Interventions to minimise hospital winter pressures related to discharge planning and integrated care: a rapid mapping review of UK evidence

Article history

The research reported in this issue of the journal was funded by the HSDR programme or one of its preceding programmes as award number NIHR130588. The contractual start date was in November 2022. The draft manuscript began editorial review in January 2023 and was accepted for publication in January 2024. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HSDR editors and production house have tried to ensure the accuracy of the authors’ manuscript and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this article.

Copyright statement

Copyright © 2024 Cantrell et al. This work was produced by Cantrell et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2024 Cantrell et al.

Background

The term ‘winter pressures’ refers to ‘how hospitals cope with the challenges of maintaining regular service over the winter period’. 1 Although attention typically focuses on additional demands on accident and emergency (A&E) services, pressures are exerted in terms of increased demand across the entire health and social care system. Increased prevalence of respiratory and cardiovascular illnesses during the winter places severe demands upon systems that already face difficulties in matching service provision to demand. Better or earlier planning and/or increasing service provision across integrated care (IC) systems and related sectors, such as housing, constitute just two of the possible responses to increased demand.

The COVID-19 pandemic response illustrates many issues that need to be addressed in planning for hospital discharge when the health system faces specific expected or unexpected system pressure. While very large numbers were rapidly discharged to care home beds in the early months of the pandemic, the discharge process generated significant concerns about whether the associated risks to both the discharged patients and the care homes had been sufficiently considered in the discharge planning process. 2

Winter pressures are a familiar phenomenon within the National Health Service (NHS) and represent the most extreme of many regular demands placed on health service provision. The impact of winter pressures is pervasive and operates along a continuum from public health, in mitigation via immunisation campaigns, through to discharge into social care and the community. Interventions target multiple points along this continuum, as well as whole-system approaches.

This mapping review focuses on one stage of the pathway, considered particularly problematic, namely the discharge process from hospital. Studies of discharge interventions and transfers of care are plentiful and international initiatives are reviewed in a recent scoping review. 3 However, a considerable challenge lies in identifying interventions that specifically are articulated as an explicit response to ‘winter pressures’. There is strong rationale for focusing on the subgroup of winter pressure interventions because these may represent interventions implemented in response to particularly acute and severe system pressure. This contrasts to interventions that seek to improve discharge within a system in ‘steady state’. While commentators on the current context articulate that the system experiences ‘winter’ pressures all year round, until recently, responses have targeted the context of winter pressures. Potentially, this set of studies therefore represents a highly relevant evidence base on discharge related interventions to inform current NHS planning.

This mapping review aims to chart and document the evidence in relation to winter pressures in the United Kingdom, together with either discharge planning to increase discharge (both a reduction in patients waiting to be discharged and patients being discharged to the most appropriate place) and/or IC. For the purposes of this review, ‘integrated care’ involves partnerships of organisations that come together to plan and deliver joined-up health and care services in NHS England.

Objective of the review

The primary objective of the mapping review is to address the question:

• Which interventions in relation to discharge planning/IC have been suggested, tried or evaluated in seeking to address winter pressures in the United Kingdom?

The secondary review question is:

• Which research or evaluation gaps exist in relation to service- or system-level interventions as a response to discharge planning/IC in the context of winter pressures?

In addressing this question the mapping review will seek:

• to identify a potential winter pressures research agenda in relation to discharge planning/IC to inform commissioning of future research.

Setting the scope and inclusion criteria

This mapping review focuses on interventions designed to reduce clinically unnecessary occupancy of hospital beds and demand on health services through discharge and transfer to social care or the community (e.g. ‘hospital at home’). Such interventions typically require improved linkages and communication between various health and social care services, captured in the concept of IC. Integrated care systems seek to capitalise on the benefits of such ‘joined-up’ systems.

Key to feasibility is the requirement for study authors to have linked their research or data explicitly to winter pressures. Numerous interventions have been proposed to handle discharge planning but not all have been explored within the context of a pressurised health or social care system. Insisting on this requirement helped in identifying interventions or mechanisms that offer promise within other pressurised care environments. Specifically, we based eligibility on the following aspects – population, exposure, comparative, outcome(s), study types:

• users of UK health and/or social care systems (population)

• winter pressures impacting on discharge, to social care and the community and IC (exposure)

• other foreseeable, unusual or exceptional periods of demand (if appropriate) (comparison – may or may not be present)

• increased smart discharge (both a reduction in patients waiting to be discharged and patients being discharged to the most appropriate place), system effects, health and health service outcomes, effects on patients, carers and staff (outcomes)

• eligible types of study design: primary research study, evidence synthesis or research report (study types).

Searching for evidence

The mapping review took a two-pronged approach to identifying the evidence for this review. The first-stage search identified the interventions through explicit references to winter pressures. The second stage of further searching was undertaken of named interventions extending beyond winter pressures on Google Scholar (Google Inc., Mountain View, CA, USA) to find systematic reviews and guidance. More detail is provided below.

