Structural interventions that affect racial inequities and their impact on population health outcomes: a systematic review

Racism is a system of power that manifests at all levels of society, resulting in the differential treatment and access to resources and opportunities based on one’s racial identity [1, 2]. Racial health inequities, power imbalances, and injustices act on institutional and structural levels. Jones’ framework, Levels of Racism, aligns with this perspective and purports that racism operates at three levels: institutionalized, personally mediated, and internalized [3]. Of these three, the institutional, or structural, level is the most fundamental and the necessary starting point for meaningful change [3]. Structural racism refers to the historical and ongoing pervasive reinforcement of racism within society and its interconnected, codependent institutions [4]. Specifically, structural racism is how populations are disadvantaged due to discriminatory systems and inequitable distribution of key resources at every level of government and within every sector of society [4]. At the structural level, racism is entrenched within institutional structures, processes and values, perpetuating historical injustices and restricting access to housing, education, employment and health and social services [5]. For example, the “redlining” to deny mortgage loans in predominantly Black neighbourhoods, preventing the accumulation of real estate wealth and perpetuation of wealth inequity [6‐8]. Schools within redlined neighbourhoods receive fewer resources resulting in long-term educational inequities [4, 9, 10]. Developers were less likely to invest in redlined neighbourhoods, leading to reduced access to healthy groceries and healthcare facilities [11‐13]. This ultimately created a pathway between racism and health, due to more proximal outcomes such as social deprivation and economic injustices [4]. Similarly, Williams and Mohammed’s Framework for the Study of Racism on Health outlines interconnected pathways, where basic causes (e.g., structural racism, institutions) and one’s social status (e.g., gender, socioeconomic status, race), influence proximal pathways (e.g., opportunities, resources, psychosocial stressors), which result in various responses and behaviours that can lead to a spectrum of negative population health outcomes [14].

Racism affords privileges for individuals and groups deemed superior based on actual or perceived proximity to Whiteness and leads to explicit and implicit maltreatment and disenfranchisement of those deemed inferior [5]. Critical Race Theory positions race as a social construct and outlines the interplay between race and racism, the means of power and domination [15]. Critical Race Theory seeks to eliminate the racial imbalance of power through the analysis of the pervasiveness of racism and its influence on society [15, 16] . Structural determinism, a key tenet of Critical Race Theory [1] posits that macro-level forces have a crucial role in creating and maintaining inequities and that racism and intersecting systems of power function to preserve the power of the dominant group [1]. The application of Critical Race Theory can improve public health policy and practice as it encourages researchers, practitioners, and policymakers to act upstream with interventions that impact the root causes of racialized inequities [1]. A key element of this work is the need to disrupt White supremacy and acknowledge the pervasiveness of Whiteness through all levels and settings within society [17, 18].

There is an extensive body of research demonstrating that structural racism is an important determinant of health [19]. Structural racism impacts health directly as well as through persistent racial income inequities and socioeconomic status, which are known to drive racial health inequities. Data from the 2017 Pan-Canadian Health Inequalities data tool found the prevalence of diabetes in Black people in Canada was 2.1 times higher than in their White counterparts; similarly, Black people are more likely to report having fair or poor health [20]. Indigenous populations in Canada endure disproportionately higher rates of infant mortality, tuberculosis, obesity, diabetes, youth suicide and environmental contaminants, resulting from ongoing colonialism and racism [21].

Despite the role of structural racism in health outcomes, there are limited studies that evaluate structural racism and related interventions [22, 23]. Research is largely focused on individual and interpersonal racism, rather than how racism embedded in systems at the structural level affects racial health inequities and suboptimal health outcomes [23]. Consequently, there is a need to measure the impacts of racism at not only the individual level, but at the structural levels [24]. Upstream interventions, which extend beyond those at the level of proximal outcomes, are necessary to improve population health inequities [14]. These include interventions that change the social, physical, economic or political environments that influence health, such as economic stability, educational and employment opportunities, discrimination and racism, access to healthy food and healthcare [25, 26].

Previous literature reviews have highlighted examples of structural interventions that have affected racial health inequities [4, 25]. Bailey et al, 2017, provides a comprehensive overview of the pathways through which structural racism can affect population health and cites several examples of interventions designed to reduce racial health inequities, but does not include a systematic search or outcome data for these studies [4]. Similarly, Brown et al, 2019, thoroughly discusses the challenges and opportunities for structural interventions to reduce racial health inequities and includes several examples of policy interventions to improve health, but again did not systematically search for studies or include outcome data [25]. These reviews help highlight the causal links between structural interventions and racial health equity, but without health outcome data it is not possible to quantify their impact for racialized populations. There are examples of systematic reviews that have been conducted to evaluate structural interventions that may affect health equity, but these have been limited to one policy domain and often do not report data for health outcomes stratified by race [27‐31]. By limiting analysis to one policy domain, it is not possible to determine the relative potential influence of different policy domains on racial health equity. At this time, a systematic review of the literature evaluating the effect of structural interventions on racial health equity that reports health outcomes stratified by raceis needed to help inform future policy development.

