Guidance and tact key to compliant, effective physician queries

When it comes to querying physicians, CDI specialists need to first recognize when it’s appropriate to query. The next step? Using tact in the wording of the query. This will help ensure compliance and elicit appropriate clarification from the physician. Although doing so does not have to be a painstaking task, it can sometimes be tricky.

“When you query a provider, it is when there is conflicting, incomplete, or ambiguous information in the health record,” said Cheryl Ericson, MS, RN, manager of clinical documentation integrity and utilization review at Medical University of South Carolina in Charleston, and ACDIS Advisory Board Member.

In the 2008 American Health Information Management Association (AHIMA) practice brief “Managing an Effective Query Process,” guidance states that the following situations may warrant physician queries:

• Clinical indicators of a diagnosis but no documentation of the condition
• Clinical evidence for a higher degree of specificity or severity
• A cause-and-effect relationship between two conditions or organism
• An underlying cause when admitted with symptoms
• Only the treatment is documented (without a diagnosis documented)
• Lack of present on admission (POA) indicator status

Note that AHIMA is not saying coders and CDI professionals may query only in these situations, but rather these are scenarios for which it is acceptable to query, said Ericson, who spoke during ACDIS’s March 18 audio conference “Physician Queries: Apply Industry Guidance to Improve Procedures and Data Tracking.”

“I don’t think that they are limiting when queries can be made,” she said. “I look at it more as a permissive idea rather than a restrictive idea.”

“I don’t think that they are limiting when queries can be made,” she said. “I look at it more as a permissive idea rather than a restrictive idea.”

Also, queries should not question a provider’s clinical judgment. Rather, Ericson said, queries should to clarify physician documentation when it does not meet the following criteria:

• Legible
• Complete
• Clear
• Consistent
• Precise

Query to clarify use of the term ‘history’

The fact that the terminology physicians use often differs from that which ICD-9-CM requires for compliant coding is a major hurdle that often triggers queries. For example, there are cases for which you may need to query the physician to clarify the use of the term “history.” Consider the following.

A physician admits a patient with pyelonephritis. The past history states right lower lobe lung carcinoma proximal lesion. The CDI specialist submits the following query:

Dear Dr. X,
A patient was admitted with pyelonephritis. The PHx states, “RLL lung CA – proximal coin lesion.” If possible, please clarify if this has resolved (ex: after complete resection), or if the patient still under treatment and/or still carries this diagnosis. Thank you.

This type of query often frustrates both physicians and coders. In clinical speak, past history includes conditions that are current, said Andrew Rothschild, MD, MS, MPH, FAAP, CCDS, director at FTI Consulting in Atlanta.

“A patient can have a history of AIDS. It doesn’t mean it went away; it just means that in the history of the patient’s life, he or she contracted the illness,” explained Rothschild, who also spoke during the March 18 audio conference. “Whereas in coding, history means that it happened in the past—these are totally different concepts.”

For a case in which the physician has already provided the diagnosis and the coder/CDI specialist is just trying to determine whether the diagnosis is relevant at the present time, Rothschild believes that it can be appropriate to ask this type of “A or B” query. He cautioned, though, that the only official AHIMA-accepted use of this format is when coders query for POA status. A POA query specifically determines whether a condition is hospital-acquired. This is often confused with querying for principal diagnosis, a completely separate concept, Rothschild noted.

“A lot of coders are using the same wordings to query both things,” Rothschild said. “Querying to find out if something was present at the time the patient was admitted is very important and that satisfies [POA] requirements, but it does not necessarily satisfy the definition of a principal diagnosis. A urinary tract infection, for example, may have been [POA], but it may not have been significant enough to have clearly warranted admission. This may require a separate query, if not already sufficiently documented.”

Examine potential query abuses

Take a look at potential query abuses, particularly when those queries influence the billing of a chart and trigger excessive charges.

A common example is when coders refer to a chart in which a radiologist who read the x-ray but did not actually see the patient notes aspiration pneumonia in his or her assessment. Routinely coding based on the radiologist’s documentation alone is a type of Medicare fraud, Rothschild said.

“You need confirmation from a ‘treating physician,’ one that is clinically caring for the patient,” he explained. Auditors have identified that coders many times up-code pneumonia, so it is particularly important to look at applicable claims and queries to ensure compliance.

For example, a physician admits a patient with pneumonia. The documentation (i.e., history of present illness) indicates:

Emesis with subsequent cough. The chest x-ray noted RML and RLL. Different lobes of the lung infiltrate consistent with aspiration, and the patient was treated with Clindamycin.

Consider the wording of the following query: “If you feel this patient has aspiration pneumonia, please clarify on the progress notes.”

The query is clearly noncompliant because the question implies a specific answer instead of leaving it open-ended, said Rothschild. He explained that many times this type of leading query results in the up-coding of pneumonia. It’s important to be able to ask this question in a more compliant format.

When crafting queries, use clinical indicators from the patient’s health record to support that this inquiry is individual to that patient, Ericson said.

“Queries that appear to lead the provider to document a particular response could result in allegations of inappropriate up-coding,” Ericson said. “The real question is determining if the hospital is entitled to higher reimbursement based on rendered services or would additional documentation artificially increase reimbursement, which could be considered fraudulent.”

Use tact when querying physicians

Take care in the way you approach queries. The American Hospital Association’s Coding Clinic, second quarter, 1998 states, “The physician should be queried when documentation is ‘suggestive’ of a condition, but not clearly documented by the attending physician.”

But Rothschild cautions against using this exact language in a query to the physician, as many physicians may find it irritating or insulting.

“To them, it is very clearly documented [clinically speaking]. It’s necessary to finesse that difference between what you know (that if not well documented in the eyes of the government, it is not codable in the eyes of the coding manager)—and the fact that the physician thinks the note is clinically well documented,” Rothschild said. “I often say, ‘You have documented this very clearly clinically, but we have to be able to translate that into codes in order to capture the appropriate diagnosis and severity’.”

Some physicians believe many queries are simply a matter of semantics, however, Rothschild explained that “this is inescapable because ICD-9 uses terms that physicians don’t have a habit of [using in documentation.]”

It is appropriate and necessary to query a physician when:

• The intended meaning of a diagnosis is uncertain due to inconsistent uses of similar terms (e.g., septicemia, bacteremia, and sepsis)
• The complex nature of a diagnosis may obscure accurate coding (e.g., unclear association between sepsis and an acute organ failure)
• It’s unclear if more than one principal diagnosis meets criteria
• It’s unclear if a condition is a complication
• It’s unclear if a remote chronic condition is newly diagnosed, resolved, or healing (e.g., history of lung cancer)

Even though physicians may perceive that they documented something clearly, sometimes different physicians use terminology, such as “bacteremia,” differently.

“If doctors are using different definitions, part of the answer is often to involve a physician champion or to have a departmental decision that aids in defining a term. This also opens doors to increase our communication, which really is the primary goal,” Rothschild said.

Editor’s note: This article first appeared online at www.justcoding.com.