The context of embedded researcher interventions to support local policies and practices

Embedded researchers are thought to be a promising approach for enhancing the production and use of research within policy and practice settings across a range of public and commercial sectors. [1–4]. They are theorised to bridge organisational siloes and to address gaps between research production and research engagement and use, with the intention of making research more beneficial and accessible to inform decision-making [1,2,5,6]. This is facilitated by embedded researchers being co-located across organisations. They may work ‘inside host organisations as members of staff, while also maintaining an affiliation with an academic institution’ [7,8]. However, variations exist around how co-location and co-affiliation are operationalised in practice, particularly given the growth in remote working arrangements since the COVID-19 pandemic [3].

There is growing interest in utilising embedded researcher roles within local public health contexts to support leaders in developing evidence informed local policies and practices for improving population health [3,6,9,10]. Within England, local authorities are responsible for planning and commissioning most local public health services. These services and measures are intended to improve the health and wellbeing of local populations, reduce health inequalities, ensure health protection functions are in place, and provide healthcare advice to the population. Services include those that are mandated (e.g., provision of sexual health testing services), and others that are discretionary depending on local public health needs (e.g., smoking cessation services). Decisions are made by elected officials, who are supported by practitioners and influenced by politics and public perceptions [11]. Research evidence that influences their decisions needs to be useful, accessible and overcome the barriers of dissemination, interpretation and implementation between academic research production and research use [12]. There is a need for public health evidence to have relevance to local populations, which may differ from national-level evidence and policies [13]. Co-produced research with stakeholders is encouraged so that evidence is more aligned with practice [14]. Embedded researchers can be viewed as embodying a form of co-production, where research is conducted in-situ with potential users [2]. However, other modes of the role exist and they might also influence and facilitate research conducted by others, or enhance the use of research in decision-making [3]. For example, embedded researchers may be the designated adviser on undertaking and using research [5,9], or undertake activities to build an organisation’s capacity to conduct, influence or use research [8,15].

We previously identified 35 evaluations of embedded researcher interventions in public health settings published between 2000 and 2021 [3]. These included evaluations of researchers embedded into practice (or policy) settings and, less commonly, evaluations of practitioners or policy-makers embedded within research settings. For example, health policy-makers in a regional government being seconded into a university [16]. The UK context was the most represented country setting (with 13 evaluations), followed by Australia, Canada and the USA (5–6 each). While the intervention is difficult to define [3], has wide variation, and there is limited understanding of how it is implemented in practice [17], there are nevertheless some common activities undertaken and mechanisms enacted that support the aim of bridging the gap between research and policy across diverse decision-making contexts. While the focus of the research presented in this study is in England, the processes described below are likely to have broader cross-country salience.

Conceptual framework to understand embedded researcher interventions

Our earlier work [1,5] developed a framework to describe common processes and broad outcomes of embedded researcher interventions through six elements, drawn from mixed-method primary and secondary research. This is summarised as follows:

1. 1). The intervention begins before a researcher is actually embedded, through the development of relationships between the organisations affiliated with the intervention (e.g., a policy or practice setting and research organisation) in order to co-create the role between stakeholders. Co-creation is not necessarily with the embedded researchers themselves.
2. 2). An embedded researcher’s initial duties involve undertaking activities to maximise their visibility and help them become an insider within the host organisation.
3. 3). Embedded researcher activities may involve a balance of conducting, brokering, and facilitating research. ‘Conducting research’ is understood as activities that directly involve the production of research; ‘brokering research’ is understood as activities involving the gathering, using and interpretation of research to inform decision-making; and ‘facilitating research’ is understood as activities that support others to lead or and conduct research.
4. 4). Early and incremental successful changes include the development of networks and infrastructure, such as improved collaborative networks and recognition of the embedded researcher as a local advocate of research. These represent important precursors to cultural changes of organisations around conducting, brokering and facilitating research.
5. 5). Longer-term impacts and signs of cultural change include instances where research is conducted (or commissioned) in a way that is more applicable to the stakeholder context and is used to inform local decision-making.
6. 6). Although generally not observed in the literature, longer term systemic changes are theorised to follow, leading to strengthened cultures of evidence informed decision-making and, ultimately, improvements in population health.

The above trajectory is moderated by a range of contextual factors as well as the individuals within the embedded researcher roles. The potential for one person to achieve such culture change is limited within the large and complex remit of local population health within the context of societal challenges (e.g., the COVID-19 pandemic, austerity to public services). In a newly-created embedder researcher role, both role holders and the public health teams they work with must navigate ambiguity of tasks and performance expectations, and together understand the issues that research might be able to address before appropriate actions are taken. Furthermore, while ambiguity can provide autonomy for tailoring embedded researcher roles to individual contexts, skills and interests [18], it may bring challenges to roleholders working in isolation [18] and navigating their professional identity [19–21]. Moreover, the challenge of working within a position that has ‘culture change’ as its long-term goal is likely to create substantial workplace and wellbeing challenges.

Study aims

Previous research investigating how embedded researcher interventions are implemented in public health, describe some activities that are undertaken though do not probe into what activities they undertake on a daily basis [2,6,22]. Allowing embedded researchers the space to record and reflect on their experiences could provide valuable insight into their daily activities and experiences and expand our understanding of such interventions. Using evidence from a diary study we aimed to understand: 1) how embedded researchers intersecting academia and policy and practice spend their time; 2) how these activities and achievements map to our conceptual framework of how such roles enhance the production and use of research; and crucially 3) how the embedded researchers perceive the importance of the work that they are conducting and the milestones they achieve.

Study context

In this paper, we explore the daily activities and perceived challenges and enablers of embedded researchers within public health local authority settings in England. This study was a component of a mixed-methods research project that examined the influence of embedded researchers within public health settings [1]. The embedded researchers involved in this study were supported by the National Institute for Health and Care Research (NIHR) Clinical Research Network (CRN) and were embedded within English local authority (LA) public health teams. All posts were expected to contribute to enhancing research production and use within public health teams, although the design of the role differed across settings depending on the needs of the LA and on the needs of the NIHR and other organisations involved (e.g., universities). We refer to these embedded researchers as Public Health Local Authority Research Practitioners (PHLARPs), and use the terms PHLARP and ‘embedded researcher’ interchangeably in the remainder of the paper. Two began their posts in March 2020, and the remaining cohort started during spring 2021. The posts were initially for one year, though typically were extended by up to three years. Their posts were developed locally and across English regions. Physical placement and line-management varied between local authorities, regional Clinical Research Networks (CRNs) and Universities [1]. Although the intervention was implemented heterogeneously, the use of a diary study will help us to understand heterogeneity in daily practice.

Ethics approval and consent to participate

The study was reviewed and approved by the UCL IOE Research Ethics Committee (REC 1540) and all participants gave written informed consent.

Sample

All PHLARPs were invited to take part in the study including through an online PHLARPs network event, and email invitations. A £50 incentive was offered on completion of the diary study for a week (5 daily entries; although these were occasionally spread across weeks depending on the working pattern of the embedded researcher). Nine participated in wave 1 (including one pilot participant), and they all were invited by email to wave 2, with non-response followed up at least once. Five participated in wave 2. Two participants had left the role before the follow-up study. Most participants reported being predominantly based in a local authority public health team, although two reported being affiliated to local authority teams while being predominantly based in a university (these could be considered to be embedded remotely, see [3]); all were affiliated with a regional CRN (seven CRNs were represented across the 9 participants in wave 1, and four in wave 2).

Data collection

Diary studies are gaining prominence for collecting data on behaviours, experiences and perceptions near the time they took place and allowing sufficient detail of recall [23–25]. They are particularly useful for data that is difficult to observe and enables comparison at differing timepoints [25]. However, diaries rely on commitment and input of the participants [25]. Diaries have flexibility to capture different types of data, for example, qualitative data on work activities [25] and quantitative data on time use [26]. Challenges to qualitative data capture include the varying detail of responses, potential for under-reporting by participants, and labour-intensive data analysis [27]. In quantitative data collection there are challenges of interpretation of pre-specified categories by participants and having a lack of depth of data [26].

In this study, a one-week online diary survey (via Qualtrics) was used for mixed-method data collection and aimed to balance sufficient data capture without becoming too onerous on participant’s time. It was undertaken in two waves during 2022 (May to August) and 2023 (April to May) following piloting on one participant (the first wave had a longer timeframe to boost participation). Participants were provided with an information sheet describing the purpose of the study, the expectations on the time they should allocate to completing the study, and brief instructions on the content expected within their response next to the diary questions (see S1 Appendix). Participants were encouraged to fill in an online survey towards the end of their working day for 10–15 minutes for five consecutive working days, with an additional 10 minutes for the last day. They were informed that their responses would be confidential and anonymous and that their words would only occasionally be quoted to support explanations of our findings. The study information sheet included a weblink to resources on managing work stress in case participation raised anxieties about work.

The diary comprised closed and open-ended questions about the work activities undertaken during the day, feelings about the day and any informal connections they had with local authority colleagues. Additional questions at the end of the week prompted reflections on time spent on different activities, planned work that had not been achieved, interactions with local authority colleagues and how participants viewed their job role. The formatting of the second survey was slightly adjusted to prompt provision of more information around working activities, and on the enablers and challenges of the job role. Closed questions were asked as a way of prompting participants to recall tasks and feelings about their day, prior to open-ended questions requiring examples and descriptions of activities, informal connections and feelings. Each day participants were asked to indicate which activities they had been doing in their job role, and there was a choice of nine responses, including ‘other’ (the categories were developed by the research team). At the end of the week, they were asked to estimate the portion time spent on each activity, with totals summing to 100 and using the same nine responses. Each day participants were asked how much they agreed with statements about their day (1–5 scale): familiarity with the activities they encountered; high interest in work; high time pressure; encountering conflict with others. The daily open-ended questions asked participants to describe in greater detail (i) up to two activities, by providing a description of the activity, purpose and context; (ii) informal or social connections with colleagues and their perceptions of these; and (iii) how they feel about their day. At the end of the week, they were asked to reflect on their activities, interactions, and perceptions of their role.

Data analysis

Analysis was undertaken in Excel and NVivo 14 by one researcher (CS), with checking by a second researcher (DK). Quantitative data was analysed through means, frequencies and comparisons. Qualitative data were thematised using template analysis (King 2012), which involved iteratively developing a coding template based on an initial set of 4 diary entries from wave 1 and applying this to subsequent diary entries in wave 1 and reviewing the codes again (supported by a descriptive codebook). One researcher (CS) developed the initial template and applied the codes. Two researchers (DK, RE) each tested the template on two different diary entries, and provided input to the final template. The template was also informed by collective discussions of findings across the wider research project. Wave 2 entries were coded to the template from wave 1. Brief sketches were developed of each participant to aid analysis of themes, and support comparisons of responses at the two time-points. Template analysis is a technique to provide a structured approach to data handling (King 2012) and is based on developing a set of codes from analysing a subset of data, which are used as a template for further coding and analysis. The codes are modified as analysis progresses further (Howard et al. 2008).

A week in the life of an embedded researcher

Embedded researchers (PHLARPs) were provided with a framework for recording distinct types of activities, and the results underscore that each day is different, and most undertook a variety of different activities throughout the week. Table 1 shows that attending meetings was an activity undertaken by all embedded researchers on most days, and on average accounted for over a quarter of their time (26%; ranging from 10–60% of time spent in a week). Other common activities that embedded researchers spent an average of 10–15% of their time on included presenting evidence and related preparation activities, providing information or research expertise, requesting information, organising information and collecting or analysing data (range 2–40%). Self-learning, training others and other activities were less common and comprised 1–6% of the time (range 0–13%). Table 1 also shows that the average time spent on various activities were broadly similar for both waves.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Activities undertaken over the week.

https://doi.org/10.1371/journal.pone.0328996.t001

Activities of significance for embedded researchers

The following eight themes of activities were developed from the participants’ responses when prompted to describe up to two activities: building linkages; applying for funding; planning research; public engagement; undertaking research or supporting research use; supporting skill development; self-development; and general administration (including time to catch up and general team meetings). These themes are described with examples in Table 2. Unlike the data presented in Table 1, which were collected through a pre-determined framework, these data allow us to examine activities of significance for embedded researchers. However, it was not always possible to determine the activity undertaken by the PHLARPS from their descriptions; activities that were planned and not undertaken that week also contributed to the themes. Where meetings were described by participants, we tried to code these by the type or purpose of the activity. For example, a meeting discussing early ideas for a funding bid, was coded as “applying for funding”

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Key activities described by PHLARPs and example statements.

https://doi.org/10.1371/journal.pone.0328996.t002

In terms of the most common activities, building linkages is a key activity taking a number of forms (described later in the text). Most of the nine participants provided details about attending a team meeting during the week (n = 8) and provided training or support to others (n = 8), of which half also enabled others to use research (n = 4). (Data corresponds to a participant mentioning this one time or across either wave). A majority described undertaking an activity related to research funding (n = 7) and planning research (n = 7), and other activities related to research facilitation and research production (n = 7). Six reported involvement in public engagement activities. Other activities described by fewer than half include activities to support data or research use, data governance, data preparation, and self-development activities.

Our analysis of these activities (Table 2) compared with the stages within our conceptual framework shows the that embedded researchers were (i) undertaking activities to maximise their visibility and become embedded (stage 2); and (ii) undertaking a complex balance of activities around the generation, brokering and facilitation of research (stage 3); but also (iii) were operating in a highly relational role where early gains were being observed in the steps towards involvement in research production and use (stage 4). We describe these in further detail below.

The roles were also observed to be working across a wide variety of public health initiatives from those related to the social determinants of health (e.g., early years development and learning, housing, and food banks) to those on particular health conditions (e.g., social isolation, dementia and obesity)

Becoming embedded and addressing needs

The choice of activities described in Table 2 underlines the importance not only of activities that could be considered core to the role (conducting and facilitating research and research use), but illustrate the diffuse process of becoming embedded. Despite having been in post for at least a year, several activities appear to involve a PHLARP’s engagement with colleagues and networks to establish the nature of their embedded researcher role (see later section) and to understand the culture of the local authority. For example, while we might expect a needs assessment/situational analysis to be conducted in the early stages of ‘becoming embedded’, analysis of diary entries revealed that assessing needs was a more continuous process than a discrete starting point. The PHLARPs described supporting skill development through identifying training needs of LAs, teams or individuals. This could be from established meetings, for example a quarterly meeting to discuss training and support, through to more discrete activities. However, it was also clear that the PHLARPs progressed beyond assessing needs to addressing needs through, for example, identifying or organising training activities such as webinars, creating guides for data analysts and researchers, and updating online resources to raise awareness of training. Capacity building took the form of sharing knowledge among peers and colleagues, providing support for those new to research. This could be informal or developed strategically: “The meeting helped identify a mutual objective in identifying what training/learning opportunities already exist and what gaps are needing to be bridged so research delivery can be further enhanced.”

Undertaking a complex balance of activities around conducting, facilitating, brokering and using research

Early gains in shifting organisational culture towards research

Three themes of activities are described here that demonstrate early and incremental gains in shifting organisational cultures towards research: Championing research across multiple stakeholders; building links and maximising visibility as core activities and processes; and developing infrastructure to support new research activities.

Joys and frustrations of the embedded researcher

Challenges, enablers and feelings encountered by the PHLARPs are described here within the themes of: In-person and remote working, the ambition of the role and daily joys and frustrations.

Daily joys and frustrations of the role.

