Preventive Care in the Modern Era

A practice manager pulls the patient panel report. Five hundred Medicare beneficiaries qualify for Advanced Primary Care Management under the new codes CMS finalized eighteen months ago. The practice has never submitted a single APCM claim.

She runs the calculation. Evidence-based panel distributions place 5% in Level 1, 82.5% in Level 2, and 12.5% in Level 3. At 2026 reimbursement rates, those 500 patients represent $359,850 in annual revenue. Sitting on the table. Unbilled.

The practice already provides care management. Monthly check-ins. Medication reviews. Care coordination. But the time logs never cooperate. A nurse spends 18 minutes on a complex call. The billing team holds the CCM claim because the threshold is 20 minutes. The work is documented. The patient benefited. The revenue is zero.

APCM was designed to solve exactly this problem. No time thresholds. No stopwatches. No claims lost because a clinically valuable 17-minute call does not reach an arbitrary billing minimum. Yet most practices have not transitioned. Not because APCM is obscure, but because the operational lift feels heavy and the regulatory framework is unfamiliar.

This chapter breaks down what APCM actually is, how it differs from CCM, what the tier structure means in revenue terms, and how practices can operationalize it without doubling their administrative burden.

Where This Code Came From

CMS finalized Advanced Primary Care Management in the Calendar Year 2025 Medicare Physician Fee Schedule final rule, published in the Federal Register on November 1, 2024. The codes took effect January 1, 2025.

The regulatory foundation sits in 42 CFR § 414.50 through § 414.52, which govern payment for physician services. CMS created APCM under its existing authority to establish payment for evaluation and management services, not through separate legislation. The stated policy goal: address chronic underutilization of care management services despite clear evidence of clinical benefit and cost savings.

The data driving the policy shift was stark. Published research using Medicare claims found CCM utilization among eligible fee-for-service beneficiaries rose from 1.1% in 2015 to just 3.4% in 2019. A separate ASPE analysis showed only 4.0% of potentially eligible beneficiaries received CCM services in 2019. Fewer than 10% of primary care physicians billed CCM codes during the program's first years.

The adoption barrier was structural. Basic CCM (CPT 99490) reimburses approximately $66 per month but requires documented evidence of at least 20 minutes of clinical staff time. A modeling study estimated that practices would need over one hundred Medicare patients consistently enrolled in CCM to recoup the salary of a single full-time registered nurse. Very few practices reached that threshold.

CMS designed APCM to remove the time-tracking barrier entirely while expanding eligibility to patients who do not meet CCM's two-chronic-condition requirement. The hypothesis: if the administrative friction is the problem, and the clinical model is sound, then eliminating the friction should increase adoption. Whether that hypothesis holds at scale remains to be tested.

Three Codes, Five Tiers

APCM uses three base HCPCS G-codes stratified by patient complexity and social risk, plus two behavioral health add-on codes introduced in the CY 2026 final rule. Each base code is billed once per patient per calendar month.

The 2026 rate increases reflect the new dual conversion factor structure finalized in the CY 2026 PFS final rule. CMS set separate conversion factors for the first time: $33.57 for qualifying Alternative Payment Model participants and $33.40 for non-qualifying participants, up from the single $32.35 conversion factor in 2025.

The increase incorporates a temporary 2.5% statutory bump from the One Big Beautiful Bill Act passed by Congress in July 2025, plus smaller permanent MACRA updates of 0.75% for APM participants and 0.25% for non-participants. The 2.5% increase is one-year only. Practices should plan for rate reversion in 2027 absent further Congressional action.

What Changed From CCM

The differences between APCM and CCM are structural, not cosmetic. Understanding them determines whether a practice can operationalize APCM or will stumble into compliance exposure.

The financial decision is not about maximizing per-patient revenue for every patient. It is about capturing revenue across the full panel, including patients who fall through CCM's time threshold and patients with fewer than two chronic conditions who were never CCM-eligible.

The 13 Service Elements

To bill any APCM code, the practice must be capable of providing all 13 service elements codified in 42 CFR § 414.50(b). CMS does not require every element to be delivered every month. It requires the practice to have the infrastructure and staffing to deliver any element when clinically appropriate.

Most practices currently providing CCM already meet the majority of these requirements. The elements that tend to require new infrastructure are performance measurement (Element 13), community service coordination (Element 10), and documented 24/7 access protocols (Element 3).

The Revenue Model

The financial case for APCM depends on three variables: panel size, tier distribution, and enrollment rate. Realistic ramp-up assumptions matter more than peak-state projections.

These are revenue projections, not profit figures. The cost to deliver APCM includes staff time for care management activities, technology for documentation and outreach, consent workflows, quality reporting infrastructure, and ongoing compliance monitoring. Practices should model their per-patient delivery cost against these revenue figures to determine true margin.

The per-patient revenue comparison to CCM varies by tier. For Level 1, APCM at $16 per month is lower than basic CCM (99490) at approximately $66 per month. But Level 1 patients would not have been eligible for CCM at all, since they have fewer than two chronic conditions. This is net new revenue.

