Moderate Levels of Concordance With Telephone Classification Interviews in Classifying Common Chronic Headache Disorders

Nonheadache and headache specialists have moderate levels of agreement in classifying common chronic headache disorders after using the telephone classification interview model, according to a study.

Many nonheadache specialists have a difficult time differentiating different types of chronic headaches, such as migraine. As a result, many patients are misdiagnosed and receive inappropriate treatments. To address this issue, researchers created a telephone classification interview logic model so nonheadache specialists can correctly classify common headache disorders.

In this study, investigators contacted neurologists, headache specialist nurses, patients with chronic headaches, and general practitioners with specialized interest in headaches.

A total of 26 delegates attended the headache classification conference and developed a logic model. The logic model consists of 5 steps that are used to exclude patients with symptoms suggestive of secondary headaches and primary headaches that are not migraines or chronic tension type headaches. Patients with significant headache changes in the past 3 months or patients with headaches triggered by coughing, sneezing, or straining are screened out and classified as having secondary headaches. Headaches that are continuous and strictly unilateral and have any of the autonomic features or restlessness and headaches lasting <4 hours only occurring at night or strictly unilateral and having any cranial autonomic features are also excluded as having headaches other than chronic migraine and tension type headaches.

Patients with an aura ≥8 days a month for ≥3 months are classified as definitive chronic migraine. Patients with aura 1 to 7 days per month with 2 of 4 headache characteristics (unilateral, pulsating, moderate/severe pain, aggravated by physical activity), and 1 of 2 associated symptoms (nausea, vomiting, phonophobia, photophobia) for ≥8 days a month for ≥3 months are also classified as definite chronic migraine. Patients fulfilling only the headache characteristics or the symptoms or having neither criteria are classified as probable chronic migraine.

Patients whose headaches have no aura but have both the required headache characteristics and associated symptoms criteria are classified as probable chronic migraine; if the headaches are ≥8 days a month for ≥3 months, then the classification switches to definite chronic migraine. Patients with no aura, and either 2 of 4 headache characteristics or 1 of 2 associated symptoms are classified as probably chronic migraine; patients who did not fulfill either criteria are classified as chronic tension type headache.

If the headache medication was taken in 10 days per 3 months and nonsteroidal anti-inflammatory drugs/acetaminophen were used ≥15 days per month for ≥3 months, then the headache is associated with medication overuse. Other drug combinations used for ≥10 days/month for ≥3 months is also associated with medication overuse.

In a sensitivity agreement test between chronic migraine and chronic tension-type headache, there was a good agreement between nurses and doctors when using this model (prevalence-adjusted bias-adjusted κ [PABAK] = 0.76). A sensitivity test between chronic migraines and other types of headaches also revealed a very good agreement between doctors and nurses (PABAK= 0.81). Agreements between medication overuse headaches and nonmedication overuse headaches between nurses and doctors were moderately good (PABAK= 0.41).

The use of the telephone classification interview logic model resulted in moderate to very good levels of agreement between nurses and doctors. Future randomized-controlled trials should test the logic model and how well it works.

Reference:

Potter R, Hee SW, Griffiths F. Development and validation of a telephone classification interview for common chronic headache disorders. J Headache Pain. 2019;20(1):2. doi: 10.1186/s10194-018-0954-z.