Key functions as conceptualized within the FIC model: Method and terminology

The FIC model provides a framework to deconstruct interventions in order to determine which core functions can be transferred across contexts to support the intervention’s logic across contexts – key functions – and which activities – forms – must be adapted [15–17]. This approach enables a deeper understanding of interventions for transfer to new contexts. As conceptualized within the FIC model, the present analysis focused on identifying the following components:

• Key functions are the essential mechanisms and steps through which an intervention achieves its objectives, and they may be consistent across different implementation contexts [16]. A key function represents a potential component of the overall intervention strategy. Distinguishing between key functions and forms challenges the traditional notion of intervention “fidelity,” based on the assumption that an intervention can remain faithful to its key functions even when the specific activities or forms differ across contexts [17]. Key functions can influence health determinants at multiple levels, including micro-, meso-, and macro-social dimensions [18];
• Forms represent the translation of key functions into concrete activities [16]. Theoretical forms refer to activities designed by project leaders for implementation, whereas observed forms emerge from the actual adaptation and evolution of these activities in different (school and municipal) contexts. Unlike key functions, forms may differ depending on the context.
• Contextual factors are not passive “backdrops”; they actively shape the underlying processes that determine how and why an intervention is effective. Because contexts are complex and unique [19], it is essential to consider them alongside the forms of the intervention.

This study focused on identifying the key functions and forms of the Alliance intervention to inform and support its transferability for future implementations. It did not involve applying the FIC model in its entirety. As such, social inequalities in health were not explicitly considered in the present study, although they are an important concern within PHIR and constitute an important dimension of the FIC model [20].

Participants

Qualitative data were collected through semi-structured interviews conducted with professionals from schools and municipalities involved in the Alliance intervention during phase 1 (2019–2022) or phase 2 (2022–2023). A wide range of stakeholders were interviewed, including intervention recipients (teachers and after-school care staff), and institutional actors involved in implementation, such as representatives from the French National Education system (health advisors at the rectorate and departmental levels, district pedagogical advisors), and municipal authorities (elected officials, employees responsible for school affairs).

All schools and municipalities that participated in phases 1 or 2 were contacted for recruitment (from 20 March to 15 April 2022 for phase 1, from 15 March to 15 June 2023 for phase 2). In schools, interviews were primarily conducted with school directors rather than with teachers, as directors were the main contacts for the project and held the school’s generic contact email address. Regarding representatives of the French National Education system, all health advisors who had been designated as Alliance project referents (one per department and one per academy) were invited to participate. The total number of interviews was predefined based on the number of schools and municipalities that had implemented the Alliance training, with the aim of achieving a balanced distribution across territories.

Participants’ gender, profession and area of practice are summarized in Table 1. In phase 1, 31 professionals were interviewed: 18 from the school sector (among 63 eligible professionals) and 13 from the municipal sector. In the school sector, three-quarters (6/8) of school directors were women, while the majority of district pedagogical advisors were men (7/10). In the municipal sector, the vast majority of interviews were conducted with local elected officials (9 out of 13), most of whom were women (6 out of 9). In phase 2, 30 professionals were interviewed (among 93 eligible professionals): 25 from the school sector and 5 from the municipal sector, including 20 women. In the school sector, most of school directors and district pedagogical advisors were women (8/9 and 5/9, respectively). The most senior positions – corresponding to inspectors and health advisors – were mainly held by women (5/7). In the municipal sector, the interviews were conducted with local elected officials (n = 5), most of whom were women (4 out of 5). In both phases, professionals from all four intervention departments were represented, across both the school and municipal sectors.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of interview participants by phase and overall.

https://doi.org/10.1371/journal.pone.0348403.t001

Data collection

Semi-structured interviews were selected as the most appropriate method to address specific questions aligned with the study objectives, while letting participants the flexibility to share their experiences with the intervention.

Data were collected during two separate periods. The first period involved participants engaged in the intervention during phase 1. Interviews were conducted in 2022 by two investigators (AG and LV, both female) as part of their master’s degree theses in Public health and Health promotion and education, respectively, under the supervision of MG, an expert in qualitative research. Participants had no prior relationship with the interviewers. The second period focused on participants involved in phase 2. Interviews were conducted from April to June 2023 by IS (Master’s degree, female, project coordinator). Some participants, primarily representatives of the French National Education system, were already acquainted with the interviewer and familiar with the project’s goals and methods.

