Social exclusion and its impact on health over the life course: A realist review protocol

Background Social exclusion is a process whereby certain individuals are born into or pushed to the margins of society and prevented from participating in social, cultural, economic, and political life. People who experience social exclusion are not afforded the same rights and privileges as other population groups. Socially excluded people often experience poorer outcomes in a variety of domains including health, education, employment, and housing than people with socio-economic privilege. People experiencing social exclusion frequently have higher and more complex health needs and poorer access to healthcare than the general population. The aim of this study is to better understand and explain how social exclusion occurs and how it impacts health over the life course. Methods A realist review will be undertaken. Data will be collected via a systematic search of databases of peer-reviewed literature and further iterative searches of peer-reviewed and other literatures as needed. The following data bases will be searched: MEDLINE, Embase, CINAHL, and ASSIA, using both indexed subject headings in each database and relevant key words. Grey literature will be searched via Google Scholar and relevant websites of organisations that work with populations affected by social exclusion. Conclusion A realist review will be conducted to explain the underlying societal mechanisms which produce social exclusion and related health outcomes in particular contexts affecting excluded population groups across the life course. The study has the potential to inform policy makers and service managers of how and why social exclusion occurs and potential key intervention points to prevent exclusion from happening.


Introduction
Social exclusion is defined by the United Nations (UN) as 'a state in which individuals are unable to participate fully in economic, social, political and cultural life, as well as the process leading to and sustaining such a state' 1 .Further, according to the UN, social exclusion is 'multidimensional' and 'not limited to material deprivation' even though 'poverty is an important dimension of exclusion, albeit only one dimension' 1 .
In fact, populations experiencing social exclusion often lack adequate access to healthcare, education, social capital, communal connection, housing, services, and more, in addition to often experiencing poverty 2,3 .Characteristics associated with social exclusion include substance dependence, homelessness, severe and enduring mental illness, incarceration, institutionalisation, and belonging to certain minority groups (e.g.Travellers, Aboriginal people) 1,4 .Often these identities intersect, where people having one experience frequently have other additional experiences of social exclusion 4 .
It is well-recognised that health is affected by poverty and other forms of deprivation.As the Marmot Review 5 has shown, there is a social gradient in health across society.Those with a high social position have better health outcomes than those with a lower one.Marmot links the cause directly to inequality: 'Inequalities in health arise because of inequalities in society -in the conditions in which people are born, grow, life, work, and age' 5 However, Marmot also points out that social exclusion is an extreme form of low socio economic status, as he says: 'social exclusion is deprivation upon stilts' 6 .A meta-analysis published in the Lancet in 2018 by Aldridge et al., bore this out when they found a ten-fold increase in all-cause standardised mortality ratios in people who experience social exclusion 7 .
Similarly, Irish research has demonstrated increased morbidity and mortality among people experiencing social exclusion.Kiernan et al. found dramatically increased levels of frailty among people who experience social exclusion in Dublin 8 .Meanwhile, Ivers and Barry found that the median age at death for men who accessed homelessness services in Dublin was 42 while for women it was even lower at 37 9 .
Disadvantage can start before birth with the foundations for physical, intellectual, and emotional child development laid at the earliest stage of life.Additionally, the unequal distribution of resources such as living conditions, education, wealth, networks, supportive family, parenting skills, and social capital across families in society accumulates across the life-course 5 .Conversely, identifying and acting on systematic differences in health can reduce health inequalities early and can have a profound effect on health outcomes later in life 5 .Social exclusion is not a static experience.As mentioned above, it is an intersectional phenomenon where experiences and identities intersect and can amplify each other.Additionally, the expressions and effects of social exclusion accumulate across the life course -a single individual may experience imprisonment, homelessness, and drug dependence across their life course with each potentially compounding the effect of each other deepening social exclusion and causing worse health and life outcomes 5,7 .Furthermore, social exclusion is associated with exposures and behaviours which can be damaging to health including excessive alcohol use, smoking of substances including heroin and crack cocaine, injecting drug use, rough sleeping, survival sex, and sex work 4,6,[10][11][12][13] .
A body of research has shown which population groups are most likely to experience social exclusion and how it contributes to causing poor health outcomes 4-10,14-27 .In response, the Inclusion Health approach to research, practice, and policy has emerged with the aim to understand the causes and consequences of social exclusion as well as to characterise and address health inequity arising from it 4 .However, there is less research explaining the underlying social mechanisms that cause social exclusion to take place at a societal level.The pathways into various and often intersecting expressions of social exclusion such as poverty, homelessness, substance dependency, sex work, etc, have been studied [28][29][30][31][32][33] but the process by which people who end up in difficult circumstances are then cast as 'excluded' within society has not been studied adequately.This study will build on and add to what is already known about stigma, internalised stigma, cultural and social barriers to inclusion, and more, by seeking to explain the underlying causal societal mechanisms that produce social exclusion and its many resulting health effects.

