Restigmatizing Mental Illness

I was reading a Reddit thread for, by, and about Zoomers, and how the defining trait of their generation is anxiety and depression. The thread quickly derailed into the typical polarization of factions: group A saying GenZ has got the short end of the stick and how important it is for society to treat mental illness more seriously and compassionately, and group B saying GenZ has just fetishized depression and society is enabling this by being too indulgent. No group C, and nothing in the middle. This here is my attempt at the nonexistent group C post. My start point is to ask, what is the best way to think about the fact that more young people are now depressed than ever before, at a time where cultural acceptance of mental illness is higher than ever before? Is that correlation significant? Are these below the only 5 possible mutually exclusive options to make sense of this statistic?

1. Negative Spirals: It isn’t true that cultural acceptance is higher than ever before. If anything, cultural acceptance is lower, the stigma is higher and the costs of depression are more severe, negatively reinforcing people handicapped by this illness, which handicaps them even more, which causes more negative reinforcement etc etc, a horrible trap they can’t escape from. It’s true that awareness of depression/anxiety is higher than before, but does that automatically translate to lower stigma? Would I have been able to come out as depressed as a 16yr old in the 2000s? Not likely, society is more ready to talk about this, and it’s going to be an uphill task to square this readiness to talk with a simultaneous increase of stigma. Or even if we tortuously maintain this cognitively dissonant claim, we’d expect the increased stigma to disincentivize further self-reporting of depression and cases should drop off a cliff and go into hiding. Still, it’s not a 0% claim, if nothing else we could segment stigma according to different cultures, socioeconomic bands, countries, ages, institutions etc and find that in some of them at least the stigma has increased. Confidence: 5%
2. Incidence vs Reporting: The same proportion of young people are depressed, we hear about their stories more now because they come forward. Either more of them are coming out and acknowledging their illness, or the same number of them are coming out but much louder. Either way, this reframes the issue slightly from one of increasing depression instead to increasing importance of depression. This is joint best case scenario for destigmatizing depression, and bears out at least a little in how many middle-aged people are now coming out with stories of how they’ve been depressed all their lives but have only now been able to talk about it. Unfortunately, this is anecdotal and by definition never going to be statistically significant to a level that bridges the gap between today’s numbers and those from decades ago. Especially when you have even more old people grumbling about snowflake youth being too fragile these days, which shows that the stigma is lowered but doesn’t say too much about whether that’s responsible for better uncovering of concealed cases. Still, I’m partial to this claim, and in general we’d be better off if it were the null hypothesis to be assumed until proven wrong, as opposed to the other way round. Confidence: 40%
3. Reporting vs Incidence: The same proportion of young people are depressed, but non-depressed people are now self-diagnosing and artificially inflating the numbers. This is the joint worst case scenario for destigmatizing depression, where society actively loses out on human potential. If this were true, some data we might see: no increase in serious depression cases (hospitalization, chronic medication), no increase in suicide rate despite increased depression rate (or maybe we’re just better at treating, mitigating its consequences). Ultimately, this claim is weakened by its counterfactual nature of causality, which is what differentiates it from claim 2 and makes the arrow of causality run in the oppose direction. Ultimately it’s a trade-off game where we accept this loss if it means a higher number of real-cases being uncovered in claim 2. Confidence: 10%
4. Apples to Oranges: More young people are depressed. The environment is more depressing now, if the Boomers were growing up today, they’d be depressed too. The other part of the joint best case scenario, the first part was prevention (catch it early, get more people to report quickly and get treatment), and this is the cure part, where conditions are terrible but thankfully we have treatment and cultural support otherwise the situation could have been much worse. Most macroeconomic indicators support this claim, from housing prices to student debt to air pollution to quality and cost of living indices, life is more competitive and harder in many ways today. Plus, this is the first generation that grew up with social media, that favorite whipping boy for all social maladies we don’t want to take responsibility for. Unfortunately life is also easier in many more ways, prosperity is at an all-time high, access to goods and services is incomparable, and political stability is unprecedented, so it’s difficult to be certain about this claim. Confidence: 40%
5. Snowflakes: More young people are depressed. They shouldn’t be. This is the mollycoddle argument, the favorite argument of the ideological Right, which, wholly unintentional I’m sure, is actually somehow more charitable than claim no.3 where we’re simply calling depressed people frauds. At least here the claim is they are actually suffering from physical symptoms, but given the strongly psychosomatic nature of this illness, they did it to themselves, or had it done to them through weak incompetent parents and adult role models during developmental stages. The other half of the joint worst case scenario for destigmatizing, in fact this one is even worse because if there is an element of causality involved it actually decreases physical well-being and experience, as opposed to just causing erroneous self-reporting and some minor negative externalities like claim 3. If we accept that a larger proportion of young people are depressed, then any uncertainty we have about claim 4 (it is environmental) spills over to more certainty about this claim 5. On the flip side, it’s also really hard to pull apart this claim from claim 4, where does environmental effect end and personal responsibility begin? If rich people are more depressed than poor people, does that automatically make them snowflakes because the environment is objectively better? I’m not at all certain, yet this claim needs to be its own, because personal responsibility does start somewhere, though that dividing line is nebulous. Confidence: 5%

