A woman walks into an Alcoholics Anonymous meeting. A man ca

STIGMATIZATION AND SELF-ESTEEM OF PERSONS IN RECOVERY FROM MENTAL ILLNESS: THE ROLE OF PEER SUPPORT

MIEKE VERHAEGHE, PIET BRACKE & KEVIN BRUYNOOGHE

ABSTRACT

Background: Persons with mental health problems often experience stigmat- ization, which can have detrimental consequences for their objective and subjective quality of life. Previous research seeking for elements buffering this negative association focused on coping strategies and revealed that none of the most often used strategies is successful. Aims: This article studies whether peer support among clients can moderate this negative link, and to what extent. Following the buffering hypothesis on stress and social support, it was expected that the association between stigmatization and self-esteem would be less among persons experiencing greater peer support. Methods: This research problem was studied by means of ordinary least squares regression analysis using quantitative data from structured questionnaires com- pleted by 595 clients of rehabilitation centres. Results and Conclusions: The results confi rm that stigmatization is negatively related to self-esteem, while peer support is positively linked with it. Furthermore, they show that peer support moderates the negative association between stigmatization and self-esteem, but not in the expected way. These fi ndings sug- gest that peer support can only have positive outcomes among clients with few stigma experiences, and that stigmatization itself could impede the formation and benefi cial consequences of constructive peer relationships among persons receiving professional mental healthcare.

Key words: stigma, peer support, mental illness

INTRODUCTION

Numerous studies have demonstrated the existence of negative attitudes towards persons with mental health problems (Huxley, 1993a & b; Hayward & Bright, 1997; Crisp et al., 2000; Phelan et al., 2000; Angermeyer & Matschinger, 2005). According to the original labelling perspective on mental illness (Scheff, 1966), the label ‘mentally ill’ activates these negative attitudes, resulting in social rejection of clients of mental health services. Furthermore, the modifi ed labelling theory (Link et al., 1987, 1989) describes how people know that the negative attitudes become relevant once they are labelled, which leads to fear of rejection and devaluation. Leaning on these arguments,

International Journal of Social Psychiatry. Copyright © 2008 SAGE Publications (Los Angeles, London, New Delhi and Singapore) www.sagepublications.com Vol 54(3): 206–218 DOI: 10.1177/0020764008090422

VERHAEGHE ET AL.: STIGMATIZATION AND SELF-ESTEEM OF PERSONS IN RECOVERY 207

several authors revealed that there are detrimental consequences of stigmatization to the self-esteem of labelled persons (e.g. Link et al., 2001; Wright et al., 2000; Hayward et al., 2002). Others, however, disputed this link. One of their counterarguments is that labelled persons could seek coping strategies to deal with stigmatization (e.g. Miller & Major, 2000). A study of Link et al. (1991) showed that three of the most often used coping strategies among persons with mental health problems – secrecy, education and avoidance/withdrawal – are not effective. In this study, we examine whether the same is true for another possible buffer: social support. We are especially interested in whether peer support can modify the assumed negative link between stigmatization and self-esteem. This is of importance for several reasons. First, it contributes to the discussion about the existence of a negative association between both variables, as opinions differ on this topic. Furthermore, it links the literature on the consequences of stigmatization with the general literature on the social stress process and the role of social support. Despite the fact that some authors argue that stigma should be considered as a stressor (e.g. Miller & Major, 2000), and the fact that the role of coping strategies is studied (e.g. Link et al., 1991), the role of social support as a buffer has not yet been studied. Finally, this study also has practical implications. As self-esteem could be affected by stigmatization, it is important to seek for elements impeding this negative relationship.

Self-esteem can be considered as a central component of subjective wellbeing and mental health in general (Arns & Linney, 1993; Rosenfi eld, 1997; Markowitz, 2001), and its enhancement is one of the crucial goals in the rehabilitation of persons with psychological problems (Anthony et al., 1990). Self-esteem theory (Rosenberg et al., 1989) sees self-esteem as a fundamental human motive: all persons strive for self-maintenance or self-enhancement. According to this approach, self-esteem has three sources: refl ected appraisals, self-perceptions and social comparisons (Gecas, 1982, 1989). This means that perceiving that others appreciate and like you, perceiving oneself performing effi cacious actions, and comparing oneself favourably with others, enhances self-esteem.

