Common Myths Associated with Multiple-Choice Physician Queries

This week I would like to address myths about multiple-choice physician queries that persist, despite guidance from the American Health Information Management Association (AHIMA) and the Association of Clinical Documentation Integrity Specialists (ACDIS) Guidelines for Achieving a Compliant Query Practice Brief (2022 Update).

My goal with this article is to address a few common myths associated with multiple-choice response options and “leading.” As an industry, we need to be careful not to overgeneralize the concept of “leading,” which has been defined in prior query practice briefs as steering a provider to a particular conclusion.

The brief also specifically states that when using multiple-choice query templates the author should “remove imbedded answer options that are not clinically credible or relevant.” (p. 7). Failure to remove clinically credible choices is a common error with templated queries. Please keep this concept in mind as you read the following myths.

Myth #1: A new diagnosis cannot be added through a multiple-choice query because it is introducing new information.

Incorrect. A new diagnosis can be added through a multiple-choice query. The purpose of a physician query is to help translate clinical data into terms that can be captured by coding. A CDI review typically involves reviewing the record for missing, vague, or incomplete documentation that cannot be captured within the ICD-10-CM/PCS code set. The Query Practice Brief states, “Diagnosis answer options that are not already documented in the health record must be supported by clinical indicators sourced from the medical record. These clinical indicators must be included within the query.” (p. 6). There are several references within the Query Practice Brief that discuss the importance of only including query choice options that are supported by the clinical evidence within the health record.

We must remember that providers are diagnosticians. It is their responsibility to translate clinical indicators, e.g., patient presentation, diagnostic results, response to treatment, etc., into an appropriate diagnosis that can be accurately reflected within ICD-10-CM/PCS. Multiple-choice queries arose from providers lacking sufficient knowledge of coding terminology. In fact, this is one of the main reasons most hospitals have CDI staff, to bridge the gap between “clinical speak” and “coding terminology.” Multiple-choice queries offer providers the correct terminology to reflect the clinical scenario, nothing more. The provider query is not a test of their coding knowledge; “Providing a new diagnosis as an option in a multiple-choice list – as supported and substantiated by referenced clinical indicator(s) from the health record – is not introducing new information.” (p. 12).

Myth #2: Providing only one response option, e.g., sepsis, acute respiratory failure, etc., makes a query non-compliant or leading.

From a clinical standpoint, there are no other diagnoses that denote these conditions. If the clinical evidence supports either of these diagnoses, it is the only reportable diagnosis option. This is why the Query Practice Brief states, “Multiple-choice query formats should include clinically significant and reasonable option(s) as supported by clinical indicator(s) in the health record, recognizing that occasionally there may be only one reasonable option.” (p. 12).

Some may argue that to be compliant, the query response choices must include the local infection as well as sepsis when the desired response is sepsis. There is no requirement that a condition that is already documented be included as a response choice. The provider does not have to confirm an already documented diagnosis in a query that is asking for a diagnosis not already documented in the health record.

Including documented signs and symptoms that support the more definitive diagnosis as query response choices is not compliant. For example, it is not necessary when asking if sepsis is an appropriate diagnosis based on the supportive clinical evidence to include the diagnosis of pneumonia without sepsis or any similar variation unless it is a clinical validation query. With a sepsis clinical validation query, the local infection (i.e., pneumonia in this example) is the clinically supported option. It would also not be compliant to add the diagnosis of systemic inflammatory response syndrome (SIRS) as an option on a sepsis query. A code for SIRS can only be assigned with the documentation supports a systemic inflammatory response to a non-infectious disease process, per the ICD-10-CM Official Guidelines for Coding and Reporting. If the patient has a documented infection, like pneumonia, SIRS would not be a clinically supported response option.

To address the concern about “leading” providers, the Query Practice Brief requires multiple-choice queries to contain the option of “other or similar terminology) to allow the provider to customize their response.” (p. 3). The industry standard is that as long as the provider is given the option of “other or similar terminology,” only including one diagnosis as a choice is not leading, as long as it is clinically supported by the health record and those indicators are included in the query.

Additionally, there is no requirement to include an option for the provider to respond that the query is unnecessary. Page 24 of the Query Practice Brief states, “The option of ‘other, please specify’ allows the provider to clarify their disagreement or impression of necessity related to the query intent. Organizations may also choose to include options such as ‘no further clarification is needed’ to track this occurrence.”

Myth #3: Choices within a multiple-choice query must include diagnoses not classified as a complication/comorbidity (CC) or major complication/comorbidity (MCC) to avoid “leading” the provider when the desired response is a diagnosis classified as a CC or MCC.

There is no such requirement. Although this specific scenario isn’t directly addressed in the Query Practice Brief, it does state that “there is no mandatory maximum or minimum number of diagnosis/procedure answer options necessary to constitute a compliant multiple-choice query.” (p. 3). Additionally, “multiple-choice answer options are not required to be in any particular order.” (p. 4).

Myth #4: If a diagnosis is only documented on a query, it cannot be reported.

If the query is included as part of the health record, the query response can be a documentation source for coding. “There is no specific direction as to where diagnoses must be documented or how often a diagnosis must be documented to allow it to be reported. Organizations may need to develop facility-based policies reflecting reportability of information that is clarified only within a query response versus elsewhere in the record.” (p. 22).

Connecting query responses to clinical evidence within the health record is a key concept for compliant query construction. Only those choices supported by clinical evidence should be included as a query choice.

The Query Practice Brief states, “all multiple-choice queries answer options should only include clinically relevant options (meaning those that are supported by clinical indicators within the health record) and exclude clinically irrelevant options.” (p. 3).

Programming note:

Listen to Cheryl Ericson every Tuesday when she reports the latest CDI news on Talk Ten Tuesday with Chuck Buck and Angela Comfort, 10 Eastern.