Our framing of multiple long-term conditions and analysis of the portfolio of evaluations

Frameworks for evaluating service innovations for people with multiple long-term conditions

Given the high proportion of people in the population who live with multiple long-term conditions, it is likely that any healthcare service innovation will have some effect on that group. The essence of this overarching study can be thought of as always asking the question whether and how providing effective and efficient care for a person with two or more long-term conditions might differ from providing care for people each living with just one of those conditions. This difference is where the significance of multimorbidity lies for patients, healthcare professionals, NHS providers and commissioners, and policy-makers.

The high degree of specialisation in health care exists for good reasons, including the ways in which medical practice has evolved over time to consist of many specialties that are reflected in training, professional practice and identity, and the broader organisation of health services.32 Patients want to be treated by healthcare professionals who are experts in the specific relevant fields of medicine or surgery. But focusing on single diseases risks missing the combined impact they have on an individual living with multiple conditions and often overlooks the need for co-ordination of a patient’s care. A patient quoted from a consultation with the Patient and Carer Network of the Royal College of Physicians33 captured the point succinctly:

Patients want to see a specialist in the bit of you that is going wrong but what would be welcomed would be seeing a doctor who sees a patient as a whole person.33

As noted earlier a member of our research team (AN) undertook a review of the literature to identify frameworks to help to structure thinking about evaluating health and care services for people with multiple long-term conditions. We found two frameworks to be particularly helpful: the SELFIE framework31 and the House of Care Model.30 There are numerous other frameworks and models that offer additional insights (see Report Supplementary Material 1).

The SELFIE framework

The SELFIE framework was originally created to support the development, description, implementation and evaluation of integrated care for people with multiple long-term conditions. The patient and their environment are at the core of the framework and concepts of integrated care for people with multimorbidity are described at the micro, meso and macro levels respectively on the basis of six WHO components: service delivery, leadership and governance, workforce, financing, technologies and medical products, and information and research (Figure 1).31 The SELFIE framework is comprehensive, and we note that the 50 individual elements identified within it could just as well be used to characterise integrated care for single morbidities.

FIGURE 1 The SELFIE framework for integrated care of people with multiple long-term conditions. Reproduced from Leijten et al., 2018.31 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) license, which permits others to distribute this work provided the original work is properly cited, the use is non-commercial, and it is not remixed, transformed or built upon. See https://creativecommons.org/licenses/by-nd-nc/1.0/. (Figure not reproduced here; see the original article.)

The House of Care Model

The House of Care Model30 provides a simplified framework, which helpfully complements SELFIE, having been developed to structure thinking about services for people with long-term conditions, explicitly including those with multiple long-term conditions. The House of Care Model focuses on achieving personalised, holistic care for each patient and comprises four main elements that form the ‘walls, floor and roof’ of the House of Care within which personalised care planning can be achieved: healthcare professionals committed to partnership working; engaged, informed patients, to which we would add their informal or formal carers; organisational processes; and responsive commissioning of health services.30 The House of Care metaphor is used to emphasise that a whole-system approach is required, recognising the interdependency of each part and the various components that need to be in place to hold it together.

We refer to both the SELFIE and House of Care Model frameworks when discussing our research and its findings in the rest of this report.

The BRACE portfolio of evaluations

Our core research material for addressing the ORQs for this study is the findings of the 10 individual rapid evaluations undertaken by the BRACE during the 5 years from April 2018 to March 2023. The 10 evaluations undertaken by the BRACE team are set out in Table 1 in the Introduction, along with the corresponding references where their findings can be read in full.

In the event, none of the BRACE evaluations was suitable to address the sixth research question ORQ6 (see Chapter 2, Aims, objectives and research questions of the BRACE overarching study), that is about outcome measures specifically. This reflects the fact that none of the service innovations we evaluated had been designed with people with multiple long-term conditions as a particular target group in the population for whom the innovation was intended. All the innovations evaluated would affect people with multiple long-term conditions as well as other service users. Nevertheless, considering the first research question ORQ1 (see Chapter 2, Aims, objectives and research questions of the BRACE overarching study) about the health and care requirements of people with multiple long-term conditions also has implications for the types of outcome measures that might be most important to people living with multiple long-term conditions.