For Level 2, APCM at $54 per month is lower than basic CCM. But the practice does not need to document 20 minutes. For patients who receive valuable care management but whose documented time falls short of the CCM threshold, APCM captures revenue that CCM leaves on the table.

For Level 3, APCM at $117 per month exceeds basic CCM rates and approaches complex CCM (99487) reimbursement levels. For QMB patients with high disease burden, APCM is the most favorable billing option available.

RHCs and FQHCs can now bill these add-on codes as well, following the sunsetting of the consolidated G0511 care management code on September 30, 2025. This is a meaningful expansion for safety-net providers who previously could not separately bill for behavioral health integration within care management.

The Evidence

APCM is new. The codes took effect January 1, 2025. No peer-reviewed studies have yet evaluated APCM-specific outcomes, cost reduction, or quality improvement. Any practice or vendor claiming APCM outcome data is projecting from adjacent programs, not reporting from APCM itself.

The evidence base that supports the model comes from CCM research and from the structural arguments CMS made in the CY 2025 final rule.

The Mathematica evaluation of CCM's first 18 months, commissioned by CMS, found that CCM recipients had significantly higher rates of advance care planning, more primary care visits, lower hospitalization rates, and lower emergency department utilization compared to a matched control group. After excluding patients with only one month of CCM, the CCM group cost Medicare approximately $95 less per month. CMS estimated net savings of over $38 million after accounting for CCM payments.

These results were promising but came from a period of early adoption by motivated practices. Whether the same associations hold at broader scale remained unclear.

A separate study found CCM adoption rates remained below 4% of eligible beneficiaries through 2019, suggesting that the time-based requirements functioned as a barrier to the very program that had demonstrated positive results.

CMS cited this adoption gap explicitly in the CY 2025 PFS final rule as the rationale for APCM. The policy argument: a program that works in motivated practices but fails to scale because of administrative burden is not fulfilling its purpose. Removing the time threshold and expanding eligibility to all Medicare beneficiaries is CMS's structural response.

Whether APCM achieves the same or better outcomes than CCM at broader scale remains unknown. The hypothesis is sound: removing administrative friction should increase adoption, and broader adoption of care management should improve outcomes for more patients. But hypotheses are not evidence. Practices should build their financial model on the revenue opportunity, which is concrete, and track their own clinical outcomes to determine whether the program produces the quality improvements they expect.

The evidence on cost-sharing is also relevant here. The American College of Physicians, the American Academy of Family Physicians, and the National Academies of Sciences, Engineering, and Medicine have all urged CMS to remove cost-sharing for APCM, arguing that any copay could suppress enrollment, particularly among lower-income patients who do not qualify for QMB. No published data yet quantifies this effect for APCM specifically, but CCM studies consistently show that cost-sharing reduces utilization of preventive services among Medicare beneficiaries.

In the CY 2026 proposed rule, CMS issued a Request for Information on whether APCM services should be designated as preventive services, which would allow cost-sharing to be waived under existing statutory authority. CMS also solicited comment on whether preventive services like the Annual Wellness Visit should be bundled into APCM codes. These are signals, not policy. As of February 2026, cost-sharing remains in place. But the direction of the conversation matters. Practices should track this closely. If CMS acts on the RFI in the CY 2027 proposed rule, the enrollment barrier drops significantly.

What Ownership Looks Like When Technology Supports the Workflow

Here is a concrete model for operationalizing APCM without doubling administrative burden.

Who This Model Misses

Every model has boundaries. Naming them protects the people who trust it.

Cost-sharing blocks enrollment for some patients. Non-QMB Medicare patients owe 20% coinsurance for APCM services. For a Level 2 patient, that amounts to approximately $11 per month, or $132 per year. For patients on fixed incomes who are not quite poor enough to qualify for QMB, that amount may deter consent. The highest-risk patients who are not dually eligible face the steepest barrier.

CMS has actively sought comment on whether APCM should be designated as a preventive service to waive cost-sharing, and the National Academies formally recommended this step. Until CMS acts, enrollment will be suppressed among the patients who need care management most. This is the single most consequential policy decision remaining for APCM's trajectory.

CCM disparities will likely persist in APCM. Published data on CCM showed lower utilization among Black and Hispanic beneficiaries, rural populations, and dually eligible patients. CMS has not published APCM utilization data stratified by race, ethnicity, or geography. Given that the structural barriers to care management enrollment—including trust, language, transportation, and health literacy—are the same regardless of the billing code, practices should assume that APCM will reproduce existing disparities unless they actively design outreach to counter them.

AI-based outreach does not reach everyone. The phone-based model assumes patients can be reached by phone, understand the conversation in English or Spanish, have sufficient health literacy to process the information, and trust the caller enough to consent. Patients who are transient, unhoused, hearing impaired, or disconnected from telecommunications infrastructure require different approaches. Patients with cognitive impairment may need caregiver involvement at every step.

