Journal Club: Are Mental Disorders Real Categories? A New Framework Suggests a Different Way Forward

Name: Elyx
Author: Aion

Sarah Marchand-Lacoursière
Jun 23
4 min read

The Problem with Psychiatric Classification

For decades, psychiatry has relied on diagnostic systems such as the DSM and ICD to organize mental disorders. These systems have undoubtedly improved communication, research, and clinical practice. Yet many longstanding challenges remain:

Two patients with the same diagnosis may look entirely different clinically.
Many patients meet criteria for multiple diagnoses simultaneously.
Boundaries between disorders are often unclear.
Biological research has struggled to identify definitive biomarkers for most psychiatric conditions.

A recent review published in JAMA Psychiatry by Eiko Fried (2026) proposes a compelling explanation: perhaps mental disorders are not discrete disease entities waiting to be discovered, but rather complex configurations of interacting properties that tend to cluster together.

From Biological Species to Mental Health

Fried draws an analogy from biology.

Historically, scientists believed species represented fixed natural categories. Modern evolutionary biology has largely moved away from this view. Instead, many philosophers of science now describe species as Homeostatic Property Clusters (HPCs).

Under this framework:

Species do not possess a single defining essence.
They consist of many properties that tend to occur together.
These properties cluster because some reinforce or favor the presence of others.
Boundaries are often fuzzy rather than absolute.

A giraffe is not a giraffe because of one defining characteristic. Rather, a collection of anatomical, genetic, behavioral, developmental, and ecological features tend to co-occur.

Fried suggests that psychiatric disorders may function similarly. Rather than being single diseases with single causes, mental disorders may represent statistical clusters of interconnected biological, psychological, social, developmental, and environmental properties.

What Does This Mean for Psychiatry?

Under the HPC framework:

Depression is not simply low serotonin.

ADHD is not simply dopamine dysregulation.

Anxiety is not simply excessive fear.

Instead, these conditions may emerge from complex networks involving:

Symptoms
Personality traits
Developmental experiences
Lifestyle factors
Social determinants of health
Biological vulnerabilities
Cognitive patterns
Environmental influences

Different individuals may arrive at similar clinical presentations through very different pathways.

This concept aligns with established observations in psychiatry:

High heterogeneity within diagnoses
High rates of comorbidity
Variable treatment responses
Overlapping symptom profiles across disorders

Rather than viewing these phenomena as failures of current classification systems, the HPC framework suggests they may simply reflect the true complexity of human mental health.

The Mental Health Atlas

One of the most intriguing ideas proposed in the paper is the creation of a "Mental Health Atlas."

Instead of organizing people into rigid diagnostic boxes, researchers would map thousands of mental health-related properties and study how they cluster together.

These properties could include:

Psychiatric symptoms
Trauma history
Sleep patterns
Personality characteristics
Cognitive functioning
Social support
Medical conditions
Socioeconomic factors
Treatment response patterns
Behavioral data

The goal is not merely to classify individuals but to understand the structure and dynamics of mental health itself.

Why This Matters for AI

Traditional digital mental health tools often focus on a single disorder or a narrow symptom checklist.

However, if mental disorders truly emerge from interacting systems, then meaningful assessment requires a broader perspective.

Future mental health technologies will need to:

Integrate multiple domains simultaneously
Capture context rather than isolated symptoms
Incorporate longitudinal information
Recognize individual variability
Support clinicians in understanding patterns rather than simply generating labels

This represents a shift from diagnostic reductionism toward systems-based psychiatry.

How This Aligns With Elyx

At Aion, our vision has been shaped by a similar observation: meaningful psychiatric assessment requires more than symptom counting.

Elyx was designed around the principle that mental health emerges from the interaction of multiple dimensions, including:

Symptoms
Developmental history
Risk factors
Social determinants of health
Functional impairment
Behavioral patterns
Longitudinal trajectories
Patient narratives

Rather than focusing on a single disorder, Elyx collects and organizes multidimensional clinical information to support integrated clinical reasoning. The platform currently evaluates a broad range of psychiatric domains while incorporating contextual and longitudinal information that extends beyond traditional screening approaches.Importantly, Elyx does not replace clinical judgment. Instead, it seeks to improve the completeness, structure, and accessibility of information available to clinicians before the clinical encounter.

Looking Ahead: Psychiatry as a Dynamic System

One of the most powerful implications of Fried's framework is the idea that mental health is dynamic.

People move through different states over time.

Mental disorders may represent self-reinforcing configurations that emerge, stabilize, and sometimes dissolve as life circumstances, biology, relationships, and behavior change.

This perspective resonates strongly with emerging fields such as:

Network psychiatry
Complexity science
Precision mental health
Longitudinal digital phenotyping
Systems medicine

As psychiatry evolves, understanding patterns may become as important as identifying diagnoses.

Final Thoughts

The Homeostatic Property Cluster framework does not eliminate psychiatric diagnoses. Rather, it encourages us to view them as useful tools rather than immutable truths.

By focusing on relationships among symptoms, biology, behavior, and context, this framework offers a scientifically grounded path toward more personalized and clinically meaningful mental health care.

At Aion, we believe the future of psychiatry will increasingly move toward multidimensional, data-informed, and person-centered approaches. Fried's work provides a compelling theoretical foundation for that future—and reinforces why comprehensive, contextualized assessment may be one of the most important challenges for the next generation of mental health innovation.