Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Racism, social resources and mental health for Aboriginal people living in Adelaide

Racism, social resources and mental health for Aboriginal people living in Adelaide R acial discrimination (hereafter described as racism) is the process by which a member or members of a socially defined racial group is, or are, treated unfairly because of membership of that group. There is a growing international body of evidence that experience of racism is detrimental to mental health. However, despite enormous health inequities between Indigenous and non‐Indigenous Australians (Indigenous Australian males and females have substantially lower life expectancy, 12 and 10 years respectively, and significantly higher levels of psychological distress ) there is relatively little known about how racism contributes to these health differentials. In particular, there are very few published studies that examine the experience of racism by Aboriginal and Torres Strait Islander people (hereafter referred to as Aboriginal people) and mental health in an urban setting, which is where the majority of Aboriginal people live. The research literature also indicates that social resources such as social connections and social support may buffer individuals from the negative effects of experiences such as racism. Whether this is the case for Aboriginal people has not been researched. In this paper, we examine the association between reported experiences of interpersonal racism (in interactions between individuals either within their institutional roles or as private individuals) and mental health for Aboriginal people living in Adelaide, and explore whether social resources help buffer individuals’ mental health from the detrimental effects of racism. Research has linked experience of racism with a range of poorer mental health and wellbeing outcomes including depression, anxiety and obsession‐compulsion and life satisfaction. There is evidence that there is a dose‐response relationship whereby the more frequently someone experiences racism the worse the health consequences. There is also evidence that experiences of racism through the life course can have lasting health implications. There are a number of ways through which experiences of racism might lead to negative health outcomes. These include stress and negative emotional reactions, through health‐related behaviours where those experiencing discrimination turn to drug and alcohol use, and the contribution of racism to creating unequal access to resources required for health (for example, employment, housing, health care). Experiences of discrimination may also lead to internalised negative evaluations and unfavourable self‐evaluations that affect psychological well‐being. There has been a call for research to consider ways that the relationship between racism and health might be ameliorated, and there is evidence that the ways that people cope with their experiences of racism can affect its health impacts. One of these ‘coping mechanisms’ that has been considered is that of social connections and the social support they potentially give rise to. The link between social connections and social support, and positive mental health outcomes is well established for a range of population groups. It is argued that social participation may distract individuals and provide a range of alternative positive experiences, and that social support can help the individual see discrimination in a collective context, can activate racial identity and can provide role models for coping with discrimination. However, a recent review of this literature found inconsistent findings, depending on measures used (including whether measures were of the availability of resources versus people reporting actually using resources in response to racism), barriers to using support, and evidence of both positive and negative outcomes from talking to others. Whether social resources might ameliorate the health affects of racism has not been explored for Australian Aboriginal people. Aboriginal people are the most disadvantaged group in Australia in terms of unemployment, income, housing and incarceration indicators. Aboriginal people also suffer a high burden of ill health and mortality and have higher rates of risk factors such as smoking and substance misuse. The extent of racism reported by Aboriginal people has varied, with single item measures indicating 16–40% experiencing racial discrimination and multiple item measures ranging from 58% to 79%. There is limited research into the mental health effects of racism experienced by Aboriginal people. One study by Larson and colleagues in a rural setting found a relationship between experiencing racism that caused ‘emotional upset’ and worse mental health outcomes. We found only one study that examined the relationship between racism and mental health for Aboriginal people living in urban areas, which found an association between experiencing racism, and depression and poorer mental health. This study addresses two research questions: First, is the experience of racism associated with poor mental health outcomes for Australian Aboriginal people? Second, do social resources help buffer negative mental health effects? Methods Sample and recruitment We conducted face‐to‐face interviews with 153 people across five Adelaide metropolitan Indigenous Locality Boundaries in 2006/07. This study was overseen by a predominantly Aboriginal Advisory Committee, and conducted in accordance with principles and practices determined by this Committee to be culturally safe. This included ensuring that Aboriginal researchers had input into the instrument's design, conducted all interviews, and assisted with analyses and interpretation of data. Findings were also reviewed by Aboriginal researchers and/or Aboriginal community members prior to publication or release. There is no complete list of Aboriginal and Torres Strait Islander people within suburban Adelaide and response rates for Aboriginal people to data collection methods such as telephone and postal survey has been found to be poor. Supported by the Advisory Committee, we employed face‐to‐face interviews and recruited people through a range of ‘snowball’ techniques including: through the Advisory Committee and Aboriginal services and organisations to identify potential participants; posters and flyers in community organisations; pamphlets at local Aboriginal agencies; residential letter box drops; and asking interviewees to identify other