Rethinking Psychiatry: A Shift Towards a Clinicopathological Model

Insights from Recent Articles

by Jonathan Bauman, M.D., LFAPA, ABAP Newsletter Editor

Over the past few months, three articles in the Psychiatric Times by Nassir Ghaemi, MD, and Mark Ruffalo, MSW, DPsa have presented intriguing ideas about the diagnostic foundation of our specialty. They argue that significant changes are needed. These articles are titled The Concept of Mental Illness and Why the DSM Approach Is Wrong (Psychiatric Times 12/3/25), What Happened to Psychopathology (12/5/26), and Diagnostic Validators and the Question of What Really Exists (3/23/26).

These discussions made me ponder what psychiatry might look like if we made major changes to our diagnostic systems in the near future. How would this affect our relationship with the ICD, payment systems for services, and treatment methods? Lacking the extensive knowledge to form a coherent vision of these changes, I decided to consult my friend, Chat-GPT. Here’s my query:

S. Nassir Ghaemi, MD, and Professor of Psychiatry at Tufts University School of Medicine, writes extensively about how the DSM's reliance on subjective distress and functional impairment diverges from scientific medicine’s clinicopathological approach to diagnosis. How would psychiatric nosology change if psychiatry adopted a clinicopathological approach to diagnosis? How would such a change impact the ICD? How would it affect insurance coverage for psychiatric disorders? How might it impact treatment?

And here’s Chat’s response, lightly edited for clarity:

What a Clinicopathological Approach Would Actually Mean

In scientific medicine, diagnoses are ideally based on:

• Etiology (cause)

• Pathophysiology (mechanism)

• Biomarkers or objective findings

• Course and prognosis

  
  

In psychiatry, diagnoses are largely:

• Syndromic (clusters of symptoms)

• Cross-sectional

• Validated by reliability rather than biological validity

  
  

A shift towards a clinicopathological model would require:

• Identifying biologically distinct disease entities

• Moving away from purely descriptive categories

• Replacing symptom thresholds with pathophysiologic subtypes

  
  

How Psychiatric Nosology Would Change

Fragmentation of Existing Disorders

Many existing disorders would likely fragment or disappear. For example:

• “Major depressive disorder” could evolve into multiple biologically distinct subtypes.

• “Schizophrenia” might become a spectrum of different neurodevelopmental or neurodegenerative diseases.

• “Bipolar disorder” could survive intact due to a stronger biological signal, aligning with Emil Kraepelin’s longitudinal/course-based approach to diagnosis.

  
  

Emergence of Biologically Anchored Subtypes

We might see the rise of biologically anchored subtypes such as:

• “Inflammatory-mediated depression”

• “Glutamatergic dysregulation syndrome”

• “Synaptic pruning disorder with psychosis”

  
  

This resembles initiatives like the NIMH’s Research Domain Criteria.

Reduced Role of Distress and Impairment

Currently, a condition is not classified as a disorder unless it causes distress or impairment. In a clinicopathological model, a disease could exist regardless of distress. Some psychiatric conditions might be diagnosed before subjective suffering emerges, similar to asymptomatic hypertension or early multiple sclerosis. Conversely, conditions like mild anxiety might not qualify as diseases at all.

Impact on the International Classification of Diseases

The ICD serves a broader global and administrative function than the DSM. Therefore:

• The ICD would likely lag behind DSM shifts.

• A dual-track system may emerge:

- A clinical/pragmatic track (symptom-based, for global use)

- A research/biological track (for advanced systems)

• Eventually, there could be convergence if robust biomarkers emerge.

  
  

Impact on Insurance Coverage

In the current system, insurance coverage depends on:

• Coded diagnoses

• Demonstrated distress or impairment

  
  

Under a clinicopathological model:

1. Some patients could lose coverage, such as those with distress but no identifiable pathology (e.g., adjustment reactions, mild anxiety, existential distress). These might be reclassified as “non-disease states” or non-reimbursable medical conditions.

