Opening Minds in Canada: Targeting Change

Abstract

Objective:

To summarize the ongoing activities of the Opening Minds (OM) Anti-Stigma Initiative of the Mental Health Commission of Canada regarding the 4 groups targeted (youth, health care providers, media, and workplaces), highlight some of the key methodological challenges, and review lessons learned.

Method:

The approach used by OM is rooted in community development philosophy, with clearly defined target groups, contact-based education as the central organizing element across interventions, and a strong evaluative component so that best practices can be identified, replicated, and disseminated. Contact-based education occurs when people who have experienced a mental illness share their personal story of recovery and hope.

Results:

Results have been generally positive. Contact-based education has the capacity to reduce prejudicial attitudes and improve social acceptance of people with a mental illness across various target groups and sectors. Variations in program outcomes have contributed to our understanding of active ingredients.

Conclusions:

Contact-based education has become a cornerstone of the OM approach to stigma reduction. A story of hope and recovery told by someone who has experienced a mental illness is powerful and engaging, and a critical ingredient in the fight against stigma. Building partnerships with existing community programs and promoting systematic evaluation using standardized approaches and instruments have contributed to our understanding of best practices in the field of anti-stigma programming. The next challenge will be to scale these up so that they may have a national impact.

In this supplement, Stuart et al¹ previously outlined the grassroots, community development approach taken by the MHCC’s OM Anti-Stigma Initiative. Our paper briefly summarizes the anti-stigma activities and outcomes across the 4 target groups (youth, health care providers, workplaces, and media), identifies some of the key challenges faced, and outlines some of the lessons learned.

The Measurement Challenge

One of the first major tasks was to standardize the measurement approach across programs. The youth team developed and tested two 11-item scales.² The first scale measured attitudes and focused on stereotypic attributions pertaining to perceived controllability of the illness, potential for recovery, and potential for violence and unpredictability. The second scale measured behavioural intentions related to social acceptance of people with a mental illness, such as desire for social distance and feelings of social responsibility related to mental health issues. Questions were worded to be accessible at the Grade 6 level, and an easy-to-read version was created for younger students. Psychometric evaluation showed good factor structure and internal consistency.

Clinical Implications

• OM has created a foundation of methods, knowledge, tools, and expertise to support stigma reduction efforts in Canada.

• Contact-based education has emerged as a best practice in the field of anti-stigma programming.

• Contact is best when it is in person, engaging, when a story of recovery and hope is told, and when speakers have received training and support to succeed in this teaching role.

Limitations

• This initiative was largely based on nonrandomized studies.

• Certain subpopulations, such as ethnic minorities or Aboriginal peoples, were not represented in these studies, leaving important gaps in our knowledge.

• The outcomes assessed were attitudinal or behavioural intentions rather than the stigmatizing behaviours themselves.

For the health care provider programs, special attention was paid to aspects of stigma that were specific to the health care field, including attitudes and behavioural intentions related to phenomena, such as diagnostic overshadowing and prognostic negativity.³ It was also important for the scale to address the following: perceptions of competence and personal control among people with a mental illness, perceptions of the extent to which health care providers felt they had a role or responsibility to advocate for people with a mental illness; social distance; and inclinations toward disclosure and help seeking. Original testing yielded a 20-item scale that was used for evaluation of programs, while an updated psychometric assessment suggests a more efficient 15-item version with 3 subscales measuring attitudes (6 items), disclosure and help seeking (4 items), and social distance (5 items) can be used without loss of performance.⁴

The workplace teams created 2 scales. The OM Scale for Workplace Attitudes is a 22-item measure assessing stigmatizing attitudes, beliefs, and behaviours in the workplace. This measure has been initially validated on a student sample, and is being further evaluated in an employed community sample. Second, a 12-item Scale for Supervisor Attitudes was derived from existing items from various studies.⁵ Psychometric evaluation of this scale is in progress.

Youth Programs

The first programs in the Youth Pilot Program Network began collecting data in 2011 using a one-group preand posttest design. To date, 27 evaluations have been completed, representing about 10 000 students from schools across Canada. Two additional evaluations are scheduled for completion in the early fall of 2014.

Even though most of the youth programs were collecting data, we sought formal approval from school boards and (or) their designated research committees. Each school district had different requirements and procedures, thus multiple iterations of proposals and documents were required. Many schools required active consent, where a permission slip would be sent home to parents for their signature. Only children with approvals could be included in the evaluation. When active consent was attempted, less than 20% of the students returned an approval. Therefore, only schools that allowed passive consent were included. A letter of explanation was sent home to parents with a signature form to be completed only if they did not want their child to be included in the evaluation. The passive consent procedure did not result in a single parental refusal.

To match pre- to posttests for each student, we used a unique anonymous identifier composed of the initials of the first and last name, sex, and date of birth or age. The matching process was laborious and, in some cases, matching was not possible. Unmatched analysis can exaggerate statistical significance, especially in large data sets such as ours. To offset this problem, we defined, a priori, a threshold score and compared the proportion of students who passed this threshold on the pre- and posttest as the main outcome. If a student answered items in a nonstigmatizing way 80% of the time or greater (corresponding to an A grade for an educational intervention), they were considered to have passed this threshold.

