The misdiagnosis of cluster headache: a nonclinic, population-based, Internet survey

We conducted the first nonclinic, Internet-based survey of cluster headache to investigate this population with regard to diagnostic problems encountered, effective and ineffective medications, problems obtaining medications through third-party payers, and symptoms as they relate to International Headache Society criteria.

Previous cluster headache surveys have been at specialty centers. These patients might be different from cluster headache sufferers in the general population. An Internet-based population of cluster headache sufferers who connected to a Web site responded to the questionnaire, and e-mailed it back to our site to be analyzed. We analyzed a total of 789 respondents, 76% men and 28% women.

Eighty-seven percent of respondents qualified as having cluster headache according to International Headache Society criteria. However, diagnosis was delayed an average of 6.6 years from the onset of symptoms. The average number of physicians seen before the correct diagnosis was made was 4.3, and the average number of incorrect diagnoses was 3.9. Seventy-one percent of respondents had undergone unnecessary magnetic resonance or computed tomography scans, and 4% had unnecessary sinus or deviated septum surgery. We found that many inappropriate medications such as propranolol, amitriptyline, and antibiotics were prescribed and that successful medications for clusters such as sumatriptan and oxygen were often denied due to a failure to understand the nature of this disorder. Seventy-seven percent of respondents were smokers. Seventy-four percent stopped smoking in an attempt to improve their condition; however, only 3% experienced relief.

The most alarming finding was the delay in diagnosing cluster headache in this population--an average of 6.6 years. The selection of medications demonstrated to be successful in the treatment of clusters proved effective for the majority of this population. Many respondents reported being denied some of these effective medications by their physicians or third-party payers. Using International Headache Society criteria for cluster headache, 87% of the respondents should have been correctly diagnosed by the first physician seen.