Protocol
This study will undertake a realist review to understand and explain how social exclusion occurs over the life course.Specifically we will address questions below: 1. How, why, in what circumstances, and for whom is social exclusion produced by underlying, often intersecting, societal mechanisms triggered in particular contexts over the life course resulting in inequitable health outcomes?Additionally, we will provide recommendations for health and social care policy makers about the contexts in which causal processes occur which create social exclusion.

Amendments from Version 1
Following reviewer feedback, the resulting changes have been made: updates to the language in the abstract for precision; changes to language in the body of the article including abbreviations/spelling out names of abbreviations (eg United Nations (UN), Non-Governmental Organisation (NGO)) and proper spelling of Aboriginals with a capital A; adding more information about the expert panel in the Methods section; and further details about how quality assessment of evidence included in the review will take place.

Methods
We have chosen to conduct this study as a realist review in the school of Pawson and Tilley 34-37 because it is a useful approach to understanding societal mechanisms which cannot be observed, such as stigma or social exclusion, but which nevertheless produce real life outcomes in specific contexts.To understand the causal mechanisms at play, realist approaches make use of theory and theorising to explain and synthesise data.
This realist review will be conducted using the six iterative steps they outline (Figure 1).

Initial programme theory
An initial programme theory (Figure 2) was built based on prior knowledge and robust discussion among the full research team and with input from an expert panel which includes patient and public involvement (PPI).It serves as the starting point for the study showing important factors that create social exclusion over the life course and the cumulative relationship between these.The expert panel includes people with lived experience of social exclusion, healthcare practitioners, and academics who all work in the area of inclusion health and bring various perspectives and expertise on the area, therefore being able to provide robust feedback and support during two formal meetings of the full group and via smaller consultations throughout the research process.

Searches
A search strategy has been developed by the research team informed by our initial programme theory (Figure 2), in consultation with the study's expert panel and a subject librarian.Two search clusters have been identified, one related to 'social exclusion' and the second related to the 'life-course'.We have developed search strings which have been piloted with changes made as needed based on the accuracy and appropriateness of various keywords.The search terms are as follows: ("Social exclu*" OR "social inclu*" OR "social marginal*" OR "social rejection" OR "Low socioeconomic status" OR "low SES" OR Depriv* OR "Severe multiple deprivation" OR "Low income" OR "early life advers*" OR "adverse childhood experiences" OR "ACEs" OR "early life stress" OR "Social class" OR "social factors" OR "Inequalities") AND ("life course" OR "life span" OR "life time") A systematic search of peer-reviewed databases will be conducted first of the following databases: MEDLINE, Embase, CINAHL, and ASSIA and using both indexed subject headings in each database and relevant key words.Additional iterative searches of peer-reviewed and other literatures will be conducted when and if needed.Such iterative searches will be performed if a gap in the literature resulting from the first search is apparent and further policy literature, grey literature and/or peer-reviewed literature sources are needed to explore aspects of the research questions.Grey literature will be searched via Google Scholar and relevant websites of organisations that work with populations affected by social exclusion.

Inclusion and exclusion criteria
Our search will seek to identify information, data, or studies which explain aspects of how social exclusion comes to be and its impact on health outcomes.In particular, for this type of study, we will be identifying sources or parts of sources which show how it happens, to whom, why, and to what extent.
Studies of any design written in English will be included.

