CPS: Clinical Problem Solving

Models the NEJM Clinical Problem-Solving format. Multi-persona diagnostic rounds apply Bayesian reasoning with likelihood ratios to systematically narrow a differential diagnosis, supported by textbook extraction and literature evidence.

Quick Start

/cps path/to/SCENARIO.md
/cps discover chest pain
/cps round ./case/45f-chest-pain 6
/cps review ./case/45f-chest-pain

Subcommands

Full Workflow (/cps SCENARIO.md)

Copy this checklist and track progress:

- [ ] Phase 1: Case Intake & Setup → Validate (Red Flag History HARD GATE)
- [ ] Phase 2: Symptom Mapping & DDx → Validate (MNM coverage, Syndromic Screen)
- [ ] Phase 3: Initial DDx (Attending) → Validate (LR completeness, Rare Cause Search)
- [ ] Phase 4: Multi-Persona Rounds → Validate (Hypothesis Space Audit after EACH round)
- [ ] Phase 5: Evidence Search (Round 4)
- [ ] Phase 6: Bayesian Probability Update
- [ ] Phase 7: Final Diagnosis → Validate (convergence, MNM exclusion, unification test)

Validation: After each phase, complete the corresponding checklist from [validation-checklist.md](references/safety/validation-checklist.md) and append to VALIDATION.md. Phase 1 Red Flag History is a HARD GATE — must pass before proceeding.

Phase 1: Case Intake & Setup

1. Read the SCENARIO.md file
2. Extract a URL-safe slug from the case (e.g., 45f-chest-pain-dyspnea)
3. Initialize case directory:

``bash python .claude/skills/cps/scripts/init_case.py "{slug}" "path/to/SCENARIO.md" --base-dir ./case ``

1. Parse structured data from the scenario: demographics, chief complaint(s), HPI, PMH, medications, vitals, PE, labs, imaging

Phase 2: Symptom Mapping & Textbook Extraction

1. Identify chief complaint(s) and map to textbook chapters using [references/core/chapter-map.md](references/core/chapter-map.md)
2. Read the relevant pre-distilled chapter references (typically 1-3). For example, chest pain + dyspnea:

- Read [references/chapters/ch09-chest-pain.md](references/chapters/ch09-chest-pain.md) - Read [references/chapters/ch15-dyspnea.md](references/chapters/ch15-dyspnea.md)

1. Focus on each chapter's:

- Differential diagnosis framework and pivotal findings - Likelihood ratios (LR+/LR-) for key findings - Diagnostic algorithms and decision points - Must-not-miss diagnoses and red flags

1. Always read [references/chapters/ch01-diagnostic-process.md](references/chapters/ch01-diagnostic-process.md) — it contains the core clinical reasoning methodology

Phase 3: Initial DDx — Round 1 (Attending Physician)

Adopt the Attending Physician (Internal Medicine) persona. See [references/core/personas.md](references/core/personas.md).

1. Write a problem representation: "This is a [age][sex] with [key PMH] presenting with [duration] of [chief complaint], associated with [key features], in the setting of [relevant context]"
2. Generate Top 10 DDx with pre-test probabilities:

- Use epidemiologic prevalence for the symptom in the relevant clinical setting - Adjust for demographics (age, sex, risk factors) - Flag all must-not-miss diagnoses regardless of probability — see [references/safety/ddx-framework.md](references/safety/ddx-framework.md) - Use both anatomic and pathophysiologic (VINDICATE) approaches

1. For each diagnosis, list key supporting and opposing evidence from the scenario
2. Write output to round-1.md using the template from [references/templates/output-templates.md](references/templates/output-templates.md)

Phase 4: Diagnostic Testing — Rounds 2-3

Round 2: Radiology & Pathology

Activate personas based on available data (see [references/core/personas.md](references/core/personas.md)):

• Radiologist: If imaging data present (CXR, CT, MRI, US, etc.)
• Pathologist: If lab/biopsy data present (CBC, CMP, UA, cultures, histology)

For each finding:

1. Provide structured interpretation
2. Look up the likelihood ratio — use [references/core/bayesian-reasoning.md](references/core/bayesian-reasoning.md) for common LRs
3. Apply LR to update the probability for each relevant diagnosis
4. Write output to round-2.md

Round 3: Subspecialty Consultation

Select 1-2 subspecialists based on the case presentation:

Each subspecialist:

