Evidence synthesis – Structural determinants of stigma across health and social conditions: a rapid review and conceptual framework to guide future research and intervention

Correspondence: Alexandra Blair, Dalla Lana School of Public Health, University of Toronto, 155 College Street, Toronto, ON  M5T 1P8; Email: alexandra.blair@utoronto.ca

Abstract

Introduction: Stigma has been identified as a key determinant of health and health inequities because of its effects on access to health-enabling resources and stress exposure. Though existing reports offer in-depth summaries of the mechanisms through which stigma influences health, a review of evidence on the upstream drivers of stigma across health and social conditions has been missing. The objective of this review is to summarize known structural determinants of stigma experienced across health and social conditions in developed country settings.

Methods: We conducted a rapid review of the literature. English- and French-language peer-reviewed and grey literature works published after 2008 were identified using MEDLINE, Embase, PsycINFO, Google and Google Scholar. Titles and abstracts were independently screened by two reviewers. Information from relevant publications was extracted, and a thematic analysis of identified determinants was conducted to identify broad domains of structural determinants. A narrative synthesis of study characteristics and identified determinants was conducted.

Results: Of 657 publications identified, 53 were included. Ten domains of structural determinants of stigma were identified: legal frameworks, welfare policies, economic policies, social and built environments, media and marketing, pedagogical factors, health care policies and practices, biomedical technology, diagnostic frameworks and public health interventions. Each domain is defined and summarized, and a conceptual framework for how the identified domains relate to the stigma process is proposed.

Conclusion: At least 10 domains of structural factors influence the occurrence of stigma across health and social conditions. These domains can be used to structure policy discussions centred on ways to reduce stigma at the population level.

Keywords: stigma, discrimination, structural determinants, social conditions, health conditions

Introduction

Stigma has been defined as a process enabled by social, economic and political power inequities, through which negative labels, beliefs and perceived differences between groups can culminate in discrimination and status loss.^{Footnote 1} As Link and Phelan wrote, “…stigma exists when elements of labelling, stereotyping, separation, status loss and discrimination occur together in a power situation that allows them.”^{Footnote 1,p.377} A key social determinant of health, stigma is a cause for concern in many substantive areas of public health practice.^{Footnote 2} Stigma-related discrimination and status loss influence health by restricting affected populations’ access to health-enabling resources such as housing, employment, social ties and health care, and by increasing exposure to stress.^{Footnote 2}

In Canada, the Public Health Agency of Canada’s Chief Public Health Officer (CPHO)’s 2019 report Addressing Stigma: Towards a More Inclusive Health System^{Footnote 3} has proposed a conceptual summary of the myriad pathways through which stigma can affect health and health inequities. This and previous frameworks^{Footnote 4Footnote 5} are helpful for understanding the theoretical underpinnings of the effects of stigma on health and for identifying potential areas for health and social policy intervention.

However, with their in-depth focus on the downstream effects of stigma on health, existing reports typically lack a thorough exploration of the upstream factors that drive stigma,^{Footnote 6} particularly those operating at a structural level. Existing literature on structural determinants of stigma tends to focus on stigma pertaining to specific stigmatized experiences, identities, behaviours or health conditions.^{Footnote 7Footnote 8} A general summary of determinants of stigma across affected populations is currently missing from the literature. This review seeks to fill these gaps in the extant literature by contributing a summary of known structural determinants of stigma across stigmatized populations.

Described as “contextual factors” in the World Health Organization’s Social Determinants of Health Model,^{Footnote 9} and structural “practices” in the CPHO’s 2019 report’s Stigma Pathways Model,^{Footnote 3} structural determinants can be defined as factors that operate outside the locus of control of individuals,^{Footnote 10} such as elements of physical, social, policy or legal environments.^{Footnote 11} For example, structural determinants can include various forms of legislation (or lack thereof) to protect individuals’ rights,^{Footnote 12} or wealth redistribution policies.^{Footnote 9Footnote 13} Structural factors are distinct from but tightly influence more proximal, individual-level determinants of health, such as individuals’ access to income, housing, food or safe working conditions.^{Footnote 9}

The scope of this review is restricted to examining the structural determinants of stigma for several reasons. First, according to public health research and theory, structural factors are considered to be those that create and perpetuate social and economic stratification within societies.^{Footnote 9} They are often identified as “root causes” of negative health and social outcomes and health inequities, and therefore merit particular attention from the perspective of population health and health equity promotion.^{Footnote 9}

Second, in the context of public health practice, structural factors tend to exert influence across multiple social contexts and populations.^{Footnote 14} Structural factors are therefore particularly relevant to consider when aiming to understand the determinants of stigma occurring across a multitude of health and social conditions—particularly when many forms of stigma intersect.^{Footnote 3}

Third, though it is difficult to achieve and often requires intersectoral collaboration,^{Footnote 9} structural determinants can theoretically be modified through changes in health and social policy.^{Footnote 15} When successful, structural-level interventions are often more impactful and far-reaching than more proximal (i.e. individual-level) interventions at reducing population-level health inequities.^{Footnote 9}

This review therefore was intended to provide a knowledge summary that, in the Canadian context, can complement the knowledge synthesis of the Canadian CPHO’s report on stigma’s effects on health^{Footnote 3} and, more broadly, can be used to structure policy discussions on ways to orient public health interventions to reduce stigma in Canada and abroad. The specific objective of this rapid review was to identify and summarize known structural determinants of stigma in Canada and in similar settings, such as those in other member countries of the Organisation for Economic Co-operation and Development (OECD).

Methods

We used a rapid review design.^{Footnote 16} Used most frequently within governmental policy contexts when time-related resources are limited,^{Footnote 17} a rapid review consists of an evidence review strategy that follows the structure of a systematic review process with abridged components to allow research questions to be addressed in a shorter time frame than is typically needed for a systematic review.^{Footnote 16} A defining feature of the rapid review design is its restricted evidence search component,^{Footnote 17} which involves a non-exhaustive search of available evidence pertaining to the research question. First, a search strategy to identify presynthesized evidence summaries (reviews, summary reports, conceptual frameworks) is applied. If identified synthesis documents are insufficient, because identified publications are not sufficiently recent or methodologically rigorous, a search for less synthesized evidence, including individual studies, is then conducted based on study relevance. Individual studies are collected until additional works fail to offer new information needed to address the research question or until other time or resource constraints prohibit future searches.^{Footnote 16}

For this review, a first search strategy was designed to prioritize the identification of presynthesized evidence such as conceptual summaries and literature reviews (Table 1, rows A and B).^{Footnote 18} A secondary search strategy was then applied to identify relevant individual studies, using general title and abstract search terms pertaining to the theme of “structural determinants of stigma” (Table 1, row C).^{Footnote 18} This secondary search was non-exhaustive. It was done to fill potential data gaps and identify domains missing from identified syntheses, and was ceased after the search strategy no longer yielded studies that reported new types of structural determinants or forms of stigma.

Table 1. Summary of components and search strings used to identify relevant references in the literature search on structural determinants of stigma

Components of search	Search string in titles or abstracts	French terms
A. Conceptual frameworks of determinants of stigma and mechanisms explaining stigma’s effect on health	(stigma OR stigmatization OR stigmatisation OR “stigmatized status”) AND (determinants OR cause OR “social factors” OR theory OR process OR pathway OR mechanism) AND (“conceptual framework” OR framework)	Stigmatisation, stigmatisé déterminant, cause, « facteur social » théorie, processus, mécanisme, « cadre conceptuel »
B. Systematic reviews of determinants of stigma	(Stigma OR stigmatization OR stigmatisation OR discrimination OR “stigmatized status”) AND (determinants OR cause OR “social factors” OR theory OR process OR pathway OR mechanism) AND (“systematic review” OR review)	« Revue systématique », revue, « recension des écrits » stigmatisation, stigmatisé déterminant, cause, « facteur social »
C. Individual studies of structural, upstream, system-level, population-level social determinants of stigma	(Stigma OR stigmatization OR stigmatisation OR “stigmatized status”) AND (determinants OR cause OR factor) AND (structural OR upstream OR “population-level” OR “system-level” OR structure OR structural OR infrastructural OR legal OR legislation OR policy OR “social environment” OR “built environment”)	Structurel, structure, amont, populationnel, système, systémique, infrastructure, légal, législation, « environnement social », « environnement bâti »
Note: The first search strategy was designed to prioritize the identification of presynthesized evidence such as conceptual summaries and literature reviews (rows A and B). A secondary search strategy was then applied to identify relevant individual studies, using general title and abstract search terms pertaining to the theme of “structural determinants of stigma” (row C). This secondary search was non-exhaustive. It was done to fill potential data gaps and identify domains missing from identified syntheses, and was ceased after the search strategy no longer yielded studies that reported new types of structural determinants or forms of stigma.

