Review methods

Rationale for using realist synthesis

Realist synthesis was considered the most appropriate method to use for this review in order to provide a theory-driven understanding of what needs to be included in UTI prevention programmes to support their successful implementation in care homes for older people in the UK. Previous systematic and literature reviews on UTI prevention have established what is known about UTI prevention strategies but have not addressed how the active components of intervention programmes work in care homes. Moreover, the evidence base for some of these strategies (e.g. promoting hydration) is heterogeneous with few randomised controlled trials (RCTs) to include in a systematic review,39 necessitating a more inclusive approach to sourcing evidence.

A realist synthesis was therefore undertaken, employing a systematic approach to the searching of evidence, since this has an explanatory focus and seeks to understand how complex programmes work in particular contexts and settings.40 The review examined the relationship between interventions to prevent UTI and the context in which they were applied to identify causal explanations of phenomena of interest, their mechanisms of action and what outcomes they produced.41 Different sources of evidence were sought to construct a programme theory through a process that included several rounds of stakeholder engagement, an overview of relevant extant theory42 and examination of primary research relating to the topic.

In realist terms, a programme theory represents the underpinning mechanism of action, rather than the intervention40 and realist synthesis results in context–mechanism–outcome configurations (CMOcs) that explain how an intervention might or might not work. These configurations ensure external validity as they enable theory building to a level of abstraction that means the theories are useful in other contexts.43

Context–mechanism–outcome configurations are often referred to as mid-range theories and afford the appropriate level of closeness to practice enabling the testing of initial programme theory propositions.44 Providing a clear account of the mechanisms of action was considered essential to understanding how interventions to prevent UTI can and should be delivered in UK care homes. This is because mechanisms of action explain the way in which the resource element of an intervention might work. In scientific realism, mechanisms are a combination of the resources offered by an intervention (e.g. a risk assessment tool) and the reasoning or behaviour that is required to implement them (e.g. how this changed the reasoning of stakeholders to bring about the desired outcome).40 CMOcs therefore provide explanations of what works, for whom and in which contexts and circumstances.

Changes in the review process

Minor changes were made to the published protocol relating to adaptations required to its delivery during the COVID-19 pandemic. More specifically, the project team and advisory group meetings were held online rather than in person and for shorter time periods than originally planned. The quality of communications and interactions may therefore have been affected by the limitations of online communication, although meeting virtually did facilitate attendance. Members of the core research team were able to meet in person at key points during the study to deliberate on theory development, which enhanced theory formulation. Similarly, the stakeholder workshops and teacher–learner interviews were held virtually rather than in person and for a shorter duration than planned, which may have affected the nature and quality of interactions.

Our original intention was to create a range of resources to support the prevention and recognition of UTI and CAUTI in the care home setting, such as care planning and decision-support tools. Our consideration of the literature and engagement with stakeholders helped us to identify a wide range of resources that were already in use in this setting and led us to question the value of adding further tools. Instead, we conducted a gap analysis (see Appendix 7, Table 5) to identify what types of resource are already in use in the care home setting and whether or where additional resources might be useful.

Review strategy

The review drew on evidence from health and social care, including primary research relating to UTI prevention and recognition in older people in care homes and improvement project reports in grey literature. It was conducted in an iterative rather than linear way over four stages (Figure 2).

FIGURE 2 Stages of the project. Reproduced from Prieto et al.45 Published by the BMJ Publishing Group Limited. This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See http://creativecommons.org/licenses/by/4.0/. (Figure not reproduced here; see the original article.)

Project advice and oversight

A Project Advisory Group (PAG) was established, comprising three patient and public involvement (PPI) representatives with experience of supporting older people living in care homes and individuals with relevant subject or professional expertise, including HCPs, a nursing home manager, a representative of the independent regulator for health and social care in England and researchers in the field of UTI/CAUTI prevention. The PAG was chaired by one of the PPI members and met on four occasions during the project to advise on priority issues within the review theory areas, comment on iterations of the plausible hypotheses and CMOcs and to provide challenge to the research team through their expert and lay perspectives.

