Driven by policy, shaped by context: A complexity-informed multiple case study of the six-month review for stroke survivors

Driven by policy, shaped by context: A complexity-informed multiple case study of the six-month review for stroke survivors | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Driven by policy, shaped by context: A complexity-informed multiple case study of the six-month review for stroke survivors RJ Holmes, S Ackerley, D Goodwin, LA Connell This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7490223/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Background: The six-month review is a policy recommendation in the United Kingdom aimed at identifying and addressing the unmet needs of stroke survivors. Differences in the provision of this complex intervention may arise from variations in context. Our study aimed to explore the dynamic interaction between the six-month review and its associated context. Methods: A qualitative multiple case study was conducted across three contrasting six-month review services in England selected primarily to reflect variation in provider organisation. Data collection included semi-structured interviews with three stakeholder groups ( Service Users, Service Providers, and Service Influencers ) , direct observations of the review process, and service-related documents. Data analysis utilised a combined deductive and inductive approach. Using the Context and Implementation of Complex Interventions framework, contextual interactions were mapped at the micro, meso and macro levels across the three cases. A cross-case synthesis, guided by complexity theory, identified key patterns of interaction between the six-month review and its context, which were summarised narratively. Results: Data were collected from 36 stakeholders, 17 hours of observation, and 26 service-related documents. Five key patterns of interaction between the six-month review and its context were identified: (1) Access is a dynamic negotiation between service design and contextual barriers, (2) Equitable service provision requires proactive adaptation , (3) Hidden needs stay hidden unless actively unmasked, (4) System levers may trigger unpredictable consequences, and (5) Outcomes are shaped by interdependence with the wider system. Conclusions: The findings demonstrate the multi-level influence that context has on the implementation and delivery of the six-month review. These contextual interactions shape outcomes in varied, unpredictable, and sometimes unintended ways, reflecting the six-month review’s position within a complex system. The identified patterns of interaction provide insight into the six-month review’s underlying mechanisms and may guide future implementation efforts. Stroke rehabilitation Life after stroke Follow-up care Implementation science CICI Complexity theory Complex interventions Context Health inequities Multiple case studies Contributions to the literature We present a novel method for exploring the influence of context on complex interventions by combining the Context and Implementation of Complex Interventions framework with a complexity-informed taxonomy to generate real-world insights. The findings highlight the importance of considering context as an active agent in shaping implementation efforts. This study contributes to theory-building by illustrating how complexity thinking can deepen understanding of the dynamic interactions between context and complex interventions. Background Stroke survivors often face long-term challenges in functional, cognitive, psychological and quality of life domains[ 1 , 2 ]. With UK prevalence of stroke expected to increase from 1.4 million to 2.1 million over the next decade, and associated societal costs estimated to reach £75 billion[ 3 ], there is an urgent need to ensure that stroke services effectively prevent, manage and mitigate the disabling sequelae of stroke. This need is heightened by the additional impact of an aging population with increasing levels of multimorbidity[ 4 ] and frailty[ 5 ]. The six-month review (6MR) for stroke survivors is recommended in current UK guidelines as a key point in the stroke pathway to identify and address unmet needs[ 6 ]. However, implementation varies significantly[ 7 ], and access remains limited, with only a minority of stroke survivors receiving a structured review[ 8 ]. The limited evidence base for the 6MR’s effectiveness may reflect the absence of a clear theoretical model articulating its function and expected outcomes[ 9 ]. The 6MR aligns with the Medical Research Council’s definition of a complex intervention, characterised by multiple interacting components, flexible delivery and a heterogenous target group[ 10 ]. The Medical Research Council’s framework emphasises the importance of considering context at all stages of the development, refinement and evaluation of complex interventions. The implementation of such interventions is influenced by the context in which they are situated, and vice versa, potentially causing variation in delivery across different contexts[ 11 ]. This context-dependent nature can, therefore, significantly influence implementation success or failure[ 12 ]. However, recent critiques highlight that context is often treated as a static backdrop in health research, rather than as having a complex and dynamic interplay with the intervention, thereby potentially obscuring its influence[ 13 ]. Developing a deeper understanding of the interaction between context and intervention is crucial to theorising the mechanisms through which outcomes emerge and clarifying why implementation varies in practice. To better understand how context influences the 6MR, we explored its interactions across different service settings. By identifying recurring patterns of interaction, we aim to inform future implementation efforts and contribute to the development of a programme theory that reflects the complexity of long-term stroke care. Methods Design This study used a multiple case study design and forms part of a mixed methods project (BE MoRe: Exploring the Benefits and Expectations of the 6-Month Review for Stroke Survivors). The quantitative phase has been reported elsewhere[7]. Adopting a pragmatic stance, we assert that context-dependent knowledge is constructed through the experiences of key stakeholders. Accordingly, we sought to develop contextually grounded, actionable insights to inform future implementation. Case study design aligned with the aims of the project, given the likelihood of uncovering multiple, interacting contextual factors that may shape the intervention. We drew on Merriam’s approach [14] which supports a pragmatic stance by encouraging methodological flexibility and an emphasis on generating practical, context-sensitive understanding across multiple real-world cases[15,16]. The 6MR was observed and analysed through the lens of complexity theory, viewing it as a complex adaptive system. This approach enabled a deep understanding of the dynamic relationships within each system[17], and prevented the urge to explain findings through simplistic, linear patterns of causation[18]. Nieuwenhuijze et al’s[19] taxonomy of the key components of complexity theory relevant to healthcare (see Table 1 ) was used to frame the analysis and interpret the findings. Table 1. Components of complex systems To situate the 6MR within its context and develop an understanding of the interaction between the two, a definition of context was required that accounts for this dynamic relationship. As such, the following definition by Pfadenhauer et al[20] was used: “ Context reflects a set of characteristics and circumstances that consist of active and unique factors. As such, context is not a backdrop for implementation, but interacts, influences, modifies and facilitates or constrains the intervention and its implementation”. Alongside this definition, the Context and Implementation of Complex Interventions (CICI) framework[20] was used as a determinant framework to ensure the consideration of a wide range of contextual factors that may interact with the 6MR. The CICI framework considers context at the micro, meso and macro levels across seven key domains: geographical, epidemiological, socio-cultural, socio-economic, ethical, legal, and political. The CICI framework was used to categorise contextual features, while the complexity taxonomy provided an interpretive lens to examine dynamic interactions. Given the significance of context and complexity within our case study approach, this article has been structured in accordance with the TRIPLE C reporting principles[21]. Positionality The lead researcher (RH), a physiotherapist and PhD candidate, brings extensive experience across the stroke pathway including clinical leadership and commissioning of 6MR services. This practice-informed perspective enriched the study but may have introduced bias in favour of clinically led models. Familiarity with system failings within stroke pathways may also have made some observations seem ‘normal’ or uneventful, leading to unintentionally underemphasising their importance during analysis. Reflexivity was maintained through journaling and regular analytical debriefing meetings with co-authors. The research team included two clinical academic physiotherapists (SA and LC) with expertise in stroke rehabilitation and implementation science research, and a social scientist (DG) with expertise in qualitative methods, providing complementary perspectives and balanced interpretation. Case Selection Three sites were purposively selected from the quantitative stage, primarily to reflect the difference in provider organisations. This was identified to be a key factor in the observed variation in practice[7]. Sites were also chosen to provide diversity in terms of selected contextual factors (i.e. geographical location, level of rurality, and level of deprivation) and system variances (i.e. service delivery method, professionals involved, embedded nature of the service within the stroke pathway). Merriam[14] stresses the importance of clearly defining the boundaries of a case to ensure methodological rigour and coherence. Accordingly, each case was defined as a service responsible for delivering 6MRs, either as its sole function or as part of an integrated service. The boundaries of each case included the direct interactions with other services and the broader influence of contextual factors at a micro, meso and macro level. Bounding each case in this way enabled a focussed yet contextually rich exploration of how systemic and contextual dynamics shape the functioning of the 6MR. Recruitment Three key stakeholder groups were included at each site: Service Users (stroke survivors and their carers or family members); Service Providers (staff involved in directly delivering the 6MR); and Service Influencers (individuals who may have some influence over the delivery of the 6MR such as commissioners, managers, and regional leaders in stroke care). The intended sample size at each site was pragmatically determined as six to eight Service Users , two to three Service Providers , and two to three Service Influencers . However, the final sample size was ultimately a reflexive decision by the research team, based on when sufficient data had been gathered to fully capture the nuances of each case[22]. Service Users were included if they were stroke survivors aged 18 or over, or the carer or family member of a stroke survivor, who received a 6MR within the previous three months. Efforts were made to ensure the inclusion of individuals with communication and cognitive difficulties. Measures included allowing extra time during interviews, providing study-related materials in a communication-accessible format when required, and involving family members to support communication or memory when appropriate. However, stroke survivors who were unable to provide their consent or did not have a consistent method of communication were excluded. Service Users were provided with the study information sheet and invited to take part in the study by the Service Providers at each site. Interested participants either contacted the research team directly or consented to be approached by the lead researcher. Service Providers were purposively approached during initial conversations about using their site as a case study. It was important that agreement was secured before site set-up to ensure it would be possible to collect sufficient data. Service Influencers were also purposively approached during the course of each case study, and were selected based on the specific context and nuances of each case. All consent procedures were undertaken by the lead researcher either face-to-face or, when conducted virtually, using a verbal consent procedure. Data Collection In keeping with robust case study methods, data were collected from multiple sources, including semi-structured interviews, observations, and service-related documents. Doing so allowed a richer understanding of each case through the triangulation of information. All data were collected between June 2024 and April 2025. Semi-Structured Interviews Within each case, the lead researcher conducted all semi-structured interviews, either virtually or in participants’ homes. Different interview schedules were developed for each stakeholder group, with the content and structure guided by the CICI framework. All interviews were audio-recorded, transcribed verbatim and anonymised. Observations Non-participatory observations were undertaken by the lead researcher using a semi-structured data collection sheet guided by the CICI framework. Observations included telephone and virtual appointments, and face-to-face reviews taking place in either a clinic setting, the stroke survivor’s home, or within a care home. All observations were done with signed consent from the Service Provider and with verbal consent of the stroke survivor. No personal data were collected; instead, the focus of the observations was on the content of the review, the interactions between individuals during the appointment, and the outcomes of the review process. All individuals involved were informed that they could ask the lead researcher to leave at any point, without needing to provide a reason. Documents Documents related to each case were also included in the analysis, acquired either from the Service Providers at each site, internet searches, or freely accessible government databases. Documentary data included: service specifications; demographic data of the local population; Sentinel Stroke National Audit Programme (SSNAP) data; service evaluations; and service-related paperwork (e.g. letters/information to Service Users , data collection tools). Data Analysis All data were uploaded to NVivo 14 for analysis. Principles of reflexive thematic analysis were drawn upon to manage coding decisions and guide the generation of themes/patterns[23]. Regular reflexive meetings were held with the research team at all stages of analysis to critically question and refine analytical decisions. Data were coded by the lead researcher independently using a combined deductive and inductive approach. Deductively, data were coded to contextual factors using the CICI framework and to the complexity taxonomy. Inductively, new codes were developed as repeated patterns were identified in the data. In the initial phase of analysis, all codes were used to build a rich description of each case. Analysis progressed iteratively across cases, with insights from one informing interpretation in others, rather than interpreting each case sequentially. Following the descriptive analysis, codes were reviewed across cases to identify contextual subdomains