- Study protocol
- Open access
- Published: 28 March 2019
![](http://serviteca.online/777/templates/cheerup2/res/banner1.gif)
The impact of racism on the future health of adults: protocol for a prospective cohort study
- James Stanley ORCID: orcid.org/0000-0002-8572-1047 1 ,
- Ricci Harris 2 ,
- Donna Cormack 2 ,
- Andrew Waa 2 &
- Richard Edwards 1
BMC Public Health volume 19 , Article number: 346 ( 2019 ) Cite this article
237k Accesses
23 Citations
266 Altmetric
Metrics details
Racial discrimination is recognised as a key social determinant of health and driver of racial/ethnic health inequities. Studies have shown that people exposed to racism have poorer health outcomes (particularly for mental health), alongside both reduced access to health care and poorer patient experiences. Most of these studies have used cross-sectional designs: this prospective cohort study (drawing on critical approaches to health research) should provide substantially stronger causal evidence regarding the impact of racism on subsequent health and health care outcomes.
Participants are adults aged 15+ sampled from 2016/17 New Zealand Health Survey (NZHS) participants, sampled based on exposure to racism (ever exposed or never exposed, using five NZHS questions) and stratified by ethnic group (Māori, Pacific, Asian, European and Other). Target sample size is 1680 participants (half exposed, half unexposed) with follow-up survey timed for 12–24 months after baseline NZHS interview. All exposed participants are invited to participate, with unexposed participants selected using propensity score matching (propensity scores for exposure to racism, based on several major confounders). Respondents receive an initial invitation letter with choice of paper or web-based questionnaire. Those invitees not responding following reminders are contacted for computer-assisted telephone interview (CATI).
A brief questionnaire was developed covering current health status (mental and physical health measures) and recent health-service utilisation (unmet need and experiences with healthcare measures). Analysis will compare outcomes between those exposed and unexposed to racism, using regression models and inverse probability of treatment weights (IPTW) to account for the propensity score sampling process.
This study will add robust evidence on the causal links between experience of racism and subsequent health. The use of the NZHS as a baseline for a prospective study allows for the use of propensity score methods during the sampling phase as a novel approach to recruiting participants from the NZHS. This method allows for management of confounding at the sampling stage, while also reducing the need and cost of following up with all NZHS participants.
Peer Review reports
Differential access to the social determinants of health both creates and maintains unjust and avoidable health inequities [ 1 ]. In New Zealand, these inequities are largely patterned by ethnicity, particularly for Māori (the indigenous peoples) and Pacific peoples, and intertwined with ethnic distributions of socioeconomic status [ 2 , 3 ]. In models of health, racism is recognised as a key social determinant that underpins systemic ethnic health and social inequities, as is evident in New Zealand and elsewhere [ 4 , 5 ].
Racism can be understood as an organised system based on the categorisation and ranking of racial/ethnic groups into social hierarchies whereby ethnic groups are assigned differential value and have differential access to power, opportunities and resources, resulting in disadvantage for some groups and advantage for others [ 4 , 6 ]. Historical power relationships underpin systems of racism [ 7 ], which in New Zealand relates specifically to our colonial history and ongoing colonial processes [ 8 ].
Racism can be expressed at structural and individual levels, with several taxonomies describing different levels of racism. Institutionalised racism, for example, has been defined as, “the structures, policies, practices, and norms resulting in differential access to the goods, services, and opportunities of society by race[/ethnicity]” (p. 10) [ 6 ]. In contrast, personally-mediated racism has been defined as, “prejudice and discrimination, where prejudice is differential assumptions about the abilities, motives, and intents of others by ‘race[/ethnicity],’ and discrimination is differential actions towards others by ‘race[/ethnicity]’” (p. 10) [ 6 ].
The multifarious expressions of racism can affect health via several recognised direct and indirect pathways. Indirect pathways include differential access to societal resources and health determinants by race/ethnicity, as evidenced by long-standing ethnic inequities in income, education, employment and living standards in New Zealand, with subsequent impacts on living environments and exposure to risk and protective factors [ 4 , 6 , 9 , 10 ]. At the individual level, experience of racism can affect health directly through physical violence and stress pathways, with negative psychological and physiological impacts leading to subsequent mental and physical health consequences. In addition, racism influences healthcare via institutions and individual health providers, leading to ethnic inequities in access to and quality of care. For example, ethnic disparities in socioeconomic status can indirectly result in differential access to care, while health provider ethnic bias can influence the quality and outcomes of healthcare interactions [ 11 ].
There has been considerable recent growth in research supporting a direct link between experience of racism and health. A recent systematic review and meta-analysis summarised the evidence for direct links between self-reported personally-mediated racism and negative physical and mental health outcomes [ 12 ], with the strongest effect sizes demonstrated for mental health. Related work has also shown that experience of racial discrimination is associated with other adverse health outcomes and preclinical indicators of disease and health risk across various ethnic groups and countries, including in New Zealand [ 9 , 13 , 14 , 15 ]. Experience of racism has also been linked to a range of negative health care-related measures [ 16 ].
However, most studies have used cross-sectional designs: very few of the articles in a recent systematic review [ 12 ] used prospective or longitudinal designs ( n = 30, 9% of total, including multiple articles from some studies), limiting our ability to draw strong causal conclusions as the direction of causality cannot be determined when racism exposure and health outcomes are measured at the same time. Additionally, cross-sectional studies may give biased estimates of the magnitude of association between experience of racism and health: for example, bias may occur if experience of ill health (outcome) increases reporting or perception of racism (exposure) [ 12 ]. This is suggested by meta-analyses where effect sizes for the association between racism and mental health were larger for cross-sectional compared to longitudinal studies [ 12 ]. Longitudinal research on the effects of racism has been particularly limited with respect to physical health outcomes and measures of healthcare access and quality [ 12 , 16 ]. Finally, existing prospective studies have largely been restricted to quite specific groups (e.g. adolescents, females, particular ethnic groups), with a limited number of studies undertaken at a national population level and few with sufficient data to explore the impact of racism on the health of Indigenous populations [ 12 ].
In New Zealand, reported experience of racism is substantially higher among Māori, Asian and Pacific ethnic groupings compared to European [ 3 , 17 ]. In our own research, we have examined cross-sectional links between reported experience of racism and various measures of adult health in New Zealand using data from the New Zealand Health Survey (NZHS), an annual national survey by the Ministry of Health including ~ 13,000 adults per annum [ 2 , 18 , 19 ]. In these studies [ 17 , 20 , 21 , 22 ] we have shown that both individual experience of racism (e.g. personal attacks or unfair treatment) and markers of structural racism (deprivation, other socioeconomic indicators) are independently associated with poor health (mental health, physical health, cardiovascular disease), health risks (smoking, hazardous alcohol consumption) and healthcare experience and use (screening, unmet need and negative patient experiences). Other New Zealand researchers have reported similar findings including studies among older Māori [ 23 ], adolescents [ 24 ], and for maternal and child health outcomes [ 25 ]. However, evidence from New Zealand prospective studies is still limited. The NZ Attitudes and Values study showed that, among Māori, experience of racism was negatively linked to subsequent wellbeing [ 26 ], and the Growing Up in New Zealand study reported that maternal experience of racism (measured antenatally) was linked to a higher risk of postnatal depression among Māori, Pacific and Asian women [ 27 ].
While empirical evidence of the links between racism and health is growing in New Zealand, it remains limited in several areas. There is consistent evidence from cross-sectional studies for the hypothesis that racism is associated with poorer health and health care. This study seeks to build on existing research to provide more robust causal evidence using a prospective design that helps to rule out reverse causality, in order to inform policy and healthcare interventions.
Theoretical and conceptual approaches
Addressing racism as a health determinant is intrinsically linked to addressing ethnic health inequities. In New Zealand, Māori health is of special relevance given Māori rights under the Treaty of Waitangi [ 28 ] and the United Nations Declaration on the Rights of Indigenous People [ 29 ], and in recognition of the inequities for Māori across most major health indicators [ 28 ]. We recognise the direct significance of this project to Māori and understand racism in its broader sense as underpinning our colonial history with ongoing contemporary manifestations and effects [ 8 ]. As such, our work is informed by critical approaches to health research that are explicitly concerned with understanding inequity and transforming systems and structures to achieve the goal of health equity. This includes decolonising and transformative research principles [ 30 ] that influence our approach to the research question, data collection, analysis and interpretation of data, and translation of research findings. The team includes senior Māori researchers as well as advisors with experience in Māori health research and policy.
Aims and research questions
The overall aim is to examine the relationship between reported experience of racism and a range of subsequent health measures. The specific objectives are:
To determine whether experience of racism leads to poorer mental health and/or physical health.
To determine the impact of racism on subsequent use and experience of health services.
Study design
The proposed study uses a prospective cohort study design. Respondents from the 2016/17 New Zealand Health Survey [ 2 , 18 , 19 ] (NZHS) provide the source of the follow-up cohort sample and the NZHS provides baseline data. The follow-up survey will be conducted between one and two years after respondents completed the NZHS. Using the NZHS data as our sampling frame provides access to exposure status (experience of racism), along with data on a substantial number of covariates (including age, gender, and socioeconomic variables) allowing us to select an appropriate study cohort for answering our research questions. Participant follow-up will be conducted by a multi-modality survey (mail, web and telephone modalities).
This study explores the impact of racism on health in the general NZ adult population (which is the target population of the NZHS that forms the baseline of the study).
Participants
Participants were selected from adult NZHS 2016/17 interviewees ( n = 13,573, aged 15+ at NZHS interview) who consented to re-contact for future research within a 2 year re-contact window (92% of adult respondents). The NZHS is a complex-sample design survey with an 80% response rate for adults [ 18 ] and oversampling of Māori, Pacific, and Asian populations (who experience higher levels of racism), which facilitates studying the impact of racism on subsequent health status. Participants who had consented to re-contact ( n = 12,530) also needed to have contact details recorded and sufficient data on exposures/confounders to be included in the sampling frame ( n = 11,775, 93.9% of consenting adults). All invited participants will be aged at least 16 at the time of follow-up, as at least one year will have passed since participation in the NZHS (where all participants were aged at least 15).
Exposure to racism was determined from the five previously validated NZHS items [ 31 ] asked of all adult respondents (see Table 1 ) about personal experience of racism across five domains (verbal and physical attack; unfair treatment in health, housing, or work). Response options for each question cover recent exposure (within the past 12 months), more historical exposure (> 12 months ago), or no exposure to racism.
Identification of exposed and unexposed individuals
Individuals were classified as exposed to racism if they answered “yes” to any question in Table 1 , in either timeframe (recent or historical: referred to as “ever” exposure). This allows for analysis restricted to the nested subset of individuals reporting recent exposure to racism (past 12 months) and those only reporting more historical exposure (> 12 months ago). The unexposed group comprised all individuals answering “No” to all five domains of experience of racism. We selected all exposed individuals for follow-up, along with a matched sample of unexposed individuals. Individuals missing exposure data were explicitly excluded.
Matching of exposed and unexposed individuals
To address potential confounding, we used propensity score matching methods in our sampling stage to remove the impact of major confounders (as measured in the NZHS) of the causal association between experience of racism and health outcomes. Propensity score methods are increasingly used in observational epidemiology as a robust method for dealing with confounding in the analysis stage [ 32 , 33 , 34 , 35 , 36 ] and have more recently been considered as a useful approach for secondary sampling of participants from existing cohorts for subsequent follow up [ 37 ].
All exposed NZHS respondents will be invited into the follow-up survey. To find matched unexposed individuals, potential participants were stratified based on self-reported ethnicity (Māori, Pacific, Asian, European and Other; using prioritised ethnicity for individuals identifying with more than one grouping) [ 38 ] and then further matched for potential sociodemographic and socioeconomic confounders using propensity score methods [ 39 , 40 ]. Stratification by ethnicity reflects the differential prevalence of racism by ethnic group, and furthermore allows ethnically-stratified estimates of the impact of racism [ 22 ].
Propensity scores were modelled using logistic regression for “ever” exposure to racism based on major confounder variables of the association between racism and poor health (Table 2 ), with modelling stratified by ethnic group. Selection of appropriate confounders was based on past work using cross-sectional analysis of the 2011/12 NZHS (e.g. [ 21 , 22 ]) and the wider literature that informed the conceptual model for the project. Some additional variables were considered for inclusion in the matching process but were removed prior to finalisation (details in Table 2 ).
Within each ethnic group stratum, exposed individuals were matched with unexposed individuals (1:1 matching) based on propensity scores to make these two groups approximately exchangeable (confounders balanced between exposure groups). The matching process [ 41 ] used nearest neighbour matching as implemented in MatchIt [ 42 ] in R 3.4 (R Institute, Vienna, Austria). As the propensity score modelling is blind to participants’ future outcome status, the final propensity score models were refined using just the baseline NZHS data to achieve maximal balance of confounders between exposure groups, without risking bias to the subsequent primary causal analyses [ 39 ]. Balance between groups was then checked on all matching variables prior to finalisation of the sampling lists.
Questionnaire development
Development of the follow-up questionnaire was informed by a literature review and a conceptual model (Figs. 1 and 2 ) of the potential pathways from racism to health outcomes (Fig. 1 ) and health service utilisation (Fig. 2 ) [ 4 , 10 , 16 , 43 , 44 ]. The literature review focussed on longitudinal studies of racism and health among adolescents and adults that included health or health service outcomes. The literature review covered longitudinal studies post-dating the 2015 systematic review by Paradies et al. [ 12 ], using similar search terms for papers between 2013 and 2017 indexed in Medline and PubMed databases, alongside additional studies from systematic reviews [ 12 , 16 ].
![racial discrimination research paper figure 1](https://media.springernature.com/lw685/springer-static/image/art%3A10.1186%2Fs12889-019-6664-x/MediaObjects/12889_2019_6664_Fig1_HTML.png)
Potential pathways between racism and health outcomes. Direct pathway: Main arrow represents the direct biopsychosocial and trauma pathways between experience of racial discrimination (Time 1) and negative health outcomes (Time 2) Indirect pathways: Racial discrimination (Time 1) can impact negatively on health outcomes (Time 2) via healthcare pathways (e.g. less engagement, unmet need). Racial discrimination (Time 1) can impact negatively on physical health outcomes (Time 2) via mental health pathways
![racial discrimination research paper figure 2](https://media.springernature.com/lw685/springer-static/image/art%3A10.1186%2Fs12889-019-6664-x/MediaObjects/12889_2019_6664_Fig2_HTML.png)
Potential pathways between racism and healthcare utilisation outcomes. Main pathway: Main arrow represents the pathway between experience of racial discrimination (Time 1) and negative healthcare measures (Time 2), via negative perceptions and expectations of healthcare (providers, organisations, systems) and future engagement. Secondary pathway: Racial discrimination (T1) can impact negatively on healthcare (Time 2) via negative impacts on health increasing healthcare need
We used several criteria for considering and prioritising variables for the questionnaire. The conceptual model also informed prioritisation of variables for the questionnaire. For outcome measures, these included: alignment with study aims and objectives; existing evidence of a relationship between racism and outcome; New Zealand evidence of ethnic inequities in outcome; previous cross-sectional relationships between racism and outcome in New Zealand data; availability of baseline measures (for health outcomes); plausibility of health effects manifesting within a 1–2 year follow-up period; and data quality (e.g. validated measures, low missing data, questions suitable for multimodal administration). Mediators and confounders were considered for variables not available in the baseline NZHS survey, as was recent experience of racism (following the NZHS interview) to provide additional measurement of exposure to recent racism. A final consideration for prioritising items for inclusion was keeping the length of the questionnaire short in order to maximise response rates (while being able to fully address the study aims). The questionnaire was extensively discussed by the research team and reviewed by the study advisors prior to finalisation.
