Conceptualising doctors’ and nurses’ experience of formal and informal solidarity: A Meta-Ethnography Protocol

Introduction & background

1.1.

Healthcare workforce retention is a global population health challenge. As argued by the World Health Organization (WHO), there is ‘no health without a workforce’, highlighting the need for research exploring frontline healthcare staff workplace experiences.¹ Negative healthcare professional (HCP) experiences of work (such as prolonged exposure to occupational stress, high workloads, unfavourable conditions, emotional trauma, a lack of preparedness in the face of newly emerging diseases, or horizontal violence) are a chronic health system issue undermining system stability, being consequential for healthcare access, patient care quality and continuity – further perpetuating the workforce crisis.^2–5 This crisis is multifaceted, and exists internationally, comprising of labour (recruitment and retention), mental health (high levels of anxiety, stress and burnout), education (inadequate institutional training and development to meet system demands) gender equality (pay gap, sexual harassment), and financial (austerity, unstable economies and a lack of funding) crises.^3,4 It is reflected in staff shortages, burnout and poor retention.^3,4,6–8 Increased workplace pressure faced by HCPs from the demands of ageing populations, and the coronavirus (COVID-19) pandemic especially, have highlighted the importance of ‘solidarity’ in shaping their working experience.^4,5

Recent studies have highlighted two key points of relevance to health research: the exacerbation of the unfavourable working conditions (extensive working hours and unreasonable shift patterns, resource shortages, pressure, stress, unsafe staffing) faced by HCPs,^4–9 as well as the role of solidarity in the management of these conditions.^4,5,9 Solidarities in work have been broadly described as a collective sense of representation, shared vision and purpose that develop within work environments. Beck and Brook¹⁰ define solidarity, broadly, as a phenomenon of ‘fellow-feeling’, which develops informally in the workplace as a collective sense of togetherness and mutually shared experiences and understanding. Solidarity (especially informal) refers to peer support or camaraderie – organic and mutually beneficial interpersonal support between colleagues with a shared sense of identity,^5,10–13 an example being illustrated by Korczynski’s communities of coping,¹⁴ where two or more healthcare practitioners support one another in the workplace. Specific definitions of solidarities which will guide our research are presented in Table 1.

Solidarities can also be formal – traditionally, as labour mobilisation or trade unions, commonly involving union membership of organised workers with similar material interests, ideas, or a sense of (in) justice.¹⁰ With respect to formal solidarity, appropriate management of labour relations via engagement of healthcare workers’ trade unions and professional representative bodies has a documented influence on addressing and safeguarding workers’ wellbeing.^3,15 Trade unions, e.g., the Irish Nurses and Midwives Union (INMO) and the Irish Medical Organisation (IMO) representing Irish nurses and doctors, respectively, act as key stakeholders of formal, collective representation. Having the power to negotiate reform recommendations for more sustainable working contracts, conditions and pay through industrial action, workers’ unions shape the industrial and labour relations landscape, and thus, the experience of work for doctors and nurses.^15,16

In the Irish context, Byrne et al.¹⁷ and Creese et al.^18,19 highlight how doctors’ experiences of work are shaped by informal peer support in a situation of a perceived ‘lack of voice’. In the US, Andrews et al.,²⁰ describe “co-worker support” as a mediating factor to turnover nurses, influencing job satisfaction. Similarly, a survey of GPs in England conducted by Jefferson et al.²¹ found that a lesser ‘sense of belonging’ (described as a factor of ‘job satisfaction’) negatively influenced retention. However, evidence coherency as to the influence of informal solidarity on the working experience of doctors and nurses remains limited. As with informal solidarity, there is a lack of coherency as to the perceived influence trade unions and other collective forms of representation have on the experience of work from the viewpoint of doctors and nurses.^3,15

Following the consultation of the ‘RETREAT’ criteria developed by Booth et al.,²² we identified qualitative evidence synthesis (QES) as the best methodology to explore this topic and address this evidence gap. Doctors and, in particular, nurses, were chosen due to historically high unionisation rates^23,24 and a tendency for these HCP unions to be single-profession (al) unions (as opposed to unions representing other allied HCPs, which tend to be mulit-professional). Additionally, the COVID-19 pandemic has highlighted the importance of solidarity for both professions.^4,5 As established, negative work experiences have the potential to exacerbate the workforce crisis by undermining HCP retention, and contributing to burnout and shortages.^4–9 Meanwhile, research has shown that solidarity shapes the working experience of HCPs, particularly that of doctors and nurses profoundly.^17–21 However, there is no universal consensus on whether doctors and nurses perceive both informal or formal solidarity (mainly trade union support) as effective in mediating negative work experiences. The specifics of the nature of the influence of solidarity for this group of professionals are also not fully conceptualised. It is worthwhile, therefore, to synthesise existing literature exploring these perceptions to better inform and support health workforce policies. The objective of this QES is to synthesise existing primary, qualitative literature defining the experience of solidarity (formal and informal) as perceived by, and from the viewpoint of doctors and nurses, to describe how exactly do both forms of solidarity influence their working experience.

