Racial discrimination and health: a prospective study of ethnic minorities in the United Kingdom

Discrimination is defined as the differential treatment of an individual based on a socially ascribed characteristic [1]. In the United Kingdom (UK), the 1965 Race Relations Act [2] outlawed discrimination on the grounds of colour, nationality and ethnic or national origins. Race remains a protected characteristic under contemporary equality law [3]. Despite this legislative effort, ethnic inequalities in education, work, health and criminal justice remain [4].

Discrimination on the basis of ethnic origin is regarded as the most common type of prejudice in Europe, with 64% of adults perceiving racial discrimination to be widespread in a survey of 27,718 people [5]. In Britain in 2017, 26% of a representative sample described themselves as racially prejudiced [6], and race continues to be the most common motivator for hate crime incidents [7, 8]. Against the backdrop of the vote to leave the European Union (Brexit), hostility towards migrants and the growth in right-wing nationalist movements [9], these figures reflect a rise in reported racial discrimination in both the UK and Europe [5, 6].

A growing body of research has investigated discrimination as a determinant of mental health [10‐12] and to a lesser extent physical health [11]. In an early meta-analysis of 110 studies, discrimination was linked with poor mental health, including psychological distress and decreased life satisfaction [11]. A sub-set of 36 studies in the review investigated associations with physical health. Significant associations were detected in a pooled analysis with various outcomes including hypertension and acute cardiovascular responses to laboratory discrimination protocols. A more recent meta-analysis of 328 studies focusing on discrimination and mental health outcomes alone, again observed that those who perceived discrimination had poorer mental health [12]. This finding was also detected in an independent analysis of 211 cross-sectional studies linking racial discrimination with poor mental health [12].

Racism is a recognised social determinant of health and a driver of ethnic inequities in health [13]. It can be understood as a complex, organised system embedded in socio-political and historical contexts, that involves classifying ethnic groups into social hierarchies. These groups are ideologically assigned differential value, which drives disparities in access to power, resources and opportunities [14, 15]. It occurs at both structural and individual levels (self-reported experiences of racial discrimination) [14, 15].

Several reviews and meta-analyses have focused solely on perceived racial discrimination and health outcomes [13, 16‐18]. The largest study to date meta-analysed the results from 293 studies and assessed both mental and physical health outcomes [16]. In this analysis, racial discrimination was associated with poorer overall mental health including greater psychological distress, poorer life satisfaction and poorer general mental functioning in independent analyses. Racism was also linked with poorer general health and poorer physical health overall, though few effects remained significant when looking at specific physical health outcomes in separate analyses.

Racial discrimination at the structural and individual level is theorised to impact health through several mechanisms [15]. At the structural level racial discrimination may operate through the unfair allocation of societal resources that are determinants of health (e.g. education, employment, housing) [14, 15] and through differential access to healthcare, as well as perceived poorer quality of care [19]. Another mechanism linking racial discrimination and health could be through the dysregulation of stress-related biological processes [20]. Frequent exposure to racial discrimination is a chronic stressor and has been linked with dysregulated cardiovascular, neuroendocrine and inflammatory processes [21, 22] which in turn impact both physical and mental health. Individual health risk (e.g. smoking, alcohol consumption) could link perceived racial discrimination and health, as means of coping with or avoiding discrimination [23, 24].

Although a growing number of studies have investigated the link between racial discrimination and health, there are still areas where more research is required. In the 2015 racism meta-analysis of almost 300 studies, only 9% of the data included were prospective [16]. The authors aimed to compare the effect sizes of the cross-sectional and prospective studies included in their review but were unable to conduct this analysis for the physical outcomes data, emphasising the need for more prospective studies on physical health outcomes in particular.

Further, the literature is dominated by United States (US)-based studies drawn from convenience samples [12, 16]. In the latest racism and health meta-analysis, over one third of the articles included were drawn from student samples and only nine (2.7%) of the included studies were UK-based [16]. This is important as the makeup of ethnic minority groups in the UK differs from that of the US, with those of South Asian backgrounds forming the largest minority group [25]. In addition, all of the UK studies were cross-sectional in nature and focused on mental health, with physical outcomes such as the number of physical illnesses [26] and self-rated health [27] included in only two of the studies.

To date, one UK study has assessed the relationship between racial discrimination and health prospectively. In an analysis of the UK Household Longitudinal Study (UKHLS), the authors found that those who reported racial discrimination had poorer mental functioning scores 4 years later [28]. They also reported a dose-response relationship between the experience of racial discrimination and mental health, with those who reported racial discrimination at more than one timepoint over a 3-year period experiencing a greater deterioration in mental functioning.