This systematic review was conducted alongside a larger scoping review on interventions to address structural determinants that affect population health outcomes, with a narrower focus on racial health inequities, to inform policy development in Canada. For the scoping review, structural determinants of health were defined according to the Pan American Health Organization (PAHO) Equity Commission’s Conceptual Framework for Structural Drivers of Health Equity and Dignified Life [32] and the World Health Organization’s Commission on Social Determinants of Health (CSDH) conceptual framework (Fig. 1) [33]. According to this framework, structural drivers, which include governance, economic policies, social policies, public policies and cultural or societal values, influence an individual’s status within the social hierarchy [33]. Social hierarchies systematically allocate unequal power and resources according to the socioeconomic and political context, resulting in inequity [33]. This systematic review is novel in its investigation of structural level interventions for any policy type that affect structural drivers through the socioeconomic and political context to affect inequities in racial health outcomes.

Database searching for the broader, related scoping review retrieved 24,311 records and citation tracking retrieved an additional 261, for a total of 26,055 records. After removing duplicates, 23,799 records were screened by title and abstract, resulting in 776 records for full text review for inclusion in the broader scoping review. Of those 776 records, 292 articles were screened for health outcomes by race for eligibility for this systematic review, resulting in 29 included articles. Studies excluded at this stage are listed in Additional file 2. See Fig. 2 for a PRISMA flow chart illustrating the article search and selection process.

This systematic review explores evidence for structural-level interventions that affect racial health inequities through determinants outlined by the WHO CSDH framework [33] (Fig. 1). Among the 29included studies, only 18 interventions specifically address race- or ethnicity-based inequities; 13 of which act through improving socioeconomic circumstances. Most interventions were targeted toward low-income populations, except for immigration policies and policies for Indigenous populations. There were no studies of policies designed to mitigate anti-Black racism, a notable gap that suggests anti-Blackness within governments and academia.

Overall, studies reported mixed effects of interventions that affect the structural determinants of health, as defined by the WHO CSDH framework. Studies of financial and nutrition safeguard policies largely found no or mixed effects, with the notable exception of two moderate-quality studies of minimum wage policies which were shown to reduce HIV incidence and improve birth outcomes for Black populations. Discriminatory policies, such as anti-immigration enforcement and abortion restrictions were shown to negatively impact birth outcomes and mental health outcomes for affected Hispanic and Black populations. Conversely, two moderate-quality studies of DACA found improvements to birth and mental health outcomes for DACA-eligible populations. Findings from policies for Indigenous populations were mixed but provided some evidence that enhancing self-governance may lead to improved outcomes. While all included interventions had potential to affect structural determinants of health according to the WHO CSDH framework (Fig. 1), findings were inconsistent for different racial and ethnic populations. Findings from this review support Critical Race Theory’s tenet that racism functions differently for different races according to a racial hierarchy [1], and support the need for population-specific interventions rather than broader, non-targeted interventions. Policies that address systemic barriers encountered by different racialized groups may be more effective than those applied to broader populations, as suggested by Assari, 2018, who proposes that the intersection of race, socioeconomic status and gender shape exposures to risk and protective factors [78]. For example, findings of a policy analysis by Carvalho et al, 2021, support a targeted approach, reporting positive impacts on health equity by policies directly addressing Black maternal health [79]. The studies included in this review that targeted specific populations were also more likely to show positive effects on inequities, for example DACA for Hispanic populations in the USA [55, 60] and the Indigenous Land and Sea Management Programs in Australia [66]. Based on the results of this systematic review, policies designed to address racial health inequities experienced by specific populations are more likely to reduce health disparities than broad policies that target populations based solely on socioeconomic status.

This review illustrates a scarcity of evidence evaluating the health impact of interventions addressing structural racism [4, 22]. In a recent systematic review of studies of institutionalized racism in the top 50 highest-impact journals in the USA between 2002 and 2015, the term was included in the article abstract or title of only 25 papers and as a critical concept in only 16 papers [23]. Some limitations of the body of evidence included in this review may be due to limited data available regarding race. Many included studies were conducted in the USA, where racial and ethnic data in public systems has limited consistency, reliability and comprehensiveness across states [80‐83]. These data challenges have been reported in the Canadian context as well, where race-based data are either not collected, reported or account for small proportions of the dataset, raising issues of privacy and limiting analyses of intersectional identities such as ethnicity, and immigration status [84‐86]. Some critical race scholars have argued that the lack of data reflects an ongoing denial of the salience of racism as a determinant of health in the Canadian context.

