progress and was asking for feedback on it before it is finalized, and the Works in Progress website was just a space for authors to release drafts of their essays. However, the About page for Works in Progress makes it clear that publication venue for complete pieces. Sharing this in case other readers were confused in the same way.
A common theme that I find in a lot of analyses is that factors that seem obviously important fall apart on a closer look. You do that with multiple factors here. I also get the same feeling when I listen to historical podcasts by Mike Duncan, like the Revolutions podcast on the French Revolution. Factors that seem important and fundamentally causal in an almost common-sensical way almost always fall apart on closer inspection.
This sometimes leads me to suspect that causality is mostly a mirage in the social sciences, and that any and every factor will fall apart on a "really" close look. Events happen not because of any overarching causal patterns, but only because of sheer dumb chance. This lack of any causal relations has also been stretched to science by Stephen Wolfram in his recent article- https://writings.stephenwolfram.com/2021/03/what-is-consciousness-some-new-perspectives-from-our-physics-project/
Wolfram says that any causal factors that we may find for events are only a result of our limited cognition, and not an inherent description of the universe itself. Although I am apprehensive about stretching the metaphor to that extent, do you also agree that any kind of causality, at least in the social sciences, is mostly illusory?
Sorry, but how can someone research suicide and not have "G-d", "morality", "religion", "value system" come into play? I am not saying that those thing would explain everything, of course not - but excluding them from the picture is beyond pale... Yes, many had a chance to reexamine their convictions and their views on acceptability of suicide.
> Could it be that people have trouble getting the means to commit suicide? For example, maybe they would like to shoot themselves, but all the gun shops are closed?
You already reject this argument in the essay, but another bit of relevant evidence here is that gun sales shot to record highs in 2020, so it doesn't seem like suicidal people would have had much difficulty buying a gun last year.
"When COVID started spreading, life got more depressing, people became more depressed, but suicide rates went down. Why?
"First, are we sure all of that is true? I won’t waste your time listing the evidence that life got more depressing, but what about the other two?"
Maybe you should waste our time with point one. At least in my case, after about the first month when I became sure I'd still be able to get groceries (more or less) without issue, my life took a profoundly positive turn. Since the pandemic started, I finally bought my first house, got a dog, and realized my career-long dream of working from home. I know many people have had great emotional struggles during this time, but how many others like myself have had the opposite happen? Is it a large enough group to be meaningful?
Since my employer announce their return-to-office plan, I've noticed a steep rise in my anxiety levels—and not from fear of COVID, against which I'm now vaccinated.
"I won’t waste your time listing the evidence that life got more depressing" is too dismissive - in exactly what way did things become more depressing and what relationship would you expect these individual factors to have to suicide?
I'm not sure "scoring high on a depression index" is the same as actually being depressed. I went and took the depression questionnaire used by Ettman et al. - the Patient Health Questionnaire–9 - and it would have reported me as depressed even when I definitely wasn't depressed. Back when covid first got to my country of residence, I experienced some anxiety which resulted in taking what I called "anxiety naps" where it felt like I would just shut down and sleep for an extra four hours most every day after work. I was also stress-eating a lot more than I usually did. I also had trouble concentrating on things - my mind would keep drifting back to the horror stories I'd read from hospitals in countries where the pandemic was hitting the hardest. All of these were specific responses to stress and anxiety over the threat of coronavirus - related, probably, to my history of respiratory illness and my specific fears about how my young children would fare if I died. At no point in the pandemic did I consider myself "depressed", and over time I built up better coping mechanisms.
But of course the nature of those specific problems - extreme but somewhat rational fear of dying horribly and leaving my kids without a father - contraindicates suicide. Put another way, while fear of death might lead you to sleep more, or overeat, or lose focus on your work, or fail to take joy in things that you otherwise would have enjoyed, you wouldn't expect someone to commit suicide out of a fear of dying.
In fact it might be worth looking at whether fear of death decreases suicide in general - do people who live more dangerous lives kill themselves with less frequency? Or something like that.
Also, I wonder if being stuck inside with your family members brought more people closer together and reduced the kind of isolation that sometimes precipitates suicide.
I'm particularly sympathetic to the "change of pace" line of thinking. On Twitter @PaulSkallas always talks about how people desire a certain amount of volatility in their lives. They get bored going to work every day, the same routine, following a traditional life trajectory, etc...sometimes they secretly hope for something radical to happen, even if it's objectively "bad" for them. I've certainly felt this way myself at times. It similar to a joke made by one of the hosts of the Red Scare podcast. "Don't kill yourself because something retarded might happen and you'd miss it". Obviously not a super polite articulation of the point. But I think the insight is the same as above.
The pandemic was something new. What if it's the apocalypse? What if society emerges from the crisis totally remade? Maybe you're depressed in your email-office job, but you'll be happy in your new post-apocalyptic role as a community garden tender. Maybe UFOs show up tomorrow and show us the wonders of the universe. Gee, maybe suicide can wait for another day, at least until the conclusion of the most recent species-wide narrative arc. You never know!
I wonder if part of the decrease is because people are less likely to commit suicide when another one of their family members is physically present in the same home. This doesn't explain the black/white divergence though.
On the black/white divergence: I wonder it the reported increase in gun sales during the pandemic varied by race?
> Second, one pillar of Aaron Beck’s triad of depression is “this is endless.” A sense that nothing can ever change, so why bother to wait and see? The pandemic was the opposite of that.
This is strongly, strongly, strongly inconsistent with the lived experiences of literally every person I know. I cannot think of a single person who was like "yeah this is totally temporary it's only two weeks then it's back to normal".
One thing I noticed was this line "Second, the US has a waiting period on guns, so most gun suicides are by people who already owned the gun when they became suicidal.". While some places do have waiting periods, some don't. For example my state has no waiting period, I walk in wanting a gun, I walk out with a gun.
Besides that one question I would have is around the kind of people who would be more prone to committing suicide. I mean if you are depressed and everything shuts down, does that impact you that much? If you were depressed were you going out to bars partying and visiting family? Maybe now that everyone is a shut in you feel some sort of kinship, "Hey welcome to my world normies, enjoy the ride". All the newly depressed people know it is temporary and rally around beating covid.
I always go into this with the thought: "yeah, but I wonder if he's thought of THIS!", and I always find out you have. I have to appreciate how thoroughly you consider things. Here are the two ideas that I thought of, and my thoughts on your points:
1. Overdoes vs. suicide. I've had one sad personal experience with a young man where the question as to whether this was an overdose or a suicide was unclear. My feeling was that we all agreed it was an overdose out of kindness for the surviving family, and maybe to lessen the shock. I wonder if, in this pandemic, there was either less investigation of these deaths happening, or whether there was just an unconscious nudge that way.
2. Meaning. I have never seriously considered suicide, but when I'm at my most dark points, it's because of the bleak expanse of unchanging, bland, dreary life ahead. The pandemic was at least interesting and engaging in a lot of ways, as grim as that is. I think people are itching to fight for something meaningful, or work for something meaningful. Most of our careers aren't very meaningful: for example, my job is helping big companies track how you behave online so they can market to you more effectively. Kind of depressing. So you've got this pandemic, and we've got the kind of existential moment, and we can all kind of "fight" for something.
At least at first. Then it starts to get political, with your mask adherence becoming a tribal thing, and everyone back to the same routines, but with social distancing. So there's that initial interesting time of the crisis (that Zunin and Myers graphic is awesome), but then it's back to the usual slog of meaninglessness. And it might be that we come out into a weird economic period that doesn't rebound into "reconstruction" the way a different crisis might, because the crisis of the pandemic was so odd, and we've already kind of been "reconstructing" through the whole thing, and now we'll have the bill to pay for the unemployment and supply chain problems and kids having just terrible school years, so they're practically a year behind. . .
Anyway, the Durkheim/"meaning" thing seems most likely to me based on my own observations.
as a person with medium to severe depression before the pandemic and very severe depression including suicidal ideation during the pandemic, I would be a data point for your last two hypotheses - the change of pace and making the whole shebang feel at least a bit different (before it turned into endless hopelessness again) + not ever being alone to actually do anything. and the reason I would more likely feel inclined to kill myself in April-ish is easy: between October and March you can (in the northern hemisphere) blame it on winter and tell yourself that it'll get better in spring, because sunshine. Then spring comes around, and it doesn't actually get better... then it definitely must be forever this bad, so what's the point. At least that's always been my reasoning. Hooray for anti depressant medication.
I remember when it first started and millions were losing there jobs my firm did a few layoffs. Ordinarily that would make me anxious and depressed. But somehow the thought that millions of other facing the same thing made it much easier to deal with. Maybe part of suicidal depression is the idea that bad things are happening just to you. You've been singled out.
On the other side of it, having a job, when everyone else has a job, doesn't feel amazing. But having a job when millions are losing theirs makes it feel special and like a real accomplishment.
Did suicide attempts fall too? Are people just less successful since there is a greater proximity to people at home?. I also wonder if people who couldnt handle the situation might have just decided to ignore the pandemic as a way of coping instead of Suicidal behavior.
> So it’s weird to see drug overdose deaths follow such a different trend from suicide. I’m tempted to wonder if people are misclassifying suicides as overdoses, but I don’t see any evidence of this or any reason why it would happen.
One theory: we always misclassify some drug-related suicides as overdoses, and the pandemic caused some suicidal people to shift their suicide method to drug-related, and as a result we increased the number of suicides misclassified as overdoses.
We could check this theory by checking if the increase in overdose deaths cancels out the decrease in suicides.
"But most people stayed home even in places without legally mandated lockdowns"
-Citation needed.
One observation of possible relevance to apparent difference in the black/white suicide directionality gap: I live in a majority-black neighborhood, and I found that for the first several months of the pandemic, almost *nobody* in my neighborhood observed basic safety measures outside the home. People would wear masks in stores once it became legally mandatory, but outdoors, only a fraction of people wore masks, and only a fraction of those people wore them properly covering their noses and mouths. I constantly saw people approaching each other within distances of 2 feet or so (this is a small personal space bubble neighborhood) and taking their masks off to talk to each other, often touching each other on the shoulders, sometimes even each other's faces in the process.
I noticed practically no difference in terms of outdoor socialization during the pandemic versus pre-pandemic, until around last Winter, after the election. People actually wear masks outdoors now, and wear them properly, and fewer people socialize at physical contact distance, but it took a *long* time for this change to take.
"Also, this explanation implies that people were financially better off with the coronavirus and relief package than they were with neither, which doesn’t really seem to match reality."
Why not? CARES was uncharacteristically generous in that many people actually did end up better off financially. Someone who worked a shitty service job slinging popcorn at a movie theater for minimum wage and got laid off by their employer probably wasn't that invested in concessions as a career and would've been quite happy to get more money for less effort.
"Third, maybe people became more worried about the effect that suicide attempts would have on their family. Part of this is increasing social closeness—people suddenly had to spend every waking moment with family members, or else bubbled together with people they had previously lived apart from. Another part is that their family members were also suffering, and depressed people decided it was unfair to ask them to bear the burden of a family member’s suicide at the same time as everything else. Or what if you just literally don’t have the personal space, out of earshot of your family, to attempt suicide without someone else noticing?"
This was also my theory upon first hearing that suicides went down during the pandemic, and before I looked at further data. More specifically, I predicted:
(a) that the drop in suicides was due to fewer suicides among young people, who were newly forced back into their parents' home
(b) that deaths of despair among adults had increased.
I don't really know how to verify my predictions -- is there an age-breakdown of the suicides anywhere? I consider the increase in overdoses to be weak confirmation of (b) for now.