within the CICI framework. The influence that each subdomain had on each case was mapped at the micro, meso and macro levels, when sufficient data was available. These insights were then synthesised across cases using a complexity lens to guide the interpretation and deepen the understanding of how contextual factors interacted to shape implementation, processes and outcomes. Finally, patterns were explored within the data to look for key interactions between context and the 6MR intervention. This was done iteratively by repeatedly checking interpretations against the raw data within each case and refining these concepts. Ethical Considerations The lead researcher is an active clinician who is trained in assessing capacity to obtain valid consent in stroke patients. Participants with communication or cognitive problems were assessed individually, with appropriate adjustments made to support understanding and decision-making. Anonymity was preserved through the removal of identifiable job titles and the redaction of location-specific details in quotes. Each participant was assigned a unique code that indicating their site and stakeholder category. For example, participant BP1 was a Service Provider from Site B, while CU1 was a Service User from Site C. Ethical approval for the study was obtained from the Health Research Authority (REC Reference: 24/WA/0059). Results Service Characteristics Table 2 summarises the characteristics of each service and the demographic context of the areas they serve. Detailed descriptions of each case study can be found in Additional file 1 . Table 2. Comparison of service characteristics and catchment areas across case studies Table 2 highlights key differences across the three services, illustrating the heterogeneity of models implemented in practice. The varied geographical and demographic profiles of the areas served also reflect a diverse set of contextual factors, creating distinct conditions that underpin and influence how each service operates. Data Sources Across the three cases, data were collected from 36 participants, 26 service-related documents, and 17 hours of non-participatory observations. Participants comprised 22 Service Users (19 stroke survivors and three family members), eight Service Providers (four stroke specialist nurses, two therapists, and two third sector stroke co-ordinators), and six Service Influencers . To preserve anonymity, the specific roles of Service Influencers are not detailed; however, all held regional leadership positions and had influence over the implementation of stroke services (via commissioning, strategic or managerial oversight). The documents set comprised one service specification, three demographic reports, three SSNAP reports, two service evaluations, and 16 service-related documents. The latter category consisted of nine pre-review information documents, two data collection documents, and five documents related to post-review actions. As Site B used a single practitioner who performed only telephone appointments observation totalled approximately three hours. The rest of the time was split evenly between Sites A and C. Observations added contextual depth, clarified the different approaches used, and confirmed consistency between stated and observed practices. Analysis of Contextual Factors Cross-case analysis of context using the CICI framework identified 12 subdomains: Geographical (Rurality, Transport Infrastructure); Epidemiological (Disease burden, Age); Socio-Cultural (Language/Culture, Values); Socio-Economic (Affordability, Funding); Ethical (Autonomy); Legal (National audit); and Political (Power distribution, Primary care). The impact of each of these factors within each case is mapped in Additional file 2 . The cross-case synthesis of how these contextual factors interacted with the 6MR at each site was informed by the features of complex systems and can also be seen in Additional file 2. Key Patterns of Interaction The cross-case synthesis highlighted five patterns which illustrate the key ways that context and the 6MR interact to shape implementation, service user experience, and outcomes. An overview of these key patterns is provided in Table 3, with each pattern explored further in the narrative that follows. Table 3. Summary of key interactions [ Please note Table 3 is large therefore has been added as an additional file ] Access is a dynamic negotiation between service design and contextual barriers Stroke survivors encountered a range of contextual barriers that shaped their ability to access the 6MR. The interaction between these barriers, and the structure of the service, influences whether, and in what way, individuals choose to engage with the review process. Important barriers included the degree of rurality, the limitations of transport infrastructure, and the associated cost to the individual. Participants emphasised that the decision to access the 6MR was a consideration between these barriers and the perceived benefit of the review. Contextual barriers could be reduced when service users were given options around how the 6MR could be accessed. …it's a bus and then a taxi (laughing). So, it's, it's easier if it's just straight questions over the phone. - BU5 When choice of access wasn’t available, services saw a reduction in uptake of the review. When choice was offered, service users considered the aforementioned barriers alongside other factors, such as their culture and values, to choose the most suitable method. However, some service users opted for ease rather than selecting a method that most complemented their specific needs. In such instances, needs may be missed. Accessibility appeared to work most effectively, from a user experience perspective, when decisions around access were shared between the stroke survivor and the service provider. …I suppose that that is a mixture of my decision making and theirs. So, if I've read their notes, and the person is elderly, hard of hearing, maybe got issues around... additional issues, you know, whatever those issues are, but issues which are going to make a phone conversation difficult. – CP2 Access to the service was also hindered by barriers inherent in its structure. This was particularly evident for individuals who spoke languages other than English, and in some instances, certain patient groups were excluded without clear justification. Our particular service doesn't work with people in nursing homes. So, we are not commissioned to do that. – CP2 Equitable service provision requires proactive adaptation The heterogeneity of impairments faced by stroke survivors necessitates an individualised approach to care. Consequently, a ‘one-size-fits-all’ approach to 6MR provision can leave service users feeling excluded or perceiving the review as irrelevant to their needs. I think, given that each patient is so different and their needs are so different, I think there needs to be flexibility, like, in appointment times and things to be able to do that because it shouldn't be about just ticking a box and saying it's been delivered. It's supposed to be meaningful and add value to the individual – BI1 Key contextual factors that risked individuals experiencing inequitable service provision include their level of disease burden, their culture and language, and their age. In these contexts, some providers adapt their service to lessen health inequalities. Examples of this include using proxies or interpreters to support the review, or automatically booking face-to-face appointments for stroke survivors who may struggle with the process of organising their 6MR appointment (e.g. those with communication and cognition difficulties). …we do an optional now, but, of course, it doesn't pick up in people with aphasia or people with just difficulties, you know, all of those, or people that just are very old. So, what we've done recently is we've decided we're just going to offer appointments to care homes and residential homes or people that we know from ESD [the early supported discharge pathway] will struggle… - AP3 When services were unaware of a stroke survivor’s specific needs, or aware but unable to adapt their provision accordingly, access and support were compromised. To facilitate effective adaptation, services needed to have a degree of knowledge about the individuals accessing their service. When services were embedded within the pathway this was easier to achieve. Hidden needs stay hidden unless actively unmasked A key role of the 6MR is to identify unmet needs. However, in certain contexts, needs are stubbornly hidden and do not emerge passively; their visibility is reliant on the structure of the service and the skills of the providers. Based on the experiences of participants, this phenomenon was particularly relevant where individuals spoke a language other than English, had specific cultural needs, or lacked capacity to advocate for themselves. When services are rigid and unable to adapt to these micro-level contexts, the ability for individuals to access the 6MR may be inhibited, or their experience degraded, as unmet needs may be missed. Examples of this include non-English speakers not understanding paperwork they have been sent, or stroke survivors in nursing homes having their 6MR completed via a proxy rather than including them. So, we realised quite quickly it wasn't working and obviously they are missing out and they shouldn't have to miss out just because they can't speak the language - BP2 Because I think we have as well, previously in the past, done them with the nursing home staff over the phone. And so, then it's a second hand. - BP1 Some participants expressed core values that influenced their interaction with the 6MR. They reported: not wanting to waste the time of the provider; deferring to the provider’s opinion; and feeling it was sometimes inappropriate to discuss certain needs (e.g. sexual relations), especially during phone consultations. That it wasn't face-to-face, it seemed a bit strange, but I was told that's how it's going to be. It's alright. - BU2 In such instances, needs were withheld by the stroke survivor. Some also expressed dissatisfaction with phone-based reviews if it was perceived to hinder the provider’s ability to make an informed decision. …[the provider said] “you could go to this group or that group, providing we could get you transport”, which I can't do at the moment. I can't go to any groups because I can't go anywhere. - BU3 When services are cognisant to the diversity of sociodemographic factors and able to adapt the structure of the 6MR in response, the needs of stroke survivors are more likely to be unmasked. Examples include: establishing a dedicated clinic aligned to the language and cultural needs of specific communities; the effective and consistent use of translator services; the provision of information prior to the 6MR to explain its purpose and to highlight unmet needs the stroke survivor could consider; and the appropriate involvement of family members to support the identification of need. This unmasking seemed to work most effectively when the provider's knowledge and skills enabled a flexible approach, creating a space where the stroke survivor felt able to discuss issues freely. …experience has given us that freedom to go with the flow and allow the patient to lead the session rather than the clinician leading the session, which is just, yeah, just experience isn’t it? And kind of not feeling out of depth when they throw you curve balls! – AP1 System levers may trigger unpredictable consequences Some system-level contexts may act as levers for change in efforts to raise the priority of the 6MR. Examples of this include the national stroke audit and financial incentives linked to performance. However, these levers can inadvertently produce perverse incentives, where efforts to optimise performance metrics may occur at the expense of person-centred care. This appeared evident in the increased uptake of reviews in response to financial incentives in 2019, which was coupled with an increase in telephone appointments. This concept of perverse incentives is also exemplified by services using SSNAP reporting requirements to guide the timing of the review, working within the boundaries of what is required, rather than focusing on what the individual actually needs. ...it is obviously a priority, but I know if we are very busy that we've got a three-month window [on SSNAP]. So, they do... we have put them on hold for a month and then followed up. - BP3 All sites were observed to collect data required for SSNAP but made limited use of it, collecting information primarily for audit compliance rather than to inform or enhance the review process. It probably does feel a little bit like a tick box in some circumstances that we're just doing it because we're told that it should be done. – BP2 The degree of managerial oversight was another lever that could influence the focus of the review process. Higher levels of managerial scrutiny seemed to trigger the service to focus on efficiency and cost-effectiveness, though this could lead to a loss in patient-centred care. Conversely, a lack of oversight enabled providers greater freedom to adhere to their patient-driven ethos. I suppose that we kind of can do our own thing that we’re not tied to delivering it in a certain way and they’ve sort of left us to be fairly free as long as we delivery it. - AP1 Outcomes are shaped by interdependence with the wider system The success of the 6MR relies on its interactions with the wider care system. When the 6MR was embedded within an integrated pathway, information flowed more easily, improving continuity and responsiveness. This was evident in several ways. Firstly, integrated computer systems enabled providers to have the stroke survivor’s history at their fingertips. This gave the stroke survivor reassurance and stopped the provider from needing to repeat aspects of the review. I can see if they've had a community therapist involved, so I can do some reading back and have an idea of what's happened. – AP2 Secondly, when 6MRs were completed by the same team who delivered their rehabilitation, there was a belief that stroke survivors’ needs had already been met and, therefore, the 6MR could be streamlined. I think that quite often we give them such a good service from an ESD point of view that they don't tend to want a stroke follow-up because they feel their needs are met, which is good. – BP3 When services were not embedded, there was a need to ensure that communication within the wider system was optimised. Sometimes this required a proactive approach to build connections across organisations. …we have had to build those relationships and it's not always been easy. – CP1 Building links with other services helped 6MR providers stay informed about incoming stroke survivors and tailor their approach accordingly. These links also supported awareness of local services and facilitated easier onward referrals. The structure of the 6MR was further shaped by how the service interacted with primary care. In some services, providers undertook additional tasks beyond the review itself to ease the burden on GPs and improve the experience for stroke survivors. In contrast, others (often constrained by financial or staffing limitations) took a more minimalist approach, handing responsibility for follow-up actions to the GP. This happened even when providers questioned whether the additional work may be too much for the GPs to handle. We would put it back to the GP to make the referral. So, say they needed, I don't know, community physio or something. We would ask the GP to do that rather than do it ourselves. And I guess there is that worry that will that