Guest post: Using the MUSIC mnemonic for documentation precision

by James S. Kennedy, MD, CCS, CDIP, CCDS

We know we need to document better so that our ICD-10-CM/PCS coding is complete. What are some systems that help do this?

I would like to credit my chairman of medicine at the University of Tennessee in Memphis during the 1970s, Dr. Gene Stollerman; those of you trained in Boston may remember him. Dr. Stollerman taught me to think critically as I listen to a patient’s history, construct a differential diagnosis, and further the patient’s evaluation in narrowing this differential toward what the patient likely has.

All of our patients' conditions have five components that we must evaluate and document completely in ICD-10-CM’s language if they are to be properly coded. These components can be remembered using the mnemonic MUSIC. They are as follows (using a patient with chest pain as an example):

• Manifestations. Most conditions have a manifestation that tells me an underlying disease is probably present. While any nursing or medical student can attest that a patient has chest pain, my diagnostic decision-making narrows this down as to whether it represents heartburn, pleuritic pain, angina pectoris, radicular pain, biliary colic, or “atypical” for any known chest pain syndrome. “Noncardiac chest pain” is like saying “non-pulmonary fever;” it states what the patient doesn’t have, not what he or she does have. I must be more specific.
• Underlying cause. Let’s presume that the patient’s presentation is consistent with angina pectoris. Underlying causes include not only atherosclerotic coronary artery disease but also ischemia with non-obstructive coronary arteries, coronary spasm, non-atherosclerotic coronary dissection (as in young women), coronary emboli or vasculitis, in-stent stenosis, thrombosis of a coronary artery bypass graft, aortic stenosis, and the like. Just because the cardiac cath is negative in the setting of a positive noninvasive study doesn’t mean the patient does not have myocardial ischemia.
• Severity or specificity. Angina pectoris can be resolved and, therefore, is not being treated (known in ICD-10-CM as a history of angina pectoris); stable (though asymptomatic) while being treated with nitrates, beta blockers, or calcium channel blockers; unstable or accelerated, meaning that the underlying cause has likely destabilized; or progressed to acute myocardial infarction, manifested by a rise and/or fall of a troponin value in the setting of ischemic symptoms or ECGs.
• Instigating or precipitating causes. Angina pectoris can worsen or be precipitated by anemia, shock, vasoactive drug use (e.g., cocaine, meth), thyrotoxicosis, a coronary or cardiac procedure, severe hypoxemic respiratory failure, and other causes. I must be explicit in stating this.
• Consequences. Angina or unstable angina/MI can result in right or left systolic heart failure, arrhythmias, systemic inflammatory response syndrome, cardiogenic shock, acute respiratory failure, Dressler’s syndrome, and sudden cardiac death.

As such, what I must do is refine the description of a documented diagnosis (e.g., pleuritic pain instead of just chest pain), put it in one of the five MUSIC categories (e.g., M for pleuritic pain), and construct differential diagnoses revolving the other four. This may prompt me to think of the pleuritic pain’s underlying cause (e.g., pneumonia), increased specificity or severity (e.g., due to community-acquired staph aureus), instigating cause (e.g., recent influenza A infection) and consequences (e.g., sepsis due to pneumonia, pleural effusion requiring thoracentesis, empyema). If I document the interrelationship of these MUSIC diagnoses (e.g., “due to,” “resulting in”), the coder will have a more complete picture from which to assign ICD-10 codes, appropriately affecting my risk-adjustment denominator.

Try this yourself with the following documented diagnoses: jaundice, altered mental status (be specific about the alteration first), abdominal pain, and hematuria.

One final note: Unlike outpatient billing, we in inpatient facilities are allowed to code uncertain conditions documented at the time of discharge as if they existed. If I suspect that my patient with an unprovoked deep venous thrombosis likely has an antiphospholipid syndrome that will be evaluated at a later date, and I document that in my discharge summary, the coder can code it and the risk-adjustment methodology will factor in credit for it.

Editor’s note: This article was adapted from the original in JustCoding. Dr. Kennedy is a general internist and certified coder, specializing in clinical effectiveness, medical informatics, and clinical documentation and coding improvement strategies. Contact him at 615-479-7021 or at jkennedy@cdimd.com. Opinions expressed are that of the author and do not necessarily represent HCPro, ACDIS, or any of its subsidiaries.