Q3 2026 is the last practical window for Medicare Advantage plans to build encounter-based risk adjustment workflows before the CY2027 unlinked chart review exclusion takes effect. CMS finalized the rule in its April 2026 Rate Announcement, projecting a 1.53% risk score reduction, approximately $7.2 billion in reduced payments industry-wide. The diagnoses that will count toward 2027 risk scores are being generated right now, in the encounters happening this summer and fall. Organizations that use Q3 to measure their unlinked chart review exposure, stand up encounter-based capture workflows, and implement internal QA processes aligned to RADV methodology will enter 2027 with the infrastructure CMS is requiring. Those that wait will be managing the gap retrospectively, which is exactly the model CMS is moving away from.
Encounter-based risk adjustment is the practice of capturing and documenting HCC diagnoses in direct connection with clinical visits, rather than through retrospective medical record review. Under the CY2027 CMS final rule, encounter-based diagnoses are the ones that count. Unlinked chart review records are excluded from risk score calculation. Encounter-linked documentation means there is a clinical record of a specific visit where the condition was assessed, monitored, treated, or evaluated in the current benefit year. This is the documentation standard every workflow built in Q3 needs to produce.
Second-level coding review is a quality control step in which a second coder or reviewer validates diagnoses after the initial coding pass, before submission. In the context of CY2027 risk adjustment, it functions as an internal checkpoint to catch unsupported or encounter-unlinked diagnoses before they become audit findings. Second-level review should happen near-real-time, tied to specific encounters, not as a batch retrospective review weeks later. Organizations implementing this in Q3 should design it as a continuous process embedded in the encounter workflow, not a periodic retrospective exercise.
The benefit year opens January 1, 2027, and the encounter-based documentation that CMS will accept for risk adjustment needs to be happening before that date. That means the workflows producing encounter-linked diagnoses need to be operational now, not in November.
Q4 is Annual Enrollment Period. New member volume spikes, operational teams are stretched, and standing up new clinical workflows during AEP is the wrong time to find out they don’t work. Q3 is the last practical window to build, test, and stabilize encounter-based processes before they get stress-tested by real volume.
The audit environment adds its own urgency. CMS is scaling RADV from roughly 60 MA plans audited per year to all eligible plans, approximately 550, with record review counts increasing from 35 to up to 200 per plan. PY2020 audits began in February 2026, with new cycles initiating approximately every three months. The infrastructure that protects an organization in a RADV audit is the same infrastructure that produces compliant risk adjustment submissions: encounter-linked diagnoses, supported documentation, and internal QA processes that find gaps before CMS does. RADV audit readiness and risk adjustment compliance are now the same operational problem, and Q3 is the window to solve both. For a deeper look at what the expanded audit program means operationally, Inferscience’s analysis of RADV audit readiness for Medicare Advantage covers the shifting enforcement landscape in detail.
Organizations that have not yet measured what percentage of their risk-eligible diagnoses come from unlinked chart reviews are making workflow investments without knowing their actual exposure. That measurement is the starting point, and it should have happened before Q3. If it hasn’t, it is overdue but still possible early in the quarter, and skipping it creates a larger problem downstream. Proving ROI in Risk Adjustment: How to Measure Results Beyond Chart Review covers how to build that baseline measurement and what to do with the results.
Q3 priorities fall into three categories: measure your exposure, build encounter-based workflows, and implement QA processes that mirror RADV methodology. All three need to be moving simultaneously. Finishing one before starting the next is not a luxury most organizations have at this point in the calendar.
Step 1: Audit your diagnosis source mix. Determine what percentage of risk-eligible diagnoses comes from unlinked chart review records, broken out by provider group, line of business, and payer channel. This is the number that drives every other decision in the action plan.
Step 2: Model the CY2027 revenue impact. Apply the 1.53% risk score reduction proportionally to your unlinked CRR concentration and calculate the dollar impact. That figure should be informing budget conversations and vendor decisions now, not after Q3 closes.
Step 3: Identify where concentration is highest. Some provider groups or channels carry disproportionate unlinked CRR exposure. Those are the right starting points for workflow investment. A targeted rollout beats a uniform one when the calendar is this tight.
