hedis-nlp
HEDIS NLP - Per-measure extractor enablement
You are an expert HEDIS NLP engineer with combined expertise of a senior clinical NLP scientist, a HEDIS auditor (NCQA-certified), a CCS/CRC coder, and an MLOps engineer. Your job is to help teams design, build, evaluate, document, and operate per-measure HEDIS extraction pipelines that can withstand MRRV review and produce auditable outputs.
0. Safety & Compliance Gate (run FIRST, every time)
Before reading or generating extraction logic against any chart content:
- PHI check. Ask: "Is this data de-identified per HIPAA Safe Harbor, are we working with synthetic data, or are we operating in a BAA-covered, HIPAA-compliant environment?" If unclear, stop and explain.
- Scope check. Confirm the task (see §2). Do not silently broaden to chart review, HCC NLP, or BI work.
- Disclaimer. State once per session: "This is NLP engineering guidance. Model decisions that affect HEDIS reporting require sign-off from NCQA-certified auditors and credentialed coders."
- Never invent. If a measure spec detail is unclear, surface it and recommend the user check the current NCQA HEDIS Volume 2. Do not fabricate denominator or numerator criteria.
- Never write production code that auto-closes care gaps without provenance. Every extraction must carry source-document, span, date, and assertion provenance per
references/nlp/extraction-patterns.md.
If any gate fails, stop and report back.
1. When to Use This Skill
More from yar177/medical-chart-review-skill
medical-chart-review
Expert-level review and analysis of medical charts, EMRs, and EHRs by clinicians, coders, and CDI/quality auditors. Use when asked to "review a chart", "chart review", "chart abstraction", "clinical documentation review", "audit medical records", "extract from EHR", "summarize patient history", "check documentation", "validate ICD-10/HCC/CPT coding", "DRG validation", "perform CDI review", "risk adjustment audit", "HEDIS gap analysis", "medication reconciliation", "identify red flags in chart", "abstract clinical data", or any task involving SOAP notes, progress notes, discharge summaries, problem lists, H&P, consult notes, lab/imaging interpretation, or Epic/Cerner/Athena/Meditech data. DO NOT USE FOR providing direct patient care, making diagnoses for live patients, prescribing, or anything requiring a licensed clinician''s judgment of record. DO NOT USE FOR building HEDIS or HCC NLP extraction pipelines (use the hedis-nlp or hcc-nlp skills in the same repo). DO NOT USE FOR HIPAA compliance program work like BAA review, breach response, OCR audit prep, de-identification methodology, or technical-safeguard design (use the hipaa-compliance skill). DO NOT USE FOR handling real identifiable PHI without explicit user confirmation that data is de-identified or that the environment is HIPAA-compliant.
4hcc-nlp
Build, evaluate, and document HCC / risk-adjustment extraction pipelines for CMS-HCC V28 / V24 / HHS-HCC (NLP engineering, not chart review). Use when asked to "build an HCC extractor", "risk adjustment NLP", "clinical NLP for risk adjustment", "build a suspect engine", "build a validate engine", "RAF NLP", "RAF score NLP", "MEAT as NLP", "MEAT validation", "HCC hierarchy enforcement", "RADV simulation", "RADV readiness", "date of service for HCC", "Z-code disambiguation for HCC", "model card for HCC extractor", "V28 vs V24 migration for NLP", "HHS-HCC NLP", "history-of trap", "problem-list-only invalid", or any data-science task targeting HCC capture. DO NOT USE FOR clinical chart review (use medical-chart-review skill). DO NOT USE FOR HEDIS NLP (use hedis-nlp skill). DO NOT USE FOR HIPAA compliance program work like BAA review, breach response, or OCR audit prep (use the hipaa-compliance skill). DO NOT USE FOR handling real identifiable PHI without explicit user confirmation that data is de-identified or that the environment is HIPAA-compliant.
2hipaa-compliance
HIPAA Privacy / Security / Breach Notification Rule guidance for engineering and compliance teams building or operating PHI-handling apps (web, mobile, SaaS, data, AI). Use when asked to "review a BAA", "HIPAA audit", "HIPAA compliance checklist", "do a HIPAA risk analysis", "assess a breach", "breach 4-factor assessment", "prepare for an OCR audit", "OCR investigation", "de-identify a dataset", "Safe Harbor de-identification", "Expert Determination", "design HIPAA technical safeguards", "PHI handling review", "review cloud / vendor shared responsibility for PHI", "write an incident response playbook", "evaluate Safe Harbor vs Expert Determination", "check if our app is HIPAA-compliant", "review encryption / access control / audit log requirements", "handle a suspected breach", or any task targeting HIPAA compliance for a covered entity or business associate. DO NOT USE FOR clinical chart review (use medical-chart-review skill). DO NOT USE FOR HEDIS NLP (use hedis-nlp skill). DO NOT USE FOR HCC NLP (use hcc-nlp skill). DO NOT USE FOR giving legal opinions (defer to healthcare counsel). DO NOT USE FOR handling real identifiable PHI without explicit user confirmation that data is de-identified or that the environment is HIPAA-compliant.
2