The research team met at scheduled meetings to discuss progress and undertake in-depth discussions. Due to the restrictions imposed during the COVID-19 pandemic in 2020 and 2021, most of these meetings were via Microsoft Teams^TM.

Ethics

The study was identified as a service evaluation/service development following submission to the UK Health Research Authority and subsequent review by the University of Southampton Faculty Ethics Committee. However, we sought ethics approval from the University of Southampton Faculty Ethics Committee to engage with and interview stakeholders, which was granted on 15 October 2020 (submission ID 53942.A1).

Stage 1: concept mining and theory development

This stage of the realist synthesis used input from content experts in the research team, the PAG and stakeholders in addition to the preliminary scoping search.

Stakeholder engagement, including patient and public involvement

The project was supported by stakeholders, including PPI representatives, throughout the process to inform the development of initial and refined programme theories (CMOc) and establish their practical relevance and real-world potential.

A stakeholder analysis was undertaken by the research team and the PAG to identify a range of people who needed to be involved in the various aspects of the review process to impact on its success (Figure 3). This informed our approach to engaging with stakeholders in our initial theory-building workshops and at different points during the research to guide its development.

FIGURE 3 Stakeholder matrix. (Figure not reproduced here; see the original article.)

Stakeholders contributed to development and refinement of the programme theory, advising on the applicability of evidence and its relevance in specific contexts. In so doing, they helped to ensure that the synthesis was underpinned by multiple perspectives and focused upon what is important to care home residents and those who provide care.

Scoping interviews

Our initial plan was to conduct three face-to-face workshops with key stakeholders, in three separate locations across the country, to gain an insight into how the prevention and recognition of UTI/CAUTI was currently managed, the barriers and facilitators, decision-making processes, multidisciplinary involvement and staff and resident perspectives on the impact of UTI/CAUTI. However, the impact of the COVID-19 pandemic on the residential care sector resulted in us conducting the workshops online and undertaking one-to-one interviews with a small number of carers and residents. The interviews were conducted by three members of the research team via Microsoft Teams. They were recorded digitally and transcribed. An interview schedule was developed to ensure that all relevant aspects of the prevention and recognition of UTI/CAUTI were addressed. It covered the following areas:

• residents and their family’s experience and involvement:
  • what are their priorities.
• importance of ‘knowing’ the resident
• experiences of caring for residents to prevent UTI/CAUTI:
  • what influences best practice
  • identification of residents at risk
  • recognition/communication of subtle changes in a resident’s condition
  • levers and drivers, for example, policies, pathways, guidance followed
  • experience–intuition–knowledge–training–peer influences.
• involvement of GPs and other members of the multidisciplinary primary care team:
  • how the care team communicate between each other on residents’ symptoms.
• what consequences people see in missing a UTI/CAUTI.

Interview with resident and family

An approach was made by the project lead to the PPI liaison officer in Wessex. They agreed for members of the research team to present the study at one of their PPI coffee group meetings to see if there were members of the PPI coffee group who may have experience as a relative of someone within a home. Two interviews were secured with members of this group.

An interview with a care home resident was secured following one of the care home workshops. The resident was interviewed with the support of the care home manager and a member of the care home staff.

Workshops

Workshops with care home managers and staff

Our initial approach to recruiting care homes was through the Research Delivery Manager for Social Care and Patient and Public Engagement [National Institute for Health and Care Research (NIHR) Clinical Research Network Wessex] and the Deputy Director of Quality and Nursing for West Hampshire Clinical Commissioning Group (CCG). It was agreed by them that the research team could present the study to care home managers at one of their regular virtual CCG meetings in January 2021. We asked care home managers who attended if they were interested in participating to share their details with the team, who would make contact to arrange a dedicated workshop with their home.

Two care homes were recruited via this route. The first was a residential home with a 33-bed capacity, which at the time of the workshop had 25 residents. The workshop was held via Microsoft Teams, with four members of the home in January 2021. The second care home was a 20-bed residential home with dementia specialist care. A workshop was held via Microsoft Teams, with five members of the home in March 2021.

