The 2026 ACDIS/AHIMA query standard is final: the outpatient and HCC query rules it sets

If you write outpatient or HCC queries, the standard that governs them just changed. ACDIS and AHIMA published the 2026 update to their *Guidelines for Achieving a Compliant Query Practice* in August 2026. It supersedes the 2022 edition, it is the version your next external reviewer will cite, and it applies across inpatient, outpatient, ambulatory, and professional-fee settings, with considerations for outpatient and ambulatory contexts noted throughout.

An earlier draft went out for public comment in the spring, and we covered the piece of it that got the most attention: the rule that an AI-generated query has to clear the same bar as a human-written one. That provision carried into the final as Section XI. This piece is about the part of the standard that lands on an outpatient risk-adjustment desk, and about the fact that it is now the operative version, not a draft you can wait on.

What the move from draft to final settles

The draft told you not to treat it as in effect yet. The final closes that gap. It's published, it replaces the 2022 edition, and ACDIS and AHIMA state that "substantial compliance" with its core principles, rather than rigid technical adherence to any single element, is the standard an external reviewer should apply. The core principles are the familiar ones: a nonleading query, clinically relevant and sourced indicators, no reference to reimbursement or a quality measure, and room for the provider to exercise independent judgment.

There's a defensive line in here worth reading twice. The brief says it is not intended to be used as a stand-alone basis for denying a claim, for post-payment recovery, or for an adverse audit finding. That gives a coder something to point to when a payer tries to weaponize the standard against a query that was otherwise sound.

The outpatient and professional-fee query rules, in one place

Section III of the 2026 brief sets out how the query process bends in the outpatient, ambulatory, and professional-fee world instead of assuming an inpatient stay. Three modifications matter to anyone querying in a physician group.

The brief also states that where it references an inpatient-specific standard, that standard does not carry into outpatient or professional-fee contexts unless the brief says so. If your query templates were built off an inpatient CDI program, that's the sentence that tells you they need a second look.

An HCC query meets the same bar as an inpatient query

The brief names risk-adjustment work directly. Its definition of a query professional includes "Hierarchical Condition Category (HCC) coders operating in physician group or risk adjustment settings, and any vendor or technology platform acting in a query-generating capacity." And in its scope section it uses an outpatient HCC program as a worked example: an outpatient program running HCC prompts in the EHR, it says, must meet all requirements of the brief, and those prompts are "subject to the same nonleading and clinical indicator standards as inpatient queries."

So the specificity a compliant outpatient query chases is the same specificity that drives accurate risk capture, reached without steering. The brief's own multiple-choice examples read like an HCC coder's day: clarifying CKD stage 4 due to type 2 diabetes (E11.22 with N18.4), pinning down the type of heart failure behind an unspecified I50.9, or specifying chronic respiratory failure in a documented COPD patient.

Each one is a documentation gap resolved by a query that offers a clinically supported option and an alternative. What it can't be is a query whose title or answer set points at the code you want. Line the clinical indicators up in the evidence builder and check them against MEAT before the query goes out.

Problem-list mining stays out of bounds

Section X is the rule most likely to catch an outpatient desk that leans on the problem list. It says systematically searching prior records for diagnoses to import into the current encounter, without a documented clinical trigger in the current encounter, is inappropriate. The brief calls the practice "mining" and defines it by the absence of any current-encounter clinical indicator that prompted the review.

Prior-encounter information is still fair game, but only in support of the present visit. The brief lists the appropriate uses: establishing further specificity of a currently documented diagnosis (the type of heart failure, the stage of CKD), or establishing a patient baseline, such as a prior creatinine to support an acute kidney injury. This is the same discipline behind the rule that a problem list alone does not validate an HCC in a RADV review. A carried-forward diagnosis on the problem list gets you a starting point for a prospective query. It can't be the query's justification on its own.

What solo and small-practice coders have to do

The brief is explicitly scalable. It says smaller organizations, including solo and small-group physician practices, critical access hospitals, and rural health clinics, should apply its principles in proportion to their operational capacity, while its core compliance requirements apply universally regardless of size. That's a direct answer to the coder who reads a CDI standard and assumes it was written for a hospital team they will never have.

The scalable pieces are the governance ones: a defined process to create and approve query templates, a version-control and review schedule for those templates, and a periodic audit of query output. A one-coder shop runs a smaller version of each, but the brief expects all three to exist.

Where the technology-query rules landed

Section XI, the technology and AI-query section, sits in the final intact: a query generated by CAPD, computer-assisted coding, an LLM, or an EHR advisory is subject to the same compliance standards as a manually written one, and a human query professional stays accountable for it. If you use a query tool, that section, plus the draft coverage we published earlier, is the pre-flight checklist. The outpatient rules above are the part of the final that will get read hardest on a risk-adjustment desk.