Background. Little is known which factors influence the blinding in acupuncture studies. Aim. To investigate if blinding varied between patients with different characteristics receiving verum or sham acupuncture. Methods. We randomised cancer patients to verum (n = 109) or sham acupuncture (n = 106) with a nonpenetrating telescopic sham needle for nausea. Level of blinding was compared between different sub-groups of patients using Bang's blinding index (BI) ranged -1 to 1 (-1 = all state the opposite treatment, 1 = all identify treatment). Results. Most patients in the verum (74 of 95; 78%, BI 0.72) and the sham (68 of 95; 72%, BI -0.60). acupuncture group believed they had received verum acupuncture. The probability for a patient to believe he/she received verum acupuncture was related to the received needling type (P = .003) and to the patient's belief in received treatment effects (P = .008). Hospital (P = .425), therapist (P = .434), previous acupuncture experience (P = .578), occurrence of nausea (P = .157), gender (P = .760), and age (P = .357) did not affect blinding. Conclusions. Blinding was successfully achieved irrespective of age, gender, acupuncture experience, treatment effect, or in which hospital or by which therapist the patient received treatment. Patients with higher belief in the effect of the treatment were more likely to believe they had received verum acupuncture.