people who may be interested. All participants recruited were Aboriginal – one individual was both Aboriginal and Torres Strait Islander. Structured and semi‐structured interview questions were asked about neighbourhood life, social activities and participation, experiences of racism and health, and a range of demographic characteristics. Interviews took place in people's homes and, where this was not convenient, an alternative venue such as a community health centre or a local park was arranged. All interviews were conducted by Aboriginal researchers. Interviews ranged from three‐quarters of an hour to two hours and included a mix of closed and open‐ended questions. Only closed questions are reported on here. Ethics approval was granted from the Aboriginal Health Council of South Australia's Aboriginal Health Research and Ethics Committee, and the Flinders University Social and Behavioural Research Ethics Committee. Variables included in the analysis Demographic variables A number of variables were included in the model to control for demographic characteristics. ‘Gender’ was included in the models. ‘Age’ was measured in years. ‘Education’ was measured as up to secondary school (reference category) versus further education. Perceived ‘financial situation’ was measured as finding it very difficult/difficult (reference category) versus getting by/comfortable/very comfortable. Health behaviours Two variables were included in the model to control for health behaviours. ‘Smoking status’ was coded as smoker and non‐smoker. ‘Alcohol intake’ was measured in terms of ‘safe’ and ‘unsafe’ drinking levels. ‘Unsafe’ drinking (the reference category) was >14 alcoholic beverages a week for women and >28 for men (following National Health and Medical Research Council guidelines at the time). Experience of racism Substantial debate exists over how to measure the experience of racism, and it is likely that methodological variations are responsible for differing rates of discrimination found. Respondents were asked how often they were treated unfairly in 10 settings because they were Aboriginal. Questions were drawn from the Measure of Indigenous Experience of Racism, though one setting in the MIRE ‘at home, by neighbours or at someone else's home’ was separated into ‘at home or at someone else's house’ and ‘neighbours or people in the neighbourhood’. The MIRE was chosen as it has been specifically validated with Australian Aboriginal people. The response options were never, hardly ever, sometimes, often, very often or this does not apply to me. The category ‘this does not apply to me’ was coded with never and also as missing and the analysis undertaken with each. The results of the analyses were identical so to reduce the missing data ‘does not apply’ was coded as ‘never’. This study used the MIRE to measure racism in two ways. First, the average regularity of racism reported across the 10 settings, and, second, whether an individual had experienced racism often in any one setting. The rationale for the second measure was that if racism was experienced often, but in only one setting, this could mean racism was still a regular feature of an individual's life. This information could be lost in an average measure, particularly where the settings may include those not used by individuals (e.g. work environments for those unemployed), as the MIRE does not assess the extent to which an individual is engaged in a setting. A third analysis (not reported here) used a measure of the number of settings that an individual had experienced racism at all, to account for the fact that even rare experiences of racism could be cumulative if they occurred in a number of settings. The findings were identical to the other two measures in that racism remained significantly associated with mental health after introducing the social resource variables. While the settings where racism might occur included in MIRE were designed to be mutually exclusive, its developers found a high level of internal consistency between items. ‘Average reported racism’ was the average regularity of racism reported across the 10 settings (Chronbach alpha=0.81)). ‘Regular racism in at least one setting’ was coded as 1 for those who reported experiencing racism often/very often in at least one setting, and zero for those who did not. Social connections and support Three social connection and support variables were included in the analysis and these were drawn from our previous research in this area. The first was ‘socialising’ which refers to regularity of face‐to‐face contact with friends and family. Response choices were ‘every day’, ‘most days’, ‘once or twice a week’, ‘once or twice a month’, ‘every couple of months’ and ‘less often’. Reflecting the spread of the data these were recoded as at least once a week and less than once a week (reference category). The second was ‘group membership’ referring to whether respondents had participated in a community group in the last 12 months. The third was ‘emotional support’ referring to number of people (none, 1–2, 3–4, 5 or more) respondents could call on to talk about a personal problem or help making an important decision. This was coded as ‘none’ and ‘at least one person’. A fourth variable was also included which was drawn from the MIRE (with the same response format). The question asked how often after experiencing racism participants ‘talk to other people like family or friends about it’ (it also included whether they wrote, draw, sang or painted’ about it, which is discussed in the study limitations). This variable (‘talking and expressing self about racism’) was coded as ‘at least sometimes’ (1) if participants used this method at least sometimes, and ‘less than sometimes’ (0) if they did not. Health ‘Mental health’ was measured using the SF‐12 self‐report health measure (scored according to Ware and colleagues ) which allows the calculation of a mental health summary score ranging from 0 to 100 with higher scores indicating better mental health. Statistical analysis Multiple linear regression using SPSS Version 17 was employed with two final models developed using the two different measures of racism. A staged modelling process was conducted with the first stage including demographic and health behaviour variables. The second stage for each of the models included one of the two racism variables. In the third stage, the social connection and support variables were included. In the final stage, the talking or expressing self after racism variable was entered. In each case, assumptions of multiple linear regression were met including tolerance indicators of 0.8 and above for variables in all models, and appropriate Normal P‐P Plots and Scatterplots. Models were significant in each step. Missing data was dealt with pairwise. Results Study participants Table 1 provides the demographic and health behaviour information for the study participants, including missing data. There were more women than men, and the average age of respondents was 41. The vast majority did not have a tertiary degree or diploma and few were financially comfortable. Most were non‐drinkers or drank moderately, and more than two‐thirds smoked. 