Integrated headache was established to provide multidisciplinary treatment to improve treatment for difficult-to-treat headache patients. Prospectively collected observational data of a large cohort of headache patients treated in our tertiary headache center showed overall improvement of headache frequency in 57.8% of the patients. Thirty-six percent of patients achieved a ≥ 50% reduction of headache days per month irrespective of primary headache subgroup. In contrast no improvement (20.9%) or worsening (21.3%) was reported in remaining patients. Headaches usually show a fluctuating headache frequency influenced by different factors (for example psychiatric comorbidity, stress, live-events) it is expectable that a part of them did not improve or wosen during time course. As shown in a recent study of our group adherence to live style modifications predict primary outcome [9
]. It is not surprising that short or single interventions in a headache center did not result in improvement of headaches in all patients. In absolute terms reduction of headache days in TTH was more pronounced than in migraine. Multivariate analysis revealed best outcome for patients older than 40 years compared to the younger. We could not explain this finding. Another important variable is the number of headache days per month at baseline. Patients with migraine and migraine and TTH who suffered from nearly daily headache (> 25 headache days per month) showed the best outcome, whereas patients with the lowest number of headache days (0-5) at baseline showed almost no change in headache frequency.
Regarding offerd consulation by neurologist and if possible once in addition by a physical therapists and/or a psychologist differentiation on outcome one year later can not be expected. Data does not suggest that additional MTP resulted in a better outcome. However, the study was not designed to compare these strategies. MTP treatment was indicated by headache frequency and burden of disease, but it was only available if treatment costs were covered by the health insurance. In addition treatment in the MTP depends on motivation of the patient and living distance to the headache center. Therefore treatment allocation was not only determined by medical aspects. Independet from this we recommend MTP especially in patients suffering from MOH. Succesful treatment of MOH resulting in reduction of the reported hiph relapse rate was shown in recent published data due to MTP [9
]. This might be explained by specific education and main emphases on MOH during the MTP.
Jensen et al. performed a 2-year systematic follow-up in the Danish Headache Centre in order to characterize patients and treatment results. They identified predictors for good outcome as female gender, migraine as primary headache, triptan-overuse and a mean headache frequency of 10 days/month, whereas tension-type headache and overuse of simple analgesics predicted poor outcome [11
]. These results are in some aspects contradictory to our findings. We found no significant influence of gender or headache type on primary outcome (≥ 50% reduction of headache frequency). In a population-based longitudinal study Bigal et al. assessed the influence of baseline body mass index on the response to headache preventive treatment. Patients suffering from episodic, chronic or transformed migraine who sought care in a headache clinic were included. Baseline information included headache frequency, number of days with severe headache and headache-related disability. The same information was obtained after three months of preventive treatment. After treatment, headache frequency declined in the entire population but no significant differences were found among BMI groups. Furthermore, BMI did not account for changes in disability, headache frequency, or in the number of days with severe headache per month [12
]. These results are in line with our data in showing no significant difference in BMI groups and reduction of headache days with categorization of BMI into only two subgroups (< 25 vs. ≥ 25) and a longer follow-up (twelve months). High BMI was a risk factor for increasing headache frequency in some studies [13
] but not others [15
]. Moreover, we could not find an association between smoking and reduction of headache days. These findings are supported by a metanalysis of three studies focusing on the association of lifestyle factors (BMI, alcohol, smoking and physical activity) and headache prevalence in Germany which also found no association between migraine and obesity or smoking [17
In order to investigate the association between level of education and reduction of headache frequency, the education level was split into a high and low education group. There was no significant difference between level of education and reduction of headache frequency (≥ 50% reduction). However, 91.4% who were diagnosed with MOH were in the low-education group, indicating that a lower level of education may be a risk factor for overuse of medication. This is supported by observations from Atasoy et al. who evaluated headache characteristics, socioeconomic and educational variabilities in subgroups of MOH and migraine patients. Their data showed that the frequency of migraine attacks as well as the duration of MOH and lower income was more frequent in low-educated migraine-patients [18
]. Scher et al. identified level of education as factor of headache chronification [13
], which has been shown for MOH, too [19
]. Furthermore, we did not find an association of marital status and reduction of headache days. Being married was associated with better prognosis in another study [13
]. The consequence of these observations is that no patients should be excluded from integrated headache care based on baseline variables.
We showed cost effectiveness for integrated headache care treatment [20
]. A formal socioeconomic analysis was not part of the present study and will be the focus of future research.
Strength of our study is the prospective design, the large cohort and the long follow-up period (12 months). It is important to investigate long-term effects of headache therapy beyond the 3 months observation time in clinical trials. Changes in medical treatment as well as changes in lifestyle and behaviour need time to show an impact on headache frequency. One of the major limitations of the study is the significant number of patients lost to follow-up and the high rate of dropouts, as well as the fact that we could not present a parallel group design with a cohort receiving standard care from general practitioners. Moreover this is a prospective observational study and not a controlled trial, which has impact on completeness of the data therefore missings were indicated in the tables. There is a clear need for future studies randomizing patients to different treatment modalities to prove the best care. Moreover, the study population was biased and taken from a tertiary headache centre taking care of severely affected and more chronic headache patients. Thus, study results may not be easily generalized to the headache population.