Show Summary Details
Page of



David L. Penn

, Abigail Judge

, Patrick Jamieson

, Joyce Garczynski

, Michael Hennessy

, and Daniel Romer

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2021. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 24 September 2021

Reducing the stigma of mental illness is among the goals of those seeking to increase the diagnosis, improve the treatment, and enhance the well-being of those with mental disorders in the United States. Both the U.S. Surgeon General's report (U.S. Department of Health and Human Services, 1999) and the first stated goal of President George W. Bush's New Freedom Commission on Mental Health advocate “a national campaign to reduce the stigma of seeking care” (President's New Freedom Commission, 2003, p. 7).

Stigma occurs when a person or group is labeled in a pejoratively categorized way that sets them apart from the majority and, as a result, is treated in ways that mark the person as socially unacceptable. Stigma has serious consequences for individuals with mental illness. Those with severe mental illness (SMI) are less likely to have apartments leased to them (Page, 1995), be given job opportunities (Farina & Felner, 1973; Link & Phelan, 2001), or be provided with adequate health care (Lawrie, 1999) relative to individuals without a mental illness. Furthermore, stigmatization is associated with a lowered quality of life (Mechanic, McAlpine, Rosenfield, & Davis, 1994), reduced self-esteem (Link, Struening, Neese-Todd, Asmussen, & Phelan, 2001; Wright, Gronfein, & Owens, 2000), and increased symptoms and stress (Markowitz, 1998). To manage stigma, individuals with mental illness may avoid others or engage in secrecy (Link, Mirotznik, & Cullen, 1991), strategies that may lead to social isolation (Perlick et al., 2001), which in turn could lower social support and increase the likelihood of relapse. Therefore, stigma poses a significant threat to the recovery of persons with SMI.

The purpose of this chapter is to propose methods for reducing stigmatization of mental illness in adolescents. It begins with a brief summary of community attitudes toward persons with mental illness, including a discussion of a recent nationwide survey of young people. We argue that addressing stigma in adolescence is important for two primary reasons. First, stigma appears to have an adverse effect on the course of mental illness once the person has been diagnosed with a disorder. Second, concerns about stigma may delay the seeking of and continuing in treatment. This section is followed by a discussion of factors that contribute to stigma and methods for reducing it. We conclude by recommending strategies for addressing mental illness stigma in adolescents.


In both Western (Crisp, Gelder, Rix, Meltzer, & Rowlands, 2000) and Eastern societies (Chou & Mak, 1998; Tsang, Tam, Chan, & Cheung, 2003), individuals with serious mental illness, such as schizophrenia and other psychotic disorders, are stigmatized by society (Farina, 1998). The extent of this phenomenon may be increasing (Phelan, Link, Stueve, & Pescosolido, 2000). Stigmatizing attitudes toward persons with SMI have a number of recurring themes: persons with SMI are viewed as dangerous, unpredictable, irresponsible, and childlike (Brockington, Hall, Levings, & Murphy, 1993; Levey & Howells, 1995), and unable to manage their own treatment needs (Pescosolido, Monahan, Link, Stueve, & Kikuzawa, 1999). These attitudes are not only held by those in the community but also may be present among mental health professionals and trainees. Specifically, there is evidence that some mental health professionals, including psychiatrists (Chaplin, 2000; Miller, Shepard, & Magen, 2001), social workers (Dudley, 2000; Minkoff, 1987), general mental health providers (Ryan, Robinson, & Hausmann, 2001; Sartorius, 2002), and medical and mental health graduate students (Hasui, Sakamoto, Sugiura, & Kitamua, 2000; Mukherjee, Fialho, Wijetunge, Checinski, & Surgenor, 2002; Werrbach & DePoy, 1993) may hold stigmatizing beliefs about those with mental disorders. The effects include “discrimination in housing, education and employment, and increased feelings of hopelessness in people with schizophrenia” (Hocking, 2003, p. S47). These findings suggest that efforts to destigmatize mental illness should not be limited to community members, but also should include mental health and medical training professionals.

The tendency to stigmatize individuals with mental illness is not a byproduct of adult experiences but has its roots in childhood. In a recent review of the literature, Wahl (2002) concluded that negative attitudes toward persons with mental illness are evident as early as in third grade. In general, those with mental illness are viewed more negatively and with more fear than are individuals with physical disabilities (Wahl, 2002). Wahl reported that there is evidence that these negative attitudes increase over time, suggesting a longitudinal process in which negative stereotypes become increasingly ingrained, culminating in potentially discriminating behaviors in adulthood.