Workshops with infection prevention and control practitioners

A member of the research team contacted the co-ordinator for the Infection Prevention Society Care Home Special Interest Group (IPS SIG), who agreed that the research team could deliver a workshop at an upcoming SIG meeting. This was delivered virtually by two members of the team and recorded digitally via Microsoft Teams in February 2021. Twenty members of the SIG contributed to the discussion.

Workshop with healthcare practitioners in clinical commissioning

A workshop was convened in April 2021 via Microsoft Teams with seven representatives from CCGs across Hampshire. The aim was to capture system-level perspectives from individuals working in commissioning. Initial contact was through a colleague working as a commissioning manager for NHS West Hampshire CCG. Through their contacts, others in similar roles were recruited from across four CCGs (NHS Southampton, Fareham and Gosport, North Hampshire and West Hampshire).

Discussions with continence advisors and general practitioners

The issue of RUTI and its management in the care home setting required discussion with clinicians working in the field of urology and bladder and bowel continence. These were conducted in a series of conversations held via Microsoft Teams by a member of the research team.

Analysis of scoping interviews and workshops

The recordings of stakeholder interviews and workshops in stage 1 were transcribed and organised thematically using the topic guide areas and a modified qualitative descriptive process. This involved familiarisation, constructing themes and defining and naming themes aligned with the topic areas. This analysis informed initial theory formulation (see ‘Conceptualising what happens’ and ‘If … then tentative hypotheses’).

Literature scoping

Search methods

We conducted a broad scoping search to identify literature on UTI and CAUTI in care homes (or similarly termed facilities) using the two concepts of disease and environment. English language and 10-year date limits were used, as the focus on UTI/CAUTI in the care home sector has been more prominent in the imperative to prevent antimicrobial resistance and prevent BSI caused by Gram-negative organisms. Articles were stored in EndNote^TM reference management software and transferred to Covidence^TM review management software, where they were deduplicated prior to screening.

The starting point for the review was a scoping or ‘background’ search,46 which extended a preliminary search of MEDLINE, CINAHL and grey literature undertaken during development of the research proposal. This extended generic topic-based multipurpose search was undertaken in September 2020 and focused on evidence that directly addressed the prevention and recognition of UTI in older people in long-term care facilities. Additional databases searched at this stage were EMBASE, Cochrane Library, Web of Science Core Collection (including the Social Sciences Citation Index), Sociological Abstracts, Bibliomap and NIHR Journals Library. Supplementary searches, for example, using key index studies (highly cited) to find ‘sibling’ studies/papers (contemporaneous papers/studies that share a context), were also undertaken at this stage, using Google Scholar and ‘Publish or Perish’ software.47 In addition, an exploratory search was undertaken in October 2020 for relevant literature relating to the US Agency for Healthcare Research and Quality (AHRQ) Safety Program to reduce CAUTI in long-term care given its relevance to the review.

The scoping search identified key empirical studies, systematic and narrative reviews and guidelines, organisations with grey literature of interest and reports of quality improvement initiatives (e.g. interventions to increase fluid intake in older people living in care homes). The results of the searches are reported in Chapter 3 with details of the search strategy in Appendix 1.

Screening and appraisal

Title and abstract screening of each article was undertaken by two project team members independently of each other to identify articles within scope and out of scope for the review (Box 1). Any conflicts were resolved by the cochief investigators.

Tags were applied to those articles screened as relevant to assist in organising the data into the following broad categories:

• UTI/CAUTI prevention
• UTI/CAUTI recognition
• AMS intervention
• quality, safety, workforce or cultural context
• other tentative theory elicitation.

The research team worked as two subgroups to identify key papers from each of the categories for theory building. From these papers, key terms, concepts and potential explanations about UTI/CAUTI prevention and recognition among older people living in care homes were identified as possible mechanisms and outcomes and data were extracted (Box 2).