Eligibility criteria

Works included were those documenting conceptual frameworks, reviews and individual quantitative, qualitative or mixed-methods studies of structural determinants of stigma. We restricted our review to works in English or French (due to the authors’ languages of expertise), published since January 2008, in peer-reviewed or grey-literature sources and set in Canada or other OECD nations. We excluded works without a research design (e.g. commentaries, letters to editors, fact sheets), as well as those that were not available through the Health Canada Health Library network.

Search strategy

MEDLINE, Embase, PsycINFO, Google and Google Scholar databases were searched to ensure appropriate coverage of international health research and social science literature.^{Footnote 19} Table 1 presents a summary of the search string components. Search strings were also applied in French in Google and Google Scholar, of which the first three pages of results were reviewed. A snowball search approach^{Footnote 20} was also used to identify reviews or conceptual summaries that were mentioned in reviewed studies but were not captured through our search strategy.

Study identification, data extraction and quality appraisal

Two reviewers independently screened titles and abstracts of all identified works, and, based on eligibility criteria, selected works for full-text review through consensus. Four reviewers conducted full-text screening and extracted data on publications’ authors, year of publication, title, country setting, study population and identified structural-level determinants of stigma (Table 2). All data elements were recorded in the data extraction table to enable synthesis of study features and thematic analysis of structural determinants. Two reviewers evaluated the quality of included works using an adaptation of Dixon-Woods et al.’s framework for critical appraisal of works with varied design.^{Footnote 21} Seven questions requiring yes-or-no answers were applied to evaluate the quality of each publication (Table 3).

Table 2. Summary of reviewed publications (N = 53)

