Explore the vast range of titles available.

Complexity is much talked about but sub-optimally studied in health services research. Although the significance of the complex system as an analytic lens is increasingly recognised, many researchers are still using methods that assume a closed system in which predictive studies in general, and controlled experiments in particular, are possible and preferred. We argue that in open systems characterised by dynamically changing inter-relationships and tensions, conventional research designs predicated on linearity and predictability must be augmented by the study of how we can best deal with uncertainty, unpredictability and emergent causality. Accordingly, the study of complexity in health services and systems requires new standards of research quality, namely (for example) rich theorising, generative learning, and pragmatic adaptation to changing contexts. This framing of complexity-informed health services research provides a backdrop for a new collection of empirical studies. Each of the initial five papers in this collection illustrates, in different ways, the value of theoretically grounded, methodologically pluralistic, flexible and adaptive study designs. We propose an agenda for future research and invite researchers to contribute to this on-going series.

Many promising technological innovations in health and social care are characterized by nonadoption or abandonment by individuals or by failed attempts to scale up locally, spread distantly, or sustain the innovation long term at the organization or system level. Our objective was to produce an evidence-based, theory-informed, and pragmatic framework to help predict and evaluate the success of a technology-supported health or social care program. The study had 2 parallel components: (1) secondary research (hermeneutic systematic review) to identify key domains, and (2) empirical case studies of technology implementation to explore, test, and refine these domains. We studied 6 technology-supported programs-video outpatient consultations, global positioning system tracking for cognitive impairment, pendant alarm services, remote biomarker monitoring for heart failure, care organizing software, and integrated case management via data sharing-using longitudinal ethnography and action research for up to 3 years across more than 20 organizations. Data were collected at micro level (individual technology users), meso level (organizational processes and systems), and macro level (national policy and wider context). Analysis and synthesis was aided by sociotechnically informed theories of individual, organizational, and system change. The draft framework was shared with colleagues who were introducing or evaluating other technology-supported health or care programs and refined in response to feedback. The literature review identified 28 previous technology implementation frameworks, of which 14 had taken a dynamic systems approach (including 2 integrative reviews of previous work). Our empirical dataset consisted of over 400 hours of ethnographic observation, 165 semistructured interviews, and 200 documents. The final nonadoption, abandonment, scale-up, spread, and sustainability (NASSS) framework included questions in 7 domains: the condition or illness, the technology, the value proposition, the adopter system (comprising professional staff, patient, and lay caregivers), the organization(s), the wider (institutional and societal) context, and the interaction and mutual adaptation between all these domains over time. Our empirical case studies raised a variety of challenges across all 7 domains, each classified as simple (straightforward, predictable, few components), complicated (multiple interacting components or issues), or complex (dynamic, unpredictable, not easily disaggregated into constituent components). Programs characterized by complicatedness proved difficult but not impossible to implement. Those characterized by complexity in multiple NASSS domains rarely, if ever, became mainstreamed. The framework showed promise when applied (both prospectively and retrospectively) to other programs. Subject to further empirical testing, NASSS could be applied across a range of technological innovations in health and social care. It has several potential uses: (1) to inform the design of a new technology; (2) to identify technological solutions that (perhaps despite policy or industry enthusiasm) have a limited chance of achieving large-scale, sustained adoption; (3) to plan the implementation, scale-up, or rollout of a technology program; and (4) to explain and learn from program failures.

Background Failures and partial successes are common in technology-supported innovation programmes in health and social care. Complexity theory can help explain why. Phenomena may be simple (straightforward, predictable, few components), complicated (multiple interacting components or issues) or complex (dynamic, unpredictable, not easily disaggregated into constituent components). The recently published NASSS framework applies this taxonomy to explain N on-adoption or A bandonment of technology by individuals and difficulties achieving S cale-up, S pread and S ustainability. This paper reports the first empirical application of the NASSS framework. Methods Six technology-supported programmes were studied using ethnography and action research for up to 3 years across 20 health and care organisations and 10 national-level bodies. They comprised video outpatient consultations, GPS tracking technology for cognitive impairment, pendant alarm services, remote biomarker monitoring for heart failure, care organising software and integrated case management via data warehousing. Data were collected at three levels: micro (individual technology users), meso (organisational processes and systems) and macro (national policy and wider context). Data analysis and synthesis were guided by socio-technical theories and organised around the seven NASSS domains: (1) the condition or illness, (2) the technology, (3) the value proposition, (4) the adopter system (professional staff, patients and lay carers), (5) the organisation(s), (6) the wider (institutional and societal) system and (7) interaction and mutual adaptation among all these domains over time. Results The study generated more than 400 h of ethnographic observation, 165 semi-structured interviews and 200 documents. The six case studies raised multiple challenges across all seven domains. Complexity was a common feature of all programmes. In particular, individuals’ health and care needs were often complex and hence unpredictable and ‘off algorithm’. Programmes in which multiple domains were complicated proved difficult, slow and expensive to implement. Those in which multiple domains were complex did not become mainstreamed (or, if mainstreamed, did not deliver key intended outputs). Conclusion The NASSS framework helped explain the successes, failures and changing fortunes of this diverse sample of technology-supported programmes. Since failure is often linked to complexity across multiple NASSS domains, further research should systematically address ways to reduce complexity and/or manage programme implementation to take account of it.