When you use phrases like “studies from Norway, England, Germany, Sweden, and New Zealand” it is unclear what this means. My best guess just, like, grammatically, would be studies from research institutions in those countries; whereas my best guess based on what information would actually be relevant to the reader is studies whose study populations were from those countries. Maybe it happens to be both simultaneously for these examples.
Anecdotally, my therapist friend had the exact same observation of anxiety patients. All of a sudden, it's like they were in their element, or like their life outlook had suddenly been vindicated.
Re drug overdoses: there are two plausible explanations I've heard:
1. More people doing drugs alone, with no one to call an ambulance or administer Naloxone
2. Erratic supply causes people to buy drugs in bulk, which leads to overdose.
It should actually be easy to figure out if 1. is true, just based on the details of individual overdoses.
I think the "two types of depression" is on the correct track. There's the 'normal' type of depression, which is the depression in the other post. That's not going to change very much during Covid-19 and may even, weirdly, become slightly better. Those are the people who, if they're going to kill themselves, are going to kill themselves whether or which.
The second type is "kinda down, kinda blue, kinda not feeling good" which happens when people not suffering from the first type of depression have their usual "go out and enjoy myself" activities curtailed, along with worries about the pandemic, maybe laid off from their job, etc. Ordinarily they're not thinking about killing themselves and things have to get really, really bad to push them to that stage. When the restrictions are limited, life goes back to normal, they'll bounce back.
What I mean for the first set of people, those who are suffering from depression before, "weirdly become slightly better" is that now *everyone* is in the same boat. I'm introverted, I don't like people, and I think I might be mildly agoraphobic. That means currently I *love* the lockdown conditions, because this is normal life for me *plus* bonus "working from home? no interaction with other people? can slob around in my night clothes all the day? oh yeah baby!"
Also having to wear a mask is great when you have (a) plain to ugly features and (b) really can't be bother slapping on makeup. When three-quarters of your face is covered between mask and glasses, it's really liberating.
Nobody can travel, nobody can socialise, everyone is confined to the same restrictions. This is awful for people who have social lives and like going out and mixing in the world. For depressed people who already have the tendency to stay inside, avoid people, and have lost interest in formerly pleasurable activities, suddenly this state of affairs is the new normal and now they're virtuous for not doing any of the "just go out, have fun, meet people, force yourself to interact" activities. This is a sort of relief of pressure, funnily enough.
Drug use (and as a result of that overdoses) going up? Yeah, I'd expect that too, the same way alcohol consumption goes up. People can't go out and do their normal activities, they get bored with being confined indoors with TV and games, they consume their usual fun substances to cheer themselves up, but now they're doing more of them, regularly, and possibly upping doses as well. They don't intend to kill themselves, it's not suicide, it's just a bit too much of your usual fun substance/mixing several recreational substances together and then - oops.
My first thought was to look at suicide during war. You gave a link saying that suicide in Britain went up during WW2, but the linked article only said that suicides in Scotland declined less during the war years than one would predict given a longer-term data sample.
I say, look for more data on this. "Suicides went down less than predicted" probably depends critically on what window you use to set your "background effect". Also, the Scots weren't uniting against a threat so much as they were fulfilling military obligations imposed on them by a foreign nation (England) which many of them (like, maybe half) saw as their oppressors.
I applaud and appreciate your methodical refusal to come to any conclusion. Measured rationalism, while less than satisfying as a narrative, is a refreshing approach to potentially torture-prone statistics. It’s like the rhetorical equivalent to publishing negative studies to insure against publication bias. Thanks for sharing your still-baking thought process. The world needs more out-loud examples of such things.
One of my kids (M 20) is a front line worker and was therefore not locked down. He has bipolar disorder, which is managed with Aripiprazole and psychotherapy. He experienced an episode of suicidal ideation over the winter that seemed to come from nowhere. He explained to his therapist (and me) that what was depressing him was his failure to progress at work, the sense of "same shit, different day" and that he didn't see a way to get back to being a college student (which he currently can't handle). So, N = 1 but I go along with the "hopelessness" theory.
Overall, very well written and helpful - as always.
Scott you're at your strongest when you discuss about stuff related to your job and I wish you'd continue doing it instead of reviewing books or hinting about HBD
Since most of the death statistics are going to be based on CDC death certificate data, I suspect that who completes the death certificate is somehow a factor. Death certificates of violent suicides or completed overdoses found in the field with an accompanying note are going to be completed by the local coroner or an emergency personnel trained in and certified in death pronouncement. Most overdoses of the kind you describe here, "There’s a thin line between a drug overdose death and a suicide. Something really terrible happens, you can’t stand the thought of dealing with it, so you take ten times your normal dose of painkiller in order to fall into a warm, pleasant sleep. At some level of misery, “but what if you take too much and never wake up?” stops feeling like a flaw in this plan. For some people, it’s a bonus—all the oblivion of suicide, with none of the guilt—after all, you weren’t trying to kill yourself, not exactly.", they will be managed in an ICU somewhere by a critical care physician, and unless there is an accompanying clear suicide note found at the scene that was transported with the patient and makes its way into the ICU and the patient's chart, those deaths are not likely to be classified as suicide at all and will instead be accidental overdoses with the death certificate being completed by the attending physician at the time of the patient's death. I have taken care of many of such ICU overdose admits and completed many death certificates concerning those cases and have rarely indicated suicide on the death certificate. It is very important to be accurate when completing death certificates because they do have life insurance implications etc. It does not surprise me that overdoses increased during those early months of lock downs, and I suspect that a fair number of those overdoses were also suicides of the type you describe.
Hypothesis on drug overdoses: with borders largely closed, it's getting more difficult to get certain foreign-made drugs, so addicts are switching to other even worse drugs; e.g. if you can't get heroin you try fentanyl.
Did I misunderstand something, or does the CDC say that 2020 10.7% of Americans considered suicide? This number seems absurdly large... Or does "thinking about suicide" mean something different?
I notice you are not too fond of the Durkheimian "during a crisis people seek together" hypothesis. A reflection here:
A test of Durkheim's theory should probably use something else than suicide rates as an operationalisation/indicator. Since (if you follow Durkheim) there are different suicide motivations, and not all of them are of the type that is influenced by "getting closer together/becoming more distant from each other".
...A better operational indicator to test the Durkheimian thesis is probably to check the rates of marriage/co-habitation break-up. The Durkheimian thesis straightforwardly leads to the assumption that the breakup rate will go down during an epidemic/disaster.
...I have not checked if that has been the case during the present pandemic (worth doing!), but here is an old article finding this effect after the 1995 Oklahoma bombing: Nakonezny et. al. (2004): Did divorces decline after the Oklahoma city bombing? Journal of Marriage and the Family vol 66 (1) p. 90-100.
Ok, I know that checking if Durkheim was on to something is not what you are interested in here: you are interested in theories that can explain variation over time in suicide rates, regardless of where these theories come from. But let me offer an elaboration that might be relevant, although admittedly speculative:
There are two types of suicide that are dominant in modern societies, if one follows Durkheim: Egoistical/self-interested suicide (calculating that the costs of continued living are higher than the benefits: Gary Becker’s & Richard Posner's rational choice theory of suicide is in this mould); and anomic suicide/"existential" suicide (Seeing/finding no meaning/purpose in life; acute Weltschmerz; that Kirkegaardian/Nietszchean stimmung).
Durkheim arguably claims that the odds of committing egoistic suicide is influenced by having friends and relatives that cares about you (and hence this type of suicide should go down during a crisis that brings people closer together, similar to breakup rates); while anomic suicide results from a deeper-level cultural ennui that is not necessarily influenced by how distant/close you feel to others.
Which might explain why drug overdoses have gone up; they may be related to the feeling "life is meaningless anyway, and it feels even more meaningless these days", rather than to "I thought no-one cared about me, but in this present crisis I somehow feel that I was wrong".
Ok, perhaps egoistical versus anomic seems like a hair-splitting difference in suicide motivations. But the distinction may potentially at least explain why suicide rates and drug overdoses move in different directions during the present covid crisis. And it might perhaps also explain why meta-studies on suicide do not find consistent results supporting an assumed Durkheimian approach (since such studies can hardly distinguish between Durkheim's different types of suicide motivations, and hence different types of suicides).
….The above is perhaps a bit rambling and disorganised, but I'll post it anyway.
Great piece! Summary: Suicides declined possibly cuz a) misery loves company, b) curiosity, c) and responsibility towards others. And we don't know why drug overdoses rose when suicides declined.
Curious what’s the story behind posting this there and not on your own blog? Is this a business/partnership/commeercial arrangement thing? (i.e. you’re posting it there it helps giving their website more exposure?)
Dostoyevski's notion of finding "happiness in the midst of despair" is distinct from "people pull together during a crisis" although at times it looks similar. Under Dostoyevski's formulation it's the crisis itself -- which may be personal -- which, although causing one to feel bad, also causes one to feel like they are *supposed* to feel bad. If feeling bad at least makes sense it's easier to tolerate.
I think of this idea as similar to Adam Smith's observation that people suffer from small embarrassments more than big ones.
Random thought that I have no emotional attachment to...a lot of people who commit suicide are "Zoom class" or "affluenza" types whose lives actually got better during this period since they still made money, got to stay at home, and could actually do all the stuff they normally want to do?
The increase in overdoses might not necessarily reflect an increase in suicidality/indifference to death. I think hard drug use in a social setting might actually be safer than solitary hard drug use. This would be due to having help on hand to administer antidotes or call an ambulance if someone takes too much. I think it could also be due to social pressures leading to people using less drug overall. The two biggest ones would be avoiding the shame/stigma of taking a way higher dose than anyone else and wanting to remain lucid enough to socialize.
I don't use opiates and I don't know anyone that does, but I do see the same pattern of people using lower doses in social settings and higher doses alone across all sorts of other types of alcohol and drug users.
"Also, there were slightly more black suicides in the months just before and after the crisis, so that if you look at total black suicides during 2020, it’s about the same as any other year—in fact, lower than two years ago."
This passage was opaque to me. Did you mean slightly *fewer* black suicides? There are two small bumps on the chart that might be the ones you're referring to, but I don't see how slightly more suicides before and after the crisis, added onto high suicides during the crisis, could lead to the year being normal in total.
> Also, [the 'increased welfare' explanation] implies that people were financially better off with the coronavirus and relief package than they were with neither, which doesn’t really seem to match reality.
For what it's worth, real personal income per person shot up sharply during the pandemic: https://fred.stlouisfed.org/graph/?g=EiOd (sorry for weird units, don't know how to fix that). The worst month throughout the entire pandemic on that metric, March 2020, was only as bad as July 2019, at around ≈$51,000. April 2020 it shot up to ≈$58,000 (going by the trendline, that wouldn't have happened for ≈5 years by default) and hasn't come back down to pre-pandemic levels since. This March was even more, at ≈$65000.
But this doesn't address your counterargument that suicides dropped off in March, before the stimulus.
Also, personal consumption expenditures per person went down even more drastically, from February 2020's $45,000 to March's $37,000, only fully recovering recently, and I guess that's a better measure of well-being overall: https://fred.stlouisfed.org/graph/?g=EiOr. (Though, question for econ people: does personal consumption include paying off debt? Does it depend on what you're paying off debt *for*? I'm trying to get a sense of where people's stimulus checks went if not on consumption — sitting in their bank account with everything closed, paying off debt, or what?)
Overall I'm pretty unclear on what people were going through financially during the pandemic outside of my bubble (and since everyone in the news media and in my social media bubbles is talking about zoom calls and whatnot, I can't really expect them to inform me).
My first instinct, and it might very well be wrong, is that some populations at risk for suicide (poor, old, sick) are also at increased risk for COVID, hell depression itself has effect on the immune system, so maybe the virus snatched suicide candidates before they could get around to it (same way it ate "market share" of other infectious diseases). And by snatched I do not even necessarily mean killed : like Joe Depressed is really hitting rock bottom and would kill himself, maybe tomorrow, but before he knows it he's in an intensive care unit where it's more or less impossible to commit suicide, and by the time he's discharged the worst of the crisis is behind him.
