In Europe, the mental health field is undergoing a paradigm shift from a hospital‐centered, institutionalized and segmented model toward a community‐based, patient‐centered and more integrated model. From 2010 onwards, Belgian policymakers availed themselves of new policy instruments to complete this shift, having been hampered by strong professional and cultural barriers over the four previous decades. However, the reform objectives have only partially been achieved. Assuming that an instrument perspective on policy implementation would illustrate why the reform does not achieve its priority objectives, the article questions the relationship between the types of instruments used and the type of change induced. Drawing on the analysis of three policy implementation processes, we argue that these “soft” instruments are by nature not suitable for initiating any type of change and may have limited effects when used in certain contexts and under certain conditions. The article ends with a discussion of the three limitations of these instruments.

Coordination is described as a widespread function emerging in relation to policy plans inducing collaboration between different sectors, organizations and professions. This paper suggests seeing the implementation phase as a translation process, one where the content of policy plans is reinvented primarily through discussion rather than linearly transferred from the political to the professional arena. It focuses on the function of coordinator with a view to examining how this function is performed and questions its influence on the local translation of both policy plans. The data collection was part of two research projects focusing on the reform of Belgian mental healthcare and the creation of care pathways for forensic patients, combining document analysis, interviews (n = 82) and observations (n = 58). The results highlight the inherent ambiguity of the coordinators’ working environment, the socially-disputed nature of their function and define the coordinators as connection-makers who exert power over processes rather than people or structures. It demonstrates that coordinators influence the policy process by inducing discussions at meetings and the documents subsequently produced. In conclusion, this paper defines coordinators as process managers whose work largely consists of translating policy plans through event connectivity and contextualizing practices. Given the importance of translation in policy implementation, this paper calls for a reconsideration of policy evaluation as well as of the coordinators’ recruitment and training procedures.