Lessons from the results of three national antistigma programmes.

Introduction

This supplement brings together results of evaluations of three national antistigma programmes: Time to Change in England 1, 2, 3, 4, 5, Opening Minds in Canada 6, 7 and ‘Hjärnkoll’ in Sweden 8. Started within a few years of each other, these programmes share several common features. They make extensive use of contact with people with mental health problems as an evidence‐based stigma‐reduction method, whether that contact is direct (face to face), or indirect (virtual such as through videos). All have focused on one or more target groups, whether these be young people (Koller in this supplement), employers 8, 9, police 10, health professionals 6 or medical students 11. All have undergone evaluation by academic researchers. The two European programmes use a public health approach to defining stigma, namely in terms of problems of knowledge, attitudes and behaviour, and in addition to target groups include the general population as a target (Hansson; Henderson in this supplement). However, ‘Hjärnkoll’ was initially delivered much more intensely in three regions of Sweden before the rest of the country was included, while in England, the aim from the start has been to target the general population as evenly as possible throughout the country. In Canada, the programme uses Link's definition of stigma as the co‐occurrence of its components: labelling, stereotyping, separation, status loss and discrimination 12, and selected four specific target groups (media, workers, youth and healthcare providers). Following this definition, the Canadian programme also emphasises structural change as a stigma‐reduction strategy. To reach members of the target groups, the programme has worked with large numbers of community partners who deliver interventions in communities across the country. Although all programmes are delivering contact‐based education, there is considerable heterogeneity in delivery, which the evaluation team has exploited to identify the active ingredients of interventions, thus paving the way for fidelity criteria for future use 13 as well as toolkits to support a national scale‐up. These differences among programmes mean that consideration of all three collectively allows for greater learning compared to consideration of each individually. In this supplement, we present papers from each evaluation and use this editorial to draw lessons across the results.

Both population and target group delivery can be effective at the national level

The positive results from all three evaluations demonstrate emphatically that there is more than one way to deliver an effective national antistigma programme. For Sweden and England, which have used mass media social marketing campaigns 8, 14 in addition to local initiatives and work with target groups, it is not possible to disentangle the different influences of these different components on the general population. However, it is possible that many people have been exposed only to the social marketing campaign and that the population level changes in the outcomes at least in part reflect this exposure. In England, stigma‐related knowledge was relatively slow to emerge as a positive outcome 3 while in Sweden this was not the case 8. This may reflect differences in programme content, especially that of the social marketing campaign. After several years, however, it is clear that positive change has been sustained in all outcomes in both countries. In Canada, work with specific target groups has been delivered across the country in preparation for a national scale‐up, which is currently underway for youth groups through national summits. Population‐level change is not expected to occur until full scale‐up has occurred. At this point, the evaluation has contributed considerably to the evidence base for contact‐based education, in terms of measurement 15; identification of the key aspects of stigma to be targeted within specific populations 6; and identification of key effective programme ingredients 13.

There may be gender differences in responses to antistigma programmes

Within the general population in England, the lack of change among men compared to women in terms of reported contact with people with mental health problems is striking 3. It suggests that while there may be an impact on men's behaviour of the campaign, this is not detectable to other men; for example, they may disclose only to female friends or partners, so that only women report an increase in contact. Another possibility is that women have increased in their ability to recognise others’ mental health problems, leading to greater reported contact, while this has not occurred for men. Among Canadian youth, a positive response to the contact‐based intervention among males was highly influenced by whether they self‐reported experience of a mental health problem, whereas this was not an influential factor among females 7. Consistent with the population in England, young males were significantly less likely than females to report prior contact with a person with a mental illness. As familiarity is a consistent predictor of positive outcomes 14, 16, 17, it seems possible that lack of contact could hinder further positive change among men in terms of any or all of stigma‐related knowledge, attitudes and desire for social distance. These findings suggest that gender‐based approaches to stigma reduction may be necessary.

Campaigns and their evaluation must attend to structural discrimination

The positive change in public attitudes mirrors the changes in mental health service users’ experiences of discrimination on the part of people with whom they have informal relationships, as friends, family members and when dating 1. These changes contrast in reported experiences of discrimination in several important domains including physical health care, welfare benefits 2 and housing. While education, both with and without contact 13, 18, can be effective at least in the short term with respect to health professionals’ attitudes, other studies in health care settings suggest organisational level changes are needed 19, 20, 21, 22. Likewise, positive changes in the way editors and journalists cover mental health topics are small and do not yet show a consistent pattern. Within coverage broadly categorised as stigmatising, they may be replacing one form of negative coverage for another by shifting their focus from violence to focussing on people with mental health problems as ‘hopeless victims’. This stereotype risks fuelling within the public the same kind of therapeutic pessimism shown to be a strong component of health professionals’ stigma 6, 18. Service users’ experiences when seeking or in work lie somewhere between these patterns of positive and lack of change. After an initial improvement 23, the changes in the domain of employment became non‐significant, re‐emerging later 1. In the UK, other evidence suggests that employers are increasingly aware of the need to consider employees’ mental health and to comply with the Equality Act, for example with respect to providing reasonable adjustments 9. However, it is also possible that people with mental health problems were particularly adversely affected by the economic recession in the UK and the stagnation that followed 24. Taken together, the results of our qualitative and quantitative work with mental health service users suggest that organisations that provide services need to consider discrimination both as it affects those of their service users with mental health problems and those of their employees.

Funding

Time to Change programme, grants from UK Department of Health and Comic Relief.

Declaration of interest

CH has received a consulting fee from Lundbeck. MK and HS have received funding from the Mental Health Commission of Canada.