Direct evidence for the presence of stigmatizing beliefs about peers with mental illness was obtained from a recent national survey of young people (ages 14–22) conducted by the Adolescent Risk Communication Institute of The University of Pennyslvania's Annenberg Public Policy Center (APPC). The National Annenberg Risk Survey of Youth (NARSY) was conducted by telephone in the spring of 2002 with 900 respondents who were selected by means of random digit dialing procedures. The response rate for the survey (52.7%), taking into account those who could not be reached as well as those who refused, was comparable to the rates achieved by the Centers for Disease Control and Prevention (CDC; about 49%) in its national surveys of risk behavior (CDC, 2003; see Romer, 2003, for a description of the survey methods and sample).

Respondents were asked if they were aware of four mental disorders: major depression, bipolar disorder, schizophrenia, and eating disorders. Awareness was highest for eating disorders (89%) and depression (86%), but high levels of awareness were registered for the other disorders as well (81% for schizophrenia and 73% for bipolar disorder). Respondents who were aware of at least one disorder were asked a series of questions concerning two disorders selected at random from among those with which they reported familiarity. In particular, they were asked to imagine someone their age who had the disorder and to indicate whether they thought this person was “more likely, less likely, or about as likely as other people to be” (a) “violent,” (b) “prone to committing suicide,” and (c) “good in school.” The order of the traits was randomized across respondents.

Responses to the three items for each of the disorders indicate (Table 27.1) that large proportions of young people believe that persons with these disorders are different from other people. Over half thought that persons with major depression are more likely to be violent, whereas over 90% said that such persons are more prone to suicide, and about three quarters said they are less likely to be good in school than other people. Although all of the disorders displayed the same pattern, violence was most associated with schizophrenia, and doing badly in school was linked the most with major depression. Eating disorders were somewhat of an exception in that they were less associated with violence than the other disorders.

Table 27.1 Percentage of Respondents Who Said That Three Characteristics Are More, Less, or About as Likely in a Peer with a Mental Disorder as in Other Persons Without the Disorder



Major Depression (%)

Bipolar Disorder




Eating Disorder








As likely
















As likely










Good in School






As likely










Respondents (n)





Data were weighted by age, gender, racial-ethnic background, and region of the country to match national demographic profiles.

There were few differences in these stereotypes by age, gender, or education. African-American and Latino youth were more likely than others to think that the persons with the disorders were not different from others (data not shown).

Correlational analyses using gamma coefficients (to compensate for the skewed distributions) indicated that the three items were highly intercorrelated for each disorder (with “good in school” reverse scored). Coefficient alphas based on the gamma correlation coefficients were .64 for depression, .61 for bipolar disorder, .68 for schizophrenia, and .53 for eating disorders. These findings suggest that each disorder had an underlying stereotype that could guide respondents' reactions to persons with mental disorders. In addition, the stereotypes were correlated across disorders, from a low of .35 to a high of .52. It appears, therefore, that young people hold a general stereotype of mental disorder that includes a heightened risk for suicide and violence as well as the inability to function in an adaptive manner.

The survey also included questions to identify youth with potential symptoms of depression. In particular, respondents were asked if (a) “during the past 12 months have you ever felt so sad or hopeless for 2 weeks or more in a row that you stopped doing your usual activities,” and if (b) “during the past 12 months have you ever seriously considered attempting suicide.” Nearly a quarter of the sample (N = 221) responded affirmatively to one of these questions. However, youth who had experienced these conditions were as likely to report the stereotypes as those who had not.


Stigmatization also plays a role in whether an individual initiates and adheres to treatment. Interestingly, failure to initiate treatment may not only be due to having a mental illness but also result from the stigma attached to seeking help. For example, Ben-Porath (2002) found that a case vignette describing someone with depression was rated most negatively when the target individual sought help for the disorder. This suggests that individuals with a mental illness have to wrestle with the fear of both being stigmatized for having a disorder and for seeking help for it.

There is growing evidence that stigma affects individuals with mental disorders in ways that minimize the likelihood that they will be successfully treated. For example, concerns that others will learn that their child is receiving mental health treatment is a common worry for parents of children between the ages of 5 and 19 (Richardson, 2001). In addition, higher levels of stigma are associated with greater treatment delay among individuals with SMI (Okazaki, 2000), parents of rural children with emotional disorders (Starr, Campbell, & Herrick, 2002), athletes (Ferraro & Rush, 2000), individuals with HIV (or concerns of having HIV) (Chesney & Smith, 1999; Fortenberry et al., 2002), general community participants unselected for presence of mental illness (Cooper, Corrigan, & Watson, 2003), and women with alcohol or drug addictions (Copeland, 1997).

