The Art of the Compliant Physician Query

Every coder eventually faces the gap: the clinical picture in the record clearly suggests something the documentation never quite says. The bridge across that gap is the physician query — and how you build it determines whether you’re practicing documentation integrity or leading the witness. The line between the two is well-defined, teachable, and worth mastering.

When a query is warranted

Industry practice briefs from AHIMA and ACDIS converge on the same triggers. Query when the record shows:

Conflict — the attending says one thing, the consultant another, and the codes depend on which
Ambiguity — “urosepsis,” “borderline findings,” and other terms that don’t map cleanly to a code
Incompleteness — a diagnosis treated but never stated (the classic: IV antibiotics, blood cultures, and fluids for a documented infection with hypotension, but no sepsis language anywhere)
Clinical indicators without a diagnosis — or, increasingly important, a diagnosis without clinical indicators, which merits a validation query rather than silent acceptance

Don’t query when the answer wouldn’t change the code, the DRG, a quality measure, or the accuracy of the record. Query volume that doesn’t move anything is provider-relationship capital spent on nothing.

The anatomy of a compliant query

A defensible query has consistent features:

It presents the relevant clinical indicators from this record — labs, medications, imaging, vitals — concisely and accurately, without cherry-picking only the evidence favoring one answer.
It asks an open question or offers multiple clinically reasonable options, always including escape routes such as “other,” “unable to determine,” or “clinically insignificant.” A multiple-choice query with one real option is a leading query wearing a costume.
It never mentions reimbursement impact. “This would change the DRG” has no place in a query — that single sentence can convert documentation improvement into evidence of improper inducement.
It’s directed to the right provider — the one whose clinical judgment the question requires — and it becomes part of the record or a retained business document per your organization’s policy, so the trail survives an audit.

Leading vs. legitimate: the feel test

A legitimate query asks the provider to exercise judgment; a leading query asks the provider to ratify yours. “Please clarify the type and acuity of heart failure, if any, supported by the clinical findings above” invites judgment. “Do you agree the patient has acute systolic heart failure?” supplies the answer and requests a signature. When in doubt, imagine the query read aloud in an audit: would it sound like a question or a suggestion?

Process habits that make queries work

Turnaround expectations. Queries answered weeks later, after discharge and coding deadlines, help no one; agreed response windows keep the process alive.
Track patterns, not just answers. The query log is a curriculum: the same gap recurring across one provider or one diagnosis is an education opportunity that ends the queries entirely.
Escalation paths. A non-response process, agreed with medical staff leadership, prevents queries from dying in inboxes.
Never pressure for a particular answer. The provider’s “no” is a legitimate outcome — and documentation that your process respects it is itself a compliance asset.

Queries done well are invisible: the record ends up saying what was clinically true, codes follow documentation, and no one can tell where the coder’s careful question made the difference. That invisibility is the craft.

Follow your organization’s query policy and current AHIMA/ACDIS practice guidance; this article is educational, not policy.

——————————————————

Originally Published On: Medical Coding News

Photo courtesy of: Getty Images

Follow Medical Coding Pro on Twitter: www.Twitter.com/CodingPro1