Most programs were successful in improving students’ stereotypical views. The most successful programs showed improvements of up to 30% in the proportion of students who reached the a priori threshold score. For many programs it was more difficult to improve social acceptance. These findings support current social psychological research showing that improved attitudes may be poor predictors of stigmatizing behaviours and underscore the importance of targeting anti-stigma programs to behavioural change.⁶

Several short, one-off classroom programs performed well when evaluated. However, research involving other target groups showed that positive change resulting from brief programs deteriorated over time when there were no follow-up interventions. When similar short programs occurred in larger groups, such as assemblies, student engagement was difficult to achieve and evaluation showed poorer results. Though teachers often request shorter single class interventions, longer programs evolving over several class days are an emerging best practice. Finally, the cornerstone of contact-based education is the story that is told. The best storytellers shared a story with a recovery theme and without lengthy digressions into the signs and symptoms of their illness, were psychologically ready to share their experiences to help students learn (rather than to achieve a personal therapeutic goal), were able to engage their audience, managed active participation through questions and discussion, and modelled recovery. The most effective programs recognized that this takes considerable training and support, which they provided on an ongoing basis.

Qualitative inquiry helped us gain a better understanding of the critical ingredients associated with success. The resulting logic model identified 4 structural domains (referring to staffing, infrastructure, and other inputs), and 6 process domains (governing how the contact-based interventions were delivered). We used the model to create 37 fidelity criteria, against which all programs will be rated. This is the first step toward validating a fidelity scale that could be used to promote best practices in contact-based education aimed at youth.

Health Care Provider Programs

Twenty-four health care programs have been evaluated and 3 are ongoing. With the exception of 2 randomized controlled studies,^7,8 evaluations used a pre- and posttest study design. Where feasible, a follow-up measure (usually 3 months) was incorporated. Programs represent a broad spectrum of health care providers, such as students in various health-related disciplines (nursing, pharmacy, medicine, occupational therapy, and psychology), and practising health care providers (physicians, emergency department personnel, nurses, allied health providers, or administrative and support staff). The nature and amount of contact-based education varied across programs. For example, in 2 of our programs targeting students in health care fields, students were required to meet with a person who had experienced a mental illness throughout the term and to construct a life and recovery narrative based on the client educator’s personal experiences and perspectives. Five programs provided less intense contact, with one-time interventions ranging from 1 to 3 hours, typically delivered in a classroom as part of a regular course.

More detailed results are presented in Knaak et al⁹ in this issue. Briefly, programs were generally effective, though varied in their impact. Follow-up data showed diminishing effects over time, most commonly observed in the shorter (1- to 3-hour) single-session programs.^1,10 As a result, 3 programs incorporated booster sessions to provide another opportunity to build on the lessons learned in the original workshops. These were short (1 hour or less), and included some form of contact-based education or skills training. Three-month follow-up data for the programs with booster sessions showed that the positive changes initially realized were sustained.¹¹ The ability to sustain positive change by providing ongoing programming was a key theme that emerged from our qualitative research.¹² Thus multiple sessions delivered over a period of time are likely to be more effective at sustaining improvements in attitudes and behavioural intentions than a single session or workshop.

Programs emphasizing skill acquisition also tended to show robust follow-up results. For example, a 4-hour workshop that provided primary care physicians and other front-line staff with knowledge and skills to support clients with severe or persistent mental illness showed continued improvements at both 3 and 6 months with no booster sessions. Similarly, a 3-module program designed to help family physicians increase their capacity and comfort in diagnosing and engaging patients in the management of their mental health showed continued improvement for those physicians who had completed the program 1.5 years earlier. These results suggest that, as health care providers put these skills into practice, they become more comfortable and confident in their ability to interact with, and care for, people who have a mental illness. Because both of these programs emphasized patient empowerment and partnership, gaining a greater understanding of recovery could be an important element of change. Setting these dynamics into motion may translate into improved attitudes and behaviour.¹³

Workplaces

The first step in developing an evidence-based approach to the workplace was a scholarly review of relevant anti-stigma intervention programs to explore existing workplace anti-stigma programs and to consider various program types and formats.¹⁴ A follow-up scoping study identified and described principles and characteristics of anti-stigma initiatives.¹⁵ Twenty-two anti-stigma initiatives were identified from peer-reviewed, grey, and other relevant literatures. Conclusions included the need for standardized interventions, validated evaluation tools, more scientific rigour in evaluation and implementation, and greater attention to sustainability.

A major activity of the workplace OM group has been to engage employers to implement one or more programs and then evaluate outcomes. In each setting, an evaluation framework is designed in consultation with the employer. Issues addressed include the following: confidentiality and recruitment, the potential increase in mental health workplace claims associated with increased awareness of concerns, and practical issues, such as time and personnel costs to conduct research.