Once in treatment, perceived stigma may be a barrier to medication compliance and treatment continuation (Buck, Baker, Chadwick, & Jacoby, 1997; Pugatch, Bennett, & Patterson, 2002; Sirey et al., 2001a, 2001b). Buck et al. (1997) reported that feelings of being stigmatized were associated with lower adherence to antiepileptic drug treatment. Pugatch et al. (2002) reported similar findings; fears of being stigmatized were associated with poorer adherence to drug regimens among young adults (ages 16–24) who were HIV positive. In a series of studies, Sirey and colleagues found that perceived stigmatization predicted medication adherence and treatment continuation (in older adults only) among individuals with depression. These findings indicate that stigma may impact the course of the illness by interfering with commitment to treatment.

Reluctance to seek psychological or psychiatric treatment is especially relevant to adolescents, because numerous disorders, such as major depression, bipolar disorder, anxiety disorders, anorexia and bulimia, and schizophrenia begin in late adolescence or early adulthood. Such delays in seeking treatment have important prognostic implications for individuals in this age group. Post, Leverich, Xing, and Weiss (2001) found that individuals with a greater number of affective episodes prior to receiving pharmacotherapy had a less favorable prognosis than individuals who began medications after fewer episodes. In schizophrenia, it has been hypothesized that the duration of untreated psychotic episodes prior to illness onset is associated with poorer long-term prognosis (Lieberman et al., 2001; Norman & Malla, 2001). Thus, the role of stigma in delaying treatment may contribute to greater severity of illness among adolescents with a newly diagnosed disorder.


To determine the role of unfavorable stereotypes on help seeking behavior, we examined the extent to which youth who had experienced symptoms of depression or suicidal ideation in the past 12 months had failed to seek help from a variety of sources as a function of negative beliefs about mental health treatment. A revised version of the NARSY conducted a year after the first survey (2003) included questions to examine this question. In this survey, we again identified a large number of respondents (N = 211; 23.4%) who had experienced symptoms of depression or suicidal ideation. Hence, we focused on the stigma associated with this disorder.

We also assessed perceptions of the effectiveness of treatment. We did so under the hypothesis that the stigma associated with mental illness should be reduced if the disorder has been successfully treated (see discussion of interventions below). If stigma continues to be attached to those who have been successfully treated, then persons with symptoms could be inhibited from seeking help. Hence, perceived treatment effectiveness could be an important moderator of stigma's effects on help seeking.

To determine the effects of potential treatment, we asked respondents the following question: “If you had major depression, do you think you could get the help you need from any of the following:” (a) “a doctor,” (b) “your friends,” (c) “your parents,” (d) “a counselor,” (e) “an Internet site,” and (f) “a telephone helpline.”

Actual help-seeking behavior on the part of respondents who had experienced depressed symptoms or suicidal ideation was assessed by asking, “Have you ever done any of the following to try to get help?” Responses were grouped into four categories based on a factor analysis that indicated that help seeking tended to occur in the following ways:

  • Gone to a doctor or nurse; or, taken medication prescribed by a doctor;

  • Gone to a counselor;

  • Talked to a friend; or talked to a parent, and

  • Gone to an Internet site; or used a telephone helpline.

Respondents were scored as having tried any category if they had sought help from any of the sources within the category. In addition, treatment effectiveness scores were created for each category on the basis of ratings of each source.

Perceptions of treatment effectiveness (Table 27.2) indicated that seeing a doctor was viewed as most effective, whereas going to an Internet site or using a telephone helpline were viewed as the least effective. Although the differences are small in magnitude, seeing a doctor was viewed as significantly more effective in the entire sample than seeing a counselor, t(814) = 3.02, p < .01, or talking to a parent, t(814) = 2.53, p < .02. Other differences that are larger in magnitude are even more statistically discernable.

Table 27.2 Perceptions of Efficacy of Sources of Help for Depression by Those With or Those Without Symptoms Associated With Depression

Respondents Who Reported

Help Source

Symptoms of

Depression or


(N = 180)


No Symptoms

of Depression

or Suicidality

(N = 635)





















Telephone helpline




Internet site




Data include only those who were aware of major depression and were weighted to match national demographics; unweighted Ns were 191 for depressed group and 619 for others (total = 810).

