Background. The goal of more integrated care has been on the Belgium health care policy agenda since 2007 with the start of disease specific care trajectories, reports on chronic care and pilot projects at local level. While there is consensus on what needs to be done, the process of how it should be done is much more uncertain. Scale-up of integrated care is complex, because it requires multi-stakeholder action and intersectoral coordination at individual healthcare practice, organizational, and political/system levels, and tailoring to the context in which it is delivered. There is little research on how to scale-up such complex and adaptive interventions.
 The SCUBY project aimed to contribute to scale-up of integrated care through development of evidence-based roadmaps and policy dialogues. This evaluation study examines the SCUBY scale-up intervention, by looking at its implementation outcomes, mechanisms of impact and the interaction with the political, policy and societal circumstances in the Belgium context.
 Methods. The ‘process evaluation for complex interventions’ by the MRC was the evaluation framework. Policy dialogues were defined as the SCUBY-team engagements with key stakeholders to develop and implement roadmaps. They comprised structured formal events, one-to-one interactions with key stakeholders, workshops, consultations, and joining ongoing dialogues of stakeholder networks. The roadmap was a scale-up strategy and included activities, a timeline and a description of the evidence base and key partners/stakeholders involved (change team). Data collection tools included policy dialogue reporting forms and surveys, key informant interviews, policy document review and a project diary. Descriptive and thematic analyses are conducted, together with a policy mapping. 
 Results. The SCUBY team conducted 79 meetings classified as policy dialogues, most of them being small group meetings or network meetings organized by partners. SCUBY organized 6 major policy dialogues and invited a mix of beneficiaries, health system administration representatives, policy makers (from federal and regional level), health care providers and their representative organizations, people involved in organization and coordination of care, care users, and scientific stakeholders. Three themes emerged as being the most important areas for the roadmap: A) primary care practice organization (change management and advocating for primary care nurses); B) data and monitoring (chronic care indicators at primary care practice level and for population health management in Primary Care Zones); and C) healthcare financing (budget for chronic care that stimulates quality and new provider payment models). The roadmap comprised subthemes, with internal objectives (to be reached with our project resources) and external objectives (beyond out sphere of control) and activities. 
 Discussion. Our findings emphasize the importance of adopting a collaborative, networking approach via the organization of various small-size and bigger-size policy dialogues, in order to create real-world impact, by building synergies and coalitions and our researcher team generating boundary spanning skills, as more proximal outcomes. The variation in progress made per roadmap theme was anticipated because of our level of involvement, expertise and resources available. The change team had to find the right balance in the trade-off between relevance and feasibility of certain scale-up strategies.