1 Study participant characteristics. n (%)*/ mean (SD) Gender Male 60 (39) Female 93 (61) Age 41.00 (13.16) Missing 3 (2) Education Up to secondary school 91 (60) Trade/business/university degree or higher 60 (39) Missing 2 (1) Financial management Getting by/comfortable 111 (73) Very difficult/difficult 41 (27) Missing 1 (1) Alcohol intake ‘Safe’ 134 (88) ‘Unsafe’ 18 (12) Missing 1 (1) Tobacco intake Smoker 103 (67) Non‐smoker 49 (32) Missing 1 (1) * Percentages sometimes do not add up to 100 due to rounding In terms of the social resource variables the majority of people (n=134, 88%) socialised at least once a week, with the remainder socialising less regularly. The majority were also in a community group (n=112, 73%), and had a least one person for social support (n=141, 92%). In terms of responding to racism 76% (n=116) a least sometimes talked or expressed themselves in other ways after experiencing racism, with 22% doing this less often (and n=4 missing data). Reported experiences of racism A majority of people reported experiencing racism at least sometimes in most of the formal settings ( Table 2 ). Experience of racism was particularly regular within the justice and educational settings. In general, racism was less often experienced in informal settings than formal settings. However, in a number of settings (particularly service and general public settings) there were still substantial numbers of people reporting racist treatment. 2 Experience of racism by setting, N (%*). Setting Very often Often Sometimes Hardly Ever Never Doesn't apply Formal Justice 26 (17) 18 (12) 47 (31) 18 (12) 38 (25) 5 (3) # Educational 20 (13) 21 (14) 47 (31) 22 (14) 33 (22) 10 (7) Government 7 (5) 13 (9) 51 (33) 29 (19) 48 (31) 5 (3) Employment 12 (8) 10 (7) 44 (29) 23 (15) 42 (28) 22 (14) Health 4 (3) 16 (11) 42 (28) 26 (17) 64 (42) 1 (1) Informal Staff in service settings 12 (8) 23 (15) 62 (41) 24 (16) 29 (19) 3 (2) In general public settings 11 (7) 16 (11) 56 (37) 33 (22) 34 (22) 3 (2) Sporting/recreational 9 (6) 9 (6) 42 (28) 32 (21) 45 (29) 16 (11) People in neighbourhood 9 (6) 10 (7) 40 (26) 32 (21) 56 (37) 5 (3) # At home/other house 3 (2) 13 (9) 42 (28) 30 (20) 60 (39) 5 (3) * Some percentages do not add up to 100 due to rounding. # One case missing. The mean score for average reported racism was 2.39 (range=1.00–4.80, SD=0.75) and averages were similar for men (2.45) and women (2.36). Sixty‐three per cent of respondents experienced racism often or very often in at least one setting, and again proportions reporting regular racism were similar for men (65%) and women (62%). Racism and mental health The median mental health score was 42.38 (SD 10.20), well below the average for the South Australian urban population of 55.20 for the same year (unpublished data from the SA Health Monitor Survey, South Australian Department of Health). The first stage for each of the models for mental health was the same with identical demographic and health behaviour variables entered ( Tables 3 and 4 ). Two variables were significant – financial situation and smoking, with mental health increasing with more positive assessments of financial situation and better for non‐smokers than for smokers. Using an unadjusted r‐square measure these variables accounted for 15% of the variance in mental health. 3 Regression models for mental health, average racism. 4 Regression models for mental health, racism often. In the second stage of the first model, average reported racism was included and was found to be negatively significantly associated with mental health – those experiencing more racism had worse mental health ( Table 3 ). Financial situation and smoking remained significant. The variance accounted for increased to 22%. In the third stage socialising, group membership and social support were entered ( Table 3 ). Average reported racism, smoking and financial situation remained significant, but none of the measures introduced in this stage were. The model accounted for 24% of variance in mental health. The introduction of the talking or expressing self variable in the final stage did not reduce the impact of racism or improve the model, nor was it significant in its own right ( Table 3 ). In the second model the second stage involved entering experience of racism often in at least one setting to the first stage model with demographic and health behaviour variables. As with average reported racism, experience of racism often in at least one setting was significantly negatively associated with mental health ( Table 4 ). Again, smoking and financial situation were significant. The model accounted for a similar amount of variance in mental health (21%). In the third stage, the social connection and support variables were again entered and were again not significant nor did they remove the association between experience of regular racism and mental health ( Table 4 ). Twenty‐two per cent of variance was accounted for by this model. Again, the talking or expressing self variable in the final stage did not reduce the significance of the racism variable, nor was it significant itself. Discussion This study found a high prevalence of experience of racism among participants, and that experience of racism was negatively associated with mental health outcomes. Social resources did not attenuate the negative affects of racism. Smoking and perceived financial situation were both also associated with mental health. These findings are discussed in more detail below in relation to the research questions. Is the experience of racism associated with poor mental health outcomes for Aboriginal people? Very high levels of racism were reported by participants in the study, with regular racism experienced in a wide range of settings, and almost two thirds reporting regular racism in at least one setting. These rates are at the upper end of other studies of racism experienced by Aboriginal