Skip to content

aizech/clinical-skills / skills/workflow-coordination/care-gap-closure

care-gap-closure

Verified

Ensure recommended imaging is completed and close care gaps in radiology. Also use when optimizing imaging completion rates, tracking screening compliance, or identifying patients overdue for recommended imaging studies.

FileKindSize
SKILL-Computer.mdmarkdown9.1 KB
SKILL.mdskill md9.1 KB
evals/evals-Computer.jsonconfig1.2 KB
evals/evals.jsonconfig1.2 KB

SKILL-Computer.md

View on GitHub
---
name: care-gap-closure
description: Ensure recommended imaging is completed and close care gaps in radiology. Also use when optimizing imaging completion rates, tracking screening compliance, or identifying patients overdue for recommended imaging studies.
---

# Care Gap Closure

You are an expert in radiology care gap management. Your role is to help identify patients missing recommended imaging and facilitate closure of these gaps.

## Care Gap Types

### Screening Gaps

| Screening | Population | Modality | Frequency |
|-----------|------------|----------|-----------|
| Lung cancer | 50-80yo, 20+ pack-year smokers | Low-dose CT | Annual |
| Breast cancer | Women 40-75 | Mammography | Annual |
| Colorectal cancer | Adults 45-75 | Colonoscopy/CT colonography | Every 10 years |
| Cervical cancer | Women 21-65 | Pap smear | Varies |
| Abdominal aortic aneurysm | Men 65-75, smokers | Ultrasound | One-time |

### Follow-up Gaps

- Incidental findings not followed
- Abnormal screening results pending resolution
- Prior imaging recommendations incomplete

### Diagnostic Gaps

- Imaging ordered but not completed
- Referral placed but no appointment scheduled
- Prior test results requiring action

## Care Gap Identification

### Patient Cohort Query

```python
CARE_GAP_QUERIES = {
    "lung_cancer_screening": {
        "criteria": {
            "age_range": [50, 80],
            "smoking_history": ">=20 pack-years",
            "smoking_status": ["current", "quit_within_15_years"]
        },
        "exclusion": {
            "prior_lung_cancer": True,
            "prior_chest_ct_12months": True
        }
    },
    "mammography_screening": {
        "criteria": {
            "gender": "Female",
            "age_range": [40, 75]
        },
        "exclusion": {
            "bilateral_mastectomy": True
        },
        "frequency": "Annual",
        "lookback_period": "12 months"
    }
}
```

### Gap Detection Logic

```python
def identify_care_gaps(patient_data, screening_guidelines):
    """Identify care gaps for a patient population."""
    
    gaps = []
    
    for patient in patient_data:
        patient_gaps = []
        
        # Check each screening guideline
        for guideline in screening_guidelines:
            if patient_meets_criteria(patient, guideline.criteria):
                if not patient_has_recent_screening(patient, guideline):
                    patient_gaps.append({
                        "patient_id": patient.id,
                        "gap_type": guideline.type,
                        "gap_reason": guideline.description,
                        "due_date": calculate_due_date(patient, guideline),
                        "urgency": guideline.urgency,
                        "intervention": guideline.recommended_action
                    })
        
        gaps.extend(patient_gaps)
    
    return gaps
```

## Intervention Strategies

### Outreach Tiers

```python
OUTREACH_TIERING = {
    "tier_1_immediate": {
        "criteria": "STAT or urgent finding",
        "methods": ["Direct phone call", "Urgent message"],
        "timeframe": "Same day",
        "escalation": "If no response in 4 hours"
    },
    "tier_2_scheduled": {
        "criteria": "Routine screening due",
        "methods": ["Patient portal", "Letter", "Phone reminder"],
        "timeframe": "30 days before due",
        "escalation": "If no response in 14 days"
    },
    "tier_3_overdue": {
        "criteria": "Past recommended timeframe",
        "methods": ["Phone call", "Provider notification"],
        "timeframe": "On due date",
        "escalation": "Weekly for 4 weeks, then provider escalation"
    }
}
```

### Patient Communication Scripts

```markdown
# Care Gap Closure Phone Script

"Hello, may I speak with [Patient Name]?

My name is [Name] from [Facility]. I'm calling about your 
healthcare.

Our records show that you are due for a [screening type] 
[as part of your routine healthcare / based on your health history].

This screening is important because [brief reason].

How would you like to schedule this?

If now is not a good time, I can help you find a time that 
works better for you.

[If patient asks why]: This test helps [reason]. It is 
recommended for people with [criteria] and is covered by most 
insurance plans.

[If patient resistant]: I understand. Would you like me to 
have your healthcare provider reach out to discuss whether 
this screening is right for you?"

CLOSING:
"Great, let me help you schedule that now. [Proceed to 
scheduling] 

Or, if you'd prefer, I can send you information through the 
patient portal to schedule when you're ready.

Thank you for your time."
```

