Diagnosis Code Overload
Know the difference between being thorough and going overboard.
By Jeremy Reimer, CPC
Recently, a client asked me to review the medical records of a patient involved in a relatively minor motor vehicle accident (MVA). Three days after the accident, the patient went to see a chiropractor. The patient complained of neck pain, and some tingling and numbness in his shoulders. He said this pain was causing him to have difficulty sleeping and giving him headaches, and that bending, stretching, and walking made the pain worse. After an initial visit, his provider diagnosed him with the following:
847.0 Sprain of neck
723.4 Brachial neuritis or radiculitis NOS
739.1 Nonallopathic lesions, cervical region
728.4 Laxity of ligament (cervical)
728.85 Spasm of muscle (cervical)
729.1 Myalgia and myositis, unspecified
847.1 Sprain of thoracic
724.4 Thoracic or lumbosacral neuritis or radiculitis, unspecified
739.2 Nonallopathic lesions, thoracic region
728.4 Laxity of ligament (thoracic)
728.85 Spasm of muscle (thoracic)
719.7 Difficulty in walking
780.5 Sleep disturbances
Red Flags Rise High
The first thing that jumped out at me about this claim was the sheer number of diagnoses rendered. This type of coding is commonly referred to as “kitchen sink” coding. Typically, you expect to see three or four diagnosis codes—or perhaps a fifth diagnosis if the injuries are severe, and/or cover multiple body areas. Because that was not the case in this accident, I decided to take a closer look at the 14 different codes this provider cited.
Take a Closer Look
Several of the reported codes struck me as odd. For instance, I’ve never encountered 728.4 (ligament laxity) to diagnose neck or back pain (or a neck or back injury). Further investigation uncovered the reasons why: ICD-9-CM codes 725-729 are used to diagnosis rheumatism, excluding the back. The cervical and thoracic regions are both areas of the back, so using 725-729 would be improper in this case. Ligament laxity is typically used to describe a chronic condition (>30 days after the accident), whereas this diagnosis was rendered three days after the MVA.
Similar reasoning applies to 728.85 (cervical and thoracic myospasms), which also falls within category 725-729 (rheumatism excluding the back), and should not have been used to describe this patient’s injuries. Neither 728.4 nor 728.85 are applicable diagnoses in this case.
The use of 719.7 (difficulty walking) also stood out. Applied correctly, 719.7 is for patients who suffer from difficulty walking, typically as a result of degenerative and chronic joint disease, which clearly was not the case here. The notes did not substantiate the patient could not walk, had developed a limp, or other walking abnormality; the records merely indicated that the patient claimed to experience more pain while walking (among other activities). In short, nothing in the documentation warranted reporting 719.7. As well, it would be premature to diagnose the patient with abnormal gait (781.2 Abnormality of gait). Should the symptoms persist, however, the more appropriate codes would be E813.1 Motor vehicle traffic accident involving collision with other vehicle injuring passenger in motor vehicle other than motorcycle, as well as E929.0 Late effects of motor vehicle accident.
Read the Guidelines
Although I’ve seen “headache” listed as a diagnosis plenty of times, this was the first time I had encountered 784.0 as a result of an MVA. ICD-9-CM codes 780-799 are for “Symptoms, Signs, and Ill-Defined Conditions.” The ICD-9-CM manual explains that these codes are to be used:
a. for cases in which a more specific diagnosis cannot be made even after investigating all the facts bearing on the case;
b. indeterminate or transient signs and symptoms;
c. provisional diagnoses in a patient who failed to return;
d. for cases referred elsewhere for investigation or treatment before a diagnosis was made;
e. when a more precise diagnosis was not available for any other reason; and/or
f. for certain symptoms which represent important problems in medical care and which might be desired to classify in addition to an unknown cause.
In this case, using 784.0 to describe the patient’s headaches does not meet the criteria for this range of codes. The facts of the case are pretty clear: The patient claims to have headaches as a result of the accident three days prior. The appropriate diagnosis would be 339.21 Acute post-traumatic headache.
Similar reasoning applies for 780.5 (sleep disturbances). The patient’s difficulty sleeping is not an isolated symptom of unknown origin; it is a direct result of the pain in his upper back caused by the accident. The patient stated his neck pain makes it hard to get comfortable enough to fall asleep at night. Based on the documentation, 780.5 is not warranted.
Double Check “Other” and “Unspecified” Codes
In my experience, the use of “other” and “unspecified” diagnosis codes can be viewed as a red flag. In this case, codes 780-799 are red flags because the origins of the patient’s symptoms are well known. Current ICD-9-CM, as well as ICD-10, contains a plethora of available classifications and subclassifications to describe a patient’s condition(s) in remarkable detail.
When it comes to diagnosis codes, there is a fine line between being thorough and going overboard. Physicians may diagnose the kitchen sink in an attempt to justify all the treatments rendered, or to exaggerate a patient’s injuries. Kitchen sink diagnosis coding can hurt patients in the long run by assigning diagnoses to their health history that may not have been applicable.
When trying to establish best practices and avoid kitchen sink diagnosis coding, ask yourself these questions:
- Are all of these codes medically indicated by the patient’s records?
- Are any of the rendered codes merely symptoms of other codes?
- Are the rendered codes as accurate and precise as possible, given the supporting medical documentation?
As a coder, you have the opportunity to educate the medical community about proper coding. Look for ways to show the benefits of being precise with the codes you choose. Diagnosis coding is always a case in which quality is more important than quantity.
Jeremy Reimer, CPC, is the president of Medical Coding Litigation Services, which provides medical coding and billing anlysis to insurance carriers, legal counsel, and companies involved in personal injury litigation. He is an author and frequent lecturer on medical coding and billing fraud. Jeremy is also an adjunct professor at Hillsborough Community College, where he teaches Advanced Medical Coding. He can be reached at: firstname.lastname@example.org.