Cardiology Coding Alert

Many cardiology coders have the false impression that they must use a pretest diagnosis when they bill for a test, such as an echocardiogram or EKG.

As long as the medical necessity requirements of the tests are met, either signs/symptoms or final results of tests may be used to obtain reimbursement. But using the post-test diagnosis is preferable because, according to the Health Care Financing Administration (HFCA) guidelines, diagnoses should be coded to the highest degree of specificity.

Suppose for example a family physician sends a patient with a diagnosis of syncope (780.2) to the cardiologists office for an echo. The test uncovers an ischemic cardiomyopathy. Nevertheless, some coders will attach the syncope diagnosis to the procedure code for the echo. However, that code may not provide medical necessity for the test and, therefore, reimbursement will be denied.

The rationale in the coders mind for the incorrect coding is that the diagnosis that prompted the test to be ordered in the first place should be the one attached to the charge for the test.

However logical this may seem, it contradicts the HCFA guidelines and often will result in denials for tests that should have been reimbursed. Some coders call using the diagnosis from the results of a test back-coding and think it is improper.

The uncertainty over diagnosis codes for tests has been getting worse in the last 18 months, says Dari Bonner, CPC, CPC-H, CCS-P, president of Xact Coding and Reimbursement in Port St. Lucie, FL.

Bonner points to a proverbial mountain of guidelines from Medicare and the American Medical Association, all of which maintain that it is perfectly legitimateand usually preferableto use the results of the test as the diagnosis to accompany the charge for it.

Post-Test Diagnosis

According to Medicare guidelines on the use of ICD-9 codes revised on Oct. 1, 1996, physicians may not use rule outs or suspected as a reason for performing a diagnostic test. In other words, the test cannot be billed simply to rule out a disease, or because an illness is suspected. Section H of the official ICD-9 CM guidelines for coding and reporting states: Do not code diagnoses documented as probable, suspected, questionable, rule out, or working diagnosis. Rather, code the condition(s) to the highest degree of certainty for that encounter/visit, such as symptoms, signs, abnormal test results or other reasons for the visit.

Susan Callaway-Stradley, CPC, CCS-P, an independent coding consultant in North Augusta, SC, says it is significant that the following paragraph states this is contrary to the coding practices used by hospitals and medical record departments for coding the diagnosis of hospital inpatients. According to Callaway-Stradley, the inpatient coding customs may well account for some of the confusion about which diagnosis code to use for outpatients.

If the test results return negative, then signs or symptoms would be the only appropriate diagnosis for the test, since rule outs have been, well, ruled out. But what if
the test returns positive? According to section D of the same guidelines, codes that describe symptoms and signs, as opposed to diagnoses, are acceptable for reporting purposes when an established diagnosis has not been diagnosed (confirmed) by the physician.

But if the test returns positive, then an established diagnosis has been confirmed. Clearly, then, the post-test diagnosis provides the highest degree of certainty required by HCFAs own ICD-9 guidelines.

What it all boils down to, Bonner explains, is the determination of medical necessity. For example, if a patient enters the cardiologists office complaining of chest pain and is given an EKG that returns with a diagnosis of gastritis (not an acceptable diagnosis for an EKG), then the ICD-9 code for chest pain (786.50, chest pain, unspecified) would have to be used to provide medical necessity for the EKG.

But if the EKG showed the patient had myocardial infarction (410.9, acute myocardial infarction, unspecified site, NOS), then the test result canand shouldbe used, because the medical necessity requirements for an EKG include MI.

Screening Tests are the Exception

Hospital and physician coders operate under different guidelines which, in the case of screening tests, may add to the confusion about which diagnosis code to use for tests of any kind. According to Coding Clinic, the ICD-9 publication used by hospital coders, services that are performed for screening may use the diagnosis discovered during the test as the code for the test.

But Medicare guidelines state something quite different, Bonner says. In the absence of illness, injury or symptoms, Medicare says any test performed is a screening exam that it will not reimburse, regardless of the outcome of the test.

In other words, Bonner says, physicians need to indicate medical necessity for performing the test, whereas the hospital doesnt. For example, a patient who has extreme anxiety may self-refer to the cardiologist to have an EKG. There is no organic problem, only the patients worry. The EKG turns out normal, and the patient is reassured. But even if the test had turned out positive, a diagnosis based on the test cannot be used because the test started out as a screening.

Differentiating between a screening and other tests, however, is not difficult. If the patient is asymptomatic and receives a test, it is classified as a screening; however, if the patient exhibits signs or symptoms, the test is diagnostic, and either signs/symptoms or the final results may be used for the diagnosis code, depending on which is more medically necessary.

Note: Callaway-Stradley says that regardless of whether the test in question was a screening or one that started out with a patient presenting symptoms, practices must code appropriately. And that means coming to terms with the fact that insurance carriers sometimes are not going to pay for the tests and the patient will have to pay the bill, she says.

Above all else, Callaway-Stradley cautions practice coders not to fall into the trap of using diagnosis codes simply to get the carrier to pay for the test. Instead, she recommends that all cardiology practices be pro-active in efforts to inform their patients that some tests will not be covered.