Contextual Analysis for Practical Action (CAPA): a framework for understanding how context works and how to apply it for practical learning and implementing change in complex systems [version 1; peer review: 2 approved with reservations]

Background: This concept paper presents Contextual Analysis for Practical Action or ‘CAPA’ as an action-oriented tool to understand and implement interventions in real-life service delivery contexts. Understanding how contexts work is a critical challenge for Health and Social Care Systems seeking to become more integrated. CAPA is a tool for understanding and empowering agents in local contexts. The local context is understood here as a key factor in implementation success. This concept paper offers rich theorising to underpin the translation of systems reform initiatives into meaningful local engagement and change. Methods: Analysis of the ‘implementation conundrum’ highlights the problem, while a review of the organisational learning literature identifies ‘sensemaking in organisation’ as a powerful mechanism to address it. Based on earlier phenomenological research the dynamics of organisational sensemaking are presented.  Finally, applications for CAPA are explored and next steps outlined. Results: CAPA makes tangible the implicit and interpretive nature of organisational life as ‘sensemaking in context’. This phenomenon is analysed as a determinant of implementation success. As such, CAPA opens a path to working with contextual complexity by framing it as a dialectic pattern of topical, ecological, dialogical, and pragmatic modes that ‘contextualise’. CAPA is a tool for researchers and practitioners to explore the complexity of local contexts and their implementation challenges, and working-with local agents for learning and change. Conclusions: Health reform needs new approaches for understanding implementation in context. CAPA is a tool for understanding and working-with contextual factors central to the complexity of policy or knowledge translation for example. This paper outlines its rationale, principles, and functionality. Possible applications are explored for coOpen Peer Review


Introduction
This concept paper presents a practical learning approach to implementing change in complex systems and organisations using a new framework for analysing the functional or modal dynamics of contextualisation. Contextual Analysis for Practical Action, or CAPA, is based on earlier phenomenological research into the everyday sensemaking work in a statutory public organisation in Ireland 1 . Many of the characteristics of public organisation and its challenges in implementing change are common across service delivery domains (e.g. health, education, and housing). These include phenomena such as institutionalism, bureaucracy, and embedded cultures that can become intransigent and resistant to change. Siloed languages, identities and meanings reinforce bounded rationalities, unexamined values, obscure path dependencies and fixed organising principles as 'hidden but real' mechanisms 2 .
Such phenomena constitute much of the complexity that implementation science and health services research in particular must learn to address with 'rich theorising' and 'generative learning' 3 . Given the pervasive nature of these challenges, this paper explores the interpretive or everyday sensemaking aspects of that complexity and defines how contextualising factors emerge as important drivers and critical determinants of change management and implementation success. As such, the paper is relevant for change and implementation managers, researchers, organisation leaders, health system reformers and policy makers in general.

Methods
The paper analyses how an 'implementation conundrum' inhibits progression and presents a review of organisational learning literature that highlights 'sensemaking' as an important and complex mechanism of organisational life in this regard. In order to better understand the nature and functionality of sensemaking, its modes and contextualising patterns are explored using earlier phenomenological research and conceptualisation 1 . A contextualising pattern of modal dialectics is thus described, and possible applications for CAPA as an implementation tool are explored.

