Where's the line between psychiatric and neurological disorders?

[MysteryDiagnosis]

It seems there is so much overlap between neurological and psychiatric disorders as both are seated in the brain.

It's confusing to me where the line is and how some are classified and neurological while other are considered psychiatric? Furthermore, even more confusing are that some are considered both (adhd, alzheimers, etc).

I tried to research, and what I found said that neurological disorders have structural impairments like lesions in the brain and psychiatric doesnt... also, that neurological disorders usually involve gross motor or physical symptoms and psychiatric doesnt and are more emotional/ behavioral... but that cofuses me because many neurological disorder's are "functional" and have no structural pathology in the CNS... and also many neurological disorder also present with emotional/ behavioral symproms like "psychiatric" disorders.

Also, another thing I read said that psychiatric disorders usually have a psychosocial/ environment trigger or component (such as childhood abuse, cultural factors, trauma, etc) and neurological doesnt... but many neurological disorders do involve those factors such as lifestyle and trauma. For instance, conversion disorder and psychogenic pain.

So where is the line? Like schizophrenia which has demonstrated several neurobiological/ brain abnormalities... what makes it psychiatric and not neurological?

Or PTSD which has shown some structural changes in the brain such enlarged hippocampus and amygdala... what makes that psychiatric and not neurological?

Does it get classified according to what's easier to treat with psychological interventions like psychotherapy? Even then, something like schizophrenia is very difficult if not impossible to treat with just psychotherapy.

---

Members don't see this ad.

 

[Electricity]

From first principles it is obvious that all psychiatric illnesses also involve dysregulation of the brain, because the mind is entirely dependent on the brain. So pointing to imaging studies showing that some traditionally psychiatric disorders have imaging biomarkers is certainly not an argument that they should be treated by neurologists.

There's no person or set of people who decide that one set of disorders is associated with one specialty and one set is associated with another specialty. The way this is decided in practice is a complex, historical system of licensing, case law, turf wars between specialties, insurance reimbursement patterns that reimburse some specialties for certain aspects of care better than others, the perceptions and preferences of patients, and many other factors. If you want to understand how this has happened with respect to a particular disorder, probably the best way is to learn about the history of the disorder. Usually it is way more messy than you'd expect.

That said, here are two of the relatively more sane divisions I've seen:

- Associated with a cellular/tissue-level lesion that is detectable under the microscope by a neuropathologist (neurology) vs not (psychiatry). This gives more of the cognitive disorders to neurology.
- Involves a defect related to the sensorimotor nervous system (neurology) vs not (psychiatry). This gives more of the cognitive disorders to psychiatry.

It also might be wise to consider the disorders in terms of clustering, as opposed to types (as these are the two major forms of categorization). As an example, from this perspective, schizophrenia is much closer to other diagnoses that are considered traditionally psychiatric than it is to diagnoses that are traditionally neurologic. For example, the IRR of schizoprenia vs schizoaffective disorder, BPAD, and unipolar depression has been found to only be about 0.6 in one study: The diagnoses of schizophrenia, schizoaffective disorder, bipolar disorder and unipolar depression: interrater reliability and congruence between D... - PubMed - NCBI. You could imagine clustering all disorders that are either neurologic and/or psychiatric based on symptomatology, comorbidity, difficulty in distinguishing them as diagnoses, and then doing k-means with k = 2 and splitting everything that way.

From the research perspective, there is way more flexibility in what you can do. Freud was a neurologist. Helen Meyburg, one of the major figures in treating depression with DBS, is a neurologist. If you're interested in the research, you don't really need to worry about it.

I feel your pain on this, because I'm a student also trying to decide between neurology and psychiatry. I'm frustrated by the division, which at times seems very arbitrary, especially for people interested in behavioral neurology/neuropsychiatry. Ultimately I'm trying to think first and foremost about the patients, who for the most part don't care about the divide, and just want to get the best treatment possible.

