utilization-review-assistant
Support utilization review decisions by evaluating medical necessity, level-of-care appropriateness, and length-of-stay justification against InterQual and Milliman criteria. Use when performing concurrent or retrospective utilization reviews, preparing peer-to-peer appeals, or assessing admission and continued-stay criteria.
Best use case
utilization-review-assistant is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Support utilization review decisions by evaluating medical necessity, level-of-care appropriateness, and length-of-stay justification against InterQual and Milliman criteria. Use when performing concurrent or retrospective utilization reviews, preparing peer-to-peer appeals, or assessing admission and continued-stay criteria.
Teams using utilization-review-assistant should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/utilization-review-assistant/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How utilization-review-assistant Compares
| Feature / Agent | utilization-review-assistant | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Support utilization review decisions by evaluating medical necessity, level-of-care appropriateness, and length-of-stay justification against InterQual and Milliman criteria. Use when performing concurrent or retrospective utilization reviews, preparing peer-to-peer appeals, or assessing admission and continued-stay criteria.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Utilization Review Assistant ## Overview Assist utilization review (UR) nurses and physicians in evaluating the medical necessity and appropriateness of healthcare services. This skill applies standardized criteria (InterQual, Milliman Care Guidelines, CMS guidelines) to patient clinical data to assess admission appropriateness, level-of-care determinations, continued-stay justification, and discharge readiness — supporting both concurrent and retrospective review workflows. ## When to Use - Evaluating whether an admission meets inpatient vs. observation criteria - Justifying continued inpatient stay for payer review - Preparing clinical summaries for peer-to-peer discussions - Performing retrospective utilization review for denials management - Assessing level-of-care appropriateness (ICU, step-down, med-surg, SNF) - Supporting discharge planning with readiness assessment ## Required Inputs | Input | Description | Format | |-------|-------------|--------| | Patient clinical data | Diagnoses, vitals, labs, treatments, clinical status | Structured object | | Admission details | Date, source, admitting diagnosis, attending | Structured object | | Current treatment plan | Active orders, medications, therapies | Structured list | | Review type | Concurrent, retrospective, pre-service | Enum string | | Payer information | Insurance type, plan specifics | Structured object | ## Methodology ### Step 1: Clinical Data Synthesis Compile the clinical picture required for criteria application: 1. **Severity of illness (SI)** indicators: vital sign instability, lab abnormalities, acute symptoms 2. **Intensity of service (IS)** indicators: IV medications, monitoring requirements, nursing intensity 3. **Current clinical trajectory**: improving, stable, worsening 4. **Comorbidity burden**: conditions affecting recovery and resource needs ### Step 2: Criteria Application Apply standardized utilization criteria systematically: **InterQual Framework:** Admission Criteria: - Severity of Illness (must meet threshold) - Intensity of Service (must match SI) Continued Stay Criteria: - Ongoing acute care needs - Active treatment requiring inpatient level - Clinical instability preventing safe discharge Discharge Criteria: - Clinical stability parameters met - Safe discharge plan in place - Follow-up arranged **CMS Two-Midnight Rule:** - Inpatient admission appropriate when physician expects patient to require hospital care spanning at least two midnights - If less than 2 midnights expected, observation status typically appropriate - Exceptions: certain procedures on the CMS Inpatient-Only List ### Step 3: Level-of-Care Assessment Determine the appropriate care level: | Level | Criteria Indicators | |-------|-------------------| | ICU | Mechanical ventilation, vasopressors, continuous monitoring, hemodynamic instability | | Step-down/Telemetry | Cardiac monitoring, frequent assessments, IV drips (non-vasopressor) | | Med-Surg Inpatient | IV medications, skilled nursing, diagnostic workup requiring observation | | Observation | Expected less than 2 midnights, diagnostic uncertainty, short-course treatment | | SNF/Rehab | Medically stable, requires daily skilled services, 3-midnight qualifying stay | | Home with services | Stable, homebound, needs skilled nursing/therapy visits | ### Step 4: Justification Documentation Build the clinical justification narrative: 1. **Medical necessity statement**: Why this level of care is required 2. **Alternative consideration**: Why lower-level care is insufficient 3. **Clinical evidence**: Specific findings supporting