On the differential suicide rate for black people in the US, we're not in an all else being equal scenario here. There's the issue that if you are black in the US there's a dominant media narrative that you are a victim of a racist society. That's going to have an effect on people, increasing feelings of helplessness and isolation (if this makes people more conscious of their perceived differences), and there seems no reason this could not be as big an effect as COVID.
When you write "there are two types of depression" I think you mean "there are two modes of depression". The statements that follow support a change in mode for those who are already in one, whereas two types could exist simultaneously.
Best practice for spelling Māori these days either has a macron over the ā or uses two as - Maaori. Both are more accurate, which one is used depends on the tradition/preference of the local iwi (Māori traditional group).
Do you have access to international overdose data? If the suicide dip/overdose surge pattern does not appear in other countries, that would be strong evidence against the cause of death misclassification hypothesis.
Well, the last explanation certainly rings most true to my experience. The chronically mildly -- and I emphasize "mildly" -- depressed people I know perk right up when a genuine existential crisis comes along, when it's root, hog, or die. Maybe there's even some weird bit of pride in there, as in, I may well kill myself but I'm damned if I'll just let this stupid hurricane/mugger/scary skin cancer kill me. But I have zero experience with severely depressed people, and an intution is worth bupkis when assessing broad social trends.
One thing I didn't see addressed there is the precision of the "depressed" categorization. Is it possible people who were stressed and scared were lumped in?
what portion of suicide is attention seeking? if you are motivated by attention, this may have been absent during covid. i dont know how much of this is true or a myth however.
This article makes me wonder how many suicidal people died of COVID-19 before they managed to commit suicide. Something can technically reduce suicide... by killing people another way.
Excellent, thank you. Highlighting the slipperiness of statistics and their danger in the hands of most people, is so important these days. I really like your conclusions on this one.
Haven't seen what seems like the intuitive answer to me. Committing suicide generally means you are not only miserable in the present but expect to be miserable in the future. I have never considered suicide when I had food poisoning, but if I were diagnosed with "chronic food poisoning" with a 0% chance of recovery, suicide might become more compelling.
Depression often arises with the belief that one is irredeemably worthless, unlovable, etc. and suicide may appear to be the best decision in light of this distorted view. When someone's depression is clearly caused or heavily influenced by an event they perceive to be temporary, I would speculate they would be far more likely to wait and see how things go before making such a permanent decision.
In fact, I would not be surprised if this "wait and see how things go" effect even applies to people who were already depressed before the pandemic. When your life gets shaken up in drastic and unforeseen ways, there is a sense that the cards might land in your favor once the dust settles. You can't just end on a cliffhanger.
My dear friend Kieran Latty, who is an economist working on subjective well-being, has always told me that recessions are associated with a paradoxical fall in the suicide rate. If this is true it would seem to explain it (while opening up another mystery- why do recessions reduce suicide). However a brief Google turns up a number of papers alleging the opposite- that recessions in fact cause suicide- which would just compound the mystery here. No idea what the truth is.
Kieran's explanation for the purported phenomena of recessions reducing suicide was that when many people are down and out economically, being down and out economically feels less bad.
I think you gave up on the overdose explanation too early, and am confused as to why you treated it as you did. It seemed like the one place where you had concrete numbers that could provide an explanation. You move on from it and instead focus on more speculative theories. While those theories sound anywhere from plausible to accurate, they are not data-driven. Ignoring the data driven answer is out of line with my expectations of your approach (and of the rationalist sphere approach in general). I also had a previous expectation that the line between what is reported as a suicide vs an overdose is a moving target subject to much personal judgment, weak circumstantial evidence, and policy directives. While I think that prior expectation made the change in approach stand out more starkly to me, upon reflection I do still think that the approach you took is surprising.
A (quite possibly wrong) hypothesis that has enough of a chance of being right I'll register it now for gloating rights.
What if internet, video calls, and to some degree improved phone service, were the drivers of the low suicide rates.
First up, why would we expect this to be the case? Well, most of the depressed people were probably otherwise healthy people whose social lives got destroyed by COVID and quarantining. However, access to communication allowed them to maintain some degree of a support network. Eating every meal alone for a year might still be depressing, but when you feel suicidal you still have your weekly zoom calls with each of your friends. Most of this is anecdotal, but I'll do why this might match the evidence next.
Firstly, any hypothesis should at least comment on the US racial disparity. This actually does that. According to wikipedia, the US has 80% access for whites, and only 60-65% for blacks and Hispanics. Considering that Hispanic suicide rates didn't appear to go up, it's hard to call this a victory, but whatever.
Secondly, this would help distinguish between past pandemics/really bad stuff and COVID. Widespread internet, and even free long distance calling, are relatively new things.
Thirdly, it explains why we see the same results across multiple studied countries; what do Germany, Sweden, Japan and the US all have in common? We're rich and we have good internet (by global standards).
Fourth, this helps with ODs. From what I've read (but mostly I read about it casually because it's depressing so I freely accept the possibility of me being wrong) the opiate epidemic is hitting rural america a lot harder. Anyone who's ever lived in a serious rural area knows that it's very hard to get decent unlimited internet. Then we might suspect that opiate ODs are a substitute for suicides in rural regions.
So, falsifiable predictions? Basically, we should see results that depend upon internet access, even when we control for both race and income. Once we get suicide rates for less developed countries, we should expect them to have rates drop less than in the US, or possibly even rise. I guess we'll see how this holds up once data comes out.
It is brutally, incredibly weird to me that it didn't occur to you to question the difference between "people say they are depressed" and "people are depressed" here. Not that you didn't settle on that as an explanation, or that you didn't dismiss it as silly - it just doesn't come up at all.
“this explanation implies that people were financially better off with the coronavirus and relief package than they were with neither”
The numbers certainly bear this out. Spending was way down, savings were way up, debt was way down, markets were way up, and stimulus on top of all of it. For many low wage workers enhanced unemployment paid better than working, and millions who lives in debt were able to wipe it away via net savings and/or by taking advantage of rates going to zero.
I do think personal financial insecurity is a well documented motivator of suicide. 2020 solved many more financial problems than it created — so far, at least.
What about esteem dynamics? One notion I have seen floated in relation to the curiously high suicide rate in Australia is that a factor driving people over the edge is the humiliating thought that "I live in the Lucky Country, so I have no excuse for my personal inadequacy." In a pandemic, everyone has an excuse for failing, so depressed people do not also have to cope with (their subjective perception of) humiliation.
When I have thought about suicide (never very seriously, in case someone I know is reading), it's never just about my being depressed. It's about disappointing people I care about.
My guess is that the pandemic actually provided a sort of life raft to depressed people... sure I'm letting my family down, but I don't feel as personally responsable.
Related to: why people kill themselves at all, instead of running away.
Perhaps killing oneself over "that stupid virus" is perceived as a status loss in a way that killing oneself over more personal troubles isn't. It's about the story you want told. Not every suicide is a storytelling act, but I assume many are.
And yes, yes I do think people often care more about status than survival.
Oh thank goodness. I'd already read the one on Works in Progress and when I saw this email come in I thought "aw, man, he's written it up for ACX too? Now I'm gonna have to read basically the same thing again just in case there are any differences!"
So I was relieved to find this was just a link to the one I'd already read. Maybe I need to get a life...
“Make enough comparisons like this and eventually you have” -- here I thought you were going to say "two subgroups which differ from each other just because of a statistical fluctuation".
I'm curious to see the data by age. Per Wikipedia, men over the age of sixty five are the highest risk for suicide. With them being in the higher risk for covid-19, is it possible that covid-19 was directly competing with the suicide numbers?
There is one point you make that really spoke to my circumstance:
"I’ve seen surprisingly many instances of depressed or anxious people, who can’t cope with everyday life, cope perfectly well with emergencies. Finally there’s something obvious and worthwhile they need to do that breaks through the fog of meaninglessness, and finally they have the motivation to try to handle it (because they could die if they don’t). But also, I’ve seen a weird number of anxiety patients who are completely calm in an emergency. Or maybe they have the same level of anxiety as usual, but it finally feels completely appropriate to the situation and so they can respond as reasonably as anyone else would, plus they have more experience with it. "
I have been battling what I was say is moderate depression for about 3 years. In 2017 I took a job on the road that required I leave behind everything I loved to hop around rather miserable rural towns in return for ALL of my financial concerns being handled (the "golden handcuffs" as they've been described to me). I knew I had to get out of the job but couldn't muster my exit and it was causing me great grief. My greatest concerns were the loneliness and boredom I was experiencing being away from friends and loved ones.
Fast forward to March 2020, I get let go from work due to their workload decrease and it completely changed my life for the better. I moved back to the city that I felt I belonged in, and suddenly everyone around me was now experiencing the feeling of isolation and loneliness that I had been going through for years. I very much fall in that bucket of people who were moderately depressed before COVID but found great comfort in the disaster.
I don't believe I would ever have considered myself contemplative of suicide, but can certainly agree with the idea that some function better in crisis. No longer were those I observed around me having the times of their life while I was locked away working in some random town (FOMO), and now I watch people I would normally look to as strong and brave in good times begin to crack under the pressure of the new status quo.
> Or what if you just literally don’t have the personal space, out of earshot of your family, to attempt suicide without someone else noticing?
Aren't suicide attempts tracked as well? This possibility should be discernable if attempts stayed constant but successes fell.
Although I admit thie "dissonance" hypothesis has some intuitive appeal. Feeling depressed in depressing circumstances feels appropriate, but being depressed when everyone else is happy and moving on seems like it would be more distressing. Dissonance seems to be a common theme with mental health problems.
Yet another hypothesis is that people weren't necessarily staying home with family and friends, but they were more frequently *staying in touch*. Depressed people might hear more stories of sadness, fear and depression from family and friends, and the external awareness of other people's circumstances might make them feel less lonely, and they might have shared in turn. Shared misery and all. I could see that helping.
Black people are disproportionately represented among the poor, so maybe they didn't have internet or phone access after losing their jobs, so many of them perhaps didn't see those benefits and possibly that's why suicides didn't fall or even went up.
Something that felt missing to me was more details about the profile of a typical person who commits suicide (or better - a distribution). Are they young or old? Do they have families or roommates or neither? Are they rural/urban, religious/not, educated/not? In what circumstances do they commit suicide (after a long fight with depression as measured by questionnaires?)? Where and how do they do it (e.g. hotel suicides mentioned in another comment. Guns/pills/...)?
I think this would help zero in on what might have changed for them. Ideally we would want the change in suicide rate for each group, like there is for race. But absent that, a change of 10% is pretty large, so can't come from a small group of suicide victims.
Anecdotally, the idea that people with high anxiety do well in crises rings true. I am chronically depressed and anxious, and experienced a large increase in well being during the pandemic (which I mostly kept to myself). The way I thought about it was that a lot of my anxiety comes from feeling like my internal models/interpretations are out of whack with the outside world. I get alarmed by the sense of "why isn't everyone as worried about this as I am?" because it feels like the two options are that most people are misperceiving dangerous situations as safe or that I'm uniquely broken, both of which are frightening in their own ways. When the pandemic hit, it was like the outside world lined up with my emotional state for the first time in memory, and I felt hope that I was relatively sane as a result. I experienced something similar when people started worrying about Trump trying to steal the election a few months before it happened, because I'd had a growing fear for months before that that I might just be paranoid about the prospect.