Background There is a growing need for methods that acknowledge and successfully capture the dynamic interaction between context and implementation of complex interventions. Case study research has the potential to provide such understanding, enabling in-depth investigation of the particularities of phenomena. However, there is limited guidance on how and when to best use different case study research approaches when evaluating complex interventions. This study aimed to review and synthesise the literature on case study research across relevant disciplines, and determine relevance to the study of contextual influences on complex interventions in health systems and public health research. Methods Systematic meta-narrative review of the literature comprising (i) a scoping review of seminal texts ( n  = 60) on case study methodology and on context, complexity and interventions, (ii) detailed review of empirical literature on case study, context and complex interventions ( n  = 71), and (iii) identifying and reviewing ‘hybrid papers’ ( n  = 8) focused on the merits and challenges of case study in the evaluation of complex interventions. Results We identified four broad (and to some extent overlapping) research traditions, all using case study in a slightly different way and with different goals: 1) developing and testing complex interventions in healthcare; 2) analysing change in organisations; 3) undertaking realist evaluations; 4) studying complex change naturalistically. Each tradition conceptualised context differently—respectively as the backdrop to, or factors impacting on, the intervention; sets of interacting conditions and relationships; circumstances triggering intervention mechanisms; and socially structured practices. Overall, these traditions drew on a small number of case study methodologists and disciplines. Few studies problematised the nature and boundaries of ‘the case’ and ‘context’ or considered the implications of such conceptualisations for methods and knowledge production. Conclusions Case study research on complex interventions in healthcare draws on a number of different research traditions, each with different epistemological and methodological preferences. The approach used and consequences for knowledge produced often remains implicit. This has implications for how researchers, practitioners and decision makers understand, implement and evaluate complex interventions in different settings. Deeper engagement with case study research as a methodology is strongly recommended.

Background The need for better methods for evaluation in health research has been widely recognised. The ‘complexity turn’ has drawn attention to the limitations of relying on causal inference from randomised controlled trials alone for understanding whether, and under which conditions, interventions in complex systems improve health services or the public health, and what mechanisms might link interventions and outcomes. We argue that case study research—currently denigrated as poor evidence—is an under-utilised resource for not only providing evidence about context and transferability, but also for helping strengthen causal inferences when pathways between intervention and effects are likely to be non-linear. Main body Case study research, as an overall approach, is based on in-depth explorations of complex phenomena in their natural, or real-life, settings. Empirical case studies typically enable dynamic understanding of complex challenges and provide evidence about causal mechanisms and the necessary and sufficient conditions (contexts) for intervention implementation and effects. This is essential evidence not just for researchers concerned about internal and external validity, but also research users in policy and practice who need to know what the likely effects of complex programmes or interventions will be in their settings. The health sciences have much to learn from scholarship on case study methodology in the social sciences. However, there are multiple challenges in fully exploiting the potential learning from case study research. First are misconceptions that case study research can only provide exploratory or descriptive evidence. Second, there is little consensus about what a case study is, and considerable diversity in how empirical case studies are conducted and reported. Finally, as case study researchers typically (and appropriately) focus on thick description (that captures contextual detail), it can be challenging to identify the key messages related to intervention evaluation from case study reports. Conclusion Whilst the diversity of published case studies in health services and public health research is rich and productive, we recommend further clarity and specific methodological guidance for those reporting case study research for evaluation audiences.

