Factors affecting recruitment and retention of nurses who deliver clinical research: A qualitative study

Abstract Aim To provide a better understanding of the factors affecting recruitment and retention of clinical research nurses. Design Qualitative exploratory design. Methods An on‐line questionnaire comprising open‐ended and fixed‐choice questions was completed by 121 clinical research nurses. Seven focus groups were held with a subgroup of 26 participants. Data were analysed using inductive thematic analysis. Results Participants were attracted to a research nurse post by an interest in research itself, a desire for a change or to achieve personal objectives. The majority expected to continue in a research post for the next 5 years, while others expected to move on to research management, a clinical post or retirement; few had ambitions to become an independent researcher. Factors identified in focus groups as leading to intentions to leave research included desire for further change, concern about loss of clinical skills, rebalancing family/work responsibilities, short‐term contracts, unsupportive employers and limited career progression.

, few have looked at what initially attracts nurses to a research post or what their current career intentions are. Published accounts of individual nurses suggest that their first research post was something they "fell into" rather than sought out and that their career in research was not consciously planned but evolved as opportunities arose over time (Grove, 2015;Sprinks, 2015). However, these are selected case examples which may not be representative of research nurses more generally.
Research on factors that affect retention of research nurses is also rare, though studies in clinical nursing are likely to have some relevance. In a study of three generations of nurses, for example, Robson and Robson (2015) found that intention to continue working in the NHS was associated with the combination of attachment to work, importance of work and (low) work-family conflict. Other studies have identified the following factors as associated with intention to remain in practice: perceived support from their organization (Masters & Liu, 2016), institutional initiatives to support their professional development (Bruyneel, Thoelen, Adriaenssens, & Sermeus, 2017;Duffield, Baldwin, Roche, & Wise, 2014;Kenny, Reeve, & Hall, 2016;Nowrouzi et al., 2015), a good work environment (Abou Hashish, 2017;Kenny et al., 2016;Laschinger, 2012) and success in or satisfaction with their career (Masters & Liu, 2016;Osuji, Uzoka, Aladi, & El-Hussein, 2014). Factors associated with high staff turnover include exhausting workloads (Havaei, MacPhee, & Dahinten, 2016), difficulties in managing conflicting demands of work and family (Chen, Brown, Bowers, & Chang, 2015;Shacklock & Brunetto, 2012;Yamaguchi, Inoue, Harada, & Oike, 2016) and insufficient opportunities for professional and career development (Tummers, Groeneveld, & Lankhaar, 2013).
While these studies provide valuable insights regarding nursing in general, how they relate to research nurses in particular remains unclear. To address these limitations, a qualitative exploratory study was undertaken.

| Aim
The aim of this paper is to provide a better understanding of the factors which affect recruitment to and retention of nurses in clinical research posts.

| Design
A qualitative exploratory design was adopted, drawing on qualitative elements of a larger mixed methods study which included other healthcare professionals (Boulton & Hopewell, 2017). The underpinning conceptual framework (Larson et al., 2013) sees recruitment and retention as dynamic social processes experienced by an individual in a wider social context. Key aspects of the social context affecting behaviour include motivations and satisfactions at the individual level; family responsibilities at the micro level; and policies and practices of employing organizations at the meso level.

| Sample
Sampling and recruitment strategies are described fully in a previous publication (Boulton & Hopewell, 2017). Recruitment was through five NHS Hospital Trusts (including one University Hospital), two NHS Community Trusts and an NIHR Primary Care Partnership (11 GP practices) spread across three counties.
For the first stage of the study, a purposive "census" sampling strategy was used (Bryman, 2001). To be invited to take part in the study, all of the following inclusion criteria had to be met: nonmedical researcher, delivers (rather than initiates or leads) research, has direct contact with study participants, funded through an NIHR infrastructure organization, employed by an NHS Trust or GP practice and employing organization located in one of three counties of south-east England. Exclusion criteria were Principal or Chief Investigator on a study or funded directly by a research grant.
A total of 280 individuals met the inclusion criteria and were sent an on-line questionnaire and participant information sheet. Questionnaires were used to reach a large, diverse sample and to provide more information power for qualitative analysis, including simple counts and proportions (Malterud, Siersma, & Guassora, 2016;Maxwell, 2010).
For the second stage, a convenience sampling strategy was used to recruit a subsample to focus groups. An invitation was included at the end of the questionnaire; those who expressed interest were sent an information sheet and invited to choose from a list of dates and places for the focus groups. Focus groups were chosen to facilitate discussion and debate around professional issues of common interest to research nurses (Green & Thorogood, 2009;Kitzinger, 1994).

