Today's Food for Thought: The Depression Causality Problem

This post stems from a thought I had the other day while perusing tumblr and seeing all the “you have flawed brain chemistry, not character” posts in support of depression sufferers. I had to ask myself, is this true? And the answer is not necessarily. But before you get angry at me, let’s discuss this further.

Read More

Just skimmed this, but I mean, I agree (as most of my followers know by now). The known biological components of most mental disorders are vastly inflated, meaning, we know far less than we think we do, and the importance we assign to biological variables without sufficient evidence to do so is kind of overwhelming. Of course biological correlates exist (I mean, biological correlates exist for everything you do), but the emphasis placed on said correlates, which are largely unknown, is astounding.

All this said, it’s a shame you have to be cautious to post something like this. I totally understand why, and I’ve had my posts misconstrued in serious ways in the past, but it’s a shame we can’t be scientifically honest. It’s like this construct has been built around psychological disorders that relies on biomedical explanations, and the moment you question that, it’s as though you are questioning other things (moral judgments about people, evidence behind a diagnosis’ legitimacy, etc.)

But is it though?

Depression is an illness not a moral failing

The first four statements of that image are entirely accurate. The last one is not.

Spreading the idea that depression is caused by neurobiological factors increases negative stigma against mental health sufferers. There isn't even close to being enough evidence to suggest serotonin plays any kind of role in the development or maintenance of depression at all.

See here.

cognitivedefusion reblogged your post and added:

There isn’t any formal experimental/scientific validity in the studies so far, however. You’re right - the experimental and control groups have no baseline for how suicidal the people in the trials are, but it’s not like antidepressants are going to cause a non-suicidal person to all of a sudden become suicidal (from the research that we know of so far).

-Ali xx

I don't know what you're referring to when you say "experimental/scientific validity," because the study I just referenced has lots of scientific validity. I'm not sure if you're familiar with the TADS, but I would look it up before discounting it. Ironically, it's the study that has commonly been used to challenge the FDA's blackbox label against SSRIs, proposing that they are safe for adolescents. It's considered to be a very reputable study.

I think the last statement you make is risky ("it’s not like antidepressants are going to cause a non-suicidal person to all of a sudden become suicidal").

First off: we don't know the truth to that. Whether it makes them suicidal or reduces inhibition/increases risk-taking behaviors, it may still increase risk of suicidality. As such, to me, the mechanism isn't all that important when you're dealing with that kind of effect. Meaning, I don't really care if it causes suicidal thoughts where there were none, or it's enhancing willingness to engage in suicidal behaviors; the effect is the same. Basically, I don't think comforting the family members of suicide victims by explaining "well, the medications didn't create suicidal thoughts, they just increased the willingness to follow through" is going to do much of anything; it still increases risk for suicide, period.

Anyways, I would consider looking more into this issue. The study I referenced has very intriguing data. I will summarize some of the relevant pieces below:

Just a short article about a thing my therapist and I talked about today: why antidepressants come with the “antidepressants might cause an increased risk for suicide”. This statement is a little misleading.

Here’s the thing. Many people with depression have...

Actually there is quasi-experimental evidence to suggest antidepressants could possibly have a causational role. The data is within the TADS study, though the official suicide data was not released upon initial publication. It showed up a few years later.

Fight stigma

I agree with the conclusion but not the premise. Illness implies something malfunctioning in the body. The notion that a depressive state is inherently a malfunctioning one is really a philosophical notion rather than a medical/empirical one, really stemming from a society which has a tendency to promote hedonism. None of this is to say depression is fun, or that those struggling with it deserve blame or are not worthy of sympathy, but just to say that its etiology is categorically distinct from the medical conditions described, and from most things we consider physical illnesses.

Ironically the notion that depression is an illness or disease only increases negative stigma rather than reducing it. Additionally, treating depression in this manner (i.e., fighting against it as though it were a cancer) does not seem to yield very good outcomes.

So, why do mental disorders have to be illnesses at all? Why can't we promote them for what they are rather than trying to make them conform to social standards of what constitutes something as worthy of sympathy? Like if it's not a physical illness then it's not worth feeling sorry over? I vote we change that rather than trying to fit mental disorders into a nice little misleading box just so ignorant people can better digest it.