The PHLARPS were asked to rate their daily feelings across a number of items (Table 3), and the daily feelings of PHLARPs were largely high familiarity and high interest with work, without being over-pressured and low conflict, though time pressure and conflict had greater fluctuations (as shown in the standard deviations). These observations were broadly consistent across both waves.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Ranking of daily average feelings across five days (High = 5; Low = 1).

https://doi.org/10.1371/journal.pone.0328996.t003

The diary narratives (open text entries) did not always correspond with reported agreements, particularly for time pressure and conflict. Within their diary narratives, all participants reported positive feelings of satisfaction or enjoyment, and most commented on feeling connected and supported (n = 8). However, the same number also reported some frustrations or disappointments and three described isolation or lack of connectivity in the role: “Working within the LA is very frustrating. There is no connectivity and a lot of duplication going on….I feel there needs to be a culture shift”. It was acknowledged within diary entries that creating change within public health teams would take time and needed wider support and system-level changes. Nevertheless, overall, there appeared to be a positive expectation that the activities of PHLARPs made an incremental contribution to enabling research production and use within LA public health teams: “Really happy as people are genuinely enthusiastic about our work and the possibilities”. Returning to the first sets of results, daily variation in the types of tasks undertaken appeared to be a core theme in the reflections of our embedded researchers, with descriptions of ‘busy’ days and comprised of ‘a range of activities’ frequently mentioned as components which helped to sustain interest in the role and ‘prevent burnout’, and as reflected by one embedded researcher: “The variety involved is both hugely enjoyable but equally terrifying.”

Strengths and limitations

Participants in the PHLARP scheme, and diary study, differed in the type of experience they brought to the role (such as research or practice or policy experience. A strength of this study is the insight gained into common activities and experiences that can exist among diverse embedded researcher roles. Further, linking broad types of activities and their purpose to a framework of how embedded researcher interventions are thought to work [1] also increases the applicability of the finding to informing similar roles within different contexts. Another strength of this study is the focus on the activities that were actually undertaken during two working weeks, at different timepoints and from the participants perspectives. However, the nature of a diary study means the data is self-reported, and participants differed in writing style, depth of description, and explanation for their responses. We interpreted the free-text responses and assigned to our coding template and others may have interpreted these differently (e.g., such as the type of research activity). At least one participant declined to describe any feelings when asked how they felt about their day, though at the end of the week shared challenges and benefits they identified around about the role. Some participants recorded brief entries that were uninformative for our analysis, for example, a two-word response of ‘Hump day’ (a colloquial term for Wednesday). There were also gaps in the quantitative data on feelings. The quantitative data was collected mainly as a prompt to facilitate reflection by the participants prior to them providing their own descriptions. It also provided a way to efficiently capture data in a short time each day. However, occasionally there was little triangulation between the quantitative data and free text entries. For example, one respondent rated their interest that day as high but stated “Felt okay today. Motivation to work was low”, suggesting that the diary entries captured both professional and personal reflections. Some respondents could potentially feel cautious in how candid they could be in free text entries or be unclear on the level of detail to write. Furthermore, we didn’t provide any ‘training’ on completing the diary and relied on instructions presented in the diary itself. Participants could describe two activities undertaken each day, which could have been notable for a time, importance or another reason. Sample size was an issue with this study, and there was substantial attrition between both waves of the diary. Among those who completed both waves, there were few perceptible differences between the perceptions of the roles and activities undertaken; we cannot discount the possibility that this pattern would change had the full sample participated in both waves. We could not infer statistical significance from the quantitative data owing to the small sample size. We also asked participants to reflect on the process of completing a diary. Most comments on participation were favourable, and negative feedback was confined to IT issues. As one PHLARP reflected: “It’s been a useful experience; it has always been the last thing that I’ve done of a day which I’m afraid may impact on the quality of response. But it’s a well-tailored way of getting insight. And it’s been rewarding for me also”.

The researchers conducted this research as part of a work programme of researching embedded researchers generally, and specifically the PHLARP scheme [1] and interacted with an advisory group for the project, presenting interim findings at meetings]. We believe these experiences increase the credibility of our findings.