Table 3 summarises the outcome measures by topic domain and original source (with references). The final questionnaire content can be found in the Additional file 1 , and includes: health outcome measures of mental and physical health (using SF12-v2 and K10 scales); health service measures (unmet need, satisfaction with usual medical centre, experiences with general practitioners); experience of racism in the last 12 months (adapted from items in the NZHS); and variables required to restrict data (e.g. having a usual medical centre, type of centre, having a General Practitioner [GP] visit in the last 12 months) or potential confounder and mediator variables not available at baseline (e.g. number of GP visits).
Recruitment and data collection
Recruitment is currently underway. The sampling phase provided a list of potential participants for invitation, and recruitment for the follow-up survey uses the contact details from the NZHS interview (physical address, mobile/landline telephone, and email address if available). Recruitment will take place over three tranches to (1) manage fieldwork capacity and (2) allow tracking of response rates and adaptation of contact strategies if recruitment is sub-optimal.
To maximise response rates, we chose to use a multi-modal survey [ 45 ]. Participants are invited to respond by a paper questionnaire included with the initial invitation letter (questionnaire returned by pre-paid post), by self-completed online questionnaire, or by computer-assisted telephone interview (CATI, on mobile or landline.) A pen is included in the study invitation to improve initial engagement with the paper-based survey [ 46 ]. Participants completing the survey are offered a NZ$20 gift card to recognise their participation. The contact information contains instructions for opting out of the study.
Those participants not responding online or by post receive a reminder postcard mailed out two weeks after the initial letter, containing a link to the web survey and a note that the participant will be contacted by telephone in two weeks’ time.
Two weeks after the reminder postcard (four weeks post-invitation) remaining non-respondents are contacted using CATI processes. For those with mobile phone numbers or email addresses, a text (SMS) or email reminder is sent two days before the telephone contact phase. Once contact is made by telephone, the interviewer asks the participant to complete the survey over the telephone at that time or organises a subsequent appointment (interview duration approximately 15 min). Interviewers make up to seven telephone contact attempts for each participant, using all recorded telephone numbers. Respondents who decline to complete the full interview at telephone follow-up are asked to consider answering two priority questions (self-rated health and any unmet need for healthcare in the last 12 months: questions 1 and 8 in Table 3 and Additional file 1 ).
Past surveys conducted in NZ have frequently noted lower response rates and hence under-representation of Māori [ 47 , 48 ]. Drawing on Kaupapa Māori research principles, we are explicitly aiming for equitable response rates of Māori to ensure maximum power for ethnically stratified analysis. This involves providing culturally appropriate invitations and interviewers for participants, and actively monitoring response rates by ethnicity during data collection to allow longer and more frequent follow-up of Māori, Pacific and Asian participants if required [ 48 , 49 ]. The use of a multi-modal survey is also expected to minimise recruitment problems inherent to any single modality (e.g. lower phone ownership or internet access in some ethnic groups).
We have contracted an external research company to co-ordinate recruitment and data collection fieldwork under our supervision (covering all contact processes described here), which follows recruitment and data management protocols set by our research team.
Statistical analysis
Propensity score methods for the sampling stage are described above: this section focuses on causal analyses for health outcomes in the achieved sample. The sampling frame selects participants based on “ever” experience of racism, which is our exposure definition.
All analyses will account for both the complex survey sampling frame (weights, strata and clusters from the NZHS) and the secondary sampling phase (selection based on propensity scores). Complex survey data will be handled using software to account for these designs (e.g. survey package [ 50 ] in R); propensity scores will be handled in the main analysis by using inverse probability of treatment weights (IPTW) combined with the sampling weights [ 51 ].
Linear regression methods will be used to compare change in continuous outcome measures (e.g. K10 score) by estimating mean score at follow-up, adjusted for baseline. Analysis of dichotomous categorical outcomes (e.g. self-rated health) will use logistic regression methods, again adjusted for baseline (for health outcomes). We will conduct analyses stratified by ethnic group to explore whether the impact of racism differs by ethnic group. Models will adjust for confounders included in creating the propensity scores (doubly-robust estimation) to address residual confounding not fully covered by the propensity score approach [ 52 ]. Analysis for other outcomes will use similar methods.
As we hypothesise that some outcomes (e.g. self-reported mental distress) will be more strongly influenced by recent experience of racism, we will also examine our main outcomes restricted to those only reporting historical (more than 12 months ago) or recent (last 12 months) racism at baseline. These historical and recent experience groups (and corresponding unexposed individuals) form nested sub-groups of the total cohort, and so analysis will follow the same framework outlined above. Experience of racism in the last 12 months (measured at follow-up) will be examined in cross-sectional analyses and in combination with baseline measures of racism to create a measure to examine the cumulative impact of racism on outcomes.
Sensitivity analyses
While the sampling invitation lists are based on matched samples, we have no control about specific individuals choosing to participate in the follow-up survey, and so the original matching is unlikely to be maintained in the achieved sample. We will conduct sensitivity analyses using re-matched data (based on propensity scores for those participating in follow-up) to allow for re-calibration of exposed and unexposed groups in the achieved sample.
To consider potential for bias due to non-response in our follow-up sample, we will compare NZHS 2016/17 cross-sectional data for responders and non-responders on baseline sociodemographic, socioeconomic, and baseline health variables.
Sample size
Based on NZHS 2011/12 responses, we anticipated a total pool of 2100 potential participants with “ever” experience of racism, with approximately 1100 expected to be Māori/Pacific/Asian ethnicity, and 10,000 with no report of racism (at least 2 unexposed per exposed individual in each ethnic group).
For the main analyses (based on “ever” experience of racism) we assumed a conservative follow-up rate of 40%, giving a final sample size of at least 840 exposed individuals. This response rate includes re-contact and agreement to participate, based on past experience recruiting NZHS participants for other studies and the relative length of the current survey questionnaire.
Initial projections (based on NZHS2011/12 data) indicated sufficient numbers of unexposed individuals for 1:1 matching based on ethnicity and propensity scores. This gives a feasible total sample size of n = 1680, providing substantial power for the K10 mental health outcome (standard deviation = 6.5: > 95% power to detect difference in change of 2 units of K10 between groups.) For the second main health outcome (change in self-rated health), this sample size will have > 85% power for a difference between 8% of those exposed to racism having worse self-reported health at follow-up (relative to baseline) compared to 5% of unexposed individuals.
For analyses of effects stratified by ethnicity, we expect > 95% power for Māori participants ( n = 280 each exposed and unexposed) for the K10 outcome (assumptions as above); change in self-rated health will have 80% power for a difference between 12% of exposed individuals having worse self-reported health at follow-up (relative to baseline) compared to 5% of unexposed individuals. Stratified estimates for Pacific and Asian groups will have poorer precision, but should still provide valid comparisons.
Ethical approval and consent to participate
The study involves recruiting participants who have already completed the NZHS interview (including questions on racial discrimination) The NZHS as conducted by the Ministry of Health has its own ethical approval (MEC/10/10/103) and participants are only invited onto the present study if they explicitly consented (at the time of completing the NZHS) to re-contact for future health research. The current study was reviewed and approved by the University of Otago’s Human Ethics (Health) Committee prior to commencement of fieldwork (reference: H17/094). Participants provided informed consent to participate at the time of completing the follow-up survey depending on response modality: implicitly through completion and return of the paper survey which stated “By completing this survey, you indicate that you understand the research and are willing to participate” (see Additional file 1 : a separate written consent document was not required by the ethics committee); in the online survey by responding “yes” to a similarly worded question that they understood the study and agreed to take part (recorded as part of data collection, and participation could not continue unless ticked), or by verbal consent in a similar initial question in the telephone interview (since written consent could not be collected in this setting). These consent methods were approved by the reviewing Ethics committee [ 53 ]. Ethical approval for the study included using the same consent processes for those participants aged 16 to 18 as for older participants.
This study will contribute robust evidence to the limited national and international literature from prospective studies on the causal links between experience of racism and subsequent health. The use of the NZHS as the baseline for the prospective study capitalises on the inclusion of racism questions in that survey to provide a unique and important opportunity to build on and substantially strengthen the current evidence base for the impact of racism on health using data spanning the entire New Zealand adult population. In addition, our use of propensity scores in the sampling phase is a novel approach to prospective recruitment of participants from the NZHS. This approach should manage confounding while reducing the need (and cost) of following up all NZHS participants, without compromising the internal validity of the results. The novel methods developed for using the NZHS as the base for a prospective cohort study will have wider application to other health priority areas. One general limitation of this approach is that baseline data (for both propensity score development and baseline health measures) is limited to the data captured in the existing larger survey. We anticipate that this study will assist in prioritising racism as a health determinant and inform the development of anti-racism interventions in health service delivery and policy making.
Current stage of research
Funding for this project began October 1st 2017. The first set of respondent invitations was mailed out on July 12th 2018; fieldwork for the final tranche of invitations was underway at the time of submission and is expected to be completed by 31 December 2018. Analysis and reporting will take place in mid-to-late 2019.
Abbreviations
Computer Assisted Telephone Interview
General Practitioner
General Social Survey
Index of Multiple Deprivation
Inverse Probability of Treatment Weights
- New Zealand
New Zealand Deprivation Index
New Zealand Health Survey
12/36-Item Short Form Survey
short message service
Commission on Social Determinants of Health. Closing the gap in a generation: health equity through actions on the social determinants of health. Geneva: World Health Organization; 2008.
Google Scholar
Ministry of Health. Annual Update of Key Results 2016/17: New Zealand Health Survey 2017 [Available from: https://www.health.govt.nz/publication/annual-update-key-results-2016-17-new-zealand-health-survey . accessed 13/09/2018].
Ministry of Social Development. The Social Report 2016: Te pūrongo oranga tangata. Wellington: Ministry of Social Development; 2016.
Williams DR, Mohammed SA. Racism and health I: pathways and scientific evidence. Am Behav Sci. 2013;57(8). https://doi.org/10.1177/0002764213487340 .
Reid P, Robson B. Understanding health inequities. In: Robson B, Harris R, editors. Hauora: Maori standards of health IV a study of the years 2000–2005. Wellington: Te Ropu Rangahau Hauora a Eru Pomare; 2007. p. 3–10.
Jones C. Confronting institutionalized racism. Phylon. 2002;50:7–22.
Article Google Scholar
Garner S. Racisms: An introduction. London/Thousand Oaks CA: Sage; 2010.
Book Google Scholar
Becares L, Cormack D, Harris R. Ethnic density and area deprivation: neighbourhood effects on Maori health and racial discrimination in Aotearoa/New Zealand. Soc Sci Med. 2013;88:76–82. https://doi.org/10.1016/j.socscimed.2013.04.007 .
Article PubMed PubMed Central Google Scholar
Paradies YC. Defining, conceptualizing and characterizing racism in health research. Crit Public Health. 2006;16(2):143–57. https://doi.org/10.1080/09581590600828881 .
Krieger N. Methods for the scientific study of discrimination and health: an ecosocial approach. Am J Public Health. 2012;102(5):936–44. https://doi.org/10.2105/AJPH.2011.300544 .
Jones CP. Invited commentary: "race," racism, and the practice of epidemiology. Am J Epidemiol. 2001;154(4):299–304; discussion 05-6.
Article CAS Google Scholar
Paradies Y, Ben J, Denson N, et al. Racism as a determinant of health: a systematic review and meta-analysis. PLoS One. 2015;10(9):e0138511. https://doi.org/10.1371/journal.pone.0138511 .
Article CAS PubMed PubMed Central Google Scholar
Lewis TT, Cogburn CD, Williams DR. Self-reported experiences of discrimination and health: scientific advances, ongoing controversies, and emerging issues. Annu Rev Clin Psychol. 2015;11:407–40. https://doi.org/10.1146/annurev-clinpsy-032814-112728 .
Gee GC, Ro A, Shariff-Marco S, et al. Racial discrimination and health among Asian Americans: evidence, assessment, and directions for future research. Epidemiol Rev. 2009;31:130–51. https://doi.org/10.1093/epirev/mxp009 .
Williams DR, Mohammed SA. Discrimination and racial disparities in health: evidence and needed research. J Behav Med. 2009;32(1):20–47. https://doi.org/10.1007/s10865-008-9185-0 .
Article PubMed Google Scholar
Ben J, Cormack D, Harris R, et al. Racism and health service utilisation: a systematic review and meta-analysis. PLoS One. 2017;12(12):e0189900. https://doi.org/10.1371/journal.pone.0189900 published Online First: 2017/12/19.
Harris R, Cormack D, Tobias M, et al. The pervasive effects of racism: experiences of racial discrimination in New Zealand over time and associations with multiple health domains. Soc Sci Med. 2012;74(3):408–15. https://doi.org/10.1016/j.socscimed.2011.11.004 .
Ministry of Health. Methodology report 2016/17: New Zealand health survey. Wellington: Ministry of Health; 2017.
Ministry of Health. Content guide 2016/17: New Zealand health survey. Wellington: Ministry of Health; 2017.
Harris R, Cormack D, Tobias M, et al. Self-reported experience of racial discrimination and health care use in New Zealand: results from the 2006/07 New Zealand health survey. Am J Public Health. 2012;102(5):1012–9. https://doi.org/10.2105/AJPH.2011.300626 .
Harris RB, Cormack DM, Stanley J. Experience of racism and associations with unmet need and healthcare satisfaction: the 2011/12 adult New Zealand health survey. Aust N Z J Public Health. 2018. https://doi.org/10.1111/1753-6405.12835 published Online First: 2018/10/09.
Harris RB, Stanley J, Cormack DM. Racism and health in New Zealand: prevalence over time and associations between recent experience of racism and health and wellbeing measures using national survey data. PLoS One. 2018;13(5):e0196476. https://doi.org/10.1371/journal.pone.0196476 published Online First: 2018/05/04.
Dyall L, Kepa M, Teh R, et al. Cultural and social factors and quality of life of Maori in advanced age. Te puawaitanga o nga tapuwae kia ora tonu - life and living in advanced age: a cohort study in New Zealand (LiLACS NZ). N Z Med J. 2014;127(1393):62–79.
PubMed Google Scholar
Crengle S, Robinson E, Ameratunga S, et al. Ethnic discrimination prevalence and associations with health outcomes: data from a nationally representative cross-sectional survey of secondary school students in New Zealand. BMC Public Health. 2012;12:45. https://doi.org/10.1186/1471-2458-12-45 .
