Abstract

Background

Conventional frameworks of population health typically implicate a range of social, environmental and economic factors as key determinants of health [1]. However, there are historical and contemporary factors such as dispossession, discrimination and racism, that are uniquely experienced by minority and Indigenous populations and are known to impact on mental and physical health [2–4]. Racism, in particular, is being cited more commonly in the international literature as having serious adverse consequences for health [5–10], and is widely acknowledged as playing a crucial role in the formation of health disparities [11–14].

Racism takes many forms and can be expressed at interpersonal, systemic and internalised levels [15–18]. Accordingly, the pathways by which racism impacts on health are multifaceted and complex, for all populations [7, 19]. They include the direct impacts of racism on health (e.g. racially motivated physical assault), and the indirect effects that stem from reduced and unequal access to medical, social and economic resources required for good health, increased exposure to risk factors associated with ill health, such as withdrawal from health care and health promoting activities, and the stress of racism and its ill effects [20, 21]. In addition, poor health in childhood has been linked to the experience of vicarious racism—that is, the experiences of others, including parents, carers and family members [22]. Few studies have examined how vicarious racism impacts on child health and wellbeing, although the hypothesised mechanisms are centred around the stress associated with parental experiences of (direct and vicarious) racism and the resultant impact on parenting practices [23–25].

Despite a considerable increase in the number of studies on racism and health in recent years, relatively few have focused on younger age groups [5, 26, 27]. The existing research has primarily examined the impacts of direct forms of racism, and suggests that younger people are more vulnerable to racism’s harmful effects. These effects extend to negative outcomes in general health, life satisfaction, behaviour problems, health-related behaviours (alcohol, drug use and smoking), and healthcare utilization [22]. In addition, a small number of studies have shown that parent-experienced racial discrimination is associated with pregnancy and birth outcomes [28–32]. While this signifies that racism can influence development in the very early stages of life, few studies have examined the ongoing consequences of early exposure to this form of stress [33]. The sparse evidence from studies that draw on longitudinal data indicates that racism precedes poor health but provides few insights about the sensitivity of early exposure, the importance of duration of exposure, or the latency period between exposure to racism and onset of illness; as a result, the causal pathways through childhood and later life are not fully understood [5, 33].

Most studies of the impact of racial discrimination in childhood have been conducted in the United States among African American, Latino and Asian groups, with few studies in other minority populations and national contexts [22]. The quantitative empirical evidence for Indigenous populations is particularly sparse, despite the fact that racism is commonly cited as underpinning Indigenous people’s experiences of exclusion from aspects of social, political, and economic life [8, 34]. In Australia, for example, there have only been a handful of studies on populations of Indigenous (Aboriginal and Torres Strait Islander) children and these have focused on a restricted set of mental health, health risk behaviour and physical health indicators using primarily cross-sectional data for discrete population groups [35–38]. While the results in Australian Indigenous settings are broadly consistent with the US literature [22], much more research is needed in order to understand the role of racism as a determinant of Indigenous child health and the underpinning pathways.

The objective of this study is to use national longitudinal data on Indigenous children and multiple-indicators of racial discrimination to provide further insights on the links between persistent racism and health in early life. We use three dimensions of carer-reported racial discrimination that capture the direct experiences of children themselves and also vicarious exposure by their primary carer and family. Six waves of data from the Footprints in Time study (the Longitudinal Study of Indigenous Children; hereafter referred to as LSIC) are drawn upon to examine the associations between racial discrimination and a range of health outcomes that have been shown to be associated with racism in prior research on ethnic population groups. These outcomes include mental health [39], general health [37], sleep difficulties [40, 41] and obesity [42]. In addition, we examine injury as an outcome given the direct threats to health posed by racial violence and the indirect pathways from racism-related stress to mental impairment and unintentional injuries [7], and asthma, to further test the relationship found between psychosocial stress and the risk to poor child respiratory health [43]. In addition, we use population attributable risks (PARs) to further assess the significance of racism as a determinant of Australian Indigenous child health. PAR estimates consider both the level of exposure to a risk factor and its impact on a health outcome, and provide a measure of how much disease can be averted if the risk factor could be eliminated [44].