actually get done? – BP Discussion Our study took a novel approach to investigate the interactions between the 6MR and its context. We found that context is not just the backdrop to the setting of the 6MR. Instead, it has a dynamic interplay with the service, the intervention, and the individuals involved, and it does so across multiple nested levels within the complex system. These interactions were observed to shape stroke survivors’ experience within the service and influence providers’ delivery of the 6MR. Understanding the potential mechanisms behind these dynamics requires attention to the layered and complex nature of health systems. Contextual factors identified in this study influenced how stroke survivors interacted with the 6MR and, thereby, influenced their outcomes. This was evident in terms of how they accessed the review, what needs they highlighted, how services were set up to deal with these needs, and how those needs were interpreted and acted upon by providers. The diversity and unpredictability of needs expressed further confirms the complexity of the 6MR. Ensuring that these needs are identified requires a personalised and adaptable approach. When providers were better informed about the person’s individual circumstances, they were able to make adaptations to their provision of care to optimise the accessibility and experience for the stroke survivor. While this flexibility appeared to increase the likelihood that unmet needs would be surfaced and addressed, outcomes were non-linear and some barriers persisted despite provider efforts. This aspect was especially important for marginalised groups who might need additional support to access the 6MR. When this flexibility was lacking, the chance of these groups experiencing inequity increased. This was evident in cases of stroke survivors who spoke languages other than English or care home residents having their review completed via a proxy. As such, our findings suggest that the standardisation of reviews may inadvertently undermine equity. Without a model that can adapt to support individual need and contextual differences, the 6MR risks becoming a procedural formality, one that records symptoms rather than enabling recovery. At the meso level, contextual factors shaped how organisations interpreted and operationalised the 6MR, influencing both the structure of services and their underlying ethos. The variation in service models observed nationally[7] can be understood as a response to the differing combinations of contextual factors faced by providers in each locality. Self-organisation in response to context is a feature of complex systems and variation of this kind should be embraced[24]. Doing so enables the emergence of service models that are responsive to the needs of the local population, and that operate optimally within their available resources[25]. However, context may also drive services to adopt models that could negatively impact clinical outcomes. In our study, this was shaped by funding constraints and the distribution of decision-making authority. All sites had to adapt service provision in response to limited resources, but the extent to which they could innovate depended on the decision-making structure within the service. A clear example of this dynamic can be seen at Site B where an increase in demand and limited resources triggered a strong organisational focus on efficiency. In this case, the pursuit of efficiency using a telephone model of delivery, while necessary, appeared to be at the expense of patient experience and needs may be missed. Levesque and Sutherland[26] highlight that strong performance in one measurement domain may cause detrimental effects in another; as such, a focus on efficiency to meet targets and manage demand may come at the expense of overall quality. Our study found that macro-level levers, while effective in driving national uptake of the 6MR, may have negative effects on the implementation of services and the experience of stroke survivors. This finding is not unique; examples of other national improvement initiatives report similar unintended consequences[27,28]. Mannion and Braithwaite[29] highlight a number of these potential consequences, including: tunnel vision (focusing primarily on the elements of measurement); gaming (adopting altered behaviours to gain a reporting advantage); and measurement fixation (emphasising data collection over the spirit of the measure). With regards to the 6MR, these adverse consequences were mirrored in the propensity for providers to spend time collecting data that they did not use; altering the timing of the review to optimise performance measures; and the use of proxies to complete reviews for marginalised groups. Providers strive for a person-centred ethos but are sometimes necessitated to act in such ways as to abide by measurement requirements, even when this necessity doesn’t fit comfortably into their local context and service delivery. These findings suggest the need for future measures of performance to move beyond procedural compliance and instead focus on tangible outcomes that are meaningful, equitable, and valued by service users. Strengths and limitations This study is strengthened by the use of theory to underpin both data collection and analysis. The CICI framework enabled consideration of a broad range of factors across nested contextual levels. This ensured that context was not narrowly conceptualised as only the sociodemographic factors of the immediate service environment, but also encompassed the beliefs and motivations of individuals within the system, and the wider influence of macro-level policies and drivers. The use of the complexity taxonomy further strengthened our analysis, allowing deeper insights into the mechanisms of complex systems and ensuring that findings were grounded in the ‘real’, rather than idealistic, world. However, the study is limited by the extent of data collected. Firstly, our study protocol prevented the inclusion of stroke survivors who declined the 6MR. Their perspectives, especially in relation to their decision-making process, could have provided further insight into how context influences uptake and highlighted factors not considered by those who undertook the review. Secondly, the limited timeframe for data collection constrained the ability to observe how services responded to evolving contextual factors. A longer period of observation at each site would have enabled deeper insight into service adaption over time. Instead, we relied on service providers’ recall of historical events which may be prone to bias. Longitudinal engagement could also have provided insight into how stroke survivors’ attitudes towards the 6MR changed over time and how context continued to influence their outcomes. Finally, in the context-dependent reality of complex systems (characterised by non-linearity, emergence, and unpredictability) the concept of universally generalisable findings becomes problematic, regardless of the research method used. Because the 6MR is shaped so strongly by context, the conditions that give rise to findings in one setting are unlikely to be fully replicated elsewhere. The value of case study research in complex systems lies not in producing generalisable claims in the traditional sense, but in its ability to uncover underlying mechanisms, identify recurring patterns, and generate context-sensitive insights. These can inform understanding of how similar dynamics might unfold in other settings, and may be used to support decision-making for current, and future, 6MR services. Conclusion This study has highlighted key patterns of interaction between the 6MR and its context, observed through the lens of complexity theory, across three 6MR services. These dynamic interactions shape how stroke survivors experience, and how providers deliver, the 6MR. The complexity and multifaceted nature of interacting elements of context make prediction of outcomes inherently uncertain. It is the unique mixture of contextual factors, combined with how the 6MR service responds to said context, that influences whether the intervention will be successful or not. However, recurring patterns of interactions identified in this study offer valuable insights that may guide reflection and consideration in other 6MR services, aiding their development or reconfiguration. These findings contribute to the development of programme theory by clarifying how contextual factors shape the delivery and function of 6MR services. This evolving theory will support the generation of actionable recommendations, which will be explored in future work. Abbreviations 6MR – six-month review CICI – Context and Implementation of Complex Interventions ESD – early supported discharge GM-SAT© – Greater Manchester Stroke Assessment Tool NHS – National Health Service ONS – Office for National Statistics SSNAP – Sentinel Stroke National Audit Programme Declarations Ethics approval and consent to participate Ethical approval for the study was obtained from the Health Research Authority (REC Reference: 24/WA/0059). Informed consent was obtained from all participants. Consent for publication Not applicable. Availability of data and materials All data generated or analysed during this study are included in this published article [and its supplementary information files]. Competing interests The authors declare that they have no competing interests. Funding RH (Clinical Doctoral Research Fellow, NIHR 302163) is funded by the NIHR for this research project. The views expressed in this publication are those of the authors and not necessarily those of the NIHR, NHS or the UK Department of Health and Social Care. Authors' contributions RH contributed to the study conceptualisation, design, data acquisition and analysis, and drafting of the manuscript. SA contributed to the study design, data analysis, and editing of the manuscript. DG contributed to the study design, data analysis, and editing of the manuscript. LC contributed to the study conceptualisation, design, data analysis, and editing of the manuscript. All authors read and approved the final manuscript. Acknowledgements We gratefully acknowledge study participants who provided their time to take part in the study. We are also extremely grateful to the three study sites for allowing access to their services and supporting the recruitment of participants. References Crichton SL, Bray BD, McKevitt C, Rudd AG, Wolfe CD. Patient outcomes up to 15 years after stroke: survival, disability, quality of life, cognition and mental health. Journal of Neurology, Neurosurgery & Psychiatry. 2016 Oct 1;87(10):1091-8. De Wit L, Theuns P, Dejaeger E, Devos S, Gantenbein AR, Kerckhofs E, Schuback B, Schupp W, Putman K. Long-term impact of stroke on patients’ health-related quality of life. Disability and rehabilitation. 2017 Jul 3;39(14):1435-40. King D, Wittenberg R, Patel A, Quayyum Z, Berdunov V, Knapp M. The future incidence, prevalence and costs of stroke in the UK. Age and ageing. 2020 Feb 27;49(2):277-82. Gallacher KI, Jani BD, Hanlon P, Nicholl BI, Mair FS. Multimorbidity in stroke. Stroke. 2019 Jul;50(7):1919-26. Evans NR, Todd OM, Minhas JS, Fearon P, Harston GW, Mant J, Mead G, Hewitt J, Quinn TJ, Warburton EA. Frailty and cerebrovascular disease: concepts and clinical implications for stroke medicine. International Journal of Stroke. 2022 Mar;17(3):251-9. Intercollegiate Stroke Working Party. National Clinical Guideline for Stroke for the United Kingdom and Ireland [Internet]. 2023 [cited 2025 Jul 16]. Available from: https://www.strokeguideline.org/ Holmes R, Ackerley S, Fisher RJ, Connell LA. Exploring variation in the six-month review for stroke survivors: a national survey of current practice in England. BMC Health Services Research. 2025 Jan 28;25(1):159. Sentinel Stroke National Audit Programme (SSNAP). SSNAP Annual Report 2024 [Internet]. London: Healthcare Quality Improvement Partnership; 2024 [cited 2025 Jul 16]. Available from: https://www.hqip.org.uk/wp-content/uploads/2024/11/Ref.-466-SSNAP-Annual-report-2024-FINAL.pdf Rutter H, Savona N, Glonti K, Bibby J, Cummins S, Finegood DT, Greaves F, Harper L, Hawe P, Moore L, Petticrew M. The need for a complex systems model of evidence for public health. The lancet. 2017 Dec 9;390(10112):2602-4. Skivington K, Matthews L, Simpson SA, Craig P, Baird J, Blazeby JM, Boyd KA, Craig N, French DP, McIntosh E, Petticrew M. A new framework for developing and evaluating complex interventions: update of Medical Research Council guidance. bmj. 2021 Sep 30;374. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, Moore L, O’Cathain A, Tinati T, Wight D, Baird J. Process evaluation of complex interventions: Medical Research Council guidance. bmj. 2015 Mar 19;350. Grant A, Bugge C, Wells M. Designing process evaluations using case study to explore the context of complex interventions evaluated in trials. Trials. 2020 Nov 27;21(1):982. Murdoch J, Paparini S, Papoutsi C, James H, Greenhalgh T, Shaw SE. Mobilising context as complex and dynamic in evaluations of complex health interventions. BMC health services research. 2023 Dec 18;23(1):1430. Merriam SB. Qualitative research and case study applications in education. 2nd ed. San Francisco: Jossey-Bass Publishers; 1998. Yazan B. Three approaches to case study methods in education: Yin, Merriam, and Stake. The Qualitative Report. 2015 Feb 1;20(2):134-53. Harrison H, Birks M, Franklin R, Mills J. Case Study Research: Foundations and Methodological Orientations. FQS [Internet]. 2017 Jan. 24 [cited 2025 Jul. 16];18(1). Available from: https://www.qualitative-research.net/index.php/fqs/article/view/2655 Braithwaite J, Churruca K, Long JC, Ellis LA, Herkes J. When complexity science meets implementation science: a theoretical and empirical analysis of systems change. BMC medicine. 2018 Apr 30;16(1):63. Tsoukas H. Don't simplify, complexify: From disjunctive to conjunctive theorizing in organization and management studies. Journal of management studies. 2017 Mar;54(2):132-53. Nieuwenhuijze M, Downe S, Gottfreðsdóttir H, Rijnders M, du Preez A, Rebelo PV. Taxonomy for complexity theory in the context of maternity care. Midwifery. 2015 Sep 1;31(9):834-43. Pfadenhauer LM, Gerhardus A, Mozygemba K, Lysdahl KB, Booth A, Hofmann B, Wahlster P, Polus S, Burns J, Brereton L, Rehfuess E. Making sense of complexity in context and implementation: the Context and Implementation of Complex Interventions (CICI) framework. Implementation science. 2017 Feb 15;12(1):21. Shaw SE, Paparini S, Murdoch J, Green J, Greenhalgh T, Hanckel B, James HM, Petticrew M, Wood GW, Papoutsi C. TRIPLE C reporting principles for case study evaluations of the role of context in complex interventions. BMC Medical Research Methodology. 2023 May 13;23(1):115. Braun, V., & Clarke, V. To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. Qualitative research in sport, exercise and health. 2021 Mar 4;13(2):201–216. Braun V, Clarke V. Reflecting on reflexive thematic analysis. Qualitative research in sport, exercise and health. 2019 Aug 8;11(4):589-97. Braithwaite J, Churruca K, Ellis LA. Can we fix the uber-complexities of healthcare?. Journal of the Royal Society of Medicine. 2017 Oct;110(10):392-4. Sutherland K, Levesque JF. Unwarranted clinical variation in health care: definitions and proposal of an analytic framework. Journal of evaluation in clinical practice. 2020 Jun;26(3):687-96. Levesque JF, Sutherland K. Combining patient, clinical and system perspectives in assessing performance in healthcare: an integrated measurement framework. BMC health services research. 2020 Jan 8;20(1):23. Mannion R, Davies H, Marshall M. Impact of star performance ratings in English acute hospital trusts. Journal of Health Services Research & Policy. 