## Gap Closure Workflow

### Closure Documentation

```python
CARE_GAP_CLOSURE = {
    "patient_id": "123456",
    "gap_type": "lung_cancer_screening",
    "identified_date": "2026-03-01",
    "outreach_attempts": [
        {
            "date": "2026-03-01",
            "method": "patient_portal_message",
            "result": "no_response"
        },
        {
            "date": "2026-03-08",
            "method": "phone_call",
            "result": "scheduled",
            "appointment_date": "2026-03-20"
        }
    ],
    "closure": {
        "status": "closed",
        "closure_date": "2026-03-20",
        "method": "completed",
        "study_type": "Low-dose CT Chest",
        "result": "Lung-RADS 2 - benign findings"
    },
    "notes": "Patient scheduled after one outreach call"
}
```

### Provider Escalation

```markdown
SUBJECT: Care Gap Escalation - Patient Not Responsive

Patient: [Name], MRN [Number]
Care Gap: [Type of screening/follow-up]
Due Date: [Date]
Days Overdue: [Number]

Intervention History:
- [Date]: Patient portal message - No response
- [Date]: Phone call - No answer
- [Date]: Letter sent - No response

Recommended Action:
[ ] Provider phone call to patient
[ ] Discuss at next visit
[ ] Remove from reminder list (patient declined)
[ ] Other: [Notes]

Patient Contact Information:
Phone: [Number]
Email: [Email]

Please advise on next steps.
```

## Quality Metrics

### Care Gap Dashboard

```python
CARE_GAP_METRICS = {
    "identification_rate": {
        "description": "% of eligible patients with identified gaps",
        "calculation": "Patients with gaps / Eligible patients",
        "target": "Measure and report"
    },
    "closure_rate": {
        "description": "% of identified gaps that are closed",
        "calculation": "Gaps closed / Gaps identified",
        "target": ">80%"
    },
    "timeliness": {
        "description": "% of gaps closed within timeframe",
        "calculation": "Closed within standard / Total closed",
        "target": ">75%"
    },
    "patient_contact": {
        "description": "% of gaps with documented patient contact",
        "calculation": "Contacted / Gaps requiring action",
        "target": "100%"
    }
}
```

## Reporting Template

```markdown
# CARE GAP CLOSURE REPORT
## [Month/Quarter/Year]

### Executive Summary
- Total care gaps identified: [Number]
- Care gaps closed: [Number]
- Closure rate: [Percentage]
- Average time to closure: [Days]

### By Gap Type

| Gap Type | Identified | Closed | Rate | Avg Days to Close |
|---------|-----------|--------|-------|-------------------|
| Lung cancer screening | 50 | 42 | 84% | 21 |
| Mammography | 75 | 68 | 91% | 14 |
| Incidental findings follow-up | 30 | 24 | 80% | 28 |

### Interventions Used

| Method | Attempts | Successful | Rate |
|--------|----------|-----------|------|
| Patient portal | 100 | 35 | 35% |
| Phone call | 80 | 50 | 63% |
| Letter | 25 | 5 | 20% |
| Provider escalation | 15 | 12 | 80% |

### Outcomes
- Abnormal findings detected: [Number]
- Cancers diagnosed: [Number]
- Patients educated: [Number]

### Recommendations
1. [Priority improvement area]
2. [Secondary improvement area]
```

## Related Skills

- **followup-tracking**: For incidental finding follow-up
- **patient-results-letter**: For patient communication
- **imaging-referral**: For referral management
- **guideline-integration**: For evidence-based screening criteria

## Examples

### Example 1: Identify Lung Cancer Screening Gaps

```
Find patients due for lung cancer screening who haven't been screened
```

```python
query = {
    "screening_type": "lung_cancer_screening",
    "criteria": {
        "age": {"min": 50, "max": 80},
        "smoking_history": ">=20 pack-years",
        "quit_date": "none or <15 years ago"
    },
    "exclusions": {
        "prior_lung_cancer": True,
        "prior_chest_ct": {"months": 12}
    },
    "lookback": "12 months"
}

# Returns: List of patients meeting criteria but without recent screening
```

### Example 2: Close a Care Gap

```
Help close the care gap for a patient who missed their screening mammogram
```

```python
closure_workflow = {
    "patient_id": "123456",
    "gap": "mammography_screening",
    "steps": [
        {"action": "contact_patient", "method": "phone_call"},
        {"action": "schedule", "study": "digital_mammography"},
        {"action": "remind_prep", "info": "No deodorant day of"},
        {"action": "document_result", "status": "completed"}
    ],
    "outcome": {
        "status": "closed",
        "appointment_completed": "2026-04-15",
        "result": "BI-RADS 1 - Negative"
    }
}
```
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall