gms | German Medical Science

In an epidemiological study we examined paediatric headache in children and adolescents, using a representative regional sample of the German population. The study aims at assessing headache prevalence, analysing the distribution of headache frequency and identifying risk factors. Parental and child questionnaires were developed and administered annually over a three year period (2003-2005). The questionnaires were sent to a randomly drawn sample of 8800 households with a child aged 7 to 14, in southern Lower Saxony and the city of Hannover. In the first year 5586 (63.5%) households responded and were asked to participate in the following two years, which 4164 (74.4%) and 3714 (66.4%) households, respectively, did.

The present results are based on the parental questionnaires from the first three years. Of the 3245 children, for which questionnaires were returned in each wave, 1833 (56.5%) showed stability in their headache report. These children either reported being headache-free (48%) or having headaches ('headache stability'; 52%) at each of the three assessment points. The Binary Logistic Regression procedure was used to find out whether headache stability could be predicted by headache type ('tension-type headache', 'migraine with or without aura' and 'unclassifiable headache'), headache frequency ('less than monthly', 'monthly' or 'weekly'), headache onset ('childhood' or 'adolescence'), age group ('child' or 'adolescent') and gender ('male' or 'female').

Multivariate analysis, using the above variables, revealed that stability of experiencing headache over a period of three years was more likely among females (OR=1.41; CI=1.1-1.8), among children and adolescents reporting monthly (OR=2.44; CI=1.8-3.3) or weekly (OR=3.20; CI=2.0-5.0) headaches, and among those with the diagnosis of migraine (OR=1.52; CI=1.0-2.3). However, the model explains a meagre 10% of the variance (Nagelkerke's R² = .098).