Neurology practices bill a concentrated set of high-value diagnostics and procedures, which means the same denial reasons show up across every patient for a given payer. That concentration cuts both ways: denial patterns are predictable and repeat at volume, but one systemic fix can clear a large backlog all at once. Four patterns account for the majority of neurology denial volume, and each has a different root cause and a different path to recovery.
The EMG and nerve conduction bundling trap
Nerve conduction studies (NCS) and needle electromyography (EMG) are separate diagnostic procedures with separate CPT codes, but payers routinely bundle one into the other when both appear on the same claim. The result is a CO-97 on the second service, with the payer treating it as included in the first.
Whether the bundling is correct depends on the specific code pair and the payer's NCCI edit for it. When the modifier indicator is 1, a modifier 59 or an X-modifier (EPSU) on the second service line is the correct override, and the appeal needs to document that each study was a distinct diagnostic act: different nerve segments or muscle groups, each generating separate tracings and a separate clinical interpretation. When the indicator is 0, the codes are mutually exclusive and the bundling stands.
Bilateral nerve conduction studies add a second layer. Some payers accept modifier 50 on a single line; others require separate LT and RT lines. Using the wrong approach produces CO-4 on the bilateral lines. Keep a payer-specific reference for bilateral NCS billing requirements, because the rules are not uniform.
Botox for chronic migraine: authorization and J-code gaps
OnabotulinumtoxinA (CPT 64615 for the injection, J0585 for the drug units) for chronic migraine is one of the more authorization-intensive procedures in neurology. Most commercial payers require prior authorization for every treatment cycle, not just the first one. Practices that obtain auth for the initial session and then bill subsequent sessions without renewing it generate CO-197 denials on every follow-up cycle until the pattern is caught.
The drug J-code is a separate issue. Many payers require independent authorization for the J-code, distinct from the procedure authorization. A practice that has procedure auth but not drug auth will see every J0585 line deny while the 64615 pays. The two authorizations need to be verified and attached separately before each treatment date.
- Preventive documentation baseline. Most payers require failure of at least two preventive medications before approving Botox for migraine. Document those trials specifically: drug name, dose, duration, and outcome. Vague notes that list medications without failure detail lose auth reviews at a higher rate.
- Headache day threshold. The clinical criteria for chronic migraine require 15 or more headache days per month. If the visit note does not document headache frequency explicitly, the auth reviewer cannot verify the threshold is met.
- Unit count precision. Coverage criteria typically specify a dose range, commonly 155 to 195 units. Document the exact units injected per site and the total units used. A vague "Botox administered" note without unit count fails the documentation standard.
EEG and neuroimaging medical necessity
CO-50 medical necessity denials on EEG and MRI brain requests are common and usually preventable. Payers apply clinical criteria that specify which diagnoses and clinical presentations qualify, and appeals that simply resubmit the original note without addressing those criteria consistently lose.
For EEG, the documentation needs to establish why the study was ordered: a first unprovoked seizure, characterization of a known seizure disorder, evaluation of episodic altered consciousness with an epilepsy differential. A visit note that records the symptom and orders the EEG without a clinical reasoning statement gives the payer's reviewer nothing to evaluate. Include the prior workup, the specific clinical question, and the physician's explicit rationale for the study.
For brain MRI, payers with more aggressive utilization review tend to deny on headache unless the note documents red flags: new or changing headache pattern, neurological signs or symptoms, failure of appropriate empiric treatment, or a specific differential requiring imaging to exclude. Routine headache management without any of those elements is a frequent CO-50 target.
Sleep study benefit routing
Attended polysomnography (PSG, CPT 95810 or 95811) and home sleep apnea testing (HSAT, CPT 95800 to 95806) route to different benefit tracks on many commercial plans. A claim submitted under the wrong benefit produces a denial that reads as a coverage exclusion but is actually a routing error. The fix is re-coding to the study type actually performed and resubmitting to the correct benefit.
A related issue: many payers require documentation of a clinical sleep evaluation before approving HSAT, particularly for Medicare patients. The clinical note needs to document the assessment of sleep-related symptoms and the indication for the study. Missing that documentation step produces a CO-197 on the HSAT claim even when the study was ordered appropriately.
Treating denials as patterns, not one-off claims
The single most effective thing a neurology billing team can do is sort denied claims by CARC code and payer before working any individual claim. A CO-97 cluster on NCS codes from one carrier means a systemic NCCI edit issue. A CO-197 cluster on Botox J-codes means an auth process gap. Working those as batches, with a single root-cause fix that prevents the next round, recovers more revenue faster than treating every claim as isolated.
The revenue leakage is real in neurology: smaller diagnostic claims that never enter a work queue because the cost to appeal each one manually exceeds its value. A CO-97 on a $65 NCS line is economically invisible one claim at a time but becomes significant once you count every instance across a year of the same payer and code.