• What is the intervention (nature and content)?
• In what circumstance or context has it been used?
• What are its policy intentions and objectives?
• What terms and concepts are articulated?
• What was the outcome?
• What explanations are offered for its success or failure?
• What appears to be the active ingredients of the intervention?
• What questions/issues are there in relation to the intervention?
• Possible theory areas (broad)

Concept mining and theory development

A process of concept mining48 was used to map evidence about approaches to recognising and preventing UTI in older people living in care homes, how they might work and any reported enablers or barriers to their successful implementation. This involved searching different bodies of evidence and consulting with stakeholders to develop the scope of the study and identify information, key terms and concepts that could help with theory building.

Research team meetings

Through group discussion, we drew up a list of questions about interventions to recognise and prevent UTI/CAUTI and sought to clarify the meaning of terms and concepts used. This assisted our thinking in bringing together evidence on the efficacy of interventions to prevent UTI with evidence on the context-dependent active elements of intervention programmes in care homes. It also informed our ideas around alignment with external programmes, services and the wider policy context.

Project Advisory Group consultation

Consultation with our PAG and stakeholders was vital to defining the scope of the review to reflect practice and priorities in the real world. We met with the PAG in stage 1 to seek their feedback on our first iteration of programme theory areas and to gain their advice on additional stakeholders to involve and further evidence to include.

Conceptualising what happens

The research team met to consider feedback from the stakeholder interviews and workshops, scoping literature and PAG input. Discussions highlighted that many of the activities and interventions that had been highlighted could be labelled as ‘core’ or fundamental care. The concept of fundamental care49 was further explored as a potential overarching theory, since the prevention of UTI and CAUTI in older people living in care homes requires attention to key aspects of fundamental physical and psychosocial needs of individuals within the context of supported living. Figure 4 conceptualises how physical care needs are focused around eight aspects of fundamental care49 and how these align with specific strategies to prevent and recognise UTI/CAUTI in older people. As the delivery of care within care homes was described by stakeholders as being focused on fundamental care needs rather than infection prevention specifically, linking the two may highlight to staff the importance of their role in UTI/CAUTI prevention rather than this being viewed as something outside of their influence. Moreover, this may enable care home managers to align UTI/CAUTI prevention strategies to their wider organisational goals in supporting a culture of safety and person-centredness and promoting the implementation of changes and knowledge mobilisation relating to UTI/CAUTI prevention.

FIGURE 4 How fundamental care interventions to meet the physical care needs of older people are aligned with strategies to prevent and recognise UTI/CAUTI. (Figure not reproduced here; see the original article.)

If … then tentative hypotheses

A series of if … then statements were developed (see Appendix 2) using the information gathered from our stakeholder workshops and interviews, PAG members and scoping searches. These statements made a tentative link between activities or interventions (as described by stakeholders and reported in the literature) and outcomes. They suggested contexts within which staff resources and responses (mechanisms) may operate. The statements covered seven broad theme areas:

• meeting hydration needs
• preventing RUTI
• preventing CAUTI
• understanding how a UTI presents in a resident
• recognising UTI and communicating concerns
• active monitoring and watchful waiting
• involving the resident and family.

The statements were tested against the evidence that was found in stage 2 and later refined into CMOc.

Stage 2: searching process

The purpose of this stage of the project was to determine whether the initial programme theories were supported by evidence within the literature and to ensure all relevant literature had been identified. The research team began stage 2 by using the publications identified in stage 1 to further develop and test the seven broad theme areas postulated and to identify gaps in the evidence retrieved. The stage 1 searches provided a large number of articles identified to be within scope for the review (n = 1141), each requiring an assessment of the full text to determine their potential to inform theory building. Therefore, we adopted a more targeted approach to the additional searches undertaken in stage 2 to address gaps in evidence and inform the realist synthesis.46 These purposive, supplementary searches focused specifically on continence care and UTI, hydration and UTI, non-antimicrobial strategies for RUTI, the recognition of soft signs and family involvement in older peoples’ care in care homes.

