BACKGROUND: 20-90% of all women suffer from dysmenorrhea. Standard therapy of primary dysmenorrhea (PD) are NSAIDs and oral contraceptives, effective but not without possible side effects. OBJECTIVE: To examine the efficacy of rhythmical massage (Anthroposophic Medicine) and heart rate variability biofeedback compared to usual care (control group) on pain intensity in women with primary dysmenorrhea. METHODS: This was a three-arm randomized controlled study. Both interventions (rhythmical massage once a week or HRV biofeedback 15 min daily) were carried out over a period of three months. The third group (control) applied usual care. The primary outcome were between-group differences in mean pain intensity (detected by a Numeric Rating Scale, NRS) during menstruation after three months (post-assessment, t2). Secondary outcomes were the use of analgesics, quality of life (SF-12) and heart rate variability. RESULTS: The study involved 60 women, mean age 29.7 years, SD 8.0 (n = 23 rhythmical massage, n = 20 biofeedback, n = 17 control). For the primary outcome there was a significant difference between the groups after three months (p = .005). Bonferroni adjusted post-hoc tests revealed a significant difference between rhythmical massage and control group (mean difference: -1.61; 95 CI: -2.77/-0.44; p = .004; ES: -0.80). No significant differences were found between rhythmical massage and biofeedback (mean difference: -0.71; 95 CI: -1.82/ 0.40; p = .361; ES: -0.34) and between biofeedback and control group (mean difference: -0.90; 95 CI: -2.10/-0.30; p = .211; ES: -0.51). For the secondary outcomes no significant differences were found between the groups at t2. The drop-out rate was higher in the biofeedback group (n = 6) than in the massage (n = 2) or the control group (n = 4). CONCLUSION: Preliminary evidence suggests that rhythmical massage might improve pain intensity after 12 weeks compared to usual care.