2005 Jan;10(1):18-24. McLean G, Sutton M, Guthrie B. Deprivation and quality of primary care services: evidence for persistence of the inverse care law from the UK Quality and Outcomes Framework. Journal of Epidemiology & Community Health. 2006 Nov 1;60(11):917-22. Mannion R, Braithwaite J. Unintended consequences of performance measurement in healthcare: 20 salutary lessons from the English National Health Service. Internal medicine journal. 2012 May;42(5):569-74. Tables Table 1. Components of complex systems Component Description Interconnection Complex systems have multiple connected elements, and their make-up continually changes. Outcomes emerge from the interactions within, and external to, the system. Open “Fuzzy” boundaries Ideas, information, and people often move freely across boundaries. Individuals can be members of multiple interacting systems at the same time with multiple influential interactions. Initial conditions (historicism) System history and initial conditions have a strong influence over future behaviour. New practices that are consistent with historical context may be more readily embraced than changes that dramatically deviate from previous practice. Self-organisation Complex systems have a tendency to organise spontaneously, in response to contextual influence, without the need for hierarchical control. Simple rules Individuals within a complex system follow a series of internalised values or principles, either explicitly or implicitly known, that drive self-organisation and emergent outcomes. Feedback loops Changes within the system gives rise to feedback that either reinforce the new pattern of behaviour or seek to suppress it. Co-evolution Complex systems evolve in line with other systems that they are closely linked with or embedded within. Emergence Interactions within a complex system produce outcomes whereby the mechanisms of action are not clearly understood. These outcomes are more than just the sum of their parts such that emergent properties cannot be explained just by exploring the relevant parts of the system. Non-linearity Effects produced within a complex system can be non-proportional or paradoxical. Small changes can lead to dramatic effects; equally, they can fail to produce the intended outcome despite a large, multifaceted intervention. Unpredictability Processes and outcomes cannot be known in advance. Accurate understanding of anticipated outcomes is not possible. Attractors Attractors are the states or patterns that a complex system is drawn to over time. Attractors can dictate how the system behaves whether that be a period of stability following a period of turmoil, repeated cyclical patterns, or chaotic patterns of repeated change. Table 2. Comparison of service characteristics and catchment areas across case studies Case A Case B Case C Service Characteristics Provider Organisation Community NHS Trust Acute NHS Trust Third-Sector Organisation Current Staffing Level within 6MR 3x Stroke Specialist Nurses (part-time within 6MR service) 1x Therapist (part-time within 6MR service) 3x Stroke Co-ordinators (part-time within 6MR service) Method of Review Face-to-face in clinic setting and home visits (including care homes) or telephone appointments Majority are telephone appointments (face-to-face in special circumstances) Home visits or telephone appointments (option for virtual) Position of 6MR within Stroke Pathway Embedded within ESD team Embedded within ESD team Delivered as part of a third-sector-led stroke support service Data collection tool In-house tool Adapted GM-SAT Adapted GM-SAT Local Area Profile Region East of England South East North West Local population* 662,000 439,000 328,300 Rurality (ONS Rural Urban Classification) Mixed rural and urban: Mostly rural town/fringe surrounding urban city and town Predominantly urban city and town Mixed urban and rural: Urban city and town with adjacent rural town and fringe areas Deprivation levels (% Households deprived in one or more dimension)* NB. National average = 52% 52% 44% 48% Ethnicity (% White)* NB. National average = 81% 94% 86% 95% *Data taken from the ONS 2021 census. (ESD – Early Supported Discharge, a model of home-based, multidisciplinary rehabilitation that enables early discharge from hospital; ONS – Office for National Statistics; GM-SAT – Greater Manchester Stroke Assessment Tool©; NHS – National Health Service). Table 3. Summary of key interactions Please see Additional file 3. Supplementary Files Add1.docx Additional File 1. Add 1.docx Descriptive analysis of the three case studies – Additional File 1 presents a detailed description of the service and context within each of the three case studies. Add2.xlsx Additional File 2. Add 2.xlsx Mapping of contextual factors and complexity-informed cross-case synthesis – Additional File 2 shows the identified subdomains of context under the CICI framework. How these contextual factors interact with the intervention at each site is highlighted alongside evidence from each case. Finally, the table shows the cross-case synthesis that was informed by components of the complexity taxonomy. TripleCreportingguideline.docx Additional File 3. Add 3.docx Table 3. Summary of key interactions – Table 3 has been included as an additional file as it is too large as per the submission guidance. We have highlighted with the main text (line 265) where the table should be inserted. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 12 Nov, 2025 Reviewers invited by journal 12 Nov, 2025 Editor assigned by journal 04 Sep, 2025 First submitted to journal 03 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7490223","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":543700674,"identity":"0c6dc7bc-bc12-40b5-953b-e8b33e60f9e1","order_by":0,"name":"RJ Holmes","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYDACZgY2MGnAwNwAZNjIMUgQr4URpCXNGKrFAJ8mFC2HExsIaTFvZ3724OceawZz9sbWDR/+pKVvl25/wFxQ8QenFpnDbOaGPc/SGSx7DrbdnNlmk7tzzhkD5hlncNsiwczDJsFz4DCDwY3Ettu8DWm5G27kMDDztuHXIvkHpuXPn8PpBjfSHzDz/sOvRRpuCwPb4QSDGwkGzLwN+LSwmUnLHEjnMTgD9EtvW5oh0GEGh3mOGePWwn/4meSbA9ZyBsebj9348cdGHuiwh495auRwaoEBHhTeAYLqR8EoGAWjYBTgBQDQwVP6aWzePAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-1703-5563","institution":"University Hospitals Sussex NHS Foundation Trust","correspondingAuthor":true,"prefix":"","firstName":"RJ","middleName":"","lastName":"Holmes","suffix":""},{"id":543700675,"identity":"3d337d8b-e721-4a67-ae90-ca350bf264d4","order_by":1,"name":"S Ackerley","email":"","orcid":"","institution":"Lancaster University","correspondingAuthor":false,"prefix":"","firstName":"S","middleName":"","lastName":"Ackerley","suffix":""},{"id":543700676,"identity":"1a34a7a9-18e7-4327-9ddb-2667832da73d","order_by":2,"name":"D Goodwin","email":"","orcid":"","institution":"Lancaster University","correspondingAuthor":false,"prefix":"","firstName":"D","middleName":"","lastName":"Goodwin","suffix":""},{"id":543700677,"identity":"259f050c-d6d5-474e-9553-a9c5249b5fd3","order_by":3,"name":"LA Connell","email":"","orcid":"","institution":"Lancaster University","correspondingAuthor":false,"prefix":"","firstName":"LA","middleName":"","lastName":"Connell","suffix":""}],"badges":[],"createdAt":"2025-08-29 16:08:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7490223/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7490223/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96914330,"identity":"f32ed201-e8b2-44c0-865b-877b5b628cb3","added_by":"auto","created_at":"2025-11-27 14:05:46","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18865,"visible":true,"origin":"","legend":"","description":"","filename":"Add3.docx","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/0d08f3664cdeb94cf3e23b96.docx"},{"id":96914304,"identity":"a0a8a42a-5fdf-49eb-99ee-d2822cec4732","added_by":"auto","created_at":"2025-11-27 14:05:41","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10443,"visible":true,"origin":"","legend":"","description":"","filename":"iscmISCMD2500348.xml","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/1a353bc29425a09452b30a47.xml"},{"id":96746882,"identity":"a0b166cc-67ca-4cf0-b797-47757a1f5bf7","added_by":"auto","created_at":"2025-11-25 16:09:43","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1087,"visible":true,"origin":"","legend":"","description":"","filename":"ISCMD25003486323.go.xml","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/22aa221c67e64d3f91bc1b65.xml"},{"id":96746878,"identity":"9b491bd5-1edd-4bbe-847c-97cc28bd1d2b","added_by":"auto","created_at":"2025-11-25 16:09:43","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":924,"visible":true,"origin":"","legend":"","description":"","filename":"ISCMD2500348Import.xml","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/dec2f1964a01f1ead93bc846.xml"},{"id":96746886,"identity":"f554fabc-a343-42d8-baed-6cd94fefb876","added_by":"auto","created_at":"2025-11-25 16:09:43","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117579,"visible":true,"origin":"","legend":"","description":"","filename":"ISCMD25003480enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/eb8a6ccd793aba8a05cbb0dd.xml"},{"id":96746888,"identity":"3694e0b5-f6b1-4bfd-89b0-cd1609e3d5c7","added_by":"auto","created_at":"2025-11-25 16:09:43","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115172,"visible":true,"origin":"","legend":"","description":"","filename":"ISCMD25003480structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/60e776734a1b4a6e06f15543.xml"},{"id":96915090,"identity":"46c876cf-fc15-4d20-bb30-f12aa0b7494a","added_by":"auto","created_at":"2025-11-27 14:06:50","extension":"html","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127881,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/73544cd699d107b1ed552f8a.html"},{"id":97249130,"identity":"f919989a-dcf7-4e66-a3b1-d88a8765c27b","added_by":"auto","created_at":"2025-12-02 13:10:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1019164,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/a98cd6d8-6cbb-430d-bc2b-7703f0095e7c.pdf"},{"id":96914159,"identity":"6f48df97-8a2f-4a64-bfc3-90b0871d7033","added_by":"auto","created_at":"2025-11-27 14:05:32","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27942,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdditional File 1.\u003c/strong\u003e Add 1.docx\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDescriptive analysis of the three case studies\u003c/em\u003e – Additional File 1 presents a detailed description of the service and context within each of the three case studies.\u003c/p\u003e","description":"","filename":"Add1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/4f9d6ed76f073fdca09f5f2b.docx"},{"id":96914614,"identity":"f016c57f-ee3b-4d77-a9d4-e4d7d446d42f","added_by":"auto","created_at":"2025-11-27 14:06:10","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":29495,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdditional File 2.\u003c/strong\u003e Add 2.xlsx\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMapping of contextual factors and complexity-informed cross-case synthesis \u003c/em\u003e– Additional File 2 shows the identified subdomains of context under the CICI framework. How these contextual factors interact with the intervention at each site is highlighted alongside evidence from each case. Finally, the table shows the cross-case synthesis that was informed by components of the complexity taxonomy.\u003c/p\u003e","description":"","filename":"Add2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/82f9381f39bac98f042bdadd.xlsx"},{"id":96746885,"identity":"08c97387-b722-4adc-87be-53b7caed066e","added_by":"auto","created_at":"2025-11-25 16:09:43","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":18918,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdditional File 3\u003c/strong\u003e. Add 3.docx\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 3. Summary of key interactions\u003c/em\u003e – Table 3 has been included as an additional file as it is too large as per the submission guidance. We have highlighted with the main text (line 265) where the table should be inserted.\u003c/p\u003e","description":"","filename":"TripleCreportingguideline.docx","url":"https://assets-eu.researchsquare.com/files/rs-7490223/v1/61a31fe33e0ece8008c90edc.docx"}],"financialInterests":"","formattedTitle":"Driven by policy, shaped by context: A complexity-informed multiple case study of the six-month review for stroke survivors","fulltext":[{"header":"Contributions to the literature","content":"\u003cul\u003e\n \u003cli\u003eWe present a novel method for exploring the influence of context on complex interventions by combining the Context and Implementation of Complex Interventions framework with a complexity-informed taxonomy to generate real-world insights.\u003c/li\u003e\n \u003cli\u003eThe findings highlight the importance of considering context as an active agent in shaping implementation efforts.\u003c/li\u003e\n \u003cli\u003eThis study contributes to theory-building by illustrating how complexity thinking can deepen understanding of the dynamic interactions between context and complex interventions.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"Background","content":"\u003cp\u003eStroke survivors often face long-term challenges in functional, cognitive, psychological and quality of life domains[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. With UK prevalence of stroke expected to increase from 1.4\u0026nbsp;million to 2.1\u0026nbsp;million over the next decade, and associated societal costs estimated to reach \u0026pound;75 billion[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], there is an urgent need to ensure that stroke services effectively prevent, manage and mitigate the disabling sequelae of stroke. This need is heightened by the additional impact of an aging population with increasing levels of multimorbidity[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] and frailty[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe six-month review (6MR) for stroke survivors is recommended in current UK guidelines as a key point in the stroke pathway to identify and address unmet needs[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. However, implementation varies significantly[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], and access remains limited, with only a minority of stroke survivors receiving a structured review[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The limited evidence base for the 6MR\u0026rsquo;s effectiveness may reflect the absence of a clear theoretical model articulating its function and expected outcomes[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe 6MR aligns with the Medical Research Council\u0026rsquo;s definition of a complex intervention, characterised by multiple interacting components, flexible delivery and a heterogenous target group[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. The Medical Research Council\u0026rsquo;s framework emphasises the importance of considering context at all stages of the development, refinement and evaluation of complex interventions. The implementation of such interventions is influenced by the context in which they are situated, and vice versa, potentially causing variation in delivery across different contexts[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. This context-dependent nature can, therefore, significantly influence implementation success or failure[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. However, recent critiques highlight that context is often treated as a static backdrop in health research, rather than as having a complex and dynamic interplay with the intervention, thereby potentially obscuring its influence[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Developing a deeper understanding of the interaction between context and intervention is crucial to theorising the mechanisms through which outcomes emerge and clarifying why implementation varies in practice.