News: Latest CMS compliance newsletter could contain conflicting advice

Under advice regarding Recovery Audit Contractors (RAC) findings related to Acute Respiratory Failure (ARF), on p. 2, the auditor finding states that:

 

“It was determined that the clinical evidence in the medical record did not support respiratory failure, despite physician documentation of the condition.”

 

Typically the physician gets the last say, according to Glenn Krauss, BBA, RHIA, CCS, CCS-P, CPUR, C-CDIS, CCDS, an ACDIS Advisory Board member and independent consultant based in Madison, WI. “Neither CDI specialists nor coders get to second guess the physician, and CDI staff would rarely (if ever) go back to the physician to query for ARF if the physician documented it several times,” he says.

 

However, fellow ACDIS Advisory Board member Robert S. Gold, MD, CEO of DCBA Inc., in Atlanta, says that some CDI teams are trained to ask that physicians document ARF “in virtually every patient's chart” if the chart included acute exacerbation of chronic obstructive pulmonary disease (COPD) or pneumonia and either a low partial oxygen pressure (pO2) or high carbon dioxide partial pressure (pCO2).

 

Where signs of over documentation for ARF exist, “it behooves the hospital… to train their staff in matters of ethical documentation based on nationally recognized definitions by the medical authorities. If the patient doesn't have it [ARF], it shouldn't be coded as though it does exist,” Gold says.

 

The newsletter’s guidance regarding gastroenteritis with hemorrhage with CC/MCC, found on p. 4, raises additional concerns, Krauss says. The document states that:

 

“All medical documentation entries must be consistent with other parts of the medical record (assessments, treatment plans, physician orders, nursing notes, medication and treatment records, etc….”  

 

The worry, Krauss says, is that CDI specialists/coders generally do not review nursing notes when the physician clearly documents a condition. “It is generally only in instances where the physician doesn’t document a particular diagnosis, or where documentation doesn’t represent codeable findings, that the CDI specialist would interrogate the medical record further looking for clinical indicators that might support a query,” he says. “In my mind, this represents conflicting guidance.”

 

“CDI folks usually use appropriate discretion in looking at nurses' notes and should continue to do so,” says Gold. “It's not necessary to look at all of them all the time but CDI staff should know what items to look for, and where in the medical record, including nurses’ notes, they can find clinical evidence.”

 

Krauss also took issue with the newsletter’s next bullet point (also on p. 4) which states:

 

“The hospital’s claim must match both the attending physician’s description/diagnosis and the information contained in the beneficiary’s medical record.”

 

However, Krauss points out that this is in conflict with CMS Special Edition MLN Matters article SE 1121) which states the following:

 

“Remember that the “Coding Clinic, First Quarter 2004” states, if there is conflicting physician documentation, and the coder fails to query the attending physician to resolve the conflict, hospitals are encouraged to code the attending physician’s version. However, the failure of the attending physician to mention a consultant’s diagnosis is not a conflict. So, if the consultant documents a diagnosis and the attending physician doesn’t mention it at all, it is acceptable to code it. A conflict occurs when 2 physicians call the same condition 2 different things – for example, the attending physician documents a sprained ankle and the orthopedist refers to the same injury as a fracture.”

 

“If one provider gives inadequate specificity and the specialist gives appropriate added specificity, then there is no reason to go with the specialist's opinion,” says Gold. “After all, that's why he was asked to see the patient. It's only when there is a disagreement between the two [that additional information should be sought]. A disagreement is different than a disparity.”

 

Additional concerns were raised regarding intracranial hemorrhage or cerebral infarction on p. 7 where auditors noted that “coding for CC/MCC of hemiparesis is often miscoded in both directions” (over or undercoded) and that “many records lack concise documentation to support either the diagnosis” of TIA or CVA.

 

Krauss noted that in situations where the physician is uncertain of a diagnosis and treats both conditions equally the coders could code either condition.  

 

Finally, guidance offered regarding other vascular procedures with MCC in the left column of p. 11 asks providers to review Coding Clinic 1^st Quarter 1993, p. 19. However, “this Coding Clinic has been superseded with the introduction of the new heart failure codes that introduce further specificity to the equation,” Krauss states. “The instructional notes for 402.XX that state ‘Use additional code to specify type of heart failure (428.0-428.43, if known)’.”

 

Editor's Note: At the time of publication, Gold had reached out to the editors of MLN Matters to discuss the concerns raised. ACDIS will continue to monitor the situation and provide any updates in a future issue of CDI Strategies.