The semi-structured interviews were conducted by telephone or videoconference, enabling participation from stakeholders across all four departments where the intervention was implemented. All interviews were audio recorded and transcribed verbatim. Field notes were also taken. Participants did not provide feedback on the transcripts or final analysis.

The interview guide (S1 File) was initially developed to identify barriers and levers to implementation. Its construction was informed by the literature on school-based health promotion and by the RE-AIM framework [21] to ensure comprehensive coverage of implementation-related dimensions. The qualitative material collected on the intervention, its implementation, and contextual factors, provided sufficient depth to support a subsequent a posteriori secondary analysis to identify the key functions and forms and contextual factors (as conceptualized within the FIC model), although they were not originally designed for this purpose. The guide was adapted for each stakeholder group.

Data analysis

The FIC model analysis relies on joint reflection among various stakeholders [15,16]. First, a thematic content analysis was carried out using QDA Miner Lite and ATLAS.ti software to identify barriers and facilitators to implementation [22]. Subsequently, these qualitative data were analyzed to characterize key functions, forms, and contextual elements as conceptualized within the FIC model. In line with established methods [16,17,23], the analysis was informed not only by qualitative data but also by internal project documentation (e.g., training materials, steering committee reports, and operational notes) and feedback from health promotion consultants (members of the research team), who were responsible for operational coordination, field implementation in schools and municipalities, and direct contact with recipients. These sources supported the triangulation of interview data by providing complementary insights into the theoretical forms of the intervention and clarifying certain organizational and implementation processes. They also contributed to ensuring consistency with field realities and offered additional perspective on the intervention’s evolution over time.

The analysis was carried out collaboratively with members of the research team that developed the FIC method (LF, CD), who provided methodological expertise to identify the key functions and forms, and project leaders (IS, VR), who had in-depth knowledge of the intervention. Six working meetings were held between January and March 2024, during which an iterative process of description and validation was followed until consensus was reached. The analysis was further refined through discussions with the health promotion consultants. This process ensured that the description remained faithful to field realities and employed appropriate terminology. Consequently, the findings were the result of a description process based on the integration of multidisciplinary academic, professional, and experiential knowledge.

Data from phase 1 interviews, internal documentation, and consultants’ feedback were used to generate the “theoretical forms” as conceptualized within the FIC model, whereas data from phase 2 interviews were used to produce “observed forms”. The “theoretical forms” of phase 1 are therefore distinct from the “observed forms” of phase 2. Phase 2 data did not lead to the emergence of new key functions; rather, they allowed us to confirm their relevance across extended implementation settings and to document how their forms evolved in practice. Only KF1 was modified between phases, shifting from “Multi-institutional and political support” to “Institutional support from the National Education system,” as territorial political support declined over time. Comparing theoretical forms with observed forms made it possible to identify and document concrete evolutions in the Alliance intervention across implementation phases.

Ethical considerations

This study received approval from the French Inserm Ethics Evaluation Committee (CEEI / IRB00003888) under reference numbers 19–600 (July 8, 2019) for phase 1 and 23–981, (March 7, 2023) for phase 2 and was conducted in accordance with the principles of the Declaration of Helsinki. Prior to the interviews, all participants were provided with a written description of the study and were informed of their right to withdraw at any time. Oral informed consent was obtained at the start of each interview for phases 1 and 2, after the audio recording had started; participants verbally confirmed their agreement to participate and to be recorded. Consent was thus documented and witnessed through the recording. This procedure was approved by the IRB. To ensure confidentiality, participants’ identities and interview content were fully anonymized.

Key functions of the Alliance intervention in relation to existing school-based health promotion program

The identification of key functions, as conceptualized within the FIC model, is not intended to determine the components of intervention effectiveness, but rather to distinguish transferable theoretical processes that ensure fidelity to the intervention from those aspects of implementation that can be adapted across contexts [14]. However, some key functions – particularly interprofessional training (KF2) and the provision of methodological support and resources (KF4) – are theoretically more related to intervention effectiveness, as they correspond to correspond to key levers identified in the literature as influencing professional practices and the implementation of health promotion activities. The Centers for Disease Control and Prevention’s systematic review of school-based health education curricula identifies teacher training, along with enhancing teachers’ interest and sense of competence, strengthening alignment with the teaching curriculum, and support for the implementation of evidence-based pedagogical strategies in health education, as key factors associated with effectiveness [24]. Similarly, a meta-analysis of school-based interventions for the prevention of overweight and obesity found that programs were more likely to be effective when teachers had received preparatory training [25]. Although no causal inference can be drawn, the encouraging results observed in phase 1 at both the professional practice and child levels are consistent with the expected role of KF2 and KF4 in supporting changes in professional practices.

