Files
magnus919_agent-skills/dsm5/references/02-assessment-and-differential.md
Magnus HedemarkGitHubfactory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
d346970bf8 feat(skill): add dsm5 — evidence-based companion to the DSM-5-TR (#276)
Adds the dsm5 skill: an evidence-based conversational expert grounded in
the DSM-5-TR (American Psychiatric Association, 2022) for clinicians,
practitioners, patients, and family members.

- SKILL.md: safety-first conversation workflow (triage -> clarify ->
  route -> compare criteria -> differentials -> calibrated conclusion),
  reference routing table, crisis protocol, audience adaptation
- references/: 28 files — foundation (00-02), all 22 DSM-5-TR diagnostic
  classes (10-31), Part III measures/culture/AMPD/conditions-for-further-
  study (32-33), and cross-cutting differentials (40). Criteria are
  paraphrased with exact counts, durations, specifiers, and ICD-10-CM
  codes, plus per-disorder clinician and patient/family conversation
  guides
- scripts/lookup.py: stdlib keyword search across the reference library
  (--json/--list/--max/-q)
- evals/evals.json: 9 output-quality cases (schema v1)
- README.md: human-facing overview, install notes, and APA attribution
- Catalog entries and generated artifacts (llms.txt, marketplace
  plugins) regenerated; all repo validators pass

Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
2026-08-04 21:30:11 -04:00

20 KiB

Assessment and Differential Diagnosis — Skill-Level Reference

Summarized and paraphrased from DSM-5-TR (American Psychiatric Association, 2022). This file is for orientation and education; verify exact criteria wording, codes, and recording procedures against the official DSM-5-TR before any formal clinical, legal, insurance, or research use.

What this file is

This file explains how a serious clinical assessment is structured, what the DSM-5-TR Part III assessment measures are for, how to use the cultural formulation tools, and how to reason through differential diagnosis. It supports the conversation workflow described in 00-overview-and-method.md and complements the per-chapter "Assessment considerations" and "Differential diagnosis" sections. For the full measure-by-measure detail, see 32-assessment-measures-and-cultural-formulation.md.

The clinical assessment approach

A DSM-5-TR-informed assessment is an evaluation, not a questionnaire. Its components:

Component What it contributes
Clinical interview (structured or semistructured) Systematic coverage of symptoms, onset, duration, course, and context
Observation Mental status: appearance, behavior, speech, mood, thought process/content, cognition
Collateral information Reports from family, partners, teachers, or other informants; records from prior clinicians
Functional assessment How symptoms impair social, occupational, or other important functioning
Rule-outs Substance/medication effects, another medical condition, and other mental disorders that better explain the picture

The manual is explicit that checking off symptoms is not sufficient for diagnosis: the relative severity and salience of signs and symptoms, and whether the combination of predisposing, precipitating, perpetuating, and protective factors has crossed into psychopathology, require clinical judgment. Many criteria sets ask for information from family or other third parties to judge distress and impairment.

The universal exclusion: substance and medical causes

Almost every criteria set contains an exclusion along the lines of "the disturbance is not attributable to the physiological effects of a substance or another medical condition." Concretely:

  • Substance/medication-induced disorders: symptoms that arise from the physiological effects of an exogenous substance on the central nervous system — typical intoxicants (alcohol, inhalants, hallucinogens, cocaine), psychotropic medications (stimulants, sedatives/hypnotics/anxiolytics), other medications (e.g., steroids), or environmental toxins — including symptoms during withdrawal.
  • Another medical condition: DSM-5-TR uses "another medical condition" (not "general medical condition") to stress that mental disorders are medical conditions and can be precipitated by other medical conditions (e.g., a neurocognitive disorder due to Alzheimer's disease).
  • Independent mental disorders are those not due to substances or medical conditions; the term "independent" does not imply the disorder is free of psychosocial or environmental causal factors.

A good assessment therefore always asks about substances, medications, and physical health before attributing symptoms to a primary mental disorder.

Assessment measures (DSM-5-TR Part III)

Part III ("Emerging Measures and Models") provides dimensional tools to complement the categorical criteria. They are aids, not diagnostic tests: clinical judgment guides their use and interpretation, and they are intended to be given at the initial evaluation to establish a baseline and repeated over time to track change (measurement-based care). All are available at www.psychiatry.org/dsm5.

