Evidence Method
How We Keep Claims From Getting Ahead Of The Evidence
Diagnostic Psychiatry needs a wide differential and a narrow claims standard. A medical contributor can belong in the workup before it is proven to explain every symptom.
Evidence Tiers
Established
Strong guideline, replicated review, or clear disease mechanism. Still interpreted in context.
Supported
Good evidence and clinical plausibility, but not definitive for every patient or setting.
Proposed
Framework-level reasoning or emerging mechanism. Useful for the differential, not proof.
Speculative
Too early for patient-facing action unless it is clearly labeled and bounded.
What Counts As A Strong Source
- clinical practice guidelines from professional societies or public health agencies
- systematic reviews and meta-analyses with clear inclusion criteria
- randomized controlled trials when the question is treatment response
- large cohort, case-control, or cross-sectional studies for associations and risk
- mechanistic or expert interpretation only when it is labeled as such
What We Do Not Let Sources Do
- A single lab value does not diagnose a psychiatric condition.
- An association does not prove cause.
- A case report does not create a general treatment rule.
- A normal reference range does not end clinical reasoning when symptoms fit.
- A proposed framework does not replace evaluation by the appropriate provider or specialist.
How Pages Are Reviewed
- Claims are checked against the registry, guidelines, and source list.
- Patient-facing language is edited for safety and scope.
- Condition pages are scheduled for review at least annually, sooner if major guidance changes.
- Corrections prioritize patient safety, source accuracy, and clear boundaries.
The Bottom Line
The point is not to make every psychiatric symptom medical. The point is to stop closing the case before thyroid disease, iron deficiency, B12 deficiency, sleep apnea, medication effects, metabolic disease, and other relevant contributors have been considered.