The rhetoric surrounding clinical artificial intelligence (AI) often exaggerates its effect on real-world care. Limited understanding of the factors that influence its implementation can perpetuate this. In this qualitative systematic review, we aimed to identify key stakeholders, consolidate their perspectives on clinical AI implementation, and characterize the evidence gaps that future qualitative research should target. Ovid-MEDLINE, EBSCO-CINAHL, ACM Digital Library, Science Citation Index-Web of Science, and Scopus were searched for primary qualitative studies on individuals' perspectives on any application of clinical AI worldwide (January 2014-April 2021). The definition of clinical AI includes both rule-based and machine learning-enabled or non-rule-based decision support tools. The language of the reports was not an exclusion criterion. Two independent reviewers performed title, abstract, and full-text screening with a third arbiter of disagreement. Two reviewers assigned the Joanna Briggs Institute 10-point checklist for qualitative research scores for each study. A single reviewer extracted free-text data relevant to clinical AI implementation, noting the stakeholders contributing to each excerpt. The best-fit framework synthesis used the Nonadoption, Abandonment, Scale-up, Spread, and Sustainability (NASSS) framework. To validate the data and improve accessibility, coauthors representing each emergent stakeholder group codeveloped summaries of the factors most relevant to their respective groups. The initial search yielded 4437 deduplicated articles, with 111 (2.5%) eligible for inclusion (median Joanna Briggs Institute 10-point checklist for qualitative research score, 8/10). Five distinct stakeholder groups emerged from the data: health care professionals (HCPs), patients, carers and other members of the public, developers, health care managers and leaders, and regulators or policy makers, contributing 1204 (70%), 196 (11.4%), 133 (7.7%), 129 (7.5%), and 59 (3.4%) of 1721 eligible excerpts, respectively. All stakeholder groups independently identified a breadth of implementation factors, with each producing data that were mapped between 17 and 24 of the 27 adapted Nonadoption, Abandonment, Scale-up, Spread, and Sustainability subdomains. Most of the factors that stakeholders found influential in the implementation of rule-based clinical AI also applied to non-rule-based clinical AI, with the exception of intellectual property, regulation, and sociocultural attitudes. Clinical AI implementation is influenced by many interdependent factors, which are in turn influenced by at least 5 distinct stakeholder groups. This implies that effective research and practice of clinical AI implementation should consider multiple stakeholder perspectives. The current underrepresentation of perspectives from stakeholders other than HCPs in the literature may limit the anticipation and management of the factors that influence successful clinical AI implementation. Future research should not only widen the representation of tools and contexts in qualitative research but also specifically investigate the perspectives of all stakeholder HCPs and emerging aspects of non-rule-based clinical AI implementation. PROSPERO (International Prospective Register of Systematic Reviews) CRD42021256005; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=256005. RR2-10.2196/33145.

Background Primary care has been described as the ‘bedrock’ of the National Health Service (NHS) accounting for approximately 90% of patient contacts but is facing significant challenges. Against a backdrop of a rapidly ageing population with increasingly complex health challenges, policy-makers have encouraged primary care commissioners to increase the usage of data when making commissioning decisions. Purported benefits include cost savings and improved population health. However, research on evidence-based commissioning has concluded that commissioners work in complex environments and that closer attention should be paid to the interplay of contextual factors and evidence use. The aim of this review was to understand how and why primary care commissioners use data to inform their decision making, what outcomes this leads to, and understand what factors or contexts promote and inhibit their usage of data. Methods We developed initial programme theory by identifying barriers and facilitators to using data to inform primary care commissioning based on the findings of an exploratory literature search and discussions with programme implementers. We then located a range of diverse studies by searching seven databases as well as grey literature. Using a realist approach, which has an explanatory rather than a judgemental focus, we identified recurrent patterns of outcomes and their associated contexts and mechanisms related to data usage in primary care commissioning to form context-mechanism-outcome (CMO) configurations. We then developed a revised and refined programme theory. Results Ninety-two studies met the inclusion criteria, informing the development of 30 CMOs. Primary care commissioners work in complex and demanding environments, and the usage of data are promoted and inhibited by a wide range of contexts including specific commissioning activities, commissioners’ perceptions and skillsets, their relationships with external providers of data (analysis), and the characteristics of data themselves. Data are used by commissioners not only as a source of evidence but also as a tool for stimulating commissioning improvements and as a warrant for convincing others about decisions commissioners wish to make. Despite being well-intentioned users of data, commissioners face considerable challenges when trying to use them, and have developed a range of strategies to deal with ‘imperfect’ data. Conclusions There are still considerable barriers to using data in certain contexts. Understanding and addressing these will be key in light of the government’s ongoing commitments to using data to inform policy-making, as well as increasing integrated commissioning.