Analysis
Addressing the implementation conundrum Implementing change at macro, meso or micro levels is challenging due to an 'implementation conundrum' in which universal fear of change, government failure to plan, and a lack of method in working with the social context all play a part 4 . Even with policy in place many implementation projects fail because they are top-down in nature, are siloed in their scope, have unintended consequences, are not evidence-informed, are insufficiently evaluated or do not take enough account of the knowledge translation process 5-8 .
Implementation models often display a project-managementtype or results-based linear rationale that limits consideration of relevant complexities 8,9 . Although newer constructions of more complex causality between interventions and outcomes on realist terms have been developed they are early stage in application and, in using causal logic remain somewhat positivist in their guiding assumptions 10-12 . Of particular use in the newer models is a focus on process outcomes 13 and the need for a consolidated framework to inform change-work across multiple and diverse contexts 14 . This includes understanding to a better degree how innovation can be linked to performance for scalability 15 .
Increasingly the focus is on the diffuse, seemingly chaotic and uncontrollable aspects of organisational life and change implementation in response to health service and system failures 16,17 . Traditional change implementation approaches insufficiently focus on organisation politics for example 18 , or too easily attribute outcomes and knowledge transfers 19 . These types of constraint make measurement, evaluation and controlled progression hard 20 . In a paradoxical twist, models and methods that frame change on its own terms are required 21,22 . Organisational learning offers some relevant resources for this trajectory, and in particular, a conceptualisation of practical learning that models a dialectic pattern of change in the everyday sensemaking (formal and informal) of normal organisational life 1,23 . Modelling this pattern is a step towards generating new ways of working-with everyday sensemaking and its contextualising effects, and offers a path towards addressing some of the challenges highlighted above.
Learning for a change Early organisational learning literature defined the learning process as knowledge creation 24,25 and a construct for 'improving actions through better knowledge and understanding' 26 , but also recognised the complexity of the knowledge-toaction-to-knowledge link [27][28][29] . Initial theorising highlighted the processes of knowledge acquisition, information distribution, interpretation and organisational memory 30 and sought to address organisational boundaries as part of the challenge 31 from a systems perspective 28 . More recently boundary spanning can be linked to overcoming the 'silo effect' 32,33 .
Dynamic interpretations of how knowledge is generated, communicated, integrated and applied were developed 34-36 . Social constructionist approaches posited organisations as languagebased interpretive systems 37-42 , characterised by metaphorical thinking 43 . From this perspective 'systems learning' is knowledge generation, transfer and linkage to action 19 so that knowledge when understood as a possession is ineffective or incomplete 44,45 .
The linguistic-turn in organisation science 46-49 highlighted the power of specific groups and discourses for making organisational knowledge for action 38,40,50 . In application the shift towards 'learning organisations' 51 and 'communities of practice' 52 maps onto more recent constructions of the organisation as a community of meaning 53,54 . For organisations as communities of meaning, readiness for implementation is judged by variables such as absorptive capacity, organisational eloquence, organisational capability for knowledge management, culture, institution and relationships 55-57 .
Although technical or professional knowledge may generate status and identity, it is situational knowledge 58 and understanding of the local (i.e. particular politics, relationships, culture, story, identity, ways of working) that matters for meaning 59 . Situated meaning defines the nature of a particular context as an essential element of the complexity that must be workedwith for implementation success 60 . This focus on situatedness and context is partly in response to system failures with accountability in the health and social care domain (e.g. 61) and generation of the common good. From a systems perspective, there is a pressing need to understand how the successful pilot or pioneer project can spread or scale; and to take this scalability as an indicator of general readiness for change in complex systemic environments (e.g. 62).
Situatedness and the nature of context Situatedness highlights the importance of local context, a term still under theorised 63 . Context has been understood as the backdrop to the real action or intervention, but less frequently as the primary determinant of change 21, 64,65 . Context has also been conceptualised through internal and external influences 66 and organisational levels 67 . A relational approach is taken here for which context is understood and generated in local social coordinations. These are made on the basis of what is known, by the underpinning principles and values, and characterised by the language-used for communication 46,47,68 . In a reflexive process of context-making 'texts' are fixed (for a given time/ space) as coordinated meanings that become 'real', i.e. they are agreed as practical facts or become identifiable phenomena, including artefacts. This happens through multiloging, which is a creative interactive process of making common or shared understandings 69 .
Texts as coordinated meanings are never ultimately fixed; they have no ultimate origin and are always in the process of making, their only limits are the 'socio-cultural context[s] in which [they are] made since these are the reference points to which people coordinate' 39 . This relational view of context coheres with a complex adaptive systems approach in realising that meaning is always emergent, changing, up for reinterpretation, self-organising, far from equilibrium, and path-dependent 70 . With the inherent open-endedness implied this approach makes practical learning a useful driver of implementation and change 28 .
Sensemaking and learning in four interpretive modes CAPA or Contextual Analysis for Practical Action is a practical learning framework emerging from qualitative phenomenological research that explores the nature of organisational learning practice in a large government organisation 1 . It is based on earlier modelling of empirically identified complex management and change processes as everyday organisational sensemaking. It does this using an organisational learning lens 1 .