the determination (vitals, labs, functional status) 4. **Treatment plan rationale**: Active treatments requiring this care setting 5. **Discharge barriers**: What must change before step-down or discharge is safe ### Step 5: Determination and Recommendation Produce the UR determination: Determination Options: - APPROVED: Meets criteria for requested level of care - APPROVED WITH MODIFICATION: Meets criteria at a different level - PENDING CLINICAL: Additional information needed for determination - REFERRED TO PHYSICIAN REVIEWER: Does not clearly meet criteria, requires MD review - NOT MEETING CRITERIA: Does not meet medical necessity for this level ## Output Specification The output includes: **review_metadata**: review_type, review_date, reviewer, payer **patient_summary**: admitting_diagnosis (description, icd10), admission_date, current_los, current_level **clinical_indicators**: severity_of_illness indicators with values and threshold status, intensity_of_service items with frequency and inpatient requirement, clinical_trajectory (improving/stable/worsening) **criteria_evaluation**: criteria_set used (InterQual/Milliman/CMS), admission_criteria_met, continued_stay_criteria_met, discharge_criteria_met, level_of_care_appropriate, recommended_level, supporting_evidence **determination**: decision, rationale narrative, next_review_date, estimated_discharge, discharge_barriers with resolution plans **peer_to_peer_prep**: key_talking_points, anticipated_payer_concerns, supporting_clinical_evidence ## Analysis Framework ### Medical Necessity Evaluation Matrix | Factor | Meets Inpatient | Observation | Outpatient | |--------|----------------|-------------|------------| | Vital instability | Persistent abnormality | Transient, improving | Stable | | IV medication need | Continuous or frequent | Short-course (under 24h) | Oral equivalent available | | Monitoring | Continuous/q1-2h | q4-6h | Self-monitoring | | Procedure recovery | High-risk, anesthesia | Minor, moderate sedation | Office-based | | Fall/safety risk | High, requires supervision | Moderate | Low | | Functional status | Cannot perform ADLs | Limited ADLs | Independent | ### Length-of-Stay Benchmarking Compare actual LOS against geometric mean LOS (GMLOS) for the MS-DRG: - **Below GMLOS**: Efficient, ensure not premature discharge - **At GMLOS**: Expected, document ongoing medical necessity - **Above GMLOS**: Requires strong justification; identify and document barriers ## Examples **Input**: 72-year-old admitted for community-acquired pneumonia, day 3. On IV ceftriaxone/azithromycin. O2 4L NC, SpO2 94%. Temp 99.1F (down from 102F on admission). WBC 11.2 (down from 18.5). **UR Assessment**: - SI: Meets — supplemental O2 requirement, resolving leukocytosis - IS: Meets — IV antibiotics, supplemental oxygen - Trajectory: Improving (temp, WBC trending down) - Continued stay: APPROVED — still requiring supplemental O2 and IV antibiotics; transition to oral not yet safe (O2 dependent) - Estimated step-down: Day 4-5 when O2 weaned and tolerating oral antibiotics - Next review: Day 4 ## Guidelines 1. **Apply criteria objectively** — base determinations on documented clinical findings, not clinical intuition 2. **Document contemporaneously** — review findings must reflect clinical status at the time of review 3. **Consider the whole patient** — comorbidities, functional status, and social factors impact care needs 4. **Prepare for peer-to-peer** — document the strongest clinical justification points upfront 5. **Track avoidable days** — identify days where care could have been provided at a lower level ## Validation Checklist - [ ] Clinical data is current (within 24 hours for concurrent review) - [ ] SI and IS indicators are specific and measurable (not vague) - [ ] Criteria set used is specified and appropriate for the payer - [ ] Level-of-care recommendation is justified with specific clinical rationale - [ ] Discharge barriers are identified with resolution plans and timelines - [ ] Peer-to-peer talking points address likely payer objections - [ ] Review complies with UMCRA (Utilization Management Certification and Review Act) if applicable ## HIPAA Compliance Notes - UR data shared with payers must comply with minimum necessary and TPO (Treatment, Payment, Operations) permitted uses - Peer-to-peer discussions should be documented but limited to relevant clinical information - UR determinations stored in the medical record are subject to patient access rights - Third-party UR vendors must operate under a BAA - Restrict UR system access to authorized UR staff, physicians, and care management personnel
Related Skills
jetbrains-marketplace-reviews
Fetch and visualize reviews for any JetBrains Marketplace plugin. Use when (1) analyzing plugin review trends, (2) getting review statistics for a time period, (3) visualizing rating distributions, (4) monitoring user feedback. Triggers on requests like "get JetBrains reviews", "copilot plugin feedback", "JetBrains marketplace reviews", "visualize plugin ratings", "analyze JetBrains plugin reviews".
