Association of pre- and post-treatment expectations with improvements after acupuncture in patients with migraine

Objective To study whether a higher expectation of acupuncture measured at baseline and after acupuncture is associated with better outcome improvements in patients with migraine. Methods We performed a secondary analysis of a previous published trial in which 476 patients with migraine were randomly allocated to three real acupuncture groups and one sham acupuncture control group. All the participants received 20 sessions of acupuncture over a 4-week period. The primary outcome was the number of days with a migraine attack (NDMA) assessed at 5–8 weeks after randomisation. The secondary outcomes were visual analogue scale, headache intensity and quality of life assessed at 4, 8 and 16 weeks after randomisation. Expectations of the acupuncture effect were assessed at baseline and at the end of treatment and categorised into five levels, with 0% the lowest and 100% the highest. Outcome improvement was first compared among the participants with different expectation levels using an analysis of variance model. The association between expectations of treatment and outcome improvement was then calculated using a logistic regression model. Results Patients with 100% baseline expectations did not report significantly fewer NDMA than those with 0% baseline expectations after adjusting for the covariates (at 5–8 weeks, 1.7 vs 3.9 days, p=0.987). High baseline expectations had no significant impact on improvement of the primary outcome (100% vs 0%: OR 8.50, 95% CI 0.89 to 191.65, p=0.682). However, patients with 100% post-treatment expectations reported fewer NDMA than those with 0% expectations (primary outcome: 1.3 vs 5.0 days, p Conclusions A high level of expectation after acupuncture treatment rather than at baseline was associated with better long-term outcome improvements in patients with migraine. Clinical trial number NCT00599586.