Screening evidence

Study screening and selection was undertaken in Microsoft Excel® (Microsoft Corporation, Redmond, WA, USA). A team of three reviewers independently screened the titles and abstracts of the 723 references from the searches. Articles meeting the inclusion criteria for titles and abstracts or items with insufficient detail on content relating to discharge planning or IC were reviewed at full text.

Coding and database production

Metadata extraction and coding for studies were undertaken using Microsoft Excel. For mapping purposes, references were categorised at title and abstract stage according to study design, characteristics and broad thematic content. A data coding spreadsheet was used to code data for study type, population, intervention (if present), outcomes or findings and conclusion. Following granular coding, we then classified related studies within broader thematic categories that were be used to present study characteristics and to structure an accompanying narrative. Due to the time and resource constraints of the proposed work, missing or unclear information was not solicited from authors.

Describing and visualising the findings

Findings: taxonomy

The final taxonomy is presented in Table 1. The taxonomy reflects the focus of the review on supporting early discharge where safe and appropriate, but also includes interventions aimed at avoiding hospital admission through rapid response or delivering care in other settings. The ‘cross-cutting’ category groups interventions that could be delivered at different stages of the patient pathway and/or different applications of a particular technology (e.g. digital and data).

Findings: map

The final map (Figure 1) is provided below and is also available as an accompanying PDF. The map lists key evidence sources for candidate interventions and is colour coded under the headings of the taxonomy S, CSB, CCP, IC, and TC. Modelling and workforce planning were excluded from the scope. Interventions delivered at different stages of the patient pathway are not differentiated in the map but are identified in Table 1.

FIGURE 1.

Map of winter pressures research organised by taxonomy headings. Note: a list of references for the report is available in the supplementary material (see Report Supplementary Material 1).

Findings: taxonomy headings/interventions

Findings are presented in the following tables following the order of the taxonomy presented in Table 1. Definitions and sample interventions are presented first, followed by a separate table of supporting evidence and identified research gaps under each taxonomy heading. For brevity within the report, the tables for the taxonomy headings with limited evidence are provided in the appendices.

Structural interventions

Findings for taxonomy headings classified as structural are summarised in the tables below. The tables are organised by the different stages of the patient pathway and first cover hospital avoidance then alternative delivery sites, followed by facilitated discharge and then Tables 18–34 cover cross-cutting headings. Detailed information can be found in the tables, with brief overviews before and after each group.

Changing staff behaviour

Findings for taxonomy headings classified as CSB are summarised in Appendix 5, Tables 65–74. The taxonomy headings were clinical audit, education of staff, protocols/guidelines, quality improvement programmes and quality management systems. Within this taxonomy section, we identified limited initiatives or research studies. CSB can take time and initiatives to minimise winter pressures are generally implemented rapidly without the time for thorough evaluation or to measure all potential outcomes. Initiatives to minimise winter pressures could potentially be too short to make changes to staff behaviour and they might not measure staff behaviour as an outcome. Further research studies could plan evaluations that consider a variety of outcomes including changes in staff behaviour, although it can be difficult to measure.

Changing community provision

In the first version of the taxonomy, as employed during the initial winter pressures mapping process, all CCP entries were subsumed under a single overarching heading. Subsequently, as we sought evidence on the specific intervention types, the review team decided to differentiate community provision within three subcategories. We believe that this is the first and most important step towards starting to think how different types of intervention attempt to respond to winter pressures in advance of a realistic focus on mechanisms. The three subcategories are therefore labelled as CCP – hospital avoidance, CCP – alternate delivery site, and CCP – facilitated discharge. Each is considered, together with specific interventions under each subcategory in the tables below.

Integrated care

Findings for the IC heading are presented in detail in Tables 44 and 45 and Appendix 7, Tables 83 and 84 and the associated summary paragraphs. Probably even more than any other section of the taxonomy, we found considerable overlap between IC interventions and other interventions. As mentioned above, a focus on ‘delivery site’ elsewhere in the taxonomy has meant that the section on IC tends to focus on initiatives that are distinctively branded as ‘integrated’. Furthermore, this particular category within the taxonomy, a specific focus identified during the commissioning of the review, is distinctive in that it operates across multiple levels; from integration at an organisation level, through shared pathways of care to frontline integration of health, social care and voluntary services staff within individual teams. As a consequence, coverage of this category is notably heterogeneous. As a combined front for research activity it would particularly benefit from exploration of mechanisms to establish whether these operate at multiple levels as a feature of integration or whether they operate exclusively within a single level or strata of the organisation.

Targeting carers

Findings from the targeting carers initiatives are presented in detail in Tables 46 and 47 and associated summary paragraphs. Four overall approaches were classified as targeting carers, carer education, carer preparation, carer respite and carer support. As with other sections of the taxonomy, we found considerable overlap both within carer interventions and between other interventions and carer provision.