Stigmatization can affect the self-esteem of persons with mental health problems in a direct and indirect way. Direct rejection can lead to perceptions of being negatively evaluated and these negative appraisals diminish self-esteem. However, more subtle mechanisms can also be at work. Being offi cially labelled as mentally ill leads to expectations of devaluation and discrimination (Link, 1987) and so-called incidental stigma reminds labelled persons of their devalued status (Link & Phelan, 1999). As labelled persons perceive that they belong to a devalued social category, they may devalue themselves, even when no direct rejection takes place. Their expectations of being discriminated against or devalued may enhance feelings of shame or a belief that they are set off from others and thus are very different, and may therefore cause them to re-evaluate and re-conceptualize themselves. Three other ways in which stigmatization could affect self-esteem are unfavourable social or temporal comparisons, the loss of valued roles and relationships, and lowered perceptions of personal control (Camp et al., 2002). Several studies have reported empirical evidence about a negative link between stigmatization and self-esteem (e.g. Rosenfi eld, 1997; Link et al., 2001; Wright et al., 2000; Hayward et al., 2002; Verhaeghe, 2003). On the other hand, several other authors argue that stigmatization and self-esteem are not necessarily related (Crocker, 1999; Camp et al., 2002; Herman & Miall, 1990). Therefore, the fi rst research question of this study is whether stigmatization and self-esteem are (negatively) linked in the sample under study.

One of the several arguments against the negative link between stigmatization and self-esteem is the possible attenuating effect of coping strategies (Miller & Major, 2000; Miller & Kaiser, 2001).

208 INTERNATIONAL JOURNAL OF SOCIAL PSYCHIATRY 54(3)

The general reasoning behind this position is that the stress linked with stigmatization will only be detrimental for wellbeing if persons are not able to deal with it successfully. The effectiveness of these coping strategies, however, is also a subject of discussion (Thoits, 1995; Miller & Major, 2000; Herman, 1993). Link et al. (1991) conducted empirical research to examine the effectiveness of three coping strategies in the case of stigmatization of persons with mental health problems, which are prominent in the literature on stigmatization and are recommended by health professionals to their clients: secrecy, education and withdrawal/avoidance. The results of their study revealed that none of these mechanisms moderates the negative consequences of stigmatization. Another important buffer in the general stress literature – social support – is paid less explicit attention to, when studying the impact of stigmatization of persons with mental health problems on wellbeing.

The general literature on stress and social support reveals that persons experiencing more sup- port will suffer less from the negative effects due to stressors (Cohen & Wills, 1985). Certain types of support are especially important: having people to talk about problems with (appraisal or informational support), and having people who make you feel better about yourself (self-esteem or esteem support). Socio-emotional support – defi ned as assertions or demonstrations of love, caring, esteem, value, empathy, sympathy and/or group-belonging – from signifi cant, or primary, others appears to be the most powerful predictor of reduced psychological distress (Thoits, 1985), because it bolsters one or more aspects of self that have been threatened by objective diffi culties. Furthermore, Thoits (1985) states empathetic understanding as a precondition for emotional support to be effi cacious. According to her, this is most likely to come from socially similar others facing or having faced the same stressors, as similar others have more detailed knowledge of the situation and are more likely to identify and address the problems perceived by the distressed individual. Furthermore, as shared experience teaches persons that they are not the only ones with diffi culties, it reduces self-blame (Rosenfi eld & Wensel, 1997). In mental health services, other clients or peers can be considered as similar others.

Peer support is considered as an important element in the recovery of persons with mental health problems (Anthony, 1993; Davidson et al., 1999). Concerning its specifi c role in the stigma process, it is suggested that peer groups can help to ‘gain insight, support and ideas for action to address their stigma problems’ and to ‘consider new, more effective ways of confronting stigma by sharing their experiences, supporting each other, and rehearsing various ways to handle their stigma encounters’ (Dudley, 2000: 452), but the empirical evidence about its functioning is scarce. Corrigan (2003), for instance, reveals that stigmatization can increase in-group feelings, which improve self-esteem, but meanwhile, he acknowledges that few studies exist about the usefulness of this general reasoning for persons with mental health problems. Link et al. (1991) showed that withdrawal to insiders, which constitutes, together with social isolation, one of the three most often mentioned stigma- coping strategies, is not effi cacious. However, avoidance and withdrawal are taken together in that study, and what Goffman (1963) calls the ‘own’ and ‘wise’ are not distinguished. Therefore, it remains useful to study the particular effect of ‘withdrawal’ to the ‘own’. Based on this previous research, the second research question concerns the link between peer support and self-esteem: does peer support enhance self-esteem among clients in rehabilitation centres? Furthermore, this study focuses on the specifi c relationship between stigmatization, peer support and self-esteem. We are especially interested in whether peer support can moderate the negative link between stigmatization and self-esteem. This constitutes our third research question.