Background Mental ill-health in health professionals, including doctors, is a global and growing concern. The existing literature on interventions that offer support, advice and/or treatment to sick doctors has not yet been synthesised in a way that considers the complexity and heterogeneity of the interventions, and the many dimensions of the problem. We (1) reviewed interventions to tackle doctors’ and medical students’ mental ill-health and its impacts on the clinical workforce and patient care—drawing on diverse literature sources and engaging iteratively with diverse stakeholder perspectives—and (2) produced recommendations that support the tailoring, implementation, monitoring and evaluation of contextually sensitive strategies to tackle mental ill-health and its impacts. Methods Realist literature review consistent with the RAMESES quality and reporting standards. Sources for inclusion were identified through bibliographic database searches supplemented by purposive searches—resulting also from engagement with stakeholders. Data were extracted from included articles and subjected to realist analysis to identify (i) mechanisms causing mental ill-health in doctors and medical students and relevant contexts or circumstances when these mechanisms were likely to be ‘triggered’ and (ii) ‘guiding principles’ and features underpinning the interventions and recommendations discussed mostly in policy document, reviews and commentaries. Results One hundred seventy-nine records were included. Most were from the USA (45%) and were published since 2009 (74%). The analysis showed that doctors were more likely to experience mental ill-health when they felt isolated or unable to do their job and when they feared repercussions of help-seeking. Healthy staff were necessary for excellent patient care. Interventions emphasising relationships and belonging were more likely to promote wellbeing. Interventions creating a people-focussed working culture, balancing positive/negative performance and acknowledging positive/negative aspects of a medical career helped doctors to thrive. The way that interventions were implemented seemed critically important. Doctors and medical students needed to have confidence in an intervention for the intervention to be effective. Conclusions Successful interventions to tackle doctors’ and students’ mental ill-health are likely to be multidimensional and multilevel and involve multiple stakeholders. Evaluating and improving existing interventions is likely to be more effective than developing new ones. Our evidence synthesis provides a basis on which to do this. Study registration PROSPERO CRD42017069870. Research project webpage http://sites.exeter.ac.uk/cup/

Disseminating innovation across the healthcare system is challenging but potentially achievable through different logics: mechanistic, ecological, and social, say Trisha Greenhalgh and Chrysanthi Papoutsi

Background Artificial intelligence (AI) technologies are expected to “revolutionise” healthcare. However, despite their promises, their integration within healthcare organisations and systems remains limited. The objective of this study is to explore and understand the systemic challenges and implications of their integration in a leading Canadian academic hospital. Methods Semi-structured interviews were conducted with 29 stakeholders concerned by the integration of a large set of AI technologies within the organisation (e.g., managers, clinicians, researchers, patients, technology providers). Data were collected and analysed using the Non-Adoption, Abandonment, Scale-up, Spread, Sustainability (NASSS) framework. Results Among enabling factors and conditions, our findings highlight: a supportive organisational culture and leadership leading to a coherent organisational innovation narrative; mutual trust and transparent communication between senior management and frontline teams; the presence of champions, translators, and boundary spanners for AI able to build bridges and trust; and the capacity to attract technical and clinical talents and expertise. Constraints and barriers include: contrasting definitions of the value of AI technologies and ways to measure such value; lack of real-life and context-based evidence; varying patients’ digital and health literacy capacities; misalignments between organisational dynamics, clinical and administrative processes, infrastructures, and AI technologies; lack of funding mechanisms covering the implementation, adaptation, and expertise required; challenges arising from practice change, new expertise development, and professional identities; lack of official professional, reimbursement, and insurance guidelines; lack of pre- and post-market approval legal and governance frameworks; diversity of the business and financing models for AI technologies; and misalignments between investors’ priorities and the needs and expectations of healthcare organisations and systems. Conclusion Thanks to the multidimensional NASSS framework, this study provides original insights and a detailed learning base for analysing AI technologies in healthcare from a thorough socio-technical perspective. Our findings highlight the importance of considering the complexity characterising healthcare organisations and systems in current efforts to introduce AI technologies within clinical routines. This study adds to the existing literature and can inform decision-making towards a judicious, responsible, and sustainable integration of these technologies in healthcare organisations and systems.