The sheer breadth and scope of this taxonomy section made a comprehensive survey of research prohibitive. The heading is included in the taxonomy for completeness, as mentioned in the discussion targeting carers sustains its own research agenda. Nevertheless, we have identified the four key areas required within the targeting carers topic and have identified limited initiatives or research studies. The role of carers has not only been identified as pivotal but also represents a hidden cost alongside formal systems of care, often compensating for pressures within the overall health system. However, reliance on carers exerts its own impact on the carers themselves. ‘Winter pressures and the implications on carers’ would seem to constitute a promising area for further exploration. It remains to be seen whether benefits of, for example, carer respite during less-pressured times can produce a dividend during subsequent pressure points in terms of carer resilience and community support. Because of its often invisible nature, carer contributions can prove very challenging to measure.

Overview of discussion

The discussion starts with an overall summary of common findings over all intervention types. It then proceeds to brief summaries of the five overarching types of interventions: (1) structural interventions, (2) CSB, (3) CCP, (4) IC, and (5) targeting carers. Finally, the discussion summarises the research priorities and the potential for a realist review.

Overall summary

Undoubtedly the foremost limitation of this review is the absolute shortage of studies that describe and evaluate winter pressures in a specific way. As mentioned above, there is no shortage of studies that describe achievements against a backdrop of winter pressures or that describe quality improvements with multiple collateral benefits including an impact against winter demands. However, as a review team, we agreed that this is substantively different from interventions that specifically target winter pressures. Even following winter funding to identify specific targets for development could not guarantee that each intervention was conceived against the specific requirements particular to winter pressures as opposed to relieving capacity pressures more broadly. As a team, we found ourselves asking should ‘winter’ simply privilege the time when initiatives take place, conventionally ‘winter pressures’ have been attributed to a specific time period or even a specific date, or are we looking for something related to specific pressures on the system as characterised by the pressures observed at winter. This discussion is particularly relevant, given recent acknowledgement that in recent years winter pressures seem to be taking place ‘all the year round’. 186 This realisation had a considerable bearing on our decision, following the preliminary mapping specific to winter pressures, to expand to the evidence base on wider responses to health and social care system pressures.

Overall, the evidence base is characterised by large numbers of case studies, often published online or presented at conferences, and relatively few peer-reviewed journal articles. Case studies are often accompanied by guidance to support implementation of changes to services. This distribution of evidence probably reflects the urgent need to develop and implement solutions to the ever increasing winter (and increasingly year round) pressures on the health and care system. The majority of evaluations report positive effects on important outcomes, such as length of hospital stay, but many are uncontrolled or based on small samples, meaning that they need to be interpreted with caution.

A further interpretation of the evidence base, fuelled by the expansion of winter pressures to other times of the year, acknowledges that ‘winter pressures’, although undoubtedly real, tend to assume extra narrative significance by presenting a continuous systemic challenge in binary form. Politically, ‘winter pressures’ may become a device used to spotlight specific attention on health system under-resourcing so that these issues assume special significance over and above general awareness as highlighted by politicians, the popular media and public campaigns.

Summary of structural interventions

Structural responses to winter pressures were classified as ‘hospital avoidance’, ‘alternative delivery site’, ‘facilitated discharge’ and ‘cross-cutting’. In line with the review’s emphasis on timely hospital discharge and IC post discharge, most taxonomy headings fell into the latter two categories. Nevertheless, it is clear that avoiding admission in the first place (where clinically appropriate and feasible) and exploring care delivery in different settings both play important roles in the overall response.

In the area of hospital avoidance, we identified many UK initiatives around SDEC, mostly reported as case studies or evaluations of low methodological quality, suggesting a research gap. Surgical hubs are a more recent development and we have identified research and implementation gaps in this area. Extra service delivery takes a variety of forms and some successes have been reported in both emergency and elective care but the evidence base remains limited.

In terms of alternative delivery sites, acute medical units were defined by the Royal College of Physicians of London as long ago as 2007. 187 Recent systematic reviews have demonstrated their effectiveness. Other initiatives have involved developing specialist units using winter funding but these remain to be fully evaluated.

Some aspects of facilitated discharge are relatively well researched, particularly models based on D2A (also ‘Home First’ and others). In addition to case studies and guidance, we identified a range of Healthwatch reports providing a patient/carer perspective on how D2A services operate in real-life clinical practice. Interventions with little UK evidence were sometimes supported by systematic reviews, although the quality of these varied. Some taxonomy headings (e.g. ‘bed management’ and ‘discharge co-ordinators’) were often evaluated as part of a broader process of ‘discharge planning’. ‘Patient flow’ is another broad heading with some overlap with both ‘bed management’ and discharge planning. The concept of ‘patient flow’ is also broader than facilitated discharge, although its ultimate goal is ensuring safe discharge as soon as is clinically appropriate. This highlights an issue we found with the taxonomy of the overlap within the taxonomy headings.