With the exception of seeking help from parents, the rank order of perceived effectiveness of the help sources was comparable between those with and without symptoms of depression. Nevertheless, there were differences between the groups. With the exception of one help source (the Internet), persons with symptoms were less likely to perceive any of the help sources as effective. Therefore, perceived treatment ineffectiveness is a source of concern in itself for youth with symptoms of mental disorder (Shaffer et al., 1990).

Analysis of Help Seeking

To determine the role of stigma on help seeking, we conducted regression analyses of the relation between stereotypes of depression and reported help seeking for respondents with symptoms of major depression or suicidality. Stereotypes were defined as the average of three beliefs associated with mental disorder in the first NARSY: violence, suicidality, and good performance in school (reversed scored). The scale ranged from (1) less likely than the average person to have the undesirable trait to (3) more likely than the average person to have the trait. We used probit regression because it is appropriate for dichotomous outcomes and because it also provides an estimate of the percentage change in the probability of the outcome given a unit change in the predictor for an average respondent, e.g., as the change in probability of seeing a counselor vs. not seeing one for a unit change in stereotypes (Agresti, 1990; Greene, 1993).

Table 27.3 has the descriptive statistics of the variables used in the analysis. The four dichotomous outcome variables indicate that talking to parents or friends is the most common help sought (88.6%), followed by seeing a doctor or taking medication (45.4%), getting counseling (40.7%), and going to the Internet or using a telephone helpline (16.1%). The expected help results show a somewhat similar ordering; most respondents with symptoms of major depression expect help from physicians, followed by counselors, parents and friends, and lastly from the Internet and helplines.

Table 27.3 Descriptive Statistics for Help Seeking–Dependent Variables Predicted by Stigma, Expectations of Positive Help by Source, and Demographic Data




Standard Deviation

Help seeking: doctor/nurse/meds (%)




Help seeking: counselor (%)




Help seeking: parents/friends (%)




Help seeking: Internet/helplines (%)




Stereotype scale




Expected help from physician (%)




Expected help from parents/friends




Expected help from counselor (%)




Expected help from Internet/helplines




Age (years)




Male (%)




Education (1 to 5 years)




African American (%)




Asian (%)




Other ethnicity (%)




Doctor/nurse/meds: seeing a physician or nurse or taking medications. Counselor: seeing a counselor. Parents/friend: talking to parents or friends. Internet/help: going to an Internet Web site or calling a helpline. Stereotype: mean stereotype score on 1 to 3 scale. Expected Help: expected help from specified sources. When expected help sources were multiple, the average of the two scores was used.

Table 27.4 summarizes the regression results for each of the four outcomes holding constant demographic variables (not shown). Except in two cases (counseling and seeing a physician/taking medications), there were no effects of any demographic characteristic on help seeking for these young respondents. Stereotypes of depression were negatively related to help seeking in three out of the four cases (the small positive effect is for seeing a doctor or nurse or taking medication). The effect of stereotypes on counseling was about −12%, and the effect of stereotypes on talking to parents or friends was about −6%, but these effects were not reliably different from zero. The strongest observed effect was for the impact of stereotypes on using the Internet or calling a helpline, which was almost −14%. Note that in two cases (counselors and Internet/helplines), expected efficacy of the treatment modality and help seeking were positively and significantly related, indicating that young people are more likely to seek help from sources that are seen as effective.

Table 27.4 Results of Probit Regressions Predicting Four Help-Seeking Outcomesa

Δ‎ in Probb

Z c


95% Confidence Interval for Prob. Δ‎

1. Doctor/Nurse/Medication

(Pseudo R-squared = .042, N = 189)







Expected help






2. Counselor

(Pseudo R-squared = .115, N = 189)







Expected help






3. Parents/Friends

(Pseudo R-squared = .097, N = 183)







Expected help






4. Internet/Helplines

(Pseudo R-squared = .212, N = 179)







Expected help






a All coefficients are adjusted for gender, ethnicity, age, and education.

b Δ‎ in Prob is the change in the probability of the outcome given a one-unit change in the predictor.

c Z = ratio of probit coefficient to its standard error.

Significance tests refer to probit estimates, not change in probability values (Δ‎) for an average respondent. For outcomes (1) and (2) there is a significant negative effect of the “other” ethnicity category. For all other outcomes and all other demographic variables, there are no discernable effects.

The results of this survey suggest that young people need not attach stigma to help seeking when the source of help is seen as particularly effective, as in the case of seeing a doctor or nurse or taking medication. For the other less effective sources of help, by contrast, stereotypes tended to be negatively related to help seeking and most reliably so for the least effective sources, the Internet and telephone helplines. The findings are also consistent with recently reported results from a large (N = 1,387) national survey conducted with adults in the United Kingdom, which found that the most common reason that individuals with “neurotic disorders” did not seek treatment was that they didn't think anyone could help them (reported by 22% of the sample; Meltzer et al., 2003).