Methods

The LSIC is the first large-scale longitudinal survey to focus on the development of Australian Indigenous children [45]. The first wave of the survey was conducted in 2008, with subsequent waves conducted annually. The LSIC employed a non-random purposive sampling design, with study children recruited from 11 sites across Australia. These sites were chosen for practical and logistical reasons, and to represent the diversity of community environments where Indigenous children live [46]. Further details regarding the rationale, sampling, recruitment, and data collection have been reported elsewhere [47].

Results

Almost a quarter (23%) of Indigenous children in Wave 6 of the LSIC had a mental health problem (high risk of clinically significant emotional or behavioural difficulties). In addition, 25% had their health assessed as less than excellent or very good, 21% had difficulty sleeping in the last month, 21% were obese, 4% had suffered an injury in their lifetime, and 11% had asthma.

Two-fifths (40%) of primary carers, 45% of families and 14% of Indigenous children aged 5–10 years were reported to have experienced racial discrimination at some point in time (Fig. ​(Fig.1).1). The majority (69%) of primary carers that had experienced racial discrimination had a time-limited exposure (reported in one wave only), while 31% experienced persistent racial discrimination (reported in multiple waves). A similar pattern was observed for family members (60% time-limited and 40% persistent) and study children (72% and 28%, respectively).

Table ​Table11 includes odds ratios from logistic regression analyses between exposure to racism and socio-demographic factors, and highlights that there were generally only weak to moderate independent associations observed, with few statistically significant effects. The results, however, indicate that it was more common for older Indigenous children and primary carers to have experienced racial discrimination. In particular, children aged 9–10 years were over 4 times more likely (OR = 4.4; 95% CI: 2.85–6.79) to be treated badly or discriminated against because they were Indigenous than children aged 5–6 years. Racial discrimination appeared to be associated with geographic location: primary carers and families living in areas of moderate and high/extreme isolation were less likely to ever be exposed to racial discrimination, with those living in the most remote locations having the lowest odds (OR = 0.43, 95% CI: 0.24–0.78 for primary carers, and OR = 0.22, 95% CI: 0.12–0.40 for families) relative to children in metropolitan areas. In contrast, children in high/extreme isolation had elevated odds of exposure to racism at school (OR = 2.2; 95% CI: 1.08–4.44). The results for SES were mixed: the evidence for a relationship between reported racial discrimination and the educational attainment of the primary carer was inconclusive in that only a relatively weak and statistically insignificant odds ratio for primary carer and family experiences of racism was observed among those with a bachelor degree or higher. However, there was a stronger and statistically significant association for family financial strain, with an increasing likelihood of exposure to racial discrimination for children in families with poorer financial circumstances.

Table ​Table22 highlights a generally positive association between primary carer and child experiences of racial discrimination and poor physical and mental health for Indigenous children. This includes statistically significant effects (p p 1: Table S1) and obesity (some missing outcome data, Additional file 2: Table S2). Effect size estimates were consistent (differed only at the second decimal place) with analyses conducted on complete cases only so this method was not pursued.

The effect sizes for the associations between primary carer experiences of racial discrimination and child health tended to be considerably larger for those exposed to persistent racial discrimination when compared with those who experienced time-limited racial discrimination, suggesting a dose-response relationship between this form of vicarious racism and mental health, sleep difficulties and asthma. For example, children were 2.6 times more likely (95% CI: 1.47–4.76) to have asthma when their primary carer experienced persistent racial discrimination, with a weak and non-significant odds (OR = 1.2, 95% CI: 0.73–2.11) when the carer reported racial discrimination in one wave only.

The relationships between racial discrimination and health are only slightly attenuated by the inclusion of other known covariates in the statistical models. This suggests that the socio-demographic circumstances of Indigenous families have little impact on the relationship between racism exposure and the health of children.

Adjusting for all variables included in Model 2, we found the highest PARs for child experiences of racism. This accounted for 16.2% (95% CI: 11.1–21.2%) of the PAR for mental health status, 19.1% (95% CI: 13.8–24.4%) of the PAR for sleep difficulties, 8.2% (95% CI: 2.2–14.1%) of the PAR for obesity and 13.2% (95% CI: 5.6–20.7%) of the PAR for asthma. Carer experiences of persistent racial discrimination accounted for a portion of the PAR for mental health (5.8%; 95% CI: 0.3–11.2%), sleep difficulties (10.8%; 95% CI: 4.6–17.0%) and asthma (17.2%; 95% CI: 7.8–26.5%), while time-limited exposure accounted for 13.1% (95% CI: 2.2–23.9%) of the PAR for obesity.