To date, the most complete evaluation has been within an Ontario municipal government that implemented an individualized version of the Canadian Mental Health Association’s Mental Health Works Program.¹⁶ Support from the senior leaders facilitated implementation and evaluation. Initially, the program was delivered to 500 supervisors. Using a scale specific to the supervisory role, attitudes were shown to become less stigmatizing. A modified program was subsequently extended to all 5000 employees, with a corresponding evaluation currently under way.

A second example involves a multi-component set of interventions to enhance call centre employees’ well-being, which was piloted in 2 call centres. This evaluation was complicated by 2 main factors: implementation of the pilot project was already in progress when OM was engaged for the evaluation, which precluded a true baseline for the study; and participation in the various components of the program was voluntary, thus each worker received a different set of interventions based on personal choice. While this provided an opportunity to investigate workplace stigma in a voluntary and unstandardized program, it did not allow for a conventional pre–post evaluation survey design. Qualitative evaluation indicated that the voluntary nature of the program was a barrier to full participation and that implementing and evaluating programs in call centres requires attention to mental health issues specific to that setting (for example, shift work, difficult customers, and work targets). The data also suggested a need to further examine the issue of self-stigma in workplace mental health.

As a final example, the Department of National Defence has developed a comprehensive program to increase resiliency and mental health in its soldiers and staff, entitled R2MR. The program uses a mental health continuum model to teach people to look for signs and behavioural indicators in themselves and others, and to take appropriate actions when they appear. Colours designate levels of severity, bypassing diagnostic labels and the stigma attached with them. OM has adapted R2MR for a general workplace audience. The Working Mind retains most of the central components and includes several additional modules focusing on the myths of mental disorders, reducing the associated stigma, and coping strategies in the workplace. The anti-stigma module uses videos and live presentations (when possible) to offer a robust contact-based educational experience. At present, there are 3 versions of the program: 1 for managers, 1 for front-line workers, and 1 for training the trainers. Evaluations are under way at numerous settings nationally and include universites and colleges, health networks, corporations, and government ministries and agencies. In addition, numerous police forces are scheduled to receive a combination of R2MR and The Working Mind, and these evaluations will be ongoing.

Compared with other target groups, engaging employers has been considerably slower. They often show initial interest in a partnership, and then interest fades. Several factors affect the courtship process. One of the largest seems to be the extent to which the executive leadership acknowledges the potential impact of mental illness–related stigma and the importance of psychological health and safety in the workplace.

One of the most important lessons learned is that voluntary interventions have low enrolment rates, sometimes less than 50%. Programs embedded as part of ongoing training have comparatively higher uptake. Embedding anti-stigma programs into the normal training cycle, and thus making them expected, appears ideal for facilitating change. Similarly, given the voluntary nature of evaluation research, organizational support and a culture of expectation of participation are also important to increase the participation in evaluations.

Several methodological issues have also been encountered. While strategies to create unique and anonymous identification codes have been employed to match questionnaires over time, in many cases these failed to yield matches. A related problem was participant attrition, in both completing multiple questionnaires and across data collection periods. A dedicated person is needed to organize and coordinate the research at each site, to ensure proper completion of questionnaires and to reduce attrition rates. In the absence of this, clear communications are needed to reduce both nonmatches and attrition. Such a plan can have the added benefit of creating enthusiasm for the initiative, explicitly making the evaluation part of the continuing actions to address employee mental health.

Media and Journalists

To promote more balanced reporting, a contact-based educational symposium was delivered to journalism students. Because there were no Canadian data, this initial symposium used an expert from the United States and examples from US newspapers. Although the intervention resulted in statistically significant improvements in stigma,¹⁷ students and their professors remained skeptical that the US reporting practices applied to Canada. As a result, OM launched a large media-monitoring project. These data have provided Canadian examples of stigmatizing newspaper reports for journalism seminars and have been used to assess changes in media practice over time and conduct subanalyses of specific events.¹⁸ This project has also contributed new standardized abstraction methods to this field.¹⁹ OM has since organized 5 additional mental illness awareness symposiums for journalism students across Canada. They are now serving as a model for an online program to be shared broadly with journalism schools.

In addition, OM contracted the Canadian Journalism Forum on Violence and Trauma to research and write a new media resource guide called Mindset: Reporting on Mental Health.²⁰ The Forum also partnered with the Canadian Broadcasting Corporation to create a companion website.²¹ The guide was released in April 2014 and contains information for reporters on the problems created by stigmatization of mental illnesses, myths perpetuated about mental illnesses, quick facts, suggestions on how to cover stories about violence, language use, contacts who can help journalists who are preparing new stories, and examples of stories that help reduce stigma.

Scaling Up

Considerable effort has been devoted to identifying the critical components that make programs work. Once active ingredients are identified, they can be used to develop fidelity criteria to help assess the extent to which programs meet best practices. Supporting resources (such as toolkits and resource guides) can be distributed widely to promote the dissemination of best practices on a national (or international) level. At this point, advertising and knowledge translation become important tools—not as a means of changing attitudes—but as a way to raise awareness that effective anti-stigma resources exist.

Abbreviations

MHCC

Mental Health Commission of Canada

OM

Opening Minds

R2MR

Road to Mental Readiness