he stigma and accompanying discrimination associated with mental illness emanates from societal structures and individuals’ knowledge, attitudes, and behaviors (1). Strategies to combat stigma and discrimination associated with mental illness are considered necessary to lessen the social injustice experienced by those suffering with mental illness (1—5). These strategies have been designed to diminish stigma via educational, contact, and/or protest methods (5, 6). Educational methods offset stigmatizing beliefs regarding mental illness by providing accurate knowledge. Contact methods offset stigmatizing knowledge and attitudes via opportunities for interaction with persons with mental illness. Protest methods offset stigmatizing attitudes and behaviors by suppressing their expression.

Attention has recently focused on the developmental expression of mental illness stigma and the subsequent design of age-appropriate stigma reduction strategies. Wahl reviewed the research literature regarding child and adolescent mental illness stigma (7). He reported that young children may hold negative attitudes toward individuals with mental illness and that these attitudes may intensify with ongoing development. Wahl was cautious, however, in making this statement, as the reviewed studies used a range of methodologies that rendered a limited generalizability of specific findings, and specific methodologies that did not incorporate sufficiently well designed, reliable, and age-appropriate measures. Watson et al. (4) discussed the contradictory nature of recent research regarding the attitudes held by adolescents and young people toward those with mental illness. Furthermore, research exploring the impact of contact and familiarity with mental illness on adolescent mental illness stigma has been contradictory, with both ameliorating (4) and accentuating effects (2) reported. There is some contention, however, that the nature of this contact may be a mediating variable, with contacts that challenge stereotypes being ameliorative and contacts that reinforce stereotypes being accentuative (2). Ongoing investigation regarding adolescent expression of stigma-related attitudes and behaviors is needed for various reasons, including the extrapolation of knowledge about adult-related stigma to adolescent-related stigma is inappropriate, to enable development of age-appropriate stigma reduction programs to ameliorate the developmental transition of negative attitudes from adolescence into adulthood, and for evaluation of the beneficial influence of stigma reduction strategies upon help-seeking behaviors of adolescents suffering from mental illness (2—4).

The stigma-related behaviors of healthy adolescents are of obvious importance to adolescent psychiatrists and their patients. As a result, psychiatric education may present a previously unrecognized opportunity to influence adolescent stigma-related knowledge, attitudes, and behaviors. In psychiatric education, healthy adolescents are commonly recruited to work as adolescent standardized patients (ASPs), simulating a range of psychiatric case conditions and adverse psychosocial experiences. Perhaps resulting from stigmatization, ASPs have reported anticipation of discomfort with simulating the psychiatric role of depression and suicidality and for sexual roles including HIV/AIDS (8). Paradoxically, however, portraying such roles as simulated patients may present a stigma reduction strategy as it creates the opportunity for both education and contact. Indeed case simulations have been used to increase awareness regarding those with physical disabilities and mental illness and to reduce the stigma associated with these conditions (5, 9). While we certainly respect the right of ASPs to refrain from participating in simulations in which they are uncomfortable, standardized patient training methods include active learning strategies such as case-related information sharing with accompanying discussion and case simulations which embody the principles of effective stigma reduction strategies reported in the literature (5). Furthermore, involvement in psychiatric case simulation work represents an arms-length mental illness contact experience for ASPs.

In summary, we hypothesize that carefully designed ASP training in psychiatric and adverse psychosocial scenarios may comprise a stigma-reduction strategy for the adolescent simulators. We designed and piloted ASP training methods that incorporated a carefully and accurately written adolescent psychiatric case simulation, case-related educational materials, and active learning methods with discussion and feedback provided by both a standardized patient trainer and adolescent mental health specialist. We assessed the effects of training and portrayal of a patient with elevated suicide-risk and depression, previously shown to be an uncomfortable role (8), versus a pediatric cough condition. As a measure of stigma-related attitudes, we surveyed ASPs posttraining regarding their comfort with future portrayals of a range of psychiatric and adverse psychosocial conditions that may be associated with mental illness stigma. We report the results of this study and preliminary investigation of this hypothesis.