Author, year	Setting	Design	Population	Identified structural determinants of stigma	Structural determinant domains	Critical appraisal score
Clair et al. 2016Footnote 46	United States	Mixed methods sequential design	People living with HIV, African Americans, and people labelled as obese	Framework proposed: cultural resources and actors contributing to de-stigmatization Exclusionary laws and policies that do not protect individual rights. Pedagogical initiatives that shift causal attributions (i.e. remove blame) and refute stereotypes.	Legal frameworks Pedagogy	Moderate
Coreil et al. 2010Footnote 48	United States and Haiti	Mixed methods sequential design	Haitian immigrants (N = 95), TB patients (N = 126), health care providers (N = 126)	Exclusionary immigration policies. Conditions of poverty and malnutrition. Media content discriminating against migrants, reinforcing negative stereotypes. Lack of privacy in health care services. Availability of effective treatment for TB influencing risk perception. Infection and control policies mandating screening during immigration process.	Legal frameworks Social & built environment Media & marketing Health care practices Biomedical technology	Strong
Henderson et al. 2017Footnote 68	United States (Alabama)	Mixed methods sequential design	College students aged 18–25 years, N = 38 to 212 based on study phase	“War on Drugs”–related policies (policing, mass incarceration, severe sentencing). Antidrug education curriculums (e.g. Drug Abuse Resistance Education [DARE]) that frame substance use as human weakness.	Legal frameworks Pedagogy	Strong
MacLean 2018Footnote 47	Canada	Mixed methods sequential design	Individuals with STBBI experience (N = 20 individual interviews, N = 3 focus groups), provider survey (N = 410)	Framework proposed: conceptual framework of STBBI stigma (structural stigma: stigma confronted at policy/legal, institutional, community levels)	Legal frameworks	Weak
Arrey et al. 2017Footnote 55	Belgium	Qualitative	Adult HIV+ women who migrated from sub-Saharan Africa to Belgium (N = 44)	Limitations in health insurance coverage for migrants. Negative media messaging regarding migrants. Lack of training among health care professionals. Absence of policies to limit HIV status disclosure in health settings.	Welfare policy Media & marketing Pedagogy Health care policy	Strong
France et al. 2015Footnote 41	Ireland	Qualitative	Persons living with HIV (N = 17)	Framework proposed: conceptual framework of HIV self-stigma (social factors, self-factors, contextual factors) Availability of social spaces fosters sense of belonging, which can help tackle HIV self-stigma. Health professional interventions using inquiry-based stress reduction techniques may help address self-perceptions and beliefs relating to HIV.	Social & built environment Health care practices	Strong
Hansen et al. 2014Footnote 73	United States (New York City)	Qualitative	Individuals who received psychiatric diagnoses and qualified for disability benefits (N = 4)	Welfare reforms that make benefits contingent on the medicalization of disabilities.	Welfare policy	Strong
Paterson et al. 2013Footnote 65	Canada (Nova Scotia)	Qualitative	Professionals in ERs (N = 25) or NGOs (N = 25) serving those who use illicit substances or living with hepatitis C	Framework proposed: analytic framework of structural stigmatization (institutional policies, cultural, protocols) Lack of privacy in health care services due to physical settings or communication practices. Limitations in services availability due to wait times, lack of trained staff.	Health care practices	Strong
Woodgate et al. 2017Footnote 40	Canada (Winnipeg)	Qualitative	Indigenous people living with HIV since age 15–29 years (N = 21); service providers, elders (N = 14)	Framework proposed: social ecology of stigma and discrimination for Indigenous people living with HIV in Manitoba, Canada Child welfare systems as a source of discrimination in HIV status. Traditional educational initiatives and programming help to develop community support networks. Lack of safe health services on- and off-reserve; potential for breaks in patient confidentiality. Personal outreach initiatives (home visits) may help reduce stigma.	Welfare policy Pedagogy Health care practices	Strong
Jorm et al. 2008Footnote 79	Australia	Quantitative (cross-sectional)	Adult (> 18 years), general population surveyed on mental health–related stigma (N = 3998)	Attribution of disease to genetic factors may lead to greater perceptions of dangerousness. Receipt of a medical diagnosis suggesting a “real medical illness” may be associated with lower social distancing from individuals living with mental health issues such as schizophrenia.	Biomedical technology Diagnostic frameworks	Moderate
Min et al. 2017Footnote 39	South Korea	Quantitative (cross-sectional)	Persons living with mental illness who were users of community mental health centres (N = 532)	Perceived neighbourhood disorder (feelings about the prevalence of graffiti, noise, vandalism, abandoned buildings, dirty streets and poor maintenance of buildings) associated with mental health stigma (Devaluation and Discrimination Scale and Experiences of Rejection Scale)	Social & built environment	Strong
Pachankis et al. 2017Footnote 69	Europe	Quantitative (cross-sectional)	Immigrants to Europe who identify as MSM surveyed on HIV risk (N = 23 371)	Anti-gay structural stigma measured using the International Lesbian, Gay, Bisexual, Trans, and Intersex Association’s Rainbow Map Scale for the measurement of criminalization, protection, and recognition of same-sex relationships, based on states’ laws and policies (death penalty for same-sex sexual acts, lifetime incarceration of sexual minorities for same-sex sexual acts; laws against the promotion of homosexuality; recognition of same-sex marriage)—associated with higher HIV risk.	Legal frameworks	Strong
Stringer et al. 2016Footnote 78	United States (Southern states)	Quantitative (cross-sectional)	Health care workers surveyed on HIV-related stigma (N = 651)	Prevalence of policies to protect patient living with HIV against discrimination. Prevalence of policies that ensure consequences to providers that do not follow policies to protect patients living with HIV. Prevalence of availability of HIV PEP and PrEP across health facilities.	Health care policy Biomedical technology	Strong
Stuber et al. 2008Footnote 64	United States (New York City)	Quantitative (cross-sectional)	Current and former smokers in New York City (N = 816)	Self-reported difficulties renting an apartment or finding housing; having been refused a job for which they were qualified; or having been refused or charged more for health insurance because of smoking—associated with smoker-related stigma measure (absence of rights protections)	Welfare policy Economic policy Social & built environment	Strong
Angermeyer et al. 2014Footnote 75	Germany	Quantitative (longitudinal)	Adults (> 18 years) (Cycle 1, N = 5025; Cycle 2, N = 1232) surveyed on mental health stigma	Lack of funding for mental health care or research as a form of structural discrimination. Increases in prevalence of affective disorders believed to influence perceived risk and population support for mental health care funding.	Economic policy Social & built environment	Strong
Hatzenbuehler et al. 2015Footnote 70	United States	Quantitative (longitudinal)	Youth (9–14 years) from the Growing Up Today Study cohort surveyed on illicit substance use (N = 12 723)	State-level structural stigma index based on (1) policies preventing sexual orientation discrimination (e.g. same-sex marriage, employment non-discrimination); (2) density of same-sex partner households; (3) prevalence of Gay–Straight Alliances among public high schools; and (4) public opinion towards sexual minorities associated with sexual orientation disparities in illicit substance use.	Legal frameworks Social & built environment	Moderate
Arboleda-Florez et al. 2012Footnote 29	No restriction	Review (narrative)	Persons living with mental health issues	Structural approaches are identified to curb attributions of blame, dangerousness, unpredictability and thereby reduce stigma: Legislation to prohibit discrimination and offer accommodation in social domains such as housing, education and employment. Educational initiatives (e.g. Mental Health First Aid, contact-based education) can improve awareness of mental health symptoms.	Legal frameworks Pedagogy	Moderate
Aste 2016Footnote 53	United States	Review (narrative)	Adults with chronic pain	Insufficient knowledge and training among providers about effective pain management (pain viewed as imagined or psychological) can lead to discriminatory beliefs and practices.	Pedagogy	Strong
Bell et al. 2016Footnote 63	No restriction	Review (narrative)	Mothers of children born with fetal alcohol syndrome disorder (FASD)	Framework proposed: stigma “loads” experienced by those affected by FASD A child’s diagnosis of FASD can lead to mother’s experience of blame in clinical and social settings. Public health social marketing campaigns aiming to reduce maternal drinking can unintentionally increase blaming of mothers; threat of child protective services removing the child inhibits disclosure and treatment-seeking behaviour.	Diagnostic frameworks Public health interventions	Moderate
Benoit et al. 2018Footnote 60	No restriction	Review (narrative)	Sex workers	Derogatory labels (prostitute, hooker, whore) used in legislation, research and media associated with human trafficking, exploitation and victimization. Criminalization of sex work based on moralistic/paternalistic principles vs. decriminalization (e.g. New Zealand removing prostitution from its criminal code and regulating the industry within a public health and safety framework). Lack of police protection for sex workers; failure to respond to reports of violence; responding to reports of violence with criminal prosecution of the victim. Over-policing of sex workers (verbal harassment, invasive searches, excessive force, unwarranted arrests) even when not working. Delegitimization of sex work as an economic activity contributes to marginalization. Media narratives are morally driven (rather than empirical), reducing diverse experiences and perspectives to stereotypical portrayals, sex workers identified as blameworthy for the harms they experience. Denial of care, breach of confidentiality, lower quality of care in clinical settings following disclosure of occupation (lack of protective policies). Community-led program to increase HIV screening among sex workers.	Legal frameworks Economic policy Media & marketing Health care practices Public health interventions	Moderate
Bonsack et al. 2013Footnote 31	No restriction	Review (narrative)	Individuals living with mental illness	Legislation promoting self-efficacy and independence (anti-discrimination laws, advance directives, laws mandating reasonable accommodation at work) as protective against discrimination, especially when individuals with lived experience are integrated in the legislative decision-making process. Insufficient training of law enforcement and health personnel about mental health can contribute to discriminatory clinical practices. Lack of patient confidentiality in health care settings can contribute to negative experiences and discrimination.	Legal frameworks Pedagogy Health care practices	Weak
Brewis 2014Footnote 62	No restriction	Review (narrative)	Individuals labelled as obese	Unequal access to career and educational opportunities. Elements of the built environment, such as chairs that fit only smaller bodies, can represent sources of weight-based discrimination.	Economic policies Social & built environment	Moderate
Carroll 2017Footnote 61	United Kingdom	Review (narrative)	Single mothers	Various forms of legislation can contribute to the discrimination and status loss of single mothers, including policies regarding the legalization of divorce, and access to birth control and abortion. Evolving welfare policies can create perceptions of deserving or undeserving single mothers, and policies that promote return to work (e.g. the United Kingdom’s “Welfare to work” policies) can cause tension between single mothers’ breadwinner and caregiver roles. Social environments can contribute to the discrimination against and status loss of single mothers, including residence in more economically deprived areas due to lack of affordability of housing, and general community-level beliefs that mothers should provide care for their children instead of working. Negative portrayals of single mothers in media can perpetuate negative beliefs and stereotypes, which can lead to discrimination and status loss.	Legal frameworks Welfare policy Social environment Media & marketing	Strong
Chaudoir et al. 2013Footnote 66	No restriction	Review (narrative)	Individuals with visible or concealable stigmatized attributes	Framework proposed: the stigma mechanisms in health disparitiesframework Legislation that restricts access to health and dental care, education, movement, marriage and employment, and leads to differential criminalization (e.g. “War on Drugs”–related policies) represents a source of discrimination and status loss. Inequalities in high-quality health care access (insurance, a regular provider, transport time to hospital) represent a source of discrimination. Differential exposure to hazardous environmental conditions (pollution, violence, infectious disease, unsafe work conditions) by sex, socioeconomic position and across racialized groups also represents a source of discrimination.	Legal frameworks Welfare policy Social & built environment	Moderate
Collins et al. 2013Footnote 32	United States	Review (narrative)	Individuals living with mental illness	Framework proposed: conceptual model for reducing stigma associated with mental illness (individual, social, and policy/practice change) Media campaigns presenting information about illness causes, effective treatments, and experiences led by individuals with lived experience can help shift population beliefs and stereotypes; media content can represent a proxy for direct social contact with stigmatized group. Training promoting direct contact and providing factual information about mental illness (causes, symptoms, treatability, experiences), especially for key power groups (health professionals, employers, landlords, criminal justice, policy makers) may also help address negative beliefs and stereotypes.	Media & marketing Pedagogy	Weak
Livingston, JD (Commission de la santé mentale du Canada), 2013Footnote 37	Canada	Review (narrative)	Individuals living with mental health issues	Welfare policies that de-incentivize work, given that disability benefits are adjusted according to income can lead to internalized stigma. Further, parents living with mental health issues can often face more intensive scrutiny by social and child protective services. Absence of policies that protect against discrimination in employment or housing access, participation in civil society (e.g. holding public office), family panning (e.g. adoption disqualification), health (e.g. forced sterilization), or immigration can represent structural determinants of stigma. Geographic segregation of individuals living with mental health issues can contribute to status loss and inequities in resource access. Media content can shape opinions and interpretations of mental illness. Supportive policies within educational systems (e.g. specialized support services, accommodation policies) can help reduce stigma experiences. Insufficient health care funding leading to gaps in care, policies disqualifying individuals for health insurance, and violations of patient privacy can represent structural determinants of discrimination. Deficiencies in health services contribute to the use of law enforcement.	Welfare policy Economic policy Social & built environment Media & marketing Pedagogy Health care practices	Moderate