Evidence support for initiatives defined as ‘cross-cutting’ varied widely. The winter pressure search identified UK initiatives to support improved teamwork, together with associated guidance (see Table 19). Various initiatives have applied data sharing and digital technology, but we found a relative lack of research studies. Actions at the organisational level (e.g. involving changes in governance or formulation and implementation of policy) play an important part in responding to winter pressures but this was not reflected in research. Further research studies are needed in the area of seven-day services, although promising case studies exist. Staff redeployment has achieved successes within the pharmacy profession and research involving other staff groups would be useful. Volunteers have been used within the NHS in many different areas but there needs to be further evaluation of these initiatives in relation to winter pressures.

We identified significant numbers of recent systematic reviews relevant to, but not targeted at, responses to winter pressures. These are potentially valuable but the review questions rarely overlap completely with the needs of addressing winter pressures, suggesting that further targeted evidence synthesis (perhaps using realist approaches to focus on the UK context) could be valuable. The diverse range of literature identified by this mapping review suggest that such an approach would be feasible for at least some approaches (e.g. D2A, reablement, virtual wards and optimising patient flow).

Summary of changing staff behaviour

Changing staff behaviour responses to winter pressures were all classified as ‘cross-cutting’.

There was recent guidance available for documenting local clinical audit from the Healthcare Quality Improvement Partnership. A case study from Shrewsbury and Telford Hospital used clinical audit to investigate compliance with the discharge process. The study found that, in 2019, patients were still being kept waiting for the same reasons in the discharge lounge and few patients were transferred to the discharge lounge.

Within the education of staff heading, there was one small research study on teaching staff about patient flow, which found that multimodal teaching approaches were the most successful at the ward level to make it relevant to staff. 27 Patient flow is one of the headings within structural interventions. The interventions introduced could also include education of staff, which again highlights the overlap within the taxonomy headings. Education of staff could be part of larger initiatives with multiple components but might not be the outcome that is been measured, more likely that the number of patients discharged or patient length of stay would be main outcomes measured. The importance of looking after the workforce and creating the best possible conditions for staff to do their best for patients was highlighted in a recent poster of a rapid review. 198

There was no evidence for the use of protocols or guidelines specifically to mitigate winter pressures. A systematic review from 2015 researching how to improve the quality of discharge summaries found that guidelines were useful for providing the key components of discharge summaries. 32

The lack of evidence for the use of quality improvement programmes to mitigate winter pressures could be due to the nature of winter pressures, with a need for implementation of rapid solutions without the time for thorough evaluation. However, the NHS Confederation has drawn six lessons from the pandemic experiences of the Q Community that would be useful to apply as winter pressures increase. Quality improvement programmes could be useful for introducing initiatives to mitigate winter pressures and would allow for more thorough planning and evaluation.

There was no evidence of the use of quality management systems to mitigate winter pressures. Within Scotland, there is a move from the use of quality improvement to the use of quality management to support better quality health and social care, indicating that quality management could become more important.

Changing staff behaviour can take time and initiatives to minimise winter pressures may need to be implemented rapidly without the time for thorough evaluation or to measure all potential outcomes. Initiatives could potentially be too short to make changes to staff behaviour and they might not measure staff behaviour as an outcome. Further research studies could plan evaluations that consider a variety of outcomes including changes in staff behaviour, although it can be difficult to measure.

Summary: changing community provision

Numerous approaches were classified as CCP. Community responses either seek to keep those who need not be admitted to a hospital within the community, whether through preventive or early warning measures or through substituting community for acute services as the point of delivery, or they seek to facilitate and accelerate hospital discharge. In the latter case, two overall approaches can be observed, either by co-ordinating the patient’s needs to remove obstacles to discharge or, again, by offering services traditionally delivered within acute care from a community setting. ‘Hospital avoidance’ from the community has been artificially separated from hospital avoidance within the structural category, simply on the basis of the setting from which avoidance is enacted. We considered that attributing initiatives and evidence to the setting is more helpful than aggregating all hospital avoidance schemes together. So, for example, a scheme that seeks to avoid hospital admission by meeting patient needs as an outpatient within acute services is distinguished from a scheme that seeks to avoid placing any demand on acute services.

Findings are presented in detail in Tables 20–23 and Appendix 6, Tables 75–78 and associated summary paragraphs. As with structural interventions, we found considerable overlap between definitions and interventions. In some cases, we found ourselves placing artificial demarcations between interventions; for example, in focusing telecare on ‘non-health’ care needs and in distinguishing ‘virtual wards’ where a single set of related functions (e.g. rehabilitation) takes place from a ‘virtual hospital’ that seeks to replicate the full range of functions offered by a physical hospital. Equally common are packages that can be coded for multiple components.