Seems almost no effect there to explain. I suspect natural disasters in general don't have any consistent effect on suicide because they're both obviously temporary and obviously not your fault. Being born into poverty in some third world country also doesn't seem to have any effect on suicide rates (the only consistent effect I see on the map is a HUGE effect in favor of muslim countries regardless of their wealth or lack thereof.)
I see an obvious explanation that unifies all the weird things about suicidality, that I've never seen proposed before. Someone feel free to tell me what's obviously wrong with it:
1. Committing suicide requires depression _plus motivation_. You need to both think everything sucks and won't get better, and _also_ to exert some momentary burst of willpower to get you from "do nothing about it" to "commit suicide about it."
2. Things that give people more motivation (i.e. increase dopamine) — like certain antidepressants, or like the bright-light-therapy analogue of getting more sunlight in spring, or like Christmas in the middle of winter — increase suicidal ideation in depressed people. Because — at least at first — having more undirected motivation doesn't make depressed people any less depressed (they still think everything is pointless and there's no way to fix anything), but it _does_ allow them to get over the big energy-cost hurdle that was preventing them from committing suicide.
3. The opposite is likely true as well: things that take away people's motivation decrease their likelihood to commit suicide. Anti-psychotics. Winter. Stress (the kind that makes you too tired to do anything at the end of the day.) Eating poorly. Being forced to stay in your house where you experience no novel sensory/social stimuli, and potentially get little sunlight even in summer.
4. Different suicide methods require different amounts of willpower. If there are "easy" methods of suicide available — especially those that don't require _making a decision_ to commit suicide (like drug overdose!) then you'll expect unmotivated depressed people to preferentially enact those forms of suicide, while motivated depressed people don't prefer them over-and-above any other method of suicide. As people get more motivated/manic during depression, more difficult, choice-requiring, and novel methods of suicide will "come into scope" for them — much like how having more motivation left in the day when choosing dinner will lead to being willing to try cooking a new dish.
(This all seems obvious to me, as someone who got diagnosed with ADHD as an adult, and so has experienced both long periods with/without motivation in my life, and sees the contrast in my own thinking every day as my medication kicks in and then later wears off. Modelling depressed people who _don't_ commit suicide, as having the same problem I have when I'm hungry but have run dry of the willpower to decide _what_ to eat, seems pretty intuitive.)
Depression and, at times, suicidal thoughts are no strangers to me.
The concept that there may be two types of depression, one correlating more strongly with suicidality and the other more with "current and presumably transient adverse events", for lack of a better description, resonates strongly with me. A chronic versus a more "acute" one, so to speak, for a definition of "acute" that would run from maybe months to single-digit years.
I would have made a similar self-diagnosis some years ago, and on my odyssey to figure myself out I finally came across the concept of cPTSD; and 'depression' as one of its common symptoms.
Complex Post-Traumatic Stress Disorder in a nutshell means that people, in childhood formation, were conditioned that certain adverse events and situations are things they can do nothing about, and as such severely stunts autonomy and self-efficacy. At the same time, those individuals will have learned incredibly complex coping and survival strategies, so that they will often present as highly functioning even to a trained psychiatrist / neuropsychologist.
Here, I'd like to invite you all, as you feel drawn, to take a look at these concepts. I'll leave a few starting points that got me into it:
* Bessel van der Kolk's book, "The Body Keeps the Score: Brain, Mind, and Body in the Treatment of Trauma"
... from an "I don't care if it is methodically complete and sufficient, it leads to a workable model for everyday dealings with one's emotional undulations" viewpoint, as in the Polyvagal Ladder:
My take on the COVID situation (shituation?) is: It sucks, doubly so for people with fragile psyche or a fragile social support network in the first place. But at the same time, while it will present as an external factor the individual can do very little about, at also presents as something with a lot of change and movement - there are bouts of "we're going to be stuck in this forever", but then there's with some regularity some change happening that again directly affects everyone's everyday life to some degree, so it doesn't feel completely stuck. There continue to be 'surprises', one prime antidote to the cPTSD stuck loop.
Of course, I have a lot of perception bias with all this - once you have a hammer, every problem starts to look like a nail. Going through town, interacting with people, I see trauma and fragmentation and hints of cPTSD everywhere.
I agree with your conclusion. I had multiple patients state outright that they felt better because everyone else was going through the same thing and they finally felt justified in their negative feelings. Social anxiety patients and OCD patients did especially well (and are now doing worse that things are opening up). Even the borderlines improved. Young women with garden-variety depression and anxiety did especially poorly. There was a lot of talk about wasted time in finding a mate and falling behind where they perceived they should be in their lives. Also mothers did terribly for all the obvious reasons. I got a lot of new patients who had no history of previous depression in these categories. Maybe these are reasons why the dip is not reflected as much for women.
I would ask that you research and comment on depression and suicide rates in children and teens. My mother patients frequently expressed that their child was going through depression for the first time. My nanny said over 1/2 of the students in her daughter's class were depressed and the guidance counsellors were reaching out to all students for wellness checks. The first article that comes up from the APA suggests high increases in suicide attempts among teens.
"Results showed recent suicidal ideation was 1.60 and 1.45 times higher in March and July 2020, respectively, than in March and July 2019. Odds of recent suicide attempt were 1.58, 2.34, 1.75 and 1.77 times higher in February, March, April and July 2020 than the same months in 2019, respectively."
similar to your theory that some suicides were still happening but masked/categorized as accidental overdoses, I wonder if the they same thing happened with traffic deaths? Because those were up quite a bit in the US. Which is at first counterintuitive, since there were so many less miles driven in 2020. The explanation I've heard, is that with less traffic, people were able to drive faster, and thus deadlier. Makes sense. And given the drop in suicides, I wonder if also some of that fast/reckless driving was suicidal
There’s one point that feels missing from the final speculation section. Suicide is inherently an act of hopelessness, which you hit on with the Aaron beck “this is endless” point. On the other hand, the pandemic was expected to end, as interminable as it felt in any given moment. When people have a point of light in the horizon, something to really hang hope of improvement on for the long term, it can provide so much fuel to continue because people can plausibly believe life will improve when the bad situation ends.
As someone who lives with someone who is chronically depressed (who, btw, has never felt as comfortable or as satisfied with life as she has during the pandemic), the "this is endless" dynamic I see as a substantial contributor to depressing leading to suicide, and conversely why pandemic depression doesn't lead to as many suicides.
Depression due to the pandemic was a different kind of psychic pain... it was a shared pain, and a pain that by definition has some kind of begin-and-end time period. And, importantly, pandemic depression has an identifiable reason.
I know we can't do exit interviews, but I do wonder if suicides from depression are driven more by the dual factors of not knowing why they feel this way (nothing to point to) and therefore grow convinced that there's no hope in ever getting relief.
Maybe potential suicides were covered by excess mortality eg people who had no will to live succumbed to Corona more by seeking help late and responding much worse to placebo etc.
That would entirely hide suicides in all countries that had excess mortality but not in the ones that had not. Someone looked it up?
I've skimmed through all the comments here but didn't see my pet theory offered as an explanation: misery loves company.
In a normal year you are expected to earn a living, maintain a social circle and have fun hobbies. If you are seriously depressed and failing (or just believe you're failing) to live up to those expectations, you feel like a major fuck-up.
Contrast that to our COVID year. Lost your job? So has everyone else, to the extent that the government has started handing out mass relief without the usual judgmental chatter about the lazy poors. Don't have any friends? Everyone's stuck inside, so it's not like you're missing out. No fun hobbies? Anything worth doing has been cancelled, anyway.
Life may have gotten worse for most people, so depression rates have gone up. But if you're _already_ feeling useless or socially isolated, lockdowns don't change your situation! And at least now there's no pressure to act like everything's fine!
> I won’t waste your time listing the evidence that life got more depressing
I don't know, life got *significantly* better for me, being able to work from home. I know my experience doesn't generalize, but it also doesn't seem like an insignificant factor across the board. Increased unemployment too, applying especially to those with the most shitty jobs.
There's also some thing about covid representing a tangible threat (vs the malaise and ennui of modern life) while at the same time not actually hurting all that many people.
In effect, many ended up safe and comfy at home, and never got sick.
What if COVID and suicides are competing hazards and strongly so for a certain subset of people?
Two of those correlates are age and trait:anxiety. Both are positively correlated (age much stronger than anxiety and anxiety is early research, still seems to be valid and has a large effect) with worse covid outcomes - hospital stays, long covid, death and with suicide in other years.
May be covid took away more of the people who would have otherwise committed suicide.
1) why most suicides in spring? My understanding (personal theory) is this: human bodies do have a built in hibernation program as many other species. This program is controlled by daily sunlight exposure. In preparation for the coming winter our serotonine level and our drive decreases, we eat more and build up fat storage to survive hard winter. Reduced drive (depression?) helped our ancestors (both human and animal) to look for a cave and just sleep there over the winter with least fuel consumption possible. In spring light exposure increases and with it serotonine levels. Drive (and sex drive) and motivation come back. In some individuals the depression stays longer and then they suddenly have the drive to kill themselves. Starting giving Antidepressants (SSRI) is always combined with the warning to tightly monitor the patient the first weeks so that it not kills itself due to this process.
2) why less depressed after a natural crisis or desaster? You mentioned "sense of meaning" and I guess this could be a key. Why there are so many people depressed and suicidal in "happy countries" (I guess you mean western wealthy countries)? Because these sociaties removed a lot of meaning of the lives of their people which were the norm for millions of years for our ancestors: tight daily human connections and interactions with close and loose family members, purposeful activities all the day around like feeding the family, the children, growing food, hunting food, building purposeful things like houses, tools, everything else, singing and dancing together and so on. A lot of these things are gone in the everyday lifes of a lot of western people. Instead they often life alone in anonymous apartments, doing a long daily commute to stupid companies sitting in boring and stupid meetings, producing useless things or services ("what would happen to the world if I, my job or this company would suddenly not exist anymore - nothing - nobody would take notice"). So they feel depressed. Now let's happen a natural desaster Like a flood or a hurricane or a pandemic. Suddenly all the stupid things are not porta to anymore but to help each other. No cracy meetings anymore, a lot of cooperation with formerly strangers from the neighborhood, suddenly a lot of purpose, humankind, success, self realisation asf.
I think it could have to do with jobs. Richer individuals got to work from home while poorer ones were stuck going into work during the pandemic. Race and job/income correlate such that you might see more suicides among essential workers than people at home. When you're constantly surrounded by family and never leave the house you may be depressed but find it harder to plan and execute a suicide.
I read it a couple of days ago (it was posted in the subreddit). Excellent analysis as usual.
Nothing to do with the content of your essay, but I was initially confused by the title of the magazine "Works in Progress". I was thinking that, like your recent "Peer Review Request", you were posting an actual work in progress and was asking for feedback on it before it is finalized, and the Works in Progress website was just a space for authors to release drafts of their essays. However, the About page for Works in Progress makes it clear that publication venue for complete pieces. Sharing this in case other readers were confused in the same way.
Might be some confounding variables in the case of black suicides.
Always suprised that the most widely accepted theory of suicide comes from a 19th century sociologist and that it still seems to fit the data.
A common theme that I find in a lot of analyses is that factors that seem obviously important fall apart on a closer look. You do that with multiple factors here. I also get the same feeling when I listen to historical podcasts by Mike Duncan, like the Revolutions podcast on the French Revolution. Factors that seem important and fundamentally causal in an almost common-sensical way almost always fall apart on closer inspection.
This sometimes leads me to suspect that causality is mostly a mirage in the social sciences, and that any and every factor will fall apart on a "really" close look. Events happen not because of any overarching causal patterns, but only because of sheer dumb chance. This lack of any causal relations has also been stretched to science by Stephen Wolfram in his recent article- https://writings.stephenwolfram.com/2021/03/what-is-consciousness-some-new-perspectives-from-our-physics-project/
Wolfram says that any causal factors that we may find for events are only a result of our limited cognition, and not an inherent description of the universe itself. Although I am apprehensive about stretching the metaphor to that extent, do you also agree that any kind of causality, at least in the social sciences, is mostly illusory?