National Academies of Sciences, Engineering, Medicine (Committee on the Science of Changing Behavioral Health Social Norms) 2016Footnote 34	No restriction	Review (narrative)	Individuals with mental and substance use disorders	While forms of legislation may help reduce stigma, such as those accommodating students with disabilities or that ensure rights protections (e.g. Americans with Disabilities Act), other forms of legislation can be harmful, such as policies that treat substance use disorders as criminal issues rather than health concerns, or policies that restrict access to civil society and participation (e.g. to serve on juries). Segregated housing of individuals with mental illness, community rejection of mental health facilities represent structural determinants of stigma. Settings where there is inequality in access to high-quality health care services represent structural forms of discrimination and status loss. Negative media portrayals of mental health symptoms or treatment effectiveness can reinforce negative societal beliefs and stereotypes. Training for providers may help prevent misdiagnoses or improper treatment due to lack of knowledge about mental illness.	Legal frameworks Social & built environment Media & marketing Pedagogy Health care practices	Moderate
De Ruddere & Craig 2016Footnote 52	No restriction	Review (narrative)	Individuals living with chronic nonmalignant pain	Insufficient knowledge among health care providers about the nature of chronic nonmalignant pain (pain viewed as exaggerated or imagined, complaints dismissed) can influence their treatment of patients.	Pedagogy	Moderate
Désy et al. 2013Footnote 76	Canada	Review (narrative)	Stigma in the context of public health actions	Social support opportunities within communities can alleviate stigma. Mass media campaigns can influence stigma by shaping public perceptions.	Social & built environment Media	Weak
Earnshaw et al. 2015Footnote 42	United States	Review (narrative)	Individuals living with HIV	Framework proposed: Stigma and HIV Disparities Model Residential segregation contributes to discrimination and status loss. Structural interventions to improve neighbourhood-level access to resources relating to health and health promotion can help address structural discrimination.	Social & built environment Public health interventions	Moderate
Fernandes et al. 2011Footnote 51	No restriction	Review (narrative)	Individuals living with epilepsy	Forms of legislation have been identified as potentially protective against discrimination (e.g. Americans with Disabilities Act); alternatively, laws can be prohibitive against certain activities (e.g. driving or certain forms of employment). Pedagogical interventions, designed in collaboration with, and which promote contact with, individuals with epilepsy have been identified as potential structures to reduce fear and ignorance, and thereby stigma.	Legal frameworks Pedagogy	Moderate
Golub 2018Footnote 26	No restriction	Review (narrative)	Individuals who may be eligible for HIV pre-exposure prophylaxis (PrEP)	Clinical PrEP eligibility assessment interviews ask individuals about their partnership status and sexual behaviour, potentially stigmatizing condomless anal sex. Health care guidelines specifying that PrEP should be used by individuals at “very high risk of HIV infection” confers HIV-related stigma (and sexual stigma) on potential PrEP users—especially surrounding attributions of blame associated with perceived risky behaviours. Availability of PrEP impacts risk perception.	Health care practices Biomedical technology	Moderate
Groulx 2011Footnote 59	Canada	Review (narrative)	Individuals experiencing economic and social exclusion	Accessing welfare noted as associated with high levels of surveillance of economic and social activities as well as discrimination, which can impact self-esteem. Structural factors such as low minimum wage and the absence of employment policies to support a balance between work and family life can lead workers such as single mothers to turn towards welfare support, which is associated with both felt and perceived (self) stigma.	Welfare policy Economic policy	Strong
Hatzenbuehler 2016Footnote 8	No restriction	Review (narrative)	Persons living with mental health issues and sexual minorities	Legislation protecting rights to employment and marriage (or absence of these laws) influence likelihood of discrimination and stress.	Legal frameworks	Moderate
Hatzenbuehler 2014Footnote 57	No restriction	Review (narrative)	Lesbian, gay, bisexual (LGB) populations	Legislation protecting rights to employment and marriage (or absence of these laws) influence LBG discrimination. Community-level prejudice against LGB, measured in relation to social conservatism, can contribute to both perceived and felt stigma among LBG populations.	Legal frameworks Social & built environment	Moderate
Holder et al. 2019Footnote 30	United States	Review (narrative)	Individuals living with mental health issues	Insufficient health care provider training about mental illness (symptom recognition, effective treatment strategies, challenging stigmatizing attitudes).	Pedagogy	Moderate
Kerr et al. 2016Footnote 45	United States	Review (narrative)	Disparities in HIV risk across racialized communities in the context of drug policy	Framework proposed: the Drug War HIV/AIDS Inequities Model “War on Drugs”–related policies led to the discrimination and status loss of African Americans through over-policing, mass incarceration, sentencing (inequitable fees, confiscation of resources) and welfare cuts (offenders disqualified from various social protections such as public housing, Section 8 benefits, Electronic Benefit Transfer vouchers).	Legal frameworks Welfare policies	Moderate
Knapp et al. 2014Footnote 54	No restriction	Review (narrative)	Individuals living with cancer	Legislation protecting cancer patients’ rights regarding employment and housing (e.g. Americans with Disabilities Act) helps to counter potential discrimination. Media shaping perceptions about cancer (perpetuation of beliefs that certain people are to blame for their cancer, depiction of cancer as a “battle” rather than a journey) may result in more stigmatizing attitudes. Medical advances resulting in better understanding of the causes, treatments and outcomes of cancer can shape disease-related beliefs.	Legal frameworks Media & marketing Biomedical technology	Strong
Link & Hatzenbuehler 2016Footnote 72	United States	Review (narrative)	General population	Legislative factors including same-sex marriage bans, differential criminal sentencing, Jim Crow laws and immigration policies (mandatory verification of immigration documentation by police officers, restricted access to driver’s licenses or welfare benefits) contribute to discrimination and status loss. Absence of protective policies is a form of policy. Neighbourhood residential segregation and housing policies that reinforce segregation contribute to social beliefs and discrimination. Potentially protective pedagogical policies include antibullying policies in schools, whereas reliance on standardized testing for admissions (SAT) can produce gaps in admission according for racialized and vulnerable populations.	Legal frameworks Social & built environment Pedagogy	Moderate
Ministère de la Santé et des Services sociaux du Québec, 2016Footnote 38	Canada	Review (narrative)	Individuals with a history of mental illness accessing health care services	Insufficient training for providers can result in stigmatizing behaviour (e.g. overlooking physical concerns of those with mental health issues). Training opportunities for providers, especially those that promote contact with individuals with a history of mental illness, may help increase awareness and empathy, and shift clinical behaviours.	Pedagogy	Weak
Mirabito et al. 2016Footnote 67	No restriction	Review (narrative)	Stigma occurring in the marketplace	Framework proposed: The Stigma Turbine of individual, societal and marketplace-based determinants of stigma. Legislation that protects against marriage, employment, or housing discrimination on the basis of social identity can help reduce stigma by legitimizing stigmatized identities and ensuring equal resource access (e.g. same-sex marriage protections, Equal Pay Act, Fair Housing Act). Social environments such as workspaces or residential areas with greater social diversity allow for greater social interaction and contact, which can mitigate negative stereotypes and beliefs. Media-based marketing tactics and consumer segmentation strategies (e.g. advertisers engaging in objectification and fat-shaming to increase sales) can potentially perpetuate stigma whereas others (e.g. ads featuring unretouched models, or diverse family experiences) can attenuate negative social beliefs and shift cultural norms. Public health social marketing campaigns (e.g. regarding breastfeeding, weight control, tobacco use, alcohol abstinence) can perpetuate victim blaming. Similarly, social programs that require group separation (i.e. separate cafeteria line for students accessing school lunch programs) can also perpetuate discrimination and self-stigma.	Legal frameworks Social & built environment Media & marketing Public health interventions	Moderate
Morey 2018Footnote 56	United States	Review (narrative)	Immigrants and racialized communities in the United States	Policies limiting immigrants’ eligibility for health and social services, whether undocumented (Social Security, federal education benefits, Medicaid, Affordable Care Act) or documented (5-year waiting period to qualify for public benefits) and discriminatory raids, deportation or detention represent structural forms of discrimination. Xenophobic media-based rhetoric during election campaigns popularizing anti-immigrant attitudes, increasing bullying and violence against visible minorities.	Legal frameworks Welfare policy Media & marketing	Moderate
Mukolo et al. 2010Footnote 28	No restriction	Review (narrative)	Children experiencing mental illness	Framework proposed: framework of the relationships among child mental disorder stigma dimensions, contexts and targets Media portrayals of mental illness that reinforce negative stereotypes of dangerousness, criminality or unpredictability can promote discrimination. Media coverage of celebrities with mental illness may normalize mental health issues as issues pertaining to health rather than weakness. Lack of mental health professionals available to children, compared to professionals specializing in physical health, as indicative of the devaluing of mental health.	Media & marketing Health care practices	Moderate
Nairn et al. 2011Footnote 27	No restriction	Review (narrative)	Individuals living with mental health issues	Media portrayals (language, content) of mental health issues in relation to crime, violence and social incompetence, contribute to stigma by influencing society-level fear.	Media & marketing	Moderate
Pescosolido et al. 2008Footnote 33	No restriction	Review (narrative)	Individuals living with mental illness	Framework proposed: framework integrating normative influence on stigma Welfare eligibility shapes norms regarding entitlement to care, legitimizes conditions covered by health insurance policies. Negative media portrayals of individuals with mental illness (as dangerous, unstable, unpredictable), affecting judgments made in everyday life about individuals with mental illness. Innovations in treatment and advances in scientific knowledge (especially relating to biomedical or genetic causes, and availability of effective treatment) influence beliefs and practices.	Welfare policy Media Biomedical technology	Moderate
Phelan et al. 2014Footnote 77	No restriction	Review (narrative)	General population	Social segregation leads to infrequent interaction between stigmatized and nonstigmatized groups and social distance. Media portrayal of stigmatized persons can shape societal attitudes; portrayals can represent a substitute for social interaction when direct contact with stigmatized persons is rare/unlikely. Public health interventions that promote contact between stigmatized and other members of the public can potentially reduce negative beliefs and stereotypes.	Social & built environment Media & marketing Public health interventions	Strong
Schabert et al. 2013Footnote 50	No restriction	Review (narrative)	Individuals living with diabetes	Framework proposed: a framework for understanding diabetes-related stigma (structural mitigating strategies: social marketing, education, counselling, health promotion) Health promotion campaigns and media messaging based on modifying individual-level behaviours are identified as potential drivers of blame-induced stigma. Education initiatives that promote contact with individuals living with stigmatized conditions may help build empathy.	Media & marketing Pedagogy Public health interventions	Strong
White Hughto et al. 2015Footnote 58	United States	Review (narrative)	Individuals identifying as transgender	Framework proposed: modified social-ecological model of transgender stigma and stigma interventions (structural interventions: policies against discrimination, promoting access to care, or curricula on the health of trans people) Policies that fail to protect the rights of trans people in public accommodation. Inequalities in access to health insurance coverage for gender-affirming procedures. Insufficient training of health care professionals on trans persons’ health. Biomedicalization of gender nonconformity (DSM diagnosis).	Legal frameworks Welfare policy Pedagogy Diagnostic frameworks	Moderate
Williams 2018Footnote 35	United Kingdom	Review (narrative)	Individuals living with mental health issues	The United Kingdom’s Equality Act (2010) was designed to protect against disability-based discrimination in the workplace; however, limitations remain. Investment in workplace support programs for individuals with disabilities are posited as protective against stigma. Media representations that depict individuals experiencing mental health issues as dangerous or that correlate mental health issues with criminality can contribute to negative stereotypes. Use of a biopsychosocial model to understand mental health issues, rather than a purely biological model, can lead to more holistic approaches of mental health care.	Legal frameworks Economic policy Media & marketing Diagnostic frameworks	Moderate
Clement et al. 2013Footnote 36	No restriction	Review (systematic)	Individuals living with mental health issues	Mass media campaigns aiming to influence behaviour have the potential to decrease or reinforce mental health stigma (intentionally or not)	Media & marketing	Strong
Craig et al. 2017Footnote 49	Low TB–incidence countries	Review (systematic)	People living with TB in low-incidence countries (Western Europe, USA, Canada, Australia, New Zealand)	Negative media portrayals of migrants with TB can contribute to negative stereotypes and discrimination, especially if there is low knowledge among the general population regarding TB transmission and curability. Health care policies such as zero-tolerance policies regarding substance use and facilities that are not available to migrants, can lead to discrimination in relation to access to care. TB public health control policies (contact tracing, quarantine) can lead to discrimination and fears of potential deportation of migrants.	Media & marketing Health care policy Public health interventions	Strong
Darlington et al. 2017Footnote 43	United States	Review (systematic)	Women living with HIV	Absence of legal protections against job or housing loss due to HIV status also represents structural determinants of discrimination and status loss. Insufficient training regarding HIV transmission and curability among health care providers can lead to discriminatory beliefs and practices. Absence of policies to prevent denial or delay of health care due to HIV status can contribute to HIV-related stigma.	Legal frameworks Pedagogy Health care policy	Strong
Katz et al. 2013Footnote 44	United States	Review (systematic)	Individuals living with HIV	Health systems promoting social support and trust between patients and staff can alleviate negative beliefs and discriminatory practices. HIV health care–related costs can contribute to poverty and status loss.	Health care practices and policy	Strong
Abbreviations: AIDS, acquired immunodeficiency syndrome; DSM, Diagnostic and Statistical Manual of Mental Disorders; ERs, emergency rooms; FASD, fetal alcohol spectrum disorder; HIV, human immunodeficiency virus; LGB, lesbian, gay, bisexual; MSM, men who have sex with men; NGOs, nongovernmental organizations; PEP, post-exposure prophylaxis; PrEP, pre-exposure prophylaxis; SAT, scholastic assessment tests; STBBI, sexually transmitted and blood-borne infection; TB, tuberculosis.