As with the King’s Fund report Reimagining Community Services: Making the most of our assets (2018),18 we have found that many areas are trying to improve co-ordination through introducing single points of access or trusted assessor arrangements. The intention is to reduce duplication in referral and assessment processes. Sometimes these changes are associated with creation of new roles with a specific care co‑ordination function (care navigators or care co-ordinators). The King’s Fund report highlights how these roles are intended to act as a ‘workaround’ to help people navigate complex and unco-ordinated services and highlights how ‘a better solution would be to make services less fragmented, so that this additional layer of complexity is not needed’. 18

Similarly, we identified commonality in interventions that seek to better co-ordinate care by bringing professionals together in integrated community teams. We discuss team-based approaches in the community under ‘community teams’ while whole system approaches are explored under a separate integrated care section. These multidisciplinary community teams typically assemble diverse community health and social care professionals working alongside groups of GPs. The teams often target older people or other groups that have relatively high health and care needs. 188 Teams are usually based within a local area and can involve multiple professionals.

The core teams can link into a wide network of local services to meet the full range of people’s needs. By working together in one team, staff across different disciplines can communicate regularly, share knowledge and expertise and co-ordinate care planning and delivery

p. 6018

Some aspects of CCP are comparatively well researched, particularly hospital at home services for a variety of healthcare needs, where several systematic reviews exist. Other less-researched initiatives, particularly trusted assessors, step-down facilities and SPOA for social care, are well represented by UK initiatives. Indeed, so many local constituencies are experimenting with such models that it is questionable whether it is meaningful to single specific initiatives for particular attention. Clearly, it could be valuable to explore variations across different models of the same initiative and to undertake robust evaluations.

Almost all of the initiatives are not specifically ‘winter pressures’ approaches. However, we have identified their use as winter pressure measures or as recipients for winter funding. This fact has been used to identify the candidacy of initiatives in this report for which we have subsequently sought more substantive and plentiful evidence. A key evaluation issue relates to whether to undertake evaluation specifically of initiatives within a winter pressures context or whether to evaluate them more widely on the understanding that they may be particularly beneficial when the health system is under particular strain as is exemplified by winter pressures.

As with structural interventions, the evidence base is characterised by large numbers of case studies, often published online or presented at conferences, and relatively few peer-reviewed journal articles. The imperative to tackle urgent issues and to adopt initiatives that have proved promising elsewhere can be detected across the evidence base. We have identified a need for extensive and robust evaluation, which could then be targeted for consolidated learning through reviews of case studies or even systematic reviews. Again, there is methodological cause for concern in the likely presence of multiple biases such as early adopter bias, hero innovator bias and reporting bias. A further possibility relates to the fact that organisations that are encountering particularly severe problems may adopt multiple responses, meaning that the strength of the effect is not replicated across more uniform sites or that positive impact may be achieved by a ‘kitchen sink effect’ of multiple indistinguishable interventions that are introduced simultaneously.

As with the D2A literature within structural interventions, the hospital at home literature is associated with significant numbers of recent systematic reviews. Their value is again dissipated by the lack of a specific focus on addressing winter pressures, suggesting that further targeted evidence synthesis (perhaps using realist approaches to focus on the UK context) could be valuable. Certainly, the plethora of different initiatives with diverse labels suggests that a focus on common mechanisms would prove more useful than a diffuse investigation based on interventions.

Summary: integrated care

Approaches to IC operate at multiple organisational levels. Evidence from Scotland allows exploration of vertical integration between health and social care services and its impact on delayed discharges. Early experience from the vanguards also allows comparison of system-wide integration. Less transformative experiences of integration are seen when health and social care agencies seek to deliver approaches to discharge planning in a joint approach that works across the respective organisational boundaries. Finally, frontline instances of IC can be achieved when health and social care agencies are represented in joint operations. This final approach is exemplified by, for example, social care and voluntary agencies being represented at a local level in discharge ‘huddles’. Other approaches involve placing staff within a different organisational environment; for example by placing social workers within a hospital or by providing community access to acute hospital expertise. These types of initiative have tended to be discussed under the delivery site for the initiative, whether an acute hospital (covered under structural) or community provision.

Integrated care poses challenges in both achieving and measuring an impact. The complexity of interventions that work across organisational boundaries means that it can take some time to organise new pathways of lasting permanence. Alternatively, short-term initiatives may be able to achieve short wins but this impact is either not sustained or, more typically, is not reported over longer periods, making evaluation problematic. Generally, evaluation of organisational approaches to improved discharge planning is neither as wide-ranging nor as sustained as it needs to be to provide reliable evidence. Measurement of long-term outcomes is therefore required.

Positive reporting bias is particularly likely to occur within new initiatives for IC. Where multiple measures are reported, accounts focus on those indicators that have achieved a positive outcome. Where initiatives are able to report less demonstrable achievements, they may focus on qualitative aspects of the integration, in terms of both structure and outcome. In particular, reports of IC initiatives focus little attention on the hidden costs in terms of staff time in setting up initiatives and in managing organisational change.