\u003c/p\u003e\u003cp\u003eTo better understand how context influences the 6MR, we explored its interactions across different service settings. By identifying recurring patterns of interaction, we aim to inform future implementation efforts and contribute to the development of a programme theory that reflects the complexity of long-term stroke care.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eDesign\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study used a multiple case study design and forms part of a mixed methods project (BE MoRe: Exploring the Benefits and Expectations of the 6-Month Review for Stroke Survivors). The quantitative phase has been reported elsewhere[7]. Adopting a pragmatic stance, we assert that context-dependent knowledge is constructed through the experiences of key stakeholders. Accordingly, we sought to develop contextually grounded, actionable insights to inform future implementation.\u003c/p\u003e\n\u003cp\u003eCase study design aligned with the aims of the project, given the likelihood of uncovering multiple, interacting contextual factors that may shape the intervention. We drew on Merriam\u0026rsquo;s approach [14] which supports a pragmatic stance by encouraging methodological flexibility and an emphasis on generating practical, context-sensitive understanding across multiple real-world cases[15,16].\u003c/p\u003e\n\u003cp\u003eThe 6MR was observed and analysed through the lens of complexity theory, viewing it as a complex adaptive system. This approach enabled a deep understanding of the dynamic relationships within each system[17], and prevented the urge to explain findings through simplistic, linear patterns of causation[18]. Nieuwenhuijze et al\u0026rsquo;s[19] taxonomy of the key components of complexity theory relevant to healthcare (see \u003cstrong\u003eTable 1\u003c/strong\u003e) was used to frame the analysis and interpret the findings.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1. Components of complex systems\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo situate the 6MR within its context and develop an understanding of the interaction between the two, a definition of context was required that accounts for this dynamic relationship. As such, the following definition by Pfadenhauer et al[20] was used: \u0026ldquo;\u003cem\u003eContext reflects a set of characteristics and circumstances that consist of active and unique factors. As such, context is not a backdrop for implementation, but interacts, influences, modifies and facilitates or constrains the intervention and its implementation\u0026rdquo;.\u003c/em\u003e Alongside this definition, the Context and Implementation of Complex Interventions (CICI) framework[20] was used as a determinant framework to ensure the consideration of a wide range of contextual factors that may interact with the 6MR. The CICI framework considers context at the micro, meso and macro levels across seven key domains: geographical, epidemiological, socio-cultural, socio-economic, ethical, legal, and political. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe CICI framework was used to categorise contextual features, while the complexity taxonomy provided an interpretive lens to examine dynamic interactions. Given the significance of context and complexity within our case study approach, this article has been structured in accordance with the TRIPLE C reporting principles[21].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePositionality\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe lead researcher (RH), a physiotherapist and PhD candidate, brings extensive experience across the stroke pathway including clinical leadership and commissioning of 6MR services. This practice-informed perspective enriched the study but may have introduced bias in favour of clinically led models. Familiarity with system failings within stroke pathways may also have made some observations seem \u0026lsquo;normal\u0026rsquo; or uneventful, leading to unintentionally underemphasising their importance during analysis. Reflexivity was maintained through journaling and regular analytical debriefing meetings with co-authors. The research team included two clinical academic physiotherapists (SA and LC) with expertise in stroke rehabilitation and implementation science research, and a social scientist (DG) with expertise in qualitative methods, providing complementary perspectives and balanced interpretation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCase Selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThree sites were purposively selected from the quantitative stage, primarily to reflect the difference in provider organisations. This was identified to be a key factor in the observed variation in practice[7]. Sites were also chosen to provide diversity in terms of selected contextual factors (i.e. geographical location, level of rurality, and level of deprivation) and system variances (i.e. service delivery method, professionals involved, embedded nature of the service within the stroke pathway).\u003c/p\u003e\n\u003cp\u003eMerriam[14] stresses the importance of clearly defining the boundaries of a case to ensure methodological rigour and coherence. Accordingly, each case was defined as a service responsible for delivering 6MRs, either as its sole function or as part of an integrated service. The boundaries of each case included the direct interactions with other services and the broader influence of contextual factors at a micro, meso and macro level. Bounding each case in this way enabled a focussed yet contextually rich exploration of how systemic and contextual dynamics shape the functioning of the 6MR.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRecruitment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThree key stakeholder groups were included at each site: \u003cem\u003eService Users\u0026nbsp;\u003c/em\u003e(stroke survivors and their carers or family members); \u003cem\u003eService Providers\u0026nbsp;\u003c/em\u003e(staff involved in directly delivering the 6MR); and \u003cem\u003eService Influencers\u0026nbsp;\u003c/em\u003e(individuals who may have some influence over the delivery of the 6MR such as commissioners, managers, and regional leaders in stroke care). The intended sample size at each site was pragmatically determined as six to eight \u003cem\u003eService Users\u003c/em\u003e, two to three \u003cem\u003eService Providers\u003c/em\u003e, and two to three \u003cem\u003eService Influencers\u003c/em\u003e. However, the final sample size was ultimately a reflexive decision by the research team, based on when sufficient data had been gathered to fully capture the nuances of each case[22].\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eService Users\u0026nbsp;\u003c/em\u003ewere included if they were stroke survivors aged 18 or over, or the carer or family member of a stroke survivor, who received a 6MR within the previous three months. Efforts were made to ensure the inclusion of individuals with communication and cognitive difficulties. Measures included allowing extra time during interviews, providing study-related materials in a communication-accessible format when required, and involving family members to support communication or memory when appropriate. However, stroke survivors who were unable to provide their consent or did not have a consistent method of communication were excluded. \u003cem\u003eService Users\u0026nbsp;\u003c/em\u003ewere provided with the study information sheet and invited to take part in the study by the \u003cem\u003eService Providers\u003c/em\u003e at each site. Interested participants either contacted the research team directly or consented to be approached by the lead researcher.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eService Providers\u0026nbsp;\u003c/em\u003ewere purposively approached during initial conversations about using their site as a case study. It was important that agreement was secured before site set-up to ensure it would be possible to collect sufficient data.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eService Influencers\u0026nbsp;\u003c/em\u003ewere also purposively approached during the course of each case study, and were selected based on the specific context and nuances of each case.\u003c/p\u003e\n\u003cp\u003eAll consent procedures were undertaken by the lead researcher either face-to-face or, when conducted virtually, using a verbal consent procedure.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn keeping with robust case study methods, data were collected from multiple sources, including semi-structured interviews, observations, and service-related documents. Doing so allowed a richer understanding of each case through the triangulation of information. All data were collected between June 2024 and April 2025.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSemi-Structured Interviews\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWithin each case, the lead researcher conducted all semi-structured interviews, either virtually or in participants\u0026rsquo; homes. Different interview schedules were developed for each stakeholder group, with the content and structure guided by the CICI framework. All interviews were audio-recorded, transcribed verbatim and anonymised.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eObservations\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNon-participatory observations were undertaken by the lead researcher using a semi-structured data collection sheet guided by the CICI framework. Observations included telephone and virtual appointments, and face-to-face reviews taking place in either a clinic setting, the stroke survivor\u0026rsquo;s home, or within a care home. All observations were done with signed consent from the \u003cem\u003eService Provider\u003c/em\u003e and with verbal consent of the stroke survivor. No personal data were collected; instead, the focus of the observations was on the content of the review, the interactions between individuals during the appointment, and the outcomes of the review process. All individuals involved were informed that they could ask the lead researcher to leave at any point, without needing to provide a reason.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDocuments\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDocuments related to each case were also included in the analysis, acquired either from the \u003cem\u003eService Providers\u0026nbsp;\u003c/em\u003eat each site, internet searches, or freely accessible government databases. Documentary data included: service specifications; demographic data of the local population; Sentinel Stroke National Audit Programme (SSNAP) data; service evaluations; and service-related paperwork (e.g. letters/information to \u003cem\u003eService Users\u003c/em\u003e, data collection tools).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data were uploaded to NVivo 14 for analysis. Principles of reflexive thematic analysis were drawn upon to manage coding decisions and guide the generation of themes/patterns[23]. Regular reflexive meetings were held with the research team at all stages of analysis to critically question and refine analytical decisions. Data were coded by the lead researcher independently using a combined deductive and inductive approach. Deductively, data were coded to contextual factors using the CICI framework and to the complexity taxonomy. Inductively, new codes were developed as repeated patterns were identified in the data. In the initial phase of analysis, all codes were used to build a rich description of each case. Analysis progressed iteratively across cases, with insights from one informing interpretation in others, rather than interpreting each case sequentially.\u003c/p\u003e\n\u003cp\u003eFollowing the descriptive analysis, codes were reviewed across cases to identify contextual subdomains within the CICI framework. The influence that each subdomain had on each case was mapped at the micro, meso and macro levels, when sufficient data was available. These insights were then synthesised across cases using a complexity lens to guide the interpretation and deepen the understanding of how contextual factors interacted to shape implementation, processes and outcomes.\u003c/p\u003e\n\u003cp\u003eFinally, patterns were explored within the data to look for key interactions between context and the 6MR intervention. This was done iteratively by repeatedly checking interpretations against the raw data within each case and refining these concepts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Considerations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe lead researcher is an active clinician who is trained in assessing capacity to obtain valid consent in stroke patients. Participants with communication or cognitive problems were assessed individually, with appropriate adjustments made to support understanding and decision-making.\u003c/p\u003e\n\u003cp\u003eAnonymity was preserved through the removal of identifiable job titles and the redaction of location-specific details in quotes. Each participant was assigned a unique code that indicating their site and stakeholder category. For example, participant \u003cem\u003eBP1\u0026nbsp;\u003c/em\u003ewas a \u003cem\u003eService Provider\u003c/em\u003e from Site B, while \u003cem\u003eCU1\u003c/em\u003e was a \u003cem\u003eService User\u0026nbsp;\u003c/em\u003efrom Site C.\u003c/p\u003e\n\u003cp\u003eEthical approval for the study was obtained from the Health Research Authority (REC Reference: 24/WA/0059).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eService Characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e summarises the characteristics of each service and the demographic context of the areas they serve. Detailed descriptions of each case study can be found in \u003cstrong\u003eAdditional file 1\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Comparison of service characteristics and catchment areas across case studies\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e highlights key differences across the three services, illustrating the heterogeneity of models implemented in practice. The varied geographical and demographic profiles of the areas served also reflect a diverse set of contextual factors, creating distinct conditions that underpin and influence how each service operates.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Sources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAcross the three cases, data were collected from 36 participants, 26 service-related documents, and 17 hours of non-participatory observations. Participants comprised 22 \u003cem\u003eService Users\u0026nbsp;\u003c/em\u003e(19 stroke survivors and three family members), eight \u003cem\u003eService Providers\u0026nbsp;\u003c/em\u003e(four stroke specialist nurses, two therapists, and two third sector stroke co-ordinators), and six \u003cem\u003eService Influencers\u003c/em\u003e. To preserve anonymity, the specific roles of \u003cem\u003eService Influencers\u0026nbsp;\u003c/em\u003eare not detailed; however, all held regional leadership positions and had influence over the implementation of stroke services (via commissioning, strategic or managerial oversight).