people in Australia, and are comparable with the results of a study of Aboriginal people living in urban Darwin using the MIRE which found 70% of participants had experienced interpersonal racism in at least one of the settings. Mental health was lower for the study participants than the South Australian urban average. Supporting a growing body of international research, and Paradies’ and Larson's research in Australia, experience of racism was associated with poor mental health. This was found using both measures of racism. Do social resources help buffer negative mental health effects? The research literature suggests that social resources such as socialising, group participation and social support may buffer individuals from negative experiences such as racism. However, reflecting some of the inconsistencies in this literature, these resources did not remove the association between perceived experience of racism and mental health for Aboriginal participants. As noted, previous research in this area has used either the presence of networks and support or whether people sought help from their networks in response to racism. While we acknowledge the limitations of our measure of the extent to which people turned to networks in response to racism, it is noteworthy that even after the inclusion of all the social resource measures the relationship between racism and health remained. One possible explanation is that we found a high prevalence of racism so that most of our sample experienced racism at least sometimes in at least one setting (including within the home) and most commonly in more than one. This means that racism saturated the daily lives of participants in our study in such a way as these experiences could have been internalised in a way that made racism the ‘norm’. In this way it is possible that being socially connected would not offset the health impact of racism. It is also possible that the measures of social connections and support used, which were not specifically designed or validated in Aboriginal communities, did not accurately reflect the social resources available to our participants. Further research is required to validate existing measures or develop new measures that are appropriate for Aboriginal people. Other factors associated with mental health While not explicit research questions, smoking and perceived financial situation were also associated with mental health. Over a quarter of our sample reported that their financial situation was difficult or very difficult. Struggling to cope financially has been previously associated with poorer mental health status and reduced ability to deal with racism. In terms of the associations with smoking status, there is literature that illustrates how smoking can be a coping strategy in dealing with stressful situations and also that some may turn to drug and tobacco use in the face of racism. In this way there is likely to be a complex web of causality between tobacco use, experience of racism and mental health. Study strengths and limitations This study was guided by a largely Aboriginal Advisory committee and employed all Aboriginal interviewers. It was successful in recruiting urban Australian Aboriginal participants, many of whom had not previously been involved in research. There was very little missing data. However, it is acknowledged that it is a relatively small sample size and used non‐random sampling which limits the generalisability of the findings. Although tensions between ensuring scientific rigour and conducting research in a culturally safe manner are sometimes difficult to reconcile, findings from this research help to fill the gap in knowledge concerning the mental health effects of racism and how these can affect Aboriginal people living in an urban environment. Similarly, the research was cross‐sectional and as such, it could be that poor mental health may lead to higher perceptions of discrimination. However, evidence from 32 cohort studies to date indicates that, predominately, discrimination predicts illness rather than illness predicting reports of discrimination. The ‘talking or expressing self after racism’ variable did not exclusively assess whether social support networks had been used in the face of racism. However, this variable was part of the MIRE scale, and without it we would not have been able to make any assessment about whether social support had actually been used as a coping strategy. Conclusions Racism was found to be a significant determinant of mental health, with its effects not diminished by social connections and support. Our findings support a growing body of literature that suggests that racism has a strong impact on mental health. This finding is important to the current Australian government's policy goal of closing the gap between the health of Indigenous and non‐Indigenous Australians in a generation. While much of the current policy discussion concerns the impact of proximal risk factors on health, this study suggests the need to look further up the casual chain and consider those social factors that may help account for the worse health status and so lower life expectancy of Aboriginal people in Australia. Elsewhere we have outlined a policy agenda for further action on reducing racism. This agenda suggests a celebration of Aboriginal culture across Australian society and ensuring that the social and economic situation of Aboriginal people is greatly improved. It also suggests more systematic effort to directly combat racism in all social, institutional, sporting and work settings, and an approach to mental illness that appreciates and incorporates the harm done by historical policies that had such significant impacts on the lives of Aboriginal people. Without such a broad ranging policy assault on racism its insidious impacts on mental health are likely to continue. Acknowledgements The authors thank the research participants for their time and generosity in sharing their stories. They would also like to thank the project Advisory Committee for their guidance of the research and Michelle Ah Matt for administrative support. This project was funded by the Australian National Health and Medical Research Council (Project 324725) and the Australian Cooperative Research Centre for Aboriginal Health (Project SD 142). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Australian and New Zealand Journal of Public Health Wiley

Racism, social resources and mental health for Aboriginal people living in Adelaide

Download PDF
7 pages

Loading next page...