Sorry, but how can someone research suicide and not have "G-d", "morality", "religion", "value system" come into play? I am not saying that those thing would explain everything, of course not - but excluding them from the picture is beyond pale... Yes, many had a chance to reexamine their convictions and their views on acceptability of suicide.
> Could it be that people have trouble getting the means to commit suicide? For example, maybe they would like to shoot themselves, but all the gun shops are closed?
You already reject this argument in the essay, but another bit of relevant evidence here is that gun sales shot to record highs in 2020, so it doesn't seem like suicidal people would have had much difficulty buying a gun last year.
"When COVID started spreading, life got more depressing, people became more depressed, but suicide rates went down. Why?
"First, are we sure all of that is true? I won’t waste your time listing the evidence that life got more depressing, but what about the other two?"
Maybe you should waste our time with point one. At least in my case, after about the first month when I became sure I'd still be able to get groceries (more or less) without issue, my life took a profoundly positive turn. Since the pandemic started, I finally bought my first house, got a dog, and realized my career-long dream of working from home. I know many people have had great emotional struggles during this time, but how many others like myself have had the opposite happen? Is it a large enough group to be meaningful?
Since my employer announce their return-to-office plan, I've noticed a steep rise in my anxiety levels—and not from fear of COVID, against which I'm now vaccinated.
"I won’t waste your time listing the evidence that life got more depressing" is too dismissive - in exactly what way did things become more depressing and what relationship would you expect these individual factors to have to suicide?
I'm not sure "scoring high on a depression index" is the same as actually being depressed. I went and took the depression questionnaire used by Ettman et al. - the Patient Health Questionnaire–9 - and it would have reported me as depressed even when I definitely wasn't depressed. Back when covid first got to my country of residence, I experienced some anxiety which resulted in taking what I called "anxiety naps" where it felt like I would just shut down and sleep for an extra four hours most every day after work. I was also stress-eating a lot more than I usually did. I also had trouble concentrating on things - my mind would keep drifting back to the horror stories I'd read from hospitals in countries where the pandemic was hitting the hardest. All of these were specific responses to stress and anxiety over the threat of coronavirus - related, probably, to my history of respiratory illness and my specific fears about how my young children would fare if I died. At no point in the pandemic did I consider myself "depressed", and over time I built up better coping mechanisms.
But of course the nature of those specific problems - extreme but somewhat rational fear of dying horribly and leaving my kids without a father - contraindicates suicide. Put another way, while fear of death might lead you to sleep more, or overeat, or lose focus on your work, or fail to take joy in things that you otherwise would have enjoyed, you wouldn't expect someone to commit suicide out of a fear of dying.
In fact it might be worth looking at whether fear of death decreases suicide in general - do people who live more dangerous lives kill themselves with less frequency? Or something like that.
Also, I wonder if being stuck inside with your family members brought more people closer together and reduced the kind of isolation that sometimes precipitates suicide.
I'm particularly sympathetic to the "change of pace" line of thinking. On Twitter @PaulSkallas always talks about how people desire a certain amount of volatility in their lives. They get bored going to work every day, the same routine, following a traditional life trajectory, etc...sometimes they secretly hope for something radical to happen, even if it's objectively "bad" for them. I've certainly felt this way myself at times. It similar to a joke made by one of the hosts of the Red Scare podcast. "Don't kill yourself because something retarded might happen and you'd miss it". Obviously not a super polite articulation of the point. But I think the insight is the same as above.
The pandemic was something new. What if it's the apocalypse? What if society emerges from the crisis totally remade? Maybe you're depressed in your email-office job, but you'll be happy in your new post-apocalyptic role as a community garden tender. Maybe UFOs show up tomorrow and show us the wonders of the universe. Gee, maybe suicide can wait for another day, at least until the conclusion of the most recent species-wide narrative arc. You never know!
I wonder if part of the decrease is because people are less likely to commit suicide when another one of their family members is physically present in the same home. This doesn't explain the black/white divergence though.
On the black/white divergence: I wonder it the reported increase in gun sales during the pandemic varied by race?
> Second, one pillar of Aaron Beck’s triad of depression is “this is endless.” A sense that nothing can ever change, so why bother to wait and see? The pandemic was the opposite of that.
This is strongly, strongly, strongly inconsistent with the lived experiences of literally every person I know. I cannot think of a single person who was like "yeah this is totally temporary it's only two weeks then it's back to normal".
One thing I noticed was this line "Second, the US has a waiting period on guns, so most gun suicides are by people who already owned the gun when they became suicidal.". While some places do have waiting periods, some don't. For example my state has no waiting period, I walk in wanting a gun, I walk out with a gun.
Besides that one question I would have is around the kind of people who would be more prone to committing suicide. I mean if you are depressed and everything shuts down, does that impact you that much? If you were depressed were you going out to bars partying and visiting family? Maybe now that everyone is a shut in you feel some sort of kinship, "Hey welcome to my world normies, enjoy the ride". All the newly depressed people know it is temporary and rally around beating covid.
I would have loved to see more analysis on the overdose angle. If we add those to the suicides, does the counterintuitive effect mostly disappear?
I always go into this with the thought: "yeah, but I wonder if he's thought of THIS!", and I always find out you have. I have to appreciate how thoroughly you consider things. Here are the two ideas that I thought of, and my thoughts on your points:
1. Overdoes vs. suicide. I've had one sad personal experience with a young man where the question as to whether this was an overdose or a suicide was unclear. My feeling was that we all agreed it was an overdose out of kindness for the surviving family, and maybe to lessen the shock. I wonder if, in this pandemic, there was either less investigation of these deaths happening, or whether there was just an unconscious nudge that way.
2. Meaning. I have never seriously considered suicide, but when I'm at my most dark points, it's because of the bleak expanse of unchanging, bland, dreary life ahead. The pandemic was at least interesting and engaging in a lot of ways, as grim as that is. I think people are itching to fight for something meaningful, or work for something meaningful. Most of our careers aren't very meaningful: for example, my job is helping big companies track how you behave online so they can market to you more effectively. Kind of depressing. So you've got this pandemic, and we've got the kind of existential moment, and we can all kind of "fight" for something.
At least at first. Then it starts to get political, with your mask adherence becoming a tribal thing, and everyone back to the same routines, but with social distancing. So there's that initial interesting time of the crisis (that Zunin and Myers graphic is awesome), but then it's back to the usual slog of meaninglessness. And it might be that we come out into a weird economic period that doesn't rebound into "reconstruction" the way a different crisis might, because the crisis of the pandemic was so odd, and we've already kind of been "reconstructing" through the whole thing, and now we'll have the bill to pay for the unemployment and supply chain problems and kids having just terrible school years, so they're practically a year behind. . .
Anyway, the Durkheim/"meaning" thing seems most likely to me based on my own observations.
as a person with medium to severe depression before the pandemic and very severe depression including suicidal ideation during the pandemic, I would be a data point for your last two hypotheses - the change of pace and making the whole shebang feel at least a bit different (before it turned into endless hopelessness again) + not ever being alone to actually do anything. and the reason I would more likely feel inclined to kill myself in April-ish is easy: between October and March you can (in the northern hemisphere) blame it on winter and tell yourself that it'll get better in spring, because sunshine. Then spring comes around, and it doesn't actually get better... then it definitely must be forever this bad, so what's the point. At least that's always been my reasoning. Hooray for anti depressant medication.
I remember when it first started and millions were losing there jobs my firm did a few layoffs. Ordinarily that would make me anxious and depressed. But somehow the thought that millions of other facing the same thing made it much easier to deal with. Maybe part of suicidal depression is the idea that bad things are happening just to you. You've been singled out.
On the other side of it, having a job, when everyone else has a job, doesn't feel amazing. But having a job when millions are losing theirs makes it feel special and like a real accomplishment.
Did suicide attempts fall too? Are people just less successful since there is a greater proximity to people at home?. I also wonder if people who couldnt handle the situation might have just decided to ignore the pandemic as a way of coping instead of Suicidal behavior.
> So it’s weird to see drug overdose deaths follow such a different trend from suicide. I’m tempted to wonder if people are misclassifying suicides as overdoses, but I don’t see any evidence of this or any reason why it would happen.
One theory: we always misclassify some drug-related suicides as overdoses, and the pandemic caused some suicidal people to shift their suicide method to drug-related, and as a result we increased the number of suicides misclassified as overdoses.
We could check this theory by checking if the increase in overdose deaths cancels out the decrease in suicides.
"But most people stayed home even in places without legally mandated lockdowns"
-Citation needed.
One observation of possible relevance to apparent difference in the black/white suicide directionality gap: I live in a majority-black neighborhood, and I found that for the first several months of the pandemic, almost *nobody* in my neighborhood observed basic safety measures outside the home. People would wear masks in stores once it became legally mandatory, but outdoors, only a fraction of people wore masks, and only a fraction of those people wore them properly covering their noses and mouths. I constantly saw people approaching each other within distances of 2 feet or so (this is a small personal space bubble neighborhood) and taking their masks off to talk to each other, often touching each other on the shoulders, sometimes even each other's faces in the process.
I noticed practically no difference in terms of outdoor socialization during the pandemic versus pre-pandemic, until around last Winter, after the election. People actually wear masks outdoors now, and wear them properly, and fewer people socialize at physical contact distance, but it took a *long* time for this change to take.
"Also, this explanation implies that people were financially better off with the coronavirus and relief package than they were with neither, which doesn’t really seem to match reality."
Why not? CARES was uncharacteristically generous in that many people actually did end up better off financially. Someone who worked a shitty service job slinging popcorn at a movie theater for minimum wage and got laid off by their employer probably wasn't that invested in concessions as a career and would've been quite happy to get more money for less effort.
"Third, maybe people became more worried about the effect that suicide attempts would have on their family. Part of this is increasing social closeness—people suddenly had to spend every waking moment with family members, or else bubbled together with people they had previously lived apart from. Another part is that their family members were also suffering, and depressed people decided it was unfair to ask them to bear the burden of a family member’s suicide at the same time as everything else. Or what if you just literally don’t have the personal space, out of earshot of your family, to attempt suicide without someone else noticing?"
This was also my theory upon first hearing that suicides went down during the pandemic, and before I looked at further data. More specifically, I predicted:
(a) that the drop in suicides was due to fewer suicides among young people, who were newly forced back into their parents' home
(b) that deaths of despair among adults had increased.
I don't really know how to verify my predictions -- is there an age-breakdown of the suicides anywhere? I consider the increase in overdoses to be weak confirmation of (b) for now.
When you use phrases like “studies from Norway, England, Germany, Sweden, and New Zealand” it is unclear what this means. My best guess just, like, grammatically, would be studies from research institutions in those countries; whereas my best guess based on what information would actually be relevant to the reader is studies whose study populations were from those countries. Maybe it happens to be both simultaneously for these examples.
Anecdotally, my therapist friend had the exact same observation of anxiety patients. All of a sudden, it's like they were in their element, or like their life outlook had suddenly been vindicated.
Re drug overdoses: there are two plausible explanations I've heard:
1. More people doing drugs alone, with no one to call an ambulance or administer Naloxone
2. Erratic supply causes people to buy drugs in bulk, which leads to overdose.
It should actually be easy to figure out if 1. is true, just based on the details of individual overdoses.
I think the "two types of depression" is on the correct track. There's the 'normal' type of depression, which is the depression in the other post. That's not going to change very much during Covid-19 and may even, weirdly, become slightly better. Those are the people who, if they're going to kill themselves, are going to kill themselves whether or which.