\u003c/p\u003e\n\u003cp\u003eThe documents set comprised one service specification, three demographic reports, three SSNAP reports, two service evaluations, and 16 service-related documents. The latter category consisted of nine pre-review information documents, two data collection documents, and five documents related to post-review actions. As Site B used a single practitioner who performed only telephone appointments observation totalled approximately three hours. The rest of the time was split evenly between Sites A and C. Observations added contextual depth, clarified the different approaches used, and confirmed consistency between stated and observed practices.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis of Contextual Factors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCross-case analysis of context using the CICI framework identified 12 subdomains: Geographical (Rurality, Transport Infrastructure); Epidemiological (Disease burden, Age); Socio-Cultural (Language/Culture, Values); Socio-Economic (Affordability, Funding); Ethical (Autonomy); Legal (National audit); and Political (Power distribution, Primary care). The impact of each of these factors within each case is mapped in \u003cstrong\u003eAdditional file 2\u003c/strong\u003e. The cross-case synthesis of how these contextual factors interacted with the 6MR at each site was informed by the features of complex systems and can also be seen in \u003cstrong\u003eAdditional file 2.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eKey Patterns of Interaction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe cross-case synthesis highlighted five patterns which illustrate the key ways that context and the 6MR interact to shape implementation, service user experience, and outcomes. An overview of these key patterns is provided in \u003cstrong\u003eTable 3,\u0026nbsp;\u003c/strong\u003ewith each pattern explored further in the narrative that follows.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Summary of key interactions\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;[\u003c/strong\u003e\u003cem\u003ePlease note Table 3 is large therefore has been added as an additional file\u003c/em\u003e]\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003e\u003cstrong\u003e\u003cem\u003eAccess is a dynamic negotiation between service design and contextual barriers\u003c/em\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eStroke survivors encountered a range of contextual barriers that shaped their ability to access the 6MR. The interaction between these barriers, and the structure of the service, influences whether, and in what way, individuals choose to engage with the review process.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eImportant barriers included the degree of rurality, the limitations of transport infrastructure, and the associated cost to the individual. Participants emphasised that the decision to access the 6MR was a consideration between these barriers and the perceived benefit of the review. Contextual barriers could be reduced when service users were given options around how the 6MR could be accessed.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026hellip;it\u0026apos;s a bus and then a taxi (laughing). So, it\u0026apos;s, it\u0026apos;s easier if it\u0026apos;s just straight questions over the phone. - BU5\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWhen choice of access wasn\u0026rsquo;t available, services saw a reduction in uptake of the review. When choice was offered, service users considered the aforementioned barriers alongside other factors, such as their culture and values, to choose the most suitable method. However, some service users opted for ease rather than selecting a method that most complemented their specific needs. In such instances, needs may be missed. Accessibility appeared to work most effectively, from a user experience perspective, when decisions around access were shared between the stroke survivor and the service provider.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026hellip;I suppose that that is a mixture of my decision making and theirs. So, if I\u0026apos;ve read their notes, and the person is elderly, hard of hearing, maybe got issues around... additional issues, you know, whatever those issues are, but issues which are going to make a phone conversation difficult. \u0026ndash; CP2\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAccess to the service was also hindered by barriers inherent in its structure. This was particularly evident for individuals who spoke languages other than English, and in some instances, certain patient groups were excluded without clear justification.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOur particular service doesn\u0026apos;t work with people in nursing homes. So, we are not commissioned to do that. \u0026ndash; CP2\u003c/em\u003e\u003c/p\u003e\n\u003col start=\"2\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u003cem\u003eEquitable service provision requires proactive adaptation\u003c/em\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe heterogeneity of impairments faced by stroke survivors necessitates an individualised approach to care. Consequently, a \u0026lsquo;one-size-fits-all\u0026rsquo; approach to 6MR provision can leave service users feeling excluded or perceiving the review as irrelevant to their needs.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eI think, given that each patient is so different and their needs are so different, I think there needs to be flexibility, like, in appointment times and things to be able to do that because it shouldn\u0026apos;t be about just ticking a box and saying it\u0026apos;s been delivered. It\u0026apos;s supposed to be meaningful and add value to the individual \u0026ndash; BI1\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eKey contextual factors that risked individuals experiencing inequitable service provision include their level of disease burden, their culture and language, and their age. In these contexts, some providers adapt their service to lessen health inequalities. Examples of this include using proxies or interpreters to support the review, or automatically booking face-to-face appointments for stroke survivors who may struggle with the process of organising their 6MR appointment (e.g. those with communication and cognition difficulties).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026hellip;we do an optional now, but, of course, it doesn\u0026apos;t pick up in people with aphasia or people with just difficulties, you know, all of those, or people that just are very old. So, what we\u0026apos;ve done recently is we\u0026apos;ve decided we\u0026apos;re just going to offer appointments to care homes and residential homes or people that we know from ESD [the early supported discharge pathway] will struggle\u0026hellip; - AP3\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWhen services were unaware of a stroke survivor\u0026rsquo;s specific needs, or aware but unable to adapt their provision accordingly, access and support were compromised. To facilitate effective adaptation, services needed to have a degree of knowledge about the individuals accessing their service. When services were embedded within the pathway this was easier to achieve.\u0026nbsp;\u003c/p\u003e\n\u003col start=\"3\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u003cem\u003eHidden needs stay hidden unless actively unmasked\u003c/em\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eA key role of the 6MR is to identify unmet needs. However, in certain contexts, needs are stubbornly hidden and do not emerge passively; their visibility is reliant on the structure of the service and the skills of the providers. Based on the experiences of participants, this phenomenon was particularly relevant where individuals spoke a language other than English, had specific cultural needs, or lacked capacity to advocate for themselves. When services are rigid and unable to adapt to these micro-level contexts, the ability for individuals to access the 6MR may be inhibited, or their experience degraded, as unmet needs may be missed. Examples of this include non-English speakers not understanding paperwork they have been sent, or stroke survivors in nursing homes having their 6MR completed via a proxy rather than including them.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSo, we realised quite quickly it wasn\u0026apos;t working and obviously they are missing out and they shouldn\u0026apos;t have to miss out just because they can\u0026apos;t speak the language - BP2\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBecause I think we have as well, previously in the past, done them with the nursing home staff over the phone. And so, then it\u0026apos;s a second hand. - BP1\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSome participants expressed core values that influenced their interaction with the 6MR. They reported: not wanting to waste the time of the provider; deferring to the provider\u0026rsquo;s opinion; and feeling it was sometimes inappropriate to discuss certain needs (e.g. sexual relations), especially during phone consultations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eThat it wasn\u0026apos;t face-to-face, it seemed a bit strange, but I was told that\u0026apos;s how it\u0026apos;s going to be. It\u0026apos;s alright. - BU2\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn such instances, needs were withheld by the stroke survivor. Some also expressed dissatisfaction with phone-based reviews if it was perceived to hinder the provider\u0026rsquo;s ability to make an informed decision.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026hellip;[the provider said] \u0026ldquo;you could go to this group or that group, providing we could get you transport\u0026rdquo;, which I can\u0026apos;t do at the moment. I can\u0026apos;t go to any groups because I can\u0026apos;t go anywhere. - BU3\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWhen services are cognisant to the diversity of sociodemographic factors and able to adapt the structure of the 6MR in response, the needs of stroke survivors are more likely to be unmasked. Examples include: \u0026nbsp;establishing a dedicated clinic aligned to the language and cultural needs of specific communities; the effective and consistent use of translator services; the provision of information prior to the 6MR to explain its purpose and to highlight unmet needs the stroke survivor could consider; and the appropriate involvement of family members to support the identification of need. This unmasking seemed to work most effectively when the provider\u0026apos;s knowledge and skills enabled a flexible approach, creating a space where the stroke survivor felt able to discuss issues freely.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026hellip;experience has given us that freedom to go with the flow and allow the patient to lead the session rather than the clinician leading the session, which is just, yeah, just experience isn\u0026rsquo;t it? And kind of not feeling out of depth when they throw you curve balls! \u0026ndash; AP1\u003c/em\u003e\u003c/p\u003e\n\u003col start=\"4\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u003cem\u003eSystem levers may trigger unpredictable consequences\u003c/em\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eSome system-level contexts may act as levers for change in efforts to raise the priority of the 6MR. Examples of this include the national stroke audit and financial incentives linked to performance. However, these levers can inadvertently produce perverse incentives, where efforts to optimise performance metrics may occur at the expense of person-centred care. This appeared evident in the increased uptake of reviews in response to financial incentives in 2019, which was coupled with an increase in telephone appointments. This concept of perverse incentives is also exemplified by services using SSNAP reporting requirements to guide the timing of the review, working within the boundaries of what is required, rather than focusing on what the individual actually needs.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e...it is obviously a priority, but I know if we are very busy that we\u0026apos;ve got a three-month window [on SSNAP]. So, they do... we have put them on hold for a month and then followed up. - BP3\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAll sites were observed to collect data required for SSNAP but made limited use of it, collecting information primarily for audit compliance rather than to inform or enhance the review process.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIt probably does feel a little bit like a tick box in some circumstances that we\u0026apos;re just doing it because we\u0026apos;re told that it should be done. \u0026ndash; BP2\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe degree of managerial oversight was another lever that could influence the focus of the review process. Higher levels of managerial scrutiny seemed to trigger the service to focus on efficiency and cost-effectiveness, though this could lead to a loss in patient-centred care. Conversely, a lack of oversight enabled providers greater freedom to adhere to their patient-driven ethos.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eI suppose that we kind of can do our own thing that we\u0026rsquo;re not tied to delivering it in a certain way and they\u0026rsquo;ve sort of left us to be fairly free as long as we delivery it. - AP1\u003c/em\u003e\u003c/p\u003e\n\u003col start=\"5\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u003cem\u003eOutcomes are shaped by interdependence with the wider system\u003c/em\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe success of the 6MR relies on its interactions with the wider care system. When the 6MR was embedded within an integrated pathway, information flowed more easily, improving continuity and responsiveness. This was evident in several ways. Firstly, integrated computer systems enabled providers to have the stroke survivor\u0026rsquo;s history at their fingertips. This gave the stroke survivor reassurance and stopped the provider from needing to repeat aspects of the review.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eI can see if they\u0026apos;ve had a community therapist involved, so I can do some reading back and have an idea of what\u0026apos;s happened. \u0026ndash; AP2\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSecondly, when 6MRs were completed by the same team who delivered their rehabilitation, there was a belief that stroke survivors\u0026rsquo; needs had already been met and, therefore, the 6MR could be streamlined.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eI think that quite often we give them such a good service from an ESD point of view that they don\u0026apos;t tend to want a stroke follow-up because they feel their needs are met, which is good. \u0026ndash; BP3\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWhen services were not embedded, there was a need to ensure that communication within the wider system was optimised. Sometimes this required a proactive approach to build connections across organisations.\u003c/p\u003e\n\u003cp\u003e\u0026hellip;we have had to build those relationships and it\u0026apos;s not always been easy. \u0026ndash; CP1\u003c/p\u003e\n\u003cp\u003eBuilding links with other services helped 6MR providers stay informed about incoming stroke survivors and tailor their approach accordingly. These links also supported awareness of local services and facilitated easier onward referrals.