There is a high level of unmet need among individuals with mental illness (
1,
2). Unmet need refers to a situation in which an individual who is in need of psychiatric treatment does not seek care. Because mental illness is not readily visible, entering psychiatric treatment is generally the primary means by which a person is labeled as mentally ill. This label is associated with a range of culturally agreed-upon beliefs or stereotypes transmitted through socialization (
3). Thus the threat of being associated with those stereotypes (
4) and the fear of stigmatization are reasons for not seeking help (
5–
7).
Stigma is a multidimensional phenomenon. Corrigan and Watson (
8) underlined the theoretical difference between perceived public stigma and anticipated self-stigma. Perceived public stigma refers to the awareness of stereotypes held by the general public about service users (
9), and anticipated self-stigma refers to the application of such stereotypes to oneself, leading to internalized devaluation and disempowerment (
10). However, a person can be aware of stereotypes without concurring with them (
8,
11). Therefore, unlike Vogel and colleagues (
12), we consider perceived public stigma and anticipated self-stigma as two different stigma dimensions.
Perceived public stigma may lead individuals to avoid seeking help, if they expect that others discriminate against and devalue service users (
7). Before people perform a behavior, they adopt the perspective of the generalized other through the process of role taking (
13). Thus, out of fear of negative reactions of others, they adapt their behavior and might not seek help. Empirical studies have shown that perceived stigma is related to a more negative attitude toward help seeking (
14) and to less willingness to seek formal care (
15).
Anticipated self-stigma may also lead people to avoid seeking help because self-stigma affects feelings of self-esteem and self-efficacy (
16–
18). For example, endorsing the stereotype of blame (that individuals with mental illness are responsible for their condition) may trigger feelings of guilt and shame (
19). To avoid these feelings, people choose coping strategies such as secrecy or social withdrawal instead of seeking help (
11). There is empirical evidence that people with a higher level of anticipated self-stigma are more reluctant to seek help (
15,
20,
21).
Research on help seeking examines both formal and informal sources of care. First, it is important to distinguish between types of formal care providers because the impact of stigma depends on the source of care that is considered (
15,
22). The negative association between stigma and formal help seeking is generally stronger for specialist care than for general care. Second, it is crucial to take informal care into account. Obtaining informal support might facilitate entering the formal care sector (
23) and also might encourage treatment adherence. Furthermore, informal support can help prevent relapse after treatment (
24,
25). The benefits of informal support have been acknowledged by Jorm and colleagues (
26) in their model of overlapping waves of action, by Pescosolido (
27,
28) in the network-episode model, and by Milstein and colleagues (
29) in their COPE continuum. For this study, we hypothesized that the negative association between stigma and informal help seeking would be more pronounced for help from friends than from family members. Stigma has less effect on close ties because family members feel obliged to help, following the hierarchy of responsibility (
30).
In addition, research on help seeking should control for certain potential confounders. Women are more predisposed to seek help (
31–
33), whereas men are more inclined to manage emotional problems on their own (
34). Young people are more likely to turn to nonprofessional sources of care (
35), and older people are more likely to choose general care (
36). Furthermore, people with a higher level of education are more likely to seek specialist care (
37). Those who are divorced or separated are more likely than those who are married or cohabiting to rely on professional care (
38). Employment status is also correlated with formal help seeking (
39). Finally, it is important to consider whether respondents report mental health complaints themselves and whether they are familiar with mental health services (
40,
41).
In sum, we expected to find a negative association between stigma and help seeking. However, we hypothesized that perceived public stigma and anticipated self-stigma would have a differential impact on help seeking from various sources of formal and informal care. Furthermore, we hypothesized that the strength of the association would be more pronounced for specialist care than for general care and for help from friends than from family members.
Methods
Sample
This study is based on data from the 2009 survey Stigma in a Global Context–Belgian Mental Health Study, which examined the attitudes of the general public toward mental health care services and people with mental illness. Fully structured, face-to-face, computer-assisted interviews were conducted with a representative sample of the noninstitutionalized adult Belgian population. The interview included questions referring to a person in a randomly chosen hypothetical vignette; the vignettes included unlabeled psychiatric case histories describing symptoms that met
DSM-IV criteria. This vignette technique is often applied in general population research about mental illness (
42–
48). The vignettes were taken from the 1996 U.S. General Social Survey (
49) and were rewritten during a meeting with representatives from teams from all the countries participating in the Stigma in a Global Context survey and a cross-cultural psychiatrist who had been involved in World Health Organization studies (
50).
To define the target population, a multistage cluster sampling design was used based on data from the Belgian National Register. In stage 1, municipalities were weighted according to number of inhabitants. Subsequently, 140 municipalities were randomly selected; selection included the possibility of being selected more than once. In stage 2, a total of 15 respondents were randomly selected in each municipality, resulting in a target sample of 2,100 people. After complete description of the study to potential participants, written informed consent was obtained from 1,166. The response rate was 55.5%, and the cooperation rate, excluding those who were incapable because of such factors as language barriers, illness, or not being at home for the interview, was 66.9% (
51).