Data extraction
Articles identified through our systematic searches, and later through iterative searches if needed, will be extracted into Covidence where automatic deduplication will take place.Of these, RS will screen 100 percent of the sources and a random 10 precent sample of the included documents will be independently screened by CO'D and SP for quality control.Potential disagreements will be solved by discussion between RS, CO'D and SP, and the full team if necessary.RS will also review 100 percent of the sources selected for full text review with CO'D and SP again checking 10 percent for consistency.
Papers selected for inclusion will be uploaded to NVivo where RS will code data in accordance with the realist approach as described by Papoutsi et al. 38 and Tierney et al. 39 .First, high-level conceptual labels will be assigned to identify what a piece of data is telling us about the research question.Then, through iterative rounds of coding, data will be assigned context, mechanism and outcome labels where possible keeping in mind that the same piece of data can act as different parts of different context-mechanism-outcome-configurations (CMOCs).Coding will at first be inductive and deductive with new codes created as needed and subsequent related pieces of data then assigned to codes that have already been created inductively, where appropriate.The second round of coding will be retroductive where a theory-driven realist approach will be used to identify patterns of causality in the data to aid in the creation of CMOCs 40 .
Data will then be extracted into a word document as contextmechanism-outcome configurations.CMOCs will be written with all supporting data listed after each configuration.CMOCs will then be iteratively refined, combined or rejected while continuously returning to the data to ensure that the synthesis is directly building on the data.Iterative searching will take place if and when needed to fill gaps and refine CMOCs as appropriate.Ultimately the analysis will result in building an overarching programme theory based on the data collected but presented at a higher level of abstraction which is transferrable within similar contexts.

Assessing quality of evidence
Following the realist approach, we will assess the quality of the data according to the relevance, rigour and trustworthiness of a given piece of data from an included source 41,42 .Quality assessment Additionally, risk of bias will be minimised by checking a random 10 percent sample of studies by a second reviewer as discussed above.And finally, quality will be ensured by adherence to the Realist And MEta-narrative Evidence Syntheses: Evolving Standards (RAMESES) publication standards for realist reviews 41 which provide guidelines for which relevant and necessary information that should be included in a realist review and allow end users and reviewers to assess the quality and rigour of a review.

Data synthesis
Typically, realist theory is expressed using the heuristic 'context + mechanism = outcome construction' also often written as 'C+M=O' or just 'CMO construction' or 'CMOC'.A CMOC is an explanatory device which shows how an outcome is produced when a particular hidden, latent power (mechanism) is triggered in a given context.
In a realist analysis, CMOCs are constructed close to the source material using specific pieces of data directly from the sources included in a review.Crucially data must be configured into explanatory statements describing the causal relationship of a given context, mechanism and outcome to account for the unseen causal action which the theory posits is happening in the data.They are then brought to a higher level of abstraction and are often combined into programme theory/ theories which explain patterns of causation removed from the specific data to explain causality in more general terms.

Study status
The systematic search has been completed and title and abstract screening is currently under way.

Dissemination
The results of our study will be disseminated primarily through academic networks via a journal article and presentations at academic conferences.We also have plans for a dissemination event for the project which this research is funded under (Health Research Board grant SDAP-2021-029).

Ethics
Ethical approval is not needed because this study is a synthesis of published literature and no data collection will take place.

Conclusion
In this research, we will seek to explain underlying societal mechanisms which cause some population groups and individuals to experience social exclusion and negative health outcomes as a result.We will conduct a robust realist review to uncover such societal mechanisms.This study will provide guidance for policy makers by supplying high-level explanations of how social exclusion occurs so they can identify societal causes and key points where interventions can be deployed to improve health outcomes.Ultimately, our work seeks to increase knowledge to prevent social exclusion from occurring and to mitigate its effects on the health of populations and individuals.