 
/lp/wiley/racism-social-resources-and-mental-health-for-aboriginal-people-living-rucX0QTVKi

References (47)

Publisher
Wiley
Copyright
© 2011 The Authors. ANZJPH © 2011 Public Health Association of Australia
ISSN
1326-0200
eISSN
1753-6405
DOI
10.1111/j.1753-6405.2011.00681.x
pmid
21627723
Publisher site
See Article on Publisher Site

Abstract

R acial discrimination (hereafter described as racism) is the process by which a member or members of a socially defined racial group is, or are, treated unfairly because of membership of that group. There is a growing international body of evidence that experience of racism is detrimental to mental health. However, despite enormous health inequities between Indigenous and non‐Indigenous Australians (Indigenous Australian males and females have substantially lower life expectancy, 12 and 10 years respectively, and significantly higher levels of psychological distress ) there is relatively little known about how racism contributes to these health differentials. In particular, there are very few published studies that examine the experience of racism by Aboriginal and Torres Strait Islander people (hereafter referred to as Aboriginal people) and mental health in an urban setting, which is where the majority of Aboriginal people live. The research literature also indicates that social resources such as social connections and social support may buffer individuals from the negative effects of experiences such as racism. Whether this is the case for Aboriginal people has not been researched. In this paper, we examine the association between reported experiences of interpersonal racism (in interactions between individuals either within their institutional roles or as private individuals) and mental health for Aboriginal people living in Adelaide, and explore whether social resources help buffer individuals’ mental health from the detrimental effects of racism. Research has linked experience of racism with a range of poorer mental health and wellbeing outcomes including depression, anxiety and obsession‐compulsion and life satisfaction. There is evidence that there is a dose‐response relationship whereby the more frequently someone experiences racism the worse the health consequences. There is also evidence that experiences of racism through the life course can have lasting health implications. There are a number of ways through which experiences of racism might lead to negative health outcomes. These include stress and negative emotional reactions, through health‐related behaviours where those experiencing discrimination turn to drug and alcohol use, and the contribution of racism to creating unequal access to resources required for health (for example, employment, housing, health care). Experiences of discrimination may also lead to internalised negative evaluations and unfavourable self‐evaluations that affect psychological well‐being. There has been a call for research to consider ways that the relationship between racism and health might be ameliorated, and there is evidence that the ways that people cope with their experiences of racism can affect its health impacts. One of these ‘coping mechanisms’ that has been considered is that of social connections and the social support they potentially give rise to. The link between social connections and social support, and positive mental health outcomes is well established for a range of population groups. It is argued that social participation may distract individuals and provide a range of alternative positive experiences, and that social support can help the individual see discrimination in a collective context, can activate racial identity and can provide role models for coping with discrimination. However, a recent review of this literature found inconsistent findings, depending on measures used (including whether measures were of the availability of resources versus people reporting actually using resources in response to racism), barriers to using support, and evidence of both positive and negative outcomes from talking to others. Whether social resources might ameliorate the health affects of racism has not been explored for Australian Aboriginal people. Aboriginal people are the most disadvantaged group in Australia in terms of unemployment, income, housing and incarceration indicators. Aboriginal people also suffer a high burden of ill health and mortality and have higher rates of risk factors such as smoking and substance misuse. The extent of racism reported by Aboriginal people has varied, with single item measures indicating 16–40% experiencing racial discrimination and multiple item measures ranging from 58% to 79%. There is limited research into the mental health effects of racism experienced by Aboriginal people. One study by Larson and colleagues in a rural setting found a relationship between experiencing racism that caused ‘emotional upset’ and worse mental health outcomes. We found only one study that examined the relationship between racism and mental health for Aboriginal people living in urban areas, which found an association between experiencing racism, and depression and poorer mental health. This study addresses two research questions: First, is the experience of racism associated with poor mental health outcomes for Australian Aboriginal people? Second, do social resources help buffer negative mental health effects? Methods Sample and recruitment We conducted face‐to‐face interviews with 153 people across five Adelaide metropolitan Indigenous Locality Boundaries in 2006/07. This study was overseen by a predominantly Aboriginal Advisory Committee, and conducted in accordance with principles and practices determined by this Committee to be culturally safe. This included ensuring that Aboriginal researchers had input into the instrument's design, conducted all interviews, and assisted with analyses and interpretation of data. Findings were also reviewed by Aboriginal researchers and/or Aboriginal community members prior to publication or release. There is no complete list of Aboriginal and Torres Strait Islander people within suburban Adelaide and response rates for Aboriginal people to data collection methods such as telephone and postal survey has been found to be poor. Supported by the Advisory Committee, we employed face‐to‐face interviews and recruited people through a range of ‘snowball’ techniques including: through the Advisory Committee and Aboriginal services and organisations to identify potential participants; posters and flyers in community