Increasing the perceived effectiveness of appropriate treatment may be an important strategy for reducing the effects of unfavorable stereotypes associated with mental disorder and for increasing help seeking in general. It is noteworthy that our sample of youth with symptoms of depression reported poorer perceptions of treatment effectiveness for all but one of the sources of help compared to youth without those symptoms (Table 27.2). Hence increasing the perceived effectiveness of appropriate treatment among this vulnerable youth segment could be a strategy to increase their help seeking.


Given the role of stigma on help-seeking behavior in youth, it is critical to identify the factors that contribute to stigma. In this section, we summarize the research on factors that contribute to stigma in both the general population and among youth. A comprehensive review of this literature is beyond the scope of this chapter, so a brief summary is provided. In general, factors contributing to stigma include (a) labels; (b) symptoms and/or anomalous behaviors associated with labels and mental illness; (c) attributions about mental illness; (d) misinformation about mental illness and negative images promulgated by the mass media; and (e) lack of contact with persons who have been successfully treated for mental illness.

According to the modified labeling theory, labels are one of a number of factors that contribute to stigma, perhaps via their association with specific behaviors, media accounts, or experiences (Link & Phelan, 1999; Phelan & Link, 1999). This is especially true for pejorative labels, such as “schizo,” “psycho,” “wacko,” and so on, which may be linked with violent and erratic behavior. Therefore, labels may be stigmatizing in their own right.

Labels clearly do not exist in a vacuum, but derive meaning from their relationship with characteristics of the disorder, both real (e.g., hearing voices) and exaggerated (e.g., being a homicidal maniac). Thus, the behaviors associated with mental illness may be stigmatizing in their own right. Evidence in support of this hypothesis is garnered from studies showing that the social behaviors of individuals with depression can elicit negative reactions from others (reviewed in Segrin, 2000), and that the social skill deficits present in schizophrenia (Mueser & Bellack, 1998) may increase stigma, even beyond the contribution of symptoms (Penn, Kohlmaier, & Corrigan, 2000).

Research on the role of attributions on stigma has its roots in the work of Weiner and colleagues (Weiner, 1993; Weiner, Perry, & Magnusson, 1988) and asserts that our explanations for mental and physical illness (i.e., in terms of controllability and responsibility) will affect our attitudes toward these disorders. Tests of this model applied to severe mental illnesses, such as schizophrenia, indicate that when such conditions are seen as under the person's control and something for which she or he is responsible, the tendency to blame and stigmatize that individual increases (Corrigan, 2000). Interestingly, ascribing causality to biological factors (e.g., genetics) leads to lower responsibility attributions, but in addition, to beliefs that the problem cannot be changed and that family members may have similar problems (i.e., a courtesy stigma; Phelan, Cruz-Rojas, & Reiff, in press).

Perhaps the most consistent predictors of stigma are misinformation about mental illness. This is not surprising in view of the large role of the mass media in providing information about mental illness. In fact, the media are the most frequent source of information about mental illness for people in this country (Wahl, 1995). This suggests that the media have an important role in shaping public perception of mental illness.

Interestingly, there has been scant research directly linking media images of mental illness to stigma. There is evidence that greater exposure to the media, particularly television viewing, is associated with greater intolerance toward persons with mental illness (Granello & Pauley, 2000). However, there is convincing evidence that the media depict persons with SMI in a negative rather than positive light. Persons with SMI are disproportionately portrayed in films, television, and newspapers as violent, erratic, and dangerous (Angermeyer & Schulze, 2001; Diefenbach, 1997; Granello, Pauley, & Carmichael, 1999; Hyler, Gabbard, & Schneider, 1991; Monahan, 1992; Nairn, Coverdale, & Claasen, 2001; Signorielli, 2001; Wahl, 1995; Wahl & Roth, 1982; Williams & Taylor, 1995). As noted by Wahl (2002), media depictions of the violence committed by persons with a SMI are more graphic and disturbing than that depicted in persons without a SMI.

In an analysis of 31 major U.S. newspapers over a period of 2 months during the year 2000, APPC found that 64.7% of the stories about persons with schizophrenia had an association with violence. While associations with violence in stories about persons with bipolar disorder (29.1%) and depression (15.2%) were lower, these rates still exaggerate the incidence of violence in persons with SMI (Silver, 2001). These negative and inaccurate depictions of persons with SMI are not limited to adult media, but are unfortunately present in children's media as well (Wahl, 2002; 2003; Wahl, Wood, Zaveri, Drapalski, & Mann, 2003; Wilson, Nairn, Coverdale, & Panapa, 2000).