KF2 and KF4 are classic key functions in health promotion programs. For example, in the “Lycéen Bouge” program, a similar key function, “Support and training of school advisors” was explicitly identified [26]. In the literature, the empowerment of educational teams is sometimes described as a key function, and in other cases as a form [26,27]. In the present analysis, the FIC model approach provides a clear distinction between what belongs to the key function (essential processes structuring the intervention) and what belongs to the form (context-specific ways in which these processes are operationalized). Distinguishing functions from forms makes it possible to preserve the underlying logic of the intervention while allowing contextual adaptations, which is central when considering transferability across settings.

Key functions related to institutional support, coordination, internal organization create the structural and governance conditions necessary for the intervention to be delivered. Similar functions have been documented in the literature on psychosocial skills development programs in primary and secondary schools. For example, the Affective and Social Development Program (PRODAS) emphasized the importance of “harmonization, sharing of practices, and multiple team discussion opportunities” [27]. This function echoes both the KF3 (knowledge of and respect for institutional functioning) and KF1 (institutional support) of the Alliance intervention. Of note, the French educational system is centralized, with a strong role of national and regional authorities. In this context, institutional support at different administrative levels and the involvement of rectorates and inspectors played a major role in facilitating implementation. These elements may not be directly transferable to educational systems with different governance structures or levels of decentralization. International frameworks such as the WHO Health Promoting Schools standards emphasize that the implementation of school-based health interventions depends on national governance arrangements and institutional contexts and therefore requires adaptation across countries [28].

Furthermore, functions related to coordination and integration (KF5–KF7) appeared particularly important in the context of intervention research. Maintaining a strong link between the research team, institutional partners, and the field actors is essential to 1) ensure that the intervention is adapted to its implementation context, and 2) prevent research activities from being perceived as an additional burden, a key issue in intervention research. Notably, three of the seven key functions relate specifically to coordination and integration with research activities. While structured coordination has been maintained by the research team during the experimental phases of the project, future scale-up may require these responsibilities to be delegated to other stakeholders. Given the central role of coordination in sustaining interventions during scale-up [29], identifying and defining potential new forms for these three key functions in future implementations remains a key challenge.

Deconstructing the intervention into its key functions also illustrates how the Alliance intervention operates across different social levels, from micro (individual) to macro levels, to influence health determinants [18]. At the macro level, KF1 secures the recognition of health promotion as a core mission of the National Education system. At the meso level, KF2 and KF4 support the development of a shared culture and the establishment of collective working dynamics in health promotion, while KF5–7 strengthen partnerships and promote knowledge sharing among stakeholders, and foster a professional climate of trust. At the micro (individual) level, KF2 and KF4 enhance individual teachers’ and after-school care staff’s legitimacy to act in health promotion, as does KF3, for instance by ensuring that health promotion actions are tailored to local needs, contexts, and professionals’ preferences.

Finally, the identification of contextual elements influencing implementation provides important guidance for future implementation, as it enables stakeholders to anticipate the institutional and organizational requirements necessary for preserving key functions. For example, securing training time within statutory hours and ensuring adequate teacher replacement capacity emerged as structural prerequisites for maintaining KF2. Similarly, recognizing barriers such as administrative fragmentation or staff turnover allows stakeholders to anticipate challenges.

Contributions of the FIC model’s key functions to study transferability

While school-based health promotion programs are increasing in number, studies in France mainly focus on their effects at the teacher level (e.g., health education practices) and pupil level (e.g., psychosocial skills), including within the present research program and other initiatives such as Explo’Santé [30]. However, they rarely address implementation processes, the conditions required for successful implementation, or the identification of interventions’ key functions; and their transferability is not often documented. One of the main contributions of the key functions as conceptualized within the FIC approach is to make explicit dimensions that remain implicit in the daily work of project leaders and are not systematically formalized or highlighted in official documents or communications. Identifying the key functions of the Alliance intervention may help future implementers, together with local actors, to 1) be aware of key functions that should be preserved during implementation, 2) anticipate which contextual elements may influence their preservation, and 3) to select appropriate forms adapted to local conditions, according to the specificities of and the knowledge they have‌‌ about the features of their context, thereby supporting transferability.