Notwithstanding such caveats, the pursuit of IC is supported by several studies that have demonstrated positive results and by a wealth of anecdotal experience of the breakdown of health and social care communications between separate agencies. Simplifying communications and procedures carries an implicit driver towards improvement, whether or not achieved in the short term or sustained over a longer period.

The specific challenge in relation to IC once more lies in the difficulty in attributing the impact of IC interventions and initiatives specifically to benefits to acute care use during times of winter pressure. By implication, care has to be integrated as a permanent feature across all seasons and patterns of demand. Commentators have observed that the health and social care systems now face ‘winter pressures’ all year round, making IC interventions more apposite as a response to endemic health service challenges. The other implication relates to the scale of change often required for change to take place, particularly involving multiple stakeholders and organisations. While any attempts at improving information flow and communication and operation of shared processes are to be welcomed, the capacity of organisations to be able to invest time and effort in organisational change at a time of systemic overload has to be questioned. A form of ‘natural selection’ may therefore take place whereby the more easily achieved small-scale adaptation of existing systems may be privileged over more radical system-wide transformation.

A more detailed analytical review could analyse initiatives against a tightly defined and recognised classification of integration. More significantly, a realist synthesis could focus on shared mechanisms by which different levels of integration achieve improvements in outcomes relating to admission avoidance or reduction in delayed discharge.

Summary: targeting carers

Approaches to targeting carers tend to fall within two main types of response: reducing either demand per se or inappropriate demand by improving carer navigation across the community/acute margins or increasing carer resource and resilience by equipping them to handle demands that might otherwise draw upon hospital-based resources. One of the challenges of the latter approach lies in the fact that carers themselves have their own health and care needs. Placing extra demands on those who are already finding it challenging to cope may represent a short-term solution that reaps later negative dividends (e.g. in the carer requiring extra support at another time). At best, this may simply spread demands upon the health and social care system more evenly across the year – for example when the carer requires respite services at other times of the year. Whether such a ‘transferred need’ presents acutely or is managed and planned exerts a considerable difference in the success of such measures. Of more concern is when placing an extra burden on a carer results in an immediate breakdown of informal support, meaning that both patient and their carer make demands on the system at a time of winter crisis and perhaps for an extended period in contrast to an otherwise brief episode.

Carer education and carer preparation are generally targeted at the carer in advance of any challenging situation. Education may involve knowledge of what to do when faced with a particular crisis, thereby increasing carer resilience and reducing demand upon acute services, or it may involve increasing carer capacity to navigate the care system appropriately, leading to better targeting of need. Preparation may involve practical skills or competencies that reduce demand on formal carers; for example, in being able to change dressings, administer medication. In both cases, education and preparation typically take place at a time of ‘normal’ demand on the health and care system and needs can be planned and anticipated.

Carer respite may occur in response to an acute need or crisis or may form part of a planned or managed response. Respite can take place on a very short-term basis; for example, allowing carers an hour or two away from the demands of their informal care, or it can take place over an extended period, as with accommodating an annual one- or two-week holiday. Again, the assumption behind such interventions is that a planned and managed period of time will not only place less demand on the system than an acute crisis but that its overall duration will typically be shorter than otherwise accumulated. Carer support may either be episodic or ongoing. Carers may offer mutual and reciprocal support developing their own resources to relieve demands on formal care, or carer support may be provided by health professionals, with the assumption that these may have more capacity (e.g. if located within the community) than is available within the acute care system.

The challenge in relation to targeting carers lies in the difficulty in attributing the impact of interventions specifically to benefits to acute care during times of winter pressure. The value of informal care in general to the health and social care system is well recognised and has been quantified. However, this contribution is typically conceived as relating to the whole year around, not specifically to times of system pressure. Carer interventions can be attributed to winter pressures where specific winter pressure funding is targeted at carers as beneficiaries. The challenge for evaluation may lie in the reverse direction with benefits from specific winter pressure interventions for carers delivering benefits that persist beyond the critical winter pressure period. The artificial separation of budgets, both across purposes and across organisations makes attribution of benefit to carer interventions particularly challenging.

Targeting carers has its own separate research agenda aside from winter pressure responses. Almost all the initiatives are not specific ‘winter pressures’ approaches. However, we have identified their use as winter pressure measures or as recipients for winter funding. This fact has been used to identify the candidacy of initiatives in this report for which we have subsequently sought more substantive and plentiful evidence. As for other initiatives, a key evaluation issue relates to whether to undertake evaluation specifically of carer initiatives within a winter pressures context or whether to evaluate them more widely on the understanding that they may be particularly beneficial when the health system is under particular strain as exemplified by winter pressures.