organisations; pamphlets at local Aboriginal agencies; residential letter box drops; and asking interviewees to identify other people who may be interested. All participants recruited were Aboriginal – one individual was both Aboriginal and Torres Strait Islander. Structured and semi‐structured interview questions were asked about neighbourhood life, social activities and participation, experiences of racism and health, and a range of demographic characteristics. Interviews took place in people's homes and, where this was not convenient, an alternative venue such as a community health centre or a local park was arranged. All interviews were conducted by Aboriginal researchers. Interviews ranged from three‐quarters of an hour to two hours and included a mix of closed and open‐ended questions. Only closed questions are reported on here. Ethics approval was granted from the Aboriginal Health Council of South Australia's Aboriginal Health Research and Ethics Committee, and the Flinders University Social and Behavioural Research Ethics Committee. Variables included in the analysis Demographic variables A number of variables were included in the model to control for demographic characteristics. ‘Gender’ was included in the models. ‘Age’ was measured in years. ‘Education’ was measured as up to secondary school (reference category) versus further education. Perceived ‘financial situation’ was measured as finding it very difficult/difficult (reference category) versus getting by/comfortable/very comfortable. Health behaviours Two variables were included in the model to control for health behaviours. ‘Smoking status’ was coded as smoker and non‐smoker. ‘Alcohol intake’ was measured in terms of ‘safe’ and ‘unsafe’ drinking levels. ‘Unsafe’ drinking (the reference category) was >14 alcoholic beverages a week for women and >28 for men (following National Health and Medical Research Council guidelines at the time). Experience of racism Substantial debate exists over how to measure the experience of racism, and it is likely that methodological variations are responsible for differing rates of discrimination found. Respondents were asked how often they were treated unfairly in 10 settings because they were Aboriginal. Questions were drawn from the Measure of Indigenous Experience of Racism, though one setting in the MIRE ‘at home, by neighbours or at someone else's home’ was separated into ‘at home or at someone else's house’ and ‘neighbours or people in the neighbourhood’. The MIRE was chosen as it has been specifically validated with Australian Aboriginal people. The response options were never, hardly ever, sometimes, often, very often or this does not apply to me. The category ‘this does not apply to me’ was coded with never and also as missing and the analysis undertaken with each. The results of the analyses were identical so to reduce the missing data ‘does not apply’ was coded as ‘never’. This study used the MIRE to measure racism in two ways. First, the average regularity of racism reported across the 10 settings, and, second, whether an individual had experienced racism often in any one setting. The rationale for the second measure was that if racism was experienced often, but in only one setting, this could mean racism was still a regular feature of an individual's life. This information could be lost in an average measure, particularly where the settings may include those not used by individuals (e.g. work environments for those unemployed), as the MIRE does not assess the extent to which an individual is engaged in a setting. A third analysis (not reported here) used a measure of the number of settings that an individual had experienced racism at all, to account for the fact that even rare experiences of racism could be cumulative if they occurred in a number of settings. The findings were identical to the other two measures in that racism remained significantly associated with mental health after introducing the social resource variables. While the settings where racism might occur included in MIRE were designed to be mutually exclusive, its developers found a high level of internal consistency between items. ‘Average reported racism’ was the average regularity of racism reported across the 10 settings (Chronbach alpha=0.81)). ‘Regular racism in at least one setting’ was coded as 1 for those who reported experiencing racism often/very often in at least one setting, and zero for those who did not. Social connections and support Three social connection and support variables were included in the analysis and these were drawn from our previous research in this area. The first was ‘socialising’ which refers to regularity of face‐to‐face contact with friends and family. Response choices were ‘every day’, ‘most days’, ‘once or twice a week’, ‘once or twice a month’, ‘every couple of months’ and ‘less often’. Reflecting the spread of the data these were recoded as at least once a week and less than once a week (reference category). The second was ‘group membership’ referring to whether respondents had participated in a community group in the last 12 months. The third was ‘emotional support’ referring to number of people (none, 1–2, 3–4, 5 or more) respondents could call on to talk about a personal problem or help making an important decision. This was coded as ‘none’ and ‘at least one person’. A fourth variable was also included which was drawn from the MIRE (with the same response format). The question asked how often after experiencing racism participants ‘talk to other people like family or friends about it’ (it also included whether they wrote, draw, sang or painted’ about it, which is discussed in the study limitations). This variable (‘talking and expressing self about racism’) was coded as ‘at least sometimes’ (1) if participants used this method at least sometimes, and ‘less than sometimes’ (0) if they did not. Health ‘Mental health’ was measured using the SF‐12 self‐report health measure (scored according to Ware and colleagues ) which allows the calculation of a mental health summary score ranging from 0 to 100 with higher scores indicating better mental health. Statistical analysis Multiple linear regression using SPSS Version 17 was employed with two final models developed using the two different measures of racism. A staged modelling process was conducted with the first stage including demographic and health behaviour variables. The second stage for each of the models included one of the two racism variables. In the third stage, the social connection and support variables were included. In the final stage, the talking or expressing self after racism variable was entered. In each case, assumptions of multiple linear regression were met including tolerance indicators of 0.8 and above for variables in all models, and appropriate Normal P‐P Plots and Scatterplots. Models were significant in each step. Missing data was dealt with pairwise. Results Study participants Table 1 provides the demographic and health behaviour information for the study participants, including missing data. There were more women than men, and the average age of respondents was 41. The vast majority did not have a tertiary degree or diploma and few were financially comfortable. Most were non‐drinkers or drank moderately, and more than two‐thirds smoked. 