Data analysis and synthesis

We performed a narrative synthesis of extracted data involving three analytic stages.^{Footnote 22} In the first stage, two authors conducted a thematic analysis of documented structural determinants of stigma.^{Footnote 23Footnote 24} Having become familiar with the data by generating initial summaries of structural determinants reported in included works during the data extraction process, these authors performed a thematic analysis, which involved identifying themes that linked structural determinants conceptually.^{Footnote 23Footnote 24} Themes were identified semantically (i.e. by interpreting factors that were explicitly mentioned in the texts rather than identifying underlying meanings) and inductively (i.e. without a predefined coding frame), through consensus.^{Footnote 25} We considered themes for which related structural factors were mentioned in at least two works, and that were internally coherent and conceptually distinct and definable.^{Footnote 24} Hereafter, these themes are described as “domains” of structural determinants.

At the second stage, we proposed a narrative synthesis of study types and characteristics and the identified domains and their relationship to the stigma process, as well as a visual conceptual framework of the domains and their relationship to the stigma process. The structuring of the elements in the conceptual framework was based both on the findings of included works and the structure of existing conceptual frameworks of the determinants of stigma.

The third and final stage consisted of summarizing the methodological quality of the included works. In order to provide a quantitative synopsis of quality appraisal results for this review, we considered studies with less than four “Yes” responses (< 60%) as weak, those receiving six or more “Yes” responses (> 85%) as strong, and the remainder as moderate. As these thresholds have not been validated, full annotated scoring results were also provided to complement quantitative summaries.

Results

The results of the narrative synthesis are presented here, beginning with a descriptive summary of study characteristics, followed by a summary of identified structural determinants.

Descriptive results

Study selection

Overall, 657 works were identified through the search strategy. Figure 1 provides a summary of works identified. Four works—all grey literature sources—were written in French. After applying inclusion and exclusion criteria, 53 works were retained (Table 2). Most rejected works were those that did not document determinants of stigma that operate at a structural level. The retained works consisted of literature reviews (n = 37; 69%), and individual mixed-method (qualitative and quantitative; n = 4; 8%), qualitative (n = 5; 9%) and quantitative (n = 7; 13%) studies. Most reviews were not limited to a specific country setting, while individual studies were predominantly based in specific geographic regions including the United States (n = 14), Canada (n = 8) and Europe (n = 5) (Table 2).

Text description: Figure 1

Figure 1. Flow diagram of data identification, selection and extraction

Level (Top to Bottom)	Position (Left to Right)	Arrow to	Text
1 (Identification)	1	2-1	Conceptual frameworks of determinants of stigma Records identified through database searching (n = 173)
	2	2-2	Systematic reviews of determinants of stigma Records identified through database searching (n = 195)
	3	2-3	Individual studies of determinants of stigma Records identified through database searching (n = 262)
	4	2-4	Other sources Grey-literature records identified through database searching & studies identified through hand searching reference lists of identified works (n = 27)
2 (Screening)	1	3-3	Records after duplicates removed (n = 172)
	2	3-3	Records after duplicates removed (n = 183)
	3	3-3	Records after duplicates removed (n = 252)
	4	3-3	Records after duplicates removed (n = 26)
3 (Screening)	3	4-3, 3-5	Title and abstract screened (n = 633)
	5	N/A	Records excluded (n = 425)
4 (Eligibility)	3	5-3	Full-text articles assessed for eligibility (n = 208)
	5	N/A	Full-text articles excluded (n = 155) No determinants described (n = 76) Not relevant to the Canadian context (n = 17) File not available (n = 52) Type or design not eligible (n = 10)
5 (Included)	3	N/A	Studies included in narrative synthesis (n = 53)

Quality appraisal

Most publications were moderate or strong in quality. The weakest element of the publications was a lack of detail on the design processes by which findings and interpretations were produced (Table 3). This element tended to be absent from works from the social sciences, which made up a large proportion of the included literature. This is a limitation of extant reviews of structural determinants of stigma. Figure 2A provides a summary of the mean score on each of the quality appraisal questions by study type, and Figure 2B provides an overview of the quality of works supporting each identified structural domain, described later on.

Figure 2. Summary of the quality appraisal assessment of identified works (N = 53), across study types and domains of structural determinants of stigma

Text description: Figure 2a

Figure 2a. Proportion of works by study type meeting each quality appraisal indicator

Quality appraisal indicators	Proportion (%) of works meeting each indicator
	Mixed methods (n=4)	Qualitative (n=5)	Quantitative (n=7)	Review (n=37)
1. Objectives are clearly stated	100	100	86	97
2. Research design is clearly specified	50	100	100	32
3. Research design is appropriate for the research objectives	100	100	71	95
4. Includes clear account of the process by which findings were produced	50	100	100	24
5. Enough data are provided to support interpretations and conclusions	75	80	71	86
6. Method of analysis is appropriate	75	100	86	97
7. Method of analysis is adequately explicated	50	100	100	22

Text description: Figure 2b

Figure 2b. Number and proportion of works supporting each identified domain of structual determinants by overall appraisal score

Ten domains of structural determinants	Number (n) and proportion (%) of works by overall quality score
	Weak	Moderate	Strong
Legal frameworks	2 (9%)	15 (65%)	6 (26%)
Welfare policies	0 (0%)	6 (50%)	6 (50%)
Economic policies	0 (0%)	4 (57%)	3 (43%)
Social and built environments	1 (6%)	9 (53%)	7 (41%)
Media and marketing	2 (11%)	9 (47%)	8 (42%)
Pedagogical factors	3 (17%)	9 (50%)	6 (33%)
Health care policies and practices	1 (55%)	8 (40%)	11 (55%)
Biomedical technology	0 (0%)	3 (50%)	3 (50%)
Diagnostic frameworks	0 (0%)	4 (100%)	0 (0%)
Public health intervention	0 (0%)	4 (57%)	3 (43%)