ethics-reviewer
This skill should be used when the user mentions "dark patterns", "accessibility", "a11y", "privacy", "tracking", "analytics", "notifications", "user data", "GDPR", "consent", "manipulation", "sustainability", "performance budget", or when building user-facing features that collect data, send notifications, display urgency, or gate access. Addresses ethical constraints in software design — manipulation, accessibility, privacy, and sustainability.
error-debugging-multi-agent-review
Use when working with error debugging multi agent review
Directus AI Assistant Integration
Build AI-powered features in Directus: chat interfaces, content generation, smart suggestions, and copilot functionality
datahub-connector-pr-review
This skill should be used when the user asks to "review my connector", "check my datahub connector", "review connector code", "audit connector", "review PR", "check code quality", or any request to review/check/audit a DataHub ingestion source. Covers compliance with standards, best practices, testing quality, and merge readiness.
cursor-rules-review
Audit Cursor IDE rules (.mdc files) against quality standards using a 5-gate review process. Validates frontmatter (YAML syntax, required fields, description quality, triggering configuration), glob patterns (specificity, performance, correctness), content quality (focus, organization, examples, cross-references), file length (under 500 lines recommended), and functionality (triggering, cross-references, maintainability). Use when reviewing pull requests with Cursor rule changes, conducting periodic rule quality audits, validating new rules before committing, identifying improvement opportunities, preparing rules for team sharing, or debugging why rules aren't working as expected.
create-assistant
Create and configure Vapi voice AI assistants with models, voices, transcribers, tools, hooks, and advanced settings. Use when building voice agents, phone bots, customer support assistants, or any conversational AI that handles phone or web calls.
cpm:review
Adversarial review of epic docs and stories. Agents from the party roster examine planning artifacts through their professional lens, challenging assumptions, spotting gaps, and flagging risks. Triggers on "/cpm:review".
contract-review-pro
专业合同审核 Skill,基于《合同审核方法论体系》提供合同类型指引和详细审核服务
codex-reviewer
Use OpenAI's Codex CLI as an independent code reviewer to provide second opinions on code implementations, architectural decisions, code specifications, and pull requests. Trigger when users request code review, second opinion, independent review, architecture validation, or mention Codex review. Provides unbiased analysis using GPT-5-Codex model through the codex exec command for non-interactive reviews.
codex-review
Two-pass adversarial review of design documents and implementation plans using OpenAI Codex CLI. Invokes Codex to review plans section-by-section (pass 1), then holistically (pass 2), feeding critique back for revision. Use when you have a design doc, architecture plan, or implementation plan that should be stress-tested before execution.
code-reviewer
Elite code review expert specializing in modern AI-powered code analysis, security vulnerabilities, performance optimization, and production reliability. Masters static analysis tools, security scanning, and configuration review with 2024/2025 best practices. Use PROACTIVELY for code quality assurance.