status-determination
Evaluates hospital admission packets to determine if inpatient or observation status is clinically supported. Uses a two-step workflow with clinical review checkpoint. Analyzes ADT, ORU (labs/vitals), and MDM clinical notes against Medicare two-midnight rule and severity criteria. Use when reviewing admission status decisions, inpatient vs observation classification, utilization review, or two-midnight compliance audits.
Best use case
status-determination is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Evaluates hospital admission packets to determine if inpatient or observation status is clinically supported. Uses a two-step workflow with clinical review checkpoint. Analyzes ADT, ORU (labs/vitals), and MDM clinical notes against Medicare two-midnight rule and severity criteria. Use when reviewing admission status decisions, inpatient vs observation classification, utilization review, or two-midnight compliance audits.
Teams using status-determination 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/status-determination/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How status-determination Compares
| Feature / Agent | status-determination | 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?
Evaluates hospital admission packets to determine if inpatient or observation status is clinically supported. Uses a two-step workflow with clinical review checkpoint. Analyzes ADT, ORU (labs/vitals), and MDM clinical notes against Medicare two-midnight rule and severity criteria. Use when reviewing admission status decisions, inpatient vs observation classification, utilization review, or two-midnight compliance audits.
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
# Status Determination Skill
Screens inpatient/observation status decisions *after* the hospital has coded the patient but *before* claim submission—catching likely denials based on ICD-10 validation, documentation gaps, medical necessity alignment, and Two-Midnight Rule compliance.
**Target Users:** Hospital case managers, utilization review nurses, admitting physicians
## Architecture
This skill uses a **2-step workflow** with a clinical review checkpoint:
```
Step 1: Intake & Assessment
↓ (parse data, apply rubric, generate preliminary verdict)
↓ writes: waypoints/assessment.json
[User Checkpoint: Review & Confirm]
Step 2: Decision & Letter
↓ (capture reviewer input, generate report)
↓ writes: waypoints/decision.json
↓ writes: outputs/status-validation-report.md
```
## Quick Start
1. Provide paths to the admission packet (ADT, labs, vitals, MDM note)
2. Run Step 1: Intake & Assessment
3. Review the checkpoint summary
4. Confirm agreement or provide override reasoning
5. Receive the Status Validation Report
---
## Execution Flow
### Startup: Check for Existing Assessment
Before starting, check if `waypoints/assessment.json` exists:
```
IF waypoints/assessment.json exists AND status = "assessment_complete":
→ Ask: "Found existing assessment for [MRN]. Resume from Step 2? (Y/N)"
→ If Y: Skip to Step 2
→ If N: Start fresh from Step 1
ELSE:
→ Start from Step 1
```
---
### Step 1: Intake & Assessment
**Execute:** Read and follow `references/01-intake-assessment.md`
**Input:** 4 data files (ADT, ORU labs, ORU vitals, MDM note)
**Process:**
1. Parse admission data from all sources
2. Validate ICD-10 codes using MCP tools
3. Assess Severity of Illness against rubric thresholds
4. Assess Intensity of Services requirements
5. Check Two-Midnight Rule compliance (Medicare)
6. Identify documentation gaps
7. Generate preliminary verdict (SUPPORTED / AT RISK / LIKELY DENIAL)
**Output:** `waypoints/assessment.json`
**After Step 1:** Display summary and ask:
"Ready to proceed to Step 2: Clinical Review & Letter Generation? (Y/N)"
---
### Step 2: Decision & Letter
**Execute:** Read and follow `references/02-decision-letter.md`
**Input:** `waypoints/assessment.json`
**Process:**
1. Display clinical review checkpoint with assessment summary
2. Ask reviewer: "Do you agree with this assessment? (Agree/Disagree)"
3. If Disagree: Capture reasoning, flag for escalation
4. Generate Status Validation Report using `assets/letter-template.md`
**Output:**
- `waypoints/decision.json`
- `outputs/status-validation-report.md`
---
## Verdict Definitions
| Verdict | Meaning | Denial Risk | Action |
|---------|---------|-------------|--------|
| SUPPORTED | Evidence justifies status decision | LOW | Proceed to submission |