Thayer ZM, Kuzawa CW. Ethnic discrimination predicts poor self-rated health and cortisol in pregnancy: insights from New Zealand. Soc Sci Med. 2015;128:36–42. https://doi.org/10.1016/j.socscimed.2015.01.003 .
Stronge S, Sengupta NK, Barlow FK, et al. Perceived discrimination predicts increased support for political rights and life satisfaction mediated by ethnic identity: a longitudinal analysis. Cult Divers Ethn Minor Psychol. 2016;22(3):359–68. https://doi.org/10.1037/cdp0000074 .
Becares L, Atatoa-Carr P. The association between maternal and partner experienced racial discrimination and prenatal perceived stress, prenatal and postnatal depression: findings from the growing up in New Zealand cohort study. Int J Equity Health. 2016;15(1):155. https://doi.org/10.1186/s12939-016-0443-4 .
Robson B, Harris R, editors. Hauora : Maori standards of health IV : a study of the years 2000–2005. Wellington: Te Rōpū Rangahau Hauora a Eru Pōmare; 2007.
UN General Assembly. United Nations Declaration on the Rights of Indigenous Peoples : resolution / adopted by the General Assembly 2007. Available from: https://undocs.org/A/RES/61/295 . Accessed 1 Nov 2018.
Smith L. Decolonizing methodologies: research and indigenous peoples. 2nd ed. London and New York: Zed Books; 2012.
Harris R, Tobias M, Jeffreys M, et al. Racism and health: the relationship between experience of racial discrimination and health in New Zealand. Soc Sci Med. 2006;63(6):1428–41. https://doi.org/10.1016/j.socscimed.2006.04.009 .
Sturmer T, Joshi M, Glynn RJ, et al. A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods. J Clin Epidemiol. 2006;59(5):437–47. https://doi.org/10.1016/j.jclinepi.2005.07.004 .
Luo Z, Gardiner JC, Bradley CJ. Applying propensity score methods in medical research: pitfalls and prospects. Med Care Res Rev. 2010;67(5):528–54. https://doi.org/10.1177/1077558710361486 .
Weitzen S, Lapane KL, Toledano AY, et al. Principles for modeling propensity scores in medical research: a systematic literature review. Pharmacoepidemiol Drug Saf. 2004;13(12):841–53. https://doi.org/10.1002/pds.969 .
Shah BR, Laupacis A, Hux JE, et al. Propensity score methods gave similar results to traditional regression modeling in observational studies: a systematic review. J Clin Epidemiol. 2005;58(6):550–9. https://doi.org/10.1016/j.jclinepi.2004.10.016 .
Stuart EA. Matching methods for causal inference: a review and a look forward. Stat Sci. 2010;25(1):1–21. https://doi.org/10.1214/09-STS313 published Online First: 2010/09/28.
Stuart EA, Ialongo NS. Matching methods for selection of subjects for follow-up. Multivariate Behav Res. 2010;45(4):746–65. https://doi.org/10.1080/00273171.2010.503544 [published Online First: 2011/01/12].
Ministry of Health. Ethnicity Data Protocols Wellington: Ministry of Health; 2017. Available from: https://www.health.govt.nz/system/files/documents/publications/hiso-10001-2017-ethnicity-data-protocols-v2.pdf . Accessed 1 Nov 2018.
Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399–424. https://doi.org/10.1080/00273171.2011.568786 .
Caliendo M, Kopeinig S. Some practical guidance for the implementation of propensity score matching. J Econ Surv. 2008;22(1):31–72. https://doi.org/10.1111/j.1467-6419.2007.00527.x .
Ho DE, Imai K, King G, et al. Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Polit Anal. 2017;15(03):199–236. https://doi.org/10.1093/pan/mpl013 .
Ho DE, Imai K, King G, et al. MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Softw. 2011;42(8):1–28.
Sarnyai Z, Berger M, Jawan I. Allostatic load mediates the impact of stress and trauma on physical and mental health in indigenous Australians. Australas Psychiatry. 2016;24(1):72–5. https://doi.org/10.1177/1039856215620025 published Online First: 2015/12/10.
Kaholokula J. Mauli Ola: Pathways toward Social Justice for Native Hawaiians. Townsville: LIME Network Conference; 2015.
Fowler FJ, Roman AM, Mahmood R, et al. Reducing nonresponse and nonresponse error in a telephone survey: an informative case study. J Survey Stat Methodol. 2016;4(2):246–62. https://doi.org/10.1093/jssam/smw004 .
White E, Carney PA, Kolar AS. Increasing response to mailed questionnaires by including a pencil/pen. Am J Epidemiol. 2005;162(3):261–6. https://doi.org/10.1093/aje/kwi194 [published Online First: 2005/06/24].
Fink JW, Paine SJ, Gander PH, et al. Changing response rates from Maori and non-Maori in national sleep health surveys. N Z Med J. 2011;124(1328):52–63.
Paine SJ, Priston M, Signal TL, et al. Developing new approaches for the recruitment and retention of Indigenous participants in longitudinal research: Lessons from E Moe, Māmā: Maternal Sleep and Health in Aotearoa/New Zealand. MAI Journal: A New Zealand Journal of Indigenous Scholarship. 2013;2(2):121–32.
Selak V, Crengle S, Elley CR, et al. Recruiting equal numbers of indigenous and non-indigenous participants to a 'polypill' randomized trial. Int J Equity Health. 2013;12:44. https://doi.org/10.1186/1475-9276-12-44 .
Lumley T. survey: analysis of complex survey samples v3.32 2017 [R package]. Available from: https://cran.r-project.org/web/packages/survey/index.html . Accessed 1 Nov 2018.
Lenis D, Nguyen TQ, Dong N, et al. It's all about balance: propensity score matching in the context of complex survey data. Biostatistics. 2017. https://doi.org/10.1093/biostatistics/kxx063 [published Online First: 2018/01/03].
Funk MJ, Westreich D, Wiesen C, et al. Doubly robust estimation of causal effects. Am J Epidemiol. 2011;173(7):761–7. https://doi.org/10.1093/aje/kwq439 [published Online First: 2011/03/10].
National Ethics Advisory Committee. Ethical guidelines for observational studies: observational research, audits and related activities. Revised edition. Wellington: Ministry of Health; 2012.
Atkinson J, Salmond C, Crampton P. NZDep2013 index of deprivation. Dunedin: University of Otago; 2014.
Exeter DJ, Zhao J, Crengle S, et al. The New Zealand indices of multiple deprivation (IMD): a new suite of indicators for social and health research in Aotearoa, New Zealand. PLoS One. 2017;12(8):e0181260. https://doi.org/10.1371/journal.pone.0181260 [published Online First: 2017/08/05].
Kessler RC, Andrews G, Colpe LJ, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32(6):959–76.
Ware J Jr, Kosinski M, Keller SD. A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220–33.
Fleishman JA, Selim AJ, Kazis LE. Deriving SF-12v2 physical and mental health summary scores: a comparison of different scoring algorithms. Qual Life Res. 2010;19(2):231–41. https://doi.org/10.1007/s11136-009-9582-z .
Bastos JL, Harnois CE, Paradies YC. Health care barriers, racism, and intersectionality in Australia. Soc Sci Med. 2018;199:209–18. https://doi.org/10.1016/j.socscimed.2017.05.010 [published Online First: 2017/05/16].
Australian Bureau of Statistics. General Social Survey 2014: Household questionnaire. Canberra: Australian Bureau of Statistics; 2014.
Ministry of Health. The New Zealand Health Survey: Content Guide and questionnaires 2011–2012 Wellington: Ministry of Health; 2012 [Available from: http://www.health.govt.nz/publication/new-zealand-health-survey-content-guide-and-questionnaires-2011-2012 . accessed 31/10/2016 2016].
Download references
Acknowledgements
We would like to acknowledge the assistance of the Ministry of Health’s New Zealand Health Survey Team for facilitating access to the NZHS data and respondent lists, and for help with constructing the questionnaire (including providing the Helpline contact template).
We would also like to acknowledge the expertise and input of our project advisory team: Natalie Talamaivao (Senior Advisor, Māori Health Research, Ministry of Health), Associate Professor Bridget Robson (Director, Eru Pōmare Māori Health Research Centre, University of Otago, Wellington), and Dr. Sarah-Jane Paine (Senior Research Fellow, University of Auckland and University of Otago, Wellington). Thanks also to Ms. Ruruhira Rameka (Eru Pōmare Māori Health Research Centre, University of Otago, Wellington) for providing administrative support. Research New Zealand was contracted to undertake the data collection and other fieldwork for the follow-up survey.
This project was funded by the Health Research Council of New Zealand (HRC 17–066). The funding body approved the study but has no further role in the study design or outputs from the study.
Availability of data and materials
Data from the follow-up study is not available to other researchers as participants did not provide their consent for data sharing. The NZHS 2016/17 data used as the baseline for the study described in this protocol is available to approved researchers subject to the New Zealand Ministry of Health’s Survey Microdata Access agreement https://www.health.govt.nz/nz-health-statistics/national-collections-and-surveys/surveys/access-survey-microdata .
Author information
Authors and affiliations.
Department of Public Health, University of Otago, Wellington, 23a Mein Street, Newtown, Wellington, New Zealand
James Stanley & Richard Edwards
Eru Pōmare Māori Health Research Centre, University of Otago, Wellington, 23a Mein Street, Newtown, Wellington, New Zealand
Ricci Harris, Donna Cormack & Andrew Waa
You can also search for this author in PubMed Google Scholar
Contributions
JS and RH initiated the project and are co-principal investigators of the study, and jointly led writing of the grant application and this protocol paper. JS designed the sampling plan, led the development of the contact protocol, led the development of the statistical analysis plan, contributed to revising the questionnaire, and is guarantor of the paper. RH designed the questionnaire, contributed to development of the sampling and contact protocol, and co-led the statistical analysis plan. DC led the conceptual plan with support from RH. AW and RE contributed to the contact protocol. DC, AW and RE all contributed to writing the grant application, revising the questionnaire and sampling plans, and revising the draft protocol paper. All authors read and approved the final version of the manuscript.
Corresponding author
Correspondence to James Stanley .
Ethics declarations
Ethics approval and consent to participate.
The follow-up study protocol and questionnaire were approved by the University of Otago’s Human Ethics (Health) Committee prior to commencement of fieldwork (reference: H17/094). The NZHS has its own ethical approval as granted to the New Zealand Ministry of Health (NZ Multi-Region Ethics Committee, MEC/10/10/103), and consent for re-contact was gained from participants at the time of their NZHS interview. Participants provided informed consented to participate at the time of completing the follow-up survey: implicitly through completion and return of the paper survey which stated “By completing this survey, you indicate that you understand the research and are willing to participate”; in the online survey by responding “yes” to a similarly worded question that they understood the study and agreed to take part, or by verbal consent in a similar initial question in the telephone interview.
Consent for publication
Not applicable.
Competing interests
JS, RH, DC, AW, and RE report funding from the Health Research Council of New Zealand to complete this work. JS and RH report personal fees from the Health Research Council of New Zealand for service as external members on committees (neither are employees of the HRC), outside the scope of the current work.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Additional file
Additional file 1:.
Questionnaire used in follow-up survey. (PDF 919 kb)
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.
Reprints and permissions
About this article
Cite this article.
Stanley, J., Harris, R., Cormack, D. et al. The impact of racism on the future health of adults: protocol for a prospective cohort study. BMC Public Health 19 , 346 (2019). https://doi.org/10.1186/s12889-019-6664-x
Download citation
Received : 28 November 2018
Accepted : 15 March 2019
Published : 28 March 2019
DOI : https://doi.org/10.1186/s12889-019-6664-x
Share this article
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
- Prospective cohort study
- Health service utilisation
- Health inequities
BMC Public Health
ISSN: 1471-2458
- General enquiries: [email protected]
![racial discrimination research paper U.S. flag](https://www.ncbi.nlm.nih.gov/coreutils/uswds/img/favicons/favicon-57.png)
An official website of the United States government
The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.
The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.
- Publications
- Account settings
Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .
- Advanced Search
- Journal List
- Health Serv Res
- v.54(Suppl 2); 2019 Dec
Understanding how discrimination can affect health
David r. williams.
1 Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston Massachusetts
2 Department of African and African American Studies, Department of Sociology, Harvard University, Cambridge Massachusetts
![](http://serviteca.online/777/templates/cheerup2/res/banner1.gif)
Jourdyn A. Lawrence
Brigette a. davis, associated data.
To provide an overview of the empirical research linking self‐reports of racial discrimination to health status and health service utilization.
A review of literature reviews and meta‐analyses published from January 2013 to 2019 was conducted using PubMed, PsycINFO, Sociological Abstracts, and Web of Science. Articles were considered for inclusion using the Preferred Reporting Items for Systematic Review and Meta‐Analyses (PRISMA) framework.
Twenty‐nine studies met the criteria for review. Both domestic and international studies find that experiences of discrimination reported by adults are adversely related to mental health and indicators of physical health, including preclinical indicators of disease, health behaviors, utilization of care, and adherence to medical regimens. Emerging evidence also suggests that discrimination can affect the health of children and adolescents and that at least some of its adverse effects may be ameliorated by the presence of psychosocial resources.
Conclusions
Increasing evidence indicates that racial discrimination is an emerging risk factor for disease and a contributor to racial disparities in health. Attention is needed to strengthen research gaps and to advance our understanding of the optimal interventions that can reduce the negative effects of discrimination.
1. INTRODUCTION
Racial and ethnic differences in health, in which socially disadvantaged racial populations have worse health than whites, are large, pervasive across a broad range of outcomes, and persistent over time. 1 They exist for the onset of disease, as well as the severity and course of illness. Socioeconomic status (SES)—whether measured by income, education, occupational status, or wealth—is a strong predictor of variations in health and has often been viewed as the driver of racial inequities in health. Research finds that although SES predicts variations in health status within each racial group, racial disparities persist at every level of SES. 2 There is a large and growing body of empirical evidence indicating self‐reports of discrimination are race‐related aspects of social experience that can have negative effects on health. This paper provides an overview of research on self‐reported discrimination and health, as well as health care utilization. It begins by situating research on racial discrimination and health within the larger context of research on racism and health. Importantly, self‐reported experiences of discrimination are one mechanism by which racism affects health, and these exposures can be best understood and effectively addressed within the context of the role of racism in health. The paper then highlights key findings in this burgeoning literature.