Discussion

This study is one of the first to use longitudinal data to examine the effects of both direct and vicarious racial discrimination on a range of child health indicators. We have shown that racism is associated with mental health and aspects of the physical health of Indigenous children in Australia, with the most consistent results relating to mental health status, sleep difficulties, obesity and asthma. While prior studies suggest that racism poses the greatest risk to mental health [22], children experiencing racial discrimination in this study were, broadly speaking, at an equally elevated risk of having sleep difficulties, asthma and mental health problems.

In a novel finding, we highlight the negative impact that persistent vicarious racism has on child health, with reports of racial discrimination by primary carers at two earlier time points associated with a 1.7–2.6 times increased odds of poor mental health, sleep difficulties and asthma in children at age 5–10 years. This finding is consistent with a causal association between racism and health, and suggests that prolonged and/or more frequent exposure to vicarious racial discrimination that starts in the early lifecourse can impact on multiple domains of health in later life. This is likely to reflect processes of biological embedding, whereby frequent exposure to stressors in early life can impact on the development and functioning of vital organs, and contribute to systematic differences in health throughout life [58, 59]. These pathways are, plausibly, enmeshed with the experiences of racism among parents and the effect that this has on their mental wellbeing and ability to provide support that optimises child development [24, 25]. While relatively little is known about the pathways from vicarious racism to poor child health, an increased focus on longitudinal studies that begin at birth will provide a deeper understanding of the role of this form of racism in the pathways to disadvantage, including the persistent and striking inequalities in health between Indigenous and other populations.

The magnitude of elevated risk for mental health found in our study is consistent with existing studies in Australian Indigenous settings, and suggests that the direct experience of racial discrimination (assessed using single-item measures) is associated with a doubling of the risk of an Indigenous child or youth developing a mental health problem [36–38]. This includes a diverse set of mental health outcomes, such as anxiety, depression, suicide risk, low self-esteem, as well as overall mental health status. Priest et al. (2011) showed a higher risk among Indigenous youths (odds ratio of 3.35) which may reflect the older ages of study participants in the study (16–20 year-olds) or the fact that participants reported their own experiences of racism and mental health [38]. Similar studies among other minority groups also predominantly focus on adolescents and youth; one study of fifth-grade students (usually aged 10–11 years) in US metropolitan areas highlighted that children from non-White racial/ethnic groups who perceived discrimination were up to 3.9 times more likely to suffer depression and up to 2.4 times more likely to have specific externalizing disorders, with differences in relative risk between Black, Hispanic and other ethnic groups [60]. These results and the findings from our study confirm that, while children may be less likely to perceive discrimination in the period prior to adolescence, it nonetheless poses a considerable risk to the mental wellbeing of those who perceive it.

Our study highlights a relationship between racism and sleep difficulties, although the strength of this association appears to be greater than the results of previous studies in the US and Brazil [40, 41]. The findings of Bittencourt et al. (2009) and Huynh and Gillen-O’Neel (2013) suggest that both overt and subtle forms of discrimination have a moderate impact on the amount and quality of sleep among adolescents from ethnic minority populations. Our study supports the notion that stress associated with racism can also disrupt sleep in children, although further research is needed to quantify the impacts on quality and duration of sleep, particularly given the importance of sleep to brain development in childhood [61] and ongoing wellbeing [62].

Previous studies have also highlighted that parental stress exacerbates the causal risks of childhood asthma [43], although to our knowledge no studies have assessed racism specifically as a stressor. We have shown that both parental and children’s direct experiences of racism are strongly associated with asthma, with a 2–3 times greater risk for these racism exposures in multivariate models. Racism is likely to impact asthma in multiple ways—for example, the stressful effects of racism can impact on multiple bodily systems and increase a person’s susceptibility to the range of environmental exposures that are implicated in the development of asthma (e.g. air pollution and smoke exposure). In addition, the association may partly be an artefact of greater visibility of asthma responses in those experiencing racism, reflecting the connection between stress events and the onset of wheeze and asthma attacks in children [63, 64].

