The experience of mental illness–related discrimination can negatively affect a number of important areas, including close relationships, employment, and education or training (
1,
2). As a consequence, experienced discrimination may lead to poverty and social marginalization and can have a negative impact on help seeking, service use, and treatment outcome (
2–
4). However, mental illness may not be the only characteristic linked to discrimination against individuals who use mental health services (
4). People with mental health problems who belong to other disenfranchised and marginalized groups, such as racial-ethnic minority groups, are likely to experience discrimination on the basis of both characteristics—that is, dual or multiple discrimination (
5).
Although the problems associated with belonging to more than one marginalized or stigmatized category have been noted (
4), few studies have investigated the effects of the dual statuses of mental illness and race-ethnicity in mental health populations (
4,
6,
7). Two studies in the United States of individuals with serious mental illness found that just over half of participants (53% and 52%) reported experiencing discrimination and that most attributed such experiences to their mental illness (73%) (
4,
6). In one of these studies, the black and Asian groups attributed higher rates of discrimination to their race, compared with the Latino and white groups (
6). A subsequent U.S. study also found that belonging to more than one marginalized social group (for example, being black and female as well as having a diagnosis of mental illness) influenced the attributions given for perceived discrimination; 74% of the sample attributed discrimination to their mental illness and 36% to their race-ethnicity (
7). Black participants were more likely than their white counterparts to attribute discrimination to their race. However, there were no differences between the black and white groups in attributions to mental illness (
7).
Acute discrimination can be distinguished from chronic discrimination in that the former refers to major experiences of unfair treatment in domains such as employment, education, and housing over the life course, whereas chronic discrimination refers to everyday experiences, such as being treated with less courtesy or respect than others (
8–
10). In a U.S. community sample, the black group reported higher levels of psychological distress and more frequent experiences of both lifetime and everyday discrimination, compared with the white group (
9). The black group also reported unfair treatment in a greater number of life events in the previous 12 months. Similarly, black groups in a South African community sample reported higher levels of psychological distress than white groups, and perceived everyday discrimination was associated with distress (
10). Black groups were also two to four times more likely to report both lifetime and everyday discrimination on the basis of race-ethnicity.
Other quantitative studies in the United Kingdom of individuals with severe mental illness have found that black patients were as likely as white patients to experience lifetime discrimination but more likely to attribute these experiences to racism (
11). A U.K. qualitative study also reported findings suggesting racial-ethnic differences in attributions for discrimination; patients in the black group were more likely to attribute unfair treatment to their race, and the white group was more likely to cite their mental illness (
12). However, the evidence presented in these studies (
11,
12) in regard to experienced discrimination and the attributions given appear to be inconsistent. Thus it is important for future research assessing discrimination on the basis of mental illness and race-ethnicity to understand how people with mental health problems explain these experiences in their attributions. No previous study has assessed these issues in a U.K. sample of people engaging with mental health services. Furthermore, the literature describing mental illness–based discrimination (
1,
2,
13) is growing, and this study adds to the literature by examining a range of potential attributions.
The main aims of this study were to describe the prevalence of discrimination across major life domains and in everyday life and the attributions given for these experiences, assess racial-ethnic differences in reported overall experiences of discrimination, and describe the relative occurrence of the two types of attributions by racial-ethnic group.
Methods
This study, the Mental Illness–Related Investigations on Discrimination study, was a cross-sectional study of 202 individuals using secondary mental health services in South London. These services provide specialist care in community settings for adults with more severe mental health problems. Data were collected between September 2011 and October 2012.
Inclusion and exclusion criteria
Eligibility criteria for participants were as follows: at least 18 years of age; a clinical diagnosis of depression, bipolar disorder, or schizophrenia spectrum disorder; self-defined black, white, or mixed race-ethnicity; current treatment with a community mental health team; sufficiently fluent in English to provide informed consent; and well enough to participate. The selected diagnostic groups represent the most prevalent diagnoses received by users of secondary mental health services. White ethnicity was self-defined as “white British,” “white Irish,” or “white other,” and black ethnicity was self-defined as “black African,” “black Caribbean,” or “black British.” The mixed category included any combination of these ethnicities. We aimed to focus on black and mixed groups because these groups are at particular risk of experiencing aversive pathways to mental health care (for example, via compulsory hospital admission or the criminal justice system) (
14).