A single RCT was included in this systematic review [46]. This trial examined the effect of the New Jersey FDP Welfare Reform’s stricter rules on the mental health of low-income mothers receiving benefits. While the study found that Black women subject to FDP Reform had a lower incidence of clinically diagnosed anxiety or depressive disorders, the study’s findings are limited by its design. Study authors chose clinical diagnosis of a mental health disorder as their primary outcome, rather than self-reported symptoms, citing discrepancies between female self reports of mental distress and clinician reports [87]. The biases faced by women seeking health care are well-documented [88‐91] and for Black women in particular [92‐95]. Authors also cited a study that demonstrated depressive symptoms are often short-lived, but this study reported on depressive symptoms due to bereavement, which may not be transferable to the experiences of low-income mothers seeking welfare [96]. Authors did not address that Black women face barriers to accessing mental healthcare [97‐99]. Authors instead discuss so-called “welfare heritage” of Black women with more frequent and longer use of welfare services [46]. The methodological biases and racial bias in the design of this RCT highlight the need for racial equity-centered evaluations of interventions.

Notably, the interventions examined by the studies included in this review were largely limited to one policy domain, for example, interventions that impacted finances or interventions that affected immigration rights. However, calls to action to counter structural racism have emphasized the need for broad, all-policy approaches [100‐103]. It is perhaps not unexpected that individual studies of single-policy interventions found mostly mixed or no effect on health inequities, as a systemic issue requires systemic solutions. Two studies of minimum-wage levels and one study of EITC expansion [43, 45, 48] found positive effects of increased minimum wage on health inequities, supporting unconditional cash transfers that can impact daily living. Similarly, two studies of DACA found reduced inequities [55, 60], possibly because the opportunities granted by DACA affect many areas of life.

A limitation of this systematic review is that for most included studies it is difficult to attribute change in outcomes to the intervention, which is a consistent challenge reported previously in policy evaluations [104‐106]. The most rigorous evidence for the effect of an intervention would be produced by a randomized controlled trial design where study groups receive identical treatment except for the intervention under study. Due to ethical and logistical barriers however, randomized controlled trials are rarely feasible, which is illustrated in this review with nearly all included studies using a quasi-experimental design. Given the pervasive effects of structural racism, for studies that conducted comparisons between racialized groups, it is unlikely that the groups had similar baseline experiences apart from the intervention [100, 107‐109]. This limits both the effect attributable to the intervention, as well as the magnitude of the effect as the intervention is but one of many differences between groups. For these reasons, changes in outcome for the same population group prior to and after implementation of an intervention will provide more rigorous evidence for the effect of an intervention.

This review had several limitations in its design. Study selection was guided by the WHO CSDH framework (Fig. 1), to ensure that included interventions target the structural determinants of health. This review found that interventions must be tailored for specific racialized groups, but the guiding WHO framework was applied to all populations and may not truly reflect the determinants of health for specific groups. Studies were also limited to those conducted in OECD countries, to reflect Canada’s context of democratic government and high-income economy most closely. This limits the findings of this review to other high-income democracies, and results may not necessarily apply to low- or middle-income countries. Structural racism continues to impact health inequities in low- and middle-income countries, and further research is required to investigate potential interventions to mitigate these inequities [110‐112].

The findings of this systematic review contribute to an understanding of how to make meaningful change to improve racial health inequities. Findings are complementary to those of other policy interventions that have not measured health outcomes, but demonstrate impact on the social determinants of health [100]. For example, the Purpose Built Communities in Atlanta, USA, which engaged community residents in designing and implementing neighbourhood design and education programs, has shown positive effects for crime, housing, and employment [4]. This intervention was effective likely because it was multilevel and upstream, targeting living conditions to improve health outcomes [113]. The interventions in this systematic review were largely limited to one policy domain and while they had the potential for upstream impact on health outcomes (in alignment with the WHO CSDH framework, Fig. 1), most were insufficient in scope to cause measurable improvements. Structural racism functions across multiple domains which are mutually reinforcing. While policy action in one domain may reduce racial health inequities, this may not be sufficient to counter the impacts of racism in other domains. For policy interventions to effect measurable and meaningful improvements to racial health inequities, policies must be designed to synergize across domains for upstream impact on daily living conditions [2, 4, 14, 113].

Structural racism remains a pervasive issue with inequitable effects on health for racialized populations. One-dimensional policy interventions lack the impact on daily living conditions to effect measurable change in health outcomes. The WHO CSDH framework defines pathways through which interventions can address the systemic barriers faced by racialized groups and impact the structural determinants of health. Future versions of this framework should consider specific contexts for different racialized populations, to ensure that it can be applied broadly as intended. In the current state of research, few interventions that target the structural determinants of health have been evaluated for their effect on health outcome inequities. Significant and long-term investments into dedicated programs for research, specific and unique to the needs of specific racialized groups, are necessary to address the root causes of structural inequities. While the overall goal of improving health outcomes is common for all populations, concerted efforts to develop, implement and evaluate policies that address the unique contexts of structural racism facing different racialized populations are required to reduce inequities.