\u003c/p\u003e\n\u003cp\u003eThe structure of the 6MR was further shaped by how the service interacted with primary care. In some services, providers undertook additional tasks beyond the review itself to ease the burden on GPs and improve the experience for stroke survivors. In contrast, others (often constrained by financial or staffing limitations) took a more minimalist approach, handing responsibility for follow-up actions to the GP. This happened even when providers questioned whether the additional work may be too much for the GPs to handle.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eWe would put it back to the GP to make the referral. So, say they needed, I don\u0026apos;t know, community physio or something. We would ask the GP to do that rather than do it ourselves. And I guess there is that worry that will that actually get done? \u0026ndash; BP\u003c/em\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur study took a novel approach to investigate the interactions between the 6MR and its context. We found that context is not just the backdrop to the setting of the 6MR. Instead, it has a dynamic interplay with the service, the intervention, and the individuals involved, and it does so across multiple nested levels within the complex system. These interactions were observed to shape stroke survivors\u0026rsquo; experience within the service and influence providers\u0026rsquo; delivery of the 6MR. Understanding the potential mechanisms behind these dynamics requires attention to the layered and complex nature of health systems.\u003c/p\u003e\n\u003cp\u003eContextual factors identified in this study influenced how stroke survivors interacted with the 6MR and, thereby, influenced their outcomes. This was evident in terms of how they accessed the review, what needs they highlighted, how services were set up to deal with these needs, and how those needs were interpreted and acted upon by providers. The diversity and unpredictability of needs expressed further confirms the complexity of the 6MR. Ensuring that these needs are identified requires a personalised and adaptable approach. When providers were better informed about the person\u0026rsquo;s individual circumstances, they were able to make adaptations to their provision of care to optimise the accessibility and experience for the stroke survivor.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile this flexibility appeared to increase the likelihood that unmet needs would be surfaced and addressed, outcomes were non-linear and some barriers persisted despite provider efforts. This aspect was especially important for marginalised groups who might need additional support to access the 6MR. When this flexibility was lacking, the chance of these groups experiencing inequity increased. This was evident in cases of stroke survivors who spoke languages other than English or care home residents having their review completed via a proxy. As such, our findings suggest that the standardisation of reviews may inadvertently undermine equity. Without a model that can adapt to support individual need and contextual differences, the 6MR risks becoming a procedural formality, one that records symptoms rather than enabling recovery.\u003c/p\u003e\n\u003cp\u003eAt the meso level, contextual factors shaped how organisations interpreted and operationalised the 6MR, influencing both the structure of services and their underlying ethos. The variation in service models observed nationally[7] can be understood as a response to the differing combinations of contextual factors faced by providers in each locality. Self-organisation in response to context is a feature of complex systems and variation of this kind should be embraced[24]. Doing so enables the emergence of service models that are responsive to the needs of the local population, and that operate optimally within their available resources[25].\u003c/p\u003e\n\u003cp\u003eHowever, context may also drive services to adopt models that could negatively impact clinical outcomes. In our study, this was shaped by funding constraints and the distribution of decision-making authority. All sites had to adapt service provision in response to limited resources, but the extent to which they could innovate depended on the decision-making structure within the service. A clear example of this dynamic can be seen at Site B where an increase in demand and limited resources triggered a strong organisational focus on efficiency. In this case, the pursuit of efficiency using a telephone model of delivery, while necessary, appeared to be at the expense of patient experience and needs may be missed. Levesque and Sutherland[26] highlight that strong performance in one measurement domain may cause detrimental effects in another; as such, a focus on efficiency to meet targets and manage demand may come at the expense of overall quality.\u003c/p\u003e\n\u003cp\u003eOur study found that macro-level levers, while effective in driving national uptake of the 6MR, may have negative effects on the implementation of services and the experience of stroke survivors. This finding is not unique; examples of other national improvement initiatives report similar unintended consequences[27,28]. Mannion and Braithwaite[29] highlight a number of these potential consequences, including: \u003cem\u003etunnel vision\u0026nbsp;\u003c/em\u003e(focusing primarily on the elements of measurement); \u003cem\u003egaming\u003c/em\u003e (adopting altered behaviours to gain a reporting advantage); and \u003cem\u003emeasurement fixation\u003c/em\u003e (emphasising data collection over the spirit of the measure). With regards to the 6MR, these adverse consequences were mirrored in the propensity for providers to spend time collecting data that they did not use; altering the timing of the review to optimise performance measures; and the use of proxies to complete reviews for marginalised groups. Providers strive for a person-centred ethos but are sometimes necessitated to act in such ways as to abide by measurement requirements, even when this necessity doesn\u0026rsquo;t fit comfortably into their local context and service delivery. These findings suggest the need for future measures of performance to move beyond procedural compliance and instead focus on tangible outcomes that are meaningful, equitable, and valued by service users.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStrengths and limitations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is strengthened by the use of theory to underpin both data collection and analysis. The CICI framework enabled consideration of a broad range of factors across nested contextual levels. This ensured that context was not narrowly conceptualised as only the sociodemographic factors of the immediate service environment, but also encompassed the beliefs and motivations of individuals within the system, and the wider influence of macro-level policies and drivers. The use of the complexity taxonomy further strengthened our analysis, allowing deeper insights into the mechanisms of complex systems and ensuring that findings were grounded in the \u0026lsquo;real\u0026rsquo;, rather than idealistic, world.\u003c/p\u003e\n\u003cp\u003eHowever, the study is limited by the extent of data collected. Firstly, our study protocol prevented the inclusion of stroke survivors who declined the 6MR. Their perspectives, especially in relation to their decision-making process, could have provided further insight into how context influences uptake and highlighted factors not considered by those who undertook the review. Secondly, the limited timeframe for data collection constrained the ability to observe how services responded to evolving contextual factors. A longer period of observation at each site would have enabled deeper insight into service adaption over time. Instead, we relied on service providers\u0026rsquo; recall of historical events which may be prone to bias. Longitudinal engagement could also have provided insight into how stroke survivors\u0026rsquo; attitudes towards the 6MR changed over time and how context continued to influence their outcomes.\u003c/p\u003e\n\u003cp\u003eFinally, in the context-dependent reality of complex systems (characterised by non-linearity, emergence, and unpredictability) the concept of universally generalisable findings becomes problematic, regardless of the research method used. Because the 6MR is shaped so strongly by context, the conditions that give rise to findings in one setting are unlikely to be fully replicated elsewhere. The value of case study research in complex systems lies not in producing generalisable claims in the traditional sense, but in its ability to uncover underlying mechanisms, identify recurring patterns, and generate context-sensitive insights. These can inform understanding of how similar dynamics might unfold in other settings, and may be used to support decision-making for current, and future, 6MR services.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study has highlighted key patterns of interaction between the 6MR and its context, observed through the lens of complexity theory, across three 6MR services. These dynamic interactions shape how stroke survivors experience, and how providers deliver, the 6MR. The complexity and multifaceted nature of interacting elements of context make prediction of outcomes inherently uncertain. It is the unique mixture of contextual factors, combined with how the 6MR service responds to said context, that influences whether the intervention will be successful or not. However, recurring patterns of interactions identified in this study offer valuable insights that may guide reflection and consideration in other 6MR services, aiding their development or reconfiguration. These findings contribute to the development of programme theory by clarifying how contextual factors shape the delivery and function of 6MR services. This evolving theory will support the generation of actionable recommendations, which will be explored in future work.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e6MR \u0026ndash; six-month review\u003c/p\u003e\n\u003cp\u003eCICI \u0026ndash; Context and Implementation of Complex Interventions\u003c/p\u003e\n\u003cp\u003eESD \u0026ndash; early supported discharge\u003c/p\u003e\n\u003cp\u003eGM-SAT\u0026copy; \u0026ndash; Greater Manchester Stroke Assessment Tool\u003c/p\u003e\n\u003cp\u003eNHS \u0026ndash; National Health Service\u003c/p\u003e\n\u003cp\u003eONS \u0026ndash; Office for National Statistics\u003c/p\u003e\n\u003cp\u003eSSNAP \u0026ndash; Sentinel Stroke National Audit Programme\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval for the study was obtained from the Health Research Authority (REC Reference: 24/WA/0059). Informed consent was obtained from all participants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data generated or analysed during this study are included in this published article [and its supplementary information files].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRH (Clinical Doctoral Research Fellow, NIHR 302163) is funded by the NIHR for this research project. The views expressed in this publication are those of the authors and not necessarily those of the NIHR, NHS or the UK Department of Health and Social Care.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRH contributed to the study conceptualisation, design, data acquisition and analysis, and drafting of the manuscript. SA contributed to the study design, data analysis, and editing of the manuscript. DG contributed to the study design, data analysis, and editing of the manuscript. LC contributed to the study conceptualisation, design, data analysis, and editing of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge study participants who provided their time to take part in the study. We are also extremely grateful to the three study sites for allowing access to their services and supporting the recruitment of participants.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eCrichton SL, Bray BD, McKevitt C, Rudd AG, Wolfe CD. Patient outcomes up to 15 years after stroke: survival, disability, quality of life, cognition and mental health. Journal of Neurology, Neurosurgery \u0026amp; Psychiatry. 2016 Oct 1;87(10):1091-8.\u003c/li\u003e\n\u003cli\u003eDe Wit L, Theuns P, Dejaeger E, Devos S, Gantenbein AR, Kerckhofs E, Schuback B, Schupp W, Putman K. Long-term impact of stroke on patients\u0026rsquo; health-related quality of life. Disability and rehabilitation. 2017 Jul 3;39(14):1435-40.\u003c/li\u003e\n\u003cli\u003eKing D, Wittenberg R, Patel A, Quayyum Z, Berdunov V, Knapp M. The future incidence, prevalence and costs of stroke in the UK. Age and ageing. 2020 Feb 27;49(2):277-82.\u003c/li\u003e\n\u003cli\u003eGallacher KI, Jani BD, Hanlon P, Nicholl BI, Mair FS. Multimorbidity in stroke. Stroke. 2019 Jul;50(7):1919-26.\u003c/li\u003e\n\u003cli\u003eEvans NR, Todd OM, Minhas JS, Fearon P, Harston GW, Mant J, Mead G, Hewitt J, Quinn TJ, Warburton EA. Frailty and cerebrovascular disease: concepts and clinical implications for stroke medicine. International Journal of Stroke. 2022 Mar;17(3):251-9.\u003c/li\u003e\n\u003cli\u003eIntercollegiate Stroke Working Party. National Clinical Guideline for Stroke for the United Kingdom and Ireland [Internet]. 2023 [cited 2025 Jul 16]. Available from: https://www.strokeguideline.org/\u003c/li\u003e\n\u003cli\u003eHolmes R, Ackerley S, Fisher RJ, Connell LA. Exploring variation in the six-month review for stroke survivors: a national survey of current practice in England. BMC Health Services Research. 2025 Jan 28;25(1):159.\u003c/li\u003e\n\u003cli\u003eSentinel Stroke National Audit Programme (SSNAP). SSNAP Annual Report 2024 [Internet]. London: Healthcare Quality Improvement Partnership; 2024 [cited 2025 Jul 16]. Available from: https://www.hqip.org.uk/wp-content/uploads/2024/11/Ref.-466-SSNAP-Annual-report-2024-FINAL.pdf\u003c/li\u003e\n\u003cli\u003eRutter H, Savona N, Glonti K, Bibby J, Cummins S, Finegood DT, Greaves F, Harper L, Hawe P, Moore L, Petticrew M. The need for a complex systems model of evidence for public health. The lancet. 2017 Dec 9;390(10112):2602-4.\u003c/li\u003e\n\u003cli\u003eSkivington K, Matthews L, Simpson SA, Craig P, Baird J, Blazeby JM, Boyd KA, Craig N, French DP, McIntosh E, Petticrew M. A new framework for developing and evaluating complex interventions: update of Medical Research Council guidance. bmj. 2021 Sep 30;374.\u003c/li\u003e\n\u003cli\u003eMoore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, Moore L, O\u0026rsquo;Cathain A, Tinati T, Wight D, Baird J. Process evaluation of complex interventions: Medical Research Council guidance. bmj. 2015 Mar 19;350.\u003c/li\u003e\n\u003cli\u003eGrant A, Bugge C, Wells M. Designing process evaluations using case study to explore the context of complex interventions evaluated in trials. Trials. 2020 Nov 27;21(1):982.\u003c/li\u003e\n\u003cli\u003eMurdoch J, Paparini S, Papoutsi C, James H, Greenhalgh T, Shaw SE. Mobilising context as complex and dynamic in evaluations of complex health interventions. BMC health services research. 2023 Dec 18;23(1):1430.\u003c/li\u003e\n\u003cli\u003eMerriam SB. Qualitative research and case study applications in education. 2nd ed. San Francisco: Jossey-Bass Publishers; 1998.\u003c/li\u003e\n\u003cli\u003eYazan B. Three approaches to case study methods in education: Yin, Merriam, and Stake. The Qualitative Report. 2015 Feb 1;20(2):134-53.\u003c/li\u003e\n\u003cli\u003eHarrison H, Birks M, Franklin R, Mills J. Case Study Research: Foundations and Methodological Orientations. FQS [Internet]. 