Overall, there is a dearth of prospective evidence on the link between racial discrimination and health in UK samples, particularly in relation to physical health outcomes.

To address these gaps in the literature, the present study set out to assess cross-sectional and prospective associations between racial discrimination and health in a large community-dwelling UK population cohort. Specifically, we were interested in psychological distress, mental functioning and life satisfaction, as indicators of mental health, as well as self-rated health and physical functioning as markers of physical health, along with limiting longstanding illness as an indicator of impairment. We hypothesised that those who perceived racial discrimination would have poorer health across all measures both cross-sectionally and prospectively.

A total of 4883 participants were included in our analysis and of these 998 (20.4%) reported ethnicity (n = 854) or nationality (n = 144) discrimination. The characteristics of the sample at wave 1 in relation to racial discrimination are displayed in Table 1. Those who perceived racial discrimination were younger on average and were more likely to hold a university degree than those who did not perceive racial discrimination. There were no differences in sex or income, but reports of racial discrimination did vary by ethnic group. Those in the Indian (23.3%) and in the Other ethnic group (24%) were most likely to report experiences of racial discrimination. Further detail on the types of racial discrimination and the settings in which the racial discrimination occurred for the different ethnic groups can be found in Supplementary Table 1.

In this large UK-based prospective sample of ethnic minority participants, we detected associations between racial discrimination and poorer health. Cross-sectionally, those who reported racial discrimination had a greater likelihood on average of limiting longstanding illness and poor self-rated health, than those who did not report racial discrimination. Racial discrimination was associated greater psychological distress, lower life satisfaction, and poorer physical and mental functioning. In prospective analyses, those who reported racial discrimination had a greater likelihood on average of limiting longstanding illness and poor self-rated health than those who did not report racial discrimination. Racial discrimination was associated with greater psychological distress and poorer mental functioning over a two-year follow-up period, regardless of baseline health. No significant prospective associations with physical functioning or life satisfaction were detected.

To our knowledge, this is the first prospective UK-based study to investigate both mental and physical health outcomes in relation to racial discrimination. One earlier analysis of the UKHLS found that those who reported racial discrimination had poorer mental functioning over a 1–4 year follow-up period [28]. The current study also found a prospective association between racial discrimination and poor mental functioning. Our study builds upon previous findings by additionally showing that this association is independent of baseline mental functioning. We also observed a prospective association with psychological distress, another marker of mental health, with those reporting racial discrimination experiencing an increase in psychological distress over time. We did not detect a prospective association between racial discrimination and poorer life satisfaction. Mean scores trended in this direction but the association did not reach statistical significance. A 2015 longitudinal analysis of the US-based Health and Retirement Study with over 6000 participants also failed to detect a prospective association between racial discrimination and decreases in life satisfaction [45], and pooled analyses have been unable to investigate prospective associations with life satisfaction due lack of sufficient evidence [12, 16]. A possible explanation for this null finding, consistent with earlier work, is that racial discrimination is more strongly associated with negative mental health outcomes such as psychological distress than with positive outcomes such as life satisfaction [12, 16]. Another potential reason for these findings relates to duration of follow-up, as review evidence suggests that a recent experience of racial discrimination may be more strongly associated with poor mental health and more weakly related to life satisfaction measures [16]. Our follow-up period of 2 years was relatively short which may have contributed to these results.

Reviews in the field [16, 17] have highlighted the need for more prospective evidence, particularly for physical health outcomes [16]. We found that participants who reported racial discrimination were more likely to report having a limiting longstanding illness and poorer self-rated health, independent of baseline status. Meta-analytic evidence has demonstrated an association between racism and poor general health and worse physical health outcomes [16]. We built upon this predominately US-based data (a considerable portion of which used convenience sampling) to demonstrate prospective associations between racial discrimination and physical health outcomes in a representative sample of UK adults from ethnic minority groups. We failed to observe a prospective association between perceived racial discrimination and physical functioning, although participants who reported racial discrimination had slightly lower physical functioning scores prospectively than those who did not report racial discrimination. This lack of association may indicate that ongoing experiences of racial discrimination had already made an impact on physical functioning at the time of wave 1 survey, limiting the scope for further significant decreases in this measure over time, particularly as we took baseline physical functioning into account in our analyses. Another possibility, is that the etiological period involved for a decline in physical functioning may differ from that of mental functioning [14]. These outcomes were measured using the same tool (SF-12) but only mental functioning was significantly associated with racial discrimination over the follow-up period.