The 13 service elements assume infrastructure that many practices lack. Documented 24/7 access, community service coordination, electronic care plans accessible to patients, and quality reporting all require technology and workflow investment. Practices in rural or underserved areas with limited EHR functionality, short staffing, and no established community organization partnerships face higher barriers to APCM readiness.

The evidence base is thin. APCM has been billable for approximately 14 months. No published research evaluates APCM-specific outcomes. No publicly available claims data quantifies APCM adoption rates across the Medicare population. Anecdotal reports from industry vendors suggest early adoption has been concentrated among practices already participating in ACOs or CMS innovation models, but no rigorous data confirms this. Practices adopting APCM are building the plane while flying it. The regulatory framework is defined. The clinical evidence is not.

APCM does not replace clinical judgment. The AI system described above does not make clinical decisions. It does not diagnose. It does not prescribe. It does not override the physician's care plan. It enforces follow-through on decisions the physician has already made. Patients whose conditions are clinically ambiguous, whose social contexts are complex beyond what an algorithm can assess, or whose trust in the system has been damaged by prior harm need human clinicians making human decisions with human presence. Technology does not solve that. It was never meant to.

What Changes This Week

For a practice reading this and considering APCM, the sequence is short.

First, pull your Medicare panel list. Count the number of patients with zero or one chronic condition, two or more chronic conditions, and QMB status. This gives you the tier distribution that determines your revenue model. Use your EHR's problem list or claims-based chronic condition counts. Typical distributions: 5% Level 1, 82.5% Level 2, 12.5% Level 3, but your panel may differ.

Second, audit your CCM time logs from the last three months. Count how many enrolled patients fell below the 20-minute threshold and had claims held or not submitted. That number represents the immediate revenue APCM recovers. If you are losing 20% of your CCM-eligible patients to time-threshold misses, APCM captures that revenue without changing your care delivery model.

Third, check your MIPS reporting status. If you are in a Medicare Shared Savings Program ACO, ACO REACH, Making Care Primary, or Primary Care First, you already satisfy the performance measurement requirement and benefit from the higher $33.57 conversion factor. If you are MIPS-eligible but not in one of these models, confirm that your practice is reporting the Value in Primary Care MVP. The 2025 performance year is the first reporting period, and data submission is underway now for the 2027 MIPS payment year.

Fourth, assign one person to map your practice's current capabilities against the 13 service elements. Identify which elements you already deliver, which require new workflows, and which require technology or staffing investment. The gap list determines your implementation timeline. Most CCM practices already meet 9-10 of the 13 elements. The remaining gaps are addressable.

Fifth, model the revenue. Use your actual tier distribution and a conservative enrollment ramp—start with 30% enrollment in month one, scaling to 60% by month six. Compare projected APCM revenue to your current CCM revenue. Include the BHI add-on codes (G0568, G0569, G0570) if your practice screens for behavioral health conditions during AWVs or office visits. The comparison will tell you whether the transition generates net new revenue or whether a hybrid model, with some patients on CCM and some on APCM, optimizes your billing.

The calculation is straightforward. The implementation requires discipline. The revenue is sitting on the table.

What Pear Health Builds

The model described in this chapter is what Pear Health operationalizes.

Pear ingests EMR data, claims history, and payer rosters to stratify every Medicare patient by APCM tier. The AI phone agent contacts patients to obtain consent, complete medication reconciliation, review care plans, and deliver preventive care reminders. Every interaction documents directly into the EHR in real time. Claims are generated and submitted monthly without manual time tracking.

For practices currently running CCM and losing revenue to time-threshold attrition, Pear transitions eligible patients to APCM and maintains CCM for patients where the higher per-patient reimbursement justifies the documentation burden. The system applies clean eligibility logic at intake so that no patient is billed under both programs in the same month. Mutual exclusivity is enforced automatically.

The 2026 behavioral health add-on codes are integrated into the workflow. Patients flagged through AWV depression screens (PHQ-9 ≥ 10) or behavioral health diagnoses receive BHI services with documentation and billing handled automatically. The system routes patients to the appropriate level—general BHI (G0570) for screening and brief interventions, full CoCM (G0568, G0569) for patients requiring psychiatric consultation and ongoing behavioral health care management.

The physician reviews tier assignments. The physician approves care plans. The physician decides what interventions are clinically appropriate. Pear owns the outreach, the consent workflow, the monthly care management delivery, the documentation, the claim submission, and the quality reporting. The practice does not need to hire additional staff to stand up a care management program. It does not need to track minutes. It does not need to choose between doing the work and proving the work was long enough to bill for.

This is not a vision document. This is a live workflow running in early-adopter practices today, with measurable monthly revenue and documented care management activities at every stage.

If your practice is losing care management revenue to time thresholds, or has not started billing APCM because the 13 service elements feel like a compliance mountain, we should talk. Not about what APCM could be. About what changes in your workflow next week.