Home

Is asking for clarification ’leading’?

Association of Clinical Documentation Improvement Specialists, June 2, 2008

by Robert S. Gold, MD

In 2000, CMS released a note that forbade individuals from asking physicians questions and assigning codes from their answers. CMS followed this with a subsequent note stating that it was acceptable to ask physicians questions and to assign codes from their answers. And HIM professionals have been in a quandary as to how to ask questions ever since.

Eventually, a Code of Ethics was published that included a statement saying it is inappropriate to ask a “leading” question. And just as in the previous sentence, the word “leading” was in quotation marks. However, the Code of Ethics didn’t define “leading.” It left it up to the interpreter to do so.

Many of us in the consulting and HIM professions have read CMS’ guidelines as well as Coding Clinic from cover to cover, have discussed the topic with HIM gurus, and have come to the conclusion that there is no true definition for what constitutes a “leading” question.

As a result, American Health Information Management Association is returning to the drawing board to better define the ethics involved in asking questions of physicians.

To summarize our independent conclusions, “leading” means that it is inappropriate to ask a physician to provide documentation regarding a condition that has no supportive basis in the medical record.

Additionally, asking a physician to provide documentation of a symptom or of an x-ray finding as a “condition” when the diagnosis is clear in the medical record is also unacceptable.

Lastly, it is inappropriate to ask a physician to document a condition:

  • That is based on an abnormal lab test but that is irrelevant to his or her management of the patient
  • That may be a complication/comorbidity (CC) or major CC (MCC), but that is untreated
  • For which the lack of attention to, or lack of treatment of, would not lead to meeting Uniform Hospital Discharge Data Set criteria for a valid secondary diagnosis
  • Simply for the financial benefits of the bill

If a physician uses terminology that is bound to lead to incorrect code assignment, it is appropriate to guide that physician to words that would lead to a code for the condition that:

  • The patient absolutely has
  • The physician treats
  • Is a major part of the patient’s total care

A physician may use incorrect terminology because he or she doesn’t know the verbiage that is appropriate to the case and to the care of the patient. In this case, it is also completely appropriate to help a physician document a condition that is fully supported in the body of the medical record.

Some of these statements are certainly bold—and I’m sure that some of these sweeping statements are also mystifying—so I’ll provide you with a few examples.

1. A physician performs an echocardiogram for evaluation of a murmur, heart failure, or an acute myocardial infarction admission. The echocardiogram shows that the left ventricle is dilated and that the ejection fraction is 20%. The physician documents “cardiomyopathy” and “EF = 20%.”

It is inappropriate not to seek clarification if the findings meet the physician’s criteria for chronic left ventricular systolic failure as a background condition, along with whatever else the patient has in addition.

Why? It takes time for a left ventricle to dilate; thus, it’s chronic.

A left ventricular ejection fraction under 40% is defined by the American Heart Association and the American College of Cardiology as systolic heart failure. At the very least, the patient has chronic left ventricular systolic failure. You should therefore assign code 428.22 with the documentation to back that up.

However, physicians frequently call dilated hearts “cardiomyopathy.” Veterinarians only recognize dilated hearts as “cardiomyopathy.” It is almost criminal not to direct the physician’s documentation with a request to provide the verbiage “chronic systolic failure” or its equivalent.

You must ask for clarification of the condition that caused the cardiomyopathy. Was it ischemic, alcoholic, valvular, or toxic? Otherwise, the only code you could assign is 425.4, and that’s unacceptable in the Medicare population because that’s the code for congenital cardiomyopathy and that’s not what the patient has.

If the patient had acute heart failure, with shortness of breath, and elevated brain natriuretic peptide levels, it is necessary to determine whether it was acute right heart failure or acute left heart failure. If it was left heart failure, was it due to acute systolic dysfunction, acute diastolic dysfunction, or both? These are the only acceptable choices.

2. An elderly patient with a history of chronic obstructive pulmonary disease and heart failure is admitted from a nursing home with pneumonia. The patient is treated with Rocephin and Zithromax, which seems to be the routine at many hospitals for almost every patient with pneumonia, regardless of age or other risk factors.

It would be inappropriate to ask the physician to document something like aspiration pneumonia or gram negative pneumonia. Why? Because the treatment provided was not the treatment for aspiration or gram negative pneumonia. Physicians treat almost everyone who comes through the door with this exact same regimen, and the patients certainly don’t all have aspiration or gram negative pneumonia.

On the other hand, when the patient fails to respond to the original regimen, the cultures demonstrate a specific organism, and the physician alters the antibiotics to cover that specific organism, then it is very appropriate to ask for clarification if an organism caused the pneumonia.

Similarly, if the workup demonstrates aspiration events while in the hospital and the physician changes the antibiotic regimen to one that is appropriate for treatment of aspiration pneumonia, it would be irresponsible not to ask whether the physician felt it was aspiration pneumonia.

3. A patient arrives with shortness of breath, and an x-ray shows pulmonary edema. The patient had experienced atrial fibrillation with rapid ventricular response at 204/minute. Medical management of the tachyarrhythmia was successful in controlling the heart rate, the patient responded with easier breathing, and the x-ray cleared.

It is inappropriate to ask the physician to document pulmonary edema or pulmonary venous congestion so that you could assign code 514. That’s an x-ray finding, nothing else. When coders use the x-ray finding as a substitute for the condition that the patient has in order to bump the relative weight of the DRG assignment, that’s more than inappropriate—it’s immoral. If there is any suggestion that the patient had acute diastolic heart failure due to the tachyarrhythmia and acute pulmonary edema due to a cardiac condition, one wouldn’t reach under the table so much. Do not do it.

4. A patient comes in after three days of severe nausea, vomiting, and diarrhea from what a physician determines to be viral gastroenteritis. The patient also has altered mental status, dry mucus membranes, and a creatinine level of 5.8. After two days of rehydrating, the creatinine drops to 1.1. The physician documents viral gastroenteritis, nausea, vomiting, diarrhea, and the creatinine values.

There is nothing wrong with asking the physician—based on guidelines established as valid by the National Kidney Foundation and described in Coding Clinic in the references for acute renal failure—whether this case meets the physician’s criteria for acute renal failure due to dehydration (or some other cause).

It is okay to ask because the patient has acute renal failure, but the words the physician used in the documentation would not lead to the codes that describe what’s actually wrong with the patient.

Consequently the data are flawed, and it is not right to leave the documentation as is. How can you possibly ask an open ended question in such a circumstance? You have to lead the physician to words that would result in correct code assignments.

Does this always apply? No, not always. Sometimes, you just don’t know. And that’s okay. You may have to ask an open-ended question.

For example, if there’s no evidence regarding the cause of a stroke with which the patient was transferred to your facility, you might ask the physician whether the stroke is hemorrhagic or ischemic. And if it is ischemic, is it embolic or locally occlusive?

Why can you ask this? Because those are the only choices, and the physician has to select one of them.

But if the physician hasn’t provided any reason for a patient’s syncope, you might ask an open-ended question such as: Have you determined a cause of the syncope?

And you know what? Sometimes, physicians just don’t know.

Editor’s note: Dr. Gold founded DCBA, Inc., a consulting firm in Atlanta that provides physician-to-physician programs in clinical documentation improvement. The goals are data accuracy, profile management, and compliance, either in the inpatient or outpatient arenas. He can be reached by phone at 770/216-9691 or by e-mail at DCBAInc@cs.com.

Most Popular