Aim

1.2.

This QES aims to conceptualise how doctors and nurses experience both formal and informal solidarity in work. Findings of this review can inform strategies for improving health system sustainability by addressing the workforce crisis, leveraging the priorities of the World Health Organization’s (WHO) ‘Framework for action on the health and care workforce in the WHO European Region 2023–2030’, which is to safeguard the wellbeing of staff.¹ Additionally, a better understanding of working experiences of frontline health workers has the potential to influence workforce policy and improve retention. The research findings may also be beneficial in redefining or adding to the understanding of the definition or concept of solidarity, specifically in healthcare work for both doctors and nurses.

Research questions

1.3.

What are doctors’ and nurses’ experiences or perceptions of formal and informal solidarity, and how do both solidarity types influence the experience of work for these two professions.

Study design

2.1.

This QES will employ Noblit and Hare’s²⁹ 7‐step meta-ethnography, a systematic approach to the synthesis of findings of relevant qualitative studies, translating them into one another. Meta-ethnography was chosen to address our research question for its ability to help synthesise various qualitative, social and cultural interpretations (primary data) by way of producing a cohesive review. By synthesising primary interpretations, the authors of the QES themselves add to the body of knowledge by providing their own interpretations and by mapping how reviewed literature relates to one another. Literature, when synthesised using meta-ethnography, can be shown to align in agreement, but it can also be refutational, where different or conflicting sentiment can be evaluated against each other across different studies. This ensures that new insights are not simply one dimensional. As such, a meta-ethnography QES is an interpretation (by the research team) of interpretations (made by primary study authors) of interpretations (as expressed by the subjects of individual primary studies) (29). In the context of our study, meta-ethnography will address aforementioned research gaps by clearly defining the perceptions and working experience influences of formal and informal solidarity as understood by doctors and nurses – for example, whether these professions perceive solidarity as effective in maintaining positive working experiences, or conversely, whether it has no benefit or undermines working morale in some cases. The 7 steps (or “phases”, as described by Noblit and Hare) include:

The protocol for this QES is reported using the PRISMA-P protocol checklist and guidelines.³¹ The final synthesis and analysis will be reported in line with the ‘Improving reporting of meta-ethnography (eMERGe)’ guidelines.³⁰

Eligibility criteria

2.2.

Inclusion/exclusion criteria of study types, properties and study participants/population sampled are set out in the SPIDER framework²⁵ of Table 1 and addresses ‘Step 2: Deciding what is relevant’ of the 7-step Noblit and Hare meta-ethnography approach.²⁹

Studies for which there is no English language translation will be excluded. However, to avoid excluding studies with non-English titles which are otherwise supplemented with an original English translation (provided by the authors/publishing journal), no English language filter will be applied to the search. Studies not written in English (title or otherwise), without a full-text English translation from the authors of the study or the publishing journal, will be excluded manually - the rationale being the importance of context (for example, direct participant quotes) in qualitative studies, which may be impacted by poor or unoriginal manual translation by the research team. The search will not include non-English language terms. Neither the search string nor the strategy will be translated.

There will be no publication year restrictions applied to the search, the justification being that solidarity is a traditional concept in the social sciences and the likelihood of studies from the 70s and 80s in industrial relations being potentially relevant to our conceptualisation and research question. Although previously cited studies describe examples of recently emergent solidarity (for example, in the context and face of the COVID-19 pandemic), the relevance and impact of solidarity for both doctors and nurses is likely to be a concept dating back in time. Therefore, in order to broadly address the research question, to note any historical differences in sentiment towards or the manifestation of solidarity, and to minimise the risk of excluding older, but highly relevant literature, no publication year restriction filter will be applied.

Search strategy

2.3.