Sensemaking is understood broadly as the embodied and interactive ability to understand an environment and respond appropriately 71,72 . For CAPA, organisations are understood as complex, adaptive and specifically social systems characterised by self-organisation, emergence, connectivity, interdependence, feedback and path-dependence 70 . This social context is understood as a critical determinant of change, which means interpretive practice is the focus of enquiry through a social constructionist frame 41,73 .
Specifically, CAPA takes a relational perspective to highlight the interactive nature of common organisational functions such as leading, learning, planning, managing, implementing and evaluating 74 . At the heart of enquiry for CAPA (whether practical or explanatory) are the ways in which people make sense of their world and their work by relating or coordinating meanings that drive action; in so doing they generate both agency and context. For example, as agents promote a purpose their positionality is also coordinated intrinsically as an organisational outcome that becomes a contextualising factor 1 . Purpose/positionality therefore becomes a relating practice or coordinated phenomenon to focus on. This type of relating (or related) practice can be corporate, group, team-based, or individual as a somewhat fixed text, but at systemic levels given the characteristics noted above, its emergence and/or sustained nature is rarely linear, simple, nor wholly clear; the system has a life of its own.
Purpose/positionality (as an example of related practice) is made through sensemaking in at least four organisational primary-action-types or modes. These have been identified as: a. Topical, i.e. the language and discourse making, or the topics talked; b. Ecological, i.e. the institutional and cultural structuring and patterning of relations that organise and distribute power, e.g. hierarchy c. Dialogical, i.e. the use of practical tools, technologies, methods and disciplines to process (topical) knowledge to new levels of abstraction for generalisation and action; and d. Pragmatic, i.e. the practical actions and decisions-made that shift empirical parameters.
These interpretive forces generate the character of organisational situations into specific contexts and as such moderate to a significant degree the success or failure of an implementation process 1 . The relating practices emerging from contextmaking, such as purpose/positionality can be understood as mechanisms, somewhat in the line of Pawson and Tilley 75 , but as phenomena they account for their own generative nature as described below.
CAPA and modal dialectics A figure representing the functionality of CAPA (Figure 1) shows how the modal dialectic works. Learning and change are de-centred or distributed so that no single centre, individual, group or unit has full view or control of what is going on; all participants become players as the command and control paradigm is destabilised 76 . The four modes identified as topical, ecological, dialogical and pragmatic are the forces driving activity and change in a cyclical or spiraling feedback loop. Each of these four modal forces introduces different types of change that continually embody (i.e. become real) by virtue of their relatedness. The pattern dialectically generates the topical and dialogical modes through digital feedback, and the ecological and pragmatic modes through analogue feedback 77 . This happens in four core generative coordinations described below as a cycle. The functionality of the cycle is always emergent and unfixed; it has its own, and potentially flexing or deconstructive life due to its complex relating pattern.
The Topical Mode fixes topics as the matters of concern talked about in normal organisational interactions; these can be formalised to a greater or lesser degree in a particular language or discourse. These discourses create an organisational topography.
The Dialogical Mode gives shape to ordering frameworks such as tools and methods, or more embedded practices such as guidelines, rules, procedures and technologies as well as disciplines, constitutions, principles, values, and assumptions.
The topical and dialogical modes generate each other (or are coordinated) through digital feedback (i.e. code) and embody purpose/positionality as a less tangible organisational phenomenon (and force).
The Ecological Mode characterises the social arrangement and artistry of power distributed through organisational settings, structures and project arrangements. This mode is ecological, more than simply environmental because (like the pragmatic mode described below) coordination is constrained by conditions here and now (a form of bounded situation and rationality). The ecology is more or less a resource for change depending on the quality of relations generating it.
Finally, the Pragmatic Mode determines particular actions emerging in, as and for the ongoing cycle. Actions include decisions-made, contracts agreed, budgets distributed etc. In other words, actions change the landscape in ways that empirically matter and therefore generate new topics, which reboots the dialectic cycle. This mode (like the ecological) is constrained by environment and situational capacity giving it its pragmatic character. It is fundamentally characterised as itself because what is done is done, specifically in a place (space) and at a time. This pragmatic mode grounds the whole process of change.
The ecological and the pragmatic modes generate each other (or are coordinated) through analogue feedback (i.e. data) since they are experienced as one thing after another in time and space -there is no (or little) superposition of observation. The ecological and pragmatic modes are experienced as the flow of life about which (in the dialectic pattern) the digital modes (topical and dialogical) create meaning and sense.
Each of the identified modes introduces difference and variability whilst their relatedness generates embodiment or 'reality' in a dialectic pattern, i.e. Topical discourse generates dialogical frameworks through the ecological mode; Dialogical frameworks generate topical discourse through the pragmatic mode; Ecologies generate pragmatic actions through the dialogical mode and finally, Pragmatic actions generate ecologies through the topical (or propositional) mode. These complex relating patterns take shape in the meaningful coordinations, whether conceptual or practical, that participants in any social system are both ordered by, and through which they proactively order their lives.
These are complex relating patterns that function in various forms of organisational discourse (private, public and presentational), creating order, disorder and change as organisations and systems fix and flex coordinated meanings to empower objectives 1 . Organisational objectives are of course myriad and often not aligned; they include intervention and implementation, improvement and innovative practice, governance and strategy as well as more challenging and at times institutionalised phenomena such as resistance to change, silo-building and protectionism.