The role of the mass media is potentially increased because of the relative lack of direct contact with individuals who have been successfully treated for mental illness. Although there are exceptions, the research generally supports a re-lationship between greater retrospective self-reported contact with persons with mental illness and less stigmatizing attitudes (Couture & Penn, 2003). Although retrospective reports have inherent flaws, such as being influenced by memory biases, the consistency of the findings is compelling.

Most of the research on stigma has been conducted with adult and community samples. However, findings from one of our own laboratories reveal that labels, previous contact, and anomalous behaviors also influence the attitudes of older adolescents (i.e., college-age samples) toward persons with mental illness. For example, labels that varied in “political correctness” (e.g., “schizophrenic” vs. “consumer of mental health services”) exerted similar effects on the attitudes of undergraduates and a community adult sample (Penn & Drummond, 2001). Regarding contact effects, most of our studies (Penn & Corrigan, 2002; Penn et al., 1994; Penn, Kommana, Mansfield, & Link, 1999), as well as those in other laboratories (reviewed in Couture & Penn, 2003), replicate the findings with adults that greater self-reported contact is associated with lower stigma. Therefore, it appears that the factors that contribute to stigma development in adulthood are also important in adolescence.

Our review of the factors that influence stigma of mental illness suggests that one approach with potential for influencing the attitudes of adolescents toward mental illness is the use of mass media. Adolescents are heavy consumers of mass media. It has been estimated that approximately 2 3 of individuals between the ages of 8 and 18 have a television in their bedroom (Roberts, Henriksen, & Foehr, in press; Woodard & Gridina, 2000). Television viewing averages approximately 31 2 hours daily in 11-to 14-year-olds (Roberts, Henriksen, & Foehr, in press), which decreases to 21 2 hours daily by late adolescence (15–18; Brown & Witherspoon, 2001; Roberts, Henriksen, & Foehr, in press). However, these numbers may underestimate media exposure, as adolescents also devote a great deal of time (and money) to movies, video rentals, and the Internet (Brown & Witherspoon, 2001; Roberts & Foehr, in press).

The attitudes and behaviors of adolescents are influenced by exposure to the mass media. Re views of the literature indicate a relationship between media exposure and the formation of gender stereotypes (Signorielli, 2001), aggression and desensitization to violence (Bushman & Anderson, 2001), particularly among children and young adolescents (Roberts, Henriksen, & Foehr, in press), and body image in adolescent women (Groesz, Levine, & Murnen, 2002). Furthermore, the media serve as a source of information for youth; over half of teenage women report learning about sex and birth control from TV, movies, and magazines (discussed in Brown & Witherspoon, 2001). Similar findings have been observed for knowledge about mental illness. In a qualitative study with 12-to 14-year-olds, all of the participants reported that television was their primary source of information about mental illness (Secker, Armstrong, & Hill, 1999). These findings suggest that the media can influence adolescent beliefs about mental illness, perhaps in a positive way. This notion will be discussed in more detail below.


Various approaches to reducing stigma of mental disorders have been attempted, mostly in the areas of education and promoting personal contact (Corrigan & Penn, 1999). Educational interventions have included various strategies, ranging from those that are brief (e.g., fact sheets, brochures; Penn et al., 1994; 1999) to more extensive interventions (e.g., semester-long courses) that provided factual information on mental illness and dispelled myths (e.g., Holmes, Corrigan, Williams, Canar, & Kubiak, 1999; reviewed in Corrigan & Penn, 1999; Hinshaw & Cicchetti, 2000; Mayville & Penn, 1999). In general, education appears to have a short-term impact on attitudes; however, the longitudinal stability of the findings has not been adequately evaluated (Corrigan & Penn, 1999). There is some evidence that education also has an effect on helping behaviors (e.g., donating money to the National Alliance for the Mentally Ill (NAMI); discussed in Corrigan, 2002), but the impact of education on specific discriminatory behaviors (e.g., treatment at work) has to this point not been assessed.

Promoting personal contact between a stigmatized group and community members is based on the “contact hypothesis,” which has an extensive history in the study of racism (Jackson, 1993; Kolodziej & Johnson, 1996). According to this hypothesis, contact effects will be strongest when the individuals meet as equals, have a chance to work cooperatively, rather than competitively, on a task, and when the target person mildly disconfirms the stereotype. The last criterion refers to the finding that encountering someone who greatly disconfirms a stereotype may result in categorizing that target as an “exception” to the rule. Therefore, positive experiences with the target individual will not generalize to the broader group (Johnstone & Hewstone, 1992).