VERHAEGHE ET AL.: STIGMATIZATION AND SELF-ESTEEM OF PERSONS IN RECOVERY 209

METHOD AND DATA

Sample The data were collected as part of a larger sociological study of the determinants of wellbeing of clients and staff members in semi-residential professional mental health organizations in Flanders (Belgium). These services include a large variety of psychotherapeutic, vocational, relaxation and leisure activities and are aimed at day-structuring or rehabilitation of persons with more or less sustained psychological problems. The study covers most of the day activity centres connected to the psychosocial rehabilitation centres (n = 7 out of 9 centres) and a random sample of day activity centres associated with vocational and social service organizations (n = 49 out of 90 organizations). Within each participating agency, the researchers selected a random sample of clients. Size was determined as the daily average number of clients, with a maximum of 30 participants for each centre. Two researchers, who interviewed the clients using structured questionnaires, visited each facility. Due to this approach, retention rates were very high, as we had only a few refusals to participate. In this way, we collected data about 676 clients from 56 centres. However, as not all participants fi lled in the questionnaires completely, the following analyses are based on data from only 595 clients. This sub-sample consists of 380 men and 215 women. Their age varies from 16 to 80, with an average of 44. Their mean length of stay is nearly three years. In Table 1 we provide additional information about these clients and the variables used in the analysis.

Variables The dependent variable self-esteem is measured by means of a Dutch translation of Rosenberg’s self-esteem scale (Brutsaert, 1993; Bruynooghe et al., 2003). This is a frequently used Likert scale containing 10 items with scores from 1 to 5, which we averaged to obtain a total score, with higher levels expressing more self-esteem (α = 0.85). The main independent variables are stigmatization and peer support. A Likert scale that is based on Link et al. (1997) and Fife and Wright (2000) is used to operationalize experienced stigmatization. This scale (α = 0.87) consists of fi ve items, such as ‘Since I come to the centre, some people treat me with less respect’. The scores from 1 to 5 are averaged to obtain a total score, with higher scores expressing more rejection experiences. In the absence of an internationally standardized and widely used scale measuring peer support among persons with mental health problems in semi-residential settings, we used an instrument that was especially designed for this study. It is a measure of socio-emotional support and concerns experiences of appreciation and acceptance by the other clients. It is operationalized by means of a scale consisting of fi ve items, such as ‘The other clients accept me the way I am’, with scores from 1 (fully disagree) to 5 (fully agree), which are averaged to compute a total score (α = 0.67). Furthermore, an indicator of symptoms is included. Current mental health status is measured by the Brief Symptom Inventory (Derogatis, 1993), using the Dutch translation of the parent instrument (SCL-90) by Arrindell and Ettema (1986). The GSI-score is obtained by computing the mean score on 53 items with scores from 0 to 4, with higher scores indicating more symptoms (α = 0.97). We include this measure because some opponents of the labelling theory on mental illness state that stigma experiences are subjective and due to the symptoms, which are manifestations of the mental illness (e.g. Gove, 1970). Their reasoning is that the eventually occurring association between stigmatization and self-esteem could be explained by a negative, pessimistic, biased perception, typical of persons with mental illness. Furthermore, the argument that lowered self-esteem could be a symptom of the illness itself is also used in the discussion of former research linking stigmatization

210 INTERNATIONAL JOURNAL OF SOCIAL PSYCHIATRY 54(3)

with self-esteem (Camp et al., 2002). In response to these critical views, it is necessary to account for psychiatric symptoms. Finally, some background variables are included as controls: gender (men = 1), age (in years), length of stay (in years) and education (primary degree = 1 to college degree = 4). Table 1 provides a summary of the variables.

Analysis Our research problems are analyzed by means of ordinary least squares regression analyses, with self-esteem as a dependent variable. In the fi rst step, we included the background variables, followed by our measure of stigmatization in the second step. Peer support is added to our model in the third step. Finally, to test the buffering hypothesis, we included an interaction term, which is computed as the multiplication of stigmatization and the social support indicator. To prevent problems of multicollinearity, which can take place due to a strong correlation between the variables and the product terms, the variables stigmatization and peer support are centred.