| Data Collection
The questionnaire was developed by the first author on the basis of a literature review and an earlier unpublished study (Boulton, 2012 (one open-ended question) and About you (six fixed-choice questions). Questionnaires were completed anonymously; submission of a questionnaire was taken as consent to take part in the study.
As study participants were employed at NHS sites across three counties, it was important to hold several focus groups, each in a location convenient to their place of work, so that as many as possible could attend. Seven focus groups were held in five widely dispersed locations to ensure that research nurses from every employing organization could take part in one. They were held in Postgraduate Medical Education Centres or University meeting rooms and were facilitated by the first author, as numbers in each group were small. Participants were told that the purpose of the focus group was to allow them the opportunity to elaborate on the topics covered in the questionnaire and to raise any further topics they felt important.
Consent forms were signed before the start of each focus group.
A topic guide was prepared and used flexibly. It covered several areas including what their post involved and how they felt about it, whether they saw research nursing as a long-term career and why they might leave and what support they had received in developing their career. Questions were phased in terms of "Could you tell me about…" with additional prompts such as "What has been others' experience?" Data collection started in June 2013 and was completed in January 2014.

| Data analysis
As the study was largely exploratory in nature, an inductive thematic analysis approach was taken (Braun & Clarke, 2006).
Responses to open-ended questions on the questionnaire were saved in a spreadsheet, anonymized and read repeatedly by both authors. Initial codes were generated and revised following discussion by the authors. A final set of codes was agreed, the responses coded, the coded responses grouped into categories and categories grouped into higher order themes (Braun & Clarke, 2006).
Focus groups were audio-recorded, transcribed, anonymized and read repeatedly. The process of qualitative analysis as described above was followed, including generating and revising initial codes, coding transcripts, grouping coded text into categories and categories into higher order themes. Table 1 gives examples of the development of a theme from an open-ended question in the questionnaire and a theme from transcripts of focus groups.

| Credibility
The authors are an experienced qualitative researcher who had conducted a previous study of research nurses and an experienced TA B L E 1 Stages in analysis of qualitative data: Examples of codes, categories and themes What attracted you to apply for your first post in clinical research?

Examples of Codes Categories (Sub-themes) Themes
Interest  (Patton, 1999). Questionnaires were used to collect data from a large sample, and topics were explored further with a subsample in focus groups (Bryman, 2001;Green & Thorogood, 2009). Both authors independently coded the qualitative data and developed and refined categories and themes through iterative discussions (Braun & Clarke, 2006). All themes are supported by illustrative quotes.

| Transferability
Participants were recruited across a wide range organizations; response rates and sample sizes for both questionnaires and focus groups were good (Carlsen & Glenton, 2011;Morse, 2000).

| Confirmability
Key decisions in coding data and in refining categories and themes were noted, along with the rationale for them, to provide an audit trail (Bryman, 2001). All illustrative quotes are identified by a participant number (e.g., P138) or focus group number and transcript lines (e.g., FG2: 123-125).

| Dependability
Inclusion criteria and methods of sampling, data collection and analysis are described in detail (Shenton, 2004).

| Ethical Considerations
The

| RE SULTS
Completed questionnaires were received from 168 of the 280 individuals invited to take part in the wider study (Boulton & Hopewell, 2017), a response rate of 60%. Of these, 121 (72%) described themselves as nurses and constitute the questionnaire sample for this paper. Seven focus groups were conducted, ranging from 90 min to 2 hr and involving 26 individuals. Table 2 shows the characteristics of both the questionnaire sample and the subsample who participated in focus groups. The characteristics of focus group participants reflect those of the larger questionnaire sample, with the exception of professional background: 19 (73%) were nurses, one (4%) was a midwife and six (23%)

| Sample characteristics
were other healthcare professionals.