Review evidence based on US data suggests that associations between racial discrimination and health may vary depending on ethnic group [16]. In our sensitivity analysis, the cross-sectional results for life satisfaction and impairment and physical health outcomes were non-significant for the Black group. Prospectively the findings for psychological distress and mental functioning were non-significant for the South Asian group. Whereas, life satisfaction was found to significantly decline for the Other group over the follow-up period. Taken together these results suggest associations with health outcomes are strongest for South Asian and Other groups cross-sectionally, while prospectively racial discrimination appears to most consistently impact mental health outcomes in Black and Other ethnic groups. These findings should be interpreted with caution due to the likelihood that some of our analyses were underpowered.

In our cross-sectional analyses, we found that those who perceived racial discrimination had poorer mental health, with greater psychological distress, poorer mental functioning and lower life satisfaction. Previous work in UKHLS has demonstrated a cross-sectional association with psychological distress using pooled data across three waves of data collection [46]. To our knowledge no prior UK-based work has reported on cross-sectional associations with poor mental functioning and low life satisfaction. These findings are consistent with earlier work in other countries [12, 16, 45].

We detected links between racial discrimination and poor physical health and impairment. Specifically, we found that those who reported racial discrimination had poorer self-rated health, poorer physical functioning scores and a greater likelihood of having a limiting longstanding illness than those who did not report racial discrimination. Earlier work using the 1993/1994 UK-based Fourth National Survey of Ethnic Minorities survey reported associations between perceived racial discrimination and poor self-rated health [27, 47] and limiting longstanding illness [47]. Our more recent findings from 2009/2010 suggest that these deleterious associations remain an issue for minorities in the UK.

We detected stronger associations between racial discrimination and health for cross-sectional than for prospective comparisons, in keeping with earlier evidence [16]. However, cross-sectional work cannot determine whether reports of racial discrimination stimulate poor mental and physical health or whether perceptions of racial discrimination are a manifestation of feeling suboptimal mentally or physically. Our prospective findings therefore add to the field in establishing that racial discrimination predicts poor mental and physical outcomes prospectively, net of baseline associations, supporting the hypothesis that racial discrimination has adverse consequences for future health.

With regard to the pathways through which racial discrimination negatively impacts health, there are several possibilities that could help explain our results. One mechanism linking racial discrimination and health may be through the dysregulation of stress-related biological processes. In response to perceived chronic discrimination, stress processes may be frequently activated, which over time may result in disturbances across multiple biological systems, in line with the theory of allostatic load [20]. Review evidence indicates discrimination is associated with heightened cardiovascular responses to stress [11, 21], though it is unclear whether this translates into an increased risk for clinical hypertension [48]. Another biological mechanism that may link discrimination and health is through activation of the hypothalamic-pituitary-adrenal (HPA) axis. Several reviews have linked racial discrimination [21, 49, 50] with changes in various cortisol parameters, which in turn have been linked with poorer mental and physical health [51, 52]. Deleterious changes in other biological processes such as heightened inflammation [22] and alterations in DNA methylation of stress-related genes [53] have been linked with discrimination in recent studies. Alterations in these stress-related biological processes offer a plausible link to negative changes in physical [54, 55] and, mental health outcomes [51, 56]. Racial discrimination has also been associated with disturbances in neurobiological processes, with alterations observed in brain areas such as the anterior cingulate cortex, prefrontal cortex and amygdala which overlap with pathways associated with poor mental health [57].

Individual health risk (e.g. smoking, alcohol consumption etc.) could link perceived racial discrimination and poor mental and physical health, either as a method of coping with the negative psychological effect of perceiving racial discrimination (e.g. excessive alcohol consumption as a coping mechanism) or as a barrier to engaging in healthy behaviours (e.g. avoiding a health service perceived to be discriminatory). Racial discrimination has been associated with smoking [23, 58, 59], excessive alcohol consumption [23, 60], as well as substance abuse [61, 62]. Review evidence has linked discrimination with poor sleep [63] as well as weight gain in prospective studies [24]. This individual health risk offers a plausible indirect pathway linking racial discrimination with both poor mental [64, 65], as well as physical health outcomes [66].