1. Provides domain-specific deep analysis
2. Applies specialty-specific clinical decision rules and risk scores
3. Recommends additional testing with expected LRs
4. Writes output to round-3.md

Phase 5: Evidence Synthesis — Round 4 (EBM Specialist)

Adopt the EBM Specialist persona. Two evidence sources:

WebSearch (always available)

Search for current evidence on the leading diagnoses:

• "[diagnosis] [key finding] likelihood ratio" — find LR values
• "[diagnosis] diagnostic criteria guidelines 2024 2025" — current guidelines
• "[symptom] differential diagnosis evidence-based" — DDx evidence

robust-lit-review CLI (optional, requires .env)

If .env is configured with API keys:

lit-review review "[clinical question]" --target 50 --min-citescore 3.0

If .env is not configured or lit-review is not installed, note the limitation and proceed with WebSearch only.

For each piece of evidence found:

1. Assess level of evidence (systematic review > RCT > cohort > case series > expert opinion)
2. Extract relevant LR values, sensitivity/specificity data
3. Note any guideline recommendations that affect the DDx

Write output to round-4.md with an evidence summary table.

Phase 6: Bayesian Probability Update

After all rounds are complete, compile all likelihood ratios applied and run the calculator:

1. Prepare JSON input with all diagnoses, priors, and findings with LRs
2. Run the calculator:

``bash echo '{"diagnoses": [...]}' | python .claude/skills/cps/scripts/lr_calculator.py ``

1. Review the output probability table
2. Write to probability-table.md

See [references/core/bayesian-reasoning.md](references/core/bayesian-reasoning.md) for formulas and common LR values.

Phase 7: Final Diagnosis — Round 5 (Diagnostic Conference)

Synthesize all rounds into the final diagnosis:

1. Write round-5.md as a Diagnostic Conference combining all persona perspectives
2. Determine the leading diagnosis with its post-test probability
3. Explain the reasoning chain: how each key finding shifted the probability
4. Document why not for each alternative in the Top 10
5. List must-not-miss diagnoses that were ruled out, and how
6. Provide teaching points from the case
7. Write FINAL_DX.md using the template from [references/templates/output-templates.md](references/templates/output-templates.md)

Discover Subcommand (/cps discover [topic])

Search for challenging clinical cases to test the CPS skill.

1. Use WebSearch with queries from [references/discovery/case-discovery.md](references/discovery/case-discovery.md):

- "clinical problem-solving" site:nejm.org [topic] - "case records" site:nejm.org [topic] - "case report" site:casereports.bmj.com [topic]

1. Present a table of found cases with: title, source, year, difficulty estimate
2. When the user selects a case, extract key information and format as SCENARIO.md
3. Offer to run the full CPS workflow on the formatted scenario

Difficulty levels: See [references/discovery/case-discovery.md](references/discovery/case-discovery.md)

Round Subcommand (/cps round [case-dir] N)

Add an additional diagnostic round to an existing case:

1. Read the existing scenario and all prior rounds from the case directory
2. Ask which persona should lead Round N (or auto-select based on case needs)
3. The new persona reviews all prior work and contributes fresh analysis
4. Update the probability table with any new LR applications
5. Write round-N.md to the case directory

Review Subcommand (/cps review [case-dir])

Review and update an existing case:

1. Read SCENARIO.md and all round files
2. Identify gaps: missing LRs, unaddressed findings, overlooked diagnoses
3. Suggest additional rounds or evidence searches
4. Optionally regenerate FINAL_DX.md with updated reasoning

Retro Subcommand (/cps retro [case-dir] [correct-dx])

Retrospective evaluation when the correct diagnosis is known (e.g., NEJM answer reveal):

1. Read SCENARIO.md, all round files, and VALIDATION.md from the case directory
2. Copy template from [references/templates/retrospective-template.md](references/templates/retrospective-template.md)
3. Complete Diagnostic Trajectory Analysis (Q1-Q4: was correct Dx in Top 10? Which pivot? Earliest data?)
4. Grade each phase A-F with specific failure modes (anchoring, knowledge gap, trigger not fired, etc.)
5. Complete Safety Check Retrospective table (should-have-fired vs did-fire)
6. Document Lessons Learned (new safety checks, reference gaps, workflow changes)
7. Write RETROSPECTIVE.md to the case directory
8. Write PERFORMANCE.md using template from [references/templates/performance-metrics.md](references/templates/performance-metrics.md)
9. Update case/PERFORMANCE_TRACKER.md with this case's metrics row