| Personal objectives focused
About a third of participants (39, 32%) indicated that they were primarily concerned with personal objectives which were tangential to research, but which could be achieved through a research post.

| Continue in a research nurse post
The majority of those who responded (67, 61%) indicated that they wanted to continue in a research post. Almost half of these indicated that they would like to continue at more or less the same level as they were, possibly with some change in focus or a greater variety of studies: More

| Progress towards nurse researcher
A much smaller proportion (11, 10%) indicated that they wanted to take steps consistent with becoming an independent researcher.
Some phrased this in terms of gaining a higher degree: Others in terms of carrying out their own research:

Developing my own research study… (P50)
Only two, however, explicitly stated that they wanted to "Become a nurse researcher (P76)."

| Leave research
A fifth of participants (21, 19%) would like to move on from a research nurse role to something else. For some, this meant a move into research management or governance: Working in an R&D department supporting researchers.

(P71)
For others this meant a move back into a clinical role, generally at a higher professional level:

Ideally I would be working closely with patients-CNS
[clinical nurse specialist] for xxx would be my ideal post.

(P81)
And for others it meant leaving nursing altogether, either to a job outside nursing or to retirement:

Owning my own teashop (P18)
Most likely I will be retired within the next 5 years (P146)

| Uncertain
The last group comprised those (11, 10%) who wanted to remain employed but were uncertain as to what they would like to be doing in

| Factors affecting intentions to remain in or leave research nursing
Views, concerns and experiences underpinning intentions to remain in or leave research nursing were discussed in the seven focus groups.
These are grouped in terms of factors operating at individual, family (micro) and organizational (meso) levels, all of which could change over time. However, not all participants were certain that they would continue to find research nursing satisfying over the longer term: Related to this was a concern that, if they remained in a research post "too long," their clinical skills would decline and so limit their ability to return to clinical practice in the future: For these latter participants, a quest for change and new challenges or their concern to keep job options open may result in their departure from research nursing at some point in the future.

| Family level
Focus group discussions also provided an insight into the way family responsibilities framed participants' decisions when considering a research nurse post. Family friendly hours and more bounded responsibilities were significant attractions for some participants and, while their children were young, helped to maintain their intention to remain:

| Organizational level
A range of policies and practices in their employing NHS Trusts were also raised as contributing to participants' intentions both to leave and to remain in research nursing. Short-term contracts were the most common cause of concern. Over 90% of participants had a contract of employment but for most this was for 12 months or less. The uncertainty and insecurity this caused raised questions about the viability of a research nurse post as a long-term option:  [FG5: 1523-29] A further concern for some was a sense that they and their work were not adequately valued by their NHS Trust, as reflected in, for ex- The third was working in an NHS Trust with a large and lively research environment which could provide help for research nurses to develop "academically" [FG3: 840] and the "right environment" in which to progress [FG3: 869]. This was particularly evident in relation to NHS Trusts associated with a research intensive university which provided longer term prospects and a wider range of opportunities for researchers:

I'm going to be in it for the duration. The fact I'm on a 2-
year rolling contract is maybe putting some people off… But if you're making a good job of things, then somebody else will find you the next project to roll on your contract.
[  This continuity of employment and at least the potential for career progression in research gave these nurses the confidence that underpinned their intention to remain in research over the longer term.