Ana Maria Arias-Uriona
Universidad Catolica Boliviana San Pablo, La Paz, La Paz Department, Bolivia Expanding the Theoretical Framework: As you mention, "Social exclusion is not a static experience.As mentioned above, it is an intersectional phenomenon" there is a need to explicitly integrate intersectionality as a critical theoretical framework within your study's methodology.This approach acknowledges the multifaceted nature of social identities and how various forms of social stratification (e.g., race, class, gender) intersect to influence individuals' experiences of social exclusion and health disparities.Highlighting this within your protocol can deepen the understanding of how intersecting structural mechanisms contribute to health outcomes. 1.

Refinement of Search Strategy:
Adjust your search strategy to include terms and concepts related to intersectionality and structural mechanisms.This can involve including specific keywords such as "intersectionality," "structural racism," "gender discrimination," "economic inequalities," and "social determinants of health" to capture literature that specifically addresses the interplay of these dimensions in the context of social exclusion and health.

2.
Inclusion Criteria Modification: Consider revising your inclusion criteria to seek out studies that employ an intersectional analysis or focus on the impacts of intersecting structural mechanisms on health.This ensures that the studies in your review align with the expanded theoretical framework and can provide insights into the complex dynamics at play.

3.
Analysis Approach: Adapt your data analysis approach to explicitly engage with intersectionality as a lens through which to interpret findings.This might involve identifying and examining patterns in how social identities and structural mechanisms interact to exacerbate or mitigate health disparities associated with social exclusion.This could also mean looking for evidence of how policies or interventions address or fail to address these intersecting factors. 4.

Susan Devine
College of Public Health, Medical and Veterinary Sciences, James Cook University, Townsville City, Queensland, Australia Thank you to the authors for their response to my suggested changes.I am satisfied that my comments have been addressed and have nothing further to add.I wish the authors well in their submission.

Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Social determinants of health.Health promotion.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
Reviewer Is the rationale for, and objectives of, the study clearly described?Yes

Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Social determinants of health; social aspects of aging; mental health; socioeconomic inequalities and health refer to differentiated forms of organization of social inequalities.In fact, the processes that could explain the link between social inequalities and health are potentially very different if class or exclusion is considered as the explanatory dimension of inequalities.While in the first case psychosocial approaches seem to be the most appropriate, in the case of exclusion it is possible that neo-materialist explanations may better describe its relationship with health.I suggest bearing in mid this differentiation when analyzing the results of the review.
I think inclusion criterion number 4 is questionable.This is a source with very different characteristics from the other sources.Is any website eligible?What criteria must the website meet to be considered an eligible source?Will they be quality criteria based on some formal element in the production of information?Please note that consideration I think that the inclusion of this source should be discussed in detail in the paper.
Is the rationale for, and objectives of, the study clearly described?Yes

Are sufficient details of the methods provided to allow replication by others? Yes
Are the datasets clearly presented in a useable and accessible format?

Not applicable
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Social determinants of health; social aspects of aging; mental health; socioeconomic inequalities and health I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Introduction
First sentence -after United Nations, add the UN abbreviation given that you go onto abbreviate it later in the paragraph.
In the second paragraph, please use a Capital A for Aboriginal.

Methods
Under the "Initial programme theory" heading -briefly provide some further detail on the expert panel.While the names of panel members are provided in the acknowledgements section, it would be good to have further insight into the level and type of expertise they bring.
Under "inclusion criteria" -write NGO out in full.Are there any year limitations on the search?
Figure 2 is good but I would add "transition" between the last two boxes as well, as this is a very big transition stage.I also wonder if it is worth adding a couple of other things to the last circle, including: loneliness, economic factors, discrimination and ageism and loss of meaningful engagement and social roles, as there is a lot of evidence that supports the importance of these issues.This is just something for the authors to consider.
Under "Assessing quality of evidence" -write RAMESES out in full and provide a brief explanation of what it is.What aspects of rigor will be considered?You could provide some further detail on how quality will be assessed and how the process of assessing quality will be documented.

Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.

Introduction
First sentence -after United Nations, add the UN abbreviation given that you go onto abbreviate it later in the paragraph.
In the second paragraph, please use a Capital A for Aboriginal.
Thank you for pointing out these mistakes.We've made the changes.