Procedure
The study was approved by the East of England/Essex 2 Research Ethics Committee. Participants were recruited from 14 teams. Staff members were approached for their approval to allow patients in their care to participate, and their caseloads were subsequently screened on the basis of the prespecified eligibility criteria. This procedure was strategic, but it is also standard practice to gain access to potential participants and support from clinicians. Patients who were well enough were sent an invitation from the clinical team to participate if interested. It was made clear that nonparticipation had no bearing on care received. A reminder flyer was sent if there was no response within one month.
After complete description of the study to participants, written informed consent was obtained. Participants were interviewed usually over two sessions and received a small payment toward their travel and time. The interview schedule collected demographic and clinical information and contained a range of measures; those relevant to this study are detailed below.
Measures
Two comprehensive measures of experienced discrimination were used (see below). Both were developed for the Detroit Area Study (
9,
15) and have been widely used in subsequent research (
10,
16,
17). Minor adaptations to make the wording relevant for the United Kingdom included addition of the word “sacked,” change of “realtor” to “letting agent or estate agent,” and change of “apartment” to “flat.” Three other areas (mental health care, general medical care, and public transport) were added to permit a broader understanding of experiences and to ensure consistency with other ongoing research, such as the South East London Community Health (SELCoH) study (
18).
Major Experiences of Discrimination Scale.
The Major Experiences of Discrimination Scale assessed discrimination by a count of the number of major experiences of unfair treatment in 12 domains, such as employment and general medical care. For each domain, participants indicated whether discrimination was experienced, when it last occurred, and its lifetime frequency. Participants also made attributions for the experiences from a list of demographic and social categories (for example, race-ethnicity, gender, and mental illness) (
10). Both main and secondary attributions were collected, but the focus was on main attributions.
Everyday Experiences of Discrimination Scale.
The Everyday Experiences of Discrimination Scale assessed the frequency of ten types of everyday discrimination (
10). Response categories ranged from 1, never, to 5, very often, and included experiences such as being treated with less courtesy or respect than other people. Attributions for overall everyday discrimination were also collected. The items are summed to give a total score (possible range 10–50). “Ever experienced”’ represents the “very often” to “almost never” response categories.
Data analysis
Data were analyzed with SPSS, version 20. The prevalence of experienced discrimination and the main attributions given are reported as frequencies, percentages, medians, and interquartile ranges. Mental illness and race-ethnicity main attributions for each major life domain and for overall everyday discrimination are also reported descriptively. Negative binominal regression was the most appropriate model to assess racial-ethnic differences in overall major discrimination; a one-way analysis of variance test was used for overall everyday discrimination. Chi square analyses were also conducted to assess racial-ethnic differences across the major life domains and the types of everyday discrimination.
The other reasons reported were mostly not related to status (for example, the personality or ignorance of another person) and therefore not included in the analysis. No significant demographic differences were found between the black groups. These groups were merged to represent the black group, and the white British and “white other” (including white Irish) groups were merged to represent the white group. This merger fails to recognize the heterogeneity within the groups, which remains a limitation of this research. However, such merging allows for comparisons to be drawn and for examination of overall differences. Main attributions regarding mental illness and race-ethnicity were dichotomized (ever experienced and never experienced), and chi square analyses were conducted to examine the association between racial-ethnic group and discrimination attributed to mental illness or to race-ethnicity.
Results
A total of 4,233 service users were screened, 1,345 (32%) were eligible, and written informed consent was obtained from 207 (15%) eligible participants. There were no differences between eligible consenting and nonconsenting participants in diagnoses, age, gender, and race-ethnicity. Five participants were excluded after the interview because they had either an incorrect diagnosis (N=4) or incomplete data (N=1), which left 202 participants in the final sample. Their sociodemographic and clinical characteristics are shown in
Table 1.
Prevalence of experienced discrimination
Prevalence rates of reported major discrimination and the most recent occurrence (recency) and lifetime frequency, along with main attribution (mental illness or race-ethnicity), are shown in
Table 2. Most participants (88%) reported discrimination in at least one major life domain. The median number of areas of unfair treatment was two (interquartile range=1–4). The most common major domains were mental health care (44%), neighbors (42%), police (33%), employment (fired or sacked) (31%), and general medical care (31%). Among these five domains, participants were most likely to report lifetime frequency of discrimination in mental and general medical care (median=3), with almost half of these experiences occurring within the past year (47% for mental health care and 41% for general medical care).