Approve! There’s a very simple system between Tetyana and I: If we agree, there’s a 99.8% chance that what we’re agreeing on is correct. Therefore, everything she said is irrevocably true.

See:

France, C. M., Lysaker, P. H., & Robinson, R. P. (2007). The “chemical imbalance” explanation for depression: Origins, lay endorsement, and clinical implications. Professional Psychology: Research and Practice, 38(4), 411-420.

You can add or deplete serotonin to both animals or humans, those with histories of depression or not, and see zero change in mood. So: “chemical imbalance” is simplistic, and in many cases, serotonin may not be too relevant to depression at all. Even as correlations.

You’ve always got to be a bit sceptical about psychological and psychiatric studies reported in newspapers. This suggests that the deprived and troubled backgrounds of many rappers and the way this is reflected in their lyrics can make their music therapeutic for some people with depression.

Agreed about being skeptical! Based on what we know of long-term depression treatment, what can’t treat depression? The mere passage of time is still the strongest and most consistent factor in positive depression outcome, so…

Meaning: lollipops are about as effective as hip-hop would be, and about as effective as many psychotherapies currently are. It seems to be the case that you could write up a study and replace hip-hop with any activity on Earth and find results.

We need to do better in trying to identify new mechanisms of depression and how better to address them. The symptom reduction paradigm is not working, despite decades of trying.

I’m skeptical as well, but in terms of hip hop equating to just about any other activity seems a little too general. While saying that hip hop can help treat depression is, in itself, an incredibly broad statement (given the many varieties that can be found in the genre), I wouldn’t count it out as a more specific way to help those with depression. 

As an art therapist in training, I believe that the value of symptom reduction is just as important to treating the depression itself. Treating the symptom allows the client to balance and stabilize themselves so that the treatment of depression can begin. Many clients can’t jump straight into treatment, and having symptom-reduction techniques can help in day-to-day activities when you don’t always have a therapist or other helper easily on hand. The idea of using hip hop is also interesting in that we are seeing a broader cultural approach to the treatment of depression. (Not being in touch with cultural roots can be deadly for some populations.)

But, yeah, this can definitely be seen as one of those dangerous “self-treatments” that the general public uses without really knowing what’s going on. I worry about articles like these being mainstream, nothing but a headline repeated over and over without any actual research data being offered, just some inspiring quotes. Still, I’d love to get my hands on those Cambridge articles.

I think it'd be difficult to find anything in the literature that suggests symptom reduction is as important as addressing the underlying mechanism for virtually any disorder, maybe save for tangible suicidality where immediate safety is the most pressing concern. Indeed, many individuals who experience major depression do attempt suicide, but a large majority does not. But much research would tell you that the presence of symptoms does not predict quality of life once you control for other variables (e.g., what the individual does in the presence of symptoms).

That is, the most important and harmful function of psychological symptoms may not be the symptoms themselves but how they confine or constrict an individual's life. This just being based off basic mediation analyses.

Aside from that, however, symptom reduction could very well interfere with practicing the ability to engage meaningfully with life even in the presence of depressive symptoms. That is, from a behavioral standpoint, I would hate to let a client leave therapy under the belief that they cannot do important and meaningful things so long as they are feeling depressed. Unfortunately, that's often a prime message in symptom reduction paradigms. I.e., "we need to get you feeling better."

The converse to this approach is, "we need to get you doing better things," which is more my style (as per more ACT/behavioral activation approaches). That is, rather than lying in bed when feeling depressed, what can we practice doing instead even when you don't feel like it, or assume it will be a worthless endeavor. This is more aligned in working with the depression rather than against it. The former of which, as most anyone who has experienced pervasive depression can tell you, does not work.

Now when you say:

I can agree with that to some extent. Though I would never suggest a client can't jump straight into treatment so much as it is immensely difficult for them to do so. I would also suggest that symptom-reduction techniques can have their place, it's just a matter of cost-benefit analysis. Doing them for the sake of doing them isn't the most functional choice. It can be a pseudo-safety behavior just applied to depression, as some might argue.