The obvious answer is that claims 2 and 4 account for the entirety of increased depression and anxiety. As long as this is true, the case for awareness and acceptance are crystal clear. The problem is when these two claims start becoming less dominant. At what equilibrium between these 5 buckets would we start becoming uneasy about the unquestionable social utility of awareness and acceptance? It seems to me that the current answer is that there is no such equilibrium at all. That Claims 2 and 4 are all that exist, and everything else is regressive or cynical. To be fair, cases 3 and 5 are for the most part put forward by regressive cynics or stodgy conservatives, but that doesn’t mean the claims themselves possess such personalities. My confidence in these 2 claims are low, but their impacts in terms of deadweight loss is high, so the expected value even with low probability is large enough to merit study, if only to falsify the claims. Deadweight loss means there is loss to both the individual as well as to society, not some tradeoff that we’re having to weigh. The loss to society is that of skewed incentives. Some views of skewed incentives are idiotic, like corporate wisdom that more relaxed sick leave policies create incentives for employees to take more days off (a. falsely and b. unintentionally genuinely as illnesses are psychosomatic and people can psych themselves into actually becoming ill more often), despite the fact that there seems to be zero research backing this up. Generous sick leave policies end up increasing productivity and costing employers less, but it’s easy to see why the assumption of skewed incentives exist. Rubbishing the assumption requires research. There is similarly an assumption that indulging people with their depressive symptoms creates a skewed incentive. This is rubbish, but there’s research to prove that, and does a better job of proving it’s rubbish than insulting people for being stodgy regressive conservative cynics.

Proving it is rubbish isn’t a matter of telling them skewed incentives don’t exist, it’s showing that we’ve incorporated them into the model and absorbed the costs. We aren’t trying to prove that 0 people will take advantage of the sick-leave policy and reduce their output, just like we aren’t trying to prove that claim 2 will have 0% probability. There is leakage, and that is fine. Any act of compassion is vulnerable to abuse, and as a society we form mental models about the % of pilferage and misuse we’re willing to accept for any social program. In an advanced societal setup, the % of misuse we’re willing to accept will be extremely high. Today we have something like this when the stakes are literally life and death, which is why we have presumption of innocence and especially difficult processes to overcome before a prosecutor can push through a death sentence, even in a better-than-average society, we’re willing to tolerate a high % of guilty people going free as long as an innocent man being convicted is as close to zero as possible. A measure of how far we’ve come is the smallest possible stake for which we make such a generous and forgiving calculus. Unemployment insurance for instance, is kept low enough that it doesn’t incentivize workers to sit at home and collect cheques. The fact that there is unemployment insurance is of course a sign of a better-than-average society, but we’re very uncomfortable with the idea of such a program being misused by X people, meant for the Y people deserving of our welfare net, even if Y >>> X.

