The occurrence of recall bias in pediatric headache: a comparison of questionnaire and diary data.

OBJECTIVE: To investigate whether children and adolescents can recall prior headache complaints accurately and to study whether age, gender, headache severity, preferred coping strategies, depression, somatization, and trait anxiety are related to recall errors, causing recall bias. METHODS: A retrospective headache questionnaire and a prospective 4-week headache diary were filled out by 181 children aged 9 to 16 years who experienced headache at least weekly. In addition, several other questionnaires were administered, measuring coping strategies, depression, somatization, and trait anxiety. Headache frequency, intensity, and duration, as scored on the questionnaire and the diary, were compared using Wilcoxon tests. Regression analyses were performed to study whether age, gender, headache severity, preferred coping strategies, depression, somatization, and trait anxiety can predict the size of differences between the diary and the questionnaire. RESULTS: Compared with the diary, headache intensity and headache duration were overestimated on the questionnaire. At group level, median headache frequency as measured by the diary and the questionnaire was equal. Regarding headache frequency and headache intensity, age and headache severity were statistically related to errors in recall. For headache frequency, depression was also predictive of the size of recall error. CONCLUSIONS: Recall errors occur when children are asked to report their headaches on a retrospective questionnaire. As compared to a prospective diary, pain complaints are evaluated more negatively on a questionnaire. Other factors such as age, depression, and headache severity influence the way children and adolescents recall their headaches. To minimize bias, the use of a diary when studying recurrent headache complaints in children is recommended.