This review drew largely on academic papers as well as reports by international development organisations. Evidence and hence lessons on how to combat forced marriage are limited and sometimes contradictory. Overall, the literature points to a number of approaches that can be effective, notably: empowerment of girls; community approaches to change social norms and attitudes on child marriage, and economic incentives (for girls and families); and, alternative opportunities (notably education, and income generation). Legislative approaches appear to be the least effective in combating child, early and forced marriage. However, different approaches need to be implemented together in order to bring about sustained change. Child and early marriages are viewed as forced marriages because child spouses are considered incapable of giving full consent. The literature predominantly uses the term child marriage or child, early and forced marriage (CEFM). This review found negligible literature using just the term ‘forced marriage’, and hence it too uses the term CEFM. While both males and females are affected by CEFM, the literature refers almost exclusively to combating CEFM in females. CEFM occurs in many countries, but is most prevalent in South Asia and Sub-Saharan Africa, and disproportionately affects females. CEFM is a human rights violation and an impediment to sustainable global development. It is fuelled by diverse social and economic factors, the most common being poverty, gender inequality and conservative social norms. The findings from the multi-programme reviews and individual programme evaluations do not clearly identify one/some approaches as being particularly effective. In some contexts, interventions to promote empowerment had a positive impact on delaying marriage, but in others, they did not. Similarly, provision of economic incentives to families led to significant reductions in child marriage in some programmes, but had no impact in others. This highlights the need for context-specific interventions, and for multiple component programmes, which combine different approaches. Just as CEFM is driven by a range of factors, so programmes to tackle it must have components that address all the diverse factors. The findings also suggest that interventions can sometimes have unexpected, negative consequences – delaying marriage age in Bangladesh, for example, led to families having to pay higher dowries. Again, this points to the need for context-specific programme design, taking into account all possible effects. Overall, the literature shows a clear dearth of rigorous evaluations of programmes to combat CEFM: much more effort and resources need to be invested in generating evidence to inform future policies and programmes. This review found two reports on child marriage among Roma communities in Montenegro and Ukraine respectively. They highlight the high prevalence rates among the Roma, the fact that this is driven by several factors – poverty, gender inequality, patriarchal attitudes, marginalisation – and the need for a multi-pronged approach combining empowerment, sensitisation, economic incentives, schooling and legislative enforcement.