The second type is "kinda down, kinda blue, kinda not feeling good" which happens when people not suffering from the first type of depression have their usual "go out and enjoy myself" activities curtailed, along with worries about the pandemic, maybe laid off from their job, etc. Ordinarily they're not thinking about killing themselves and things have to get really, really bad to push them to that stage. When the restrictions are limited, life goes back to normal, they'll bounce back.
What I mean for the first set of people, those who are suffering from depression before, "weirdly become slightly better" is that now *everyone* is in the same boat. I'm introverted, I don't like people, and I think I might be mildly agoraphobic. That means currently I *love* the lockdown conditions, because this is normal life for me *plus* bonus "working from home? no interaction with other people? can slob around in my night clothes all the day? oh yeah baby!"
Also having to wear a mask is great when you have (a) plain to ugly features and (b) really can't be bother slapping on makeup. When three-quarters of your face is covered between mask and glasses, it's really liberating.
Nobody can travel, nobody can socialise, everyone is confined to the same restrictions. This is awful for people who have social lives and like going out and mixing in the world. For depressed people who already have the tendency to stay inside, avoid people, and have lost interest in formerly pleasurable activities, suddenly this state of affairs is the new normal and now they're virtuous for not doing any of the "just go out, have fun, meet people, force yourself to interact" activities. This is a sort of relief of pressure, funnily enough.
Drug use (and as a result of that overdoses) going up? Yeah, I'd expect that too, the same way alcohol consumption goes up. People can't go out and do their normal activities, they get bored with being confined indoors with TV and games, they consume their usual fun substances to cheer themselves up, but now they're doing more of them, regularly, and possibly upping doses as well. They don't intend to kill themselves, it's not suicide, it's just a bit too much of your usual fun substance/mixing several recreational substances together and then - oops.
My first thought was to look at suicide during war. You gave a link saying that suicide in Britain went up during WW2, but the linked article only said that suicides in Scotland declined less during the war years than one would predict given a longer-term data sample.
I say, look for more data on this. "Suicides went down less than predicted" probably depends critically on what window you use to set your "background effect". Also, the Scots weren't uniting against a threat so much as they were fulfilling military obligations imposed on them by a foreign nation (England) which many of them (like, maybe half) saw as their oppressors.
I applaud and appreciate your methodical refusal to come to any conclusion. Measured rationalism, while less than satisfying as a narrative, is a refreshing approach to potentially torture-prone statistics. It’s like the rhetorical equivalent to publishing negative studies to insure against publication bias. Thanks for sharing your still-baking thought process. The world needs more out-loud examples of such things.
One of my kids (M 20) is a front line worker and was therefore not locked down. He has bipolar disorder, which is managed with Aripiprazole and psychotherapy. He experienced an episode of suicidal ideation over the winter that seemed to come from nowhere. He explained to his therapist (and me) that what was depressing him was his failure to progress at work, the sense of "same shit, different day" and that he didn't see a way to get back to being a college student (which he currently can't handle). So, N = 1 but I go along with the "hopelessness" theory.
Overall, very well written and helpful - as always.
Scott you're at your strongest when you discuss about stuff related to your job and I wish you'd continue doing it instead of reviewing books or hinting about HBD
Since most of the death statistics are going to be based on CDC death certificate data, I suspect that who completes the death certificate is somehow a factor. Death certificates of violent suicides or completed overdoses found in the field with an accompanying note are going to be completed by the local coroner or an emergency personnel trained in and certified in death pronouncement. Most overdoses of the kind you describe here, "There’s a thin line between a drug overdose death and a suicide. Something really terrible happens, you can’t stand the thought of dealing with it, so you take ten times your normal dose of painkiller in order to fall into a warm, pleasant sleep. At some level of misery, “but what if you take too much and never wake up?” stops feeling like a flaw in this plan. For some people, it’s a bonus—all the oblivion of suicide, with none of the guilt—after all, you weren’t trying to kill yourself, not exactly.", they will be managed in an ICU somewhere by a critical care physician, and unless there is an accompanying clear suicide note found at the scene that was transported with the patient and makes its way into the ICU and the patient's chart, those deaths are not likely to be classified as suicide at all and will instead be accidental overdoses with the death certificate being completed by the attending physician at the time of the patient's death. I have taken care of many of such ICU overdose admits and completed many death certificates concerning those cases and have rarely indicated suicide on the death certificate. It is very important to be accurate when completing death certificates because they do have life insurance implications etc. It does not surprise me that overdoses increased during those early months of lock downs, and I suspect that a fair number of those overdoses were also suicides of the type you describe.
Hypothesis on drug overdoses: with borders largely closed, it's getting more difficult to get certain foreign-made drugs, so addicts are switching to other even worse drugs; e.g. if you can't get heroin you try fentanyl.
Did I misunderstand something, or does the CDC say that 2020 10.7% of Americans considered suicide? This number seems absurdly large... Or does "thinking about suicide" mean something different?
I think you mean to cite "Daly, Oswald, and Wilson" rather than "Daly, Oswald, and Wu".
I notice you are not too fond of the Durkheimian "during a crisis people seek together" hypothesis. A reflection here:
A test of Durkheim's theory should probably use something else than suicide rates as an operationalisation/indicator. Since (if you follow Durkheim) there are different suicide motivations, and not all of them are of the type that is influenced by "getting closer together/becoming more distant from each other".
...A better operational indicator to test the Durkheimian thesis is probably to check the rates of marriage/co-habitation break-up. The Durkheimian thesis straightforwardly leads to the assumption that the breakup rate will go down during an epidemic/disaster.
...I have not checked if that has been the case during the present pandemic (worth doing!), but here is an old article finding this effect after the 1995 Oklahoma bombing: Nakonezny et. al. (2004): Did divorces decline after the Oklahoma city bombing? Journal of Marriage and the Family vol 66 (1) p. 90-100.
Ok, I know that checking if Durkheim was on to something is not what you are interested in here: you are interested in theories that can explain variation over time in suicide rates, regardless of where these theories come from. But let me offer an elaboration that might be relevant, although admittedly speculative:
There are two types of suicide that are dominant in modern societies, if one follows Durkheim: Egoistical/self-interested suicide (calculating that the costs of continued living are higher than the benefits: Gary Becker’s & Richard Posner's rational choice theory of suicide is in this mould); and anomic suicide/"existential" suicide (Seeing/finding no meaning/purpose in life; acute Weltschmerz; that Kirkegaardian/Nietszchean stimmung).
Durkheim arguably claims that the odds of committing egoistic suicide is influenced by having friends and relatives that cares about you (and hence this type of suicide should go down during a crisis that brings people closer together, similar to breakup rates); while anomic suicide results from a deeper-level cultural ennui that is not necessarily influenced by how distant/close you feel to others.
Which might explain why drug overdoses have gone up; they may be related to the feeling "life is meaningless anyway, and it feels even more meaningless these days", rather than to "I thought no-one cared about me, but in this present crisis I somehow feel that I was wrong".
Ok, perhaps egoistical versus anomic seems like a hair-splitting difference in suicide motivations. But the distinction may potentially at least explain why suicide rates and drug overdoses move in different directions during the present covid crisis. And it might perhaps also explain why meta-studies on suicide do not find consistent results supporting an assumed Durkheimian approach (since such studies can hardly distinguish between Durkheim's different types of suicide motivations, and hence different types of suicides).
….The above is perhaps a bit rambling and disorganised, but I'll post it anyway.
Great piece! Summary: Suicides declined possibly cuz a) misery loves company, b) curiosity, c) and responsibility towards others. And we don't know why drug overdoses rose when suicides declined.
Curious what’s the story behind posting this there and not on your own blog? Is this a business/partnership/commeercial arrangement thing? (i.e. you’re posting it there it helps giving their website more exposure?)
Re: "The US has a waiting period to buy guns"
No it doesn't. A few individual states, including California, have waiting periods, but most of the country does not.
Dostoyevski's notion of finding "happiness in the midst of despair" is distinct from "people pull together during a crisis" although at times it looks similar. Under Dostoyevski's formulation it's the crisis itself -- which may be personal -- which, although causing one to feel bad, also causes one to feel like they are *supposed* to feel bad. If feeling bad at least makes sense it's easier to tolerate.
I think of this idea as similar to Adam Smith's observation that people suffer from small embarrassments more than big ones.
Random thought that I have no emotional attachment to...a lot of people who commit suicide are "Zoom class" or "affluenza" types whose lives actually got better during this period since they still made money, got to stay at home, and could actually do all the stuff they normally want to do?
The increase in overdoses might not necessarily reflect an increase in suicidality/indifference to death. I think hard drug use in a social setting might actually be safer than solitary hard drug use. This would be due to having help on hand to administer antidotes or call an ambulance if someone takes too much. I think it could also be due to social pressures leading to people using less drug overall. The two biggest ones would be avoiding the shame/stigma of taking a way higher dose than anyone else and wanting to remain lucid enough to socialize.
I don't use opiates and I don't know anyone that does, but I do see the same pattern of people using lower doses in social settings and higher doses alone across all sorts of other types of alcohol and drug users.
"Also, there were slightly more black suicides in the months just before and after the crisis, so that if you look at total black suicides during 2020, it’s about the same as any other year—in fact, lower than two years ago."
This passage was opaque to me. Did you mean slightly *fewer* black suicides? There are two small bumps on the chart that might be the ones you're referring to, but I don't see how slightly more suicides before and after the crisis, added onto high suicides during the crisis, could lead to the year being normal in total.
> Also, [the 'increased welfare' explanation] implies that people were financially better off with the coronavirus and relief package than they were with neither, which doesn’t really seem to match reality.
For what it's worth, real personal income per person shot up sharply during the pandemic: https://fred.stlouisfed.org/graph/?g=EiOd (sorry for weird units, don't know how to fix that). The worst month throughout the entire pandemic on that metric, March 2020, was only as bad as July 2019, at around ≈$51,000. April 2020 it shot up to ≈$58,000 (going by the trendline, that wouldn't have happened for ≈5 years by default) and hasn't come back down to pre-pandemic levels since. This March was even more, at ≈$65000.
But this doesn't address your counterargument that suicides dropped off in March, before the stimulus.
Also, personal consumption expenditures per person went down even more drastically, from February 2020's $45,000 to March's $37,000, only fully recovering recently, and I guess that's a better measure of well-being overall: https://fred.stlouisfed.org/graph/?g=EiOr. (Though, question for econ people: does personal consumption include paying off debt? Does it depend on what you're paying off debt *for*? I'm trying to get a sense of where people's stimulus checks went if not on consumption — sitting in their bank account with everything closed, paying off debt, or what?)
Overall I'm pretty unclear on what people were going through financially during the pandemic outside of my bubble (and since everyone in the news media and in my social media bubbles is talking about zoom calls and whatnot, I can't really expect them to inform me).
My first instinct, and it might very well be wrong, is that some populations at risk for suicide (poor, old, sick) are also at increased risk for COVID, hell depression itself has effect on the immune system, so maybe the virus snatched suicide candidates before they could get around to it (same way it ate "market share" of other infectious diseases). And by snatched I do not even necessarily mean killed : like Joe Depressed is really hitting rock bottom and would kill himself, maybe tomorrow, but before he knows it he's in an intensive care unit where it's more or less impossible to commit suicide, and by the time he's discharged the worst of the crisis is behind him.
No real idea ho to work out the math.
On the differential suicide rate for black people in the US, we're not in an all else being equal scenario here. There's the issue that if you are black in the US there's a dominant media narrative that you are a victim of a racist society. That's going to have an effect on people, increasing feelings of helplessness and isolation (if this makes people more conscious of their perceived differences), and there seems no reason this could not be as big an effect as COVID.
