Neurosurgery Coding Alert

Question: Does an undiagnosed new problem with uncertain diagnosis automatically translate to a moderate level of problem complexity when calculating medical decision making (MDM)?

Colorado Subscriber

Answer: You’re right that a condition that is undiagnosed as of the date of service (DOS) has to factor into the MDM calculation for the evaluation and management (E/M) service. For example, a headache could just be a headache, but if a provider has a suspicion that it could be something worse, the differential diagnosis needs to be documented. This shows the complexity of the encounter, which will very often rise to the moderate level of MDM for such conditions.

The confusion here revolves around the different ways ICD-10-CM and CPT^® operate in regard to uncertain diagnoses. With ICD-10-CM, you are bound by Guideline IV.H, which tells you not to document a condition as “as ‘probable.’ ‘suspected.’ ‘questionable,’ ‘rule out,’ ‘compatible with,’ ‘consistent with,’ or ‘working diagnosis’ or other similar terms indicating uncertainty.”

In other words, don’t code a probable or rule-out ICD-10-CM code for the diagnosis portion of the encounter. The potential for a serious problem based on the simple fact it’s an unknown is not reason enough to bump an encounter up a level in MDM. However, those details having to do with the physician’s suspicions (if any) should still be included in the documentation and therefore if those details outline sufficient complexity in that unique case, it is quite possible the encounter will warrant that higher level.

Keep in mind also that a higher level of MDM will also require appropriate level of complexity for the amount of data reviewed or the risk of the treatment plan involved.