In such a context, a truly enlightened society would extend the compassion of acceptance and active aid to the depressed/anxious even if claims 3 and 5 far outnumber 2 and 4. As long as claims 2 and 4 are non-zero, there is no level of misuse (3, 5) that would make us change our minds about the utility of compassion despite its demonstrably skewed incentives. This is a society I’d like to live in, so I’m happy to leave it at that, as long as one condition is met, minimizing individual loss. No surprise obviously but still hilarious that I, shamelessly rabid libertarian, have taken such a complex multi-faceted social issue, stripped away everything social about it, and reduced it to a single issue, of individual utility. The argument is simple, for anything to be ‘compassion’, I need to better off under that system than the one that existed before. As someone with depression, I must be better off under a system of cultural acceptance and aid for mental illness, if this is true than the system is compassionate. The non-stodgy non-regressive interpretation of claims 3 and 5 are to consider the possibility that actual wellbeing was better when there was more stigma against mental illness. Again, I’m surprised to see the conflation between the idea that such an idea doesn’t exist and such an idea exists with negligible probability and incidence. I only hear the former, and that is just asking for trouble from a gleeful conservative right, which can immediately point out that the nocebo effect is real (expecting depressive symptoms causes them), that the brain makes real what it pays attention to (law of attraction), and that a sense of control and agency have a higher effect on improving symptoms than anything else. That implies that a silver-lining approach is to say even if claim 5 is grossly factually incorrect, it would be better to behave as if it were true, and that my life is objectively better if I reframed things outside my control as being within my control.

This creates an uncomfortable cognitive dissonance for me, because I can’t disagree with that. I further have a general dissatisfaction with the concept of awareness campaigns that lead nowhere. In a classical marketing funnel, awareness is the base of the pyramid, but it’s pointless unless it leads to all the other levels, trial, repurchase, and loyalty. With issues so much more consequential than selling toothpaste, why do our marketing efforts stop with awareness? If 90% of sales comes from 10% of the aware-level of my pyramid, my biggest task is increasing conversion, not increasing awareness. Social awareness of depression has been touted as the first and most important step, and yet the nature o that awareness hasn’t been wholly devoid of self-defeating elements. There is such a huge difference between the way Jordan Peterson talks about depression (how I should understand it, how I should organize my thoughts about it, and what I should actually do about it) and how an awareness-builder like Deepika Padukone talks about it (how others should understand me, how they should treat me, and what they should actually do about it). Maybe there’s a role for both, that the former is for the individual and the latter is for broader society, but why don’t these mass campaigns that reach so many people, and are ostensibly supposed to be directly talking to those affected touch upon anything regarding the individual? The problem of destigmatizing depression is that it is considered an end in itself, that the stigma is the problem and not depression. There’s obviously a difference between saying it’s been done incompetently, and that it shouldn’t have been done at all, which is the misleading provocative title of this post, only a provocative way to reduce the primacy of destigmatization in the fight against mental illness. This is, sadly, not at all obvious, not even to me.

Why are a 5th of Americans on prescription psychiatric medication? Why is the psychology profession enjoying a boom it’s never come close to, while the % of patients with manageable or improving symptoms does not rise commensurately? My covenant is between me and my therapist, stigma has nothing to do with it, I simply want to get better, therapy/meds either work or they don’t, and they’re both being studied and reported on with the appropriate urgency. But something else is being studied too, without any reporting at all. I’m seeing so many researches correlating mood disorders with creative output, and whether I believe in the Sylvia Plath Hemingway tortured poet stereotype or not, I’ve experienced far stronger emotion, creative drive, and for some weird reason, empathy, when I’ve been most down in the dumps. Depression impacts the way we process visual information, focusing on more intricate details. It lets us divert resources to seemingly higher-return activities, increases concentration on specific tasks, and seems to make us a better judge of people. It could even make us more rational, suppressing biases like fundamental attribution error. Depressed people also perform better in sequential decision tasks. Some really sketchy study also seems to say it helps fight off some infections, which I can make absolutely no sense of given I’ve always heard that it suppresses the immune system. I’ve had to actively seek these out. In all the thousands of pieces of content I’ve consumed either directly or obliquely addressing depression, I’ve never seen a single one address these positives, I’ve never seen one that didn’t pathologize it. They aren’t inaccurate, merely incomplete, which is more insidious than inaccuracy because it’s harder to detect. After thousands of examples, incompleteness becomes inaccuracy. Our approach currently is to say increasing awareness will make the depression problem better, and when it gets worse we say good it’s working we need to increase awareness even more. This makes sense to me, but it’s also what we might say if it wasn’t actually working. What data would we need to see that might convince us destigmatization has failed in its execution while remaining noble in its intent? To say that is a hypothesis that can never be tested is just lazy given the scale of the depression epidemic today, a view that smacks very ironically of not taking depression seriously as an illness. What society could use right now is a stigma against lazy assumptions in a world that is so terribly neat in its dichotomy between thinkers and doers.