The authors followed 532 consecutive patients with episodic migraine (<15 days per month) for 1 year. Sixty-four patients (14%) developed chronic headache (≥15 days per month). The odds ratios for developing CH were 20.1 (95% CI: 5.7 to 71.5) comparing patients with a “critical” (10 to 14 days per month) versus “low” (0 to 4 days per month) and 6.2 (95% CI: 1.7 to 26.6) in patients with an “intermediate” (6 to 9 days per month) versus “low” headache frequency and 19.4 (95% CI: 8.7 to 43.2) comparing patients with and without medication overuse.

Discussion: Sixty-four (14%) of 450 patients developed headache chronicity during 1 year. We identified “intermediate” and “critical” headache frequencies and medication overuse as risk factors of headache chronicity.

It has been shown that medication overuse coexists in the majority of patients with chronic headache and that patients improve after withdrawal from the overused headache medication. This suggests a causal relationship between the medication overuse and chronicity of headache. Our study is in line with these findings showing that patients who overused their headache medication had a high risk of developing chronicity of headache. One-third of patients, however, developed chronic headache without medication overuse, suggesting that medication overuse is an important but not the only factor determining chronicity of headache.

Recently, we were able to show that patients who overused triptans developed headache chronicity faster and that they used fewer single dosages than patients who overused ergots or analgesics. In this study, we could not find any differences with regard to the type of drugs used, suggesting that usage of triptans per se does not necessarily bear a risk for chronicity of headache. This finding is, however, limited because of the relatively small sample size.

Another association was found between the initially high headache frequency and headache chronicity. Patients with an “intermediate” headache frequency had an increased risk for headache chronicity, which was further increased in patients with the “critical” frequency. The high headache frequency, however, should not be considered as a causal risk factor. The chronicity of headache could be a consequence of medication overuse or reflect a natural migraine fluctuation.

The regular intake of preventive medication was a univariate predictor for headache chronicity. One explanation for this unexpected finding might be that patients with high headache frequency received preventive headache medication much more frequently than patients with lower headache frequency. Therefore, the relationship between the regular intake of preventive medication and chronicity of headache might have been confounded by the indication of taking preventive drugs. However, inclusion of “preventive medication” into the final multivariable model did not add to the overall prediction of headache chronicity.

Our study has several strengths, including diagnosis of headache by an experienced headache specialist, standardized data collection, and a good response rate for the follow-up interviews. We used stringent inclusion and exclusion criteria for variable selection of the multivariable models, aiming to reduce model overfitting.

Several limitations must be considered. The follow-up data reflected the last 3 months and did not cover the time period of the first 9 months of follow-up. We studied a patient population from a specialized headache center, and therefore we cannot necessarily extend these results to other population. This could also explain the relatively high 14% 1-year cumulative incidence of headache chronicity. We could not control for psychological factors, which might play a role in the development of headache chronicity. Our study was not large enough to use data splitting to validate our model. Future population-based studies must determine whether the three predictors identified in our study can be generalized to other population and whether additional factors can be identified.

Comments: We found that triptans accounted for only 9% of 456 medication overuse headache patients at The New England Center for Headache, while butalbital mixtures accounted for 47%, opioids 37%, and aspirin 24%, just to put this in perspective (Bigal ME, Sheftell FD, Rapoport AM, Tepper SJ, Feleppa M. Triptan use in a migrainuer population. Determinants of preference. Neurology. 2003;60(Supp 1):A170-A171 (abs), Cephalalgia 2004, in press. Stewart J. Tepper