The contribution of the FIC model’s key functions can be better understood in relation to other frameworks commonly used in implementation and transferability research. Unlike tools that rely on prescriptive checklists to evaluate transferability, such as the ASTAIRE tool [10], the FIC model adopts a different approach, offering a reflexive framework to disentangle the key functions of an intervention (i.e., the theoretical processes that must be preserved to maintain fidelity) from the adaptable forms that can and should be tailored to specific contexts. The model assumes that intervention effectiveness is not determined by key functions alone (in this sense, they differ from what is typically considered the “active ingredients” of an intervention, i.e., the observable, replicable, and irreducible aspects of the intervention [31]) but emerges from their dynamic interaction with contextual factors during implementation. Moreover, implementation science frameworks [32–34] are widely used to analyze determinants of implementation, the relationships between implementation strategies and outcomes, and conditions for successful delivery. These frameworks help explain why implementation succeeds or fails within a given context and allow anticipation of barriers and facilitators in future implementations. However, they do not explicitly address how an intervention can be transferred across contexts. In this perspective, the FIC analytical framework provides a complementary contribution by focusing on the internal structure of the intervention and clarifying which processes must remain stable across contexts and which may vary depending on local conditions.

Limitations

This study has several limitations. First, the analysis is partly based on retrospective phase 1 data, which were originally collected to explore barriers and levers to project implementation rather than to identify key functions. This limitation is however mitigated by the inclusion of phase 2 interviews, which were specifically conducted to address this objective. Second, although the sample included a wide range of professional roles, some stakeholder groups were underrepresented. While the FIC model traditionally primarily relies on insights from implementers and researchers [14,15,23], it is important to acknowledge that the perspectives of those directly targeted by the intervention were not fully considered in this study. The absence of pupils (due to the complexity of ethical approval procedures for involving minors) and underrepresentation of certain frontline professionals (teachers other than school directors, municipal after-school care staff who were difficult to contact) limits the understanding of acceptability, feasibility, and perceived relevance of the intervention at the operational level. Third, women were overrepresented among participants. However, this reflects the actual gender distribution in education and child-focused health professions (e.g., 85% of primary school teachers, over 80% of primary school directors and teachers, 71% of pediatricians [35,36]). More broadly, sociological evidence indicates that women are more engaged than men in health, health education (both for themselves and others), and care-related activities, especially those involving children [37,38].

Regarding the FIC approach itself, challenges include the granularity of key functions and the criteria for distinguishing between key functions and forms, which are beyond the appraisal and negotiation of stakeholders. Field actors, who are best positioned to judge which intervention elements can be adapted based on their experience and knowledge of local contexts [39], were actively involved, through interviews and participation of health promotion consultants in the co-construction and revision of the key functions and forms. Nevertheless, it should be noted that some research and project partners were not included in the model’s development due to time constraints and competing priorities within the project timeline.

In addition, some researchers (IS, VR) were involved in the design and implementation of the intervention and thus also acted as stakeholders in the project. While this positioning may introduce a potential “judge and party” bias, it also represents a strength by providing in-depth, experience-based insight into the intervention from an insider perspective. This reflects a situated and reflexive approach to knowledge production [40,41]. To mitigate potential bias and circularity in identifying key functions, the analysis was primarily grounded in interview data and supported by iterative discussions with researchers (LF, CD) who were not involved in the intervention’s design or implementation. Any disagreements were resolved by consensus. This process led to several refinements: for instance, the initial distinction between municipal- and school-level key functions was removed in favor of a unified approach, where forms indicated whether components pertained to the National Education system or the municipal level; two key functions were merged to improve coherence.

Perspectives

In this study, the FIC model was applied partially, focusing on the identification of key functions to inform future implementation. The model can be further leveraged to examine social inequalities in health. Analyzing the Alliance intervention through the lens of social inequalities could help anticipate differential effects of the intervention across diverse contexts and support equity-sensitive implementation. Conducting such analyses would require additional, targeted data collection during interviews with both implementers and beneficiaries.

Future research will examine the implementation of the Alliance intervention in new groups of schools and municipalities under routine implementation conditions, as opposed to the experimental conditions of phases 1 and 2. These evaluations will generate complementary insights and inform the formalization of practical, context-sensitive recommendations to guide future replication or scale-up, thereby supporting effective and sustainable implementation.