Short-term diagnostic stability of probable headache disorders based on the International Classification of Headache Disorders, 3rd edition beta version, in first-visit patients: a multicenter follow-up study

Background: A "Probable headache disorder" is diagnosed when a patient's headache fulfills all but one criterion of a headache disorder in the 3rd beta edition of the International Classification of Headache Disorder (ICHD-3 beta). We investigated diagnostic changes in probable headache disorders in first-visit patients after at least 3 months of follow-up. Methods: This was a longitudinal study using a prospective headache registry from nine headache clinics of referral hospitals. The diagnostic change of probable headache disorders at baseline was assessed at least 3 months after the initial visit using ICHD-3 beta. Results: Of 216 patients with probable headache disorders at baseline, the initial probable diagnosis remained unchanged for 162 (75.0 %) patients, while it progressed to a definite diagnosis within the same headache subtype for 45 (20.8 %) by fulfilling the criteria during a median follow-up period of 6.5 months. Significant difference on the proportions of constant diagnosis was not found between headache subtypes (P < 0.935): 75.9 % for probable migraine, 73.7 % for probable tension-type headache (TTH), and 76.0 % for probable other primary headache disorders (OPHD). Among patients with headache recurrence, the proportion of constant diagnosis was higher for probable migraine than for probable TTH plus probable OPHD (59.2 vs. 23.1 %; P < 0.001). The proportions of constant diagnosis did not significantly differ by follow-up duration (> 3 and <= 6 months vs. > 6 and <= 10 months) in probable migraine, probable TTH, and probable OPHD, respectively. Conclusions: In this study, a probable headache diagnosis, based on ICHD-3 beta, remained in approximately three-quarters of the outpatients; however, diagnostic stability could differ by headache recurrence and subtype. Probable headache management might have to consider these differences.