2017 Jan. 24 [cited 2025 Jul. 16];18(1). Available from: https://www.qualitative-research.net/index.php/fqs/article/view/2655\u003c/li\u003e\n\u003cli\u003eBraithwaite J, Churruca K, Long JC, Ellis LA, Herkes J. When complexity science meets implementation science: a theoretical and empirical analysis of systems change. BMC medicine. 2018 Apr 30;16(1):63.\u003c/li\u003e\n\u003cli\u003eTsoukas H. Don\u0026apos;t simplify, complexify: From disjunctive to conjunctive theorizing in organization and management studies. Journal of management studies. 2017 Mar;54(2):132-53.\u003c/li\u003e\n\u003cli\u003eNieuwenhuijze M, Downe S, Gottfre\u0026eth;sd\u0026oacute;ttir H, Rijnders M, du Preez A, Rebelo PV. Taxonomy for complexity theory in the context of maternity care. Midwifery. 2015 Sep 1;31(9):834-43.\u003c/li\u003e\n\u003cli\u003ePfadenhauer LM, Gerhardus A, Mozygemba K, Lysdahl KB, Booth A, Hofmann B, Wahlster P, Polus S, Burns J, Brereton L, Rehfuess E. Making sense of complexity in context and implementation: the Context and Implementation of Complex Interventions (CICI) framework. Implementation science. 2017 Feb 15;12(1):21.\u003c/li\u003e\n\u003cli\u003eShaw SE, Paparini S, Murdoch J, Green J, Greenhalgh T, Hanckel B, James HM, Petticrew M, Wood GW, Papoutsi C. TRIPLE C reporting principles for case study evaluations of the role of context in complex interventions. BMC Medical Research Methodology. 2023 May 13;23(1):115.\u003c/li\u003e\n\u003cli\u003eBraun, V., \u0026amp; Clarke, V. To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. Qualitative research in sport, exercise and health. 2021 Mar 4;13(2):201\u0026ndash;216.\u003c/li\u003e\n\u003cli\u003eBraun V, Clarke V. Reflecting on reflexive thematic analysis. Qualitative research in sport, exercise and health. 2019 Aug 8;11(4):589-97.\u003c/li\u003e\n\u003cli\u003eBraithwaite J, Churruca K, Ellis LA. Can we fix the uber-complexities of healthcare?. Journal of the Royal Society of Medicine. 2017 Oct;110(10):392-4.\u003c/li\u003e\n\u003cli\u003eSutherland K, Levesque JF. Unwarranted clinical variation in health care: definitions and proposal of an analytic framework. Journal of evaluation in clinical practice. 2020 Jun;26(3):687-96.\u003c/li\u003e\n\u003cli\u003eLevesque JF, Sutherland K. Combining patient, clinical and system perspectives in assessing performance in healthcare: an integrated measurement framework. BMC health services research. 2020 Jan 8;20(1):23.\u003c/li\u003e\n\u003cli\u003eMannion R, Davies H, Marshall M. Impact of star performance ratings in English acute hospital trusts. Journal of Health Services Research \u0026amp; Policy. 2005 Jan;10(1):18-24.\u003c/li\u003e\n\u003cli\u003eMcLean G, Sutton M, Guthrie B. Deprivation and quality of primary care services: evidence for persistence of the inverse care law from the UK Quality and Outcomes Framework. Journal of Epidemiology \u0026amp; Community Health. 2006 Nov 1;60(11):917-22.\u003c/li\u003e\n\u003cli\u003eMannion R, Braithwaite J. Unintended consequences of performance measurement in healthcare: 20 salutary lessons from the English National Health Service. Internal medicine journal. 2012 May;42(5):569-74.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Components of complex systems\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eComponent\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDescription\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eInterconnection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eComplex systems have multiple connected elements, and their make-up continually changes. Outcomes emerge from the interactions within, and external to, the system.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOpen \u0026ldquo;Fuzzy\u0026rdquo; boundaries\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eIdeas, information, and people often move freely across boundaries. Individuals can be members of multiple interacting systems at the same time with multiple influential interactions.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInitial conditions (historicism)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eSystem history and initial conditions have a strong influence over future behaviour. New practices that are consistent with historical context may be more readily embraced than changes that dramatically deviate from previous practice.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSelf-organisation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eComplex systems have a tendency to organise spontaneously, in response to contextual influence, without the need for hierarchical control.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSimple rules\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eIndividuals within a complex system follow a series of internalised values or principles, either explicitly or implicitly known, that drive self-organisation and emergent outcomes.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFeedback loops\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eChanges within the system gives rise to feedback that either reinforce the new pattern of behaviour or seek to suppress it.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCo-evolution\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eComplex systems evolve in line with other systems that they are closely linked with or embedded within.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEmergence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eInteractions within a complex system produce outcomes whereby the mechanisms of action are not clearly understood. These outcomes are more than just the sum of their parts such that emergent properties cannot be explained just by exploring the relevant parts of the system.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNon-linearity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eEffects produced within a complex system can be non-proportional or paradoxical. Small changes can lead to dramatic effects; equally, they can fail to produce the intended outcome despite a large, multifaceted intervention.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUnpredictability\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eProcesses and outcomes cannot be known in advance. Accurate understanding of anticipated outcomes is not possible.\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAttractors\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 469px;\"\u003e\n \u003cp\u003eAttractors are the states or patterns that a complex system is drawn to over time. Attractors can dictate how the system behaves whether that be a period of stability following a period of turmoil, repeated cyclical patterns, or chaotic patterns of repeated change.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Comparison of service characteristics and catchment areas across case studies\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eCase A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eCase B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eCase C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eService Characteristics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eProvider Organisation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eCommunity NHS Trust\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eAcute NHS Trust\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eThird-Sector Organisation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eCurrent Staffing Level within 6MR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e3x Stroke Specialist Nurses (part-time within 6MR service)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e1x Therapist (part-time within 6MR service)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e3x Stroke Co-ordinators (part-time within 6MR service)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eMethod of Review\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eFace-to-face in clinic setting and home visits (including care homes) or telephone appointments\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eMajority are telephone appointments (face-to-face in special circumstances)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eHome visits or telephone appointments (option for virtual)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003ePosition of 6MR within Stroke Pathway\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eEmbedded within ESD team\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eEmbedded within ESD team\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eDelivered as part of a third-sector-led stroke support service\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eData collection tool\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eIn-house tool\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eAdapted GM-SAT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eAdapted GM-SAT\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eLocal Area Profile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eRegion\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eEast of England\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eSouth East\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eNorth West\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eLocal population*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e662,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e439,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e328,300\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eRurality (ONS Rural Urban Classification)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eMixed rural and urban: Mostly rural town/fringe surrounding urban city and town\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003ePredominantly urban city and town\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003eMixed urban and rural: Urban city and town with adjacent rural town and fringe areas\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eDeprivation levels (% Households deprived in one or more dimension)*\u003cbr\u003e\u0026nbsp;\u003cbr\u003e\u0026nbsp;NB. National average = 52%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e52%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e44%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e48%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eEthnicity (% White)*\u003cbr\u003e\u0026nbsp;\u003cbr\u003e\u0026nbsp;NB. National average = 81%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e94%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e86%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e95%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e*Data taken from the ONS 2021 census. (ESD \u0026ndash; Early Supported Discharge, a model of home-based, multidisciplinary rehabilitation that enables early discharge from hospital; ONS \u0026ndash; Office for National Statistics; GM-SAT \u0026ndash; Greater Manchester Stroke Assessment Tool\u0026copy;; NHS \u0026ndash; National Health Service).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Summary of key interactions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePlease see Additional file 3.\u003c/em\u003e\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"implementation-science-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"iscm","sideBox":"Learn more about [Implementation Science Communications](https://implementationsciencecomms.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/ISCM/default.aspx","title":"Implementation Science Communications","twitterHandle":"@ImplementSci","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Stroke rehabilitation, Life after stroke, Follow-up care, Implementation science, CICI, Complexity theory, Complex interventions, Context, Health inequities, Multiple case studies","lastPublishedDoi":"10.21203/rs.3.rs-7490223/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7490223/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eThe six-month review is a policy recommendation in the United Kingdom aimed at identifying and addressing the unmet needs of stroke survivors. Differences in the provision of this complex intervention may arise from variations in context. Our study aimed to explore the dynamic interaction between the six-month review and its associated context.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eA qualitative multiple case study was conducted across three contrasting six-month review services in England selected primarily to reflect variation in provider organisation. Data collection included semi-structured interviews with three stakeholder groups (\u003cem\u003eService Users, Service Providers, \u003c/em\u003eand\u003cem\u003e Service Influencers\u003c/em\u003e)\u003cem\u003e,\u003c/em\u003e direct observations of the review process, and service-related documents. Data analysis utilised a combined deductive and inductive approach. Using the Context and Implementation of Complex Interventions framework, contextual interactions were mapped at the micro, meso and macro levels across the three cases. A cross-case synthesis, guided by complexity theory, identified key patterns of interaction between the six-month review and its context, which were summarised narratively.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eData were collected from 36 stakeholders, 17 hours of observation, and 26 service-related documents.\u003cstrong\u003e \u003c/strong\u003eFive key patterns of interaction between the six-month review and its context were identified: (1) Access is a dynamic negotiation between service design and contextual barriers, (2) Equitable service provision requires proactive adaptation , (3) Hidden needs stay hidden unless actively unmasked, (4) System levers may trigger unpredictable consequences, and (5) Outcomes are shaped by interdependence with the wider system.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eThe findings demonstrate the multi-level influence that context has on the implementation and delivery of the six-month review. These contextual interactions shape outcomes in varied, unpredictable, and sometimes unintended ways, reflecting the six-month review’s position within a complex system. The identified patterns of interaction provide insight into the six-month review’s underlying mechanisms and may guide future implementation efforts.\u003c/p\u003e","manuscriptTitle":"Driven by policy, shaped by context: A complexity-informed multiple case study of the six-month review for stroke survivors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-25 16:09:38","doi":"10.21203/rs.3.rs-7490223/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2025-11-12T18:07:31+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-12T06:10:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-05T03:32:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Implementation Science Communications","date":"2025-09-03T05:14:01+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"implementation-science-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"iscm","sideBox":"Learn more about [Implementation Science Communications](https://implementationsciencecomms.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/ISCM/default.aspx","title":"Implementation Science Communications","twitterHandle":"@ImplementSci","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"10cef90b-d274-4632-ab84-2d6cbcb710d3","owner":[],"postedDate":"November 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-11-25T16:09:38+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-25 16:09:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7490223","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7490223","identity":"rs-7490223","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}