Development of a program theory for clinical pathways in hospitals: protocol for a realist review

We choose a realist approach to develop a program theory for CPWs in hospitals due to the heterogeneous nature of the evidence base generated by our ongoing systematic review (Rotter and colleagues: Clinical Pathways for secondary care and the effect on professional practice, patient outcomes, length of stay and hospital costs, unpublished) on the effects of clinical pathways in hospitals. The initial program theory will be derived from the preliminary findings of our on-going Cochrane systematic review update on the effects of CPWs in hospitals. International participants outside Saskatchewan will be involved throughout the theory building phase. We will identify and group the processes, strategies, resources, and outcomes resulting from the implementation of CPWs in hospitals based on the objective and theoretical accounts of the authors. The review team will construct the preliminary program theory by generating hypotheses of “how,” “to what extent,” and “in what contexts” does the implementation of clinical pathways in hospitals contribute to better health system outcomes. This process will be carried out in close consultation with a realist methodologist, Dr. Gill Westhorp. With support from the Saskatchewan Ministry of Health and the Saskatchewan Health Quality Council’s Variation and Appropriateness Working Group (VAWG) and our international stakeholders, we will consult with a group of multidisciplinary content experts including physicians, nurses, researchers, project managers, and other relevant frontline users of CPWs based in and outside Saskatchewan, to provide critical insights to refine the program theory.

To develop an initial understanding of how CPWs in hospitals are intended to translate to better health system outcomes, we will conduct a focused examination of the CPWs literature. As recommended by the RAMESES guidelines, our search strategy and search terms will be guided by hypotheses developed from the preliminary program theory [27]. We will start with the initial search terms identified from the on-going systematic review update of CPWs effects in hospitals. Search terms will include: “clinical pathways,” “care pathways,” “critical pathways,” “clinical protocol,” “implementation,” “effective,” “hospital,” “patient,” “physician(s),” “treatment,” “theory,” “models,” “strategy(s).” These search terms will be examined by the review team in consultation with an information scientist and expanded as necessary. The search process will commence with Medline and Google scholar and be further expanded to other databases such as NHSEED, CINAHL EBSCO, HMIC, and PsycINFO. All sources of information related to our review objectives will be considered. We will assess saturation iteratively by asking after each stage or cycle of searching whether the latest information has added anything new to our understanding of the intervention and whether further searching is likely to add new knowledge [21]. Since CPWs emerged in the 1980s [8], we will search for published and unpublished studies from 1980 to present with no language restrictions. See (Additional file 2, part A) for an example of the initial MEDLINE search strategy used for the on-going Cochrane systematic review on the effects of CPWs in hospitals.

Additionally, we will search gray literature and websites of organizations involved with the implementation of CPWs, such as Intermountain Healthcare [31], European Pathway Association [32], The National Institute for Health and Care Excellence [33], etc. We will consult with authors of included primary articles for other relevant publications. We will use a snowball sampling method [34] to examine references of retained articles during the initial screening phase. We will follow an iterative approach for searching the literature to refine or refute the initial program theory. Overall, the initial development and refinement of the search strategy and conducting the literature search will follow an iterative process.

Iteratively, we will use a two-step process to screen all potentially eligible retrieved studies beginning with the abstracts, followed by full-text assessment of the included abstracts. Two review members will independently screen all retrieved studies from the search process using the following inclusion criteria: (1) intervention meets the following definition for a clinical pathway (i) the intervention was a structured multidisciplinary plan of care; (ii) the intervention was used to translate guidelines or evidence into local structures; (iii) the intervention detailed the steps in a course of treatment or care in a plan, pathway, algorithm, guideline, protocol, or other ‘inventory of actions’ (i.e., the intervention had time-frames or criteria-based progression); and (iv) the intervention aimed to standardize care for a specific population. An intervention meeting all four criteria will be considered a CPW); (2) study was conducted in a hospital; (3) availability of information on implementation strategy(s) employed; (4) at least one health system outcome reported such as length of hospital stay in days, hospital costs, in-hospital mortality, adherence to recommended practice, and in-hospital complications. We will exclude studies on CPWs that are not focused on hospital interventions.

To ensure that the credibility and trustworthiness of the review are maintained, we will assess the rigor and relevance of each study’s potential contribution to the development of the program theory. Two independent reviewers will assess for rigor and relevance by examining whether the study inferences are based on evidence or author opinion. This reflexive process will use judgements and discussion to resolve conflicts. We will not exclude studies based on rigor. We will manage the screening and data extraction process using the Distiller SR software [35]. Where applicable, a third review member will resolve any disagreement during the screening process. Following a similar approach, we will assess gray literature and other sources of information relevant during the review process.