Recommendations for further research

The scoping review, published in 2021, concludes by identifying a potential need for ‘future research to consider a realist review of the literature on delayed hospital discharge to understand the context, mechanisms of impact, outcomes and theories of change, given that addressing a delayed discharge is a complex problem’ (p. 25). 3 Our own exploration has confirmed both this need and its potential contribution. In providing brief summaries of how the individual interventions and initiatives might contribute to relieving winter pressures, we have identified certain common underpinning mechanisms such as reassurance of the older person, establishment of clinical need, provision of a second/specialist opinion, transfer of venue and so on, which could be operationalised well within a realist review.

We would add that rather than focus on a ‘single sector and reactive approach to addressing delayed discharge’ (p. 26),3 our evidence summaries provide initial justification for including upstream anticipatory approaches viewed holistically across the widest possible array of sectors as they contribute within a whole system evaluation frame. However, it must be noted that, to remain both feasible and productive, the holistic lens required will necessitate a clear targeting and prioritisation of a focus for synthesis. The potential contribution of a realist approach is further attested to by identification of an existing realist synthesis in the specific area of intermediate care. 174 Another option discussed with the stakeholder is an implementation review of discharge interventions at the hospital/social care interface. The implementation review would focus on the factors that facilitate the implementation of the discharge interventions and their transferability, spread and sustainability using implementation studies where available or implementation data from other studies.

Strengths and limitations

This review benefitted from ongoing stakeholder involvement throughout the project to ensure that the deliverable is closely targeted to the requirements of the commissioner. Winter pressures is a broad topic that runs the full gamut of public health and health service provision from upstream public health approaches through to delayed discharge and overwhelmed capacity of social care provision. Detailed consultation helped to focus on the specific issues that have been found to be of strategic importance, as blockages to the entire system, that relate to health services reconfiguration and that command ongoing press and public interest and concern. Therefore, a strength of this report is its specific focus on discharge arrangements and on IC.

It soon became apparent, however, that health service and social care use takes place within a complex whole system. Focusing attention only a model that assumes admission as an inpatient as a default fails to capitalise on developments such as same-day services that seek to bypass any form of overnight stay. Furthermore, assuming presentation to an emergency department (ED) fails to take into account use of other points of access, particularly those that aim at community-based hospital avoidance. A strength of this review is its facility to extend beyond hospital-based interventions, a limitation openly acknowledged for previous scoping reviews. Indeed, inclusion of a significant body of community and voluntary service initiatives moves much closer to the concept of IC, although not necessarily badged as such. Broadening the coverage of the taxonomy, even given exclusion of any workforce or modelling interventions, necessarily holds consequences in the inevitable trade-off between breadth and depth. The review team was only able to focus on indicative items of evidence, namely systematic reviews and major UK studies. Between these items of evidence, privileged for rigour and relevance respectively, and the many identified case studies, privileged for currency and creativity, lies an extensive unacknowledged supply of studies exploring various aspects of admission avoidance, hospital discharge and IC provision. To this extent, the picture afforded by the individual intervention summaries is more a guide to evidence availability rather than a comprehensive snapshot of research activity. We encourage any research team that follows up any individual intervention summary to invest significant time in developing a more comprehensive picture of that topic. Nevertheless, the purpose of this report is to indicate promising areas for further research activity and so does not require a comparably comprehensive approach to a review that identifies effective interventions.

This review was conducted by three experienced reviewers with a collective portfolio across multiple health services and delivery review topics. This expertise was particularly important in this context because of the premium that the methodology placed on rapid identification and summary of indicative items – a particular strength of all three individuals. The three reviewers possessed the skillset to evaluate and appraise individual reviews and studies as identified but the rapid timescale and broad coverage of the review meant that this could not be performed formally and comprehensively for all retrieved items and could only be carried out opportunistically as reading of specific individual studies took place. The short timescale also meant that the team had to identify a collective point at which to draw a line under identification of further initiatives and to move on to taking stock of items already found and documenting their characteristics. In addition, time spent on each intervention type could not be standardised and so variability in coverage and depth reflects both different levels of research activity and different degrees of ease of study identification.

Finally, we must acknowledge the limitations of the taxonomy. In seeking to characterise each intervention within at least one place within the taxonomy, we have often simplified their characteristics. Interventions are frequently multicomponent and therefore typically include multiple headings from the taxonomy. In a few cases, where interventions possessed multiple defining characteristics or where discrete evidence on the intervention was less plentiful, we would duplicate entries across the taxonomy. However, our general intention was to try to make the intervention categories self-defining and discrete. A notable exception was the cross-boundary work on hospital avoidance, which necessarily appears in both structural interventions and community provision – in the former instance having the principal interpretation of admission avoidance and in the latter having the broader interpretation of acute hospital avoidance (both EDs and inpatient admission). In pursuing the further application of the taxonomy, we would encourage indexing interventions exhaustively across multiple taxonomy headings, potentially as a prequel to further work unpacking the mechanisms that underpin multiple intervention types attesting to the likely value of a planned realist approach.