Level 1 Cross-Cutting Symptom Measure

A screening instrument modeled on general medicine's "review of systems." It scans many psychiatric domains at once so clinicians notice symptoms beyond the presenting complaint — atypical presentations, subsyndromal conditions, and coexisting problems.

  • Adult version (self-rated): 23 questions covering 13 domains — depression, anger, mania, anxiety, somatic symptoms, suicidal ideation, psychosis, sleep problems, memory, repetitive thoughts and behaviors, dissociation, personality functioning, and substance use. Each item asks how much/how often the person was bothered in the past 2 weeks, rated 0 (none) to 4 (severe).
  • Parent/guardian version (child age 6-17): 25 questions covering 12 domains, including inattention and irritability; items on suicidal ideation/attempts and substance use are answered "Yes / No / Don't Know."
  • Child-rated version (age 11-17): available online; similar structure.
  • Interpretation thresholds: for most domains, a rating of mild (2) or greater suggests further inquiry; for suicidal ideation, psychosis, and substance use, a rating of slight (1) or greater does; for inattention in the child version, slight or greater. A "Don't Know" on the child suicide or substance items warrants probing the child directly.
  • The APA Practice Guidelines for the Psychiatric Evaluation of Adults endorse the Level 1 measure as a first step in the psychiatric evaluation.

Level 2 Cross-Cutting Symptom Measures

When a Level 1 domain crosses its threshold, a Level 2 measure provides a more in-depth look at that domain. Examples named in the text: PROMIS short forms for depression, anxiety, anger, and sleep disturbance; the Altman Self-Rating Mania Scale; the Patient Health Questionnaire-15 (PHQ-15) for somatic symptoms; the Florida Obsessive-Compulsive Inventory (FOCI) severity scale; the NIDA-modified ASSIST for substance use; the SNAP-IV for inattention; and the Affective Reactivity Index for irritability. Adult and pediatric versions are available for most domains.

Disorder-specific severity measures

Severity measures map closely onto a specific disorder's criteria and rate symptom intensity, frequency, duration, or overall severity; they can be used whether or not full criteria are met. The text's flagship example is the Clinician-Rated Dimensions of Psychosis Symptom Severity: 8 items covering hallucinations, delusions, disorganized speech, abnormal psychomotor behavior, negative symptoms, impaired cognition, depression, and mania, each rated 0 (none) to 4 (severe) at its most severe in the past 7 days. Similar measures exist for conditions such as generalized anxiety disorder, social anxiety disorder, PTSD, and autism spectrum disorder.

WHODAS 2.0 (World Health Organization Disability Assessment Schedule 2.0)

  • A 36-item, self-administered measure of disability in adults 18 and older, validated across numerous cultures and sensitive to change.
  • Assesses six domains: understanding and communicating; getting around; self-care; getting along with people; life activities (household, work, and/or school); and participation in society.
  • Based on the WHO International Classification of Functioning, Disability and Health; usable for any medical condition, not just mental disorders. A proxy-administered version exists for people with impaired capacity (e.g., major neurocognitive disorder).
  • In DSM-5, WHODAS replaced the DSM-IV multiaxial system's Axis V Global Assessment of Functioning (GAF) scale. DSM-5-TR is non-axial: clinical disorders, personality disorders, and other medical conditions are listed together, and psychosocial problems are recorded with Z codes (see 31-other-conditions-that-may-be-a-focus-of-clinical-attention.md).

Cultural formulation

Culture shapes the experience, expression, and threshold of every symptom. DSM-5-TR provides two tools for incorporating culture into assessment: the Outline for Cultural Formulation and the Cultural Formulation Interview (CFI), both in Part III.

Outline for Cultural Formulation

A framework for systematically assessing, for every individual (not only those whose background is unfamiliar to the clinician):

  1. Cultural identity of the individual — demographic and social characteristics, religious/spiritual affiliation, migrant status, language use, and which aspects of identity the individual prioritizes.
  2. Cultural concepts of distress — idioms of distress, explanations/perceived causes, and cultural syndromes (below), plus stigma, anticipated outcomes, and help-seeking expectations.
  3. Psychosocial stressors and cultural features of vulnerability and resilience — social determinants of mental health, exposure to racism and discrimination, supports and stressors in family and community.
  4. Cultural features of the relationship between individual and clinician — how differences in background, language, and status affect communication, trust, and the accuracy of assessment.
  5. Overall cultural assessment — a summary of the implications of all of the above for differential diagnosis, management, and treatment.