Methods
Under the "Initial programme theory" heading -briefly provide some further detail on the expert panel.While the names of panel members are provided in the acknowledgements section, it would be good to have further insight into the level and type of expertise they bring.
We've added more information about the expert panel in the text.
Under "inclusion criteria" -write NGO out in full.Are there any year limitations on the search?
Yes good point -we've spelled out non-governmental organisation.And we've added that there's no limitation on the year of publication.
Figure 2 is good but I would add "transition" between the last two boxes as well, as this is a very big transition stage.I also wonder if it is worth adding a couple of other things to the last circle, including: loneliness, economic factors, discrimination and ageism and loss of meaningful engagement and social roles, as there is a lot of evidence that supports the importance of these issues.This is just something for the authors to consider.
Thank you for this feedback!This is helpful and we will keep this in mind for the analysis but prefer not to change the initial programme theory because it has been developed with and agreed with our expert panel.Also, it is simply a rough starting point for the research and it is not meant to be exhaustive at this point.We assume that things we have not thought of, including the things mentioned here will come out in the literature synthesis.
Under "Assessing quality of evidence" -write RAMESES out in full and provide a brief explanation of what it is.What aspects of rigor will be considered?You could provide some further detail on how quality will be assessed and how the process of assessing quality will be documented.
Yes good suggestion.We have amended the manuscript.

Comments from Prof Esteban Sánchez-Moreno Author responses in italics
Thank you for the opportunity to review this interesting article ("Social exclusion and its impact on health over the life course: A realist review protocol"), which describes in detail the design of a theoretically informed systematic review protocol on the relationship between social exclusion and health.The article is remarkable for the detail with which both the rationale for the review and the procedures for its development are described.In addition, it addresses a research issue of special relevance, since the analysis of the mechanisms that operate in the existence of a social gradient in health in the case of social exclusion is understudied in the literature.This is a well written proposal and I only have a few comments and suggestions.

Thank you very much for your kind words and encouragement!
The introduction provides a detailed and rigorous analysis of the concept of social exclusion used, as well as some of the gaps in research on the relationship between social exclusion and health.It should be noted that the authors use a concept of social exclusion focused exclusively on situations of extreme social exclusion.This is a legitimate decision which, nevertheless, I believe should be clearly stated.Indeed, recent literature on social exclusion shows that this concept refers to a general process of organization of social inequalities.Therefore, its use is not limited to situations of particular hardship or socioeconomic deterioration, but constitutes a transversal axis around which social inequalities are organized.As a result, the population may be affected by more or less elements of this system of inequality, so that when we use the concept of social exclusion we refer to situations ranging from precarious integration to severe exclusion, passing through different degrees of moderate or mild exclusion.Specifically, this protocol focuses on situations of severe social exclusion, as can be seen from the introduction.This is a good point and we ourselves have had numerous conversations about terminology and finding the right word when it comes to describing the population group which we are talking about.You are correct that we are talking specifically about people who are on the margins of society when we say 'socially excluded' populations or people.We plan to continue to employ this term because across our various research projects and publications we have chosen to uniformly use the term 'social exclusion' and do not want to introduce 'extreme social exclusion' here.Additionally, the term 'extreme social exclusion' is not well established and so may mean different things to different people.
Additionally, the term 'social exclusion' is in our experience used in connection with extreme deprivation and we disagree that it should be used to characterise moderate or mild exclusion.We have cited this source in the article manuscript to be clearer about our conceptualisation of 'social exclusion'.
The protocol itself is clearly described and well documented.I am concerned that social class is included among the search terms.In the literature on social inequalities in general, and in much of the literature on socioeconomic gradient in health in particular, the concept of social class is clearly differentiated from the concept of social exclusion.Conceptually and theoretically, they refer to differentiated forms of organization of social inequalities.In fact, the processes that could explain the link between social inequalities and health are potentially very different if class or exclusion is considered as the explanatory dimension of inequalities.While in the first case psychosocial approaches seem to be the most appropriate, in the case of exclusion it is possible that neomaterialist explanations may better describe its relationship with health.I suggest bearing in mid this differentiation when analyzing the results of the review.
Thank you.We will keep your suggestion in mind as we move through the project.In selecting search terms, we sought to use keywords which would likely turn up the kinds of sources that would be useful for this analysis.Due to the association between health outcomes and both social class and social exclusion, both are included in the search.Once the analysis moves on, we will focus on explaining causal pathways in the data but at the search stage we were simply attempting to use search terms that are associated with the kinds of scholarship we want to synthesise in this piece of work.
I think inclusion criterion number 4 is questionable.This is a source with very different characteristics from the other sources.Is any website eligible?What criteria must the website meet to be considered an eligible source?Will they be quality criteria based on some formal element in the production of information?Please note that consideration I think that the inclusion of this source should be discussed in detail in the paper.
We will potentially use information from websites from reputable sources.Any website will not be eligible, rather as we say we will specifically look at NGO and professional organisation websites.These sources will be considered grey literature, which is commonly used in many kinds of research.For example, the websites of reputable homelessness organisations, the EU, the UN, the WHO etc might be useful.Any pages or reports taken from these sources, will be assessed for rigour, relevance and trustworthiness.Additionally, no findings will be based on one or a few sources so the data used in the analysis will effectively be triangulated to ensure congruence.

