Psychiatric Diagnosis Found to Be “Scientifically Meaningless”
technologynetworks.com
technologynetworks.com
> The study, led by researchers from the University of Liverpool, involved a detailed analysis of five key chapters of the latest edition of the widely used Diagnostic and Statistical Manual (DSM), on ‘schizophrenia’, ‘bipolar disorder’, ‘depressive disorders’, ‘anxiety disorders’ and ‘trauma-related disorders’.
So, basically, they read the DSM and gave their opinion of it. Which is fine, but calling this some type of "study" is ludicrous. It's a bit rich to use statements like "the main findings of this research are" when the "research" just literally was Reading The F*ing Manual.
"A detailed examination, analysis, or experiment investigating a subject or phenomenon"
Similarly, "research" is a very broad term and quite clearly can involve just reading without performing any experiments.
I would (less semantically) argue (and the editors of "Psychiatry Research" would seem to agree) that pointing out inconsistent classification criteria is important to understanding the reliability of experiments based on those criteria.
I am pretty clear about what he expected. The question is: on what basis was this expected and is it reasonable to start a semantic debate about it on that basis? What is the "big problem" here?
If I present that opinion as, “Study finds C++ more prone to bugs than other languages,” you don’t find it misleading?
I think most people assume the word “study” indicates something is backed by a rigorous method of research, like the scientific method. You can’t just make an opinion and then claim it’s a study.
I don't think it calling itself a "study" should have any bearing on your desire to see that quantitative evidence or on your general epistemic stance towards the claims.
> I think most people assume the word “study” indicates something is backed by a rigorous method of research, like the scientific method. You can’t just make an opinion and then claim it’s a study.
I get that there may be some connotation in the context of headlines that empirical evidence gathering was involved.
Does that make it ludicrous to use the term? Not at all.
The paper and the article are both clear about what was involved in the study. I really don't see how there is a "big problem" here (especially compared to issues of reproducibility in studies that are back by empirical data.)
My big concern is the assertions that papers like this are not part of a "rigorous method of research, like the scientific method" completely discount the important role that such papers play in drawing the solid conclusions from data.
https://en.wikipedia.org/wiki/Outline_of_academic_discipline...
Psychology (and psychiatry) fall under the social sciences. These do not claim to be value-free. In other words, they draw conclusions because people feel those conclusions to be "needed", or socially relevant. Not even valid, in terms of truth versus fiction.
The problem is that with psychiatry you have quite a few studies that point out it's not empirically valid at all:
1) diagnoses of mental illness are made on factors other than the patient or the symptoms:
(for example they are made on available space, how "necessary" bed & meals are for the patient, whether the referring person needs them to be admitted, ...)
https://en.wikipedia.org/wiki/Rosenhan_experiment
Worse, the same goes for involuntary mental treatment and even youth care. All of those interventions, including everything from limiting study choice to locking a kid up for 2 decades, are based on whether a social worker considers such a thing socially needed. These determinations are made not on the child or person, but on the perceived need from the people asking for such a "diagnosis" and intervention. And that's the normal practice, not cases of abuse (which happen regularly, because of course admitting a patient pays, as well as the need for validation these people have)
So, for instance, and as shown in the experiment, psychiatrists will lock up, place in isolation, medicate and use violence against patients to get them to admit/confess to a diagnosis by an earlier mental health professional. This is the normal practice, and not exceptional.
2) mental health treatment does not actually make patients better, or at least, despite 2 centuries of searching, there is no empirical or statistical evidence beyond "professionals say it works".
Or, more dramatically:
https://www.buzzfeednews.com/article/esmewwang/psychiatric-h...
2b) furthermore effectiveness studies keep, again and again, pointing out that only factors that have nothing to do with mental health theories and nothing to do with mental health treatment matter for the number of afflicted.
Chiefly it is pointed out that mental health clinics tend to get filled. So the number of afflicted is determined by supply of care rather than by patient demand.
Second studies keep pointing out that the relationship with carers matters for outcomes, and that specifically treatment, medicine, duration, ... doesn't matter at all.
Third studies keep pointing out that "care" is used as a coercive social control tool
Fourth when it comes to severity of problems the economic situation matters a lot more than anything the professionals do or know.