1 Study participant characteristics. n (%)*/ mean (SD) Gender Male 60 (39) Female 93 (61) Age 41.00 (13.16) Missing 3 (2) Education Up to secondary school 91 (60) Trade/business/university degree or higher 60 (39) Missing 2 (1) Financial management Getting by/comfortable 111 (73) Very difficult/difficult 41 (27) Missing 1 (1) Alcohol intake ‘Safe’ 134 (88) ‘Unsafe’ 18 (12) Missing 1 (1) Tobacco intake Smoker 103 (67) Non‐smoker 49 (32) Missing 1 (1) * Percentages sometimes do not add up to 100 due to rounding In terms of the social resource variables the majority of people (n=134, 88%) socialised at least once a week, with the remainder socialising less regularly. The majority were also in a community group (n=112, 73%), and had a least one person for social support (n=141, 92%). In terms of responding to racism 76% (n=116) a least sometimes talked or expressed themselves in other ways after experiencing racism, with 22% doing this less often (and n=4 missing data). Reported experiences of racism A majority of people reported experiencing racism at least sometimes in most of the formal settings ( Table 2 ). Experience of racism was particularly regular within the justice and educational settings. In general, racism was less often experienced in informal settings than formal settings. However, in a number of settings (particularly service and general public settings) there were still substantial numbers of people reporting racist treatment. 2 Experience of racism by setting, N (%*). Setting Very often Often Sometimes Hardly Ever Never Doesn't apply Formal Justice 26 (17) 18 (12) 47 (31) 18 (12) 38 (25) 5 (3) # Educational 20 (13) 21 (14) 47 (31) 22 (14) 33 (22) 10 (7) Government 7 (5) 13 (9) 51 (33) 29 (19) 48 (31) 5 (3) Employment 12 (8) 10 (7) 44 (29) 23 (15) 42 (28) 22 (14) Health 4 (3) 16 (11) 42 (28) 26 (17) 64 (42) 1 (1) Informal Staff in service settings 12 (8) 23 (15) 62 (41) 24 (16) 29 (19) 3 (2) In general public settings 11 (7) 16 (11) 56 (37) 33 (22) 34 (22) 3 (2) Sporting/recreational 9 (6) 9 (6) 42 (28) 32 (21) 45 (29) 16 (11) People in neighbourhood 9 (6) 10 (7) 40 (26) 32 (21) 56 (37) 5 (3) # At home/other house 3 (2) 13 (9) 42 (28) 30 (20) 60 (39) 5 (3) * Some percentages do not add up to 100 due to rounding. # One case missing. The mean score for average reported racism was 2.39 (range=1.00–4.80, SD=0.75) and averages were similar for men (2.45) and women (2.36). Sixty‐three per cent of respondents experienced racism often or very often in at least one setting, and again proportions reporting regular racism were similar for men (65%) and women (62%). Racism and mental health The median mental health score was 42.38 (SD 10.20), well below the average for the South Australian urban population of 55.20 for the same year (unpublished data from the SA Health Monitor Survey, South Australian Department of Health). The first stage for each of the models for mental health was the same with identical demographic and health behaviour variables entered ( Tables 3 and 4 ). Two variables were significant – financial situation and smoking, with mental health increasing with more positive assessments of financial situation and better for non‐smokers than for smokers. Using an unadjusted r‐square measure these variables accounted for 15% of the variance in mental health. 3 Regression models for mental health, average racism. 4 Regression models for mental health, racism often. In the second stage of the first model, average reported racism was included and was found to be negatively significantly associated with mental health – those experiencing more racism had worse mental health ( Table 3 ). Financial situation and smoking remained significant. The variance accounted for increased to 22%. In the third stage socialising, group membership and social support were entered ( Table 3 ). Average reported racism, smoking and financial situation remained significant, but none of the measures introduced in this stage were. The model accounted for 24% of variance in mental health. The introduction of the talking or expressing self variable in the final stage did not reduce the impact of racism or improve the model, nor was it significant in its own right ( Table 3 ). In the second model the second stage involved entering experience of racism often in at least one setting to the first stage model with demographic and health behaviour variables. As with average reported racism, experience of racism often in at least one setting was significantly negatively associated with mental health ( Table 4 ). Again, smoking and financial situation were significant. The model accounted for a similar amount of variance in mental health (21%). In the third stage, the social connection and support variables were again entered and were again not significant nor did they remove the association between experience of regular racism and mental health ( Table 4 ). Twenty‐two per cent of variance was accounted for by this model. Again, the talking or expressing self variable in the final stage did not reduce the significance of the racism variable, nor was it significant itself. Discussion This study found a high prevalence of experience of racism among participants, and that experience of racism was negatively associated with mental health outcomes. Social resources did not attenuate the negative affects of racism. Smoking and perceived financial situation were both also associated with mental health. These findings are discussed in more detail below in relation to the research questions. Is the experience of racism associated with poor mental health outcomes for Aboriginal people? Very high levels of racism were reported by participants in the study, with regular racism experienced in a wide range of settings, and almost two thirds reporting regular racism