[skissane]

[1] argues that ASD lacks validity – in other words, "ASD" is not something which actually exists in nature, it is simply a largely arbitrary grouping of symptoms, and those symptoms have a diverse array of unrelated causes. Given this, [1] argues that the use of the ASD diagnosis as a basis for research should be abandoned. Instead, researchers should look for the causes and associations of each individual symptom included in the ASD diagnostic criteria, irrespective of whether the individuals having that symptom have been diagnosed with ASD or not, and irrespective of whether they meet the overall diagnostic criteria for ASD or not. If [1] is right that the validity of ASD is dubious, then any research based on the diagnosis of ASD is also dubious, this study included.

[1] https://link.springer.com/article/10.1007/s40489-016-0085-x

[lazyguy2]

The same can be said of probably most disorders listed in the DSM.

Mental disorder diagnostics are all 'Design by committee'. People get together and vote and they are argue and they create categorizations based on that. They do their compromises every time they need to update the book.

Diagnosis is then based on observed behavior and interviews. They don't do some sort of blood test, MRI scans, or X-rays or any other kind of medical diagnosis. Yeah, sure, there is some sort of questionable neuroimaging scan they can do, but none of that stuff ever gets used to actually diagnose people.

In terms of a science it's much closer to the 'sociology' side of things then 'search for the cure for cancer'.

The original purpose of DSM was so that psychologists and the lot have categorizations handy for filling out claims with patient's insurance. It's a lot easier to convince insurance companies to pay for treatments when the diagnosis is 'Attention deficit disorder' rather then 'Little Timmy is having a hard time sitting still in class and occasionally gets completely out of control'.

That doesn't mean these disorders don't exist or that DSM diagnoses are invalid or useless. It just means that it's all a bit 'hand wavy'.

[skissane]

> The same can be said of probably most disorders listed in the DSM.

True. Although, it is plausible that some DSM-5 disorders may have greater validity issues than others, and ASD may be one of those disorders with greater-than-average validity issues. The paper I cited built an extensive case, based on 14 separate research findings, to support the conclusion that ASD lacks validity. I wouldn't presume that if you repeated the same exercise for some other DSM-5 disorder you'd necessarily get an equally negative result.

> The original purpose of DSM was so that psychologists and the lot have categorizations handy for filling out claims with patient's insurance.

Which makes its use in research especially questionable. And this is part of why NIMH announced back in 2013 [1] that they were shifting research funding away from research based on DSM categories. If research based on the DSM-5 is dubious, then funding such research is not a good use of limited research funding resources. (It is worth noting this study was funded by NICHD not NIMH; I'd question whether NIMH would fund a study like this given its current funding criteria.)

> That doesn't mean these disorders don't exist or that DSM diagnoses are invalid or useless. It just means that it's all a bit 'hand wavy'.

What do you mean by "invalid"? If by "valid" one means biological validity and/or construct validity, then the whole point of the paper I cited was to argue that ASD lacks both forms of validity, and hence that in those senses of "valid", the diagnosis of ASD is invalid.

[1] https://www.nimh.nih.gov/about/directors/thomas-insel/blog/2...