| name | Access Friction Detector |
| description | Systematically identify and quantify barriers to healthcare access across scheduling, geographic, financial, cultural, and digital dimensions using CMS access standards and health equity frameworks. |
| metadata | {"display_name":"Access Friction Detector","short_description":"Detect and quantify healthcare access barriers by type","default_prompt":"Review my access friction and highlight top risks and next actions","version":"1.0.1","tags":["healthcare"],"icon_path":"assets/icon.png"} |
Access Friction Detector
Overview
This skill detects and quantifies friction points that prevent or delay patients from accessing healthcare services. It evaluates access across five core dimensions aligned with the Penchansky-Thomas Access Framework: availability, accessibility, accommodation, affordability, and acceptability. The analysis incorporates CMS network adequacy standards, ADA compliance requirements, and health equity benchmarks to produce actionable findings with regulatory context.
When to Use
- Evaluating access barriers after patient complaint trend analysis reveals scheduling or availability themes
- Preparing for NCQA accreditation or CMS network adequacy reviews
- Assessing health equity impact of service changes (clinic closures, hour reductions, telehealth transitions)
- Investigating disparities in appointment wait times, no-show rates, or patient leakage by demographic
- Supporting Community Health Needs Assessment (CHNA) access components
- Designing patient access center workflows or digital front-door strategies
Required Inputs
| Input | Description | Format |
|---|
appointment_data | Scheduling records with request date, appointment date, type, location, provider | De-identified CSV/JSON |
patient_demographics | Aggregate demographic data (zip code, language, insurance, age, disability) | JSON array |
facility_locations | Addresses and service hours for all access points | JSON array |
payer_mix | Insurance type distribution across patient population | JSON object |
complaint_data | Access-related complaints categorized by theme | JSON array |
telehealth_utilization | Virtual visit adoption rates by service and demographic | JSON object |
referral_data | Referral completion rates and time-to-appointment | JSON object |
Methodology
Step 1: Availability Analysis
- Calculate third-next-available appointment (TNAA) by provider, specialty, and location
- Benchmark against specialty-specific access standards:
- Primary care: 7 days or fewer for routine, 48 hours or fewer for urgent
- Specialty care: 14 days or fewer for routine referrals
- Behavioral health: 10 business days or fewer (per NCQA)
- Post-discharge follow-up: 7 days or fewer (per CMS readmission reduction)
- Identify capacity bottlenecks by day of week, time of day, and provider panel saturation
- Flag providers operating above 85% panel capacity as at-risk for access degradation
Step 2: Accessibility Analysis (Geographic and Physical)
- Compute drive-time and public-transit-time isochrones from patient zip code centroids to facilities
- Apply CMS time-distance standards:
- Primary care: 30 min / 15 miles (urban), 60 min / 60 miles (rural)
- Specialty: 60 min / 30 miles (urban), 120 min / 75 miles (rural)
- Assess ADA physical accessibility for each facility (parking, entrance, exam room, equipment)
- Evaluate public transportation proximity (within 0.5 miles from transit stop)
- Map geographic deserts where access standards are not met
Step 3: Accommodation Analysis
- Evaluate scheduling flexibility:
- Extended hours availability (before 8 AM, after 5 PM, weekends)
- Same-day and walk-in capacity
- Online self-scheduling adoption rate
- Cancellation and reschedule ease (number of steps, channels available)
- Assess communication accommodation:
- Language services availability (interpreter access within 10 minutes per LEP standards)
- TTY/TDD availability for deaf and hard-of-hearing patients
- Health literacy accommodations in intake and instructions
Step 4: Affordability Analysis
- Analyze cost-related access barriers:
- Percentage of patients with high-deductible health plans (HDHP)
- Financial assistance and charity care application rates and approval rates
- Copay and cost-sharing transparency at point of scheduling
- Prescription affordability barriers flagged in clinical notes (de-identified aggregate)
- Identify insurance-related friction:
- Prior authorization denial rates and average turnaround time
- Out-of-network referral frequency
- Medicaid acceptance gaps across specialties
Step 5: Acceptability Analysis
- Evaluate cultural and trust barriers:
- Provider demographic concordance rates (race, ethnicity, gender, language)
- Patient-reported experience scores segmented by demographic group
- Trust indicators from community health assessments
- Cultural competency training completion rates among staff
- Assess digital acceptability:
- Patient portal adoption by age group and digital literacy level
- Telehealth utilization disparities by race, age, rurality, and insurance
- Digital divide indicators (broadband access rates by patient zip code)
Step 6: Friction Scoring and Prioritization
- Assign composite friction scores per dimension (0-100):
- 0-25: Low friction (meeting standards)
- 26-50: Moderate friction (approaching risk)
- 51-75: High friction (below standards, action needed)
- 76-100: Critical friction (regulatory or equity risk)
- Aggregate into an overall Access Friction Index (AFI)
- Rank barriers by population impact (number of patients affected multiplied by severity)
Output Specification
access_friction_report:
analysis_date: date
population_scope: string
overall_afi_score: number
dimension_scores:
availability: number
accessibility: number
accommodation: number
affordability: number
acceptability: number
critical_barriers:
- dimension: string
barrier_description: string
affected_population: string
patient_count_estimate: number
severity: string
regulatory_reference: string
recommendation: string
geographic_gaps:
- area: string
population: number
nearest_facility_minutes: number
standard_exceeded_by: number
equity_disparities:
- metric: string
advantaged_group_value: number
disadvantaged_group_value: number
gap_percentage: number
recommendations:
- action: string
priority: string
estimated_impact: string
timeline: string
Analysis Framework
Use the Penchansky-Thomas 5A Framework enhanced with equity stratification:
| Dimension | Core Question | Key Metrics |
|---|
| Availability | Is there enough supply? | TNAA, panel size, provider FTE |
| Accessibility | Can patients get there? | Drive time, transit time, ADA compliance |
| Accommodation | Does the system flex for patients? | Hours, channels, language services |
| Affordability | Can patients pay? | HDHP rates, PA denials, charity care |
| Acceptability | Will patients engage? | Concordance, cultural competency, digital divide |
Examples
Example: Urban Health System Access Audit
- Availability: TNAA for dermatology = 47 days, CRITICAL (benchmark: 14 days)
- Accessibility: 12% of Medicaid patients live more than 45 min from nearest PCP, HIGH
- Accommodation: Online scheduling available for only 3 of 12 specialties, MODERATE
- Affordability: PA denial rate for behavioral health = 31%, HIGH
- Acceptability: Telehealth adoption among 65+ = 8% vs. 45% overall, Equity gap flagged
- Overall AFI: 62/100, HIGH friction, priority intervention recommended
Guidelines
- HIPAA Compliance: All analysis must use de-identified or aggregate data. Geographic analysis should use zip code or census tract level, never individual addresses. Complaint data must be stripped of PHI before ingestion.
- Regulatory Alignment: Reference CMS network adequacy standards (42 CFR 438.68), ADA Title III requirements, and Section 1557 language access provisions in findings.
- Equity-First Approach: Always stratify findings by race, ethnicity, language, insurance type, disability status, and rurality. Report disparities explicitly.
- Actionability: Every identified barrier must include at least one specific, implementable recommendation with timeline and responsible party.
- Community Input: Validate findings with patient advisory councils and community health workers when possible.
Validation Checklist
