18.3 Leading Queries as a Compliance Failure
Key Takeaways
- The August 2026 ACDIS/AHIMA Guidelines for Achieving a Compliant Query Practice treat a query as leading when it directs or influences the provider toward a preferred diagnosis or result.
- Never put DRG, CC/MCC, reimbursement, relative weight, or quality-outcome language in a query, and never query to a DRG.
- Bolding, highlighting, underlining, or arrowing one answer option is leading; a yes/no query may not introduce a new diagnosis.
- A leading query that produces a higher-weighted MS-DRG can support False Claims Act exposure under 31 U.S.C. 3729; knowledge includes reckless disregard, and specific intent to defraud is not required.
- For penalties assessed after July 3, 2025 on violations after November 2, 2015, DOJ's 2025 inflation adjustment sets False Claims Act civil penalties at $14,308 to $28,619 per claim, plus treble damages.
18.3 Leading Queries as a Compliance Failure
Quick Answer: A query is leading when it directs or influences the provider toward a specific diagnosis or result. The August 2026 ACDIS/AHIMA Guidelines for Achieving a Compliant Query Practice require a nonleading question, sourced clinical indicators, no reimbursement or quality-outcome language, and independent provider judgment. Never query to a DRG. A steered inpatient diagnosis that raises an MS-DRG can become a False Claims Act problem, not just a coaching issue.
This independent OpenExamPrep section teaches the Domain VII handbook bullet: demonstrate what constitutes a leading query. Chapter 10 of this guide covers compliant-versus-leading technique in the documentation domain. This section is the ethics and compliance view of the same failure: why a leading query is not a harmless shortcut, and how it connects to federal false-claim risk. The 2026 brief supersedes the 2022 version. Do not study 2022 examples as if they were still the current joint standard.
OpenExamPrep does not claim partnership or official approval by ACDIS, AHIMA, or the U.S. Department of Justice (DOJ). Facts below about civil penalties come from DOJ's published 2025 inflation adjustment, not from guesswork.
What substantial compliance requires
The 2026 brief describes substantial compliance as a query that:
- Is nonleading.
- Includes clinically relevant, sourced indicators from the health record for this encounter (with limited, policy-directed use of prior encounters when they actually support the question).
- Does not reference reimbursement, quality outcomes, or other impact.
- Gives the provider a real chance to exercise independent clinical judgment.
A query can be noncompliant without being leading — for example, when indicators are generic, copied from a template that does not match this patient, or when the specialist inserts a personal interpretation as if it were already documented. Domain VII still starts with leading, because leading is the version of noncompliance that most clearly manufactures a preferred code.
A nonleading statement seeks clarification without pointing at the desired answer. Providing multiple-choice options that are actually supported by this patient's indicators is not, by itself, introducing new information. Listing only the MCC-level choice, or dressing that choice in bold type, is another story.
What a leading query looks like on an inpatient unit
Lead with the failure modes CCDS items actually test.
Querying to a DRG. Any language that names an MS-DRG, a relative weight, a CC/MCC, reimbursement, or a desired grouper result is leading. Classic failures include please document X so this groups to MS-DRG 291; this case is a 293 without a CC — is there a CC?; and if you add acute on chronic systolic heart failure we pick up an MCC. The provider is no longer being asked what is wrong with the patient. The provider is being asked to complete a billing worksheet. Never query to a DRG.
Quality-outcome language. Please document so the mortality risk adjustment improves, this will trigger a PSI unless you clarify, and we need this for HAC exemption are the quality twins of DRG language. The 2026 brief groups reimbursement and quality-outcome references together as content that must stay out of the query. Those topics belong in education sessions and program metrics, not in the question the attending answers in the chart.
Emphasis that picks a winner. Answer options must not use arrows, bolding, highlighting, or underlining to mark a preferred response. It is also a poor practice to highlight indicators inside the query in a way that could be read as steering, and highlighting should never appear inside the answer options. If the source record itself used bold lab values, organizational policy may allow preserving that source formatting; the specialist still must not add new emphasis to manufacture a favorite diagnosis.
Yes/no that introduces a new diagnosis. The 2026 brief allows yes/no in tight situations: POA (with an unable-to-determine option), substantiating a diagnosis already documented (for example, a pathology confirmation), or cause-and-effect between conditions that are already documented. Yes/no may not introduce a new diagnosis. Asking Does the patient have severe protein-calorie malnutrition? when no provider has named malnutrition is a leading yes/no, even if the albumin is low. Use multiple-choice with clinically relevant options and a required open-ended other, please specify (or similar) path.
Clinically irrelevant options and missing other. Options must match the indicators. Offering hypernatremia when the sodium is 122 mEq/L is not balanced; it is noise that can still steer by contrast. Multiple-choice must include at least one clinically valid option and an open-ended other path. Omitting other traps the provider inside the specialist's list.