Study characteristics

The identified works overwhelmingly cited Link and Phelan’s^{Footnote 1} conceptualization of stigma as a process driven by social, economic and political power inequities, through which attitudes, negative stereotypes and a sense of separation between groups can lead to discrimination and status loss. The reviewed literature explored many stigmatized experiences, identities, behaviours and health conditions (Table 2). Comprehensively, these were: individuals with mental health and substance use disorders;^{Footnote 26Footnote 27Footnote 28Footnote 29Footnote 30Footnote 31Footnote 32Footnote 33Footnote 34Footnote 35Footnote 36Footnote 37Footnote 38Footnote 39} individuals living with human immunodeficiency virus (HIV)^{Footnote 26Footnote 40Footnote 41Footnote 42Footnote 43Footnote 44Footnote 45Footnote 46} or other sexually transmitted or blood-borne infections (STBBI),^{Footnote 47} tuberculosis (TB),^{Footnote 48Footnote 49} diabetes,^{Footnote 50} epilepsy,^{Footnote 51} chronic pain^{Footnote 52Footnote 53} or cancer—particularly types whose etiology may be attributable to patients’ behaviours;^{Footnote 54} vulnerable subpopulations such as migrants and racialized communities;^{Footnote 55Footnote 56} lesbian, gay, bisexual, transgender, queer and other (LGBTQ+) populations;^{Footnote 57Footnote 58} individuals experiencing poverty;^{Footnote 59} sex workers;^{Footnote 60} single mothers;^{Footnote 61} individuals labelled as obese or fat;^{Footnote 46Footnote 62} biological mothers of children diagnosed with fetal alcohol spectrum disorder;^{Footnote 63} and individuals who smoke.^{Footnote 64} Some authors highlighted that, although their research may have focussed on stigma pertaining to a specific condition or identity, individuals can experience multiple sources of stigma due to the intersection of multiple complex identities and life experiences.^{Footnote 47Footnote 60} These authors acknowledged that restricted study scope could represent a potential limitation to their studies.

Lastly, among the works reviewed, 15 proposed conceptual frameworks that graphically represented at least one upstream determinant of stigma (Table 2).^{Footnote 28Footnote 32Footnote 33Footnote 40Footnote 41Footnote 42Footnote 45Footnote 46Footnote 47Footnote 50Footnote 58Footnote 63Footnote 65Footnote 66Footnote 67} However, none were intended to provide a comprehensive summary of known structural determinants of stigma across populations. Identified frameworks were heterogeneous in form and content. However, most frameworks included an acknowledgement that stigma could be enacted at many levels, such as the individual (internalized), interpersonal, community, institutional and societal levels. Many frameworks also acknowledged that stigma is influenced by historical social inequities, such as those relating to systemic racism.^{Footnote 46}

Structural determinants of stigma

Thematic analysis of the reviewed literature yielded 10 overall domains of structural determinants of stigma (Table 2): (1) legal frameworks; (2) welfare policies; (3) economic policies; (4) social and built environments; (5) media and marketing; (6) pedagogical factors; (7) health care practices and policies; (8) biomedical technology; (9) diagnostic frameworks; and (10) public health interventions. A narrative synthesis of these domains and their relationship with stigma follows.

Legal frameworks

The domain of “legal frameworks” refers to factors pertaining to enacted or proposed legislation, including broad bills of rights, as well as downstream elements of criminal justice systems such as factors pertaining to policing, courts and corrections. Twenty-four works referenced terms that fell under this domain.^{Footnote 29Footnote 31Footnote 32Footnote 34Footnote 35Footnote 43Footnote 45Footnote 46Footnote 47Footnote 48Footnote 51Footnote 54Footnote 56Footnote 57Footnote 58Footnote 60Footnote 61Footnote 66Footnote 67Footnote 68Footnote 69Footnote 70Footnote 71Footnote 72} Examples of terms describing structural determinants under this domain include (but are not limited to) “laws,”^{Footnote 32} “legislative action,”^{Footnote 43} “legal protections,”^{Footnote 46} “Acts” (e.g. the United Kingdom’s Mental Health Act or Equality Act),^{Footnote 35} “policing” and “sentencing.”^{Footnote 45} Terms related to legal frameworks were mentioned in six existing frameworks.^{Footnote 32Footnote 45Footnote 46Footnote 47Footnote 58Footnote 67}

Overall, legal frameworks were identified as potential levers to prevent the discrimination and status loss components of the stigma process. Legislation that enshrines individual rights in relation to employment,^{Footnote 57Footnote 60} housing,^{Footnote 43} marriage,^{Footnote 66Footnote 69} or immigration^{Footnote 56}—to name some of the areas referenced in reviewed works—can prevent inequitable access to health and social resources.^{Footnote 67} On the other hand, factors relating to legal frameworks can also enable discrimination. This can occur when existing legislation fails to protect the rights of certain populations (people living with HIV are one prominent example^{Footnote 43}), when legislation restricts rights for certain groups (e.g. barring individuals with mental illnesses from serving on juries,^{Footnote 34} or prohibiting individuals with epilepsy from driving^{Footnote 51}), or when components of the criminal justice system such as policing or sentencing affect certain populations more than others.^{Footnote 46} The laws introduced during North America’s “War on Drugs,” referenced in several studies, are one of the most illustrative examples of the latter. These laws were described as influencing the stigma process by perpetuating negative stereotypes of individuals who committed drug-related crimes (e.g. using or selling illicit substances)^{Footnote 34} and enabling status loss of Indigenous, Black and racialized communities through disproportionate policing and incarceration.^{Footnote 45Footnote 66}

Welfare policies

The domain of “welfare policies” refers to factors relating to the presence of, eligibility for and relative coverage or “generosity” of government-based structures, services and benefit programs that offer social, health or economic support for those in need. A non-exhaustive list of terms that led to the categorization of this domain included “welfare state,” “insurance,”^{Footnote 33} “child welfare system,”^{Footnote 40} “benefits”^{Footnote 73} and “social security.”^{Footnote 56} Twelve works used terms related to welfare policy,^{Footnote 33Footnote 37Footnote 40Footnote 45Footnote 55Footnote 56Footnote 58Footnote 59Footnote 61Footnote 64Footnote 66Footnote 73} of which two also referred to these terms in proposed conceptual frameworks.^{Footnote 33Footnote 40}

Overall, the provision and coverage of social supports (or lack thereof) was noted across the included works as contributing to the stigma process via two principle mechanisms: by shaping societal beliefs about service or benefit recipients, and by influencing vulnerable populations’ access to protective and life-sustaining resources. Two examples of these welfare-related mechanisms include policies that disqualify individuals from welfare benefits or services based on certain statuses, such as those limiting immigrants’ eligibility for social services based on documentation status or length of time since immigration,^{Footnote 56} and “War on Drugs” policies that disqualify offenders from social protections such as access to public housing.^{Footnote 45} Both types of policies were identified as reinforcing negative societal perceptions of those who do not “merit” societal support, therefore legitimizing exclusion and negative stereotypes.^{Footnote 33} Similar processes can occur if welfare coverage is restricted for certain conditions, such as when health insurance coverage is limited for mental health services^{Footnote 37} or gender-affirming procedures,^{Footnote 58} to name two examples. Policies that disqualify certain populations from access to social supports contribute to the stigma process by creating systematic gaps in care and status loss for affected populations.

Economic policies

The domain of “economic policies” refers to factors pertaining to governmental influence on features of economic landscapes, including policies relating to labour market wage, income redistribution policies or budgetary funding allocation across sectors. Though minimum wage limits are set through legislation, they are grouped under this domain because of their effects on economic conditions such as hiring practices and labour market participation—factors connected to the economic landscape of political jurisdictions.^{Footnote 74} Eight works referred to terms pertaining to this domain,^{Footnote 33Footnote 35Footnote 37Footnote 59Footnote 60Footnote 62Footnote 64Footnote 75} including one which integrated the concept of economic policies in its proposed conceptual framework.^{Footnote 33} Examples of terms relating to this domain included “funding,”^{Footnote 75} “tax,”^{Footnote 37} “investment,”^{Footnote 37} “economic development”^{Footnote 33} and “minimum wage.”^{Footnote 59}

Overall, governmental influence on economic environments was described as influencing the stigma process by determining how equitably (or inequitably) economic resources are distributed within a population—thereby both influencing socioeconomic positioning of groups and sending the implicit message to disenfranchised groups that their disadvantaged state is not worth addressing through public investment. One example of how economic policies can influence status loss was that limited budgetary resource allocation to health services and workplace support programs for those experiencing mental health issues can lead to gaps in care and social and economic exclusion for those affected.^{Footnote 35Footnote 37Footnote 75}

Social and built environments

The domain of “social and built environments” refers to the characteristics of communities and places, at an aggregate level, in which individuals live, work or play—including population prevalence of certain health, social or economic conditions or elements of physical environments. Although they are likely influenced by elements that fall under preceding domains, such as economic or welfare policies, we consider socioecological environmental characteristics as a distinct domain. Seventeen works included terms pertaining to this domain,^{Footnote 34Footnote 37Footnote 39Footnote 41Footnote 42Footnote 48Footnote 53Footnote 57Footnote 61Footnote 62Footnote 64Footnote 66Footnote 67Footnote 70Footnote 72Footnote 75Footnote 76Footnote 77} including three conceptual frameworks.^{Footnote 41Footnote 42Footnote 66} A non-exhaustive list of terms relating to this domain included “environmental hazards,”^{Footnote 66} “residential segregation,”^{Footnote 42} “neighborhood disorder,”^{Footnote 39} “prevalence” of a health or social condition (e.g. depression)^{Footnote 75} and availability of social meeting “spaces.”^{Footnote 41}

