Clarification: Anemia documentation and coding presents communication hazards

Editor’s note: The following Q&A first appeared in the August 20, 2009, edition of CDI Strategies and has been expanded on and clarified in this week’s issue.

Q: Is it true that if a physician documents a precipitous drop in hemoglobin/hematocrit (hgb/hct), then the physician can’t document “anemia” anywhere in the chart?

A: ICD-9-CM code 790.0 (Abnormality of red blood cells) contains an excludes note for anemia (other specified types 280.0-285.9), says Adrienne Gmeiner, RN, CCS, clinical documentation specialist with Lawrence (MA) General Hospital. So, if the physician documented an “anemia” that fell into that range of codes in the record, the coder would not also report symptom code 790.01. This code is intended for a patient who comes in with a precipitous drop in their hct without a determined causal diagnosis. 

Q: It’s too bad there is an excludes note regarding anemia. Sometimes a patient comes in with a very low hgb/hct that keeps trending downward. For example, the patient has a heme negative guaiac, and the physician writes ‘anemia.’ Many physicians feel uncomfortable with ‘acute blood loss anemia’ terminology as it might negatively impact his or her quality reporting if it’s not an obvious bleed (e.g. GI). 

 

At my hospital we deal with a large number of elderly with many co-morbidities. A patient might come in with a fractured hip and while he or she might spend a couple of days waiting for clearance for surgery, his or her hgb/hct trends down and then drops further after the surgery. When I leave a request for acute blood loss anemia, a number of physicians are reluctant to document this terminology.

 

I’d love to hear from any experts in the field the difference between the two, and tips on requesting documentation in the example above. 

A: The term “acute blood loss anemia” is not a specific term in coding or in physician language, says Robert S. Gold, MD, CEO/founder of DCBA Inc., in Atlanta.

 

Gmeiner points out that a coder would, upon seeing documentation of “acute blood loss anemia,” go to the ICD-9 index where he or she would look up anemia, blood loss, acute and be directed to the code 285.1. Although the code is titled “acute post hemorrhagic anemia,” the descriptor states “anemia due to acute blood loss.”

 

For true specificity, Gold agrees, the code is listed “acute posthemorrhagic anemia” for the coders, and physicians should document “anemia due to acute blood loss” or “anemia due to acute GI bleed.”  

 

Regarding the perceived negative impact on physician scorecards, William E. Haik, MD, director, DRG Review, Inc., Fort Walton Beach, FL, tries to reassure physicians that 285.1 is not a quality indicator and that only 998.11 is.

 

In fact, in its 2009 report card, HealthGrades no longer lists 285.1 as a complication of a procedure, Gold says. However, 998.11, Hemorrhage complicating a procedure is used in HealthGrades report cards. Gold notes that if anemia develops in a patient due to the condition that caused the bleed (GI bleed, menometrorrhagia, hip fracture, etc.), the 998.xx code should not be reported.

 

Also, CDI specialists should explain to physicians that the Agency for Healthcare Research and Quality (AHRQ) also excludes 285.1 from their algorithms regarding complications of surgery. Reassure physicians that 285.1 is no longer counted as a complication and their documentation of anemia and its cause justifies their utilization of resources, says Gold. “Otherwise, they gave that unit of packed cells away for free,” he says.

 

Remind physicians that the DRG system is a system of averages and means, and therefore if they only document the more severe cases (i.e., patients who undergo multiple transfusions, etc.), then they won’t average out the easy cases that result from P.O. (oral) iron, resulting in a poor and inappropriate performance profile for the physician, says Haik. “[This means] you would consider the anemia clinically significant even if only treated with oral iron, not just blood transfusions,” he says.

 

It is not uncommon for patients to lose several hundred cc’s of blood into a hip or long bone fracture site (unrelated to the surgery), Haik says. ”The drop of the hgb/hct won’t occur until after the surgery some 12 to 24 hours later as the plasma volume is restored with hydration, etc.,” he says. “And therefore, if addressed or treated, then the physician should get credit for the increased hospital use.”

 

With a femur (hip/long bone) fracture, says Gold, a patient may lose up to two units of blood whether the patient goes to the operating room or not. And, as Haik suggests, once the patient receives IV fluids, the hemoglobin level falls whether the patient goes to the operating room or not.

 

There are two considerations once that drop in hemoglobin happens, Gold says.

“Sure, ask the physician to document for consistency, but as ‘anemia due to acute blood loss from the femur fracture.’ Then, if it is treated, code it —if it’s not, don’t,” Gold says.

 

Similarly, with a GI bleed or extreme menorrhagia or severe hematuria, a person can easily develop anemia due to acute blood loss, Gold says. Here, unless the patient goes to the operating room, there’s no concern about “complications of surgery.” However, if the patient does go to the operating room (suture ligation of bleeding duodenal ulcer) or has a procedure (embolization of bleeding submucosal leiomyoma), as long as the documentation is provided as “anemia due to acute GI bleed” or “anemia due to acute blood loss from submucosal fibroid,” the 285.1 code should be assigned but not the 998.11 code. 

 

“That’s because the anemia was not due to any excessive blood loss from the procedure, but from the pathologic condition—and that’s not counted as a complication of a procedure anymore,” Gold says.