Two review authors will independently extract data from all included primary studies using a piloted draft proposition sheet to extract explanatory account propositions (EAPs) about the program theory for CPWs implementation in hospitals. The EAPs are explanatory statements that may support, refine or refute judgements about an aspect or the program theory in its entirety [27]. Areas of data extraction will include the following:

Relevance and rigor are dimensions of ‘fitness for purpose’ in realist synthesis. EAPs will be extracted from primary studies using an adapted format (“If-then-because”) for extracting propositions in realist inquiries [36]. If-then-because propositions assume that “outcome-y” will occur because “z mechanism” fire in context “x.” We will identify EAPs focused on the implementation and outcomes of CPWs in hospital settings. Subsequently, we will attempt to identify the CMOs from the extracted EAPs. Additionally, we will identify middle-range theories which are demi-regularities emerging from the data and are observable as a testable hypothesis to refine, support, or refute the program theory [27]. Where feasible, the extracted middle-range theories will be supported by formal theories identified from our ongoing systematic review update or other sectors apart from healthcare, where CPWs have been implemented. This step will be supported by the review teams’ theoretical and content expertise. All extracted data will be managed using Microsoft Excel. See (Additional file 2, part B) for the draft data extraction sheet to be used for the review.

Two review authors will extract the EAPs from the data extraction stage. Where necessary the team will consult with a review team to resolve any disagreement to improve the consistency and validity of the evidence generated. We will group the EAPs statements based on emerging themes, for example, “leadership support” or “staff empowerment.” Using abductive reasoning technique [37], we will identify the emerging CMOs configurations from thematic groupings of EAPs. Because the emerging themes from EAPs are not mutually exclusive, EAPs can be grouped into more than one thematic category. Where appropriate, we will juxtapose, reconcile, adjudicate, consolidate, and situate the evidence created within the framework of the program theory.

We will then identify the CMOs configuration existing within each thematic grouping and attempt to form CMOs configurations at the middle range theory level. Middle-range theories explain the regularities of social behavior that can be explored further by hypothesis testing [27]. The CMOs at the middle range theory level will be used to revise our initial program theory to form the initial revised program theory. All hidden and explicitly stated mechanisms [26] will be elicited with the support of content and theoretical expertise within the review team. Due to the focused scope of the review, we will only utilize key mechanisms in developing and refining the program theory; we will utilize mechanisms that yielded the largest impact on the outcomes of interest based on different contexts reported in the primary studies. This approach also makes the breadth and depth of the review manageable and tangible for translation and implementation. This process will be conducted in consultation with content specialists and other frontline users of CPWs in hospitals to refine the program theory. Since the quality assurance of our realist review is based on the reflexivity and explicitness of our review team and stakeholders involved, we cannot guarantee all fundamental mechanisms will be illuminated in the final program theory or that our results will be reproducible in other jurisdictions. Also, our literature search process may miss essential studies that may influence the final program theory. However, we will document and explain all changes conducted between the protocol and the realist review.

To ensure the relevance of the program theory, we will consult with a CPW stakeholder advisory group comprising of content experts, a realist methodologist, a patient representative, policy-makers, program managers, healthcare providers, and other CPWs frontline users. Content experts may include authors of primary studies focused on CPW implementation in hospitals or researchers who have successfully implemented CPWs in hospital settings. We will present the initial revised program theory and other study findings to the advisory group for their perspectives and to confirm, refine, or refute the initial revised program theory. The stakeholder engagement sessions will be delivered via facilitated discussion in a semi-structured format comprising of face-to-face and remote sessions to accommodate international participants. The engagement sessions will follow the guiding principles described by Manzano 2016 [38] for conducting approach for realist inquiry. The goal of the engagement sessions is to ensure the appropriateness of all aspects of the program theory and to make suggested changes where necessary. This process will further improve our understanding of the program theory for CPWs in hospital and ultimately lead to the development of the final revised program theory. We will develop a detailed knowledge translation plan with key decision-makers in Saskatchewan to facilitate the translation of the main findings to clinical practice.

Prior to consulting with stakeholders to help refine the draft program theory, ethical approval will be sought through the University of Saskatchewan Research Ethics Boards. All stakeholders will be consulted for consent to participate and publish.