Overall, studies describe how characteristics at the level of the local area and the community can influence the stigma process by contributing to both real and perceived social separation between groups and by generating social stratification in resource access. One example of these mechanisms is the way racialized and lower-income communities are segregated across residential neighbourhoods, which can reduce contact between stigmatized and nonstigmatized groups and reinforce perceptions of social differentiation.^{Footnote 42Footnote 77} These exclusionary structures can contribute to economic deprivation,^{Footnote 48} differential exposure to hazardous environmental conditions such as unsafe work conditions, pollution or infectious disease,^{Footnote 66} and inequalities in access to health care facilities^{Footnote 34} or places of education and employment for excluded groups.^{Footnote 37} In contrast, social environments such as workplaces or residential areas with greater social diversity can foster more interactions between population subgroups, shift societal perceptions and beliefs and reduce inequities in resource access.^{Footnote 67Footnote 75}

Media and marketing

The domain of “media and marketing” refers to factors pertaining to the content development and regulation of communications strategies of various forms, such as news media, broadcasting or advertising, designed for the purposes of entertainment and sales, or to promote changes in individuals’ behaviours. Nineteen works used terms pertaining to this domain.^{Footnote 27Footnote 28Footnote 32Footnote 33Footnote 34Footnote 35Footnote 36Footnote 37Footnote 48Footnote 49Footnote 50Footnote 54Footnote 55Footnote 56Footnote 60Footnote 61Footnote 67Footnote 76Footnote 77} Examples of terms relating to this domain included “media portrayals,”^{Footnote 32} “media coverage,”^{Footnote 37} “media context,”^{Footnote 33} “commercial”^{Footnote 67} and “social marketing.”^{Footnote 50} Five of the identified conceptual frameworks explicitly mentioned media- and marketing-related terms.^{Footnote 32Footnote 33Footnote 46Footnote 50Footnote 67}

Media content and social marketing endeavours can influence the stigma process by shaping societal attitudes and beliefs, and by reinforcing or countering negative stereotypes, interpretations and attributions of blame. Media content can also serve as a proxy for social interaction with stigmatized groups. If direct contact with stigmatized populations is rare, media may act as the primary or sole source of information that impacts judgments made in everyday life about these individuals.^{Footnote 32Footnote 77} To illustrate, one example is media portrayals of people with mental illnesses that depict these individuals as dangerous, unpredictable and criminal, thereby inciting fear and perceptions of social differentiation (or “othering”).^{Footnote 27Footnote 28Footnote 37} Another example is media portrayals that reinforce beliefs that certain groups are to blame for the harms they experience, and are therefore less deserving of social support and inclusion.^{Footnote 48Footnote 54Footnote 60} These narratives can legitimize discrimination towards marginalized groups.^{Footnote 27Footnote 32Footnote 35Footnote 77}

Media content can also reduce stigma by normalizing and promoting a greater understanding of certain behaviours or conditions.^{Footnote 28Footnote 48} Media content presenting positive and inclusive messaging, and including factual information about the causes, treatments and experiences of individuals with mental illness, can shift population-level misconceptions and negative stereotypes.^{Footnote 32}

Pedagogical factors

The domain of “pedagogical factors” pertains to the structure, design and implementation of educational content (e.g. curricula) and teaching initiatives, as well as educational institutions. A sample of terms relating to this domain included “trainings,”^{Footnote 58} “curricula”^{Footnote 37} and “educational programs.”^{Footnote 68} Eighteen works referenced terms that fell under this domain.^{Footnote 29Footnote 30Footnote 31Footnote 32Footnote 34Footnote 37Footnote 38Footnote 40Footnote 43Footnote 46Footnote 50Footnote 51Footnote 52Footnote 53Footnote 55Footnote 58Footnote 68Footnote 72} Pedagogical factors were referenced in three conceptual frameworks.^{Footnote 32Footnote 50Footnote 58}

These works described how both the form and the content of educational material can influence the stigma process. Overall, much like media content, pedagogical factors can influence the stigma process by shaping societal attitudes and beliefs. Across studies, it was noted that pedagogical material that promotes or enables direct contact between the public and stigmatized individuals,^{Footnote 38Footnote 50Footnote 51} that tackles misinformation^{Footnote 32} including negative stereotypes^{Footnote 29} and that normalizes stigmatized behaviours can reduce stigma by increasing awareness and empathy.

In contrast the absence of training regarding certain health or social conditions can influence how care and services are delivered by professionals working in these fields, thereby contributing to discriminatory practices.^{Footnote 38} One salient example is the way insufficient training among health care providers on HIV transmission risk can lead to insensitive and discriminatory practices towards patients living with HIV.^{Footnote 43Footnote 55}

Health care policies and practices

The domain of “health care policies and practices” refers to health system–level factors that pertain to the delivery of health care, including facility presence, accessibility and internal operational policies. Seventeen individual works referenced terms associated with this domain.^{Footnote 26Footnote 28Footnote 31Footnote 33Footnote 34Footnote 37Footnote 40Footnote 41Footnote 44Footnote 45Footnote 48Footnote 49Footnote 55Footnote 60Footnote 64Footnote 65Footnote 66} Of these, five proposed conceptual frameworks that referenced terms relating to this domain.^{Footnote 33Footnote 40Footnote 45Footnote 65Footnote 66} A non-exhaustive list of terms used included “health care system,”^{Footnote 33} “linkage to care,”^{Footnote 45} “healthcare quality”^{Footnote 66} and “institutional [health care] protocols.”^{Footnote 65}

Overall, structural determinants relating to health care policies tended to fall under two categories: factors relating to the availability and relative accessibility of high-quality health care services, and factors pertaining to internal policies on health service organization and delivery. An example of the former is systemic differences in the availability of prompt,^{Footnote 28Footnote 65} high-quality health care across communities, according to region of residence,^{Footnote 40Footnote 66} financial capacity,^{Footnote 44} and linguistic or cultural background.^{Footnote 34} Examples of health care initiatives that may reduce stigma are programs that improve the accessibility of services, such as policies that help patients navigate health systems and adhere to treatment plans,^{Footnote 44} and outreach activities (e.g. home visits) for underserved populations.^{Footnote 40}

Within clinical settings, institutional policies and structures can also contribute to discrimination against certain populations. Salient examples include physical structures, such as open-plan waiting rooms, and inter-staff communication policies that fail to protect patient’s confidentiality by allowing diagnoses to be overhead by others. These kinds of policies and structures can deter vulnerable populations from seeking care and can lead to the mistreatment of patients with stigmatized health conditions such as HIV^{Footnote 55} or TB.^{Footnote 48}

Biomedical technology

The domain of “biomedical technology” refers to structural determinants that pertain to the development, existence, use and effects of technology or medical products that are provided to patients in clinical health settings to treat diagnosed health conditions. A sample of terms relating to this domain include “advent of effective treatment,”^{Footnote 48} “control of disease”^{Footnote 33} and “treatment side effects.”^{Footnote 54} Six works—including one conceptual framework^{Footnote 33}—referenced determinants relating to this domain.^{Footnote 26Footnote 33Footnote 48Footnote 54Footnote 78Footnote 79}

Overall, studies documented how the existence of biomedical technologies that can prevent, manage or treat health conditions such as HIV,^{Footnote 26} TB,^{Footnote 48} or cancer,^{Footnote 54} to name some examples, can influence the stigma process by changing the visibility of the condition, or by shifting societal beliefs around the dangerousness, severity and permanence of the disease as well as the risk of transmission. Studies also identified how the targeted promotion of biomedical treatments to certain populations because of their behaviours or risk profiles may lead to increased stigma. For example, guidelines that recommend that individuals engaging in riskier sexual activities use pre-exposure prophylaxis (PrEP) to prevent HIV infection may cause people to conflate the use of this technology with sexual promiscuity, resulting in potential and current users fearing and experiencing discrimination.^{Footnote 26}

Diagnostic frameworks

The domain of “diagnostic frameworks” refers to structural determinants that pertain to developments in the understanding of disease etiology and classification. Seven works described determinants that relate to this domain,^{Footnote 33Footnote 35Footnote 46Footnote 50Footnote 58Footnote 63Footnote 79} two of which^{Footnote 33Footnote 46} proposed a conceptual framework that referenced terms relating to this domain. Examples of terms falling under this domain included “diagnostic practices,”^{Footnote 63} “DSM” (Diagnostic and Statistical Manual of Mental Disorders),^{Footnote 79} “genetic causal information”^{Footnote 50} and “medicalization.”^{Footnote 58}