in at least one setting. These rates are at the upper end of other studies of racism experienced by Aboriginal people in Australia, and are comparable with the results of a study of Aboriginal people living in urban Darwin using the MIRE which found 70% of participants had experienced interpersonal racism in at least one of the settings. Mental health was lower for the study participants than the South Australian urban average. Supporting a growing body of international research, and Paradies’ and Larson's research in Australia, experience of racism was associated with poor mental health. This was found using both measures of racism. Do social resources help buffer negative mental health effects? The research literature suggests that social resources such as socialising, group participation and social support may buffer individuals from negative experiences such as racism. However, reflecting some of the inconsistencies in this literature, these resources did not remove the association between perceived experience of racism and mental health for Aboriginal participants. As noted, previous research in this area has used either the presence of networks and support or whether people sought help from their networks in response to racism. While we acknowledge the limitations of our measure of the extent to which people turned to networks in response to racism, it is noteworthy that even after the inclusion of all the social resource measures the relationship between racism and health remained. One possible explanation is that we found a high prevalence of racism so that most of our sample experienced racism at least sometimes in at least one setting (including within the home) and most commonly in more than one. This means that racism saturated the daily lives of participants in our study in such a way as these experiences could have been internalised in a way that made racism the ‘norm’. In this way it is possible that being socially connected would not offset the health impact of racism. It is also possible that the measures of social connections and support used, which were not specifically designed or validated in Aboriginal communities, did not accurately reflect the social resources available to our participants. Further research is required to validate existing measures or develop new measures that are appropriate for Aboriginal people. Other factors associated with mental health While not explicit research questions, smoking and perceived financial situation were also associated with mental health. Over a quarter of our sample reported that their financial situation was difficult or very difficult. Struggling to cope financially has been previously associated with poorer mental health status and reduced ability to deal with racism. In terms of the associations with smoking status, there is literature that illustrates how smoking can be a coping strategy in dealing with stressful situations and also that some may turn to drug and tobacco use in the face of racism. In this way there is likely to be a complex web of causality between tobacco use, experience of racism and mental health. Study strengths and limitations This study was guided by a largely Aboriginal Advisory committee and employed all Aboriginal interviewers. It was successful in recruiting urban Australian Aboriginal participants, many of whom had not previously been involved in research. There was very little missing data. However, it is acknowledged that it is a relatively small sample size and used non‐random sampling which limits the generalisability of the findings. Although tensions between ensuring scientific rigour and conducting research in a culturally safe manner are sometimes difficult to reconcile, findings from this research help to fill the gap in knowledge concerning the mental health effects of racism and how these can affect Aboriginal people living in an urban environment. Similarly, the research was cross‐sectional and as such, it could be that poor mental health may lead to higher perceptions of discrimination. However, evidence from 32 cohort studies to date indicates that, predominately, discrimination predicts illness rather than illness predicting reports of discrimination. The ‘talking or expressing self after racism’ variable did not exclusively assess whether social support networks had been used in the face of racism. However, this variable was part of the MIRE scale, and without it we would not have been able to make any assessment about whether social support had actually been used as a coping strategy. Conclusions Racism was found to be a significant determinant of mental health, with its effects not diminished by social connections and support. Our findings support a growing body of literature that suggests that racism has a strong impact on mental health. This finding is important to the current Australian government's policy goal of closing the gap between the health of Indigenous and non‐Indigenous Australians in a generation. While much of the current policy discussion concerns the impact of proximal risk factors on health, this study suggests the need to look further up the casual chain and consider those social factors that may help account for the worse health status and so lower life expectancy of Aboriginal people in Australia. Elsewhere we have outlined a policy agenda for further action on reducing racism. This agenda suggests a celebration of Aboriginal culture across Australian society and ensuring that the social and economic situation of Aboriginal people is greatly improved. It also suggests more systematic effort to directly combat racism in all social, institutional, sporting and work settings, and an approach to mental illness that appreciates and incorporates the harm done by historical policies that had such significant impacts on the lives of Aboriginal people. Without such a broad ranging policy assault on racism its insidious impacts on mental health are likely to continue. Acknowledgements The authors thank the research participants for their time and generosity in sharing their stories. They would also like to thank the project Advisory Committee for their guidance of the research and Michelle Ah Matt for administrative support. This project was funded by the Australian National Health and Medical Research Council (Project 324725) and the Australian Cooperative Research Centre for Aboriginal Health (Project SD 142).

Journal

Australian and New Zealand Journal of Public Health – Wiley

Published: Jun 1, 2011

There are no references for this article.