2. BACKGROUND AND THEORETICAL FRAMEWORK
Figure Figure1 1 illustrates the multiple components of racism and the ways in which these components can affect health. Racism is viewed as a dynamic societal system that is shaped by and reshapes other social institutions such as the political, legal, and economic systems. 3 , 4 , 5 , 6 Central to racism, in the US context, is a hierarchical ideology that the dominant white group uses to categorize and rank social groups into races with whites being superior compared to other races. There are three major pathways that link racism to inequities in society and health. The first pathway by which racism operates is cultural racism. 6 This refers to the embedding of the inferiority of blacks and other nonwhites into the belief systems, images, and norms of the larger culture that leads to widespread negative beliefs (stereotypes) and attitudes (prejudice) that devalue, marginalize, and subordinate nonwhite racial populations. Cultural racism creates a larger ideological environment within which the system of racism can flourish. It initiates and sustains racial prejudice and negative racial stereotypes that can lessen support for egalitarian policies, trigger health‐damaging psychological responses in stigmatized persons such as internalized racism and stereotype threat, and facilitate explicit and implicit biases that restrict access to desirable resources, including medical care. 6
![An external file that holds a picture, illustration, etc.
Object name is HESR-54-1374-g001.jpg An external file that holds a picture, illustration, etc.
Object name is HESR-54-1374-g001.jpg](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6864381/bin/HESR-54-1374-g001.jpg)
The House that Racism Built
The second pathway is institutional or structural racism. We use these terms interchangeably to refer to societal structures and policies that reduce access of the socially stigmatized to desirable opportunities and resources in society. 5 The system of racism develops and sustains policies and structures that empower the dominant group to differentially allocate desirable societal opportunities and resources to racial groups regarded as inferior. Residential segregation is one example of an institutional mechanism of racism that adversely affects health in multiple ways. 7 , 8 The forced removal and relocation of American Indians to reservations is another example of institutionalized isolation of a marginalized racial population. Segregation is a critical determinant of SES, as it reduces access to quality elementary and high school education, preparation for higher education, and access to employment opportunities. One national study found that the elimination of segregation would erase black‐white differences in income, education, and unemployment, and reduce racial differences in single motherhood by two‐thirds. 9 SES, in turn, is a strong predictor of variation in health and risk factors that affect health. Segregation can also lead to increased exposure to multiple psychosocial, physical, and chemical stressors linked to neighborhood and housing conditions, including crime, violence, and air pollution. It can also affect access to and the quality of local services, ranging from medical care to municipal services.
The third pathway through which racism operates is through individual‐level discrimination. Stigmatized racial groups experience differential treatment (discrimination) directed at them by both social institutions and individuals. Considerable scientific evidence documents the persistence of objectively assessed individual discrimination in contemporary society. A review of audit studies—those in which researchers carefully select, match, and train individuals to be equally qualified in every respect but to differ only in race—provide striking examples of contemporary racial discrimination. 10 Discrimination has been documented in renting apartments, purchasing homes and cars, obtaining mortgages and medical care, applying for insurance, and hailing taxis. Such incidents of discrimination can lead to reduced access to a broad range of societal resources and opportunities. Figure Figure1 1 indicates that the persistence of stark racial inequities in multiple domains of society can confirm racial stereotypes and stigma, and thus serve to reinforce the system of racism. Moreover, the pathways by which racism affect are interrelated and mutually reinforcing. 11
The lower panel of Figure Figure1 1 serves to further unpack how individual‐level discrimination can affect health. The focus here is on a subset of incidents of individual discrimination that is perceived by the individual. According to social stress theory, perceived discrimination is a type of stressor that, like other psychosocial stressors, is adversely related to a broad range of physical and mental health outcomes. 12 , 13 A recent study, for example, documented that self‐reported experiences of discrimination are associated with neural functioning in ways that mirror patterns observed for other psychosocial stressors (eg, greater spontaneous amygdala activity and greater connectivity between the amygdala and other regions of the brain including the thalamus). 14 The lower panel of Figure Figure1 1 delineates how discriminatory incidents of which the individual is aware can trigger appraisal and affective reactions that can be experienced as stressful life exposures, and they have a cascade of negative effects on health. 15 They can lead to negative emotions that can adversely affect psychological well‐being, leading to symptoms of distress and increasing the risk of discrete psychiatric disorders. These negative emotions can also lead to biological dysregulation that can contribute to indicators of subclinical disease and chronic physical illness. 15 Coping with negative emotional states can also lead to increases in risky health behaviors, including declines in the utilization of and engagement with health care services. Figure Figure1 1 also acknowledges that in the face of exposure to discrimination, individuals and groups can respond in ways that can neutralize at least some of the negative effects of discrimination.
3.1. Search strategy
Reviews were identified through a search of PubMed, PsycINFO, Sociological Abstracts, and Web of Science. Reviews were eligible for inclusion if they were focused reviews or meta‐analyses, in English, published from January 2013 to the present, extending the systematic review and meta‐analysis published by Paradies and colleagues. 16 The following keywords were used: (racism* OR social discrimination*) OR (race* OR racial*) AND discriminat*)) AND (systematic*[sb] OR systematic*[ti] OR review*[ti] OR review*[sb] OR meta‐analysis*[ti]). The bibliographies of included studies were manually examined to identify additional reviews and meta‐analyses.
3.2. Inclusion criteria
Two of us (JAL, CV) reviewed titles and abstracts of the traced articles followed by a full‐text review to check inclusion criteria using the Covidence systematic review software. 17 A third author (DRW) acted as a tiebreaker regarding study selection and inclusion. A review was eligible for inclusion if it satisfies the following criteria: (a) evaluated studies examining self‐reported racial/ethnic discrimination or studies that examined perceived discrimination broadly, and (b) examined health or health‐related outcomes. This is consistent with the finding that adverse health effects of discrimination are generally evident, irrespective of whether an incident is linked to a general perception of bias or unfair treatment or to discriminatory experiences attributed to race/ethnicity or other stigmatized social statuses. 18 , 19 The outcomes were mental health, including positive psychological well‐being, indicators of physical health and risk factors, health behaviors, and health service utilization.
Of 1189 articles screened, based on the criteria for inclusion, two authors (JAL, CV) completed title and abstract screening for 922 unique studies, identifying 32 for full‐text review. An additional study was identified for inclusion (n = 33) from a review of bibliographies. A total of 29 reviews were extracted for analysis (Table (Table1 1 ).
Reviews of the research literature linking discrimination and health
Discrimination study | Focus | No. papers included | Study design | Health outcomes | Findings |
---|---|---|---|---|---|
Mental health | |||||
Britt‐Spells, AM., et al (2018) | Depressive symptoms | 12 | Cross‐sectional: 100% | Depressive symptoms, psychological distress, psychiatric symptoms | Positive ( = .290; 95% CI: 0.235, 0.343) |
Carter, RT., et al (2019) | Mental and physical health | 242 Mental: 200 Physical: 48 Cultural: 88 Substance use: 23 | Cross‐sectional Longitudinal | Adverse mental health (eg, anxiety, depression, hostility, anger, stress, psychological distress), physical health (eg, blood pressure, BMI, self‐reported health), and substance use (eg, alcohol, smoking, polysubstance use) outcomes | Overall: positive ( = .16, < .01) Mental: positive ( = .21, < .01) Substance use: positive ( = .16, < .01) Physical: positive ( = .07, < .01) |
Carter, RT., et al (2017) | Mental and physical health | 105 | Cross‐sectional Longitudinal | Adverse mental health, physical health, and substance use | Overall: positive ( = .17; 95% CI: 0.15, 0.20) Mental: positive ( = .20; 95% CI: 0.17, 0.24) Physical: positive ( = .09; 95% CI: 0.03, 0.14) Substance use: null ( = .12; 95% CI: −0.02, 0.25) |
de Freitas, DF., et al (2018) | Mental health | 51 | Cross‐sectional Longitudinal | Psychological disturbance, depression, anxiety, psychosis, perceived stress, externalizing behavior, self‐esteem, positive evaluation of life, self‐efficacy, well‐being psychological adaptation | Overall: positive ( = .17; 95% CI: 0.15, 0.20) Mental: positive ( = .20; 95% CI: 0.17, 0.24) Physical: positive ( = .09; 95% CI: 0.03, 0.14) Substance use: null ( = .12; 95% CI: −0.02, 0.25) |
Hopkins, PD., Shook, NJ. (2017) | Anxiety | 24 | Cross‐sectional | General anxiety disorder or social anxiety disorder | Positive: 100% (n = 3) One study found a positive association with general anxiety disorder among a sample of only African Americans |
Jones, KP., et al (2016) | Mental and physical health | 90 | Of the 44 primary samples obtained, 4/44 were from experimental studies | Psychological health, physical health | Positive: physical health ( = .19; 95% CI: 0.07‐0.21); psychological health ( = .30; 95% CI: 0.15‐0.36) |
Kirkinis, K., et al (2018) | Trauma | 28 papers 44 associations | Cross‐sectional: 93% (n = 26) Longitudinal: 7% (n = 2) | PTSD, dissociation, other measures of trauma, and race‐based traumatic stress symptoms | Positive: 70% (31/44 associations) |
Lewis et al (2015) | Mental disorders | 12 | Cross‐sectional | DSM‐IV disorders (depression, anxiety, eating, psychotic) | Positive association with disorders Inverse association with depression, anxiety for Asian immigrants |
Paradies, Y., et al (2015) | Mental and physical health | 293 | Cross‐sectional: 89.8% Longitudinal: 9.0% Other: 1.2% | Negative mental health (ie, depression, distress, stress, anxiety, internalizing, negative affect, PTSD, somatization, suicide ideation/ attempts, other mental health symptoms, general mental health, overall negative mental health); positive mental health (ie, self‐esteem, control, life satisfaction, positive affect, well‐being, overall positive mental health); physical health (ie, BP, heart conditions, overweight, diabetes, misc.); general health | Negative: negative mental health ( = −.23; 95% CI: −0.24, −0.21); positive mental health ( = −.13; 95% CI: −0.16, −0.10); general health ( = −.13; 95% CI: −0.18, −0.09); physical health ( = −.09; 95% CI: −0.12, −0.06) |
Potter, LN., et al (2019) | Daily mental health | 25 | Longitudinal (daily diary studies) | Poor mental health in daily life (eg, depressive symptoms, negative affect, somatic symptoms, active coping) | Positive: poor mental health (10/11; 91%) |
Schmitt, MT., et al (2014) | Psychological well‐being | 328 for the first research question and 54 for the second | Cross‐sectional and longitudinal for the first meta‐analysis and experimental studies only for the second meta‐analysis | Psychological well being, broadly constructed (ie, mood, self‐esteem, anxiety, depression, life satisfaction, affect, other measures of mental health) | Negative: psychological well‐being ( = −.21; 95% CI: −0.22, −0.19) |
Triana, MC., et al (2015) | Mental and physical health | 79 | Cross‐sectional Longitudinal | Psychological health (ie, stress, mental health, anxiety, negative affect, self‐esteem, life satisfaction, and depression); physical health (ie, blood pressure, bodily pain, general physical health, illness, drug or alcohol use) | Negative: psychological health ( = −.12, = −0.14); physical health ( = −.06, = −0.07) Positive: coping behavior ( = .17, = 0.20) |
Vines, AI., et al (2017) | Mental health | 85 | Cross‐sectional Longitudinal | Mental health was not specified, but includes PTSD, depression. mediators/ confounders: aggression, coping & personality, internalized psych response (eg, self‐esteem), external supportive buffers | Conditional/mixed: no percentage breakdown of trends |
Physical health | |||||
Bernardo, CD., et al (2017) | Adiposity | 10 | Longitudinal: 100% | Weight change; waist circumference change; BMI change; become obese; remain obese | Weight change: positive Waist circumference change: 1 positive, 3 null BMI change: 2 positive, 2 null Become obese: positive Remain obese: null |
Black, LL., et al (2015) | Physical health | 19 | Cross‐sectional Longitudinal | Heart disease risk factors (ie, CRP (C‐reactive protein); coronary calcium positive status; IMT (carotid intima‐media thickness); arterial plaque; coronary artery calcification); blood pressure (and incidence of hypertension); adverse birth outcomes; cancer/tumor incidence; weight change (and body fat distribution); other outcomes (ie, all‐cause mortality (ACM); Epstein‐Barr virus reactivation (EBV); frequency of common colds/physical illnesses (cold)) | Heart disease risk factors: null (3/3) Blood pressure: null (3/3) Adverse birth outcomes: null (1/6); positive (5/6) Cancer/tumor incidence: conditional on context of discrimination (1/2); positive (1/2) Weight change: positive (1/2); negative (1/2) Other health outcomes: ACM: null (1/1); EBV: positive (1/1); Cold: positive (1/1) |
Busse, D., et al (2017) | Stress | 27 | Experimental: 37% (n = 10) Longitudinal: 7% (n = 2) Cross‐sectional: 56% (n = 15) | Hypothalamic‐pituitary‐adrenal (HPA) axis: salivary and awakening cortisol; dehydroepiandrosterone (DHEA); corticotropin‐releasing hormone | Salivary cortisol: 1/2 positive; 1/2 negative Cortisol awakening response: 1/1 positive Null: corticotropin‐releasing hormone; afternoon DHEA |
Dolezsar, CM., et al (2014) | Hypertension | 44 | Cross‐sectional, longitudinal, experimental designs | Hypertensive status; blood pressure | Positive: hypertensive status ( = 0.05; 95% CI: 0.01, 0.09), nighttime ambulatory blood pressure ( = 0.15; 95% CI: 0.04, 0.19) Null: blood pressure (systolic: = 0.01; 95% CI: −0.01, 0.03) (diastolic: = 0.2; 95% CI: −0.01, 0.03) |
Korous, KM., et al (2017) | Cortisol | 16 | Experimental: 25% (n = 4) Nonexperimental: 75% (n = 12) | Current cortisol, diurnal cortisol, cortisol reactivity, average cortisol | Positive ( = .040; 95% CI: 0.038‐0.117) |
Lewis, TT., et al (2014) | Cardiovascular health | 38 | 26 cross‐sectional 12 longitudinal/cohort or unspecified | Lifestyle factors (eg, smoking, physical activity, alcohol intake); hypertension and blood pressure; biomeasures (eg, obesity, C‐reactive protein, coronary artery occlusion) | Conditional: lifestyle factors; resting blood pressure/hypertension; biomeasures Positive: ambulatory blood pressure |
Lockwood, KG., et al (2018) | Cardiovascular health | 21 | Cross‐sectional, longitudinal, experimental designs | Cardiovascular reactivity (ie, blood pressure, heart rate, heart rate variability, total peripheral resistance, preejection period, cardiac output); HPA axis (ie, diurnal cortisol slope); immune (ie, C‐reactive protein, interleukin, monocyte chemoattractant protein, tumor necrosis factor, interferon), neural activity | Generally positive associations for CVD reactivity, flatter diurnal cortisol slopes, systemic inflammation, and neural activity in the brain regions consistent with exposure to psychosocial stress |
Health behaviors | |||||
Desalu, JM., et al (2019) | Alcohol use | 27 | Cross‐sectional: 85% Longitudinal: 15% | Consumption; binge/heavy drinking; at‐risk drinking; alcohol use disorders (AUD); negative drinking consequences | Positive: consumption ( = .12; 95% CI: 0.08, 0.17); binge drinking ( = .06; 95% CI: 0.02, 0.10); at‐risk drinking ( = .14; 95% CI: 0.06, 0.23); negative drinking consequences ( = .25; 95% CI: 0.09, 0.42) Null: AUD ( = .10; 95% CI: −0.01, 0.20) |
Gilbert, PA., et al (2016) | Alcohol use | 97 | Cross‐sectional: 80% Longitudinal: 18% Experimental: 2% | Alcohol‐related outcomes (number of drinks per month, past 2 weeks of binge drinking, past week/ 30 days/ year of alcohol use, past 2 months of weekend drinking, drinking‐related problems, past‐year alcohol use, past 30/90 days of binge drinking, drinking debut, alcohol use disorder, lifetime alcohol use, hazardous drinking, current alcohol use, alcohol use disorder) | Positive: 45% (n = 14) Null: 32% (n = 10) Conditional: 23% (n = 7) |
Slopen, N., et al (2016) | Sleep | 17 | Longitudinal: 29% (n = 5 [1 daily diary]) Cross‐sectional: 71% (n = 12) | Poor sleep outcomes (ie, duration, efficiency, sleep latency, wake after sleep onset, REM sleep, light sleep, stage 3 and 4 sleep) | Positive: sleep difficulties or insomnia (16/16; 100%); poor sleep quality (7/7; 100%) |