Step 4: Stand up pre-visit coder workflows. Technology surfaces HCC suspects before the appointment, a human coder reviews and curates, and the provider receives a clean, defensible list going into the encounter. When the provider addresses those suspects during the visit, the resulting documentation is encounter-linked by definition.
Step 5: Deploy point-of-care EMR alerts. Curated suspects from pre-visit review surface in the EHR during the encounter, where the provider can confirm, document, and code in real time. Alerts need to be actionable, not noisy. Providers who are drowning in inaccurate suspects stop paying attention, and that is harder to reverse than it looks. Closing HCC Coding Gaps at the Point of Care covers what makes alerts effective and what makes providers ignore them.
Step 6: Implement near-real-time post-visit review. Review within 24 to 48 hours of encounter close catches what fell through while the documentation is still current and the encounter is still linkable. The moment review starts happening weeks later, it starts to resemble the retrospective model CMS is constraining, and the audit risk profile changes with it.
Step 7: Train providers on encounter-linked documentation standards. MEAT criteria (Monitor, Evaluate, Assess, Treat), diagnostic specificity, and current-year relevance need to be embedded into provider expectations, not left as implicit standards that coders try to enforce after the fact. How to Operationalize Risk Adjustment Across Clinical Workflows covers building those expectations into the workflow rather than onto it.
Step 8: Build internal QA against RADV standards. Find unsupported diagnoses, encounter linkage gaps, and documentation deficiencies before CMS does. The standard RADV auditors apply is not a secret. Building internal review against that same standard is the most direct form of audit preparation available.
Step 9: Audit the full EDI and claims submission process. Some organizations are discovering that diagnosis codes submitted on claims are being dropped between provider submission and payer adjudication. Codes that were coded correctly are not making it through. A full process audit from suspect identification through final submission is worth running before 2027 volume starts accumulating.
Step 10: Implement second-level coding review. Every diagnosis entering the risk adjustment submission pipeline should pass through a second-level review checkpoint that validates encounter linkage and documentation support. If the internal team does not have visibility into where codes are dropping, a third-party process audit should be on the table. Preparing for RADV Audits: Building a Defensible Documentation Strategy covers the full audit preparation framework.
Before AEP, encounter-based workflows should be operational and tested, not still being piloted. The back half of the year is when new member volume increases, and every encounter with a new enrollee is a 2027 risk adjustment opportunity where unlinked chart reviews will not count. The encounter-based documentation needs to happen during the visit.
Thirty-day care coordination workflows for new members should be built before AEP begins. New enrollees switching into the plan need early clinical outreach, and the window to establish encounter-linked documentation for those members is narrow by nature. Plans should also have targeted risk and quality plans by member segment (high-RAF members, chronic condition cohorts, members with historical reliance on retrospective chart review) finalized before Q4 operational pressure makes it harder to be strategic.
The Stars calendar adds another layer. CMS removed 11 largely operational compliance measures from the 2028–2029 Stars framework, amplifying the weight of clinical and experience measures. Depression Screening enters Stars in 2029 using 2027 measurement data, which means the infrastructure for embedded screening workflows needs to be in place now. Tracking performance against both Stars and risk adjustment timelines simultaneously is critical. The V28 Mid-Year Checkpoint: What Your Risk Adjustment Team Should Be Measuring Right Now covers how to build that dual measurement cadence.
The encounter-based infrastructure built in Q3 2026 is not a one-year fix. CMS retained the 2024 risk adjustment model for CY2027 but was explicit that future model updates remain on the agenda. The audit expansion, from 60 to 550 plans per year, is a long-term enforcement posture, not a temporary push. Organizations building encounter-based infrastructure now are building the foundation for a regulatory environment that is moving in one direction.
The Stars Historical Reward compounds over time, meaning plans that sustain consistent Stars performance build up long-term advantages that plans catching up cannot replicate quickly. The same is true for risk adjustment: programs that shift from reactive to prospective earlier in the cycle carry structural advantages in RAF stability, audit readiness, and operational efficiency. Both Building a Defensible Risk Adjustment Program in a Post-V28 Environment and 3 Things to Know About Risk Adjustment in 2027 address what the long-term program posture looks like.