ASP recruitment and training activities were completed by the standardized patient trainer (BH) of the Clinical Learning Centre at McMaster University using a snowball sampling procedure (i.e., spread by word of mouth through the existing standardized patient population). Consenting procedures were undertaken at standardized patient recruitment evenings. Parental consent and adolescent assent was obtained for adolescents under the age of 16. Adolescent consent was obtained for those over age 16. Upon recruitment, participants completed the Suicidal Ideation Questionnaire (10) and Reynolds Adolescent Depression Scale-2 (11). Those scoring in the clinical range were monitored by an adolescent mental health professional, but no participants were excluded on the basis of scores on these measures. Upon completion of these scales, ASPs were randomly assigned to train for simulations of either a suicide-risk depression role or a pediatric cough role.

ASP training for both conditions involved two sessions, each lasting 2 hours with four to six ASPs per group. In the first training session, ASPs were asked to review case-specific educational materials outlining the nature of the health condition to be simulated. They engaged in discussion with an adolescent mental health specialist and ASPs rehearsed their role with the standardized patient trainer who emphasized its behavioral aspects. Those in the depression/suicidal ideation condition received instruction in stress-relief techniques as part of the intervention. The second training session occurred approximately 2 weeks later during which ASPs practiced their role and received feedback from the standardized patient trainer and adolescent mental health specialist regarding their performance.

In our previous ASP research regarding simulation effects we evaluated adolescent anticipated role discomfort by means of the "Project Role Questionnaire" (PRQ) (8). The PRQ listed 10 patient simulations (a range of pediatric, psychiatric, and adverse psychosocial conditions). For each role, ASPs were asked to rate their comfort by assigning a score on a 4-point Likert scale ranging from 1 ("very uncomfortable") to 4 ("very comfortable"). These items were followed by two open ended questions: "Explain your discomfort for identified roles," and "For each of these roles, could standardized patient training address your concerns?" For the current study, the PRQ was adapted to include 16 adolescent psychiatric and adverse psychosocial conditions. It no longer included pediatric medical conditions. In this fashion, a broader range of adolescent psychiatric conditions were evaluated and some roles grouped together on the original PRQ were separated out (e.g., sexual abuse/rape/prostitution or HIV/AIDS was one standardized patient role item on the original PRQ and became multiple items on the adapted PRQ). The adapted PRQ questionnaire maintained the Likert scale scoring scheme and the two open-ended questions of the original PRQ. Table 1 includes the complete list of 16 roles participants considered. Approximately 2 weeks after ASP training, ASPs reconvened to complete the Reynolds Adolescent Depression Scale-2, the Suicidal Ideation Questionnaire, and the adapted PRQ questionnaire. Additional data were collected posttraining from all participants for another purpose. Those details are reported by Hanson et al. (12)

Analysis of variance was used, with the standardized patient’s condition as a grouping factor, to assess whether or not portraying the study simulation associated with potential mental illness stigma (depression and suicidal ideation) had an impact on ASPs’ anticipated comfort with portraying future roles. These future roles included a range of psychiatric and adverse psychosocial conditions that may also be associated with mental illness stigma relative to the more neutral pediatric cough condition. In addition, ASPs’ answers to one of the open ended questions were independently reviewed by two study investigators (MH, SJ) for common themes related to role discomfort. Each generated themes independently and then came to consensus regarding what was indicated by each statement. A 3rd author (KE) was consulted to resolve points of discussion.

This study received ethics approval from the Hamilton Health Sciences Research Ethics Board.