Table 3 presents the percentages of participants who reported ever experiencing everyday discrimination, with median response categories. Percentages ranged from 85% (“People act as if they are better than you”) to 48% (“You receive poorer services . . .” and “You are followed around in stores”). A large majority (94%, N=189) reported experiencing at least one type of everyday discrimination to some degree.
Racial-ethnic differences in overall experienced discrimination
Table 4 presents the mean number of areas of discrimination for major and everyday discrimination by racial-ethnic group. No significant differences were found between the white, mixed, and black groups in overall major discrimination. However, comparisons of the black groups indicated that the black Caribbean group reported major discrimination at a rate nearly twice that of the black African group (incidence rate ratio=1.99, CI=1.07–3.69). No significant differences were found between racial-ethnic groups in overall everyday discrimination.
Differences by domain and type of everyday discrimination
A larger proportion of participants in the black group reported unfair treatment when applying for loans or mortgages (33%, N=16 of 48), compared with the mixed group (10%, N=1 of 10) and white group (10%, N=8 of 78) (χ2=11.05, df=2, p=.004). In terms of everyday discrimination, a significantly higher proportion of participants in the mixed group reported ever experiencing being followed around in stores (88%, N=15 of 17), compared with the white group (44%, N=47 of 108) and black group (45%, N=34 of 76) (χ2=12.22, df=2, p=.002).
Attributions for experienced discrimination
Table 5 presents the main attributions for major and everyday discrimination for all participants who reported discrimination and for each racial-ethnic group. The most prevalent main attributions for major discrimination were mental illness (57%), race-ethnicity (24%), education or income (20%), and appearance (19%). For main attributions across the 12 life domains, 44% (N=89) of participants reported a single attribution and 47% (N=93) reported multiple attributions. Results showed that 25% (N=51) made dual attributions (that is, attributed the discrimination to two causes) and 22% (N=44) made multiple attributions (three or more causes). The most prevalent main attributions for everyday discrimination were mental illness (39%), appearance (17%), and race-ethnicity (11%).
Mental illness was the most prevalent main attribution for major discrimination for all racial-ethnic groups, followed by race-ethnicity for the mixed and black groups. Mental illness was also the most prevalent main attribution for everyday discrimination, followed by race-ethnicity for the black group.
Attributions to mental illness or race-ethnicity
Mental illness was endorsed more frequently than race-ethnicity for everyday discrimination and in most of the major domains except promotion, neighbors, bank loans and mortgages, and public transport. We examined differences between racial-ethnic groups for the two primary attributions. In terms of mental illness as the main attribution, we found no racial-ethnic differences for major or everyday discrimination. However, when race-ethnicity was the main reason for discrimination endorsed by a racial-ethnic group, the black group was most likely to make this attribution for both major and everyday discrimination (
Table 5).
Discussion
Most of the mental health service users in the sample experienced discrimination in at least one major life domain and in their everyday lives. The prevalence rates were higher than in related U.S. studies (
4,
6,
7), although similar rates of mental illness–related discrimination have been reported in England (
13). The most common major domains of discrimination were mental health care, neighbors, police, employment, and general medical care; participants were most likely to report lifetime frequency of discrimination in mental and general medical care.
This study found no evidence to support previous studies that have reported racial-ethnic differences in overall experiences of discrimination (
9,
10), although we did not have comparison groups from other racial-ethnic groups. The black group did not report discrimination in a greater number of life domains than the other groups; however, the black Caribbean group reported discrimination at a rate significantly higher than the black African group. Reasons for this merit further exploration in a wider context, including generational migration experiences. In this study, participants in the black African group were more likely to be first-generation immigrants, whereas those in the black Caribbean and black British groups were more likely to be second-generation immigrants.