| AT RISK | Borderline — documentation gaps exist | MEDIUM | Strengthen documentation |
| LIKELY DENIAL | Evidence does not support status | HIGH | Change status or escalate |
---
## Reference Files
| File | Purpose |
|------|---------|
| [rubric.md](rubric.md) | Clinical criteria thresholds for labs, vitals, MDM |
| [tools.md](tools.md) | MCP tool integration patterns (ICD-10, CMS Coverage) |
| [references/01-intake-assessment.md](references/01-intake-assessment.md) | Step 1 detailed workflow |
| [references/02-decision-letter.md](references/02-decision-letter.md) | Step 2 detailed workflow |
| [assets/letter-template.md](assets/letter-template.md) | Report template with placeholders |
---
## Tool Dependencies
This skill integrates with:
- `ICD10_Codes` - Diagnosis code validation, hierarchy, specificity checks
- `CMS_Coverage` - Medicare coverage policy lookup (optional)
---
## Directory Structure
```
daisy-skill-1/
├── SKILL.md ← You are here (orchestrator)
├── rubric.md ← Clinical criteria thresholds
├── tools.md ← MCP tool integration
├── references/
│ ├── 01-intake-assessment.md ← Step 1 logic
│ └── 02-decision-letter.md ← Step 2 logic
├── waypoints/
│ ├── assessment.json ← Step 1 output (resume state)
│ └── decision.json ← Step 2 output
├── outputs/
│ └── status-validation-report.md ← Final report
├── assets/
│ └── letter-template.md ← Report template
├── sample-data/
│ ├── adt-a01-admission.json
│ ├── oru-r01-labs.json
│ ├── oru-vitals.json
│ └── mdm-t02-ed-note.json
└── examples/
└── case-001-sepsis-cap.md ← Reference output
```
---
## Sample Usage
```
User: Run status determination on the sample data in sample-data/
Claude:
1. Loads sample-data/adt-a01-admission.json, oru-r01-labs.json,
oru-vitals.json, mdm-t02-ed-note.json
2. Executes Step 1 (01-intake-assessment.md)
3. Writes waypoints/assessment.json
4. Displays checkpoint summary
5. User confirms agreement
6. Executes Step 2 (02-decision-letter.md)
7. Writes outputs/status-validation-report.md
8. Displays completion with verdict and action items
```Related Skills
workflow-status
Display project workflow progress by reading handoff documents in .docs/ directory. This skill should be used when users want to check their workflow status, see what phase they're in, or when other workflow skills need to verify prerequisites. Provides reusable prerequisite-checking templates for integration with other workflow skills.
statusline-customization
Configuration reference and troubleshooting for the statusline plugin — sections, themes, bar widths, and script architecture
statuscake-automation
Automate Statuscake tasks via Rube MCP (Composio). Always search tools first for current schemas.
oatda-video-status
Use when the user wants to check the status of an asynchronous video generation task from OATDA. Returns the video URL when generation is complete.
feature-status
Count features marked as @failing and write to status JSON file. Used to determine when the autonomous coding loop should end.
hig-components-status
Apple HIG guidance for status and progress UI components including progress indicators, status bars, and activity rings.
status
Use when the user returns to a session and asks about progress, current state, what happened, or where they left off. Provides status dashboard and execution recovery.
specs-status
현재 프로젝트의 EARS 스펙 상태 현황 표시
generate-status-report
Comprehensive system status report with services, infrastructure, performance metrics, and recommendations
conductor-status
Display project status, active tracks, and next actions
architecture-status
Reports on the health and state of architecture documentation (counts of ADRs, reviews, activity levels, documentation gaps). Use when the user asks "What's our architecture status?", "Show architecture documentation", "How many ADRs do we have?", "What decisions are documented?", "Architecture health check", or wants an overview/summary of documentation state. Do NOT use for listing team members (use list-members), creating new documents (use create-adr), or conducting reviews (use architecture-review or specialist-review).
asyncredux-action-status
Checks an AsyncRedux (Flutter) action's completion status using ActionStatus right after the dispatch returns. Use only when you need to know whether an action completed, whether it failed with an error, what error it produced, or how to navigate based on success or failure.