Twenty-four ASPs (17 female), 14—17 years of age (mean = 15.5), participated in the study. Study ASPs had not worked as ASPs previously. Equal ASP numbers were assigned to each condition in a stratified random manner: three of the 24 ASPs had clinically elevated Reynolds Adolescent Depression Scale-2 and Suicidal Ideation Questionnaire scores and stratification precluded the assignment of all three to the same condition. Overall, Reynolds Adolescent Depression Scale-2 scores and Suicidal Ideation Questionnaire scores for those ASPs assigned to the depression with suicidality condition (54.3 versus 9.9, respectively) did not differ from those of ASPs assigned to the pediatric cough condition (52.7 and 9.3, respectively), p>0.05 in both cases. Post-participation, these scores did not change significantly and only two of the three "clinical" ASPs remained in the clinical range. Adolescent mental health interviews were completed for the three ASP participants identified as being in the clinical range on these instruments and none were considered to be at acute mental health risk or to have suffered deterioration in their mental status during the study period. Community mental health service referrals were discussed with these three ASPs during these interviews and an eventual referral was initiated for one ASP at study completion. The test-retest reliability of both instruments was well within the acceptable range (0.84 for the SIQ and 0.90 for the Reynolds Adolescent Depression Scale-2).

A mixed design ANOVA (with ASP condition as a between subjects factor and hypothetical PRQ role as a repeated measure) was used to compare comfort levels across groups. A significant effect of group was observed (F (1, 21)=6.5, p<0.02) indicating that participants who took part in the depression and suicidality condition anticipated greater comfort (mean=3.29, 95% CI=2.93—3.65) with portraying the various emotionally evocative roles than did those in the pediatric cough condition (mean=2.69, 95% CI=2.34—3.03). While there was also a significant effect of role (F (15, 315)=4.7, p<0.001), there was no interaction involving role and ASP condition. These latter two results indicate that participants anticipated some roles being more uncomfortable than others, but that the difference between those in the depression and suicidality condition and those in the pediatric (cough) condition was fairly consistent across question. Means for each condition are illustrated in Table 1 as a function of role (i.e., suicidality or cough).

In response to the open-ended explanatory question regarding ASP discomfort, 13/14 ASPs who endorsed discomfort for at least one PRQ role offered written explanatory responses. Of these 13 ASPs, four had portrayed the suicidal depression role, the frequency with which participants in this group indicated discomfort (4/12) being significantly less than the frequency with which participants in the simulated pediatric cough role (9/12) provided statements indicative of discomfort (χequals;4.2, p<0.05, consistent with the mean differences reported in Table 1).

For sexuality roles, ASPs typically provided explanatory statements of a general nature (e.g., "I do not like people asking me about sexual things because it is kind of a personal thing" [sic] and "It is very uncomfortable to talk to people about sex"), but a few did focus their comments on a specific adverse sexual experience (e.g., "Rape is a sickening subject that has always bothered me." and "HIV/AIDS has scared me since I learnt about it.")

For the PRQ psychiatric/adverse psychosocial roles, ASPs’explanatory statements were unique relative to the statements about discomfort with sexual roles. These statements tended to suggest personal reactions secondary to knowledge-based issues. ASPs claimed discomfort due to lack of knowledge of the condition (e.g., "not having an illness myself, I would find it hard also putting myself in the others shoes would be difficult" [sic]); inaccurate knowledge regarding what would be expected of someone portraying a psychiatric role (e.g., "also, I do not think I’d be good with yelling or something, if I was acting as a mentally ill person" [sic]); and too much knowledge (e.g., "I have a close friend who has had problems with depression that have gone onto suicide attempts and my personal experience could greatly effect my performance as a standardized patient").

Our study’s guiding hypothesis was that a contributing factor to the ASP discomfort associated with a variety of adolescent psychiatric and adverse psychosocial conditions is stigma related to knowledge (or lack thereof) and attitudes regarding these conditions. As a result, we chose to test whether or not engaging ASPs in an ASP training exercise involving one such simulation of a psychiatric/adverse psychosocial condition (utilizing education and role playing in a safe setting with adolescent mental health professionals) would help to increase comfort with simulations of these conditions. In other words, we designed ASP training to include the education and contact stigma reduction methods documented in the literature. Education took place via discussion of the ASP mental illness case and contact took place via role playing of an individual suffering with a mental illness.