With respect to mental illness stigma, contact has been provided in the context of volunteer activities, classroom experiences, job training, and simulated laboratory encounters (reviewed by Couture & Penn, 2003; Kolodziej & Johnson, 1996). The findings suggest that contact effects are especially impressive and robust (Corrigan, 2002; Couture & Penn, 2003); a recent meta-analysis of the literature reported that the average effect size of contact on attitudes was .34 (Kolodziej & Johnson, 1996). These effects were largest when the contact was provided to students, rather than professionals, especially if the contact was not a required part of the classroom or training experience (Kolodziej & Johnson, 1996). These findings indicate that promoting personal contact can reduce stigma toward persons with mental illness.

Most of these studies were conducted with either college-age student samples or with adults in the community. In addition, there have been numerous grass roots or community efforts to reduce stigma, some focused on children and adolescents (Estroff, Penn, & Toporek, in press; WPA, 2002). However, most of these community efforts either did not collect outcome data or are in the process of collecting it. Therefore, we will report on the few studies that have been conducted with younger adolescents that evaluated outcomes related to stigma.

Petchers, Biegel, and Drescher (1988) imple mented a video-based educational program in two high schools. The program included a videotape of teenagers discussing their experiences of having a person with mental illness in the family, along with a six-lesson educational supplement. The results of a posttest-only design showed that participation in the video-based program was associated with higher ratings on a measure that assessed both knowledge about and attitudes toward mental illness, relative to participants who did not participate in this program. These findings are limited by the posttest-only design and use of a nonstandard measure of stigma.

Esters, Cooker, and Ittenbach (1998) assigned two classes of rural high school students to receive either 3 days of instruction on mental health, which included an instructional video, and information pertaining to sources of help in the community, or instruction unrelated to mental health. Participants completed measures that assessed attitudes toward receiving treatment and to persons with mental illness, prior to and following the course instruction and at 12-week follow-up. The results showed that the mental health instruction was significantly associated with improved attitudes both toward persons with mental illness and with seeking professional help; findings that held at follow-up.

Schulze, Richter-Werling, Matschinger, and Angermeyer (2003) also implemented an educational intervention, but one which included a contact component. Ninety high school students signed up for a program entitled Crazy? So What, a 5-day program that involved the presentation of information about mental illness, meeting someone with schizophrenia (who discussed their personal experiences with the illness, treatment, and stigma), and group discussions. Participants in this program were compared to high school students who chose to sign up for non–mental health–related projects. The results showed that Crazy? So What produced a significant reduction in negative stereotypes and a trend toward less social distance after participation in the project. These results remained stable at 1-month follow-up. Interpretation of these findings needs to be tempered by the fact that participants self-selected into the project. Thus, the preexisting characteristics that led to participation in the mental health project may have also been the underlying mechanism by which the program exerted its effects.

Finally, Pinfold et al. (2003) evaluated the effectiveness of two mental health awareness workshops on attitudes toward and knowledge of mental illness in 472 secondary school children in the United Kingdom. An individual who worked in the mental health field led the first workshop, which included viewing a videotape about people living with mental illness and challenging negative stereotypes of mental illness. An individual who shared her or his personal experience with having a mental illness facilitated the second workshop. The results showed that the workshops had a positive impact on attitudes and knowledge of mental illness, with attitudinal changes remaining stable across 6-month follow-up. These results were especially strong for individuals who reported previous contact with someone with a mental illness. Again, the value of these findings was circumscribed by the uncontrolled design.

Although limited in number and by methodological limitations, these findings are nevertheless promising. The findings also converge on a number of themes. In particular, consistent with previous findings obtained from adult samples (Couture & Penn, 2003), promoting contact seems to be a key element in reducing stigma. In addition, there is indirect evidence that demonstrating the effectiveness of psychiatric and psychological treatments, either through direct instruction (Esters et al., 1998) or via role models (Schulze et al., 2003), may reduce stigma, a finding consistent with the survey data reported earlier in this chapter (Table 27.4). Therefore, facilitating personal contact between members of the community and individuals with mental illness, and promoting the effectiveness of treatments for mental illness, may be crucial to reducing stigma among youth.