Health care utilization | |||||
Ben, J., et al (2017) | Health care utilization | Review: 83 Meta‐analysis: 59 | Cross‐sectional: 96.4% Longitudinal: 3.6% | Health service experiences [HSE] (ie, communication; satisfaction/perceived quality of care; trust; some combination of these) Health service utilization [HSU] (ie, having examinations, screenings, checks, etc; uptake of treatments, medications, vaccinations; hospital visits and admissions to ERs; delaying health care; insurance coverage; some combination of these) | HSE: negative HSU: conditional on outcome, negatively associated with uptake of treatments and seeking health care; no association for the other measures |
Gaston, GB., et al (2013) | HIV treatment adherence | 16 | Qualitative | Antiretroviral medication or medical self‐care adherence | Discrimination serves as a barrier to medical care, poorer self‐rated health, lower self‐care adherence, less satisfaction with care, greater depressive symptoms |
Children and adolescents | |||||
Alhusen, JL., et al (2016) | Maternal and child health | 15 | Qualitative: 27% (n = 4) Quantitative: 73% (n = 11) | Preterm birth; low birth weight; small‐for‐gestational‐age newborn; access to and quality of prenatal care | Preterm birth (quant studies: 5): 3/5 null; 2/5 positive Low birth weight (quant studies: 3): 2/3 positive; 1/3 null Small‐for‐gestational‐age: positive Initiation of prenatal care (quant studies:1): null |
Benner AD. et al (2018) | Socioemotional, academic, and behavioral health | 214 | Cross‐sectional Longitudinal | Socioemotional well‐being (depression, internalizing symptoms, positive well‐being, self‐esteem); academic (achievement, school engagement, motivation); behavioral (externalizing behaviors, risky sex behaviors, substance use, deviant peer affiliations) | Racial discrimination was positively associated with depression, internalizing symptoms, externalizing behaviors, risky sex behaviors, substance use, deviant peer affiliations and negatively associated with self‐esteem, academic achievement, school engagement, academic motivation |
Heard Garris NJ et al (2018) | Child health (infant health outcomes, mental health, socioemotional health, health care utilization, physical health, cognitive development, and youth health) | 30 | Case‐control (10%), cross‐sectional (27%), and longitudinal (53%) | Infant health outcomes (preterm birth, cortisol reactivity, birthweight); mental health (depressive symptoms, anxiety, substance use, well‐being, anxiety); socioemotional health (externalizing and internalizing behavior, socioemotional difficulties, self‐esteem, positive behavior); health care utilization (frequency of sick child visits); physical health (BMI, general child illness, weight for age); cognitive development (spatial ability); youth health outcomes (depressive) | Caregiver racial discrimination is associated with preterm birth in 4/7 studies, cortisol reactivity in 1/1 study, and birthweight in 6/9 studies observing child outcomes. Within postbirth, caregiver pathway: caregiver racial discrimination is associated with depressive symptoms in 1/7 studies, anxiety in 1/3, substance use in 1/2, well‐being in 1/1, depressive symptoms in 1/7, externalizing in 7/10, internalizing behavior in 4/7, socioemotional difficulties in 2/2, self‐esteem in 1/1, positive behavior in 1/4, frequency of sick child visits in 1/2, BMI in 1/1, general child illness in 1/2, weight for age in 1/1, and spatial cognitive ability in 1/1 in child outcomes; within postbirth, other pathway: caregiver racial discrimination was associated with depressive symptoms in 1/2 studies in child outcomes |
Priest N. et al (2013) | Mental and physical health (negative and positive mental health, negative and positive general health, physical health, negative and positive pregnancy, behavior problems, well‐being, health‐related behaviors, health care utilization) | 121 | 2% Case‐control, 78% Cross‐sectional, and 20% longitudinal | Negative mental health (anxiety, depression, distress, hopelessness, loneliness, negative self‐esteem, posttraumatic stress, psychological distress, social and emotional difficulties, somatic symptoms, stress, suicide, mental health problems); positive mental health (emotional adjustment, psychological adjustment, psychological adaption, resilience, self‐esteem, self‐worth, social and adaptive functioning); physical health (blood pressure, childhood illnesses, common childhood illnesses, insulin resistance, obesity, physical symptoms); general health; negative general health (feeling unhappy, feeling unhealthy, health problems); positive general health (self‐rated health); well‐being (general health, HrQoL, life satisfaction, well‐being); negative pregnancy (LBW, preterm birth, VLBW); positive pregnancy (birth weight, gestational age); behavior problems (ADHD, aggression, behavior problems, conduct problems, delinquent behavior, deviance, emotional and behavioral problems, externalizing, internalizing, problem behavior; health‐related behavior (alcohol, drug use, smoking); health care utilization (access and cost) | Of the 121 studies and 461 associations, 46% of associations were negatively associated with reported racial discrimination, 18% were positive and 3% were conditional. 76% of the associations between racial discrimination and negative mental health outcomes were positive. 62% of the associations between racial discrimination and negative mental health outcomes were negative. 69% of the associations between racial discrimination and behavior problems/delinquent behaviors were positive. 51% of the associations between racial discrimination and health‐related behaviors were positive. 45% of the associations between racial discrimination and well‐being/life satisfaction/quality of life outcomes were negative and 50% was unrelated. 79% of the associations between racial discrimination and negative pregnancy/birth outcomes were positive. 67% of the associations between racial discrimination and physical health had no significant associations. Additionally, mental health was the most studied association among the 121 studies and 461 relationships (51% of associations were mental health‐related) |
4.1. Discrimination and mental health
A 2015 meta‐analysis by Paradies and colleagues 16 found over 300 articles on racial discrimination and health published through 2013, with the association between discrimination and mental health stronger than for physical health. Although 8 out of every 10 studies came from the United States, there were publications from 19 other countries. Discrimination was significantly associated with poorer mental health outcomes (eg, depression, anxiety, psychological stress, r = −.23) and positive mental health outcomes (eg, self‐esteem, life satisfaction, control, well‐being, r = −.13). The meta‐analysis found that the effect sizes for the association between perceived discrimination and mental health were stronger in cross‐sectional studies than in longitudinal ones and in nonrepresentative samples than in representative ones.
A meta‐analysis of 51 studies in Europe highlights growing international evidence. Across diverse ethnic populations, positive associations were found between ethnic discrimination and emotional distress, as well as inverse associations with positive markers of well‐being, such as self‐esteem and self‐efficacy. 20 Several recent reviews continue to document an inverse association between discrimination and good mental health. 21 , 22 , 23 , 24 , 25 , 26 , 27 For example, a 2014 review reported the results of two meta‐analyses focused on the association between discrimination and well‐being. 28 Discrimination, in the first meta‐analysis, was associated with poorer well‐being (self‐esteem, depressive and anxiety symptoms, psychological distress, and life satisfaction), with the association being somewhat weaker for positive outcomes than negative ones. The observed associations (effect sizes) were larger for disadvantaged groups compared to advantaged groups (eg, women vs men) and for children than for adults. They were also evident in both cross‐sectional and longitudinal analyses. In the second meta‐analysis, the researchers examined experimental data for studies relating the manipulation of discrimination to indicators of well‐being. The study found a significant negative effect ( d = −0.25) of multiple exposures to discrimination on well‐being. A single event of discrimination was not adversely related to well‐being. Research also indicates that exposure to discrimination can adversely affect the personality characteristics of adults. Longitudinal analyses in two national studies, the Health and Retirement Survey and the Midlife in the United States Study (MIDUS), found that incident discrimination was associated with increases in neuroticism (negative emotions) and declines in agreeableness (trusting) and in conscientiousness (organization and discipline). 29
One review documented that in addition to discrimination being positively associated with measures of depression, anxiety symptoms, and psychological distress, it is also associated with increased risk of defined psychiatric disorders. 18 For example, in the National Study of American Life (NSAL), among African American and Caribbean Black adults 55 years and older, both racial and nonracial chronic Everyday Discrimination was positively associated with increased risk of any lifetime (LT) disorder, as well as LT mood and anxiety disorders. 30 It was also associated with an increased risk of depressive symptoms and serious psychological distress. Similarly, in the National Latino and Asian American Study (NLAAS), Everyday Discrimination was associated with an increased risk of psychiatric disorders, but the association was stronger among Mexicans than for Puerto Ricans. 31 In the same study, Everyday Discrimination was associated, in multivariate models, with increased odds of any DSM‐IV disorder (odds ratio [OR] = 1.90), depressive disorder (OR = 1.72), and anxiety disorder (OR = 2.24) among Asian Americans. 32 Another review documented a positive association between discrimination and PTSD or other indicators of trauma in 70 percent of the associations examined. 33
Research also reveals that the accumulation of experiences of discrimination over time is associated with an increased risk of mental health problems. For example, in the Study of Women Across the Nation (SWAN), the levels of Everyday Discrimination were assessed six times over 10 years. 34 It found that women who experienced the highest accumulation of experiences of discrimination over time, domains, and attributes (race/ethnicity, sex, or other) reported the highest levels of depressive symptoms. This pattern was evident for all women (black, Chinese, Hispanic, and white), regardless of their race or ethnic group. Similarly, a study in the United Kingdom examined the cumulative, longitudinal effects of racial discrimination on mental health of ethnic minorities. 35 The study found evidence of a dose‐response relationship between the cumulative discrimination measure (number of experiences and number of time points exposed) and a scale of nonspecific psychological distress.
Most of the early studies of discrimination were cross‐sectional. In addition, the extent to which observed associations between discrimination and mental health outcomes were due to unmeasured psychological factors remained unclear. These concerns have been addressed in recent research. 18 Although the majority of studies of discrimination and health are still cross‐sectional, there are a growing number of prospective studies that link changes over time in discrimination to increases in symptoms of distress and depression. One review of 25 daily diary, longitudinal studies found that over 90 percent of the time, discriminatory events on a given day were associated with increased symptoms of distress. 36 A few studies have also documented that the association between discrimination and mental health remains robust after adjustment for potential psychological confounders such as neuroticism, social desirability, hostility, and negative affect. 18
4.2. Discrimination and physical health
In the Paradies meta‐analysis model, racial discrimination was significantly associated with poorer general health ( r = −.13) and poorer physical health ( r = −.09). 16 Research also reveals that discrimination is associated with multiple indicators of adverse cardiovascular disease (CVD) outcomes and risk factors of CVD. A 2014 paper 37 reviewed the research on self‐reported discrimination and CVD published between 2011 and 2013. It found that most studies focused on hypertension, smoking, and other health behaviors, with few studies on cardiovascular endpoints. However, one study documented that self‐reported discrimination was associated with more severe coronary artery obstruction among veterans undergoing cardiac catherization, for blacks but not whites. 38 A review of discrimination and physical health among black women found few significant associations for indicators of CVD, highlighting the need to better understand the conditions under which the stress of discrimination has adverse health effects. 39
A 2017 review of 10 longitudinal studies found evidence of a consistent association between self‐reported discrimination and body mass index (BMI), waist circumference, and incidence of obesity. 40 The associations between experiences of discrimination and adiposity were predominantly linear, and racial discrimination was also significantly associated with changes in BMI and waist circumference among women, but not men. Nonetheless, racial discrimination was significantly associated with the incidence of obesity overall.
Research has also focused on some of the specific pathways that may link exposure to discrimination to changes in health status. A meta‐analysis of discrimination and cortisol output found a small positive association. 41 Another review of 21 studies of discrimination and the HPA axis found that discrimination has both positive and negative associations with salivary cortisol. 42 An additional review of 21 studies focused on multisystem responses to discrimination and found strong consistent associations between discrimination and CVD and HPA axis reactivity, but less consistent associations for immune responses. 43
Another subclinical indicator of heart disease that has been examined in relationship to discrimination is intima‐media thickness (IMT). An early study found that discrimination was positively associated with IMT. 44 Recent analyses of data from the SWAN study assessed everyday discrimination six times over 10 years and assessed its relationship with intima‐media thickness. 45 It found that the average levels of discrimination in years 0, 1, 2, 3, 7, and 10 were associated with higher IMT levels at year 12. The association was significant only for white women and not for black, Hispanic, and Chinese women, even though black and Chinese women reported higher levels of discrimination than whites. There is a need to better understand which indicators of discrimination will be predictive of specific health outcomes, for particular population subgroups.
From the earliest studies of discrimination, there has been an increasing interest in the association between discrimination and blood pressure. A recent comprehensive review and meta‐analysis of the association between self‐reported discrimination and hypertension identified 44 studies. 46 It found a small, significant association between perceived discrimination and hypertension. Larger effect sizes observed were between perceived discrimination and nighttime ambulatory systolic (SBP) and diastolic blood pressure (DBP), especially among blacks. Prior research had found that African Americans are more likely than whites to manifest a blunted blood pressure decline during sleep, a pattern that is predictive of an increased risk for cardiovascular mortality and other outcomes. This review indicated that exposure to discrimination contributes to the decrease in blood pressure dipping during sleep, which results in elevated levels of nighttime blood pressure among blacks. It is currently not clear if the association between discrimination and SBP and DBP is independent of its association with obesity. In the SWAN study, exposure to Everyday Discrimination predicted increases in SBP and DBP over 10 years of follow‐up, even after adjusting for known sociodemographic, behavioral, and medical risk factors. However, consistent across multiple racial groups, when a measure of adiposity (either waist circumference or BMI) was added to the model, the association was no longer significant. 47
Several recent studies have examined the association between discrimination and inflammation. Among African Americans in the MIDUS study, experiences of discrimination were associated with increased emotional dysregulation (venting and denial) and with increased biological dysregulation, as measured by increases in three indicators of inflammation (interleukin‐6, e‐selectin, and c‐reactive protein). 48 Another recent study found that lifetime discrimination but not chronic everyday discrimination was associated with increased risk of four markers of inflammation in multivariate models. 49 Another recent article on discrimination and inflammation found that the associations varied by gender and the indicator of inflammation. 50
These findings highlight the need to better understand how the different types of discrimination combine to affect health.