Conclusions

We conducted a mapping review of evidence specifically related to winter pressures. It was challenging to attribute initiatives as a specific response to the winter pressures agenda, given both the lack of specificity and the artificiality of winter pressures as a binary, time-defined construct. This conception becomes progressively less useful as system pressures become a perennial issue, not a seasonal one. Few initiatives identified were specifically implemented as a response to winter pressures.

In seeking an evidence base, we extended a process of evidence summaries to general initiatives used to relieve pressures on admission and discharge. Hospital at home and D2A were well-supported initiatives but other responses, while also heavily used, were based on limited evidence. Across all intervention groups, we confirmed the lack of studies considering patient, family and provider needs when developing interventions aimed at improving delayed discharge; identified from previous reviews of discharge and of patient and care involvement. This is a particularly important omission, given the heavy dependence of formal healthcare systems on the contribution of informal caregivers, whether when in hospital or in the community.

Additionally, few studies measure the impact of interventions over a long time. Short-term results can appear promising, but evidence for longer-term sustainability is notably absent. Hospital avoidance and delayed discharge requires a whole-system approach. We conclude, alongside a previous scoping review, that it is imperative to consider the whole system to ensure that implementing an initiative in one setting does not simply move the problem to another setting.

Patient and public involvement

Stakeholder involvement was undertaken to ensure that the review was informed by the views and experiences of stakeholders at the Department of Health and Social Care and NICE, as well as with a standing advisory group of patients and cares.

The consultation aimed to:

• highlight the practical and personal challenges experienced by those planning or using health services during times of exceptional demand

• inform the second phase of our searches by identifying specific strategies, initiatives and interventions that target discharge planning and IC in the context of winter pressures

• identify gaps in the literature with a view to compiling a future research agenda.

In the review proposal, we aimed to consult with stakeholders from the Department of Health and Social Care and NICE and a standing advisory group of patient and carers. The key stakeholders were involved in the proposal for review and commented on a draft map when they asked for findings to be included from research.

A standing advisory group of patients and carers met once during the project to discuss the group’s knowledge about different interventions that could be used to mitigate winter pressures and the research team found it especially interesting to elicit group opinions on which interventions might require research funding. The rapid nature of the project meant that discussions with the standing advisory group had to be concentrated within a single meeting. For a long project, we would include the standing advisory group earlier, enabling them to input into the scope of the review and discuss review findings, in particular through discussing the research gaps identified.

Equality, diversity and inclusion

As a secondary data study, our review did not include any research participants. We were, however, inclusive in the studies we selected.

Contribution of authors

Anna Cantrell (https://orcid.org/0000-0003-0040-9853) (Research Fellow) was the lead reviewer and was involved in protocol development, information retrieval, study selection, data extraction and report writing.

Duncan Chambers (https://orcid.org/0000-0002-0154-0469) (Research Fellow) was involved in study selection, data extraction and report writing.

Andrew Booth (https://orcid.org/0000-0003-4808-3880) (Professor in Evidence Synthesis) was the methodological adviser and lead protocol developer, was involved in study selection, data extraction and report writing, and was the guarantor of the review.

All authors commented on drafts of the protocol and report.

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are available in the toolkit on the NIHR Journals Library report publication page at https://doi.org/10.3310/KRWH4301.

Primary conflicts of interest: Andrew Booth is a former member of the NIHR HS&DR Funding Committee and the NIHR Systematic Reviews Advisory Group, and Convenor of the Cochrane Qualitative and Implementation Methods Group.

Data-sharing statement

Any additional data not included in this report and its appendices are available on request. All queries should be submitted to the corresponding author.

Ethics statement

This review did not involve the collection or analysis of any data that were not included in previously published research in the public domain. It was therefore exempt from formal ethical review by the University of Sheffield Ethics Committee.

Information governance statement

NIHR is committed to handling all personal information in line with the UK Data Protection Act (2018) and the General Data Protection Regulation (EU GDPR) 2016/679. Under Data Protection legislation NIHR is the Data Processor; the Department for Health and Social Care (DHSC) is the Data Controller, and we process personal data in accordance with their instructions. You can find out more about how we handle personal data, including how to exercise your individual rights and the contact details for DHSC’s Data Protection Officer here.

Department of Health and Social Care disclaimer

This publication presents independent research commissioned by the National Institute for Health and Care Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, MRC, NIHR Coordinating Centre, the HSDR programme or the Department of Health and Social Care. This monograph was published based on current knowledge at the time and date of publication. NIHR is committed to being inclusive and will continually monitor best practice and guidance in relation to terminology and language to ensure that we remain relevant to our stakeholders.

Disclaimers

This article presents independent research funded by the National Institute for Health and Care Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, the HSDR programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, the HSDR programme or the Department of Health and Social Care.

Patient and public involvement group input

The PPI group’s knowledge and thoughts about each scenario are provided below.

CSB – Clinical audit

Changing staff behaviour: education of staff

CSB – Protocols/guidelines

CSB – Quality improvement programmes

CSB – Quality management systems