Overall, developments in societal understanding of how diseases emerge can contribute to the stigma process by influencing societal perceptions of where responsibility for disease emergence should be placed. One example of this has been research findings that certain conditions such as mental health disorders or diabetes can be linked to underlying genetic factors. These developments in scientific knowledge can shift how societies attribute blame for incidence of these conditions.^{Footnote 33Footnote 50}

Developments in societal understanding of how health and social conditions should be classified can contribute to the stigma process by influencing perceptions of what is considered abnormal. By labelling certain conditions as “disorders” or “diseases,” clinical diagnoses can imply the need for corrective treatment and facilitate the ostracizing of affected individuals. One illustrative example is how previous editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM) pathologized homosexuality and gender nonconformity, contributing to negative stereotypes of and discrimination against LGBTQ+ communities.^{Footnote 58}

Public health interventions

The domain of “public health interventions” refers to policies, programs and initiatives led or mandated and financially supported by public health stakeholders. Seven works, including one conceptual framework,^{Footnote 50} described determinants relating to this domain.^{Footnote 42Footnote 49Footnote 50Footnote 60Footnote 63Footnote 67Footnote 77} Terms used in these studies included “health promotion initiatives,”^{Footnote 50} “public health initiatives”^{Footnote 63} and “community-based outreach interventions.”^{Footnote 42}

Overall, studies described how public health interventions can influence the stigma process by shaping societal norms and beliefs both positively and negatively. For example, health promotion interventions that perpetuate messaging around the need to change certain individual behaviours (e.g. weight control, substance consumption) can have unintentional negative influences on societal beliefs by reinforcing narratives of blame and responsibility for individuals engaging in these behaviours or experiencing resulting health conditions.^{Footnote 26Footnote 50} One illustration of this was public health messaging of zero-tolerance of alcohol consumption during pregnancy. This messaging can perpetuate negative societal beliefs of pregnant individuals who drink or of mothers of children with fetal alcohol spectrum disorder, and can lead to hesitancy among pregnant individuals who do drink to consult health and social services.^{Footnote 63}

Community-based public health interventions can also influence the discrimination and status loss components of the stigma process by intervening on resource distribution. One example of such interventions is community-based public health initiatives designed to improve access to health and harm reduction services for populations that may have less trust in medical establishments due to historical discrimination, such as individuals living with HIV or sex workers.^{Footnote 42Footnote 60} These types of interventions are believed to be protective against stigma as they promote respect and inclusivity, and represent sources of empowerment for vulnerable populations.^{Footnote 42}

Conceptual framework

The ten domains of structural determinants are summarized in Figure 3. This figure is a simplified conceptual framework that depicts the structural determinant domains identified in the reviewed literature, and how they were described in relation to the stigma process as defined by Link and Phelan.^{Footnote 1} Three large arrows flow from the structural determinants to the stigma process. These arrows indicate how structural factors can influence the stigma process overall, and more precisely by shaping the psychosocial “drivers” of the stigma process, which include societal beliefs and stereotypes based in fear, normative judgment or blame as well as lived experiences of discrimination and status loss.^{Footnote 3Footnote 72}

Note: This figure provides a visual summary of the narrative synthesis presented in this review. It should not be interpreted as providing a comprehensive summary of all possible structural determinant domains, nor as depicting firm causal ties between each of the domains and elements of the stigma process.

Text description: Figure 3

Figure 3. Conceptual framework of identified domains of structural determinants and their relationship with the stigma process

This figure demonstrates the conceptual framework of identified domains of structural determinants and their relationship with the stigma process. The 10 identified structural determinants of stigma are legal frameworks, economic policies, media and marketing, health car policies & practices, diagnostic frameworks, welfare policies, social & built environment, pedagogical factors, biomedical technology, and public health interventions. These determinants are linked to the stigma process at both its stages of drivers and experiences. Within the stigma process, it starts with drivers which are labels leading to beliefs leading to stereotypes leading to separation (us vs. them). From there drivers turn into experiences of discrimination and status loss. These experiences then manifest themselves on the structural determinants of stigma. Additionally, levels of interaction at which stigma can manifest include population/systemic, institutional/social, interpersonal, and individual.

In line with the reviewed conceptual frameworks and models, including the CPHO’s Stigma Pathways Model,^{Footnote 3} Figure 3 also includes a graphical representation of the levels of population interaction at which stigma can be enacted, from the individual (internalized) to the systems level. Further, many studies and existing frameworks acknowledged that stigma processes are influenced by historical social inequities and discrimination,^{Footnote 3Footnote 46Footnote 72} thus creating feedback loops between structural practices and stigma processes through time. To represent these dynamics, Figure 3 includes a large arrow that flows from the stigma process back to structural determinants.

Since this review did not involve an exhaustive literature search and further research is needed to confirm both the causal associations between factors and the effectiveness of various stigma-reduction interventions across populations and settings, Figure 3 should not be interpreted as providing a comprehensive summary of all possible structural determinant domains, nor as depicting firm causal ties between each of the domains and elements of the stigma process. Instead, it was designed to provide a visual summary of the narrative synthesis presented in this review that can be used as a tool to structure policy discussions on ways to orient public health interventions to reduce stigma in Canada and abroad.

Discussion

This rapid review was designed to identify and summarize structural determinants of stigma in Canada and other OECD settings, in order to guide future research and intervention. An analysis of findings from 53 works from peer-reviewed and grey-literature sources, 15 of which included proposed conceptual frameworks that mentioned at least one type of structural determinant of stigma, this review is the first known summary and conceptual framework of structural determinants of stigma across health and social conditions. Applying a thematic analysis of structural-level factors documented in the literature, we identified and defined 10 distinct domains of structural determinants of stigma. These domains were legal frameworks, welfare policies, economic policies, social and built environments, media and marketing, pedagogical factors, health care practices and policies, biomedical technology, diagnostic frameworks, and public health interventions.

This summary fills an important gap in the existing literature by bringing together findings from a wide range of fields of stigma research, and elucidating types of factors that operate at the contextual level to influence societal beliefs, negative stereotypes, discrimination or status loss across multiple social contexts and populations.^{Footnote 14} This review and the proposed conceptual framework are tools that can be used to structure future policy conversations; the ten domains of factors and the governance sectors to which they relate can be systematically considered when seeking to address and prevent stigma. As structural-level factors can contribute to social stratification and health inequities,^{Footnote 9} each identified domain merits attention.

Strengths and limitations

One strength of this rapid review is its focus on determinants of stigma across health outcomes and physical or social attributes. The resulting summary is therefore applicable to a wide range of substantive domains of public health and social policy. Another strength is its focus on structures that could theoretically be modified through intersectoral public health intervention, with potential population-level impacts on the stigma process.^{Footnote 9Footnote 15}

Nonetheless, this review has certain limitations. Due to the non-exhaustive search strategy of the rapid review design, relevant studies—particularly individual qualitative or quantitative studies—may have been missed by our search strategy, and thus, certain examples of structural determinants may have been missed as well. However, given that the majority of published works included in this review are evidence summaries and frameworks of structural determinants, we expect that the contribution of missed studies is likely to be minimal.

Another limitation is that many of the works reviewed here did not include sufficient detail about the process by which their results were obtained and synthesized. The opacity of the data-generating process of these works calls into question the comprehensiveness of their findings. The resulting summary should therefore be used primarily as a conceptual guide rather than an exhaustive review. Including this element in future texts will be a necessary step for strengthening public health literature on stigma.

Finally, this review summarizes findings for the general context of OECD nations. We did not seek to explore structural determinants within a specific jurisdiction. Since the impacts of structural determinants on the stigma process may be heterogeneous across local contexts, future research and policy conversations on ways to address stigma should consider how local eco-social or political environments may influence the structural determinants of stigma or the effects of potential interventions on stigma reduction.

Conclusion

This review complements previously published summaries of the influence of stigma as a determinant of health. Here, the structural determinants of stigma as a social outcome occurring across health and social conditions were explored. A rapid review of existing evidence suggests that there are at least ten domains of structural determinants of stigma. The present review’s conceptual framework of these domains can be used as a tool to structure future policy conversation across sectors on ways to reduce stigma at a population level.

Acknowledgements

We would like to thank Lisa Glandon and the Information and Knowledge Management Division from the Health Canada Library, as well as Dr. Kimberly Gray, Michèle Boileau-Falardeau and Dr. Louis Turcotte for their contributions and feedback.

Conflicts of interest

The authors have no conflicts of interest to declare.

Authors’ contributions and statement

All authors contributed to the methodological design of the literature review. AB and BHMF conducted the literature review search and identified relevant works. All authors extracted data, analyzed identified works and drafted the article. AB and CBF revised the article.

The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.