Recent analyses have also examined discrimination in relationship to other indicators of biological functioning. Allostatic load (AL) is a measure of multisystem dysregulation. In the MIDUS study, this index sums 24 indicators of risk scores across seven physiological systems. 51 Analyses of data from African Americans in the MIDUS study found that after adjusting for demographic factors, SES, medication use, cigarette smoking, alcohol use, and mental health symptoms, Everyday Discrimination was associated with higher AL scores. Also, attributions of Everyday Discrimination to race were not more strongly linked to AL than attributions linked to other social statuses. Another recent study has shed light on the pathways that might link discrimination to AL. 52 In this study, African Americans had higher levels of allostatic load (11 indicators of physiological functioning) and discrimination than their white peers. Discrimination was associated with elevated AL scores. However, this association was fully mediated by measures of anger and poor sleep. Another recent study using national data from the HRS linked higher levels of Everyday Discrimination with lower telomere length for blacks but not whites. 53
4.3. Discrimination and health behaviors
Recent reviews indicate that there is a behavioral pathway linking experiences of discrimination to health, with exposure to discrimination predictive of engaging in more high‐risk behaviors and fewer health‐promoting activities. For example, a 2016 systematic review found 97 studies published between 1980 and 2015 that examined the association between discrimination and alcohol use. 54 Most studies focused on African Americans and most found positive associations between increased experiences of discrimination, alcohol consumption, and other drinking‐related problems. The review noted that there was considerable variation in quality across the studies and the need for more longitudinal data collection and the use of representative samples. Similarly, a 2019 meta‐analytic review of 27 studies of African Americans found a positive association between discrimination and alcohol consumption, binge drinking, at‐risk drinking, and negative consequences. 55 Discrimination was unrelated to alcohol use disorder. Earlier reviews found that experiences of discrimination were associated with increased risk of cigarette smoking and drug use. 19 , 56
A 2016 review found 17 studies that examined the association between discrimination and sleep (sleep duration and quality), and every study found at least one positive association between exposure to discrimination and poor sleep. 57 Most studies were cross‐sectional in design (12 of 17); however, three were prospective studies, one was a natural experiment, and one utilized a nine‐day diary component.
4.4. Discrimination and health care
Another pathway linking discrimination to poor health status is the potential of experiences of discrimination to lead to reduced health care‐seeking behaviors and adherence to medical regimens. A recent review and meta‐analysis of studies of racism and health service utilization identified 83 papers for review and 59 papers for meta‐analysis. 58 Major findings included that persons reporting experiences of racial discrimination had two to three times the odds of being less trusting of health care workers and systems, perceiving lower quality of and satisfaction with care, and expressing less satisfaction with patient‐provider communication and relationships. Experiencing racism was also associated with delays in seeking health care and reduced adherence to medical recommendations, although these outcomes were not frequently assessed. Findings related to the use of health services were mixed and mostly not statistically significant. The review also noted important methodological limitations in the research. Many of the measures used to assess discrimination were brief (<25 percent of papers used measures with nine or more items) and over 50 percent of the measures used did not specify a timeframe regarding exposure to racism. A review of 16 qualitative studies examined the role of discrimination in adherence to treatment among persons with HIV. 59 It was found that exposure to discrimination was associated with less adherence to antiretroviral medication, less self‐care, and lower levels of satisfaction with care.
4.5. Discrimination in children and adolescents
Although much of the early research on discrimination and health focused on adult populations, there has been an increasing attention in recent years to the role of discrimination in health outcomes for children and adolescents. A 2013 review identified 121 studies (with 461 outcomes) that examined the association between discrimination and health among persons 0‐18 years old. 60 Indicators of mental health status were the most frequently assessed. Exposure to discrimination was positively associated with symptoms of anxiety and depression, aggression, internalizing behavior, externalizing behavior, and conduct problems. Discrimination was also inversely associated with indicators of positive mental health, such as life satisfaction, resilience, self‐esteem, and quality of life. Consistent with the literature on adults, a positive association was found between discrimination and poor health practices (alcohol use, drug use, and smoking) in 51 percent of 74 tests. Discrimination was also positively related to poor pregnancy or birth‐related outcomes, such as low birth weight and preterm birth. Research also indicates that adolescents experience discrimination in online contexts. One study, for example, found that after adjustment for age, gender, ethnicity, other adolescent stress, and offline discrimination, online discrimination was positively related to depressive symptoms and anxiety symptoms among 14‐ to 18‐year olds. 61
A 2018 meta‐analysis of 214 studies examined racial/ethnic discrimination and adolescent outcomes. 62 It found that there were moderate positive associations between discrimination and multiple indicators of socioemotional distress (eg, depressive symptoms or effects) and internalizing symptoms (eg, anxiety, loneliness, and somatic symptoms). Discrimination was also inversely related to indicators of positive well‐being (eg, life satisfaction, prosocial behaviors, and self‐control), as well as general self‐esteem and self‐worth. The review also included 73 studies that examined the association between discrimination and academic performance. Small‐to‐moderate inverse associations were evident between discrimination and school engagement (eg, attendance), motivation (eg, academic efficacy), and achievement (eg, GPA). This review also documented behavioral pathways among adolescents. There were 71 studies assessing the association between discrimination and risky health behaviors. Small‐to‐moderate positive associations were evident for discrimination with substance abuse, externalizing behaviors (eg, delinquency and anger), affiliation with deviant peers, and risky sexual behaviors (eg, unprotected sex). The analysis also found that for socioemotional distress, associations were stronger for Asian and Latino adolescents compared to African Americans. Another significant moderating effect observed was for the developmental period. Associations with socioemotional distress were stronger in early adolescence (age 10‐13) than late adolescence, and for academics, they were stronger in mid‐adolescence than early adolescence.
A recent study of Latino adolescents illustrates the complex pathways between discrimination and mental health. Using three waves of data, it found that racial/ethnic discrimination predicted increases in symptoms of depression and anxiety. 63 It also found that outward anger expression was a significant mediator, with greater racial/ethnic discrimination associated with more frequent outward anger expression. Anger expression, in turn, was associated with higher levels of anxiety and depression. This study suggests the possibility that prevention and intervention efforts around managing anger could reduce at least some of the negative effects of racial discrimination on Latino youths' mental health.
A few studies have also reported that adverse effects of discrimination experienced as an adolescent are predictive of physical health outcomes in early adulthood. For example, a study of 331 black adolescents from nine rural counties in Georgia found that youth with high and stable perceived racial discrimination at age 16, 17, and 18 had higher levels of multisystem biological dysregulation as measured by stress hormones (cortisol, epinephrine, and norepinephrine), systolic and diastolic blood pressure, inflammation, and weight by age 20. 64 A recent review of 30 longitudinal studies found that vicarious discrimination (ie, experiences of discrimination that occur in the life of adults in a child's social network or others with whom the child identify) can adversely affect the health of the target child both prenatally and postbirth. 65
4.6. Discrimination and disparities in health
Most studies of discrimination and health have not examined the contribution that these exposures make to account for racial disparities in health. However, a few studies in the United States and internationally have documented that perceived discrimination makes an incremental contribution over SES in accounting for racial/ethnic inequities in mental health and self‐reported measures of physical health. This pattern has been evident in community and national studies in the United States, New Zealand, Australia, and South Africa. 56
Recent studies provide further evidence of the role of discrimination in contributing to racial inequities. One study examined SES trajectories over a 33‐year period and their relationship to discrimination and self‐rated health. 66 It found that increased SES for whites is associated with lower reported discrimination. In contrast, for blacks and Hispanics, upward mobility is associated with increased exposure to discrimination compared to their socioeconomically stable peers. Importantly, exposure to discrimination explained a large part of the black/white gap in self‐rated health (but not the Hispanic/white gap). A study in the United Kingdom also assessed the role of discrimination in ethnic inequalities in mental health. 35 In cross‐sectional and longitudinal analyses, they found that adjusting for socioeconomic disadvantage and racial discrimination eliminated ethnic inequalities in mental health for some ethnic groups in the United Kingdom but not for others.
4.7. Individual and collective protective and resilient responses
Figure Figure1 1 also indicates that targets of discrimination are not passive actors but can respond in individual and collective ways to minimize the negative effects of racism. Lewis and colleagues 18 have reviewed the limited evidence pointing to a number of resources that have been shown to cushion at least some of the negative effects of exposure to discrimination on health. For example, prospective analyses in national studies have shown that religious beliefs and behavior can reduce some of the negative effects of discrimination on health. Other evidence reviewed revealed that there is limited evidence that mindfulness (ie, nonjudgmental attention and awareness) can also reduce the negative effects of discrimination on mental health problems, as measured by depressive symptoms. Finally, research also finds emotional support from family, friends, and supportive professionals can also buffer the adverse impacts of exposure to discrimination on health.
There is still much to be learned about the full range of protective factors that can ameliorate the negative effects of discrimination on health and the conditions that maximize the health‐protective effects of such resources. Relatedly, we need a serious and sustained program of research that would guide us in identifying the interventions that enhance civility and respect for stigmatized groups in our society. There is also a serious need for societal interventions to be developed and implemented to reduce and ultimately eliminate societal prejudice and discrimination. Such research is currently in its infancy. 67 We also need more systematic attention to the extent to which efforts that seek to comprehensively address the social determinants of health can reduce exposure to racism and its negative consequences. 68
5. DISCUSSION
This review of research on discrimination and health points to many areas that would benefit from further investigation. Prior reviews indicate that methodological limitations that need to be addressed include the overreliance on cross‐sectional studies and refining the measurement approaches to maximize comprehensiveness and accuracy in the assessment of discrimination. 56 This would require greater attention to capturing the critical stressful dimensions of discriminatory experiences, including the severity, chronicity, and duration of these experiences. There is a need to expand assessment to capture discrimination in multiple domains (eg, race, sex, gender, sexual orientation, stigmatized religious status, and SES), and to extend analyses to assess how exposure in more than one domain relate to each other and combine to affect the adverse impact of discrimination on physical and mental health. 5 Emerging evidence suggests that utilizing an intersectionality framework that examines associations between discrimination and health, with the simultaneous consideration of multiple social categories, leads to larger associations than when only a single social category is considered. 69 Given the increasing evidence of the adverse impacts of discrimination early in life, there is also growing awareness of the need to better understand how discriminatory experiences emerge and accumulate over the life course and combine with other stressful experiences to affect physical and mental health. 70
6. CONCLUSION
This article has provided a glimpse of the growing empirical evidence linking self‐reported experiences of discrimination to health. This area of study is only about three decades old. While there is much that we need to learn and important limitations that need to be addressed, the range of health outcomes associated with discrimination is impressive, and the incidence of multiple populations being affected by discrimination, both domestically and globally, is striking. It is now clear that discrimination is a newly emerging risk factor for a broad range of health outcomes that may make an important contribution to understanding racial and ethnic variations in health and health care utilization. This body of research is a reminder that a broad range of psychosocial factors in homes, neighborhoods, workplaces, and schools can be critical determinants of health, and that improving health and reducing inequities in health will likely require interventions outside of the traditional domains of health policy.
Supporting information
Acknowledgments.
Joint Acknowledgment/Disclosure Statement : Sandra Krumholz for assistance with preparation of the manuscript. Preparation of this manuscript was supported in part by the W.K. Kellogg Foundation.
Williams DR, Lawrence JA, Davis BA, Vu C. Understanding how discrimination can affect health . Health Serv Res . 2019; 54 :1374–1388. 10.1111/1475-6773.13222 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
Racism, racial discrimination, and trauma: a systematic review of the social science literature
- August 2018
- Ethnicity and Health 26(3):1-21
![racial discrimination research paper Katheirne Kirkinis at University at Albany, The State University of New York](https://i1.rgstatic.net/ii/profile.image/276884274335758-1443025763930_Q64/Katheirne-Kirkinis-2.jpg)
- University at Albany, The State University of New York
![racial discrimination research paper Alex L. Pieterse at University at Albany, The State University of New York](https://i1.rgstatic.net/ii/profile.image/444972492169217-1483101117164_Q64/Alex-Pieterse.jpg)
- This person is not on ResearchGate, or hasn't claimed this research yet.
![racial discrimination research paper Alexandra Agiliga at University at Albany, The State University of New York](https://i1.rgstatic.net/ii/profile.image/993377411674116-1613851039928_Q64/Alexandra-Agiliga.jpg)
Abstract and Figures
![racial discrimination research paper . Findings of 18 empirical quantitative studies of reported racial discrimination and various measures of trauma.](https://www.researchgate.net/profile/Alex-Pieterse/publication/327344336/figure/tbl1/AS:672533351247875@1537355854714/Findings-of-18-empirical-quantitative-studies-of-reported-racial-discrimination-and_Q320.jpg)
Discover the world's research
- 25+ million members
- 160+ million publication pages
- 2.3+ billion citations
- Abdul Basid
- Muhamad Syahril
![racial discrimination research paper Moh Zawawi](https://c5.rgstatic.net/m/4671872220764/images/template/default/profile/profile_default_m.jpg)
- ARCH WOMEN MENT HLTH
- Madeline M. Curzon
![racial discrimination research paper Paulo A Graziano](https://i1.rgstatic.net/ii/profile.image/502318152273920-1496773388827_Q64/Paulo-Graziano-2.jpg)
- Emily Arcia
- Noelle L. O’Mara
![racial discrimination research paper Lorne Foster](https://i1.rgstatic.net/ii/profile.image/787841697214467-1564847506712_Q64/Lorne-Foster.jpg)
- Jamile Tellez Lieberman
- Nishita Dsouza
![racial discrimination research paper Carmen Valdez](https://i1.rgstatic.net/ii/profile.image/272685112492037-1442024605584_Q64/Carmen-Valdez-3.jpg)
- Tanya C. Saraiya
![racial discrimination research paper Denise Aimee Hien](https://i1.rgstatic.net/ii/profile.image/272751508324367-1442040435844_Q64/Denise-Hien.jpg)
- Michele Bevilacqua
- Angela Charles
![racial discrimination research paper Saran Nurse](https://i1.rgstatic.net/ii/profile.image/1138276241031168-1648397612497_Q64/Saran-Nurse.jpg)
- Int J Contemp Hospit Manag
- Anna Marie Ruef
![racial discrimination research paper Brett Litz](https://i1.rgstatic.net/ii/profile.image/431252382720014-1479829988616_Q64/Brett-Litz.jpg)
- Traumatology
![racial discrimination research paper Robert T. Carter](https://c5.rgstatic.net/m/4671872220764/images/template/default/profile/profile_default_m.jpg)
- Carrie Muchow
![racial discrimination research paper Alex L. Pieterse](https://i1.rgstatic.net/ii/profile.image/444972492169217-1483101117164_Q64/Alex-Pieterse.jpg)
- Chester M. Pierce
![racial discrimination research paper Alex Pieterse](https://i1.rgstatic.net/ii/profile.image/11431281166575975-1686305619307_Q64/Alex-Pieterse-2.jpg)
- Joe R. Feagin
![racial discrimination research paper Katheryn Roberson-Miranda](https://c5.rgstatic.net/m/4671872220764/images/template/default/profile/profile_default_m.jpg)
- David R. Williams
![racial discrimination research paper Alexander C Tsai](https://i1.rgstatic.net/ii/profile.image/836226399625216-1576383318146_Q64/Alexander-Tsai-2.jpg)
- Tyonna Adams
- Adriana Alejandre
- Anthea Gray
- Recruit researchers
- Join for free
- Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up
Introduction: The Case for Discrimination Research
- Open Access
- First Online: 09 April 2021
Cite this chapter
You have full access to this open access chapter
- Rosita Fibbi 4 ,
- Arnfinn H. Midtbøen 5 &
- Patrick Simon 6
Part of the book series: IMISCOE Research Series ((IMIS))
11k Accesses
Increasing migration-related diversity in Europe has fostered dramatic changes since the 1950s, among them the rise of striking ethno-racial inequalities in employment, housing, health, and a range of other social domains. These ethno-racial disadvantages can be understood as evidence of widespread discrimination; however, scholarly debates reflect striking differences in the conceptualization and measurement of discrimination in the social sciences. Indeed, what discrimination is, as well as how and why it operates, are differently understood and studied by the various scholarships and scientific fields. It is the ambition of this book to summarize how we frame, study, theorize, and aim at combatting ethno-racial discrimination in Europe.