Search strategy

Details of the search strategy for each of the supplementary searches can be found in Appendix 3. These searches took place between July and December 2021. Articles were stored in EndNote^TM database software and then transferred to Covidence^TM online software for deduplication, screening, selection and appraisal. Additional papers were identified by members of the research team and the PAG. The results of the supplementary searches are reported in Chapter 3 with a summary of searches and retrieval in Appendix 4.

Selection and appraisal of documents

The screening and full-text review of manuscripts were conducted using Covidence^TM software with each article being assessed by two members of the research team independent of each other. All core project team members were involved in the full-text review process, which was undertaken to determine relevance and rigour. The review of each manuscript was cross-checked by a second team member and any conflicts were resolved by the cochief investigators. The existing list of tags was applied to articles identified as relevant.

Inclusion

The test of inclusion for studies informing a realist review is based on relevance and rigour.40 Relevance is defined as the extent to which evidence can contribute to theory building, testing and refinement, and rigour as the extent to which the methods used to generate that piece of data are credible and trustworthy. As Pawson50 asserts, in a realist synthesis, the strategy is to provide a comprehensive explanation of the subjects, circumstances and respects in which a programme theory works (or not) and so there is potential for a study to offer credible explanatory evidence, even when the quality of the study may be poor. The quality (rigour) and relevance of the evidence were assessed during the research synthesis,50 and its contribution to the development of the explanatory account and to the review questions and aims highlighted.51

Included studies were organised into three broad evidence areas and data extracted:

1. UTI symptom presentation and the effectiveness of strategies to prevent UTI/CAUTI and its recurrence
2. implementation of interventions and knowledge mobilisation in care homes to recognise and prevent UTI/CAUTI
3. organisational culture and approaches in care homes in the promotion of UTI/CAUTI prevention strategies.

To focus the review on strategies of efficacy in the prevention of UTI and its recurrence, the project team further assessed the identified studies according to the degree of certainty about the potential for effectiveness in older people. This reduced the number of papers for data extraction by excluding articles reporting insufficient evidence of efficacy for UTI prevention strategies. Strategies found to be lacking in evidence included continence care and management of incontinence, hygiene and non-pharmacological treatments (e.g. cranberry formulations, chokeberry juice, Chinese teas).

Articles in the second and third categories were subjected to a further assessment, with articles excluded at this point if they offered insufficient detail to support development or testing of the programme theories.

Data extraction

Realist synthesis takes an iterative approach to the development, refinement and testing of theories. This is facilitated by the data extraction process. We used an approach developed previously by Williams et al.51 for the test of relevance and rigour, adapting a ‘relevant and good enough’ flowchart for use as the first step in the data extraction process for considering the relevance and trustworthiness of the evidence within the appraised articles (see Appendix 5). This was checked by at least two members of the team. Discrepancies were resolved through discussion among the research team, with deliberation focusing on whether the inference drawn from the research by the reviewers was sound.50

The data extraction process was undertaken by a single researcher (distributed across six team members), with a 50% proportion of those identified for inclusion being peer reviewed and checked by a second researcher. A bespoke data extraction form (see Appendix 6) was used to structure the relevant information, document insights and chart data so that the theory areas are populated with evidence on what appears to work, for whom, how and in what contexts.41 A summary table was included to record the study characteristics, including methods, settings, approach to analysis, outcomes and limitations.

Stage 3: analysis, synthesis and hypothesis testing

Analysis and synthesis processes

Realist synthesis uses triangulation to bring the different sources of evidence together in the development of theory. Abductive and retroductive analysis52 facilitates a theory-driven approach to understanding the CMOc by stimulating the consideration of how something might or might not fit a particular phenomenon (abduction) and what conditions need to exist for a concept to exist (retroduction). This process creates an iterative and critical revisiting of evidence and stakeholder input to identify patterns and contingencies and move from ‘if … then’ statements to CMOc.