Our findings suggest that ASP role discomfort is indeed readily modifiable. Two groups of ASPs completed the adapted PRQ questionnaire, one after simulation of an adolescent psychiatric depression and suicide risk condition and the other after simulation of a pediatric cough condition. ASP discomfort level (as measured by the adapted PRQ) across 16 hypothetical psychiatric/adverse psychosocial roles differed between ASP groups; the psychiatric ASP group anticipated more comfort with these roles than the pediatric ASP group. These results suggest that a single ASP work-related experience with an adolescent psychiatric case simulation may positively influence adolescents’ general perceptions of a range of psychiatric/adverse psychosocial conditions.

To further clarify our study hypothesis, potential factors contributing to ASP comfort were explored by review of ASP explanatory statements regarding their discomfort. One category of ASP explanatory responses suggested that personal reactions to and personal experiences with psychiatric/adverse psychosocial conditions contributed to their role discomfort. These reactions may relate to limited knowledge regarding mental illness, inaccurate knowledge regarding mental illness, intimate knowledge (subsequent to personal contacts) regarding the mental illness in question, or other as yet unidentified factors. Another distinct category of ASP responses focused on ASP discomfort with sexual matters in simulations ranging from sexual experiences of an appropriate developmental type to adverse events such as sexual abuse. In part, these ASP responses were similar to our previous ASP focus group research in which ASPs themselves suggest that ASPs comfortable with their sexuality be selected for sexual roles (8). However, ASP discomfort with adverse sexual roles may not be fully explained by this developmental maturity factor; stigma-related attitudes with some roles may also contribute to role discomfort. For instance, the statement "Rape is a sickening subject that has always bothered me," may be more reflective of a stigma-related attitude than a developmental factor.

Limitations of this study include, first and foremost, the fact that the adapted PRQ did not define comfort in terms of stigma-related factors such as knowledge of and attitudes toward mental illness or contacts with those suffering from mental illness. In other words, a formal direct measure of stigma itself was not used. However, the review of ASP explanatory statements suggests that such factors may contribute to role discomfort as measured by the adapted PRQ. Although these ASP statements are not definitive evidence of either stigma-related and/or developmental factors’ contribution to ASP simulation discomfort, we do believe they provide preliminary support for our study hypothesis. Second, an additional limitation is the small number of ASPs enrolled in the study. The sample did have sufficient power to reveal statistically significant effects, thus lessening this concern, but the small sample prevented us from thoroughly exploring other potential mediating factors like gender. Third, the adapted PRQ was not completed before ASP participation, making it impossible to judge the extent of the change in comfort levels pre- versus posttraining. Finally, the full extent of adolescent stigma-related knowledge and attitudinal factors may not be known because some ASPs may not have wished to openly express them on the adapted PRQ.

Further study should use reliable measures of adolescent mental illness stigmatization within ASP work with psychiatric and adverse psychosocial conditions. We may then be better positioned to determine the presence and strength of the relationship between stigma-related knowledge and attitudes and ASP discomfort, and whether or not ASP training methods provide an opportunity to modify these ASP stigma-related factors in addition to raising ASP comfort levels in portraying these roles. In addition, it would be interesting to know if raising comfort levels and decreasing stigmatization has an impact upon the actual performance of the ASPs, thus altering the quality of the educational/evaluative experience they provide the health professional trainees for whom the simulated patients are being trained. Primary and secondary schools have demonstrated interest in incorporating stigma reduction programs into their curricula (3, 13, 14). If further study with reliable measures continues to demonstrate beneficial effects of fairly minimal, but targeted interventions such as these, then the community should consider studies into the efficacy of school-based programs that would enable a broader reach to greater numbers of adolescents.