SCENARIO.md Input Format

Cases can be free-form text, but this structure is recommended:

# [Brief Case Title]

## Chief Complaint
[Main reason for presentation]

## History of Present Illness
[Narrative of the current episode]

## Past Medical History
[Comorbidities, surgeries]

## Medications
[Current medications]

## Social History
[Smoking, alcohol, occupation, travel]

## Family History
[Relevant family conditions]

## Vital Signs
T: __ HR: __ BP: __/__ RR: __ SpO2: __%

## Physical Examination
[System-by-system findings]

## Laboratory Data
[Labs with values and reference ranges]

## Imaging
[Imaging study descriptions and findings]

## Additional Studies
[ECG, PFTs, cultures, biopsies, etc.]

Reference Files (Sub-Module Structure)

Only core/ loads every run. Others load on demand to minimize context usage.

core/ — Always loaded (~386 lines)

chapters/ — Load 1-3 per case (~80-120 lines each)

| [chapters/ch01-ch33](references/chapters/) | 33 files | Evidence-based DDx, LRs, algorithms per symptom |

templates/ — Load when writing output

| [output-templates.md](references/templates/output-templates.md) | 158 | FINAL_DX, round-N, probability-table templates | | [retrospective-template.md](references/templates/retrospective-template.md) | 106 | Post-case grading (for /cps retro) | | [performance-metrics.md](references/templates/performance-metrics.md) | 56 | Cross-case metrics definitions |

safety/ — Load during validation

| [validation-checklist.md](references/safety/validation-checklist.md) | 125 | Step validation per phase | | [ddx-framework.md](references/safety/ddx-framework.md) | 144 | VINDICATE + must-not-miss lists | | [rare-causes.md](references/safety/rare-causes.md) | 63 | NF1, KD, SCAD, PXE, Fabry... |

discovery/ — Load only for /cps discover

| [case-discovery.md](references/discovery/case-discovery.md) | 80 | Finding challenging cases from literature |

Scripts

Safety Checks (Built from Case Retrospectives)

1. Red Flag History Checkpoint

When: Patient has unusual PMH (MI at <40, stroke at <50, aneurysm at <60) Action: Before generating DDx, DEMAND the etiology of the prior event. Do NOT assume the current event shares the same mechanism without evidence.

2. Syndromic Screen

When: Vascular disease etiology is unclear in a patient <50 Action: Trigger a comprehensive review beyond the chief complaint:

• Complete skin exam (café-au-lait spots, neurofibromas, xanthomas, skin laxity)
• Eye exam (Lisch nodules, lens subluxation, angioid streaks)
• Vascular exam of all territories (pulse asymmetry, bruits)
• Connective tissue screen (joint hypermobility, arm span, pectus)
• 3-generation family pedigree with inheritance pattern analysis

3. Rare Cause Search

When: Diagnostic findings (cath, imaging, biopsy) reveal a pattern not matching common DDx Action: WebSearch for comprehensive etiologic reviews before defaulting to the "most common" cause. The textbook chapters cover common diagnoses; rare causes need active search.

4. Hypothesis Space Audit

When: After EVERY round, before finalizing probabilities Action: Ask explicitly: "Is there a diagnosis NOT in our Top 10 that could explain ALL the findings?"

• Bayesian updating can only redistribute probability among existing hypotheses
• If the true diagnosis isn't in the DDx, no amount of LR application will find it
• Force consideration of unifying diagnoses that explain seemingly unrelated findings

5. Genetic Pattern Recognition

When: Family history shows autosomal dominant premature vascular/cardiac disease Action: Consider genetic vasculopathies: NF1, Marfan, vascular Ehlers-Danlos, Loeys-Dietz, FMD, ACTA2

Key Principles

1. Evidence over intuition: Every probability shift must cite a likelihood ratio with source
2. Must-not-miss first: Always identify and explicitly rule out dangerous diagnoses
3. Bayesian discipline: Pre-test → apply LR → post-test. No skipping steps
4. Multi-perspective: No single persona owns the diagnosis — the conference decides
5. Transparent reasoning: Every round documents what changed and why
6. Hypothesis humility: The DDx is never closed. Unexpected findings demand expanding the hypothesis space, not forcing them into existing categories