Another possibility at the broader structural level is that racial discrimination may impact health through differential access to societal resources such as education, employment, welfare and criminal justice [14, 15]. In the UK, a 2016 report documented persistent ethnic disparities in educational attainment, employment, access to fair pay and adequate housing, as well the over-representation of ethnic minorities in the criminal justice system [4]. Further, data from this report highlight inequalities in access to healthcare among ethnic minority groups [4]. While meta-analytic evidence indicates that racial discrimination is associated with more negative patient experiences of health services, as well as delaying/not getting healthcare and lack of treatment uptake [19]. As these factors are social determinants of health in of themselves [13‐15], they may act as a pathway through which perceptions of racial discrimination can act to negatively influence health.

The results of the current study need to be assessed in terms of strengths and limitations. There is a dearth of prospective evidence on the link between racial discrimination and health in UK samples, particularly in relation to physical health. Our large sample of ethnic minority participants allowed us to examine changes in mental and physical health over 2 years, and demonstrated both cross-sectional and prospective associations. We also adjusted statistically for factors that potentially confound associations, including age, sex, socioeconomic status and ethnicity. Although controlling for covariates does not tease out the complexity of the relationships between perceived racial discrimination and these sociodemographic characteristics [43]. For example, socioeconomic status contributes to racial inequalities in health [43], while racial discrimination can compound these disparities and can be conceptualised as an indicator of structural racism [13]; statistical adjusting for socioeconomic status does not capture these relationships.

The study of racism is a complex and contested area of research [67, 68] and our study was not without limitations. Our measure of perceived discrimination was not specifically tailored for racial discrimination, as participants in the could attribute their experience to other forms of discrimination as well (e.g. sexism, ageism). There is evidence that the exposure instrument can influence associations between racism and physical and mental health outcomes [16]. Participants were able to attribute multiple reasons for their report of discrimination, which could have helped to avoid priming and this measure has been used to assess racial discrimination in previous work [28]. However, it is possible that measures such as the Schedule of Racist Events scale [69] and the Perceived Racism Scale [70] with more specific items on racist degradation and experiences of racism in personal and professional contexts could have garnered different results. Further, the self-report individual measure of racial discrimination employed in our study does not capture the structural conditions that shape the varied ways in which racial discrimination operates [14]. We only assessed perceived racial discrimination at baseline in this study and did not investigate whether racial discrimination experiences were persistent or changed over time.

Racial discrimination was assessed by self-reports of experiences in the past year and was therefore subject to recall bias. Our findings reflect the perception of racial discrimination rather than objective encounters with racial discrimination. It is possible that objective encounters with racism and perceiving one’s self as the target of racial discrimination might have different consequences for health. Experimental studies involving exposure to discriminatory scenarios have been used to investigate the health impact of objective exposures to racial discrimination. However, these studies may not represent a gold standard for the study of the relationship between discrimination and health, as meta-analytic evidence indicates that exposure to a single negative event in a laboratory setting does not negatively influence health [12].

In conclusion, this study adds to the field by demonstrating cross-sectional and prospective relationships between racial discrimination and both mental and physical health outcomes. With the rise in racial discrimination in the UK [6] in the aftermath of the Brexit vote [9] our findings highlight the need to reduce racial discrimination, not only to promote equity, but also to potentially benefit mental and physical health and reduce health inequalities.

Racial discrimination is a complex system that involves assigning ethnic groups differential value, which drives disparities in access to power, resources and opportunities [14, 15]. Due to its multi-faceted nature, occurring at both the structural and individual level multiple interventions will be required to tackle this pervasive determinant of health. Historically, raising awareness of racial discrimination has been necessary to promote activism to bring about legislative and social change to improve the position of ethnic minority groups. In terms of public health, there are calls to integrate research about racial discrimination and health into medical teaching in an attempt to tackle structural racism and to highlight the impact racial discrimination has on health [71, 72]. As well as strategies to reduce the pervasiveness of racial discrimination in institutional contexts, action through social media may have benefits for individual health too. The Black Lives Matter campaign is an example of a recent social media movement which has drawn attention to the issue of racial discrimination. There is some evidence that campaigns may provide a source of empowerment, particularly in a time where ethnic minority youth participation in traditional civic engagement activities are in decline [73]. Evidence suggests the Twitter conversation remained Black-led [73] and that the majority of the 40 million plus tweets were supportive of the movement [73, 74]. However, whether social media campaigns positively [73] or negatively impact minority health [75] remains the subject of debate. Further, it should be acknowledged that interventions to educate and raise awareness do not tackle the structural macro-level forces that shape the position of ethnic minorities in society. Although, more challenging to address, work is required to identify socio-political processes that generate racial discrimination so attempts can be made to mitigate its effects. Research into the pathways underlying the link between racial discrimination and health are required to develop policy and to target interventions in this field.