The most common attribution for discrimination was mental illness, which is consistent with similar quantitative research from non-U.K. settings (
4,
6,
7). However, this finding contrasts with results of previous qualitative research, which suggested that white groups were more likely to attribute discrimination to their mental illness (
12). Almost half attributed major discrimination to two or more causes, indicating that dual and multiple attributions are prevalent and that future research on mental health–related discrimination should more frequently employ a multiple attribution framework. When examining race-ethnicity as the main attribution, we found that participants in the black group were most likely to endorse race-ethnicity as a main attribution, which is consistent with previous findings (
6,
7,
10). This is not entirely surprising given the role that culture and race-ethnicity play in an individual’s identity and functioning.
Strengths and limitations
This is the first U.K. quantitative study to assess discrimination on the basis of multiple identities in a sample of individuals engaging with secondary mental health services. The study included more major life domains than included in recent studies (
10) and assessed everyday discrimination, which is an important phenomenon that has not been previously examined in populations of mental health service users. Attributions were given after participants indicated whether they experienced discrimination, which made the questioning less leading than in previous studies (
6). The sample may also be considered representative across the sampling frame because no differences were found between eligible consenting and nonconsenting participants.
However, a limitation of the study is that it was carried out in an area where many racial-ethnic groups reside, and attribution results might differ in areas where there is a lower racial-ethnic density (
19,
20). Furthermore, the numbers within the specific ethnic groups were relatively small. Minority groups such as Asians were not included in this study because of anticipated low prevalence in the target area and our focus on groups most at risk of experiencing adverse pathways (for example, via compulsory hospital admission or the criminal justice system). This study did not investigate how race-ethnicity and diagnosis might combine to influence experienced discrimination. In addition, the study focused on subjective experiences rather than on the legal concept of discrimination. Subjective interpretation of events may be influenced by levels of anxiety or depression (
21). However, because of the cross-sectional nature of the research, we were unable to infer directionality of this effect.
Implications
Because this is, to our knowledge, the first quantitative study conducted in the United Kingdom assessing the multiple attributions for discrimination reported by individuals with mental illness, further research is needed in this area, particularly in clinical samples with diagnoses not included in this study and in community samples. Such data are being collected in the SELCoH study (
18). Further research is also needed in areas of low racial-ethnic density and with other racial-ethnic minority groups. Investigations of discrimination should routinely investigate how individuals attribute their experiences, because discrimination may be attributed to race-ethnicity or other social categorizations instead of, or as well as, being attributed to having a mental illness diagnosis. Interventions should be informed by these alternative or multiple attributions. Other forms of discrimination experienced by people with mental illness, such as discrimination resulting from physical disability, sexual orientation, or weight (
22–
24), should also be explored. Further research is also warranted to investigate the differences between major and everyday discrimination as well as between subjective and objective discrimination.
The high prevalence rates of experienced discrimination suggest a greater need for antidiscrimination interventions, such as campaigns and legislative changes, within the U.K. context. Interventions in the most prevalent domains are warranted to reduce levels of discrimination experienced in these key life areas. In addition, the large proportion of participants reporting everyday discrimination highlights the importance of this phenomenon in the lives of people with mental illness and merits further investigation of measures to counter it. Mental illness was the most prevalent attribution among all racial-ethnic groups, whereas attributions to race-ethnicity were prevalent for the mixed and black groups. Antidiscrimination interventions focusing on discrimination related to mental illness and race-ethnicity may be beneficial for racial-ethnic minority groups (
25,
26).
Conclusions
The study found that mental illness–related discrimination remains a common issue across various racial-ethnic groups and that discrimination based on race-ethnicity was also prevalent for the mixed and black groups. There is a need for targeted antidiscrimination strategies that combine efforts to reduce the experience of discrimination attributed to mental illness and race-ethnicity for racial-ethnic minority groups.
Acknowledgments and disclosures
This independent research was funded by the National Institute for Health Research (NIHR) under its Programme Grants for Applied Research scheme (RP-PG-0606-1053). The views expressed are those of the authors and not necessarily those of the National Health Service (NHS), the NIHR, or the U.K. Department of Health. Dr. Hatch and Dr. Thornicroft are funded through the NIHR Specialist Mental Health Biomedical Research Centre at the Institute of Psychiatry, King’s College London, and the South London and Maudsley NHS Foundation Trust. The authors thank study participants and the 14 community mental health teams in South London and Maudsley NHS Foundation Trust that facilitated recruitment and data collection. They also thank the Mental Health Research Network for assisting with screening of eligible individuals.
The authors report no competing interests