But I would elaborate on this and say that symptom reduction is, by itself, not the end-goal of treatment, which is how it's often construed (both by the article linked to this post and by many studies as well). Symptom reduction is much like motivational interviewing or most cognitive therapies: it can be used effectively as a precursor to active treatment, but it is rarely an active treatment in itself. If it's not getting individuals to something more "advanced" (e.g., meaningful behavioral practice), then it's pretty worthless long-term. Reducing symptoms for the sake of reducing symptoms is a weak therapeutic goal, in my mind, and that's my major contention with symptom reduction; when it is treated as the end-all.

Hopefully that makes sense.

You’ve always got to be a bit sceptical about psychological and psychiatric studies reported in newspapers. This suggests that the deprived and troubled backgrounds of many rappers and the way this is reflected in their lyrics can make their music therapeutic for some people with depression.

Agreed about being skeptical! Based on what we know of long-term depression treatment, what can't treat depression? The mere passage of time is still the strongest and most consistent factor in positive depression outcome, so...

Meaning: lollipops are about as effective as hip-hop would be, and about as effective as many psychotherapies currently are. It seems to be the case that you could write up a study and replace hip-hop with any activity on Earth and find results.

We need to do better in trying to identify new mechanisms of depression and how better to address them. The symptom reduction paradigm is not working, despite decades of trying.

I wouldn’t say I’m prejudiced against psychiatrists so much as the system under which American psychiatrists operate (namely, their education). How many [American] medical schools teach statistics/research methods beyond what you would find in a bachelor’s or maybe master’s program? From what I’ve seen, very few (unless students are actively seeking that out on their own through electives). And I think that’s a major issue.

Considering PhDs have enough trouble understanding pretty simple concepts like statistical significance, I’d be surprised if MD’s were any better given they have even less training.

Fair enough. I do think that most people have difficult with statistical concepts. PhDs too. 

I’m not in the US, so I don’t know, but do you have evidence for this? I actually wonder what %age of psychiatrists think that about depression and how it varies by age/when they were trained/where they were trained, etc. 

Probably. 

Power. Capitalism. Inertia. Influence of pharmaceutical industry. I generally always tend to put way more blame on institutions than on individuals. 

In Ontario, psychiatrists are covered by government insurance. I’ve paid nothing out of pocket. Psychologists are not. (I don’t agree with this, and don’t think it should be the case, but that’s how it is right now.) 

Well, we don’t have sufficiently long-term data, do we? But certainly safer than benzos and antipsychotics.

Regarding chemical imbalance hypothesis: I've seen survey estimates range from 50% to 85% - I'll see if I have any saved. Usually I just see that statistic mentioned in studies and cited, but I'm not sure I have any direct sources saved.

Agree regarding blaming institutions, which is why I think the educational system is partly to blame.

Regarding health insurance: definitely depends. My dad used to work in public health in America (now Canada!) and so I used to pick his brain about this stuff. Given, that was when he was living in the states which was around 2010-2011, so much could have changed in that time (especially with health care changing here).

The long-term data is very preliminary because, in my opinion, we've been way myopic in determining drug safety/effectiveness. Doing a 12-month pre/post design does not begin to get at long-term complications, but that's what we've basically relied on. Long-term use for adolescents doesn't look great, from the bit I've seen. I'm not sure about initiation of use in adults. I'll see what I have on that, too.

Research showing that drugs don’t work is less likely to be published. This article reports on a meta-analysis which pooled the results from several smaller studies and shows that, when you take several important factors into consideration, these drugs don’t tend to work. Of course they may work for an individual but that’s offset by the many for whom they don’t work.

It’s also important to know that even for a given individual what may be “working” could be attributed to placebo effect or something that isn’t part of the “active” antidepressant effect.

So, saying “I took antidepressants and then felt better therefore they worked,” could be misleading, because the change in mood could be due to other things and have nothing to do with the proposed mechanisms behind antidepressants.

(Correlation does not immediately infer causation.)

The studies continue to show rusty efficacy of antidepressants is lackluster and always had been. Why then are doctors still prescribing SSRI as a first resort for depressive symptoms?

