LETTER TO THE EDITOR Postconcussion headache: reply to editorial L. J. Stovner a,b , H. Schrader a,b , D. Mickevicˇ iene c , D. Surkiene c and T. Sand a,b a Norwegian National Headache Centre, Department of Neuroscience, Norwegian University of Science and Technology, Trondheim; b Department of Neurology and Clinical Neurophysiology, St. Olavs Hos- pital, Trondheim, Norway; and c Depart- ment of Neurology, Kaunas University of Medicine, Kaunas, Lithuania Correspondence: Lars Jacob Stovner, Norwegian National Headache Centre, St. Olavs Hospital, N-7006 Trondheim, Norway (tel.: +47 72575070; fax: +47 73598795; e-mail: lars.stovner@ ntnu.no). Keywords: concussion, debate, epidemiology, headache, post-traumatic Received 24 September 2008 Accepted 26 September 2008 When new scientific evidence goes against widely accepted truths, a vigorous debate is both expected and desirable. Our somewhat controversial conclusion, that long-lasting headache after concussion is not caused by head injury, is based on the finding that concussed patients in two cohort studies had similar levels of head- ache prevalence and headache frequencies 3 and 12 months post-injury as matched control groups with no head injuries [1]. We believe that the discrepancy between our results and those of previous studies is due to the fact that we have used new (in this context) and improved methodology (cohort studies with adequate control groups, performed in a country with no secondary gain and less negative expectations). We have never held the opinion, imputed to us in the editorial [2], that lack of specific defining characteristics of the headaches in itself is proof that headache is not caused by the concussion. However, we think that this corroborates and illu- minates our conclusion in an interesting way. Interpreting the acute headache after concussion as a primary headache induced by the stress of the situation explains why concussed patients and controls have similar long-term prognosis with regard to headache, and also the lack of a positive association between severity of the con- cussion and headache, clearly demon- strated in this and several previous studies (for discussion, see Ref. [1]). As to the second argument of Couch et al., that the concussion group, in con- trast to the control group, had an increase in headache prevalence from before the concussion; we have discussed this at some length in the article. We believe that the most probable explanation is that con- cussed patients in Lithuania tend to underestimate pre-injury symptoms, just as they do in the US [3,4]. If our critics were correct in their interpretation, this would imply that the concussion group, in both cohort studies, actually had signifi- cantly less headache than the controls before the accident, and became ÔnormalÕ with regard to headache prevalence after- wards. This seems highly unlikely. Our critics also state that the high baseline prevalence of CDH is a Ôchal- lengeÕ for our study, but they neither explain precisely why nor in what way they have Ôcontrolled for this high preva- lenceÕ. They also interpret our baseline CDH prevalences incorrectly. It is clearly evident from our Table 3 that CDH is not different at baseline in the concussion group and controls (P = 0.79). In com- paring our baseline prevalences with reported epidemiological data, it is important also to consider both the method and type of questionnaire and the sampling error of the reported estimates, which may vary considerably between studies. The lower 95% confidence limit of the CDH prevalence in our controls is indeed 5%, which is within the range reported in studies previously referred to by Couch et al. [5]. Recalculating our data into ratios may show the changes from baseline in another way, but ratios can not be interpreted without their associated confidence inter- vals, and they do not offer any better explanation for what we consider most likely to be underestimation of complaints before the concussion. In addition, Couch et al. have apparently not calculated correct ratios for CDH in their Table 1, as their ratios are far too high for all four columns. Moreover, we cannot see that the ratios presented in their Table 2 can possibly be calculated from our data in Table 3 [1]. Indeed, only one of the nine odds ratios (concerning headache fre- quencies) which can be calculated from Table 3 is significant, in analogy with the one significant p value reported by us for short duration headache (1–7 days per month before trauma) (spreadsheets of our calculations can be offered on request). In conclusion, these editorial recalcu- lations do not seem to add anything useful to the debate about postconcussion head- ache. We will also strongly argue that it is unacceptable and with no basis in our data when Couch et al. try to force our results to be in line with the conclusion of their own methodologically weaker study, claiming that 15% of CDH cases can be attributed to head and neck injuries [5] (our previous discussion of this study is referred in [6]). In continuation of this interesting and important debate, we believe, for the sake of clarity, that both parties should admit that there is a real difference of results and opinions. References 1. Stovner LJ, Schrader H, Mickevicˇiene D, Surkiene D, Sand T. Headache after con- cussion. European Journal of Neurology 2009; 16: 112–120. 2. Couch JR, Lipton RB, Stewart WF. Is post- traumatic headache classifiable and does it exist? European Journal of Neurology 2009; 16: 12–13. 3. Mittenberg W, DiGiulio DV, Perrin S, Bass AE. Symptoms following mild head injury: expectation as aetiology. Journal of Neurol- ogy, Neurosurgery and Psychiatry 1992; 55: 200–204. 4. Gunstad J, Suhr JA. Cognitive factors in postconcussion syndrome symptom report. Archives of Clinical Neuropsychology 2004; 19: 391–405. 5. Couch JR, Lipton RB, Stewart WF, Scher AI. Head or neck injury increases the risk of chronic daily headache: a population-based study. Neurology 2007; 69: 1169–1177. 6. Stovner LJ, Schrader H, Couch JR, Lipton RB, Stewart WF, Scher AI. Correspondence on ‘‘Head or neck injury increases the risk of chronic daily headache: a population-based study’’. Neurology 2008; 71: 381–383. e14 Ó 2009 The Author(s) Journal compilation Ó 2009 EFNS European Journal of Neurology 2009, 16: e14 doi:10.1111/j.1468-1331.2008.02359.x