Because adolescents are heavy media users, greater use of mass media to reduce, rather than augment, stigmatizing attitudes should be explored. This can be done in a number of ways. First, the media can serve as a resource for adolescents with a mental illness. For example, Gould, Munfakh, Lubell, Kleinman, and Parker (2002) found that nearly one fifth of New York adolescents reported using the Internet to help with emotional disorders. Unfortunately, over 20% of the respondents were not satisfied with the information they found, which suggests that the content and links provided by some of the Web sites were inadequate. Second, the media can be educated to report on mental illness responsibly (Gould & Kramer, 2001; Salter & Byrne, 2002). This has been advocated in the context of suicide in adolescents by addressing contagion, the increase in suicides that follow from the reporting of suicide stories in the news media (Gould & Kramer, 2001; Gould, Jamieson, & Romer, 2003). Such responsible reporting can also be extended to presenting information about mental illness with the aim of educating the public about the efficacy of treatment for mental disorders, providing the information in a more balanced manner, and avoiding attention-grabbing, pejorative headlines (Gould et al., 2003; Wahl, 1995).

Stigma reduction may also be achieved by integrating mental illness messages into the entertainment value of film and television. Anecdotally, this appears to be the case with the film A Beautiful Mind, which seems to have served the dual purpose of both entertaining and educating audiences about mental illness. Another recent example is the collaboration between Barbara Hocking from with the staff of Home and Away, the most popular soap opera in Australia. This collaboration resulted in a storyline of a current character, Joey Rainbow, who gradually develops schizophrenia. Hocking consulted with the program's staff and provided educational materials on schizophrenia to the 18-year-old actor who played the character who developed schizophrenia. This culminated in a more accurate portrayal of schizophrenia (e.g., prodromal symptoms, acute episode, residual symptoms) than typically manifest on screen. What is especially appealing about this approach is that the character was already well established on the program, so the development of his schizophrenia was arguably comparable to observing a family member or friend develop the disorder. Thus, it would appear the producers for Home and Away provided an opportunity to promote a mediated form of contact between the audience and someone with mental illness.

We recently investigated the effects of a documentary film about schizophrenia on the attitudes of undergraduates toward schizophrenia (Penn, Chamberlin, & Mueser, 2003). The documentary I'm Still Here depicts individuals with schizophrenia in a balanced light; individuals with both remitted and acute symptoms are portrayed, with both humor and sensitivity. These depictions were supplemented by interviews with family members and mental health professionals. The results showed that this documentary resulted in more benign attributions about schizophrenia (i.e., that persons with schizophrenia are not to blame for having this disorder and that they are not responsible for causing it themselves) relative to two control documentary films. The film did not have a significant effect on participant attitudes toward individuals with schizophrenia, although the pattern of performance (i.e., task means) was in the expected direction. Participant ratings revealed that they found this film enjoyable, suggesting that documentaries can be an effective and entertaining medium for delivering information about mental illness.


This chapter has reviewed evidence that addressing the stigma of mental illness in adolescence is a worthy endeavor. It is likely that adolescents who are informed about mental illness, both in terms of facts and the dispelling of myths, will be less likely to stigmatize others and more likely to seek and stay in treatment for their own symptoms. In particular, there is promising evidence in both our survey findings and the results of interventions that increasing awareness of the efficacy of treatment can reduce the role of stigma in inhibiting help seeking and in discriminating against persons with mental disorders.

This chapter has also highlighted the potential role of mass media in destigmatizing mental illness, a role that will be more effective if expressed in partnerships between media and mental health professionals. Of course, the media can only do so much to address stigma; much is also dependent on educators and the mental health field. For example, bringing persons with mental illness to the classroom as guest speakers and providing opportunities for adolescents to volunteer with persons with mental illness have shown promise as stigma reduction strategies (Couture & Penn, 2003). In sum, the high school and classroom can promote activities that may challenge negative stereotypes.

In a similar manner, mental health professionals should strive to make the process of seeking and staying in treatment as destigmatizing as possible. Part of the challenge is developing liaisons between primary-care practitioners and school counselors, the gatekeepers who may be the first to encounter adolescents with emotional problems. This will facilitate appropriate mental health referrals and reduce the time from symptom onset and treatment. In addition, there have been recent efforts to provide treatment at home or in settings that are not identified as psychiatric facilities to address the stigma or shame of seeking treatment for physical disorders (e.g., AIDS; Gewirtz & Gossart-Walker, 2000) or psychiatric disorders (prodromal symptoms; McGorry, Yung, & Phillips, 2001). These approaches, coupled with early education and contact opportunities, and working with the media to provide balanced views of mental illness, are important steps in addressing stigma and ensuring that adolescents get early treatment for mental disorders.