The pervasive health inequalities between Indigenous and non-Indigenous people in Australia demands a better understanding of the aetiology of poor health outcomes in Indigenous populations. While the current scientific literature implicates racism and its ill effects in the complex pathways to physical and mental health problems [5], there has been little scrutiny of the saliency of these factors in Indigenous children. This study adds to the sparse studies on Indigenous children, and further confirms that Indigenous children exposed to racism tend to be at a considerably greater risk of poor health [35–38]. In addition, the prevalence of racism among participants in this study is broadly consistent with the existing estimates in comparable studies of Indigenous child populations. We found that 40% of primary carers, 45% of families and 14% of Indigenous children aged 5–10 years had experienced racism at some point in time, with previous studies using single-item measures typically in the range of 15–40% [22]. However, the prevalence of direct experiences of racism among 12–17 year olds in Victoria (35%) and Western Australia (22%) are higher than what we have reported in this study (14%). This is likely to reflect differences in participant age range and respondent awareness. Prevalence rates are generally higher among adolescents than children, reflecting that adolescence is a time when children widen their spheres of influence and gain a greater awareness of discrimination [25]. Further, previous studies have asked adolescents about their experiences of racism directly whereas the LSIC asked the primary carer about their (perhaps limited) perceptions of their child’s experiences.

PAR estimates were of a low to moderate size, with notable values shown for mental health, sleep difficulties, obesity and asthma with respect to both child and primary carer experiences of racial discrimination. PARs can highlight modifiable risk factors and estimate how much disease can be averted with their elimination [44], although few studies have examined the PAR of racism, to date. The results here suggest that eradicating racism is likely to reduce the burden associated with some important health problems in childhood and therefore account for a portion of the health inequality between Indigenous and other Australians. The PAR estimates here do not fully capture the impact of racism on health inequality, given that the effect of racism on other determinants of health was not examined in this study due to limitations in the available data.

Overall, the study results suggest that children who are exposed to racial discrimination (and/or perceive it) have poorer physical and mental health, and that racism is an important modifiable risk factor that needs to be targeted to reduce disease burden. However, the impact of racism needs to be explored in more detail in order to more fully understand what types of racism have the most impact on Indigenous health. We need to know more about the prevalence of systemic, interpersonal and internalised racism for Indigenous Australians and how this varies by place and socioeconomic status, among other factors. Instruments need to be developed that measure perceptions of racism in children and youth that take into account the unique contexts and developmental levels of children. And, critically, future research will need to focus on pathways in order to deepen our understanding of detrimental effects of racism on health. All of this will help us design more effective interventions to combat racism. This, in turn, will help us to increase the pace of change toward health equality in Australia.

Conclusion

Direct and persistent vicarious racial discrimination are detrimental to the physical and mental health of Indigenous children in Australia. Racism is thus a critical social determinant of health in Indigenous and minority children. Prolonged and more frequent exposure to racial discrimination that starts in the early lifecourse can impact on multiple domains of health in later life. Tackling and reducing racism should be an integral part of policy and intervention aimed at improving the health of Australian Indigenous children and thereby reducing health disparities between Indigenous and non-Indigenous children.

Additional files

Acknowledgements

Abbreviations

Authors’ contributions

CCJS and JL conceived and designed the study. CCJS performed the primary statistical analyses and wrote the initial drafts of the manuscript. MNC conducted the imputation analyses. All authors have subsequently read, revised, and approved the manuscript.

Notes

Footnotes

Contributor Information

Carrington C.J. Shepherd, Email: ua.gro.sdiknohtelet@drehpehS.notgnirraC.

Jianghong Li, Email: ue.bzw@il.gnohgnaij.

Matthew N. Cooper, Email: ua.gro.sdiknohtelet@repooC.ttaM.

Katrina D. Hopkins, Email: ua.gro.sdiknohtelet@snikpoH.anirtaK.

Brad M. Farrant, Email: ua.gro.sdiknohtelet@tnarraf.darb.

References