2. Other conditions might gain earlier coverage, such as preclinical or biomarker-positive individuals. For example, “prodromal psychosis with neurobiological markers” could receive coverage, paralleling pre-diabetes or early-stage cancer screening.

3. There would be greater stratification of reimbursement, with treatments tied to specific mechanisms. Insurers may require biomarker confirmation or treatment matching.

   
   

Psychiatry currently captures a wide swath of human suffering, but a clinicopathological shift might push some conditions into psychology, social services, or non-medical care.

Impact on Treatment

Precision Psychiatry

Instead of relying on trial-and-error with SSRIs and antipsychotics, we would see a move toward precision psychiatry. This would involve:

• Mechanism-specific treatments

• Biomarker-guided prescribing

  
  

For example:

• Inflammatory depression could be treated with anti-inflammatory agents.

• Dopaminergic dysregulation might require targeted neuromodulators.

  
  

Fewer Broad-Spectrum Diagnoses

We would likely see fewer broad-spectrum diagnoses and generalized treatments. SSRIs might be used only in specific subtypes, and antipsychotics would be prescribed based on pathophysiology rather than just symptoms. There would also be a reduced role for non-specific psychotherapy for biologically driven illnesses.

Increased Emphasis on Longitudinal Course

The emphasis would shift toward the trajectory of the illness, as per Kraepelin, with less focus on cross-sectional diagnosis.

Conceptual and Philosophical Consequences

A More Medical Psychiatry

Psychiatry could become more aligned with neurology and less dependent on subjective reports. However, this shift risks excluding meaningful suffering that lacks biomarkers. It could undermine psychiatry’s role as a bridge between medicine and the human condition.

Bottom Line

If psychiatry adopted a clinicopathological model, we would see:

• Nosology shifting from syndromes to biologically defined diseases.

• DSM categories fragmenting or disappearing.

• The ICD adapting slowly, likely maintaining dual systems.

• Insurance becoming more biomarker-driven, potentially excluding some currently covered conditions.

• Treatment becoming more precise and less trial-and-error.

  
  

Psychiatry would gain scientific validity but risk narrowing its scope and abandoning parts of human suffering that are real but not yet biologically classifiable.

ChatGPT then contrasted Ghaemi’s position with another thinker in the field: Kenneth Kendler, MD, a professor at the Virginia Institute for Psychiatric and Behavioral Genetics. Ghaemi argues that psychiatry should approximate the rest of medicine, asserting that DSM diagnoses are merely a collection of symptoms lacking validity as discrete diseases. He believes that ‘distress/impairment’ is arbitrary and not a scientific criterion, advocating for diagnosis based on etiology, pathology, and longitudinal course.

On the other hand, Kendler asserts that psychiatric diagnoses are multifactorial, bio-psycho-social constructs that cannot yet be reduced to biological mechanisms. He acknowledges that while DSM diagnoses are imperfect, they provide useful constructs where ‘distress/impairment’ is a necessary feature. Kendler advocates for embracing multiple explanations rather than waiting for pathological mechanisms that may never be discovered.

In summary, Ghaemi pushes psychiatry towards scientific purity and medical validity, while Kendler supports pragmatic pluralism that embraces imperfect categories as the best available tools.

Future Directions for DSM

Right on time, the March edition of the Psychiatric Times focused on where we should go from the DSM-5-R. An article by Awais Aftab, MD, clinical assistant professor at Case Western Reserve University, summarized significant proposed changes to the DSM. I encourage you to read the article, but I will highlight one change: the name of the DSM would change from Diagnostic and Statistical Manual to Diagnostic and Scientific Manual. To me, ‘Scientific’ sounds aspirationally premature and pretentious, given how wide the gulf is between current practice and ‘scientific purity’. Assuming that this controversy will linger, why not simply leave the name as Diagnostic Manual of Psychiatry?

Jonathan Bauman, MD

June 2026