Cultural Formulation Interview (CFI)

A brief, semistructured, person-centered interview that obtains the individual's own view of the problem and its social context. It has three components:

  • Core CFI: 16 questions, organized into four domains — Cultural Definition of the Problem (questions 1-3); Cultural Perceptions of Cause, Context, and Support (4-10); Cultural Factors Affecting Self-Coping and Past Help Seeking (11-13); and Cultural Factors Affecting Current Help Seeking (14-16).
  • Informant version: collects the same information from a knowledgeable family member or caregiver; useful when the individual cannot provide it (e.g., children, florid psychosis, cognitive impairment).
  • Supplementary modules: expand each domain and exist for specific populations (children and adolescents, older adults, caregivers, immigrants and refugees, religion and spirituality, etc.).

The CFI is designed to avoid stereotyping: there are no right or wrong answers, and it is useful in any setting regardless of the cultural background of the individual or clinician — people who appear to share a background can still differ in ways that matter to care. It is especially helpful when: assessment is difficult due to cultural/religious/socioeconomic differences between clinician and individual; culturally distinctive symptoms do not obviously fit criteria; severity or impairment is hard to judge; clinician and individual disagree about care; or engagement and trust are limited (e.g., in communities with histories of oppression and mistrust of mainstream services).

Cultural concepts of distress

Three related constructs replaced the older term "culture-bound syndrome":

Construct Meaning Example
Cultural idiom of distress A shared way of talking about or expressing distress that need not map onto a syndrome Saying "I feel so depressed" to express low mood below a disorder threshold; "nerves"
Cultural explanation or perceived cause A culturally recognized label, attribution, or causal model for symptoms or distress Attributing illness to "stress," spirits, or failure to follow prescribed practices
Cultural syndrome A cluster of co-occurring symptoms recognized locally as a coherent pattern ataque de nervios; khyâl cap

Four features matter for diagnosis: there is seldom a one-to-one correspondence between any cultural concept and a DSM disorder (the mapping is typically one-to-many in either direction); the same term often serves more than one function (e.g., "depression" can name a syndrome, an idiom, or an explanation); cultural concepts span the full range of severity, including presentations that meet no DSM criteria; and they change over time.

Examples of cultural concepts of distress

The following are paraphrased from the manual's ten worked examples. Each overlaps phenomenologically with some DSM disorders but is not equivalent to any single one.

Concept Cultural context Key features Closely related DSM conditions
Ataque de nervios ("attack of nerves") Latinx communities Acute intense emotional upset: anxiety, anger, or grief; screaming and shouting; attacks of crying; trembling; heat rising in the chest; verbal/physical aggression; a sense of being out of control; sometimes dissociation, fainting, or seizure-like episodes. Often triggered by a family stressor (e.g., death of a relative, conflict). Reported by 7%-15% of US Latinx adults and 4%-9% of youth in community samples; some ataques are normative expressions of acute distress (e.g., at a funeral) with no clinical sequelae. Panic attack/panic disorder, other specified or unspecified dissociative disorder, functional neurological symptom disorder (conversion disorder), intermittent explosive disorder
Khyâl cap ("wind attack") Cambodia Panic-like attacks (dizziness, palpitations, shortness of breath, cold extremities) plus catastrophic cognitions that a windlike substance (khyâl) rising in the body will compress the lungs or reach the head. Triggers: worry, standing up, specific odors, crowded places, riding in a car. Usually meets panic attack criteria. Panic attack, panic disorder, generalized anxiety disorder, agoraphobia, PTSD, illness anxiety disorder
Taijin kyofusho ("interpersonal fear disorder") Japan Anxiety about and avoidance of interpersonal situations from the belief that one's appearance or actions offend others. "Sensitive" type resembles severe social anxiety; "offensive" type centers on offending others. Variants involve blushing, body odor, gaze, or facial expression/body movements. Broader than social anxiety disorder. Social anxiety disorder, body dysmorphic disorder, olfactory reference syndrome, delusional disorder, OCD
Dhat syndrome South Asia Anxiety and distress about semen loss (dhat) in the absence of identifiable physiological dysfunction, with diverse accompanying complaints (fatigue, weakness, anxiety, depressed mood, somatic symptoms). Largely a cultural explanation rather than a discrete syndrome. Major depressive disorder, generalized anxiety disorder, somatic symptom disorder, illness anxiety disorder, sexual dysfunctions
Hikikomori Japan (now described widely) Protracted, severe social withdrawal, often an adolescent/young adult remaining in the home with minimal in-person contact (Japanese ministry guideline: 6 months of withdrawal). May occur alongside an established disorder or independently. Social anxiety disorder, major depressive disorder, autism spectrum disorder, schizoid/avoidant personality disorders, schizophrenia
Kufungisisa ("thinking too much") Shona, Zimbabwe Idiom and explanation: rumination on upsetting thoughts and worries (including chronic illness like HIV), held to cause anxiety, depression, and somatic problems. "Thinking too much" appears across many regions. Major depressive disorder, generalized anxiety disorder, PTSD, OCD, prolonged grief disorder
Maladi dyab ("sent sickness") Haitian communities Explanation attributing diverse psychiatric and medical problems to envy and malice, with illness "sent" by sorcerers. Onset mode and social status matter more than symptoms. Risk of misdiagnosis as delusional disorder or schizophrenia; many disorders can be attributed to it
Nervios ("nerves") Latinx communities Broad idiom and explanation spanning normal distress to presentations resembling anxiety, depressive, dissociative, somatic, or psychotic disorders. Major depressive disorder, generalized anxiety disorder, social anxiety disorder, somatic symptom disorder
Shenjing shuairuo ("weakness of the nervous system") China Cultural syndrome blending Traditional Chinese Medicine with neurasthenia: weakness, vexation, excitement, nervous pain, sleep problems. Use has declined as idioms of depression/anxiety spread. Mood, anxiety, and somatic symptom disorders
Susto ("fright") Latin American contexts Illness attributed to a frightening event causing the soul to leave the body; variable symptoms (appetite/sleep disturbance, sadness, low self-worth, somatic complaints). Syndromic subtypes align differently with depression, PTSD, or somatic symptom disorder. Major depressive disorder, PTSD, somatic symptom disorder