When you write "there are two types of depression" I think you mean "there are two modes of depression". The statements that follow support a change in mode for those who are already in one, whereas two types could exist simultaneously.
Best practice for spelling Māori these days either has a macron over the ā or uses two as - Maaori. Both are more accurate, which one is used depends on the tradition/preference of the local iwi (Māori traditional group).
Do you have access to international overdose data? If the suicide dip/overdose surge pattern does not appear in other countries, that would be strong evidence against the cause of death misclassification hypothesis.
Well, the last explanation certainly rings most true to my experience. The chronically mildly -- and I emphasize "mildly" -- depressed people I know perk right up when a genuine existential crisis comes along, when it's root, hog, or die. Maybe there's even some weird bit of pride in there, as in, I may well kill myself but I'm damned if I'll just let this stupid hurricane/mugger/scary skin cancer kill me. But I have zero experience with severely depressed people, and an intution is worth bupkis when assessing broad social trends.
One thing I didn't see addressed there is the precision of the "depressed" categorization. Is it possible people who were stressed and scared were lumped in?
what portion of suicide is attention seeking? if you are motivated by attention, this may have been absent during covid. i dont know how much of this is true or a myth however.
This article makes me wonder how many suicidal people died of COVID-19 before they managed to commit suicide. Something can technically reduce suicide... by killing people another way.
Excellent, thank you. Highlighting the slipperiness of statistics and their danger in the hands of most people, is so important these days. I really like your conclusions on this one.
Haven't seen what seems like the intuitive answer to me. Committing suicide generally means you are not only miserable in the present but expect to be miserable in the future. I have never considered suicide when I had food poisoning, but if I were diagnosed with "chronic food poisoning" with a 0% chance of recovery, suicide might become more compelling.
Depression often arises with the belief that one is irredeemably worthless, unlovable, etc. and suicide may appear to be the best decision in light of this distorted view. When someone's depression is clearly caused or heavily influenced by an event they perceive to be temporary, I would speculate they would be far more likely to wait and see how things go before making such a permanent decision.
In fact, I would not be surprised if this "wait and see how things go" effect even applies to people who were already depressed before the pandemic. When your life gets shaken up in drastic and unforeseen ways, there is a sense that the cards might land in your favor once the dust settles. You can't just end on a cliffhanger.
My dear friend Kieran Latty, who is an economist working on subjective well-being, has always told me that recessions are associated with a paradoxical fall in the suicide rate. If this is true it would seem to explain it (while opening up another mystery- why do recessions reduce suicide). However a brief Google turns up a number of papers alleging the opposite- that recessions in fact cause suicide- which would just compound the mystery here. No idea what the truth is.
Kieran's explanation for the purported phenomena of recessions reducing suicide was that when many people are down and out economically, being down and out economically feels less bad.
I think you gave up on the overdose explanation too early, and am confused as to why you treated it as you did. It seemed like the one place where you had concrete numbers that could provide an explanation. You move on from it and instead focus on more speculative theories. While those theories sound anywhere from plausible to accurate, they are not data-driven. Ignoring the data driven answer is out of line with my expectations of your approach (and of the rationalist sphere approach in general). I also had a previous expectation that the line between what is reported as a suicide vs an overdose is a moving target subject to much personal judgment, weak circumstantial evidence, and policy directives. While I think that prior expectation made the change in approach stand out more starkly to me, upon reflection I do still think that the approach you took is surprising.
A (quite possibly wrong) hypothesis that has enough of a chance of being right I'll register it now for gloating rights.
What if internet, video calls, and to some degree improved phone service, were the drivers of the low suicide rates.
First up, why would we expect this to be the case? Well, most of the depressed people were probably otherwise healthy people whose social lives got destroyed by COVID and quarantining. However, access to communication allowed them to maintain some degree of a support network. Eating every meal alone for a year might still be depressing, but when you feel suicidal you still have your weekly zoom calls with each of your friends. Most of this is anecdotal, but I'll do why this might match the evidence next.
Firstly, any hypothesis should at least comment on the US racial disparity. This actually does that. According to wikipedia, the US has 80% access for whites, and only 60-65% for blacks and Hispanics. Considering that Hispanic suicide rates didn't appear to go up, it's hard to call this a victory, but whatever.
Secondly, this would help distinguish between past pandemics/really bad stuff and COVID. Widespread internet, and even free long distance calling, are relatively new things.
Thirdly, it explains why we see the same results across multiple studied countries; what do Germany, Sweden, Japan and the US all have in common? We're rich and we have good internet (by global standards).
Fourth, this helps with ODs. From what I've read (but mostly I read about it casually because it's depressing so I freely accept the possibility of me being wrong) the opiate epidemic is hitting rural america a lot harder. Anyone who's ever lived in a serious rural area knows that it's very hard to get decent unlimited internet. Then we might suspect that opiate ODs are a substitute for suicides in rural regions.
So, falsifiable predictions? Basically, we should see results that depend upon internet access, even when we control for both race and income. Once we get suicide rates for less developed countries, we should expect them to have rates drop less than in the US, or possibly even rise. I guess we'll see how this holds up once data comes out.
It is brutally, incredibly weird to me that it didn't occur to you to question the difference between "people say they are depressed" and "people are depressed" here. Not that you didn't settle on that as an explanation, or that you didn't dismiss it as silly - it just doesn't come up at all.
“this explanation implies that people were financially better off with the coronavirus and relief package than they were with neither”
The numbers certainly bear this out. Spending was way down, savings were way up, debt was way down, markets were way up, and stimulus on top of all of it. For many low wage workers enhanced unemployment paid better than working, and millions who lives in debt were able to wipe it away via net savings and/or by taking advantage of rates going to zero.
I do think personal financial insecurity is a well documented motivator of suicide. 2020 solved many more financial problems than it created — so far, at least.
https://www.nytimes.com/2021/01/01/upshot/why-markets-boomed-2020.html
What about esteem dynamics? One notion I have seen floated in relation to the curiously high suicide rate in Australia is that a factor driving people over the edge is the humiliating thought that "I live in the Lucky Country, so I have no excuse for my personal inadequacy." In a pandemic, everyone has an excuse for failing, so depressed people do not also have to cope with (their subjective perception of) humiliation.
BLM movement may be relevant, although according to google trends it didn't really pick up until June 1 2020
When I have thought about suicide (never very seriously, in case someone I know is reading), it's never just about my being depressed. It's about disappointing people I care about.
My guess is that the pandemic actually provided a sort of life raft to depressed people... sure I'm letting my family down, but I don't feel as personally responsable.
Related to: why people kill themselves at all, instead of running away.
Perhaps killing oneself over "that stupid virus" is perceived as a status loss in a way that killing oneself over more personal troubles isn't. It's about the story you want told. Not every suicide is a storytelling act, but I assume many are.
And yes, yes I do think people often care more about status than survival.
Oh thank goodness. I'd already read the one on Works in Progress and when I saw this email come in I thought "aw, man, he's written it up for ACX too? Now I'm gonna have to read basically the same thing again just in case there are any differences!"
So I was relieved to find this was just a link to the one I'd already read. Maybe I need to get a life...
“Make enough comparisons like this and eventually you have” -- here I thought you were going to say "two subgroups which differ from each other just because of a statistical fluctuation".
;-)
"anxiety patients who are completely calm in an emergency": similia similibus curentur?
Guess as the "two types/axes of depression":
- The world is [negative affect].
- I am [negative affect].
>Britain came together a lot during World War II, but if anything, suicides went up.
For years I'd heard that suicide rates in the UK fell during the Blitz. It's not true?
Another data point: heard a news story on the radio that suicides had not increased during the pandemic in Ireland, the most recent coverage online is here: https://www.irishtimes.com/life-and-style/health-family/suicides-have-not-increased-during-pandemic-figures-indicate-1.4492985
I'm curious to see the data by age. Per Wikipedia, men over the age of sixty five are the highest risk for suicide. With them being in the higher risk for covid-19, is it possible that covid-19 was directly competing with the suicide numbers?
There is one point you make that really spoke to my circumstance:
"I’ve seen surprisingly many instances of depressed or anxious people, who can’t cope with everyday life, cope perfectly well with emergencies. Finally there’s something obvious and worthwhile they need to do that breaks through the fog of meaninglessness, and finally they have the motivation to try to handle it (because they could die if they don’t). But also, I’ve seen a weird number of anxiety patients who are completely calm in an emergency. Or maybe they have the same level of anxiety as usual, but it finally feels completely appropriate to the situation and so they can respond as reasonably as anyone else would, plus they have more experience with it. "
I have been battling what I was say is moderate depression for about 3 years. In 2017 I took a job on the road that required I leave behind everything I loved to hop around rather miserable rural towns in return for ALL of my financial concerns being handled (the "golden handcuffs" as they've been described to me). I knew I had to get out of the job but couldn't muster my exit and it was causing me great grief. My greatest concerns were the loneliness and boredom I was experiencing being away from friends and loved ones.
Fast forward to March 2020, I get let go from work due to their workload decrease and it completely changed my life for the better. I moved back to the city that I felt I belonged in, and suddenly everyone around me was now experiencing the feeling of isolation and loneliness that I had been going through for years. I very much fall in that bucket of people who were moderately depressed before COVID but found great comfort in the disaster.
I don't believe I would ever have considered myself contemplative of suicide, but can certainly agree with the idea that some function better in crisis. No longer were those I observed around me having the times of their life while I was locked away working in some random town (FOMO), and now I watch people I would normally look to as strong and brave in good times begin to crack under the pressure of the new status quo.
> Or what if you just literally don’t have the personal space, out of earshot of your family, to attempt suicide without someone else noticing?
Aren't suicide attempts tracked as well? This possibility should be discernable if attempts stayed constant but successes fell.
Although I admit thie "dissonance" hypothesis has some intuitive appeal. Feeling depressed in depressing circumstances feels appropriate, but being depressed when everyone else is happy and moving on seems like it would be more distressing. Dissonance seems to be a common theme with mental health problems.
Yet another hypothesis is that people weren't necessarily staying home with family and friends, but they were more frequently *staying in touch*. Depressed people might hear more stories of sadness, fear and depression from family and friends, and the external awareness of other people's circumstances might make them feel less lonely, and they might have shared in turn. Shared misery and all. I could see that helping.
Black people are disproportionately represented among the poor, so maybe they didn't have internet or phone access after losing their jobs, so many of them perhaps didn't see those benefits and possibly that's why suicides didn't fall or even went up.
Very interesting data nonetheless.
Interesting and in depth as always.
Something that felt missing to me was more details about the profile of a typical person who commits suicide (or better - a distribution). Are they young or old? Do they have families or roommates or neither? Are they rural/urban, religious/not, educated/not? In what circumstances do they commit suicide (after a long fight with depression as measured by questionnaires?)? Where and how do they do it (e.g. hotel suicides mentioned in another comment. Guns/pills/...)?
I think this would help zero in on what might have changed for them. Ideally we would want the change in suicide rate for each group, like there is for race. But absent that, a change of 10% is pretty large, so can't come from a small group of suicide victims.
Anecdotally, the idea that people with high anxiety do well in crises rings true. I am chronically depressed and anxious, and experienced a large increase in well being during the pandemic (which I mostly kept to myself). The way I thought about it was that a lot of my anxiety comes from feeling like my internal models/interpretations are out of whack with the outside world. I get alarmed by the sense of "why isn't everyone as worried about this as I am?" because it feels like the two options are that most people are misperceiving dangerous situations as safe or that I'm uniquely broken, both of which are frightening in their own ways. When the pandemic hit, it was like the outside world lined up with my emotional state for the first time in memory, and I felt hope that I was relatively sane as a result. I experienced something similar when people started worrying about Trump trying to steal the election a few months before it happened, because I'd had a growing fear for months before that that I might just be paranoid about the prospect.