RESULTS

In Table 3, we present the results of our ordinary least squares regression analysis. After we introduced our background variables in the fi rst step, our measure of stigmatization is introduced. The negative coeffi cient shows us that stigmatization is negatively associated with self-esteem (β = 0.208; p = 0.000): clients who experience more rejection by their environment because of their attendance at the centre are characterized by lower self-esteem, as we could expect from our literature review. Therefore, we can answer our fi rst research question about a possible negative link between stigmatization and self-esteem affi rmatively. An important note is that symptoms are controlled for. This means that this result cannot be explained by the link of symptoms with both stigmatization (R = 0.355; p < 0.01) and self-esteem (R = –0.634; p < 0.01; see Table 2). Therefore, we disagree with the argument of some opponents of the labelling theory on mental illness that stigma experiences are subjective and that the link between stigmatization and self-esteem can be completely attributed to the psychopathology of the clients.

In the third step, we introduce our variable concerning peer contact. Table 3 reveals that peer support itself is positively related with self-esteem (β = 0.168; p = 0.000): clients who feel

Table 1 Descriptive sample characteristics

n = 595 Mean SD Minimum Maximum

Gender (men = 1) 0.64 0.48 0.00 1.00 Age 44.21 12.73 16.00 80.00 Education 2.28 0.97 1.00 4.00 Length of stay 2.97 3.52 0.00 26.00 Symptoms 1.04 0.82 0.00 3.55 Stigmatization 2.57 1.19 1.00 5.00 Peer support 3.80 0.67 1.80 5.00 Self-esteem 3.39 0.79 1.30 5.00

VERHAEGHE ET AL.: STIGMATIZATION AND SELF-ESTEEM OF PERSONS IN RECOVERY 211

appreciated and esteemed by their peers in the centre have more positive self-evaluations. This constitutes the answer to our second research question. This fi nding is consistent with the view that perceived support from similar others bolsters the self (Thoits, 1985). In addition, Table 3 shows that the regression coeffi cient of stigmatization diminishes slightly when introducing peer contact in the model. Hence, the negative relation between stigmatization and self-esteem can be partially attributed to the fact that stigma experiences seem to obstruct peer support. Table 2 already revealed that clients who experience more stigmatization receive less peer support (R = 0.362; p < 0.01). Thus, negative outside relationships seem to go together with negative inside ones. A possible explanation could be that socially rejected clients feel ashamed about their attendance at the centre and therefore hesitate to come into contact with peers.

Finally, the buffering hypothesis is tested by introducing an interaction term in the analysis as a fourth step. According to this hypothesis, wellbeing should be higher among highly supported people than among poorly supported ones, but only under conditions of high stress. Under conditions of low stress, no differences should be found (Cohen & Wills, 1985). Applying this reasoning to our third research problem, we should fi nd higher levels of self-esteem among clients receiving more peer support, but only under circumstances of high stigmatization. Table 3 reveals that the interaction term is negative and signifi cant (β = –0.069; p = 0.016), which implies that peer contact does modify the negative link between stigmatization and self-esteem. To answer whether this result supports the buffering hypothesis, we take a closer look at the direction of the interaction effect. In Figure 1, the mean scores on our self-esteem measure are compared among clients with low and high stigma experiences, receiving high or low peer support (the distinction between ‘high’ and ‘low’ is based on the median). This comparison shows us that we can only fi nd differences in self-esteem between highly and lowly supported clients in the case of low stigmatization, revealing that peer support is only effective among persons with low stigma experiences, not among those experiencing a lot of stigmatization. Translated to the general stress literature, this means that our results show that wellbeing only differs between lowly and highly supported persons in the case of low stress. Therefore, we can conclude that our data do not support the hypothesis that peer contact impedes the negative impact of stigmatization on self-esteem. This result will be discussed in the next section.

Table 2 Correlations between variables: Pearson’s correlation coeffi cients

n = 595 Age Education Length of stay

Symptoms Stigmatization Peer support

Self-esteem

Gender (men = 1) 0.040 –0.063 0.099* –0.277** –0.081* 0.047 0.275** Age –0.148** 0.335** –0.137** –0.057 0.081* 0.141** Education –0.015 0.060 0.078 –0.023 –0.037 Length of stay –0.161** –0.050 0.047 0.189** Symptoms 0.355** –0.199** –0.634** Stigmatization –0.362** –0.403** Peer support 0.333** Self-esteem –

** p < 0.01 * p < 0.05

212 INTERNATIONAL JOURNAL OF SOCIAL PSYCHIATRY 54(3)

Ta bl

e 3

T he

li nk

b et

w ee

n st

ig m

at iz

at io

n, p

ee r

su pp

or t

an d

se lf

-e st

ee m

: re

su lt

s of

o rd

in ar

y le

as t

sq ua

re s

re gr

es si

on a

na ly

se s

w it

h se

lf -e

st ee

m a

s a

de pe

nd en

t va

ri ab

le (

n =

59 5)

Unstandardized regression coeffi cient

SE

Standardized regression coeffi cient

p

Unstandardized regression coeffi cient

SE

Standardized regression coeffi cient

p

Unstandardized regression coeffi cient

SE

Standardized regression coeffi cient

p

Unstandardized regression coeffi cient

SE

Standardized regression coeffi cient

p

C on

st an

t 3.