3) there are many abuses of the system by mental health professionals, and these have persisted over centuries
"historical" (which should not be taken as "past" in most cases)
https://en.wikipedia.org/wiki/Ten_Days_in_a_Mad-House
https://en.wikipedia.org/wiki/Aktion_T4
https://en.wikipedia.org/wiki/Political_abuse_of_psychiatry_...
https://www.scmp.com/news/world/europe/article/3016435/angel...
https://www.bbc.com/news/health-48367071
https://www.bbc.co.uk/news/resources/idt-sh/norways_hidden_s... (pedofile using state power to kidnap children and place them ... with himself)
https://www.commissiegeweldjeugdzorg.nl/onderzoek/english/
https://en.wikipedia.org/wiki/Rotherham_child_sexual_exploit...
https://www.bbc.com/news/world-europe-49320260
https://www.washingtonpost.com/nation/2019/06/05/an-anguishe...
...
4) truth be told there are many abuses of the system by "patients".
Just visit a mental health facility or psychiatrist and ask about the homeless that want to be taken "into care". I guarantee details will be forthcoming.
Here's a randomized control trial that says otherwise: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4312318/
According to you it would be mere coincidence that lithium appears to work so well to treat the misdiagnosis of bipolar disorder.
Is this a joke ? Of course that's what it does.
Does it help against the actual problem ? No.
As far as I can tell from the article, the "study" is basically arguing that because most psychiatric diagnoses use circumstantial evidence--that is, they don't rely on tools like an MRI to quantitatively identify and classify a neurologically-based defect--that it's not evidence based.
But that doesn't follow. Most of medicine is like psychiatry, at least in the way that that paper seems to criticize. If a person comes into the ER with the symptoms of a cold and you treat them for a cold, you can't say that the diagnosis of a cold wasn't evidence based simply because the doctor didn't run a test to identify rhinovirus fragments. If it's known that the vast majority of such cases would be rhinovirus, that the vast majority of cases for those with something more severe (and where misdiagnosis would be dire) would present with different symptoms, then treating the patient as-if they have rhinovirus is a legitimately evidence-based, experimentally supported activity.
It may not be ideal for each particular patient. The doctor isn't searching for some objective truth about each patient's condition. But so what? We do that because we're trying to maximize population wide outcomes. If we abstained from diagnosing every cold without first identifying viral fragments, many people would not only go undiagnosed but completely untreated, including those with much more severe infections.
If it walks like a duck and quacks like a duck, it just might be a duck. And if research says that it is a duck 90% of the time, then making that inference is honestly and truly evidence based, even though it leaves open the possibility of 10% misdiagnosis rate.
It's also not surprising that the assessment criteria would differ markedly between different types of ailments. The criteria for assessing a sprained ankle wouldn't look anything like that for a cold. It would be highly context based. We can't infer that it's therefore not evidence based.
Psychiatry may or may not be evidence based, but you can't show that by an examination of the DSM. You have to affirmatively show the absence of a connection to evidence. Without doing that, you could only at best make a persuasive argument that the evidence itself sucks, or that the disease models are weak. But, again, a weak disease model doesn't imply lack of evidence. Our existing models of most diseases are going to look incredibly weak 200 years from now; but it wouldn't be reasonable to say that late 20th century medicine was unscientific, non-evidence-based hokum.
Perhaps you would have been better served by reading the linked abstract of the paper.
The paper specifically criticizes the heterogeneous and inconsistent of classification of identical symptoms in the the diagnostic criteria for similar diagnosis.
>> Selected chapters of the DSM-5 were thematically analysed: schizophrenia spectrum and other psychotic disorders; bipolar and related disorders; depressive disorders; anxiety disorders; and trauma- and stressor-related disorders. Themes identified heterogeneity in specific diagnostic criteria, including symptom comparators, duration of difficulties, indicators of severity, and perspective used to assess difficulties.
> If a person comes into the ER with the symptoms of a cold and you treat them for a cold, you can't say that the diagnosis of a cold wasn't evidence based simply because the doctor didn't run a test to identify rhinovirus fragments.
You picked a particularly odd example. There is no treatment for colds, all that can be done is to alleviate the patients symptoms.
Indeed, there a number of situations in which you would indeed want to confirm that the symptoms are caused by the common cold because there are other potential causes that can be life threatening. [0]
[0] https://www.health.com/cold-flu-sinus/signs-more-serious-com...
> And if research says that it is a duck 90% of the time, then making that inference is honestly and truly evidence based, even though it leaves open the possibility of 10% misdiagnosis rate.