Seems almost no effect there to explain. I suspect natural disasters in general don't have any consistent effect on suicide because they're both obviously temporary and obviously not your fault. Being born into poverty in some third world country also doesn't seem to have any effect on suicide rates (the only consistent effect I see on the map is a HUGE effect in favor of muslim countries regardless of their wealth or lack thereof.)
https://www.undispatch.com/map-day-suicide-global-health-priority/
I see an obvious explanation that unifies all the weird things about suicidality, that I've never seen proposed before. Someone feel free to tell me what's obviously wrong with it:
1. Committing suicide requires depression _plus motivation_. You need to both think everything sucks and won't get better, and _also_ to exert some momentary burst of willpower to get you from "do nothing about it" to "commit suicide about it."
2. Things that give people more motivation (i.e. increase dopamine) — like certain antidepressants, or like the bright-light-therapy analogue of getting more sunlight in spring, or like Christmas in the middle of winter — increase suicidal ideation in depressed people. Because — at least at first — having more undirected motivation doesn't make depressed people any less depressed (they still think everything is pointless and there's no way to fix anything), but it _does_ allow them to get over the big energy-cost hurdle that was preventing them from committing suicide.
3. The opposite is likely true as well: things that take away people's motivation decrease their likelihood to commit suicide. Anti-psychotics. Winter. Stress (the kind that makes you too tired to do anything at the end of the day.) Eating poorly. Being forced to stay in your house where you experience no novel sensory/social stimuli, and potentially get little sunlight even in summer.
4. Different suicide methods require different amounts of willpower. If there are "easy" methods of suicide available — especially those that don't require _making a decision_ to commit suicide (like drug overdose!) then you'll expect unmotivated depressed people to preferentially enact those forms of suicide, while motivated depressed people don't prefer them over-and-above any other method of suicide. As people get more motivated/manic during depression, more difficult, choice-requiring, and novel methods of suicide will "come into scope" for them — much like how having more motivation left in the day when choosing dinner will lead to being willing to try cooking a new dish.
(This all seems obvious to me, as someone who got diagnosed with ADHD as an adult, and so has experienced both long periods with/without motivation in my life, and sees the contrast in my own thinking every day as my medication kicks in and then later wears off. Modelling depressed people who _don't_ commit suicide, as having the same problem I have when I'm hungry but have run dry of the willpower to decide _what_ to eat, seems pretty intuitive.)
I read this with great interest.
Depression and, at times, suicidal thoughts are no strangers to me.
The concept that there may be two types of depression, one correlating more strongly with suicidality and the other more with "current and presumably transient adverse events", for lack of a better description, resonates strongly with me. A chronic versus a more "acute" one, so to speak, for a definition of "acute" that would run from maybe months to single-digit years.
I would have made a similar self-diagnosis some years ago, and on my odyssey to figure myself out I finally came across the concept of cPTSD; and 'depression' as one of its common symptoms.
Complex Post-Traumatic Stress Disorder in a nutshell means that people, in childhood formation, were conditioned that certain adverse events and situations are things they can do nothing about, and as such severely stunts autonomy and self-efficacy. At the same time, those individuals will have learned incredibly complex coping and survival strategies, so that they will often present as highly functioning even to a trained psychiatrist / neuropsychologist.
Here, I'd like to invite you all, as you feel drawn, to take a look at these concepts. I'll leave a few starting points that got me into it:
* Bessel van der Kolk's book, "The Body Keeps the Score: Brain, Mind, and Body in the Treatment of Trauma"
There's an enganging summary:
https://fortelabs.co/blog/the-body-keeps-the-score-summary/
* Polyvagal Theory
https://en.wikipedia.org/wiki/Polyvagal_theory
... from an "I don't care if it is methodically complete and sufficient, it leads to a workable model for everyday dealings with one's emotional undulations" viewpoint, as in the Polyvagal Ladder:
https://medium.com/age-of-awareness/how-to-use-the-polyvagal-ladder-c68835f0fe19
* The work and perspectives of Dr Gabor Maté
https://youtu.be/H9B5mYfBPlY
My take on the COVID situation (shituation?) is: It sucks, doubly so for people with fragile psyche or a fragile social support network in the first place. But at the same time, while it will present as an external factor the individual can do very little about, at also presents as something with a lot of change and movement - there are bouts of "we're going to be stuck in this forever", but then there's with some regularity some change happening that again directly affects everyone's everyday life to some degree, so it doesn't feel completely stuck. There continue to be 'surprises', one prime antidote to the cPTSD stuck loop.
Of course, I have a lot of perception bias with all this - once you have a hammer, every problem starts to look like a nail. Going through town, interacting with people, I see trauma and fragmentation and hints of cPTSD everywhere.
An excellent piece.
The strange disparity between black and white suicide rates is easy to explain:
Whites couldn't get a moments peace to off themselves because everybody was cooped up at home.
Blacks couldn't satisfy the urge to kill in the usual way because, see above, and had to off themselves.
Any questions?
I agree with your conclusion. I had multiple patients state outright that they felt better because everyone else was going through the same thing and they finally felt justified in their negative feelings. Social anxiety patients and OCD patients did especially well (and are now doing worse that things are opening up). Even the borderlines improved. Young women with garden-variety depression and anxiety did especially poorly. There was a lot of talk about wasted time in finding a mate and falling behind where they perceived they should be in their lives. Also mothers did terribly for all the obvious reasons. I got a lot of new patients who had no history of previous depression in these categories. Maybe these are reasons why the dip is not reflected as much for women.
I would ask that you research and comment on depression and suicide rates in children and teens. My mother patients frequently expressed that their child was going through depression for the first time. My nanny said over 1/2 of the students in her daughter's class were depressed and the guidance counsellors were reaching out to all students for wellness checks. The first article that comes up from the APA suggests high increases in suicide attempts among teens.
"Results showed recent suicidal ideation was 1.60 and 1.45 times higher in March and July 2020, respectively, than in March and July 2019. Odds of recent suicide attempt were 1.58, 2.34, 1.75 and 1.77 times higher in February, March, April and July 2020 than the same months in 2019, respectively."
Good post. I think it's the end in sight and reduced depression/anxiety in 'emergency' type situations.
similar to your theory that some suicides were still happening but masked/categorized as accidental overdoses, I wonder if the they same thing happened with traffic deaths? Because those were up quite a bit in the US. Which is at first counterintuitive, since there were so many less miles driven in 2020. The explanation I've heard, is that with less traffic, people were able to drive faster, and thus deadlier. Makes sense. And given the drop in suicides, I wonder if also some of that fast/reckless driving was suicidal
Hell is other people. Not having to deal with other people has improved my mood. Surely that explains a lot?
There’s one point that feels missing from the final speculation section. Suicide is inherently an act of hopelessness, which you hit on with the Aaron beck “this is endless” point. On the other hand, the pandemic was expected to end, as interminable as it felt in any given moment. When people have a point of light in the horizon, something to really hang hope of improvement on for the long term, it can provide so much fuel to continue because people can plausibly believe life will improve when the bad situation ends.
As someone who lives with someone who is chronically depressed (who, btw, has never felt as comfortable or as satisfied with life as she has during the pandemic), the "this is endless" dynamic I see as a substantial contributor to depressing leading to suicide, and conversely why pandemic depression doesn't lead to as many suicides.
Depression due to the pandemic was a different kind of psychic pain... it was a shared pain, and a pain that by definition has some kind of begin-and-end time period. And, importantly, pandemic depression has an identifiable reason.
I know we can't do exit interviews, but I do wonder if suicides from depression are driven more by the dual factors of not knowing why they feel this way (nothing to point to) and therefore grow convinced that there's no hope in ever getting relief.
Maybe potential suicides were covered by excess mortality eg people who had no will to live succumbed to Corona more by seeking help late and responding much worse to placebo etc.
That would entirely hide suicides in all countries that had excess mortality but not in the ones that had not. Someone looked it up?
I've skimmed through all the comments here but didn't see my pet theory offered as an explanation: misery loves company.
In a normal year you are expected to earn a living, maintain a social circle and have fun hobbies. If you are seriously depressed and failing (or just believe you're failing) to live up to those expectations, you feel like a major fuck-up.
Contrast that to our COVID year. Lost your job? So has everyone else, to the extent that the government has started handing out mass relief without the usual judgmental chatter about the lazy poors. Don't have any friends? Everyone's stuck inside, so it's not like you're missing out. No fun hobbies? Anything worth doing has been cancelled, anyway.
Life may have gotten worse for most people, so depression rates have gone up. But if you're _already_ feeling useless or socially isolated, lockdowns don't change your situation! And at least now there's no pressure to act like everything's fine!
> I won’t waste your time listing the evidence that life got more depressing
I don't know, life got *significantly* better for me, being able to work from home. I know my experience doesn't generalize, but it also doesn't seem like an insignificant factor across the board. Increased unemployment too, applying especially to those with the most shitty jobs.
There's also some thing about covid representing a tangible threat (vs the malaise and ennui of modern life) while at the same time not actually hurting all that many people.
In effect, many ended up safe and comfy at home, and never got sick.
What if COVID and suicides are competing hazards and strongly so for a certain subset of people?
Two of those correlates are age and trait:anxiety. Both are positively correlated (age much stronger than anxiety and anxiety is early research, still seems to be valid and has a large effect) with worse covid outcomes - hospital stays, long covid, death and with suicide in other years.
May be covid took away more of the people who would have otherwise committed suicide.
@Scott: some comments on your Covid suicide article on work in progress (https://worksinprogress.co/issue/why-didnt-suicides-rise-during-covid/)
1) why most suicides in spring? My understanding (personal theory) is this: human bodies do have a built in hibernation program as many other species. This program is controlled by daily sunlight exposure. In preparation for the coming winter our serotonine level and our drive decreases, we eat more and build up fat storage to survive hard winter. Reduced drive (depression?) helped our ancestors (both human and animal) to look for a cave and just sleep there over the winter with least fuel consumption possible. In spring light exposure increases and with it serotonine levels. Drive (and sex drive) and motivation come back. In some individuals the depression stays longer and then they suddenly have the drive to kill themselves. Starting giving Antidepressants (SSRI) is always combined with the warning to tightly monitor the patient the first weeks so that it not kills itself due to this process.
2) why less depressed after a natural crisis or desaster? You mentioned "sense of meaning" and I guess this could be a key. Why there are so many people depressed and suicidal in "happy countries" (I guess you mean western wealthy countries)? Because these sociaties removed a lot of meaning of the lives of their people which were the norm for millions of years for our ancestors: tight daily human connections and interactions with close and loose family members, purposeful activities all the day around like feeding the family, the children, growing food, hunting food, building purposeful things like houses, tools, everything else, singing and dancing together and so on. A lot of these things are gone in the everyday lifes of a lot of western people. Instead they often life alone in anonymous apartments, doing a long daily commute to stupid companies sitting in boring and stupid meetings, producing useless things or services ("what would happen to the world if I, my job or this company would suddenly not exist anymore - nothing - nobody would take notice"). So they feel depressed. Now let's happen a natural desaster Like a flood or a hurricane or a pandemic. Suddenly all the stupid things are not porta to anymore but to help each other. No cracy meetings anymore, a lot of cooperation with formerly strangers from the neighborhood, suddenly a lot of purpose, humankind, success, self realisation asf.
I think it could have to do with jobs. Richer individuals got to work from home while poorer ones were stuck going into work during the pandemic. Race and job/income correlate such that you might see more suicides among essential workers than people at home. When you're constantly surrounded by family and never leave the house you may be depressed but find it harder to plan and execute a suicide.