Most clinical documentation integrity (CDI) departments are still judged on how much they do: query rates, response rates, agreement rates. In a November 6, 2026 piece for ICD10monitor, CDI consultant Cheryl Ericson, RN, MS, CCDS, CDIP, argues that this is the wrong yardstick. A department can hit every activity target and still watch claims come back denied months later, because none of those metrics ask whether a query actually held up once a payer looked at it. Her proposed fix — measuring query accuracy instead of query volume — is a useful lens for any coding or compliance team trying to explain why “clean” claims keep failing downstream.
A Measurement Problem Hiding in Plain Sight
Query accuracy, as Ericson frames it, asks three things of every query a CDI specialist sends: was it clinically supported, was it compliant with query-practice standards, and was it timely. The mirror metric is the query error rate, whose denominator is every reviewed record rather than every query sent — which means it also captures the queries that should have gone out but never did. Under that framing, a clinical validation query that lowers a case’s severity, and therefore its reimbursement, still counts as accurate, because the goal is a documentation record that survives review, not a higher case mix index.
What the Denials Data Actually Shows
The numbers behind Ericson’s argument come from Experian Health’s 2026 State of Claims survey, released October 28, 2026, covering 204 billing and finance professionals.
Denials are climbing, not holding steady
Forty percent of respondents now report denial rates of 10 percent or higher, up from 31 percent in 2022. Sixty-five percent say submitting a clean claim is harder than it was a year ago, and 64 percent say time to reimbursement is getting longer. Eighty-nine percent call reducing denials an organizational priority for 2026 — which is a lot of attention pointed at a problem that keeps growing anyway.
Code inaccuracy is a top-three cause, but not the only one
The survey ranks code inaccuracy as tied for the third most-cited reason claims get denied. That matters for coding teams, but it also means more than two-thirds of denial exposure sits outside coding accuracy alone — in documentation support, medical necessity, and payer policy interpretation. Only 30 percent of organizations recover 80 to 100 percent of denied revenue; 52 percent recover somewhere between half and 70 percent, according to the same survey.
Why a Clean Claim Can Still Fail Clinical Validation
Part of the disconnect, Ericson notes, is that a “clean” claim and a clinically defensible claim are not the same thing. Citing coding-operations vendor GeBBS Healthcare Solutions, she points out that a clean claim has simply passed front-end edits for format and code validity. Claim scrubbers check whether a code is structurally valid and correctly formatted; they do not check whether the underlying clinical documentation actually supports that code. A claim can sail through every automated edit and still arrive at a payer’s clinical validation review with an unsupported major comorbidity or complication — and come back denied, sometimes close to a year after the patient was discharged.
That gap is widened by a feedback problem the article also flags: more than half of clinical denials, according to Sift Healthcare’s 2025 Denials Insights Report, never even show up in the remittance file coding teams normally review. Takebacks on one large health system’s clinical denials averaged roughly $5,000 each and landed about 280 days after discharge — long enough that the original encounter, and the query decisions made on it, are a distant memory by the time anyone traces the denial back to its cause.
What a Query Accuracy Program Actually Tracks
Translated into practice, a query accuracy program shifts CDI reporting toward a small set of outcome questions rather than output counts:
- Was each query clinically supported by evidence already in the record, not just plausible?
- Did the query follow the 2026 ACDIS/AHIMA Guidelines for Achieving a Compliant Query Practice, since a non-compliant query cannot be counted as accurate regardless of outcome?
- Was the query timely enough to reach the provider before discharge, avoiding post-discharge queries that leave claims sitting in discharged-not-final-billed status?
- Did the resulting diagnosis or code survive payer review, including clinical validation and RADV-style audits?
- What is the query error rate — measured against every record reviewed, not just records queried — to catch the queries that should have been sent but weren’t?
Ericson also calls out a cost most CDI leaders don’t formally track: duplicate queries reissued because the first one was lost in workflow, and blanket policies that require every comorbidity and complication to be re-stated verbatim in the discharge summary, which add clerical time without adding clinical clarity.
Closing the Loop Between Denials and Documentation
The structural fix the article proposes is a feedback loop that most organizations only run informally: denial and recoupment data, including correspondence-only denials that never hit a remittance advice, routed back to CDI by encounter, diagnosis, and reviewer. The review itself should be non-punitive, asking whether the original denial was valid on the facts and whether an earlier, better-targeted query could have prevented it. That is a different exercise than the audit-and-blame cycle many CDI programs default to, and it is the only way query accuracy becomes a trend line instead of a one-time finding.
The financial backdrop makes the case for building that loop now rather than later. Premier has put the industry’s claims-adjudication administrative cost at $25.7 billion for 2023, and the American Hospital Association’s 2026 Cost of Caring report estimates hospitals spent close to $18 billion overturning denied claims and $43 billion pursuing payments already owed to them in 2025, with denials cutting net Medicare Advantage revenue by an estimated 7 percent. Vizient’s 2026 Impact of Change forecast adds a structural pressure on top of that: slowing inpatient volume growth alongside rising patient acuity, meaning more complex diagnoses for payers to scrutinize, not fewer.
Where Automated Coding Fits Into the Accuracy Gap
None of this is an argument against automation in the revenue cycle — it is an argument for pointing automation at the right layer of the problem. A claim scrubber that only checks code validity and format will keep missing the clinical-support gaps that drive validation denials, because that was never what it was built to check. Closing the query accuracy gap takes coding and CDI tools that reason about whether a code is actually supported by the documentation in front of them, not just whether it is structurally well-formed. That is the layer Medikode’s automated medical coding platform is built to operate on, which is a more durable way to bring denial rates down than adding more queries to the pile.