A set of potential CMOcs were produced that best reflected the programme theory as refined in stage 2 of the research. For each of the included studies, the implications of its findings were explored to discern the requirements of successful programmes.50 Evidence sources were examined in respect of explanations given about the mechanisms by which interventions were considered to work (or not). The data were organised into evidence tables to enable comparison of the findings and identification of patterns across studies that offered insight about the necessary ingredients of successful interventions. The process of analysis and synthesis was undertaken by five members of the research team with input from the wider project team and PAG, including our patient and public representatives.

Programme theory testing

To test and refine the programme theory, we conducted nine ‘teacher–learner’ interviews with a range of participants (see Report Supplementary Material 1). A mixture of purposive and convenience sampling was used to gain perspectives from individuals reflective of the different audiences likely to be interested in acting on the findings from this review. Semistructured interview schedules, guided by the content of the CMOc, were used to ensure the interviews focused on participants’ perspectives of the theories as relevant to their role and expertise (see example in Report Supplementary Material 2). The theories of relevance to each participant were introduced to them by the researcher to enable them to comment on them from their own experience and using their own examples. Participants were invited to question the theories and reflect on their relevance from their perspectives of the realities of practice. All interviews were conducted online by a member of the research team and lasted between 30 and 65 minutes. Interviews were audio-recorded and fully transcribed.

Analysis of teacher–learner interviews

The recordings of teacher–learner interviews in stages 3 were transcribed and organised thematically using the topic guide areas and a modified qualitative descriptive process. This involved familiarisation, constructing themes and defining and naming themes aligned with the topic areas. These were triangulated with the CMOc to form part of the abductive and retroductive processes undertaken by the research team.

Resource gap analysis

Our consideration of the literature and engagement with stakeholders helped us to identify a wide range of resources in use in care homes to support the prevention and recognition of UTI and CAUTI. All resources identified were read multiple times and classified thematically according to the group of interest (i.e. older adults or long-term care residents), the intended end users of the resource (e.g. care home staff, prescribers, GPs), the clinical action within the care pathway (e.g. prevention, diagnosing, managing or prescribing), the domain influenced by the resource to support the user (e.g. knowledge or behaviour) and the tool type (e.g. leaflet, checklist, toolkit, algorithm). Information on publication year and publisher or developer of the resource was also collated. A summary of each tool was created, with all information cross-tabulated using Notion software (%www.notion.so) to identify any themes.

Stage 4: consultation

Stakeholder conference

Stakeholders from residential care providers, IPC services, general practice, quality improvement, integrated care systems, NHS England (NHSE) and regulators were approached to participate in a consultation event in June 2022. The aim of the event was to ensure that our theories resonated with their experience and insights of the field and to gain their advice on tailoring the research outputs for different audiences and targeting the dissemination of findings appropriately. Although we had planned a face-to-face event, it was not possible to bring the participants together for more than 2 hours due to participants’ clinical and work pressures. Among those who expressed an interest in participating, there was a preference to meet virtually. The event was therefore held via Microsoft Teams and the breakout discussions recorded and transcribed.

The evidence-informed programme theories were presented and shared with stakeholders to test clarity and fit with current strategies and quality initiatives for recognising and preventing UTI in older people in care homes. In addition, our gap analysis of the existing types of resources in use in care homes, including physical deterioration and escalation tools, decision aids, toolkits and guidelines, was discussed with stakeholders and an example of how digital tools might work in the care home setting was also shared. Report Supplementary Material 3 lists the representatives who attended the stakeholder conference and Report Supplementary Material 4 includes the presentation and breakout room topic guides.

Analysis of feedback from stakeholder event

The approach used for the analysis of scoping workshops and interviews in stages 1 and 3 was also used for the recordings of the stakeholder event, which were transcribed and organised thematically using the topic guide areas and a modified qualitative descriptive. As before, this involved familiarisation, constructing themes and defining and naming themes aligned with the topic areas. These were triangulated with the CMOc forming part of the abductive and retroductive processes undertaken by the research team.