You have full access to this open access chapter, Download chapter PDF
European societies are more ethnically diverse than ever. The increasing migration-related diversity has fostered dramatic changes since the 1950s, among them the rise of striking ethno-racial inequalities in employment, housing, health, and a range of other social domains. The sources of these enduring inequalities have been a subject of controversy for decades. To some scholars, ethno-racial gaps in such outcomes are seen as transitional bumps in the road toward integration, while others view structural racism, ethnic hostility, and subtle forms of outgroup-bias as fundamental causes of persistent ethno-racial inequalities. These ethno-racial disadvantages can be understood as evidence of widespread discrimination; however, scholarly debates reflect striking differences in the conceptualization and measurement of discrimination in the social sciences.
What discrimination is, as well as how and why it operates, are differently understood and studied by the various scholarships and scientific fields. A large body of research has been undertaken over the previous three decades, using a variety of methods – qualitative, quantitative, and experimental. These research efforts have improved our knowledge of the dynamics of discrimination in Europe and beyond. It is the ambition of this book to summarize how we frame, study, theorize, and aim at combatting ethno-racial discrimination in Europe.
1.1 Post-War Immigration and the Ethno-racial Diversity Turn
Even though ethnic and racial diversity has existed to some extent in Europe (through the slave trade, transnational merchants, and colonial troops), the scope of migration-related diversity reached an unprecedented level in the period following World War II. This period coincides with broader processes of decolonization and the beginning of mass migration from non-European countries, be it from former colonies to the former metropoles (from the Caribbean or India and Pakistan to the UK; South-East Asia, North Africa or Sub-Saharan Africa to France) or in the context of labor migration without prior colonial ties (from Turkey to Germany or the Netherlands; Morocco to Belgium or the Netherlands, etc.).
The ethnic and racial diversity in large demographic figures began in the 1960s (Van Mol and de Valk 2016 ). At this time, most labor migrants were coming from other European countries, but figures of non-European migration were beginning to rise: in 1975, 8% of the population in France and the UK had a migration background, half of which originated from a non-European country. By contrast, in 2014, 9.2% of the population of the EU28 had a migration background from outside of Europe (either foreign born or native-born from foreign-born parent(s)), and this share reached almost 16% in Sweden; 14% in the Netherlands, France, and the UK; and between 10 and 13% in Germany, Belgium, and Austria. The intensification of migration, especially from Asia and Africa, has heightened the visibility of ethno-racial diversity in large European metropolises. Almost 50% of inhabitants in Amsterdam and Rotterdam have a “nonwestern allochthon ” background (2014), 40% of Londoners are black or ethnic minorities (2011), while 30% of Berliners (2013) and 43% of Parisians (metropolitan area; 2009) have a migration background. The major facts of this demographic evolution are not only that diversity has reached a point of “super-diversity” (see Vertovec 2007 ; Crul 2016 ) in size and origins, but also that descendants of immigrants (i.e., the second generation) today make up a significant demographic group in most European countries, with the exception of Southern Europe where immigration first boomed in the 2000s.
The coming of age of the second generation has challenged the capacity of different models of integration to fulfill promises of equality, while the socio-cultural cohesion of European societies is changing and has to be revised to include ethnic and racial diversity. Native-born descendants of immigrants are socialized in the country of their parents’ migration and, in most European countries, share the full citizenship of the country where they live and, consequently, the rights attached to it. However, an increasing number of studies show that even the second generation faces disadvantages in education, employment, and housing that cannot be explained by their lack of skills or social capital (Heath and Cheung 2007 ). The transmission of penalties from one generation to the other – and in some cases an even higher level of penalty for the second generation than for the first – cannot be explained solely by the deficiencies in human, social, and cultural capital, as could have been the case for low-skilled labor migrants arriving in the 1960s and 1970s. Indeed, the persistence of ethno-racial disadvantages among citizens who do not differ from others except for their ethnic background, their skin color, or their religious beliefs is a testament to the fact that equality for all is an ambition not yet achieved.
Citizenship status may represent a basis for differential treatment. Undoubtedly, citizenship status is generally considered a legitimate basis for differential treatment, which is therefore not acknowledged as discrimination. Indeed, in many European countries, the divide between nationals and European Union (EU) citizens lost its bearing with the extension of social rights to EU citizens (Koopmans et al. 2012 ). Yet, in other countries, and for non-EU citizens, foreign citizenship status creates barriers to access to social subsidies, health care, specific professions, and pensions or exposure to differential treatment in criminal justice. In most countries, voting rights are conditional to citizenship, and the movement to expand the polity to non-citizens is uneven, at least for elections of representatives at the national parliaments. Notably, in countries with restrictive access to naturalization, citizenship status may provide an effective basis for unequal treatment (Hainmueller and Hangartner 2013 ). The issue of discrimination among nationals, therefore, should not overshadow the enduring citizenship-based inequalities.
The gap between ethnic diversity among the population and scarcity of the representation of this diversity in the economic, political, and cultural elites demonstrate that there are obstacles to minorities entering these positions. This picture varies across countries and social domains. The UK, Belgium, or the Netherlands display a higher proportion of elected politicians with a migration background than France or Germany (Alba and Foner 2015 ). Some would argue that it is only a matter of time before newcomers will take their rank in the queue and access the close ring of power in one or two generations. Others conclude that there is a glass ceiling for ethno-racial minorities, which will prove as efficient as that for women to prevent them from making their way to the top. The exception that proves the rule can be found in sports, where athletes with minority backgrounds are often well represented in high-level competitions. The question is how to narrow the gap in other domains of social life, and what this gap tells us about the structures of inequalities in European societies.
1.2 Talking About Discrimination in Europe
Discrimination is as old as human society. However, the use of the concept in academic research and policy debates in Europe is fairly recent. In the case of differential treatment of ethnic and racial minorities, the concept was typically related to blatant forms of racism and antisemitism, while the more subtle forms of stigmatization, subordination, and exclusion for a long time did not receive much attention as forms of “everyday racism” (Essed 1991 ). The turn from explicit racism to more subtle forms of selection and preference based on ethnicity and race paved the way to current research on discrimination. In European societies, where formal equality is a fundamental principle protected by law, discrimination is rarely observed directly. Contrary to overt racism, which is explicit and easily identified, discrimination is typically a hidden part of decisions, selection processes, and choices that are not explicitly based on ethnic or racial characteristics, even though they produce unfair biases. Discrimination does not have to be intentional and it is often not even a conscious part of human action and interaction. While it is clear that discrimination exists, this form of differential treatment is hard to make visible. The major task of research in the field is thus to provide evidence of the processes and magnitude of discrimination. Beyond the variety of approaches in the different disciplines, however, discrimination researchers tend to agree on the starting point: stereotypes and prejudices are nurturing negative perceptions, more or less explicit, of individuals or groups through processes of ethnicization or racialization, which in turn create biases in decision-making processes and serve as barriers to opportunities for these individuals or groups.
Although the concepts of inequality, discrimination, and racism are sometimes used interchangeably, the concept of discrimination entails specificities in terms of social processes, power relations, and legal frameworks that have opened new perspectives to understand ethnic and racial inequalities. The genealogy of the concept and its diffusion in scientific publications still has to be studied thoroughly, and we searched in major journals to identify broad historical sequences across national contexts. Until the 1980s, the use of the concept of discrimination was not widespread in the media, public opinion, science, or policies. In scientific publications, the dissemination of the concept was already well advanced in the US at the beginning of the twentieth century in the aftermath of the abolition of slavery to describe interracial relations. In Europe, there is a sharp distinction between the UK and continental Europe in this regard. The development of studies referring explicitly to discrimination in the UK has a clear link to the post-colonial migration after World War II and the foundation of ethnic and racial studies in the 1960s. However, the references to discrimination remained quite limited in the scientific literature until the 1990s – even in specialized journals such as Ethnic and Racial Studies , New Community and its follower Journal for Ethnic and Migration Studies , and more recently Ethnicities – when the number of articles containing the term discrimination in their title or keywords increased significantly. In French-speaking journals, references to discrimination were restricted to a small number of feminist journals in the 1970s and became popular in the 1990s and 2000s in mainstream social science journals. The same held true in Germany, with a slight delay in the middle of the 2000s. Since the 2000s, the scientific publications on discrimination have reached new peaks in most European countries.
The year 2000 stands as a turning point in the development of research and public interest in discrimination in continental Europe. This date coincides with the legal recognition of discrimination by the parliament of the EU through a directive “implementing the principle of equal treatment between persons irrespective of racial or ethnic origin,” more commonly called the “Race Equality Directive.” This directive put ethnic and racial discrimination on the political agenda of EU countries. This political decision contributed to changing the legal framework of EU countries, which incorporated non-discrimination as a major reference and transposed most of the terms of the Race Equality Directive into their national legislation. The implementation of the directive was also a milestone in the advent of the awareness of discrimination in Europe. In order to think in terms of discrimination, there should be a principle of equal treatment applied to everyone, regardless of their ethnicity or race. This principle of equal treatment is not new, but it has remained quite formal for a long time. The Race Equality Directive represented a turning point toward a more effective and proactive approach to achieve equality and accrued sensitivity to counter discrimination wherever it takes place.
The first step to mobilize against discrimination is to launch awareness-raising campaigns to create a new consciousness of the existence of ethno-racial disadvantages. The denial of discrimination is indeed a paradoxical consequence of the extension of formal equality in post-war democratic regimes. Since racism is morally condemned and legally prohibited, it is expected that discrimination should not occur and, thus, that racism is incidental. Incidentally, an opinion survey conducted in 2000 for the European Union Monitoring Center on Racism and Xenophobia (which was replaced in 2003 by the Fundamental Rights Agency [FRA]), showed that only 31% of respondents in the EU15 at the time agreed that discrimination should be outlawed. However, the second Eurobarometer explicitly dedicated to studying discrimination in 2007 found that ethnic discrimination was perceived as the most widespread (very or fairly) type of discrimination by 64% of EU citizens (European Commission 2007 ). Almost 10 years later, in 2015, the answers were similar for ethnic discrimination but had increased for all other grounds except gender. Yet, there are large discrepancies between countries, with the Netherlands, Sweden, and France showing the highest levels of consciousness of ethnic discrimination (84%, 84%, and 82%, respectively), whereas awareness is much lower in Poland (31%) and Latvia (32%). In Western Europe, Germany (60%) and Austria (58%) stand out with relatively lower marks (European Commission 2015 ).
These Eurobarometer surveys provide useful information about the knowledge of discrimination and the attitudes of Europeans toward policies against it. However, they focus on the representation of different types of discrimination rather than the personal experience of minority members. To gather statistics on the experience of discrimination is difficult for two reasons: (1) minorities are poorly represented in surveys with relatively small samples in the general population and (2) questions about experiences of discrimination are rarely asked in non-specific surveys. Thanks to the growing interest in discrimination, more surveys are providing direct and indirect variables that are useful in studying the personal experiences of ethno-racial disadvantage.
The European Social Survey, for example, has introduced a question on perceived group discrimination (which is not exactly a personal self-reported experience of discrimination, see Chap. 4 ). In 2007 and 2015, the FRA conducted a specialized survey on discrimination in the 28 EU countries, the Minorities and Discrimination (EU-MIDIS) survey, to fill the gap in the knowledge of the experience of discrimination of ethnic and racial minorities. The information collected is wide ranging; however, only two minority groups were surveyed in each EU country, and the survey is not representative of the population.
Of course, European-wide surveys are not the main statistical sources on discrimination. Administrative statistics, censuses, and social surveys at the national and local levels in numerous countries bring new knowledge of discrimination, either with direct measures when this is the main topic of data collection or more indirectly when they provide information on gaps in employment or education faced by disadvantaged groups. The key point is to be able to identify the relevant population category in relation to discrimination, as we know that ethno-racial groups do not experience discrimination to the same extent. Analyses of immigrants or the second generation as a whole might miss the significant differences between – broadly speaking – European and non-European origins. Or, to put it in a different way, between white and non-white or “visible” minorities. Countries where groups with a European background make up most of the migration-related diversity typically show low levels of discrimination, while countries with high proportions of groups with non-European backgrounds, especially Africans (North and Sub-Saharan), Caribbean people, and South Asians, record dramatic levels of discrimination.
1.3 Who Is Discriminated Against? The Problem with Statistics on Ethnicity and Race
Collecting data on discrimination raises the problem of the identification of minority groups. Migration-related diversity has been designed from the beginning of mass migration based on place of birth of the individuals (foreign born) or their citizenship (foreigners). In countries where citizenship acquisition is limited, citizenship or nationality draws the boundary between “us” and “the others” over generations. This is not the case in countries with more open citizenship regimes where native-born children of immigrants acquire by law the nationality of their country of residence and thus cannot be identified by these variables. If most European countries collect data on foreigners and immigrants, a limited number identify the second generation (i.e., the children of immigrants born in the country of immigration). The question is whether the categories of immigrants and the second generation really reflect the population groups exposed to ethno-racial discrimination. As the grounds of discrimination make clear, nationality or country of birth is not the only characteristic generating biases and disadvantages: ethnicity, race, or color are directly involved. However, if it seems straightforward to define country of birth and citizenship, collecting data on ethnicity, race, or color is complex and, in Europe, highly sensitive.