How culture should change the assessment

  • Use the cultural concepts to detect, not dismiss. Referring to a patient's own idiom can improve identification of psychopathology and rapport. But a cultural syndrome is not itself a DSM diagnosis: assess whether the presentation meets criteria for a specific disorder, an "other specified" category, or no disorder.
  • Avoid misdiagnosis. Socially warranted suspicion can be misread as paranoia; unfamiliar symptom presentations can be misread as psychosis. The CFI is designed to prevent exactly these errors.
  • Document the cultural material. Once a disorder is diagnosed, include the cultural terms and explanations in the case formulation; they clarify symptoms and attributions that would otherwise be confusing.
  • Need for care is separate. People whose symptoms meet no DSM criteria may still expect and require treatment; assess that on a case-by-case basis.

The differential-diagnosis method

Differential diagnosis is the disciplined comparison of candidate conditions. The same symptom — poor concentration, insomnia, irritability, panic — appears across many disorders, so the distinguishing features are pattern, timing, context, and exclusions.

  1. Assemble the presentation. Symptoms, onset, duration, course (episodic vs. persistent), distress and impairment, context and stressors, and age/developmental stage. Ask; do not assume.
  2. Screen safety and physical causes first. Rule out imminent risk (see 01-safety-and-boundaries.md) and the universal exclusion: substances, medications, and other medical conditions.
  3. Generate candidates from the symptom pattern. Use the Level 1 cross-cutting measure to scan adjacent domains; use the routing table in 00-overview-and-method.md to pick the chapter files for each candidate.
  4. Compare against each candidate's criteria. For each candidate, state which criteria appear met, unmet, or unknown (counts, durations, impairment, exclusions).
  5. Weigh distinguishing features. Typical differentiators: age of onset (e.g., ADHD requires onset in the developmental period), course (acute vs. insidious), context (e.g., substance-induced vs. independent), response to reassurance, and which symptoms are primary.
  6. Present the most likely candidates with their distinguishing features. Name what information would change the ranking — e.g., a urine drug screen, a medication review, a collateral history, or the passage of a duration threshold.
  7. Close with calibrated language and next steps. "Consistent with" framing, explicit uncertainty, and a recommendation to have a qualified clinician confirm.

The manual's per-chapter differential sections (summarized in each chapter reference) and the cross-cutting guide 40-cross-cutting-differentials.md (which systematically compares symptoms that span conditions) are the working tools for steps 3-5. Boundary cases are expected: the manual notes that categories have fluid boundaries and that boundary presentations are diagnosed "in any but a probabilistic fashion."

Cross-references