Beyond the PHQ-9

Why a Single Screener Misses Most Comorbidities

The problem with point-solution screeners

The PHQ-9 is good at what it does: identifying probable major depression in primary care populations. It was designed for fast, population-level screening in non-specialist settings. It was never intended to be the primary diagnostic tool in a psychiatric practice.

Yet many psychiatric providers, facing schedule pressure and limited intake time, default to a small set of brief screeners: PHQ-9 for depression, GAD-7 for anxiety, maybe an AUDIT for alcohol. These tools capture the chief complaint, but they are structurally incapable of revealing what else might be going on.

The result is a predictable clinical pattern: patients who arrive with a depression diagnosis but don't respond to antidepressants. Patients who have been in treatment for years without reaching remission. Patients with undetected ADHD, bipolar II, PTSD, or OCD layered underneath the presenting complaint.

A large portion of patients at RCBM had spent years in treatment with minimal relief before receiving a comprehensive screening assessment. Many had been misdiagnosed. Comprehensive baseline screening is the cornerstone of the diagnostic precision RCBM is known for.

What single-scale screeners structurally cannot do

The core limitation is not a matter of scale quality, it is a matter of scope. When you administer a PHQ-9 and only a PHQ-9, you are asking a question about one condition. Psychiatric comorbidity is the rule, not the exception. DSM-5 criteria for most conditions require that alternative diagnoses be considered and ruled out. A single-scale approach makes this impossible by design.

Brief screeners also suffer from confirmation bias at the system level. If a referral reads "depression," the provider is more likely to administer depression scales. The screener confirms what was already suspected. It rarely reveals what wasn't.

The case for cross-cutting, adaptive assessment

The DSM-5 itself recognizes this problem. Its cross-cutting symptom measures are designed to cast a wide net across symptom domains before zeroing in on specific conditions. This is the logic behind how MindMetrix is structured: begin with a broad screen across all 60+ condition areas, then adaptively administer targeted validated scales only for the areas where significant symptoms emerge.

This approach has two major advantages:

• It catches comorbidities that a targeted intake would miss, including conditions the patient has never raised and the provider never suspected.
• It generates a probability-weighted differential for each flagged condition, using Bayesian modeling across validated scale results — giving the provider a data-driven starting point rather than a clinical hunch.

Common conditions missed by single-scale approaches

In clinical practice, the following are frequently underdetected when providers rely on brief targeted screeners:

• Bipolar II: Frequently presents as recurrent depression. PHQ-9 will confirm depressive episodes but cannot detect hypomanic history. The wrong diagnosis leads to antidepressant monotherapy — which may destabilize the patient.
• Adult ADHD: Highly comorbid with depression and anxiety. Cognitive symptoms are often attributed to the mood disorder. Treatment-resistant depression with attention and executive function complaints warrants ADHD screening.
• PTSD: Underreported in self-referral. Patients rarely identify trauma as the presenting complaint. Somatic symptoms, sleep disruption, and hypervigilance overlap significantly with depression and anxiety — and require targeted screening to untangle.
• OCD: Patients often do not spontaneously disclose obsessive or compulsive symptoms. Without a direct screener, this diagnosis is routinely delayed by years.

A more defensible clinical standard

Comprehensive baseline screening is increasingly the standard the field is moving toward. Value-based care contracts, payer scrutiny, and the growing emphasis on measurement-based care all point in the same direction: diagnostic impressions need to be grounded in data.

A chart-ready MindMetrix report — generated before the first appointment, with probability scores for 60+ conditions and differential diagnosis guidance — puts the provider in a defensible clinical position from visit one.