Home  |   Subscribe  |   Resources  |   Reprints  |   Writers' Guidelines

November 7, 2011

Cause for Confusion
By Judy Sturgeon, CCS, CCDS
For The Record
Vol. 23 No. 20 P. 8

Coders whose professional responsibilities include inpatient hospital diagnosis-related group (DRG) coding are likely to agree on at least one subject: Determining how the cause of a particular diagnosis might affect the final code sequencing and DRG assignment can be particularly confusing.

While there is certainly enough overall variation in the rules for assigning ICD-9-CM codes, what makes this issue so concerning is its effect on the final DRG assignment and its subsequent impact on final payment and data reporting.

Begin with the definition of principal diagnosis: “That condition established after study to be chiefly responsible for occasioning the admission of the patient to the hospital for care.” At first glance, this seems to imply that once you know the cause of the symptoms necessitating admission status, you would simply replace the code for the symptom with the code for the cause. But could anything be that simple when it comes to coding? Not a chance.

Let’s take a look at the conflicting rules regarding cause and effect or symptom and etiology. Some rules are simple. Consider several scenarios in which each patient is admitted to the hospital for chest pain, and the plan is to work up the cause in order to determine what to do next.

Patient A: The cause of chest pain is determined to be gastroesophageal reflux disease (GERD). Treatment is with a proton pump inhibitor medication to control the acid, and the patient is discharged for follow-up with a gastrointestinal specialist.

The principal diagnosis is the cause of the chest pain: GERD. The chest pain is not coded.

Patient B: The cause of chest pain is determined to be angina. The physician documents the findings of a cardiac catheterization performed the previous month showing moderate coronary atherosclerosis (CAD) that was not severe enough for a bypass or even a stent. The angiotensin-converting enzyme inhibitor dosage is increased, and the patient is counseled to stop smoking and modify his high-fat diet. He is discharged with instructions to follow up with his primary care physician in one month. No specific cause and effect between the CAD and the angina are documented.

CAD is the principal diagnosis, followed by the code for angina. If a patient has any history of CAD and has angina, the cause is presumed to be CAD unless documented to the contrary, and CAD is coded first. Chest pain is not coded at all. (See AHA Coding Clinic, fifth issue of 1993 and second quarter of 1995.)

Patient C: The cause of chest pain is determined to be acute pulmonary edema, a result of the patient forgetting to refill his Lasix prescription and thus exacerbating his congestive heart failure (CHF). Treatment is a course of IV Lasix until both the chest pain and the pulmonary edema resolve. The patient is discharged home with strict warnings to comply with his medical regimen.

Sequence the CHF first and do not code either the chest pain or the acute pulmonary edema. (See the index and tabular conventions and instructions for edema, lung, acute, with heart failure, congestive, 428.0.)

Patient D: The cause of chest pain is found to be metastatic lung cancer. The patient, who was diagnosed some months ago, has not responded to chemotherapy or radiation therapy, and the tumors have now spread to the mediastinum. The patient is accepting palliative care only; he is put on IV morphine and discharged to home with hospice.

The pain is to be sequenced first but do not use code 786.59 for chest pain; 338.3, Neoplasm-related pain, was created for this situation. In this case, because only the pain is being treated and was the reason for admission, neoplasm-related pain is the principal diagnosis. “Since the neoplasm code will provide information regarding the specific site, an additional code for the site of pain should not be assigned” (AHA Coding Clinic, 2007, second quarter, pages 13-14).

As if these examples aren’t enough to cause your own chest pain, consider the probable cause and the multiple cause scenarios. Even they have variations in how they must be handled depending on the terminology used in the physician documentation. The following are examples based on the Official ICD-9-CM Coding Guidelines, Section II, Selection of Principal Diagnosis:

Patient E: The cause of chest pain is documented as “probably due to the patient’s GERD.” This is charted in the history and physical and the progress notes on days 1 and 2. However, on day 3, the diagnosis is listed as chest pain and the probable cause is never mentioned in subsequent documentation or on the discharge summary.

The principal diagnosis must be chest pain unless the probable cause is documented at discharge (ie, on a discharge progress note or in a discharge summary). A coder can query the physician for an addendum with the most probable cause of the chest pain. If the doctor documents this probable cause in the chart, then the principal diagnosis can be changed to GERD and the chest pain is not coded.

Patient F: The cause of chest pain is documented on discharge as follows: “MI [myocardial infarction] ruled out. CP [chest pain] might be due to GERD or MSK [musculoskeletal] strain. Discharge patient for follow-up with primary care physician for further diagnostics.”

Finally, it’s OK to code chest pain as the principal diagnosis, followed by codes for the differential diagnoses of reflux and musculoskeletal strain.
Patient G: The cause of chest pain is documented as stemming from both a community-acquired pneumonia and an exacerbation of CHF. The patient is treated with IV antibiotics and IV Lasix.

The coder may sequence either the CHF or the pneumonia first, with the other as secondary. Chest pain is not coded. Unlike a documented differential diagnosis, this case specifies that the patient has both conditions and that the chest pain is due to both.

Is coding sometimes confusing and apparently contradictory? Yes, it certainly is. The more we dig into detailed rules and diagnosis-specific rules, the worse it can seem. The rules fit the circumstances of each case or each chapter, and the ambiguities of physician documentation can make matters even more complex. But remember, if it was easy, coders would be earning minimum wage, and if it was fun, they’d be paying the hospital instead of the other way around.

Learn the rules, embrace the challenges, and you’ll find plenty of good reasons to keep loving to code.

— Judy Sturgeon, CCS, CCDS, is the clinical coding/reimbursement compliance manager at Harris County Hospital District in Houston and a contributing editor at For The Record. While her initial education was in medical technology, she has been in hospital coding and compliance for 21 years.