Repeating or shopping the question. Sending the same query until the preferred MCC appears, or walking it to a different provider who did not care for the patient in order to get a yes, is not persistence. It is pressure. The Official Guidelines still require a provider who is legally accountable for the diagnosis. The 2026 brief still requires independent judgment.
| Pattern | Why it is leading or otherwise noncompliant | Repair |
|---|---|---|
| Please document acute systolic HF so we get MS-DRG 291 | Queries to a DRG and names payment | Ask for type and acuity of heart failure using sourced indicators; no DRG text |
| Bold severe malnutrition among four choices | Emphasis selects the preferred option | Same weight for every option; include other, please specify |
| Yes/no: Does the patient have acute hypoxic respiratory failure? as a new diagnosis | Yes/no introducing a new condition | Multiple-choice with relevant respiratory diagnoses plus other |
| This will hurt our PSI rate unless you clarify POA as Y | Quality-outcome language | Ask POA with Y/N/U/W-style options and unable to determine; no score talk |
| Indicator list that already states the diagnosis the specialist wants | Specialist interpretation presented as fact | Quote sourced findings; let the provider name the condition |
False Claims Act: conceptual risk with verified penalty math
The False Claims Act (FCA), 31 U.S.C. 3729, imposes civil liability on a person who knowingly presents, or causes to be presented, a false or fraudulent claim for payment to the United States, or who knowingly makes a false statement material to a false claim. Knowingly includes actual knowledge, deliberate ignorance, and reckless disregard of truth or falsity. The statute says specific intent to defraud is not required. Treble damages (three times the government's damages) apply, plus a per-claim civil penalty.
DOJ adjusts that per-claim penalty for inflation. In the Civil Monetary Penalties Inflation Adjustments for 2025 final rule (effective July 3, 2025, 90 FR 29445), for penalties assessed after July 3, 2025 on violations occurring after November 2, 2015, the FCA range under 31 U.S.C. 3729(a) is minimum $14,308 and maximum $28,619 per claim (table 1 to 28 CFR 85.5). Earlier assessment windows use earlier inflation tables. Those figures will move again in a later annual adjustment; recheck the current 28 CFR 85.5 table before quoting them in policy. This chapter uses the July 3, 2025 DOJ table because it is the current published adjustment as of this writing.
How a leading query gets into that statute is conceptual, not theatrical. Medicare IPPS pays a hospital claim that includes diagnoses produced by the record. If a query named the MS-DRG, bolded the MCC, or refused to take no for an answer until the higher-weighted code appeared, the resulting claim can be false as to the patient's condition. Reckless disregard is enough. A specialist does not need a memo that says we are going to cheat. A pattern of DRG-directed templates can be enough for the government to argue that the organization ignored the obvious risk. Individual knowledge still matters; so does organizational culture. Domain VII wants you to see the chain before Chapter 19 walks DRG creep, the OIG Work Plan, and PEPPER in more depth.
The FCA is not limited to extra payment. A claim can be false because a quality-driven diagnosis was manufactured, or because POA was steered to avoid a HAC payment provision. The common ingredient is still the same: the documentation no longer reflects independent clinical judgment supported by the record.
Technology does not launder a leading prompt
The 2026 brief states that technology-generated queries follow the same standard as human ones. An EHR suggestion that says add MCC: acute kidney injury to capture DRG xxx is a leading query with a software vendor's letterhead. Auto-bolding the highest-weighted option is still emphasis. A generative tool that drafts please confirm sepsis so mortality risk improves has inserted quality-outcome language. The specialist who clicks send owns the query that left the system.
Candidate scenario: the working-DRG sticky note
A concurrent reviewer notes working MS-DRG 293 (heart failure and shock without CC/MCC). Indicators include ejection fraction 25%, intravenous loop diuretic, and BNP 1,840 pg/mL. The attending documented unspecified heart failure. A compliant multiple-choice query cites those sourced indicators, asks the attending to clarify type and acuity if known, lists only clinically relevant options, and includes other, please specify and unable to determine. A leading query says we need an MCC to leave 293 — please document acute on chronic systolic heart failure. The first query may still result in a higher-weighted MS-DRG if the attending independently documents it. The second query is a compliance event whether or not the attending agrees. Payment that follows the second query is the FCA fact pattern Domain VII wants you to recognize.
Exam traps for 18.3
- A higher-weighted MS-DRG after a compliant query is not fraud. Accurate capture following independent documentation is the system working. The failure is the method that produced the words, not the fact that CMI moved.
- Open-ended is not automatically safe. Please document the MCC you are treating is leading in an open-ended wrapper. The statement still has to avoid a preferred result.
- Verbal queries follow the same rule. Saying in the hallway we need systolic failure for the DRG is a leading verbal query. Record date, time, people, sourced indicators, the nonleading statement, options if any, and the verbal response, and get the response into the permanent health record before coding.
- Do not invent penalty math. If you cite a per-claim dollar figure, use the current DOJ inflation table and the assessment/violation dates that table applies to. Do not recycle a $5,000–$10,000 statutory range as if it were the current assessed amount, and do not invent a million-dollar automatic fine.
- Chapter 19 territory. DRG creep, the OIG Work Plan, and PEPPER outlier flags are the next Domain VII chapter. This section only needs the leading-query mechanism and the FCA knowledge standard.
Official pages to recheck:
Which inpatient query is leading because it queries to a DRG?
A multiple-choice query lists four clinically relevant options but bolds and underlines only severe protein-calorie malnutrition. That formatting is:
A yes/no query asks, Does the patient have acute hypoxic respiratory failure? when no provider has documented that diagnosis and the specialist wants it added. Under the 2026 query standard this format is:
How can a leading query create False Claims Act risk on an inpatient Medicare claim?