Because they likely don’t know better, for a myriad of reasons depending on the specific specialization (e.g., pediatrician vs. psychiatrist), and what training they received (e.g., many med school are not training advanced research/statistics like you would receive at a PhD program).

Much of the information that doctors get, for instance, come from a misunderstanding of the literature or pharmaceutical reps.

I think a big reason is that it is what is at their disposal. Maybe it might just work. They can prescribe anti-depressants. They are pretty safe drugs, compared to a helluva lot of other stuff that really probably shouldn’t be prescribed. And so I think when they see a patient that is in need, even though the drugs are sub-par, they might not be for that person, and what else can the psychiatrist really do? Particularly, I’d say, in the US. (My psychiatrist totally does therapy/hates just playing a role of monitoring meds, but that’s a dying breed, unfortunately.)

I don’t think it has much to do with misunderstanding the literature. Is there evidence of this? You are so damn prejudiced against psychiatrists, it is like aarghegd.

I do think pharma reps play a role, for sure, and maybe that’s where some of the misunderstanding is coming from, but we don’t have the same culture of that in Canada, at least as far as I know/from what I understand. 

I really, really, really hate that SSRIs are called antidepressants. It is such a misnomer. 

I wouldn't say I'm prejudiced against psychiatrists so much as the system under which American psychiatrists operate (namely, their education). How many [American] medical schools teach statistics/research methods beyond what you would find in a bachelor's or maybe master's program? From what I've seen, very few (unless students are actively seeking that out on their own through electives). And I think that's a major issue.

Considering PhDs have enough trouble understanding pretty simple concepts like statistical significance, I'd be surprised if MD's were any better given they have even less training.

Loads of American psychiatrists still think depression is a medical problems stemming from a chemical imbalance. And that might be heavily due to the pharmaceutical companies stressing that idea.

Further, why aren't they being taught about psychological treatments? Why aren't they being taught, "hey psychological treatments are better long-term and pose less risk of adverse effects"? Or are they and they are just neglecting it? Either way, it's a problem either within their education or within them, depending on the source.

(Disclaimer: I do not deny the distinct possibility that the individuals they are seeing simply cannot afford psychological treatment, though typically if you have insurance that would cover psychatric drugs then they often cover ~8-16 sessions of therapy, sometimes more.)

I don't think the long-term data would support the idea that most any psychotropic drug is "safe" long-term. Even SSRIs. Maybe safer than others, but I would be hesitant to label it safe.

Research showing that drugs don’t work is less likely to be published. This article reports on a meta-analysis which pooled the results from several smaller studies and shows that, when you take several important factors into consideration, these drugs don’t tend to work. Of course they may work for an individual but that’s offset by the many for whom they don’t work.

It’s also important to know that even for a given individual what may be “working” could be attributed to placebo effect or something that isn’t part of the “active” antidepressant effect.

So, saying “I took antidepressants and then felt better therefore they worked,” could be misleading, because the change in mood could be due to other things and have nothing to do with the proposed mechanisms behind antidepressants.

(Correlation does not immediately infer causation.)

The studies continue to show rusty efficacy of antidepressants is lackluster and always had been. Why then are doctors still prescribing SSRI as a first resort for depressive symptoms?

Because they likely don't know better, for a myriad of reasons depending on the specific specialization (e.g., pediatrician vs. psychiatrist), and what training they received (e.g., many med school are not training advanced research/statistics like you would receive at a PhD program).

Much of the information that doctors get, for instance, come from a misunderstanding of the literature or pharmaceutical reps.

Research showing that drugs don’t work is less likely to be published. This article reports on a meta-analysis which pooled the results from several smaller studies and shows that, when you take several important factors into consideration, these drugs don’t tend to work. Of course they may work for an individual but that’s offset by the many for whom they don’t work.

It's also important to know that even for a given individual what may be "working" could be attributed to placebo effect or something that isn't part of the "active" antidepressant effect.

So, saying "I took antidepressants and then felt better therefore they worked," could be misleading, because the change in mood could be due to other things and have nothing to do with the proposed mechanisms behind antidepressants.

(Correlation does not immediately infer causation.)