It is disheartening that there are still misconceptions about what a leading query is.
In my last article, “The Query Conundrum,” I briefly talked about the timing of an initial clinical documentation improvement (CDI) review, which can impact the timing of the first query.
Specifically, I asked “wouldn’t it make sense to be as proactive as possible when it comes to querying?” because there are many CDI professionals who wait to give the provider a “chance to review the latest findings” before sending a query. This can result in inefficiency, as the CDI specialist (CDIS) s to rereview the record to verify that the provider did not address the potential documentation issue the following day, then issue a query and then await a response. It can also artificially inflate review volumes, as I would argue that the initial review was not complete until the query was issued, so there was only one extended review (because the issue was identified during the first review of the record and not resolved until the following day, when the CDI checked the record again).
Are we engaging in activities for the sake of padding our numbers? Do our metrics really drive performance improvement? With so many CDI departments feeling understaffed, shouldn’t we be as efficient as possible with our query process? Is there really such a thing as querying too early when the clinical evidence exists to support the need for a query? Wouldn’t the record be more cohesive if a diagnosis was readily associated with the abnormal findings during the next physician note, because the CDIS proactively queried the provider to highlight these relevant clinical indicators? Would this approach improve patient care, as the medical record would be as complete and accurate as possible with each physician entry? Is it possible that providers could come to appreciate the queries that bring relevant clinical indicators to their attention as they are writing their note, rather than requiring them to amend a future note? Could we prevent future denials by having a diagnosis appear earlier, while the condition is present, and potentially more frequently in the record due to an “early” query? I see a lot of benefit to a proactive query process.
I understand not wanting to appear pushy, but what I found interesting is that some of the feedback included the concern that querying the provider before they have reviewed the latest piece of data is “leading.” I am a huge advocate of a compliant query process, and have been fortunate enough to be part of several query practice briefs, including the most recent, issued in 2019, and I appreciate this concern. However, it is disheartening that there are still misconceptions about what a leading query is. As part of the query practice brief teams, I know there was a concerted effort in 2016 to define “leading,” resulting in the following definition: “a leading query is one that is not supported by the clinical elements in the health record and/or directs a provider to a specific diagnosis or procedure.” However, it was a brief sentence that could have been easily overlooked, and it was not restated in the most recent query practice brief.
So, let’s see how we got to this place. Back in January 2001, the Centers for Medicare & Medicaid Services (CMS) told its Medicare Peer Review Organizations (PRO) program that they “are not to accept coding summary forms (e.g., physician query forms) as a substitute for documentation in the medical record for DRG validation purposes.”
The key word here is “substitute,” because CMS clarified, “we prohibited PROs from considering coding summary forms as a substitute for documentation in the medical record. We acknowledge that such forms may have limited value as a supplement to the medical record and serve as important communication tools within hospitals.” As a result, CMS adopted the position that “in conducting medical review for validating the DRG, the PRO reviewer shall use his or her professional judgment and discretion in considering the information contained on a physician query form along with the rest of the medical record. If the physician query form is leading in nature or if it introduces new information, the case shall be referred to a PRO physician for further review.”
This sentence is the first reference to “leading,” but it does not explicitly note that a query can’t be leading; it just requires that the type of query include a physician review. Additionally, the concept of “leading” was not defined by this issuance, but the summary did clarify that “we defer the promulgation of specific guidelines addressing these practices to health information management experts and organizations. Our position allows the use of the physician query form to the extent it provides clarification and is consistent with other medical record documentation. In addition, it addresses our concerns about leading questions and introducing information not otherwise contained in the medical record.”
In other words, the concern CMS had with leading was related to how the physician was asked to clarify the diagnosis, not necessarily when the provider was asked to clarify the diagnosis. It is important to note that when this statement was made, CDI wasn’t a recognized profession, and most if not all querying occurred during the coding process, when there was a record with complete documentation.
Health information management (HIM) guidance was available at this time through the American Health Information Management Association (AHIMA) query practice brief, “Developing a Physician Query Response Process,” which CMS referenced as a starting point for implementing the query process. This practice brief does refer to both a concurrent and post-discharge query process, so even though CDI professionals aren’t specifically addressed, they are most likely to query concurrently (and, consequently, to worry about the timing of a query). It is also important to note that there have been several query practice briefs since 2001, and each supersedes the next, but it is useful to revisit the language and intent of prior query practice briefs as we evolve into a compliant query practice. AHIMA recognized queries as a “necessary mechanism for improving the quality of coding and medical record documentation and capturing complete clinical data,” also calling it “an accepted tool for communicating with physicians on documentation issues influencing proper code assignment.”
“Query forms should be used in a judicious and appropriate manner,” AHIMA added. “They must be used as a communication tool to improve the accuracy of code assignment and the quality of physician documentation, not to inappropriately maximize reimbursement.”
In regard to what is a compliant query, the practice brief states that “an inappropriate query – such as a form that is poorly constructed or asks leading questions – or overuse of the query process can result in quality-of-care, legal, and ethical concerns.” The practice brief goes on to cite the query format, including what a query should and should not include.