Results
Through this analysis, contextualisation is highlighted as an important determinant of implementation success and an approach for working-with it as a distinct phenomenon established using CAPA. A range of complex interpretive organisational forces are captured through CAPA which defines them as a dialectic pattern of topical, ecological, dialogical, and pragmatic sensemaking modes. CAPA is presented as a useful conceptual and framing resource for nuanced implementation work in complex adaptive systems.

Discussion -possible applications for CAPA
CAPA is a processual approach to system functionality taking sensemaking in organisational contexts as its unit of analysis, and therefore as an important resource for implementing organisational change. It is useful for working with less tangible or implicit elements of complexity such as culture. It engages uncertainty and emergence on their own terms. CAPA enables a language of participation as a critical determinant of reform 78 . It aims to empower stakeholders collectively recognising that power is distributed according to organisational capability to topicalise issues, to progress them as legitimised knowledge, language and discourse 79,80 and to frame them in practical actions coherent with organisational ends. There is an assumption here that empowering all stakeholders contributes to successful systems.
CAPA assumes that participation in explicit sensemaking as practical learning enables decisions that make a difference for implementation. An implementing team or group can use CAPA as a reflexive tool to 'generate a view' of their own context for example; it can frame-for-action some of the otherwise tacit complexities involved. CAPA can also be used in understanding the challenges of scaling or spreading innovation through large systems at a time when local autonomy and engagement, customisation and distributed leadership emerge as critical factors for implementation success 60,81 . The question of how large public service delivery systems can distribute decisionmaking powers and resources whilst also being accountable is a large one 82,83 .
Given that stakeholder purposes and positionalities often do not align, the need to discover higher order goals to drive change can be important. CAPA offers an opportunity to socialise and locally interpret policy goals for example, drawing on the translation process itself to generate ownership, local application and meaningfulness. The 'people-centered care' discourse of the WHO is such a policy level goal that needs local and practical interpretation in order to make any outcome-difference for patients and providers 84 . Despite the varying purposes and positions of stakeholders, CAPA engages purpose/positionality as a relating practice and organisational complexity that becomes a resource for analysis and change insofar as participants are empowered to fulfill the potential of their task and situation, moving beyond the siloed lens 1 . CAPA aims to shift stakeholders from silo to system whilst remaining a 'bottom-up' approach to implementation.
From the research and practice perspectives, CAPA can potentially generate language for engaging elements of deeply embedded institutional and bureaucratic practices that have become problematic. In line with an appreciative approach 85 as a reflective practice its aim is to enable and empower local insight, ownership and buy-in. To this end CAPA is particularly well suited to the challenges of regionalising integrated care for which new organisational forms of governance, knowledge management, resource allocation, and service delivery are required 86 .
In the Irish context CAPA has been used for case study research focussed on the contextual challenges of delivering integrated care for older persons 87 .
The assumption here is that CAPA can enable a different, distributed organisational imagination. CAPA may also be useful for engaging organisational conflict, particularly across departmental or unit boundaries through a two-staged process of reflexive and multiperspective analysis. Learning how to implement change at scale is a critical challenge for health and social care systems around the world given current population health needs for which they are ill designed and ill equipped 5,88 .
The approach is relevant across several literatures. For example, CAPA can contribute to realist evaluation aimed at generating evidence for complex settings and processes and for understanding how to manage timely adaptation as it takes account of many confounding factors in organisational life. These include discursive engagement, contingency, politics, the structuring dimension, social norming and the like. As an organisational learning tool, CAPA can be useful in the context of quality assurance and risk management as an appraisal method that engages practical service delivery contexts on their own terms. It has the potential to create language and reflection space for naming risk and generating improvement by topicalising issues that traditional, linear frameworks can miss.

