Background: Many women experience fear of childbirth (FOC). While fears about childbirth may be normal during pregnancy, some women experience high to severe FOC. At the extreme end of the fear spectrum is tocophobia, which is considered a specific condition that may cause distress, affect well‐being during pregnancy and impede the transition to parenthood. Various interventions have been trialled, which support women to reduce and manage high to severe FOC, including tocophobia. Objectives: To investigate the effectiveness of non‐pharmacological interventions for reducing fear of childbirth (FOC) compared with standard maternity care in pregnant women with high to severe FOC, including tocophobia. Search methods: In July 2020, we searched Cochrane Pregnancy and Childbirth’s Trials Register, ClinicalTrials.gov, the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), and reference lists of retrieved studies. We contacted researchers of trials which were registered and appeared to be ongoing. Selection criteria: We included randomised clinical trials which recruited pregnant women with high or severe FOC (as defined by the individual trial), for treatment intended to reduce FOC. Two review authors independently screened and selected titles and abstracts for inclusion. We excluded quasi‐randomised and cross‐over trials.Data collection and analysis: We used standard methodological approaches as recommended by Cochrane. Two review authors independently extracted data and assessed the studies for risk of bias. A third review author checked the data analysis for accuracy. We used GRADE to assess the certainty of the evidence. The primary outcome was a reduction in FOC. Secondary outcomes were caesarean section, depression, birth preference for caesarean section or spontaneous vaginal delivery, and epidural use. Main results: We included seven trials with a total of 1357 participants. The interventions included psychoeducation, cognitive behavioural therapy, group discussion, peer education and art therapy. We judged four studies as high or unclear risk of bias in terms of allocation concealment; we judged three studies as high risk in terms of incomplete outcome data; and in all studies, there was a high risk of bias due to lack of blinding. We downgraded the certainty of the evidence due to concerns about risk of bias, imprecision and inconsistency. None of the studies reported data about women's anxiety. Participating in non‐pharmacological interventions may reduce levels of fear of childbirth, as measured by the Wijma Delivery Expectancy Questionnaire (W‐DEQ), but the reduction may not be clinically meaningful (mean difference (MD) ‐7.08, 95% confidence interval (CI) ‐12.19 to ‐1.97; 7 studies, 828 women; low‐certainty evidence). The W‐DEQ tool is scored from 0 to 165 (higher score = greater fear). Non‐pharmacological interventions probably reduce the number of women having a caesarean section (RR 0.70, 95% CI 0.55 to 0.89; 5 studies, 557 women; moderate‐certainty evidence). There may be little to no difference between non‐pharmacological interventions and usual care in depression scores measured with the Edinburgh Postnatal Depression Scale (EPDS) (MD 0.09, 95% CI ‐1.23 to 1.40; 2 studies, 399 women; low‐certainty evidence). The EPDS tool is scored from 0 to 30 (higher score = greater depression). Non‐pharmacological interventions probably lead to fewer women preferring a caesarean section (RR 0.37, 95% CI 0.15 to 0.89; 3 studies, 276 women; moderate‐certainty evidence). Non‐pharmacological interventions may increase epidural use compared with usual care, but the 95% CI includes the possibility of a slight reduction in epidural use (RR 1.21, 95% CI 0.98 to 1.48; 2 studies, 380 women; low‐certainty evidence). Authors' conclusions: The effect of non‐pharmacological interventions for women with high to severe fear of childbirth in terms of reducing fear is uncertain. Fear of childbirth, as measured by W‐DEQ, may be reduced but it is not certain if this represents a meaningful clinical reduction of fear. There may be little or no difference in depression, but there may be a reduction in caesarean section delivery. Future trials should recruit adequate numbers of women and measure birth satisfaction and anxiety.