8.2 CDI Workflows and Query Governance
Key Takeaways
- CDI workflows should support documentation that is accurate, complete, and compliant for care, coding, quality, and reporting.
- Query governance follows the ACDIS/AHIMA compliant query brief: non-leading format, patient-specific clinical indicators citing record location, tracking, response monitoring, and escalation rules.
- CDI metrics (query rate, response rate, agreement rate, impact rate, denial trend) must be read together because any single metric can mislead.
- RHIA leaders connect CDI findings to provider education, coding reconciliation, EHR templates, and quality reporting needs.
Governing CDI Workflows
Clinical documentation integrity (CDI) — historically "clinical documentation improvement" — is a structured process for improving the clarity and completeness of provider documentation. On the RHIA exam, Domain 3 covers CDI-related audits and workflow optimization, while Domain 4 (Revenue Cycle Management) addresses CDI's reimbursement role. That overlap matters: CDI is not only about payment. It drives quality measures, severity-of-illness and risk-of-mortality reporting, patient-safety analysis, coding accuracy, and the usefulness of the legal health record.
A typical CDI workflow runs: case selection → record review → query-opportunity identification → compliant query → provider response → documentation update → coding reconciliation → performance monitoring. The RHIA must know where each step fails: cases selected too late to influence the chart, queries lacking clinical support, non-responding providers, unresolved coder–CDI disagreement, or templates that encourage copy-forward and obscure indicators.
Query Governance and the Compliant Query Brief
Query practice is governed by the ACDIS/AHIMA Guidelines for Achieving a Compliant Query Practice, refreshed in the 2026 update. Core rules an RHIA must apply:
- A query must be non-leading — it may not direct the provider to a desired diagnosis. Educating within a query (verbal or written) is leading and noncompliant; education belongs in a separate setting.
- Every query must carry patient- and episode-specific clinical indicators, each citing the location found in the record (e.g., "lactate 4.2 mmol/L, 0600 lab").
- Multiple-choice queries must offer all clinically plausible options, including "clinically undetermined" and an option to add a free-text response — never only the highest-paying choice.
- The brief applies regardless of how the query is generated, including by artificial-intelligence or computer-assisted tools.
Governance also defines who may query, when queries are appropriate, how responses are documented in the legal record, how unanswered queries escalate, and how templates are maintained. Metrics must never pressure staff to generate unnecessary queries.
| CDI metric | What it can show | What to verify |
|---|---|---|
| Query rate | Review intensity or documentation opportunity | Case mix and selection method |
| Response rate | Provider engagement with queries | Timeliness and escalation process |
| Agreement rate | Provider acceptance of clarification | Query quality and clinical support |
| Impact rate | Effect on coded data, severity, quality | Whether the impact is appropriate and compliant |
| Denial trend | Downstream documentation vulnerability | Payer pattern, coding validation, appeal outcome |
Reading Metrics Together
No metric stands alone. A high query volume with low agreement signals weak query quality or poor case selection. A high response rate with little documentation impact suggests queries are answered but clinically meaningless. A denial drop after education may be real — or it may reflect a payer-mix shift; validate before claiming credit.
Worked example. A facility brags about a 95% query response rate, yet its case-mix index is flat and severity capture is unchanged. Reading the metrics together, the RHIA finds providers click "agree" on leading multiple-choice queries that offer only one realistic option. The fix is not more queries — it is rebuilding templates to be non-leading and re-educating CDI staff on the compliant query brief, then re-measuring agreement against impact.
Common trap. Exam answers that reward raw query volume, ignore provider burden, or accept leading queries to capture a higher-weighted DRG are wrong. The credentialed answer uses CDI analytics to find root cause, fix workflow and templates, and protect documentation integrity across revenue, quality, and safety uses.
Query Formats and the Coding–CDI Reconciliation Loop
The compliant query brief recognizes several formats, and the RHIA must know when each is appropriate. An open-ended query asks the provider to document the clinical significance of a finding in their own words and is the least leading. A multiple-choice query lists clinically supported options plus "clinically undetermined" and "other" — it is compliant only when every plausible option appears, not just the lucrative one. A yes/no query is permitted in narrow situations, such as confirming POA status or linking an already-documented condition to a manifestation; it is not used to introduce a new diagnosis.
Verbal queries are allowed but must be documented with the same indicators and non-leading standard as written ones.
Reconciliation between coding and CDI is a governance requirement, not an optional courtesy. When the CDI working DRG and the coded final DRG differ, a defined process must resolve the discrepancy before the bill drops — typically a second-level review by a coding or physician advisor. Tracking the mismatch rate and the reasons (missed query, late documentation, guideline interpretation) tells leadership whether the gap is a knowledge problem, a workflow problem, or a documentation problem.
Finally, the RHIA must guard against query fatigue and metric gaming. If staff are rewarded purely on query count or financial impact, they will generate marginal queries that burden providers and erode trust, and they may drift toward leading language. Balanced scorecards, periodic compliance review of a query sample, and education tied to specific documentation patterns keep the program both productive and defensible — the posture the exam consistently rewards.
Where the Query Lives and Cross-Functional Governance
A recurring governance decision is whether the query and its response become part of the permanent legal health record or live in a separate business record. Both approaches exist; the organization must pick one in policy and apply it consistently, because auditors and payers will look for the documentation that supports the final code. If the query response is the only place a diagnosis is clarified, that response generally must be retained and discoverable. The RHIA ensures the policy aligns with the medical-staff rules and the legal-record definition.
CDI improvement is inherently cross-functional. Providers need pattern-specific education rather than generic reminders; coding needs a documented reconciliation path; quality teams need agreement on which data elements feed each measure; EHR builders may need to remove copy-forward shortcuts and templates that hide indicators; and compliance should review query practice whenever risk signals appear. The RHIA's role is to convene these functions, translate CDI analytics into root-cause findings, validate that interventions worked, and protect documentation integrity across revenue, quality, severity, and safety uses simultaneously.
Treating CDI as a single-department payment activity is the framing the exam marks wrong.
A CDI team reports high query volume but low provider agreement. What should the RHIA evaluate?
Under the ACDIS/AHIMA compliant query brief, which query practice is correct?
Why should CDI metrics be reviewed together rather than one at a time?