Multimodal behavioral treatment of migraine: an Internet-administered, randomized, controlled trial

Abstract

Introduction: Multimodal approaches in behavioral treatment have gained recent interest, with proven efficacy for migraine. The utility of the Internet has been demonstrated for behavioral treatment of headache disorders, but not specifically for migraine. The aim of the study was to develop and evaluate an Internet-based multimodal behavior treatment (MBT) program for migraine and to test hand massage treatment as an adjunct.

Methods: Eighty-three adults, 58 women and 25 men, with at least two migraine attacks a month were recruited via advertisements. An MBT program aiming at improvements in life-style and stress coping was developed for this study and, together with a diary, adapted for use over the Internet. Participants were randomized to MBT with and without hand massage and to a control group, and were followed for 11 months. Questionnaires addressing issues of quality of life (PQ23) and depressive symptoms (MADRS-S) were used.

Results: A 50%, or greater, reduction in migraine frequency was found in 40% and 42% of participants of the two groups receiving MBT (with and without hand massage, respectively), who statistically were significantly more improved than participants in the control group. No effect of hand massage was detected, and gender did not show any independent contribution to the effect in a multivariate analysis.

Conclusions: MBT administered over the Internet appears feasible and effective in the treatment of migraine, but no effect of hand massage was found. For increased knowledge on long-term effects and the modes of action of the present MBT program, further studies are needed.

Figure 1.

Diagram of the study design with group sizes and drop-outs (MBT = multimodal behavior treatment; Q = questionnaires; R = randomization; CD = compact disc).

Figure 2.

Change in migraine frequency between base-line and end period registrations. Box plot diagrams show median value, interquartile range, and full range of number of days with migraine during the first (base-line period) and the last (end period) 56 days of the study for the respective study groups. Outliers are represented by circles and personal codes.

Figure 3.

Relative change in migraine frequency between base-line and end period registrations. Individual changes, measured as the percentage decrease or increase of migraine frequency during end period registration compared to the base-line period, are shown in order of magnitude for the respective study groups. Horizontal lines define our main outcome measure: 50%, or more, decreased migraine frequency.

Figure 4.

Early evaluation of hand massage efficacy in alleviating migraine. The changes in number of days with migraine between the first and the last 4 weeks of an initial 3-month period of hand massage (study group A) versus no treatment (study group B) are shown separately by use of box plot diagrams showing median value, interquartile range, and full range of these changes. Mann–Whitney U-test was used for statistical comparison between the groups.

Figure 5.

Self-reported compliance of hand massage. Forty sessions were recommended as indicated by the line.