For this study, we included only respondents who received the vignettes depicting major depression or schizophrenia. [The two vignettes are available in an online
supplement to this article.] Therefore, one-third of respondents—those who received the asthma vignette—were excluded from the research sample, resulting in a sample of 755 respondents. Furthermore, a poststratification weighting procedure was applied to approximate the cross-classification of the census population count within gender, age, and education. Accordingly, our weighted sample consisted of 728 cases.
Variables
Dependent variables.
To assess help-seeking attitudes, respondents were presented with a list of potential care providers whom the person in the vignette could consult to cope with the situation: general practitioner, psychiatrist, psychologist, family member, and friend. Respondents were asked to rate how important it was for the person in the vignette to consult each type of care provider to deal with the problem. Importance was rated from 1, not important at all, to 10, very important. This measure is comparable to the rating of helpfulness of diverse treatment options, which is often used in vignette studies (
22,
43–
46).
Independent variables.
Perceived public stigma was measured with Link's Perceived Devaluation and Discrimination Scale (
9). Responses are made on a 4-point Likert scale, from totally agree to absolutely disagree. Some items are reverse-scored, so higher values represent more anticipated discrimination and devaluation. In this study, we referred to former service users in general, instead of to someone who has been in a psychiatric hospital. Construct validity has been demonstrated in other studies (
52). In this study, Cronbach’s alpha was .82.
Anticipated self-stigma was derived from the social isolation subscale of Fife and Wright (
53), which has been used in other studies (
15,
54). It consists of five items that assess negative reactions toward oneself if one were to enter psychiatric treatment. Examples are “I would feel ashamed if I had received psychiatric care,” and “Receiving psychiatric care would make me feel useless.” Responses are scored on a 5-point Likert scale, ranging from totally agree to totally disagree. The scores were reversed so that the mean score would represent a higher level of anticipated self-stigma. The internal consistency of the scale is very good (Cronbach’s alpha=.90).
Control variables.
Education was measured as number of years of education (
55). To assess mental health status, the shortened General Health Questionnaire was used (
56). The response categories were coded on a 4-point Likert scale: not at all, not more than usual, a little more than usual, and a lot more than usual. Scores of positive items were first reversed, and all items were summed so that possible scores ranged from 0 to 36. Cronbach’s alpha was .79. To assess contact with mental health services, respondents were asked whether they ever received psychiatric treatment and whether a family member, friend, or acquaintance had received psychiatric treatment. They were assigned to the category representing the most direct form of contact.
Finally, analyses controlled for characteristics of the person in the vignette—gender, type of disorder, and in-group versus out-group status. Belgian nationality was considered in-group status, and Turkish nationality was considered out-group status, because the Turkish community is one of the largest ethnic minority groups in Belgium.
Analysis
This study focused on the association between perceived public stigma and anticipated self-stigma and attitudes toward formal and informal help seeking. Analyses were performed with SPSS Statistics, version 19. Multiple linear regression models were estimated because a separate model was estimated for each type of care provider. The dependent variables had a skewed distribution; we performed a logarithmic transformation to approach a normal distribution and to render the estimations of the ordinary least-squares regression analyses more accurate (
57). Subsequently, all independent variables were entered as a block. We also checked for problems of multicollinearity for the two stigma variables, but variance inflation factor scores showed that this was not the case. Unstandardized coefficients are reported, accompanied by their standard errors and the test statistic values.
Results
Table 1 presents descriptive data for the study population. It should be noted that two-thirds of respondents (N=488, 67%) stated that other people look down on former service users, and almost half of respondents (N=357, 49%) agreed that other people think that receiving psychiatric treatment is a sign of personal failure (perceived public stigma). In regard to anticipated self-stigma, one-quarter of respondents (N=182, 25%) agreed that they would feel ashamed of being in psychiatric treatment; 313 (43%) stated that they would start to doubt themselves after psychiatric treatment.
As shown in
Table 2, perceived public stigma was negatively associated with the rating of importance of informal care providers (family and friends). Respondents who reported a higher level of perceived public stigma were less likely to rate as important help from family (B=–.111, p<.01) and friends (B=–.113, p<.01). Perceived public stigma did not appear to have a significant impact on formal help-seeking options.
Anticipated self-stigma was associated with rating as less important help provided by a general practitioner (B=–.022, p<.05) or a psychiatrist (B=–.032, p<.05). Anticipated self-stigma was not significantly associated with ratings of help provided by a psychologist or ratings of informal help.
Discussion
This study found that two stigma dimensions—anticipated self-stigma and perceived public stigma—had a differential impact on survey respondents’ perceived importance of formal and informal care. Anticipated self-stigma was associated with rating help from medical care providers (general practitioners and psychiatrists) as less important. Perceived public stigma was associated with rating informal help (from family and friends) as less important.