... what? A 10% misdiagnosis rate because other possible explanation for symptoms are ignored is considered good science and medicine?
> It's also not surprising that the assessment criteria would differ markedly between different types of ailments. The criteria for assessing a sprained ankle wouldn't look anything like that for a cold.
Except the study explicitly compares similar ailments (more like "sprained ankle" vs "broken ankle" vs "torn ligament in ankle"
The main point of the criticism is that unlike other medical diagnosis, psyciactric diagnosis minimize the role of trauma. This is like a doctor trying to diagnose ankle pain while not really considering the fact that you fell off your bike.
> Pragmatic criteria and difficulties that recur across multiple diagnostic categories offer flexibility for the clinician, but undermine the model of discrete categories of disorder. This nevertheless has implications for the way cause is conceptualised, such as implying that trauma affects only a limited number of diagnoses despite increasing evidence to the contrary.
But neither science nor evidence-based medicine is synonymous with advanced; you can't say something isn't scientific or evidence-based simply because it's not sufficiently advanced. Both are processes, which by definition means they'll always fall short of what could be at any point in time--it takes time for new knowledge to travel.
[1] The fields of medicine and psychiatry are replete with examples of where supposedly revolutionary models and procedures were adopted by overly credulous practitioners, causing much harm and some of the most memorable horrors of recent history. Without better disease models practitioners will be more skeptical of new evidence, as it's more difficult to judge the quality and applicability of the evidence. It doesn't mean they're anti-evidence. Now, maybe these fields, especially psychiatry, are too skeptical and intransigent, or too indecisive, or too confused. It would be more constructive to just say that, rather than coming out swinging with conclusions that don't follow, effectively committing the same sins they're accusing of others.
Psychiatric diagnosis is at the level of pre-scientific symptom-based diagnosis, for the most part. Imagine if you went to the doctor for a cough and, without any tests, they prescribed a month long course of antibiotics with no idea how effective they would be. A friend of mine was prescribed powerful antipsychotics to "fix" a problem that was actually being caused by another psychiatric medication they had been prescribed earlier to fix something else. Their mental health kept getting worse and worse, and what finally helped was starting a DBT (dialectic behavior therapy) program, and now they're in the process of getting off all those meds that turned out to be more harmful than helpful.
We need to stop pretending that a point-scored questionnaire is a scientific way to distinguish a diagnosis, or that we understand the chemistry of the brain well enough that medication can be the first line of defense.
But there's not much you can do if it's inaccurate besides start looking for another provider. I've written a few scripts to massage my provider's directory into something usable, now lost unfortunately.
UnitedHealth is apparently working on a blockchain for provider data: https://cdn2.hubspot.net/hubfs/4801399/18-SYN-001-Synaptic-W... I guess making a web service is too hard? https://www.managedcaremag.com/archives/2019/7/how-blockchai... suggests it's mostly about getting management to sign off.
I decided to seek treatment for ADHD because my inability to focus was having a negative impact on my career and, indirectly, my well being. For context, I've lost more than one job where I've done good work on the interesting parts, but failed to carry my weight on the boring parts to the point where I simply wore out my welcome. I'm trying to start a family and I felt I was beginning to tread a familiar path with my otherwise very understanding and welcoming new boss.
This probably varies per provider, per locale, etc, but for me the process was:
- Used my insurers website to look up psychiatrists in my area. Had issues with less than great data here also and long appointment waiting times for some, just kept calling around
- called and scheduled an appointment
- told the psychiatrist that I believed I had ADHD
- went through a verbal assessment with the psychiatrist
- got blood work and an EKG done
- had my psychiatrist look over these results
- picked up prescription for first 30 days
- come back in for assessments at 3 weeks, 7 weeks, 11 weeks. Can get prescriptions over the phone afterwards
For me it has been life changing. There's a stigma within my family about mental health so for a long time I've suspected I had this problem but just tried to discipline myself and stick to a system instead of asking for help. AIUI I use a relatively small amount (10mg in the morning, then either 5mg or 10mg in the afternoon). The biggest differences for me have been a) I start the work that sounds boring in a timely manner instead of putting it off to the last minute, b) when I dig into these boring items, I find that the stuff I considered boring before actually can be interesting too sometimes - and at the very least solving them correctly gives me a little rush of dopamine, and c) I make fewer mistakes on the boring stuff because I'm not rushing through it under self-induced time pressure just trying to accept the first "looks right" explanation.