Indeed, the controversial point is defining population groups by using the same characteristics by which they are discriminated against. This raises ethical, political, legal, and methodological issues. Ethical because the choice to re-use the very categories that convey stereotypes and prejudices at the heart of discrimination entails significant consequences. Political because European countries have adopted a color-blind strategy since 1945, meaning that their political philosophies consider that racial terminologies are producing racism by themselves and should be strictly avoided (depending on the countries, ethnicities receive the same blame). Legal because most European countries interpret the provisions of the European directive on data protection and their transposition in national laws as a legal prohibition. Methodological because there is no standardized format to collect personal information on ethnicity or race and there are several methodological pitfalls commented in the scientific literature. Data on ethnicity per se are collected in censuses to describe national minorities in Eastern Europe, the UK, and Ireland, which are the only Western European countries to produce statistics by ethno-racial categories (Simon 2012 ). The information is collected by self-identification either with an open question about one’s ethnicity or by ticking a box (or several in the case of multiple choices) in a list of categories. None of these questions explicitly mention race: for example, the categories in the UK census refer to “White,” “black British,” or “Asian British” among other items, but the question itself is called the “ethnic group question.”
In the rest of Europe, place of birth and nationality of the parents would be used as proxies for ethnicity in a limited number of countries: Scandinavia, the Netherlands, and Belgium to name a few. Data on second generations can be found in France, Germany, and Switzerland among others in specialized surveys with limitations in size and scope. Moreover, the succession of generations since the arrival of the first migrants will fade groups into invisibility by the third generation. This process is already well advanced in the oldest immigration countries, such as France, Germany, Switzerland, and the Netherlands. Asking questions about the grandparents and the previous generations is not an option since it would require hard decisions to classify those with mixed ancestry (how many ancestors are needed to belong to one category?), not to mention the problems in memory to retrieve all valuable information about the grandparents. This is one of the reasons why traditional immigration countries (USA, Canada, Australia) collect data on ethnicity through self-identification questions.
The discrepancies between official categories and those exposed to discrimination have fostered debates between state members and International Human Rights Organizations – such as the UN Committee for the Elimination of Racial Discrimination (CERD), European Commission against Racism and Intolerance (ECRI) at the Council of Europe, and the EU FRA – which claim that more data are needed on racism and discrimination categorized by ethnicity. The same applies to academia and antiracist NGOs where debates host advocates and opponents to “ethnic statistics.” There is no easy solution, but the accuracy of data for the measurement of discrimination is a strategic issue for both research and policies.
1.4 Discrimination and Integration: Commonalities and Contradictions
How does research on discrimination relate to the broader field of research on immigrant assimilation or integration? On one hand, assimilation/integration and discrimination are closely related both in theory and in empirical studies. Discrimination hinders full participation in society, and the persistence of ethnic penalties across generations contradicts long-term assimilation prospects. On the other hand, both assimilation and integration theory tend to assume that the role of discrimination in shaping access to opportunities will decrease over time. Assimilation is often defined as “the decline of ethnic distinction and its corollary cultural and social difference” (Alba and Nee 2003 , 11), a definition that bears an expectation that migrants and their descendants will over time cease to be viewed as different from the “mainstream population,” reach parity in socioeconomic outcomes, and gradually become “one of us.” In the canonical definition, integration departs from assimilation by considering incorporation as a two-way process. Migrants and ethnic minorities are expected to become full members of a society by adopting core values, norms, and basic cultural codes (e.g., language) from mainstream society, while mainstream society is transformed in return by the participation of migrants and ethnic minorities (Alba et al. 2012 ). The main idea is that convergence rather than differentiation should occur to reach social cohesion, and mastering the cultural codes of mainstream society will alleviate the barriers to resource access, such as education, employment, housing, and rights.
Of course, studies of assimilation and integration do not necessarily ignore that migrants and ethnic minorities face penalties in the course of the process of acculturation and incorporation into mainstream society. In the landmark book, Assimilation in American Life , Milton Gordon clearly spelled out that the elimination of prejudice and discrimination is a key parameter for assimilation to occur; or to use his own terms, that “attitude receptional” and “behavioral receptional” dimensions of assimilation are crucial to complete the process (Gordon 1964 , 81). Yet, ethnic penalties are believed to be mainly determined by human capital and class differences and therefore progressively offset as education level rises, elevating the newcomers to conditions of the natives and reducing the social distance between groups. Stressing the importance of generational progress, assimilation theory thus tends to consider discrimination as merely a short-run phenomenon.
The main blind spots in assimilation and integration theories revolve around two issues: the specific inequalities related to the ethnicization or racialization of non-white minorities and the balance between the responsibilities of the structures of mainstream society and the agencies of migrants and ethnic minorities in the process of incorporation. Along these two dimensions, discrimination research offers a different perspective than what is regularly employed in studies of assimilation and integration.
Discrimination research tends to identify the unfavorable and unfair treatment of individuals or groups based on categorical characteristics and often shows these unfair treatments lie in the activation of stereotypes and prejudices by gatekeepers and the lack of neutrality in processes of selection. In this perspective, what has to be transformed and adapted to change the situation are the structures – the institutions, procedures, bureaucratic routines, etc. – of mainstream society, opening it up to ethnic and racial diversity to enable migrants and ethnic minorities to participate on equal footing with other individuals, independent of their identities. By contrast, in studies of assimilation and integration, explanations of disadvantages are often linked to the lack of human capital and social networks among migrants and ethnic minorities, suggesting that they have to transform themselves to be able to take full part in society. To simplify matters, studies of assimilation and integration often explain persistent disadvantages by pointing to characteristics of migrants and ethnic minorities, while discrimination research explains disadvantages by characteristics of the social and political system.
Both assimilation and integration theories have gradually opened up for including processes of ethnicization and racialization and the consequences of such processes on assimilation prospects. Most prominently, segmented assimilation theory (Portes and Rumbaut 2001 ; Portes and Zhou 1993 ) shifts the focus away from migrants’ adaptation efforts and to the forms of interaction between minority groups – and prominently the second and later generations – and the receiving society. In this variant of assimilation theory, societies are viewed as structurally stratified by class, gender, and race, which powerfully influence the resources and opportunities available to immigrants and their descendants and contribute to shaping alternative paths of incorporation. According to segmented assimilation theory, children of immigrants may end up “ascending into the ranks of a prosperous middle class or join in large numbers the ranks of a racialized, permanently impoverished population at the bottom of society” (Portes et al. 2005 , 1004), the latter outcome echoing worries over persistent ethnic and racial disadvantage. Another possible outcome is upward bicultural mobility (selective acculturation) of the children of poorly educated parents, protected by strong community ties.
The major question arising from these related fields of research – the literature on assimilation and integration, on the one hand, and the literature on discrimination, on the other – is whether the gradual diversification of Europe will result in “mainstream expansion,” in which migrants and their descendants over time will ascend the ladders into the middle and upper classes of the societies they live in, or whether we are witnessing the formation of a permanent underclass along ethnic and racial lines. This book will not provide the ultimate answer to this question. However, by introducing the main concepts, theories, and methods in the field of discrimination, as well as pointing out key research findings, policies that are enacted to combat discrimination, and avenues for future research, we hope to provide the reader with an overview of the field.
1.5 The Content of the Book
The literature on discrimination is flourishing, and it involves a wide range of concepts, theories, methods, and findings. Chapter 2 provides the key concepts in the field. The chapter distinguishes between direct and indirect discrimination as legal and sociological concepts, between systemic and institutional discrimination, and between discrimination as intentional actions, subtle biases, and what might be referred to as the cumulative effects of past discrimination on the present. Chapter 3 reviews the main theoretical explanations of discrimination from a cross-disciplinary perspective. Mirroring the historical development of the field, it presents and discusses theories seeking the cause of prejudice and discrimination at the individual, organizational, and structural levels.
Of course, our knowledge of discrimination depends on the methods of measurement, since the phenomenon is mainly visible through its quantification. Hence, Chapter 4 offers an overview of the strengths and weaknesses of available methods of measurement, including statistical analysis of administrative data, surveys among potential victims and perpetrators, qualitative in-depth studies, legal cases, and experimental approaches to the study of discrimination (including survey experiments, lab experiments, and field experiments).
Importantly, discrimination does not occur similarly in all domains of social life, and it takes different forms according to the domain in question (e.g., the labor market, education, housing, health services, and public services). Chapter 5 taps into the large body of empirical work that can be grouped under the heading “discrimination research” in order to provide some key findings, while simultaneously highlighting a distinction between systems of differentiation and systems of equality.
What happens when discrimination occurs? Chapter 6 addresses the consequences of unfair treatment for targeted individuals and groups, as well as their reaction to it. These individual and collective responses to discrimination are seconded by policies designed to tackle discrimination. However, antidiscrimination policies vary greatly across countries, and Chapter 7 provides an overview of the different types of policies against discrimination in Europe and beyond, both public policies and schemes implemented by organizations. The chapter also reflects on some of the key political and societal debates about the implementation and the future of these policies. Chapter 8 concludes on the future of discrimination research in Europe, stressing the main challenges ahead for a burgeoning scientific field.
Alba, R., & Foner, N. (2015). Strangers no more: Immigration and the challenges of integration in North America and Western Europe . Princeton: Princeton University Press.
Book Google Scholar
Alba, R., & Nee, V. (2003). Remaking the American mainstream: Assimilation and contemporary immigration . Cambridge: Harvard University Press.
Alba, R., Reitz, J. G., & Simon, P. (2012). National Conceptions of assimilation, integration, and cohesion. In M. Crul & J. H. Mollenkopf (Eds.), The changing face of world cities: Young adult children of immigrants in Europe and the United States (pp. 44–61). New York: Russel Sage.
Google Scholar
Crul, M. (2016). Super-diversity vs. assimilation: How complex diversity in majority–minority cities challenges the assumptions of assimilation. Journal of Ethnic and Migration Studies, 42 (1), 54–68. https://doi.org/10.1080/1369183X.2015.1061425 .
Article Google Scholar
Essed, P. (1991). Understanding everyday racism: An interdisciplinary theory . Newbury Park: Sage.
European Commission. (2007). Discrimination in the European Union (Special Eurobarometer, Vol. 263). Brussels: European Commission.
European Commission. (2015). Discrimination in the EU in 2015 (Special Eurobarometer, Vol. 437). Brussels: European Commission.
Gordon, M. (1964). Assimilation in American life: The role of race, religion, and National Origins . New York: Oxford University Press.
Hainmueller, J., & Hangartner, D. (2013). Who gets a swiss passport? A natural experiment in immigrant discrimination. American Political Science Review, 107 (01), 159–187. https://doi.org/10.1017/S0003055412000494 .
Heath, A. F., & Cheung, S. Y. (Eds.). (2007). Unequal chances: Ethnic minorities in Western labour markets . Oxford: British Academy/Oxford University Press.
Koopmans, R., Michalowski, I., & Waibel, S. (2012). Citizenship rights for immigrants. National political processes and cross-national convergence in Western Europe, 1980–2008. American Journal of Sociology, 117 (4), 1202–2045. https://doi.org/10.1086/662707 .
Portes, A., & Rumbaut, R. (Eds.). (2001). Legacies: The story of the immigrant second generation . Los Angeles: University of California Press.
Portes, A., & Zhou, M. (1993). The new second generation: Segmented assimilation and its variants. The Annals of the American Academy of Political and Social Science, 530 , 74–96. https://doi.org/10.1177/0002716293530001006 .
Portes, A., Fernández-Kelly, P., & Haller, W. (2005). Segmented assimilation on the ground: The new second generation in early adulthood. Ethnic and Racial Studies, 28 (6), 1000–1040. https://doi.org/10.1080/01419870500224117 .
Simon, P. (2012). Collecting ethnic statistics in Europe: A review. Ethnic and Racial Studies, 35 (8), 1366–1391. https://doi.org/10.1080/01419870.2011.607507 .
Van Mol, C., & de Valk, H. (2016). Migration and immigrants in Europe: A historical and demographic perspective. In B. Garcés-Mascareñas & R. Penninx (Eds.), Integration processes and policies in Europe (IMISCOE Research Series). Cham: Springer.
Vertovec, S. (2007). Super-diversity and its implications. Ethnic and Racial Studies, 30 (6), 1024–1054. https://doi.org/10.1080/01419870701599465 .
Download references
Author information
Authors and affiliations.
Swiss Forum for Migration and Population Studies, University of Neuchâtel, Neuchatel, Switzerland
Rosita Fibbi
Institute for Social Research, Oslo, Norway
Arnfinn H. Midtbøen
National Institute for Demographic Studies, Paris, France
Patrick Simon
You can also search for this author in PubMed Google Scholar
Rights and permissions
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Reprints and permissions
Copyright information
© 2021 The Author(s)
About this chapter
Fibbi, R., Midtbøen, A.H., Simon, P. (2021). Introduction: The Case for Discrimination Research. In: Migration and Discrimination. IMISCOE Research Series. Springer, Cham. https://doi.org/10.1007/978-3-030-67281-2_1
Download citation
DOI : https://doi.org/10.1007/978-3-030-67281-2_1
Published : 09 April 2021
Publisher Name : Springer, Cham
Print ISBN : 978-3-030-67280-5
Online ISBN : 978-3-030-67281-2
eBook Packages : Social Sciences Social Sciences (R0)
Share this chapter
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
- Publish with us
Policies and ethics
- Find a journal
- Track your research
![](http://serviteca.online/777/templates/cheerup2/res/banner1.gif)
IMAGES
COMMENTS
In defining racial discrimination, many scholars and legal advocates distinguish between differential treatment and disparate impact, creating a two-part definition: Differential treatment occurs when individuals are treated unequally because of their race.
Studies have identified a high prevalence of racist incidents faced by Black Americans, including Lee and colleagues (2019), who found that 69.5% have experienced racial discrimination from time to time or regularly.
As a legacy of African enslavement, structural racism affects both population and individual health in three interrelated domains: redlining and racialized residential segregation, mass ...
Racial discrimination is recognised as a key social determinant of health and driver of racial/ethnic health inequities. Studies have shown that people exposed to racism have poorer health outcomes (particularly for mental health), alongside both reduced access to health care and poorer patient experiences.
Between racial groups, they found prevalence rates of 23.53% for Whites, 31.88% for Blacks, 27.15% for Hispanics, 11.61% for American Indians, 18.72% for Asians, and 26.99% for mixed-race individuals.
Who participates in the psychological research that highlights race, and is the participants’ race predicted by the race of the lead author? To answer these and other questions, we surveyed five decades of publications across three major areas of psychology: cognitive, developmental, and social.
This paper provides an overview of research on self‐reported discrimination and health, as well as health care utilization. It begins by situating research on racial discrimination and health within the larger context of research on racism and health.
Objective: The aim of this paper is to provide a systematic review of studies that examine relationships between racial discrimination and trauma in order to further understand the...
In the present study, we examine the effects of race on crime at the micro-level, highlighting interpersonal discrimination as a race-specific risk factor for crime and cultural adaptations that provide African Americans with resilience.
The turn from explicit racism to more subtle forms of selection and preference based on ethnicity and race paved the way to current research on discrimination. In European societies, where formal equality is a fundamental principle protected by law, discrimination is rarely observed directly.