Developing a typology of BRACE evaluations

Some BRACE evaluations were not directly concerned with the impact of service innovations on people with multiple long-term conditions, but others were. We therefore conceptualised our BRACE studies as being of two main types, as illustrated in Table 2:

• Type 1 evaluations are those in which people with multiple long-term conditions are not the main focus of the service innovation being evaluated, but their perspective is one lens through which to view and assess the innovation. For example, BRACE undertook an initial rapid evaluation of acute hospitals managing general practice services, which is a form of vertical integration.2,34 The initial evaluation2 focused on why and how such vertical integration was being implemented across three case study sites; how services were affected; and the impact on primary and secondary healthcare workforces. Better care co-ordination, pathway redesign and sharing of patient data were all opportunities that have been linked with vertical integration and are highly relevant to care for people with multiple long-term conditions, but they were not the principal focus of the evaluation. BRACE also undertook a second phase of evaluation9 of where acute hospitals are managing general practice services, looking at the impact on patients’ experience of care and their use of hospital services, which includes a particular focus on whether these are different for patients with multiple long-term conditions compared to others. We characterise this type of follow-up evaluation as a Type 1B rapid evaluation, to distinguish it from the more broadly aimed, initial, rapid evaluation, which we label Type 1A.
• Type 2 evaluations are individual BRACE studies focused from the outset on care for people with multiple long-term conditions. Examples of such projects were the quantitative evaluation of ‘telephone-first’ primary care and the experience of this service innovation by people with multiple long-term conditions,4 and the qualitative evaluation of digital-first primary care for those with multiple long-term conditions.8

In the Type 1B evaluation and two Type 2 evaluations we undertook, we included in the respective protocols questions that specifically sought to address whether people with multiple long-term conditions were being affected by the service innovation differently from other service users. Thus, in the Phase 2 evaluation of vertical integration of acute hospitals with GP practices9 (Type 1B study), our research questions – and hence the evaluation findings – included:

• What impact is vertical integration having on secondary care utilisation (outpatient attendances, A&E attendances, all inpatient admissions, emergency inpatient admissions, inpatient admissions for ambulatory care sensitive conditions, bed days, readmission within 30 days of discharge)? Does this impact differ for people with multiple long-term conditions compared to other patients without long-term conditions, or living with a single condition?
• What impact is vertical integration having on the patient journey regarding access to and overall experience of care? How does the experience differ for people with multiple long-term conditions compared to those living with no or one long-term condition?

In the evaluation of telephone-first primary care,4 we investigated the following:

• Considering people living with multiple long-term health conditions only, does a telephone triage approach affect how quickly people can see or speak to an appropriate primary care professional?
• What is the size of that effect relative to the effect on people contacting a general practice who do not have multiple long-term health conditions?
• Are there any subgroups of the population with multiple long-term health conditions who are particularly affected (either positively or negatively) in terms of how quickly they see or speak to an appropriate primary care professional, both generally and when a telephone triage approach is used?

In respect of digital-first primary care,8 we evaluated the following:

• What is the experience of digital-first primary care for health professionals and stakeholders (including academics, policy-makers and primary care providers), both before and during the COVID-19 pandemic?
• What is the impact of digital-first primary care on the nature of consultations for unpaid carers/patients with multiple long-term conditions, from the perspectives of health professionals and stakeholders, which includes aspects such as the health professional(s) spoken to, timeliness of care and continuity of care?
• What are the advantages or disadvantages of digital-first primary care for patients with multiple long-term conditions, and their unpaid carers, as reported by health professionals and stakeholders?
• What lessons can be learnt from staff and stakeholders, for future service delivery for patients with multiple long-term conditions in primary care? Are there individual groups within the community where there is particular learning for future service provision?

In addition to these focused research questions in three BRACE evaluations, the other seven evaluations provided the opportunity for interviewees consulted in those studies, and documentation reviewed in those studies, to yield information pertinent to the care of people with multiple long-term conditions. Sometimes that happened and sometimes not, but even the latter cases can be revealing of the attention given (or not) within service innovation to the concerns of this (substantial) group of the population. Consequently, in our analysis of the BRACE portfolio of evaluations, we actively noted when mention of people with multiple long-term conditions was absent, as well as when it had occurred.