 

[MysteryDiagnosis]

From first principles it is obvious that all psychiatric illnesses also involve dysregulation of the brain, because the mind is entirely dependent on the brain. So pointing to imaging studies showing that some traditionally psychiatric disorders have imaging biomarkers is certainly not an argument that they should be treated by neurologists.

There's no person or set of people who decide that one set of disorders is associated with one specialty and one set is associated with another specialty. The way this is decided in practice is a complex, historical system of licensing, case law, turf wars between specialties, insurance reimbursement patterns that reimburse some specialties for certain aspects of care better than others, the perceptions and preferences of patients, and many other factors. If you want to understand how this has happened with respect to a particular disorder, probably the best way is to learn about the history of the disorder. Usually it is way more messy than you'd expect.

That said, here are two of the relatively more sane divisions I've seen:

- Associated with a cellular/tissue-level lesion that is detectable under the microscope by a neuropathologist (neurology) vs not (psychiatry). This gives more of the cognitive disorders to neurology.
- Involves a defect related to the sensorimotor nervous system (neurology) vs not (psychiatry). This gives more of the cognitive disorders to psychiatry.

It also might be wise to consider the disorders in terms of clustering, as opposed to types (as these are the two major forms of categorization). As an example, from this perspective, schizophrenia is much closer to other diagnoses that are considered traditionally psychiatric than it is to diagnoses that are traditionally neurologic. For example, the IRR of schizoprenia vs schizoaffective disorder, BPAD, and unipolar depression has been found to only be about 0.6 in one study: The diagnoses of schizophrenia, schizoaffective disorder, bipolar disorder and unipolar depression: interrater reliability and congruence between D... - PubMed - NCBI. You could imagine clustering all disorders that are either neurologic and/or psychiatric based on symptomatology, comorbidity, difficulty in distinguishing them as diagnoses, and then doing k-means with k = 2 and splitting everything that way.

From the research perspective, there is way more flexibility in what you can do. Freud was a neurologist. Ellen Mayburg, one of the major figures in treating depression with DBS, is a neurologist. If you're interested in the research, you don't really need to worry about it.

I feel your pain on this, because I'm a student also trying to decide between neurology and psychiatry. I'm frustrated by the division, which at times seems very arbitrary, especially for people interested in behavioral neurology/neuropsychiatry. Ultimately I'm trying to think first and foremost about the patients, who for the most part don't care about the divide, and just want to get the best treatment possible.

Thank you very much for your honest and in depth post. I'm glad I'm not the only one who finds this divide very confusing and at times arbitrary.

I also like what you said about behavioral neurology and neuropsychiatry... I feel that it would be both fields to merge and become one again since there is so much crossover... and then there can be subspecialists in select areas such as psychosis, depression etc.

 

[Thama]

The two fields evolved in parallel. Some would say that the original delineation was that neurology involved those disease processes of the nervous system where abnormalities could be seen on light microscopy in the late 19th century, while psychiatry involved diseases that could not be. It is notable that dementia has historically been managed by both specialties, for a time it was considered more a part of psychiatry but, as the molecular basis of common dementing disorders has become known, neurology has gradually taken over the field. More recently psychiatry has stepped away from Freudian pseudoscience and into biopsychiatry, so there is truly no line of what should be considered neurologic and what should be considered psychiatric. Many primary neurologic disorders involve widespread network dysfunction (e.g. idiopathic Parkinson disease), while many primary psychiatric disorders produce clear structural abnormalities that are observable on volumetric analysis or DTI. Other disorders still seem to straddle the two fields to varying degrees, such as autism spectrum disorders, tic disorders, NMDA receptor encephalitis and Lewy body dementia.