The finding that respondents who had higher levels of anticipated self-stigma had negative attitudes toward help seeking from general practitioners and psychiatrists but not from nonmedical specialists (psychologists) is striking. A comparison of respondents’ attitudes toward general practitioners and psychiatrists showed that the effect of anticipated self-stigma was strongest in regard to help provided by psychiatrists, which is in accordance with findings of Barney and colleagues (
15). This finding is not surprising, because many people have doubts about the quality and effectiveness of specialized psychiatric treatment (
58–
63).
It is remarkable that fear of anticipated discrimination and devaluation by others (perceived public stigma) appeared to prevent respondents from viewing informal care as an important coping strategy. Our finding is in line with results of the few research studies that have found links between stigma and informal help seeking; these studies found that informal help seeking was impaired when study participants feared devaluation (in the sense of being perceived as inferior or incompetent) (
64) and when young participants feared social discrimination (
22). Respondents with higher levels of perceived public stigma were equally likely to give lower importance ratings to help from friends and from family members in our study. These findings and our results call into question Perry’s (
30) argument that stigma is more pronounced with regard to peripheral network ties, compared with core network ties.
Another interesting finding is that the gender and the ethnicity of the person in the vignette had almost no impact on help-seeking attitudes, which strengthens the main effects. Moreover, respondents’ sociodemographic characteristics had a limited effect on help-seeking attitudes, which is a common finding (
65,
66).
The policy implications of these findings are fourfold. First, structural stigma related to the mental health care sector and psychiatrists in particular should be addressed. Second, it is worrying that stigma was associated with consulting a general practitioner. If patients experience barriers to disclosing mental health problems to such a familiar and accessible source of care, then general practitioners should take the initiative and screen patients for psychiatric symptoms, referring them to specialist care if necessary. Thus a gatekeeper system is recommended (
67). Third, it is interesting that anticipated self-stigma did not prevent respondents from acknowledging the importance of psychologists. The role of nonmedical specialist care in the mental health care sector should be better acknowledged. Fourth, some people seemed to fear devaluation and discrimination by their significant others. Therefore, psychoeducation of families and friends of people with mental illness should be encouraged, which would enhance the provision of lay support.
Our findings should be viewed within the limitations of the data and measures. First, our data do not allow a strict causal interpretation of the results. Nevertheless, we partially controlled for the selection hypothesis, which suggests that previous negative experiences in the mental health care sector might trigger stigmatizing attitudes, by considering whether respondents had direct or indirect contact with mental health services. Second, the study examined the association between stigma and help-seeking attitudes, more specifically the perceived importance of seeking help from various care providers. A disadvantage of this approach is that importance ratings might be confounded by the respondent’s level of mental health literacy (
68). A better strategy would have been to study actual help-seeking behavior. However, this was not possible because an analysis of the small number of service users (15.5%) in the study sample would lack statistical power. Nevertheless, a better understanding of help-seeking attitudes of the general public is helpful because it sheds light on the lay referral system (
69). The lay referral system serves as a reference system, reflecting social norms about stigma related to help seeking (
70,
71).
A third limitation is that the stigma measures referred to attitudes toward former service users, and no information was provided to respondents about severity of the psychiatric symptoms. However, it has been found that social rejection associated with having been in psychiatric treatment is independent of symptom severity (
9). People with a relatively minor mental disorder are not more likely to participate in treatment than those with more severe disorders (
72).
Within these limitations, this study extends the research in this area in several ways. First, our study examined the association between stigma and help-seeking attitudes in a sample representative of the general population. Other research has been limited to student samples (
12,
73–
78) or samples in rural areas (
14,
79,
80). Furthermore, structural barriers to seeking mental health care in Belgium are limited because of the high density of health professionals, including mental health professionals, and because access to specialist care is unrestricted and mental health visits to medically trained professionals are covered by public insurance (
81). Second, the study focused attention on stigma dimensions related to being in psychiatric treatment rather than relying on a general scale about professional help seeking, such as the Attitudes Toward Seeking Professional Psychological Help (
82). Third, the study distinguished between different sources of care, including informal care providers, because little is known about the impact of stigma on attitudes toward seeking informal help (
22).
Conclusions
Anticipated self-stigma and perceived public stigma had a differential impact on the perceived importance of formal and informal care. Respondents’ agreement with stereotypes about psychiatric treatment was negatively related to their perceptions of the importance of seeking help from general practitioners and psychiatrists. Respondents’ awareness that other people hold stereotypes about service users deterred them from acknowledging the importance of seeking informal care.
Acknowledgments and disclosures
This project was supported by grant 3G028709N from the Research Foundation Flanders and by grants 01B03308W and O1J15611T from the Special Research Fund of Ghent University. The authors thank Bernice Pescosolido, Ph.D., for the opportunity to collaborate on this project.
The authors report no competing interests.