Cross-project analysis of the portfolio of BRACE evaluations

As explained in the Methods chapter, members of the BRACE research team undertook a cross-analysis in August 2022 of the reports of the seven BRACE evaluations that had than been completed. The cross-analysis was based on the mixed-methods matrix approach described in O’Cathain et al.35 To complete the cross-analysis, the remaining three BRACE evaluations were added in January 2023 by the researcher who had led the first stage (JSu) in discussion with researchers from each of those three studies. To facilitate the cross-project analysis, we used a data extraction table (see Report Supplementary Material 2) to capture relevant information from each BRACE evaluation. Table 3 maps the data extraction headings to the research questions.

When members of the research team (JSu, BL) extracted evidence under these headings from the final reports of the first seven BRACE evaluations completed and discussed it with members of those evaluations’ respective research teams (at an internal BRACE team analysis workshop in August 2022), it became clear that issues related to people with multiple long-term conditions were seldom raised by stakeholders unless they had been specifically prompted to do so.

The findings of the first six Type 1A evaluations listed in Table 1 [the evaluation of Women’s Health Hubs (WHHs) had not been completed at that time] offered some insights, particularly into how the service innovations in question were or were not meeting the specific health and care requirements of people with multiple long-term conditions, and what some of those requirements might be. Additionally, the telephone-first primary care evaluation4 (a Type 2 evaluation) had found that all patients had benefited by speedier access to a healthcare professional, and that this improvement was similar regardless of the number of long-term conditions with which a patient was living. The research team was also expecting the ongoing evaluations of digital-first primary care8 (Type 2) and the Phase 2 study of hospitals managing general practice9 (Type 1B) to yield findings directly pertinent to care of people with multiple long-term conditions. But it was clear that some additional research focused on aspects of innovation in care for this section of the population would be valuable.

At the August 2022 overarching study workshop, the research team discussed and prioritised options for further research that could complement the work already done and add most value within the remaining time and resources available to the BRACE. Evidence from the BRACE portfolio largely fell within three of the six segments of the SELFIE framework, namely:

• service delivery
• technologies and medical products
• workforce.

We considered where within those segments a rapid literature review might yield most value and identified remote monitoring of people with multiple long-term conditions as promising: what might good remote monitoring for people with multiple long-term conditions look like? This decision was also informed by the strong policy push from NHS England to adopt remote monitoring more extensively in the NHS post-pandemic, including as part of ‘virtual wards’.36

The BRACE evaluations of COVID oximetry at home6 and AI in social care7 both examined aspects of the role of remote monitoring outside healthcare settings but found little specifically about people with multiple long-term conditions. Telecare technologies such as those are, however, potentially highly relevant to people with multiple long-term conditions. We therefore focused part of our further research on reviewing the published evidence on one specific type of service, namely remote monitoring. This would enable us to see, within the bounds of a limited and rapid additional study:

• what evidence already exists of whether and how people with multiple long-term conditions have needs or wishes (for remote monitoring) that differ from those of other people (ORQ1)
• the outcomes (relevant to remote monitoring) that evidence refers to (ORQ6)
• how far such different needs are being met (ORQ2)
• how effectively (ORQ3), and whether people were being better supported to manage their own care (ORQ4) as a result.

Given the low profile of people with multiple long-term conditions in most of the service innovations BRACE has evaluated, we sought to understand more about how far the needs of those people are driving the approach to integration being taken by NHS Integrated Care Systems and the organisations within them. We re-contacted some interviewees from previous BRACE evaluations to seek their views on this and approached a sample of national-level stakeholders (as explained in the Methods chapter) to interview them about what a system of service innovation that does prioritise people with multiple long-term conditions might look like, and how this might be achieved.

The value of the BRACE seeking to have an overarching focus on people with multiple long-term conditions was evidenced in how this increasingly served as a prompt to the individual evaluation teams when scoping a study, being questioned about the focus of the evaluation approach, and which data sources to use to try and gain insights into the experiences of this patient population. We noted that, in most cases, the needs of people with multiple long-term conditions were not presented to the evaluation team by commissioners (local or national) of the studies as being a core concern, despite the active attention accorded to this group in policy circles, as discussed earlier.8–14