Applying system support mapping combined with the matrixed multiple case study approach to identify and analyse implementation determinants: lessons from a multi-site implementation study in German neonatal intensive care units

Abstract Background Identifying and understanding implementation determinants is a critical step in the process of tailoring implementation strategies to local contexts. System Support Mapping (SSM) is a method rooted in systems thinking that offers a structured approach to exploring the responsibilities, needs, resources, and wishes of individuals within an implementation context. However, SSM has not yet been systematically described and applied as a methodology for identifying implementation determinants within multi-site implementation studies. This article introduces and illustrates the combined application of SSM and the Matrixed Multiple Case Study (MMCS) approach as a structured six-step methodology for identifying and analysing implementation determinants within and across sites, using data from the Neo-MILK project as an illustrative example. The project aimed to implement a structured lactation support programme for mothers of very low birth weight infants and establish human donor milk banks in German neonatal intensive care units (NICUs). Methods To illustrate the six-step methodology, semi-structured interviews with key persons from five German NICUs, which were conducted during the early intervention phase, served as the empirical basis for demonstrating each methodological step. Transcripts were analysed using directed content analysis, with SSM elements serving as deductive categories and context-specific codes developed inductively. The MMCS approach was used to organise, compare, and synthesise findings within and across sites. Results Applying the six-step methodology, we demonstrate the type of output each step produces, ranging from individual System Support Maps per key person and NICU (steps 1–3) to a sortable cross-site matrix (steps 4–6). The within-site analysis (step 5) illustrates how the methodology surfaces context-dependent perceptions of the same determinant, showing that a given resource category could be perceived as facilitating, inhibiting, or both, depending on individual responsibilities, needs, and local context. The cross-site analysis (step 6) demonstrates how the matrix structure makes homogeneous and heterogeneous determinant patterns visible across sites, indicating different contextual manifestations of the same determinant. Conclusion The combined application of SSM and the MMCS approach provides a structured methodology for identifying and analysing implementation determinants that accounts for the perspectives of individual interest-holders and enables systematic within-site and cross-site comparison. The distinction between homogeneous and heterogeneous determinant patterns offers a basis for differentiating between cross-site and site-specific implementation strategies. Further research is needed to evaluate the use of SSM across multiple evaluation timepoints and its integration into initial and ongoing tailoring of implementation strategies. Trial registration DRKS00025058; registered 6 May 2021.