Conclusions
Given the challenges of distributing processes such as policy interpretation, shared understanding and implementation at local levels and in context, as well as enabling greater functionality with leadership, governance and change coordination for example, this paper addresses the call for better theory to underpin practical applications. It offers a path of organisational learning with complexity and a method developed from empirical research for practical situated learning about, and with context 1 . The hope is that by framing the pragmatic, heuristic and contingent sensemaking work of everyday organisational settings, and rendering this complexity for local learning and general analysis CAPA can make a useful contribution towards overcoming the implementation conundrum and the challenges of health system reform and system change generally.

Ethics approval
Ethical approval for the source study informing the research underpinning CAPA was given by the Centre for Health Policy and Management and Centre for Global Health Ethics Committee, Trinity College Dublin.

Data availability
No data is associated with this article.
All textual data used in the analytical generation and conceptualisation of CAPA is available at: https://www.taosinstitute. net/resources/dissertations/other-noteworthy-dissertations/sarahelizabeth-dissertation 1

. Organisational Learning on "Another
Map" -phenomenological explorations with ecological epistemology as practical learning for development-aid organising: https://www.taosinstitute.net/resources/dissertations/othernoteworthy-dissertations/sarah-elizabeth-dissertation 1 . Are the conclusions about the method and its performance adequately supported by the findings presented in the article? Partly I really appreciate the opportunity to review this conceptual paper, which discusses important aspects of implementation in practice through the CAPA model.
While I read the paper with interest and applaud the author's ambition to further theory and methods relating to 'the implementation conundrum', I think there are problems with the current version of the paper. I address these and offer comments and suggestions for improvement below.
To me, the most important shortcoming is the lack of theoretical definition and discussion of the main concepts, particularly the concepts of context, contextualization, and sensemaking. Depending on the purpose of the paper (developing practical learning approach or theoretical underpinnings), this can either be fixed relatively easy or is essential.
The author states that "contextualisation is highlighted as an important determinant of implementation success and an approach for working-with it as a distinct phenomenon established using CAPA." However, I could not discern what the concept of contextualization refers to in the paper. This becomes a problem, because the lack of conceptual clarity makes it difficult for me as a reader to follow the author's argumentation and to understand what sentences such as this one refers to: "contextualising factors emerge as important drivers and critical determinants of change management and implementation success". Because this phenomenon, 'contextualization', is identified as important determinant of implementation success and change management, I need to understand what the term refers to, if I am to understand the claim made by the author. Based on my reading of the paper, my best guess is that 'organizational sensemaking' is 'contextualisation. However, the lack of definition or discussion of the term means that it is not clear to me if the author recognizes that contextualization can be individual and social -and that this has important implications for CAPA. As I read this paper, the author focuses on social processes of contextualization and does not touch upon individual processes of contexualisation. This introduces a disconnect between contextualization and sensemaking, because the author uses this latter term to refer to primarily individual phenomenon (although there are contributions to this theoretical field that address collective sensemaking in organizations).
Organizational life is also referred to as 'sensemaking in context'. But because the term 'sensemaking' is only briefly discussed and defined as 'the embodied and interactive ability to understand an environment and respond appropriately', I am not sure the author means to equate sensemaking with contextualization. If so, I suggest that the term is given more Unfortunately, the same lack of conceptual clarity characterizes the key term 'context'. The author states that 'context is understood and generated in local social coordination. These are made on the basis of what is known, by the underpinning principles and values, and characterised by the language-used for communication'. However, as a reader I am left with the pressing question of what does context mean? What does the term refer to in the paper? My best guess is that context in this paper refers to shared constructions of meaning that can become 'the background' or frame for action (i.e. the 'intervention') for a group of people and therefore an important coordination mechanisms in the organizational change process that involves organizations and