Making sense of evidence in management decisions: the role of research-based knowledge on innovation adoption and implementation in healthcare. study protocol

Health service delivery and organisation as well as clinical practice can be improved through the introduction of novel interventions whose effectiveness is backed by sound evidence. However, the uptake and implementation of innovations in healthcare has often proved challenging and, in some cases, very slow [1, 2]. As a result, research findings are not always translated into changes in clinical and managerial practice. This reality also raises the pressing question of how to spread best practice and implement promising innovations within complex settings such as the UK's National Health Service (NHS), a large, professionalised, and highly politicised organisation. The study makes a timely contribution given the recent report Innovation, Health and Wealth by the Department of Health on accelerating the pace of adoption and diffusion of innovations in the NHS [3].

Our study is theory based and grounded in the practical experience of healthcare managers to provide answers to persistent challenges. The study draws on organisational theory, largely on sensemaking in organisations, and on innovation theory of change. One of the central questions in this latter body of literature that aligns with the scope of our study is as follows: 'Why do innovations not readily spread, even if backed by strong evidence?' There is a growing body of evidence, including empirical cases in healthcare, that argues that the adoption of new ideas is a process far more dynamic and complex than previously suggested. The classic innovation diffusion model of change, which has influenced UK healthcare policy in the last decades, suggests that the adoption of innovations is a rational and linear process, which is conditioned by the interaction of perceived (by the adopter) innovation attributes with the characteristics of adopters and influences of the social system [4]. This early innovation diffusion work has been criticised, however, for taking a simplistic rational view of change, ignoring the complexities of the change process, and also focusing on individuals rather than organisations. Later work in the same tradition has partially addressed this criticism by explicitly considering adoption within organisations [5]. Other innovation and communication studies have departed from the linear model of diffusion [4] to offer conceptual notions that are more dynamic and interactive [6‐8]. Building on the latter model, it is suggested that innovation adoption is a process that is highly dependent on the multiple interactions between the innovation, local actors, and complex contextual factors [9‐12], including epistemic and social boundaries amongst different professional groups [13].

In addition, the nature or definition of 'evidence' related to innovation is often ambiguous and contested [2, 14]. Managers make decisions based on experience, personal expertise, judgment, inference, and advice and do not passively receive new knowledge even if presented as scientifically produced and validated. Research-based knowledge has to be constantly interpreted and reframed along with the local context and clinical or managerial priorities, a process that often involves power struggles among various professional groups [15]. Different professional and managerial groups may interpret evidence differently or may prioritise dissimilar types of evidence partly as a result of their disparate organisational role, education, and training during their socialisation to the profession. We employ a sensemaking perspective to gain insight to this inter- and intraprofessional level and how this plays out in innovation adoption and implementation [16]. This lens pays particular attention to the social construction and coproduction of evidence through the interaction of a range of diverse professional and managerial groups. Our empirical setting of infection prevention and control comprises a multiprofessional workforce operating in a volatile environment, with infection outbreaks coupled with high media and public attention, as well as regulatory and political influences. This environment thus provides opportunities to contribute to this body of literature, which has been useful in explaining organisational response to critical events in the healthcare setting [17, 18].

Our work addresses a significant gap in evidence-based healthcare implementation literature. We respond to the call for more sustained interpretive work, which explores the role and motives of actors and the influence of the organisational context and the social construction of evidence [19]. Overall, we aim to address issues that permeate many stages of the research innovation pathway and, more specifically, will investigate processes that relate to the stages of evaluation, adoption, and diffusion. By contributing to the theoretical and empirical discussion, our study will add to the current international and NHS policy and practice body of knowledge. Our work also complements recent and ongoing research commissioned by the National Institute for Health Research Service Delivery & Organisation (NIHR SDO) programme; in particular, it fits well the NIHR SDO 2008 call for proposals that also focused on issues of knowledge utilisation in healthcare management [20]. We complement and add to this work by looking at different types of decisions in different healthcare settings.

Our approach allows the application of interpretive methods of inquiry without abandoning the commitment to arrive at a plausible account of commonly shared 'objective' reality. Whilst we uphold that the world and reality do exist independently of the observer, it is not possible for the researcher to escape the social world in order to study the empirical reality [30]. Any given reality can be represented from a range of different perspectives, each perspective being potentially true. This stance presents a possibility of multiple, valid descriptions of the same social phenomenon. We emphasise a more interpretive stance to explore why different understandings and meanings emerge for one observation, and how this explains different views of 'scientific' evidence and when and how different sources of evidence come into play. In addition, we investigate how negotiated orders within teams and organisations validate perceptions and hence lead to sometimes 'hierarchical' views of different sources of evidence.

The potential for learning and generation of new knowledge from this in-depth qualitative study is substantial due to the policy-relevant research aims and fit with methodological approach. We employ a study design of comparative case studies and involve multiple methods, namely, in-depth semistructured interviews, observation of key meetings, short survey questionnaires incorporated in the interviews' topic guide, reflective field notes, and systematic collection and analysis of relevant secondary data from within and outside of the organisations studied. The study is grounded in organisation and management theory, thus enhancing the generalisibility of the findings. We incorporate multilevel analysis exploring influences at micro (individual), meso (organisation/organisational unit), and macro (wider context beyond the organisation) levels and interlevel interactions. Bounding our study in the empirical setting of infection prevention and control facilitates this multilevel analysis. In addition, by employing complementary theoretical approaches, we will triangulate our findings. We will be able to explain how sensemaking occurs when multiple stakeholder groups are involved and the ways in which accounts they generate are--or are not--reconciled [31]. We will situate this within the meso and macro contexts to allow generalisability of our findings.

By combining retrospective and ongoing data, we will explore the social and organisational influences on how evidence is constructed and becomes meaningful to diverse organisational actors. This will allow us to develop a better understanding than short-term and 'snap-shot' methods. As well as the above measures of methodological rigour, we will use measures of acceptability and relevance of the research as defined by key stakeholders, namely, policy makers, academics, clinicians, health managers, and patients. Active involvement of academic experts, health professionals, people with exposure in policy development within the Department of Health, and members of the public through steering-group membership will inform our research. Specifically, our steering-committee members will inform the study design, be involved in the management and oversight of the study, provide feedback on early findings, assist in dissemination of findings to diverse localities and forums, and (through public involvement) advise on 'plain English' summaries of our research outputs.

The current climate of policy and financial uncertainty does pose a challenge in getting access to trusts and informants. We have planned recruitment acknowledging this, and contingency plans have been in place (i.e., we determined an additional pool of potential trusts to replace all types of hospitals initially targeted). We are also mindful of potential challenges in securing interviews over the winter months due to seasonal pressures on hospital staff and the further potential adverse impact of infection outbreaks. We are also aware of the challenges that may arise in coordinating and completing ethical review by NHS sites (completing global and local governance checks across all nine trusts and obtaining letters of access) in time to start data collection as planned. Finally, due to the inductive nature of the study, the number of interviews may vary depending on analytic saturation as well as the nature of technologies sampled. On limitations, a qualitative process study may span a longer duration (greater than 18 months in the field); however, such a project may have posed a higher risk with diminishing returns on research investment given the empirical setting as described above. We do have a longer retrospective dimension in four of the trusts as they comprised a sample from our previous study, which looked at the adoption and implementation processes of novel technologies [27].