The search strategy was devised, using the PRESS Peer-Review process, to address ‘Step 2: Deciding what is relevant’ of the Noblit and Hare 7-step meta-ethnography approach.²⁹

An independent, expert health librarian affiliated with the 1 ^st Author’s host institution was consulted to develop the search strategy and search strings for each relevant database. An initial search was developed with the help of the host institution librarian ('searcher'). A second, independent, university librarian ('reviewer') affiliated with the 2 ^nd Author’s host institution produced a PRESS Peer-Review³²of the initial search, to finalise the search string, strategy, and increase rigour and transparency throughout the QES.

Upon meeting and discussing with both librarians, any amendments advised as part of the PRESS peer-review were clarified and implemented across the search strategy and search strings for all databases used. The finalised search strings and strategy used for the purpose of this review, as well as the PRESS peer-review form completed by both librarians, is available under the ‘Extended Data’ section. The PRESS peer-review proved to be a very beneficial exercise, both in formalising the search strategy, as well as increasing the rigour and transparency of study search methodology. Additionally, both librarians offered invaluable amendments and suggestions for a more precise database search by pointing to some minor omissions and errors.

Preliminary search terms were developed to reflect the research question, using both keywords and controlled vocabulary of MeSH terms. MeSH terms were used for term refinement. The final search string will be translated and applied to MEDLINE (Ovid; doctor focus), CINAHL (Cochrane; nurse focus), Scopus (Elsevier; a multidisciplinary database, with coverage of social sciences and humanities research), and Embase (Ovid/Elsevier). PROSPERO will be searched for similar, prospective reviews. Article authors may be contacted for clarifications.

The search terms have been designed to capture international spelling variants and titles, using MeSH terms, controlled database-specific vocabulary and appropriate terms and truncations (e.g. “physician*”).

Search results will be subject to initial title and abstract screening carried out independently by two reviewers (MD, and a second, appointed reviewer) and selected based on relevance and meeting inclusion criteria outlined above. Disagreements and conflicts will be resolved by consulting a third reviewer (JPB).

Grey literature search strategy. A two-step, grey literature search will be carried out via a targeted, domain specific Google search using the following keywords: “nurse”, “doctor”, “physician”, “midwife”, “frontline worker”, “union”, “experience”, “solidarity”, “peer support”, “employee morale”, “workplace support”, “strike”, “collective organisation”, “camaraderie” of:

Neither the search terms nor search string for the primary and grey literature search will be translated to non-English language phrasing.

The reference list of all included documents will be screened (by MD) for additional studies using forward and backward citation (citation chasing via ‘Citationchaser’), supplemented by hand-searching reference lists where necessary.³⁴

Screening/data management & selection process. Following the search, all identified citations will be collated and imported into Zotero. After deduplication, citations will be uploaded into Covidence. The title and abstracts will be screened independently by two reviewers (MD and a second, appointed reviewer) against the inclusion criteria outlined in Table 1.

Two reviewers (MD and a second, appointed reviewer) will review the first 100 titles/abstracts as part of a pilot test. Disagreements will be resolved by consulting a third reviewer (JPB).

Remaining articles will be subject to independent, full-text screening by two reviewers (MD and JPB) - disagreements will be resolved by consulting a third reviewer (SOR) (Steps 2 and 3 of the meta-ethnography approach).²⁹

The QES screening tool developed by Ames et al. will be implemented during full-text screening.²⁸ Studies containing large volumes of data addressing the research question specifically will be selected for inclusion. Following these inclusions, studies will be screened further by way of sampling (independently, by two reviewers - MD and JPB), against the QES tool on the basis of contextual thickness and conceptual richness. Contextual richness refers to the depth to which the study has been contextualised (background, population/sample characteristics, description of methodology used by the study). Meanwhile, conceptual richness refers to the level of which the study findings or data have been transformed (by way of analysis, using existing, appropriate theoretical frameworks).²⁸ Appropriate thickness and richness cut-off points will be agreed upon by the research team prior to sampling but following full-text inclusions based, at this stage, on data volume only. The tool and cut-off points will be piloted by the research team on a small number of said full-text inclusions. For increased reproducibility and transparency, exclusion decisions based on the screening tool will be noted and made available as part of the supplementary material of the published QES review.

Data extraction process and analysis

2.4.

