So, can you do a lab test for a mental disorder, and if not, what mitigating factors should be taken into account?
A mental disorder typically manifests as a behavioural or mood disorder.
See the Diagnosis of schizophrenia, for example: there isn't what you'd call a "lab test" to detect disorganized speech, disorganized behaviour, blunted affect, etc. -- these are behavioural (not neurochemical) symptoms/criteria.
OTOH a doctor can give a person some prescription medication and see whether their symptoms improve: that's a kind of test: does the medicine work, have the desired effect?
The principal claim in the rant you cited seems to be:
“There is no definition of a mental disorder. It’s bullshit. I mean, you just can’t define it.”
Well it is defined, for example by the DSM.
The rant is right about a lot of things:
- Misdiagnosis is possible (e.g. several diseases may present as similar symptoms and require Differential diagnosis)
- Skilled/experienced diagnosis is recommended (people aren't supposed to self-diagnose)
- Doctors can't be sure in advance which drugs if any will have the desired effect: it's a bit of an experiment
Perhaps this answers your question.
Incidentally, I noticed this today.
When A Patient Presents With A Painful Red Toe
The three women had superficial frostbite. No laboratory test exists for frostbite as it is a clinical diagnosis.
A diagnosis made on the basis of medical signs and patient-reported symptoms, rather than diagnostic tests
I don't think one would want to argue that "frostbite" for example doesn't exist or is an urban legend, nor blame it on the medical establishment, merely because it's diagnosed clinically rather than in a lab.