Description and Validation of an Italian ICHD-II-Based Questionnaire for Use in Epidemiological Research

Background. In the absence of biological markers, the diagnosis of primary headache in epidemiological studies rests on clinical findings, as reported through ad-hoc interviews. Objectives. The aim of this study was to validate a specially designed headache questionnaire to be administered by a physician for the diagnosis of primary headaches or of probable medication overuse headache in the general population according to the 2004 International Classification of Headache Disorders, 2nd edition (ICHD-II). Methods. The questionnaire comprises 76 questions based on the ICHD-II diagnostic criteria for migraine (codes 1.1, 1.2.1, 1.2.2, 1.2.3, 1.5.1, and 1.6), tension-type headache (codes 2.1, 2.2, 2.3, and 2.4), primary stabbing headache (code 4.1), and probable medication-overuse headache (code 8.2.7), as well as on other clinical features (eg, age at onset, relation between headache and pregnancy, etc). The answers to each question could be of the following types: (1) numbers (ie, age at onset); (2) Yes or No (eg, as in reply to Do you have nausea during headache?); and (3) predefined answers (eg, quality of pain). We assessed the validity and reliability of the questionnaire and its sensitivity and specificity for migraine and tension-type headache. Results. The study population consisted of 50 patients (37 women and 13 men) aged 18-76 years (mean, 40.7) seen for the first time on a consecutive basis at the University of Parma Headache Centre. The questionnaire was administered independently by 2 trained physicians (E1 and E2) prior to the visit performed by a headache specialist taken as the gold standard (GS). GS, E1, and E2 were blind to the diagnosis made by each others. If appropriate, more than 1 headache type were considered. When present, we defined the 2 different headache types in the same subject as Diagnosis 1 and Diagnosis 2. Questionnaire-based diagnosis was compared with the diagnosis established by GS. For Diagnosis 1 (n = 50), we found an agreement rate of 98% (K-value: 0.96; 95% confidence interval [CI]: 0.88-1.00) between E1 and GS and between E2 and GS, and of 96% (K-value: 0.91; 95% CI: 0.80-1.00) between E1 and E2. For Diagnosis 2 (n = 24), we found an agreement rate of 83.3% (K-value: 0.80; 95% CI: 0.63-0.98) between E1 and GS, of 62.5% (K-value: 0.62; 95% CI: 0.41-0.82) between E2 and GS, and of 70.8% (K-value: 0.66; 95% CI: 0.45-0.87) between E1 and E2. Sensitivity and specificity were 100% and 93.3%, respectively, for migraine without aura (code 1.1) and 100% for frequent episodic tension-type headache (code 2.2). Conclusion. Our findings support the use of this questionnaire as a valid and reliable tool for diagnosis of headaches in epidemiological studies.