71 5

0. 13

1 0.

00 0

3. 61

3 0.

12 8

0. 00

0 3.

63 0

0. 12

5 0.

00 0

3. 61

1 0.

12 5

0. 00

0 G

en de

r (m

en =

1 )

0. 17

2 0.

05 4

0. 10

4 0.

00 2

0. 17

9 0.

05 2

0. 10

9 0.

00 1

0. 18

0 0.

05 1

0. 10

9 0.

00 0

0. 18

1 0.

05 1

0. 11

0 0.

00 0

A ge

0. 00

2 0.

00 2

0. 03

4 0.

31 7

0. 00

2 0.

00 2

0. 03

3 0.

31 0

0. 00

2 0.

00 2

0. 02

4 0.

45 0

0. 00

1 0.

00 2

0. 02

0 0.

54 1

E du

ca tio

n 0.

00 9

0. 02

6 0.

01 1

0. 72

6 0.

01 9

0. 02

5 0.

02 3

0. 45

5 0.

01 7

0. 02

5 0.

02 0

0. 49

8 0.

01 9

0. 02

5 0.

02 3

0. 43

5 L

en gt

h of

s ta

y 0.

01 6

0. 00

8 0.

07 3

0. 03

1 0.

01 7

0. 00

7 0.

07 4

0. 02

3 0.

01 7

0. 00

7 0.

07 4

0. 02

0 0.

01 8

0. 00

7 0.

08 0

0. 01

3 Sy

m pt

om s

–0 .5

71 0.

03 2

–0 .5

89 0.

00 0

–0 .4

99 0.

03 3

–0 .5

14 0.

00 0

–0 .4

87 0.

03 2

–0 .5

02 0.

00 0

–0 .4

85 0.

03 2

–0 .5

00 0.

00 0

St ig

m at

iz at

io n

–0 .1

38 0.

02 2

–0 .2

08 0.

00 0

–0 .1

01 0.

02 2

–0 .1

52 0.

00 0

–0 .0

99 0.

02 2

–0 .1

49 0.

00 0

Pe er

s up

po rt

0. 19

8 0.

03 8

0. 16

8 0.

00 0

0. 20

0 0.

03 8

0. 16

9 0.

00 0

St ig

m at

iz at

io n

×

pe er

s up

po rt

–0 .0

69 0.

02 9

–0 .0

72 0.

01 6

R ²

0. 42

0 0.

45 8

0. 47

6 0.

48 0

si g

R ² c

ha ng

e 0.

00 0

0. 00

0 0.

00 0

0. 01

6

VERHAEGHE ET AL.: STIGMATIZATION AND SELF-ESTEEM OF PERSONS IN RECOVERY 213

DISCUSSION AND CONCLUSION

Persons with mental health problems are often socially rejected, which can have detrimental consequences for their wellbeing in general and their self-esteem in particular. Can this negative effect of stigma experiences be attenuated? As Link et al. (1991) argued, the issue of coping effectiveness is a matter of no small consequence. If coping works, persons with mental health problems can be trained to use effective coping strategies. The same can be said of the effectiveness of peer support. As self-esteem forms a central component of wellbeing, and as its enhancement can be regarded as a central goal of rehabilitation, we became very interested in examining whether peer contact can moderate the negative link between stigmatization and self-esteem. Using data of 595 clients from 56 rehabilitation centres, we studied the relationship between stigmatization, peer support and self-esteem.

Before turning to the discussion of the central variables, we point out some fi ndings concerning background variables. First, the results revealed that men are higher in self-esteem than women, a fi nding which is consistent in the literature on gender differences in the self-concept (Kling et al., 1999). Furthermore, a strong, negative link between symptoms and self-esteem is found, which is also revealed in other studies (e.g. Rosenberg et al., 1989). When symptoms lead persons to perform worse than before, th

Collepals.com Plagiarism Free Papers

Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.

Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS

Why Hire Collepals.com writers to do your paper?

Quality- We are experienced and have access to ample research materials.

We write plagiarism Free Content

Confidential- We never share or sell your personal information to third parties.

Support-Chat with us today! We are always waiting to answer all your questions.