For those considering the two fields, the important thing is not where the line lies exactly, but what your training perspective is. Those who go through a neurology residency will see large amounts of acute stroke, ICU management of intracranial pressure and status epilepticus, immunosuppression for neuromuscular disorders, and many other treatments of people with devastating or potentially devastating neurologic injuries. Most of neurology residency is a lot more like an internal medicine residency than it is a psychiatry residency. Those who do a psychiatry residency will be concentrating on management of psychosis, mania, and a variety of behavioral modifications.

Frankly if you are interested in the intersection point between neurology and psychiatry, a psychiatry residency is likely to be both more applicable and less demanding (for example at my institution the neurology residency is considered easily the most difficult non-surgical program). The behavioral neurologists I know are purely dementia specialists and almost all of them spend 80%+ of their time in the lab, either doing Alzheimer's trials or wet bench work on B-amyloid/tauopathies/etc. The neuropsychiatrists I know practice more along the lines of what you're discussing in this thread.

 

[kchan99]

- Associated with a cellular/tissue-level lesion that is detectable under the microscope by a neuropathologist (neurology) vs not (psychiatry). This gives more of the cognitive disorders to neurology.

For neurological disorders:
- Associated with abnormal conventional MRI.
This would be the case with PRES. The neuropathology of PRES is hard to define as most cases do not come to autopsy since patients survive.

- Response to surgical treatment.
An example would be essential tremor, which responds well to DBS. In contrast, while certain psychiatric disorders may respond to DBS, responses can be less predictable. The neuropathology of essential tremor can be controversial.

For many psychiatric disorders, especially neuroses, the talking cure, or psychotherapy, is quite effective. This is not the case with neurological disorders such as epilepsy, dystonia, and essential tremor.

while many primary psychiatric disorders produce clear structural abnormalities that are observable on volumetric analysis or DTI.

These changes on quantitative MRI are at the group level, and not the individual level.

 

[Thama]

These changes on quantitative MRI are at the group level, and not the individual level.

Whinch matters... why? Volumetrics isn't clinically validated in AD either but dementia specialists tack that analysis on regularly.

 

[kchan99]

Whinch matters... why? Volumetrics isn't clinically validated in AD either but dementia specialists tack that analysis on regularly.

Hippocampal volumes are not diagnostic of any specific disease. Hippocampal atrophy is seen in a variety of pathologies, such as Alzheimer's disease, hippocampal sclerosis of aging, primary age related tauopathy, mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS), and hypoxic-ischemic insults. If a patient with an amnestic cognitive impairment with normal appearing hippocampi on MRI, then I would think about other differential diagnoses.

Alzheimer's disease happens to be an easy diagnosis to make clinically for amnestic syndromes because amyloid and tau pathology occur frequently in aging. In addition, there are well-validated biomarkers using amyloid PET and CSF amyloid and tau/phospho-tau testing. Hippocampal volume is not one of these biomarkers. In many cases of neurodegeneration, we hardly ever know what our patients really have because of low autopsy rates that is in part related to absence of insurance and Medicare reimbursement. Consequently, many imaging biomarkers are not validated against pathology.

 

[Thama]

Hippocampal volumes are not diagnostic of any specific disease. Hippocampal atrophy is seen in a variety of pathologies, such as Alzheimer's disease, hippocampal sclerosis of aging, primary age related tauopathy, mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS), and hypoxic-ischemic insults. If a patient with an amnestic cognitive impairment with normal appearing hippocampi on MRI, then I would think about other differential diagnoses.

Alzheimer's disease happens to be an easy diagnosis to make clinically for amnestic syndromes because amyloid and tau pathology occur frequently in aging. In addition, there are well-validated biomarkers using amyloid PET and CSF amyloid and tau/phospho-tau testing. Hippocampal volume is not one of these biomarkers. In many cases of neurodegeneration, we hardly ever know what our patients really have because of low autopsy rates that is in part related to absence of insurance and Medicare reimbursement. Consequently, many imaging biomarkers are not validated against pathology.

I think you're trying to have a completely different discussion than the one occurring in this thread.