Rather than discuss all these recommendations, I’m going to focus on those that could impact the concept of “leading.” The query should “present the scenario and state a question that asks the physician to make a clinical interpretation of a given diagnosis or condition based on treatment, evaluation, monitoring, and/or services provided,” the practice brief reads. “ Queries that appear to lead the physician to provide a particular response could lead to allegations of inappropriate upcoding.” The query practice brief also states that a query should not “lead the physician” or “sound presumptive, directing, prodding, probing, or as through the physician is being led to make an assumption.” It is interesting to note that a definition of leading was not reiterated in the “Guidelines for Achieving a Compliant Query Practice” (2019 Update), so the most recent definition remains “a leading query is one that is not supported by the clinical elements in the health record and/or directs a provider to a specific diagnosis or procedure.” In other words, the concept of leading does not include the timing of a query if it is supported by clinical elements in the health record and does not direct the provider to a specific conclusion.
Because the timing of the query isn’t necessarily addressed in terms of “leading,” it is important to also examine guidance in regard to “when to query.” The query practice brief states that “physicians should be queried whenever there is conflicting, ambiguous, or incomplete information in the medical record regarding any significant reportable condition or procedure.” This same practice brief also notes that “every discrepancy or issue not addressed in the physician documentation should not necessarily result in the physician being queried. Each facility needs to develop policies and procedures regarding the clinical conditions and documentation situations warranting a request for physician clarification. For example, insignificant or irrelevant findings may not warrant querying the physician.”
The 2016 query practice brief states that the “generation of a query should be considered when the health record documentation describes or is associated with clinical indicators without a definitive relationship to an underlying diagnosis,” as well as when the health record “includes clinical indicators, diagnostic evaluation, and/or treatment not related to a specific condition or procedure.” The 2019 query practice brief updates the reasons to query to include “to support documentation of medical diagnoses or conditions that are clinically evident and meet Uniform Hospital Discharge Data Set (UHDDS) requirements but without the corresponding diagnoses or conditions stated.”
The electronic health record allows the opportunity for efficiency that was otherwise unavailable when we lived in a paper medical record world. CDI professionals can see data as it becomes available, but that should not be a barrier to the query process. If the CDI professional is following industry guidance by including relevant clinical indicators that support a reportable missing diagnosis that is not insignificant or irrelevant, there isn’t a requirement to allow the provider the opportunity to proactively make the diagnosis first. I also want to be clear that waiting for the results of a diagnostic test is not the same situation, because in this case, there are missing relevant clinical indicators that can support a particular diagnosis. Basically, if the CDI professional has sufficient clinical evidence to support a reportable condition, there should not be a compliance issue based on the timing of the query. If they do not, then it would be inappropriate to query. I realize that opens the conversation to what is enough evidence to query the provider, so that leads me to reference the 2019 query practice brief section on clinical indicators:
“’Clinical indicators’ is a broad term encompassing documentation that supports a diagnosis as reportable and/or establishes the presence of a condition. Examples of clinical indicators include: provider observations (physical exam and assessment), diagnostic findings, treatments, etc. provided by providers and ancillary professionals. There is not a required number of clinical indicators that must accompany a query because what is a ‘relevant’ clinical indicator will vary by diagnosis, patient, and clinical scenario.
While organizations, payers, and other entities may establish guidelines for clinical indicators for a diagnosis, providers make the final determination as to what clinical indicators define a diagnosis. AHA’s Coding Clinic® similarly affirms that in its first quarter 2014 issue, stating ‘Clinical information previously published in Coding Clinic, whether for ICD-9-CM or ICD-10-CM/PCS, does not constitute clinical criteria for establishing a diagnosis, substitute for the provider’s clinical judgment, or eliminate the need for provider documentation regarding the clinical significance of a patient’s medical condition. It may still be useful to understand clinical clues regarding signs or symptoms that may be integral (or not) to a condition. However, care should be exercised as ICD-10-CM has new combination codes as well as instructional notes that may or may not be consistent with ICD-9-CM.’
The purpose or type of query will also impact how much clinical support is necessary to justify the query and, when applicable, reasonable option(s). When the purpose of the query is to add a diagnosis, clinical indicators should clearly support the condition, allowing the provider to identify the most appropriate medical condition or procedure. The quality of clinical indicators – how well they relate to the condition being clarified – is more important than the quantity of clinical indicators.
Clinical indicators can be identified from sources within the entirety of the patient’s health record, including emergency services, diagnostic findings, and provider impressions as well as relevant prior visits, if the documentation is clinically pertinent to the present encounter. For example, there is care being provided in the current encounter that necessitated the review of a previous encounter to identify the undocumented condition. Compliant query practice always requires the individualization of each query to reflect the specifics of the current circumstance.”
Again, I appreciate the desire to be compliant; we must be compliant. But we must also understand the query guidelines. You don’t need every possible clinical indicator to support a possible condition (unless specifically required by your organization). I always use this rule of thumb: would other providers come to the same conclusion based on the same information? If yes, then you likely have enough clinical indicators. If not, then maybe a query is not warranted. Again, unless you have organizational requirements defining what clinical indicators are needed for a particular diagnosis, your clinical indicators will vary by patient, diagnosis, and episode of care. If you haven’t taken the time to do so, I encourage you to carefully read the 2019 query practice brief and the associated FAQs. If you read it when was first published, I encourage you to revisit it. Let’s not use misconceptions as a reason to avoid change. That’s not to say that all providers should be queried as soon as those clinical indicators appear in the health record, but CDI professionals should use their knowledge of each provider to consider how likely he or she is to make an associated diagnosis once they review the updated information and act accordingly.
Isn’t it time we stop thinking of queries as a nuisance or a bother to providers, and instead help providers embrace them as a tool that helps drive quality patient care? You may also be surprised to find that some providers like a proactive query approach, because they are used to nursing and ancillary staff highlighting information that needs their attention so they can be more efficient.