Local communities are at the forefront when disaster happens. People cannot control natural hazards, however, they can work together to strengthen their capacities and identify ways to reduce vulnerability for improved disaster preparedness, response and recovery. Community Based Disaster Risk Reduction (CBDRR) aims to give a voice to local people so they can assess their own vulnerability and capacities and act upon reducing disaster risk. Non-government organisations (NGOs), in particular, have increasingly used CBDRR to provide a bottom-up approach that complements the resources and actions developed from the top down. One common way this has been achieved is through undertaking CBDRR activities with local disaster committees. CBDRR approaches have been implemented in Vanuatu for about 30 years. In 2008 the National Disaster Management Office (NDMO) began endorsing NGOs to facilitate the development of Community Disaster Committees, now known as Community Disaster and Climate Change Committees (CDCCC), to better prepare, respond and recover from disasters. However, while CBDRR is extensively used to tackle disaster risk reduction (DRR), there is still limited understanding of both the effectiveness and sustainability of this approach. NGOs’ involvement in CBDRR activities is generally restricted to the timeframes of projects, funding conditions and pre-defined goals from donor agencies. Monitoring and evaluation reports often focus on the impacts at community level immediately after NGOs have completed a project, but outcomes thereafter are not commonly understood nor even assessed. This thesis aimed to fill this knowledge gap by examining to what extent Vanuatu’s CDCCCs are sustainably contributing towards effective DRR. This research adds to the disaster literature, in particular CBDRR, and the implications this has on policy and practices. This study used participatory methods with CDCCC members on the one hand, and key informant interviews with local NGO and government staff working with CDCCCs on the other hand. The rationale for this methodological approach was to gather the views from Ni-Vanuatu people who have taken part in CDCCCs and to compare these with practitioners’ views who were in charge of their facilitation. The research focused on Sanma and Tafea provinces because CDCCCs have been developed in these locations since 2010 and 2012 respectively. Vanuatu also has experience with multiple large-scale disasters such as the 2015 Tropical Cyclone (TC) Pam, 2017 and 2018 evacuations of Ambae Island following volcanic eruptions, and 2020 TC Harold. A total of six Focus Group Discussions (FGDs) using participatory tools were conducted in Sanma and Tafea in April 2018, and five key informant interviews were conducted in Sanma from November 2018 to December 2018. The results indicate positive outcomes including community involvement in early warning systems and activity in disaster response, such as being involved in initial disaster impact assessments and distribution of relief items alongside government and NGOs. However, the research identifies several challenges associated with the use of CBDRR. Communities advised they were unlikely to engage in formal CDCCC preparedness activities unless the implementing agency/NGO was present. Findings highlight that without adequate external support, including financial resources, CDCCCs are unlikely to function over time. There is also indication that rigid/one-size-fits-all implementation of CBDRR tool kits associated with the development of CDCCCs hinders both effectiveness and sustainability of such an approach. This includes the mandating of disaster committee roles, the prescribed forced representation of groups including women and people with disability, and the language used in the approach. There are indeed indications that the CDCCC framework could be modified to improve effectiveness including allowing community members to choose their own roles and for the CBDRR approach to build on existing socio-cultural networks such as churches and ni-Vanuatu chief and kastam systems. This research concludes that CBDRR requires a careful balance of participation, accountability and integration to be both effective and sustainable over time. Communities should have the opportunity to consider trade-offs between development goals and decide what they would like to prioritise. The guidelines and frameworks that guide CBDRR must remain flexible, so that communities take ownership over funded programmes implemented by external agencies. Disaster committee structures must align with government level arrangements and have the flexibility to incorporate local and governmental leadership systems. Finally, CBDRR activities must fit with the resources available at the local community level, and that of the local and national government. It is hoped that practitioners, researchers, donors and community members continue to consider the impact and sustainability of CBDRR activities. It is urged that all practitioners, researchers, donors and community members reflect on the opportunities, challenges and trade-offs associated with CBDRR processes and outcomes so that direction of interventions fits the communities that they serve.