For the purpose of collecting both descriptive characteristics of studies included in this review, as well as qualitative data reported in each, data extraction forms will be drafted, piloted, and finalised. Data will be extracted using said preconceived extraction forms and the qualitative data (such as interview or focus group quotes/statements, observational notes and author analysis) generated from each included study will be imported into the latest NVivo software available to the 1^st Author, for further synthesis and analysis. This will be done independently by MD, with review from JPB – disagreements, should they arise, will be resolved by consulting SOR. This will address Steps 3 and 4 of the 7-step meta-ethnography process.²⁹

Data items. Data items collected will reflect the requirements proposed in data extraction forms. Data extraction forms (Microsoft Excel format) will be piloted by the research team, and will include:

- Sample demographics and sample size.

- Country of origin of each study.

- Year of publication.

- Author name(s).

- Qualitative data generated and the authors’ interpretation of this data. This could include sections/paragraphs or transcripts, or direct study participant quotes interpretations, and analyses from grey literature. Extracted raw data will be grouped as first-order constructs (e.g. participant data such as direct quotes) and second-order constructs (e.g. primary study author interpretations of participant quotes by way of analytical themes). Both first-order and second-order constructs constitute primary data from individual articles, while third-order constructs will be the researcher’s novel, original analysis and synthesis of first and second-order constructs.

Data synthesis

2.5.

Using the latest version of NVivo available to the research team, original, full-text articles will be uploaded and stored as documents once data extraction (using piloted, Excel forms imported to NVivo) is complete. A translation table of first and second-order constructs will then be created. This will allow for constant comparison within and between first and second-order constructs, and for the research team to refer back to original data when developing third-order constructs.

Meta-ethnography as described in the 7-step process developed by Noblit and Hare will be used to analyse both first and second-order constructs, through line-by-line coding (Step 4 of the process), followed by descriptive coding of both construct groups (including any contradictory findings and addressing Step 5 of the process). This will be achieved with the help of node and memo generation features of NVivo – original articles and constructs will be indexed and linked to specific article sections or participant quotes. The research team will then generate novel analytical themes and interpretations (third-order constructs, using any existing conceptual or theoretical frameworks in this specific research field if applicable, for example Durkheim’s sociological foundations and mechanical and organic solidarity,^10,35 or the Social Exchange Theory).^24,36

Individual study results (i.e. raw data, authors’ interpretations) will be translated into one another to create/synthesise novel themes, adding to the body of knowledge. This has the effect of adding cohesiveness to pre-existing knowledge, summarising how results from one existing study can relate to another (Step 6).²⁹ Third-order constructs will be linked to quotes and other data from original articles in NVivo to help ensure novel interpretations remain grounded. To further ensure that third-order constructs are grounded in first and second-order constructs, the research team will discuss third-order interpretations. The author (MD) will maintain a data synthesis diary to challenge and reflect on their interpretations of first and second-order constructs..

The specific type of synthesis approach (i.e. reciprocal, refutational or line-of-argument synthesis) will be dependent on the alignments, similarities or differences across first and second order-constructs .^29,37 However, a gold-standard approach to meta-ethnography translation considers and applies all possible types of syntheses. The research team will pay additional attention to differences and disconfirming cases across first and second-order constructs (if applicable) in which case refutational translation will be considered for these cases, alongside reciprocal translation of aligning constructs. A line-of-argument synthesis will also be applied to translate different accounts addressing various aspects of solidarity to produce a whole greater than the sum of its parts, as described by Noblit and Hare when referring to this type of synthesis.²⁹

Outcomes & prioritisation. In line with the research question and qualitative nature of this review, outcome data will include doctors’ and nurses’ experiences and perceptions of formal and informal solidarity, as well as the impact and influence of solidarity on their working experience. Outcome measures of solidarity impact may be inductive, but could include staff expressions of engagement, intention to leave current post or health system, workplace satisfaction, satisfaction with the representativeness of trade unions or consequences of industrial action, and the perceived ability of formal and informal solidarity to enforce, promote or enable workplace wellbeing and the voicing of staff concerns.

There is also scope to examine any existing relationships or tensions between formal and informal solidarity, perceived in the context of work in medicine and nursing. However, this will largely depend on the data gathered as part of the QES.

Risk of bias in individual studies, meta bias (es)

2.6.

Studies will be independently assessed for risk of bias (in both study methodology and outcomes reporting) by MD and JPB (with input from SOR should discrepancies arise) using the’Critical Appraisal Skills Programme (CASP) Qualitative Studies Checklist’.³⁸ Studies will not be excluded or included based on quality, but as previously stated, based on the richness and thickness of qualitative data. Contextual thickness and conceptual richness will be determined using a QES assessment tool developed by Ames et al..²⁸

Confidence in cumulative evidence

2.7.

