Workflow GuideFamily Medicine

Chronic Care Risk Stratification for Family Medicine

Optimize APCM revenue and patient outcomes with our chronic care risk stratification guide for family medicine practices using AI automation.

Transitioning from legacy CCM to the new APCM risk-stratification model requires family physicians to categorize multi-generational panels accurately. This guide outlines how to leverage AI-powered call handling to identify high-risk patients, streamline outreach, and ensure compliance with the 13 APCM service elements while maintaining whole-family health.

The Challenge

Family practices manage the highest volume of chronic conditions per patient but often lack the dedicated staff to identify and stratify panels for APCM eligibility, leading to missed revenue and gaps in preventive care for complex multi-generational households.

Step-by-Step Workflow

1

Data Extraction and Panel Cleaning

Run EMR reports to identify patients with two or more chronic conditions, filtering by AAFP-recommended diagnostic codes specifically relevant to multi-generational family panels.

Best Practices
  • Use automated SQL queries to pull ICD-10 codes for hypertension and diabetes.
  • Exclude patients who have not had an Annual Wellness Visit in the last 12 months.
Common Pitfalls
  • Including acute-only patients in chronic care management counts.
  • Failing to update panel status for deceased or relocated patients.
2

AI-Powered Initial Outreach

Deploy AI voice agents to contact identified patients, verifying current health status and identifying new social determinants of health (SDOH) that impact their clinical risk profile.

Best Practices
  • Set AI triggers for patients missing their monthly care coordination call.
  • Use natural language processing to detect changes in medication adherence.
Common Pitfalls
  • Relying on manual phone tag which delays the stratification process.
  • Ignoring patient preference for communication channels in multi-generational homes.
3

Risk Level Categorization

Assign patients to Tier 1 (Low), Tier 2 (Moderate), or Tier 3 (High) based on the APCM 13 service elements and the complexity of their specific comorbidities.

Best Practices
  • Reference AAFP risk-stratification templates for consistency across the practice.
  • Factor in all emergency room visits and hospitalizations from the last 6 months.
Common Pitfalls
  • Under-stratifying high-risk patients due to a lack of recent biometric data.
  • Over-stratifying based on patient age alone rather than condition complexity.
4

Care Plan Alignment

Map each risk tier to specific intervention protocols, ensuring high-risk patients receive 24/7 access to care coordination via AI-integrated call centers.

Best Practices
  • Integrate AI call logs directly into the EMR care plan section for HIPAA compliance.
  • Automate follow-up scheduling for Tier 3 patients immediately after their risk assessment.
Common Pitfalls
  • Creating generic care plans that do not address family-wide health dynamics.
  • Failing to document the required 20 minutes of non-face-to-face care for billing.
5

Continuous Monitoring and Re-Stratification

Use real-time AI analytics to monitor incoming patient calls and health updates, triggering an immediate risk-level review if a clinical decline is detected.

Best Practices
  • Review risk tiers quarterly during staff huddles to ensure accuracy.
  • Set automated alerts for recent hospital discharges to trigger immediate re-stratification.
Common Pitfalls
  • Treating stratification as a one-time event rather than a dynamic clinical process.
  • Inconsistent documentation of risk changes within the MIPS MVP pathway.

Expected Outcomes

1

Increased APCM enrollment through automated patient identification.

2

Improved MIPS quality scores via better chronic condition management.

3

Reduced administrative burden on family practice nursing staff.

4

Enhanced patient satisfaction through proactive, AI-driven outreach.

5

Higher Medicare Shared Savings Program performance through risk accuracy.

Frequently Asked Questions

APCM focuses on a risk-stratified, whole-person approach rather than just time-based tracking, aligning better with the AAFP's advanced primary care medical home model.

Yes, AI voice agents provide immediate response and triage for patients with chronic conditions, ensuring the practice meets the 13 essential service elements required for billing.

AAFP recommends using specific APCM codes that reflect the complexity and risk level of the managed panel, ensuring that the documentation supports the medical necessity of the tier assigned.

Accurate stratification allows for more precise reporting on chronic care measures, directly improving your performance score in the Value-Based Care framework and maximizing incentives.

Ready to transform your family medicine practice?

See how Tile Healthcare's AI call center can handle scheduling, triage, and patient communication for your practice.

Schedule a Demo
Chronic Care Risk Stratification for Family Medicine | Tile Health