Since the full article is behind a paywall, I cannot see the details that the author goes into. However, I will say that psychiatry has a long way to go. One ma

Since the full article is behind a paywall, I cannot see the details that the author goes into. However, I will say that psychiatry has a long way to go. One major issue is that there's no commonly agreed upon methods of diagnoses. There's a wide range of inventories, surveys, structured clinical interviews, etc., and every psychiatrist has their preferred sets of those. It doesn't help that the questions asked are often easily interpreted different from patient to patient and that some questions are quite literal, while others are somewhat vague probes into emotion. With that said, the issue is ultimately the reliance solely on the word of the patients, whom are often lacking a degree of insight due to their mental illness (e.g. brain fog, deficient expression, cognitive dissonance, emotional suppression, etc.). It doesn't help that mental illness has some inherent difficulties compared to other areas of medicine: * Prevalently comorbid with some illnesses being symptoms of something else (e.g. anxiety and depression being symptoms of bipolar disorder) * Distinctions between illnesses being blurry and those illnesses overlapping in symptoms (e.g. General Anxiety Disorder and Major Depression Disorder, Psychosis and Mania, Panic Disorders and PTSD, etc.) * Symptoms being inherently unstable because of the number of uncontrollable internal and external variables that influence them * Medications affect every person differently to the point where medications in the same classification (e.g. SSRI) will have vastly different effects to the same patient (e.g. one SSRI alleviates symptoms, while another doesn't or even exacerbates symptoms) * Those medications having difficult onboarding effects or side effects (e.g. SSRIs typically making depressive symptoms worse for the first 3-6 weeks before symptoms improve) * Lack of physical testing methods like blood tests, biopsies, etc. * The many complexities and unknowns of the brain that we have still yet to understand compared to other organs * Diagnoses and treatments need to be personalized, but no easy, economic, objective way to do so. For example, brain activity and functional networks vary from person to person, effective stimulation treatments require personalized neurophysiological targeting derived from functional connectivity analysis, illness can derive from a wide range of causes or combination of causes like brain networks, neurotransmitter imbalance, hormone imbalance, etc. You even need to consider those with more detail, like are the involved brain networks too strong or too weak, is the neurotransmitter imbalance from deficient production or deficient uptake, etc.? --- TL;DR: It's no wonder that diagnosis of mental illness is so inconsistent across psychiatry because our current methods are almost entirely qualitative, mental illness is relatively poorly understood compared to other areas of medicine, mental illness is inherently comorbid and overlapping, and diagnosis/treatment requires a deeply personalized approach with no good way to do so (yet).