To assess the overall level of confidence to be placed in the overall findings of this review, and to increase reporting transparency, the (GRADE-CERQual) approach³⁹ will be applied to the cumulative evidence synthesis. Confidence in the cumulative evidence will be graded, assessed independently by two reviewers (MD and JPB) and will be presented and discussed by the research team using a summary table to address all four components of the guideline. The research team will determine a level of confidence to which each GRADE-CERQual component has been appropriately addressed. The ‘Methodological Limitations’ component will examine the rigour of study methodology, such as, for example, data collection and analysis. The ‘Coherence’ component will examine whether individual, primary study findings appropriately and adequately address the research question and findings of the QES, including any conflicting findings – or whether the QES as a whole is grounded in the findings of individual studies. ‘Adequacy of Data’ will be used to assess the conceptual richness (the level and depth of synthesis, using and applying appropriate frameworks to conceptualise findings), as well as the volume of data which supports the review findings. ‘Relevance’ will be used to determine the generalisability of results by evaluating whether the context of primary studies matches that of the study and research question, including, for example, the setting, population (doctors and nurses) or phenomenon of interest (solidarity).³⁹

Reflexivity

2.8.

To ensure an unbiased approach, especially during the data analysis and the presentation of study findings, the research team will reflect on individual preconceptions and past experiences which could interfere with the validity of this study. The 1^st Author (MD) has a background in microbiology and public health, with some experience in the administrative aspect of healthcare management and teaching, but little experience in sociology or industrial relations. Two of the remaining authors (JPB and SOR) have an extensive background in sociology, with experience in the field of healthcare management. JPB’s previous work has largely focused on the working conditions and experiences of frontline healthcare workers (especially doctors). All authors have carried out research in both an Irish and an international context. However, none of the authors are qualified doctors, nurses, or other allied healthcare professionals. The research team will reflect on any possible preconceptions at each stage of producing this review, which may affect the interpretation of data and research results.

Dissemination of information

2.9.

It is hoped that this review will be published in a relevant peer-reviewed journal. The published review and its findings will be incorporated as part of the 1 ^st Author’s PhD research and disseminated at relevant conferences, PhD research showcases (internal and external to the 1 ^st Author’s host institution) and presented to policymakers and stakeholders of interest as relevant (‘Step 7: Synthesis Expression’).²⁹

Limitations

4.1.

This review will examine the experiences of qualified doctors and nurses only – undergraduate student doctors and nurses, and other allied healthcare professionals (be they students or fully qualified professionals) will not be considered, as this would fall far beyond the scope of a PhD research project QES. Therefore, this review will not capture the phenomena of solidarity, formal or informal, as perceived by other types of HCPs, or future frontline health workers. Due to the current political, social, and industrial relations landscape, views of those who practice in countries where trade union membership is considered unprofessional, taboo, or perceived to have consequences for one’s employment security (e.g. USA currently – especially in terms of the possibility of newly emerging research being limited, Belarus, Hungary, Turkiye) may not be captured fully despite this study aiming to capture an international perspective. Research emerging from countries where union density tends to be high (for example, Nordic and Benelux countries), may be overrepresented in this review.

Additionally, articles not published in the English language are excluded, which, despite all efforts to include international literature, does not fully encapsulate the international perspective of doctors and nurses surrounding the concept of solidarity in the workplace. Although meta-ethnography is a powerful evidence synthesis tool for qualitative research, it often necessitates an exclusion of mixed-methods and quantitative studies such as research which utilises questionnaires or surveys with an explanatory or open-text element. Therefore, “thinner” qualitative data and quantitative data fall outside the scope of this review. However, this is partially justified with the primary aim of meta-ethnography being the synthesis of valuable rich and thick qualitative data only, which tends to be reported on and generated more transparently and rigorously in a fully qualitative study, as opposed to being an adjunct of a primarily quantitative study (which oftentimes yields ‘thin’ qualitative data).^28,29

Underlying data

No data associated with this article.

Extended data

PORSPERO Registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251150676 .

The following extended data is available for this protocol under the Open Science Framework and the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication): https://doi.org/10.17605/OSF.IO/DGMK4.⁴⁰

Protocol reporting guidelines

PRISMA-P 2015 Checklist³¹ for ‘Conceptualising doctors’ and nurses’ experience of formal and informal solidarity: A Meta-Ethnography Protocol’ is available as extended data as ‘PRISMA-P Checklist’.docx under the Open Science Framework and the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication): https://doi.org/10.17605/OSF.IO/DGMK4.⁴⁰