Figure 1 .
Figure 1.Stages of a realist review.

Inclusion criteria: 1 .
Research studies and clinical guidelines of any study design e.g.qualitative, quantitative, mixed methods 2. Review studies 3. Policy documents 4. Non-Governmental Organisation and professional organisation websites 5. Information published in English 6.Any year of publication Exclusion criteria: 1. Information not published in English
Report 13 February 2024 https://doi.org/10.21956/hrbopenres.15153.r37394© 2024 Sánchez-Moreno E. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Esteban Sánchez-Moreno Department of Sociology, Universidad Complutense de Madrid, Madrid, Community of Madrid, Spain Thank you for the opportunity to read the new version of the article.I do not have any additional comments.My warmest congratulations to the authors for this project, which is of great interest in the field of the study of health and its social determinants.

health and care for people who sleep rough: Going above and beyond
. London; 2020.Reference Source 17. FEANTSA: Health and homelessness.Homeless in Europe.Winter; 2019.Reference Source 18. St Mungo's: Health

and homelessness: Understanding the costs and role of primary care services for homeless people
. Jul, 2013.Reference Source 19.Simon Communities of Ireland: Health

and Homelessness: Health Snapshot Study of People Using Simon Services and Projects in Ireland. Dublin
; 2010.Reference Source 20.Keogh C, O'Brien KK, Hoban A, et al.: Health

and use of health services of people who are homeless and at risk of homelessness who receive free primary health care in Dublin
. BMC Health Serv Res.2015; 15(1): 58.PubMed Abstract | Publisher Full Text | Free Full Text

Is the rationale for, and objectives of, the study clearly described? Yes Is the study design appropriate for the research question? Partly Are sufficient details of the methods provided to allow replication by others? Partly Are the datasets clearly presented in a useable and accessible format? Yes
1. Collins P: Intersectionality's Definitional Dilemmas.Annual Review of Sociology.2015; 41 (1): 1-20 Competing Interests: No competing interests were disclosed.Reviewer Expertise: Social exclusión, intersectionality, migration I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.Reviewer Report 13 February 2024 https://doi.org/10.21956/hrbopenres.15153.r37395© 2024 Devine S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
strength of realist research is that information can be gleaned from many different places, sometimes ones that wouldn't be considered high quality research but which nevertheless can contain important nuggets of information.Pawson goes into more detail in (Pawson, 2006) Levitas, R. (2006).The concept and measurement of social exclusion.In Poverty and Social Exclusion in Britain (pp.123-160).The Policy Press.Pawson, R. (2006).Digging for Nuggets: How 'Bad' Research Can Yield 'Good' Evidence.International Journal of Social Research Methodology, 9(2), 127-142.https://doi.org/10.1080/13645570600595314