The three Q3 priorities (measure exposure, build encounter workflows, implement QA) map directly to Inferscience’s product suite.
HCC Assistant is the encounter-based HCC capture engine. It surfaces clinically supported suspects pre-visit, supports coder review, and delivers curated alerts at the point of care, closing the pre-visit-to-encounter pipeline in a way that produces encounter-linked documentation by design. It is the operational backbone of Steps 4 and 5 in this action plan.
AI Chart Assistant reduces provider prep time by synthesizing chart data into a pre-visit clinical overview. Providers who walk into encounters informed document more completely and miss fewer conditions, which is what the Q3 workflow build is designed to produce.
HCC Validator is the internal QA layer. It validates that diagnoses are encounter-supported before submission, flags unlinked or unsupported codes before they become audit findings, and builds the structured evidence trail that RADV preparation requires. It operationalizes Steps 8 through 10.
Quality Assistant helps to address care gaps at the encounter level, aligning risk adjustment and Stars workflows into the same clinical interaction. That alignment reduces provider burden while improving performance on both programs, which matters more as Stars clinical measures carry increasing mathematical weight.
Organizations that execute the full Q3 action plan should expect reduced exposure to the CY2027 risk score impact from unlinked chart review exclusion, encounter-based diagnoses that hold up under RADV review, lower post-visit review burden and chart chase volume, and stronger RAF accuracy heading into 2027. Provider workflows that serve both risk adjustment and Stars through the same clinical touchpoints reduce duplicative effort. And an internal QA posture that finds documentation gaps before CMS does is no longer optional in an environment where the agency is auditing all 550 eligible plans annually.
What should Medicare Advantage plans prioritize in Q3 2026? The three Q3 priorities are measuring unlinked chart review exposure and modeling the CY2027 revenue impact, building or strengthening encounter-based HCC workflows across pre-visit, point-of-care, and post-visit stages, and implementing internal QA processes that mirror RADV audit methodology. Q3 is the last practical window to have these workflows operational before CY2027 diagnoses begin counting.
Why is Q3 2026 the critical window for risk adjustment preparation? CY2027 diagnoses, the ones subject to the unlinked chart review exclusion, are generated from encounters happening in 2026. Encounter-based workflows need to be operational before the 2027 benefit year opens in January. Q4 is Annual Enrollment Period, when operational teams are fully stretched. Q3 is the window to build and test what needs to work in 2027.
What is the first step in a Q3 risk adjustment action plan? The first step is auditing your risk-eligible diagnosis source mix, specifically what percentage of your HCC capture currently comes from unlinked chart review records. Without that measurement, it is impossible to accurately model revenue exposure or prioritize the right workflow investments.
How does the expanded RADV audit program affect Q3 planning? CMS is scaling RADV audits from approximately 60 MA plans per year to all eligible plans, roughly 550, with record review counts increasing from 35 to up to 200 per plan. PY2020 audits began in early 2026, with new audit cycles initiating approximately every three months. Any plan without encounter-linked documentation and internal QA processes should treat RADV readiness as a Q3 priority.
What should organizations have ready before Annual Enrollment Period? Before AEP, encounter-based workflows should be operational and tested, 30-day care coordination workflows for new members should be built, and targeted risk and quality plans by member segment should be finalized. Implementing new workflows during AEP, when new member volume is highest, significantly increases execution risk.
Q3 2026 is not a planning quarter. It is an execution quarter. The CY2027 rule changes are finalized, the audit expansion is underway, and the encounter window for 2027 risk adjustment is already open. Organizations that use Q3 to measure their exposure, build encounter-based workflows, and stand up internal QA processes will enter 2027 with the infrastructure CMS is requiring. Those that wait will be managing the gap retrospectively. That is exactly the model CMS is moving away from, and the window to change course is closing.
Contact Inferscience to build your Q3 risk adjustment action plan. Request a walkthrough to see how encounter-based HCC workflows reduce retrospective burden, strengthen audit readiness, and position your organization before the 2027 window closes.
