care-gap-identification

name: care-gap-identification description: Identify missing or overdue care steps against HEDIS, STAR, USPSTF, and disease-specific quality measures for individual patients or populations. Use when performing care gap analysis, generating patient outreach lists, preparing for quality measure reporting, or supporting value-based care performance improvement.

metadata: display_name: "Care Gap Identification" short_description: "Find missing care steps against HEDIS and STAR measures" default_prompt: "Review my care gap and highlight top risks and next actions" version: "1.0.1" tags: - healthcare

icon_path: "assets/icon.png"

Care Gap Identification

Overview

Systematically identify missing, overdue, or incomplete care activities by comparing patient clinical records against evidence-based quality measures and preventive care guidelines. This skill evaluates compliance with HEDIS (Healthcare Effectiveness Data and Information Set), CMS Star Ratings, USPSTF recommendations, and disease-specific protocols to surface actionable care gaps for individual patients or population panels.

When to Use

• Running care gap analyses for patient panels or individual patients
• Preparing for HEDIS or STAR rating measurement periods
• Generating patient outreach lists for preventive services
• Supporting value-based care contract performance
• Identifying gaps before annual wellness visits or chronic care appointments
• Building quality dashboards with gap closure tracking

Required Inputs

Methodology

Step 1: Measure Applicability Determination

Identify which quality measures apply based on patient demographics and conditions:

1. Evaluate age, sex, and insurance type against measure denominators
2. Check active diagnoses for disease-specific measures (diabetes, hypertension, depression)
3. Apply exclusion criteria (hospice, terminal illness, denominator exclusions)
4. Generate the applicable measure list for this patient

Step 2: Service History Evaluation

For each applicable measure, check if the required service has been completed:

• Screenings: Was the test performed within the required interval?
• Immunizations: Is the vaccine series complete and current?
• Chronic disease management: Were required labs and visits completed?
• Medication adherence: Does PDC (Proportion of Days Covered) meet threshold?
• Follow-up care: Were post-event follow-ups completed within required timeframes?

Step 3: Gap Classification

Classify each gap by type and urgency:

Step 4: Priority Scoring

Score each gap by clinical impact and measure weight:

Priority Factors:

• Clinical urgency (immediate health impact vs. long-term prevention)
• Measure weight in quality programs (triple-weighted STAR measures carry more impact)
• Time sensitivity (approaching measure close date, overdue duration)
• Patient risk level (high-risk patients have amplified gap impact)
• Contractual significance (tied to value-based payment)

Step 5: Intervention Recommendation

For each identified gap, recommend closure actions:

• Specific service needed with CPT/HCPCS code
• Preferred provider or care setting
• Patient outreach method (phone, portal message, mail)
• Scheduling guidance (combine with upcoming visit if possible)
• Documentation requirements for measure credit

Output Specification

The output includes:

patient_summary: demographics, risk_level, payer, applicable_measure_count

applicable_measures: measure_id, measure_name, domain (preventive/chronic/behavioral/medication), denominator_criteria_met, exclusions_evaluated

identified_gaps: measure_id, measure_name, gap_type, gap_description, last_completed_date (if ever), due_date, overdue_by, priority_score, clinical_urgency, closure_action with CPT code and service description, estimated_effort

gap_summary_by_domain: domain, total_measures, gaps_found, gap_rate

closed_measures: measures where criteria are met (for completeness tracking)

outreach_recommendations: patient contact preferences, suggested outreach message, scheduling recommendations

Analysis Framework

Key HEDIS/STAR Measures

Medication Adherence Measures (Triple-Weighted in STAR)

• Diabetes medications: PDC threshold 80%
• RAS antagonists (hypertension): PDC threshold 80%
• Statins (cholesterol): PDC threshold 80%

PDC = (Total days covered by fills in period) / (Days in measurement period) x 100

Examples

Input: 58-year-old female with type 2 diabetes, hypertension, on metformin and lisinopril. Last HbA1c: 14 months ago. Last mammogram: 3 years ago. Last eye exam: 2 years ago. Statin not prescribed despite ASCVD risk score >20%.

Gaps Identified:

1. CDC-HbA1c: OVERDUE (14 months, annual required) — Priority: HIGH. Action: Order HbA1c lab
2. BCS: OVERDUE (3 years, every 2 years required) — Priority: HIGH. Action: Schedule mammogram
3. CDC-Eye: OVERDUE (2 years, annual required) — Priority: MEDIUM. Action: Refer to ophthalmology
4. SPC: NOT MET (statin not prescribed, ASCVD risk >20%) — Priority: HIGH. Action: Prescribe statin therapy
5. CDC-Kidney: UNKNOWN (no eGFR/uACR on file) — Priority: MEDIUM. Action: Order renal panel with uACR

Guidelines

1. Apply exclusions before flagging gaps — ensure patients are truly in the measure denominator
2. Check supplemental data sources — patients may have completed services outside the primary system
3. Combine gap closure with existing visits — maximize efficiency by bundling services
4. Prioritize triple-weighted measures for STAR rating impact
5. Track gap closure rates over time to measure program effectiveness

Validation Checklist

• All applicable measures are identified based on demographics and conditions
• Exclusion criteria are properly evaluated before flagging gaps
• Gap dates are calculated correctly against measurement period requirements
• Priority scoring reflects both clinical urgency and quality program impact
• Closure actions include specific CPT/HCPCS codes and service descriptions
• Medication adherence gaps include current PDC calculations
• Output distinguishes between "never done" and "overdue" gaps

HIPAA Compliance Notes

• Care gap data involves PHI and must be processed within BAA-covered systems
• Patient outreach for gap closure must comply with communication preferences and consent
• Population-level gap reports should be de-identified for quality improvement analysis
• Share gap data with contracted providers only under appropriate data use agreements
• Medication adherence data sourced from pharmacy claims requires appropriate authorization chains