I experienced this quite a bit earlier on in my career (and still do, to a certain degree), but I eventually arrived at the realization that this had a lot more to do with anxiety for me than with ADHD. For me, the "boring stuff" was either
1) all the little details and maintenance work that needs to be done to bring a project from complete implementation to complete deployment and rollout and
2) things that someone else could probably do better than me, or that I could ask for help with.
With respect to 1), I began to realized after working through multiple projects that the thing wasn't really done until it had been used in the wild for a while and I had gone through enough iteration cycles of gradual refinement to know it was in a stable state -- that let me launch things a lot more comfortably and be "okay" with the knowledge that it would never really be done, just good enough, and that I could always come back later and improve it more if needed.
With the latter, I used to be a lot worse at asking for help. I found it embarrassing and seemingly a tacit admission of my incompetence. I've also gotten better at that by understanding that productivity gets done through teams, and the more that I ask for help, the more the team is growing and that I'm increasing my own mental context on all the things I otherwise wouldn't know. Asking questions wasn't a sign of my incompetence; it was a sign of my curiosity, and a desire to keep improving.
I'm not sure if any of those strike a chord, but if so, these could be things to think through. As an aside, I found that when I worked in unhealthy work environments, these things came up a lot more because I reported to and worked with folks in a low trust environment. This meant that they'd simplify things to saying that I've done "good work on the interesting parts, but failed to carry my weight on the boring parts" when in all actuality, it was painful to do the parts I didn't do -- not just for me, but for everyone. I was just the squeaky wheel.
I can't fully tell, but from the way you talk about it, it's possible this is the case with your work environments. It's miraculous how much less prevalent these kinds of issues are in a supportive environment where you're working with not just competent but caring and kind team members and leaders.
> I experienced this quite a bit earlier on in my career (and still do, to a certain degree), but I eventually arrived at the realization that this had a lot more to do with anxiety for me than with ADHD.
I was pretty anxious about work, the quality of work I was doing, etc. As I've been improving at execution of the boring parts, this anxiety has evaporated for me. My psychiatrist touched on this briefly - it can be tricky diagnosing people who have characteristics of ADHD and anxiety, because ADHD medication can exacerbate anxiety. You have to try to understand if anxiety is the root of the problem (your anxiety is preventing you from being effective day to day, which sounds like might have been your case) or ancillary (being ineffective day to day is a problem for you, which is also causing you anxiety).
> With the latter, I used to be a lot worse at asking for help. I found it embarrassing and seemingly a tacit admission of my incompetence. I've also gotten better at that by understanding that productivity gets done through teams, and the more that I ask for help, the more the team is growing and that I'm increasing my own context on all the things I otherwise wouldn't know.
This is a good lesson to learn. I tend too much toward trying to understand everything in a system and preferring to understand it from first principles rather than blackboxing it and being satisfied with a high level explanation from someone else. What you say about low trust environments also resonates for some roles I've had in the past, but isn't core to my case imo.
My downfall has usually been stuff like this:
- receive client ticket on thing I know. Work on more interesting tasks until SLA period is almost over, then look into client issue too quickly and give first plausible answer which later comes back wrong (and trivially so)
- rush through a modification to the system I made because it's the end of the sprint and I spent the whole time noodling around with interesting, low priority thing <x> instead. See bug occur in testing on client site, dismiss it as "it's a really bad site though, probably unrelated" because I see it isn't everywhere on their site and I don't take the time to try to understand what is happening. Bug is real, client fires us for money the bug causes them to lose.
"Bad at the boring stuff" isn't the assessment I've been given from HR people or bosses. It's me looking back on where I make mistakes, acknowledging that I have a tendency to be sloppy on the grind work that everybody on a team needs to do but that nobody wants to, and that the extent to which I was sloppy on these sorts of tasks has had serious effects on my career.
When life forced me out of desperation to seek therapy, I went through three therapists before finding the right one for me.
I have been seeing my guy now for several years. Between talk therapy, meds, and a daily meditation practice, my life is substantially better. And, it turns out that my work has improved the lives of my wife and family. I am generally a hell of a lot easier to live with, and am a better approximation to the "sweet old guy" that I want to be as I grow into my later years.
When you are in a weakened psychological state, possibly anxious, fearful, conflict-avoidant, passive-aggressive, terrified of conflict or triggering unmanageable reactions from other people, it can be extremely difficult to terminate a relationship with a therapist. The only way I found was to write a brief note, enclose a check for payment in full for the next session and any other financial obligations incurred or committed to, inform the therapist in writing that I was no longer going to be a patient, and skip the last session.
I'm caring for a family member who's been dealing with serious mental health issues. Hearing voices, not sleeping, crying, screaming and banging and throwing stuff all day and night. It's been tough. I've been attacked twice since they got here.
Not quite two weeks ago I insisted they go see a doctor and get some prescription meds to calm them down and help them sleep at night. The doctor prescribed a generic Prozac and some kind of sedative to take at night.
They're already doing much better. Sleeping all night and no raging. Still crying a bit, but not near as much, and finally a bit more receptive to the behavior therapy I've been offering. They've been seeing a behavioral therapist once a week for a month now and haven't been very receptive to what they're doing but I think showing them some info on how "dialectical behavior therapy techniques" work might encourage them to engage with their therapist more.
So, thank you for sharing that info. I'll be spending some time learning more about that myself now.
> The doctor prescribed a generic Prozac and some kind of sedative to take at night.
I wonder, would this sedative actually be an anti-psychotic drug? As someone who has a family member with mental illness, these symptoms seem like something that would warrant such medication.
We use them more as a communication tool to give a rough picture of a patient’s symptoms to other providers than as a map onto biology. We simply don’t understand the underlying biology well enough yet to do any better. But there’s a lot of exciting research cracking this area open and we may very well be able to diagnose and prognose based on objective data such as brain imaging, genetics and blood tests.
Psych resident
As a psychiatric professional, you may find it reassuring to predict the future dominance of an objective medical model of mental illness. It would be braver to admit that the complexity of the mind and of the circumstances that lead to mental illness mean that medical research is a very long way from cracking this problem.
Yes, but the drugs actually work, which validates the prediction.
> Mental illnesses, in general, don't develop organically from biological defects.
In schizophrenic patients there are clear organic changes in the brain, you can see it on brain scans.
> These things come about because of profoundly dysfunctional psychosocial environments.
Then it would surely be possible to "fix" these issues without pharmacology treatment, which is unmistakably not the case.
When someone has psychosis (eg hearing voices), maybe it was caused by a severe trauma or maybe it was mostly genetic. In either case the person is suffering from psychosis and is unable to normally function in society (if severe enough). In both cases, in principle, we should be able to detect the problem at the level of the brain in structure and function, and should be able to intervene at the level of the brain.
The intervention required depends on the nature of the brain change. We can change brain function in many cases by intensive therapy. In some cases sensory inputs alone (via talk therapy) are insufficient to change the brain to rectify an issue. That’s when we have to consider direct intervention on the brain itself either via medications or neuromodulation technology.
I don’t think lay people understand how profoundly mentally ill some people are and that they can really have dramatic improvements with treatment. Lay people think of psychiatrists just peddling drugs to mildly depressed patients or people who are mildly oddly behaved. That’s just not the case. I’m working with people who sometimes literally stop eating because they think their food is poisoned, or attempt to kill themselves with minor frustrations, etc. And many patients, mostly not that sick, come in voluntarily because they feel they can’t handle some problem on their own. You’re right that much I mental illness is caused by psychosocial stressors, but some people have just intractably messed up life situations and they’ve just spiraled downward as a result and wouldn’t be able to get on a better path without treatment.
No psychiatrist believes this.
I agree that they are used as heuristics - and indeed were developed to be heuristics, dating back to the RDC.
Psych attending
The exciting research I care about is more along the lines of < https://www.ncbi.nlm.nih.gov/pubmed/30696271 > where we observe neural dynamics, hypothesize causality, intervene with TMS or other focal neuromodulatory tech, and see if in fact region A is causally linked to outcome B, or if neural dynamical state A is causally linked to outcome B.
I also meant that the DSM is biologically unreliable because it is only symptom clusters, and we cannot intervene at the level of symptoms, you have to intervene at the level of pathology. Symptoms are not pathology. It's the brain that is experiencing some pathology, so we need to ascertain the brain pathology and intervene at that level, and DSM, as noted by brain imaging studies, does not seem to reliably map onto that biology.
Aside, are you involved in software engineering? What brings you to HN?
Abstraction is as powerful in nature as it is in software. DNA is an amazing abstraction that essentially precludes physics from having a vote in the information content of the genes, and relies instead on an entirely new logic at a higher abstraction level, namely evolution. Grasping for other metaphors, there is a profound complementarity between a great violin and a transcendent performance, but nonetheless Stradivarius and Heifetz were geniuses in different categories. Steinway and Horowitz. Without a piano there is no Horowitz, and everything Horowitz does is translated via the piano and physics to the audience. But, the music is at a qualitatively different abstraction level than the physical piano.
So, my current (interested layperson's) view: The human mind and the human brain are complementary yet distinct phenomena. While every mind-event has a physical representation in the brain, and without doubt many physical abnormalities in the brain can manifest in the mind and can therefore be treated by treating the brain, working reductively up from the brain toward the mind as the only acceptable path for treating all mental patients will IMHO eventually run afoul of an abstraction barrier similar to the one physics runs into at the level of DNA.
https://ghr.nlm.nih.gov/condition/williams-syndrome#genes
Are we close to being able to do this for conditions like depression, schizophrenia, autism... not so much.
Sure, the DSM is basically a dumping ground for various obnoxiously imprecise syndromes that don't clearly belong to another specialty such as neurology or endocrinology. But that doesn't make those syndromes "scientifically meaningless". It's one thing to say that these definitions should not be taken as gospel for determining future research directions. It's something else entirely to say that they're "meaningless" and "not fit for purpose" based on a philosophical notion of "meaning" or "purpose" that itself isn't rigorously defined.
[1] It's funny how "biopsychiatry" is expected to identify precise molecular pathways or anatomical derangements or genetic markers, but once the paradigm is flipped to the more psychodynamic approaches, rigor suddenly doesn't matter so long as we can tell a compelling story about how our experiences shaped us.
> Almost all diagnoses mask the role of trauma and adverse events...
> The authors conclude that diagnostic labelling represents ‘a disingenuous categorical system’.
The DSM fails miserably as a manual to alleviate suffering, because it downright ignores the influence of trauma on all but one major diagnosis (acute PTSD).
Worse, complex trauma ("cPTSD") aka "developmental trauma" is still not an official diagnosis, which is absurd, given that it is well established in the research that child abuse can severely damage mental health.
The whole thing is looking at the problem of mental health from the wrong angle. The so-called epidemic of mental illness will only subside once we ditch the dizzying array of overlapping symptoms, and focus on the root causes.
Every single individual seeking professional counsel on mental health should be screened for trauma at intake. There's just no excuse for not doing that given how much trauma impacts outcomes.
• Psychiatric diagnoses all use different decision-making rules
• There is a huge amount of overlap in symptoms between diagnoses
• Almost all diagnoses mask the role of trauma and adverse events
• Diagnoses tell us little about the individual patient and what treatment they need
> • Psychiatric diagnoses all use different decision-making rules
> • There is a huge amount of overlap in symptoms between diagnoses
> • Almost all diagnoses mask the role of trauma and adverse events
> • Diagnoses tell us little about the individual patient and what treatment they need
I think the alluring thing about diagnoses is it quickly gives closure, so we don't have to think, balance facts, or ask why. Check some boxes and we can draw a type of a person by characteristics as we would with the periodic table of elements.
We've been raised in a culture (or multiple of them), have an upbringing and have a story behind that impacts how we trust others, we need to be loved, have complicated emotions, we have impulses and drives, and sometimes there is a biological/chemical aspect to it, we're made of elements.
As a society, we're here now cooperating because we care to help each other. We're complex social beings, and it's hard to put us to a definition and have it do us justice.
One of of the big papers criticizing mental illness is this: https://psychclassics.yorku.ca/Szasz/myth.htm
Another one is Rosenham Experiment: https://en.wikipedia.org/wiki/Rosenhan_experiment
https://youtu.be/1MbARnJqjnc?t=519 - Some podcast talk on diagnosing mental illness / diagnosis and how not all clinicians would come to the same agreement on someone's behavior.
1. Biochemical
2. Trauma, often made worse by social BS
3. General social BS
4. A mismatch between the person and expectations placed upon them
I think the first is something we should work on a great deal more than we really do. It's much more readily quantifiable in a scientific manner than social BS and addressing it can dramatically reduce the impact of the other three, all of which entail varying degrees of social BS.
For example, this conference on the Power Threat Meaning Framework (and alternative to diagnosis) is charging huge amounts of money for their conferences.