Ensuring that harm does not take place in the work conducted in other countries by the humanitarian and development sector, is vital. However, there still seems to be a lack of research looking into specific conflict-sensitive approaches. The Do No Harm (DNH) framework was developed by Mary B. Anderson in 1999 and in regards to conflict sensitivity tools, the DNH framework is one of the most well-known. Still, there is only a small amount of research that has looked into how the DNH framework has been implemented and the impact that the framework has had in the humanitarian and development sector.  This thesis aims to examine how the DNH framework has been implemented and used by the Swedish development sector. The research is conducted through a qualitative case study of the Swedish development sector and the findings are analyzed through the lens of different policy change theories.  The thesis finds that the Swedish development sector early on was aware of DNH and that the framework quite rapidly was adopted into the sector as a policy instrument. However, it also concludes that the DNH framework has not been fully implemented in the policy objectives or on the ground, which led to the addition of a research question that investigated why the DNH framework had not been thoroughly implemented. Upon looking into the reasons for this lack of implementation, the study finds that the theories of Punctuated Equilibrium and Path Dependency can provide some explanations for this. Further, it can also be concluded that there are improvements to be made in terms of applying the framework on the ground and considering the conflict-sensitive perspective in relation to development work, not only in the humanitarian field.