| D ISCUSS I ON
This paper has explored what attracts nurses to and retains them in a research nurse post using qualitative methods and a conceptual framework which sees recruitment and retention as dynamic social processes experienced by individuals in a wider social context. Table 3 shows the factors that were found to affect this process at the individual, family and organizational levels.
In terms of recruitment, it is striking that only a minority re- family friendly hours helped in managing conflicting demands of work and family (Powell & Greenhaus, 2010); and other features enabled the achievement of a range of personal goals, from working in a specific clinical area to gaining promotion in clinical practice (Duffield & Franks, 2002).
As in clinical nursing, the variety of attractions described may also reflect the range of career stages and personal circumstances of the study sample. For example, Pool, Poell, Berings, and ten Cate (2015) found that orientations to work changed over time in ways that resonate with those identified in this study, for example, learning more about research and using their skills more fully; seeking new challenges or a better work/life balance; and working in their chosen clinical field.
While studies in clinical nursing suggest that career planning is an important element in achieving career progression (e.g., Donner & Wheeler, 2001;Hall, Waddell, Donner, & Wheeler, 2004;Shermont, Krepcio, & Murphy, 2009), there was little evidence of career planning in participants' accounts. This is consistent with an Australian study of research nurses which observed that most participants had not planned a career in research (Rickard et al., 2011) and with published accounts of individual research nurses in the United Kingdom (Grove, 2015;Sprinks, 2015).
In terms of retention, the majority of participants reported that they would like to be still working in clinical research in 5-year time, reflecting the interest and satisfaction they currently derived from their roles (Rickard et al., 2007;Roberts et al., 2006;Rodrigo, 2013   The importance of a lively research environment as enabling research nurses to flourish and progress is another novel contribution of the current study. The continuity of employment offered by a large research active institution provides research nurses the opportunity to build experience and expertise. Their role in "the team" also affords a good basis for career progression.
Outside research intensive centres, however, jobs may feel less secure and opportunities to progress less evident. In these circumstances, a move to more administrative posts such as trial manager or into research governance may be their only options.

| Limitations and implications for further research
Research nurses who were employed by NHS Trusts and GP prac-

| Implications for policy and practice
Challenges to recruiting and retaining sufficient research nurses to meet increasing demand internationally will need to be addressed in several ways.

| To address recruitment issues
At undergraduate level, an interest in research could be encouraged by providing more opportunities for nursing students to experience research through placements with research nurses and through conducting primary research for their dissertation (Loke et al., 2014).
Support for new graduates in career planning (Byrne et al., 2014) and internships or mentoring schemes for those new to clinical research (Cleary et al., 2016;Jones-Berry 2016) are also important in attracting and retaining new nurses. To recruit experienced nurses, other nurses or health professionals could act as "research advocates," identifying talented nurses and encouraging them to take up a research post, while recognizing and supporting their diverse motivations for doing so.

| To address retention issues
To retain experienced research nurses, it will be essential for employing organizations to address policies and practices which create insecurity, appear to demean their work or dismiss their ambitions for professional development and career progression.
Fostering networks of research nurses within and across research sites, including university hospitals, may also provide greater opportunities to access a wider range of research posts and greater potential for progressing in their research careers. Nursing organizations such as the Royal College of Nursing also have a role to play in supporting "the professional recognition and status that researchers in other settings are afforded" (Rickard et al., 2011: 166) and in turning policy documents on a career in clinical research into reality, by supporting the progression of research nurses through to equivalent of Clinical Nurse Specialist, Advanced Practitioners and Nurse Consultants.

| CON CLUS IONS
The factors affecting recruitment to and retention of nurses in research posts derive from the motivations and family responsibilities of individual nurses, the policies and practices of employing organizations and how these change over time. Most nurses in a research nurse post are attracted by aspects of the role that meet their interests and concerns at the time. As these change and as the frustrations consequent to the policies and practices of their employing organization become more salient, many leave research for a clinical or administrative post that better meets their new interests and circumstances. This suggests that retention will continue to be a problem until key institutional disincentives are addressed. While specific aspects of the research nurse role may continue to attract new applicants and so replenish the workforce, recruitment of nurses with an interest in pursuing a long-term career in research may also continue to be problematic. A range of initiatives are needed in both educational and employing institutions to address these challenges.

ACK N OWLED G EM ENTS
We thank all those who took the time to complete the questionnaire and take part in a focus group and also all the R&D managers at the NHS Trusts, and the managers at the NIHR Clinical Research Networks, the NIHR Oxford BRC and BRU who identified potential participants and sent out the invitation emails, particularly Leigh Gerdes and Vasiliki Kiparoglou. Thanks also to Denise Best and Val Woods for their support and encouragement and to Joana Lopes for commenting on an earlier draft.

CO N FLI C T O F I NTE R E S T
No conflict of interest has been declared by the authors.

AUTH O R CO NTR I B UTI O N S
Mary Boulton conceived and designed the study, led on analysis and wrote the first draft of the paper. Sally Beer made a substantial contribution to the analysis and interpretation of the data. Both authors have given approval for publication and agreed to be accountable for all aspects of the work.