AIMS AND METHOD
To assess the effect of factsheets from the Royal College of Psychiatrists’ Changing Minds campaign on stigmatised attitudes of members of the general public towards those with mental illness. Participants were recruited at random from a panel of over 1200 members of the general population and presented with questionnaires containing single-page factsheets adapted from the Changing Minds campaign describing schizophrenia or substance use disorders. The Attitudes to Mental Illness Questionnaire (AMIQ) was used to measure the effect on stigmatised attitudes.
In total 200 questionnaires were distributed; 158 completed questionnaires were received (response rate 79%). The AMIQ scores for the alcoholism and schizophrenia vignettes did not differ between experimental and control groups. Fidelity questions included in the questionnaire indicated that participants had read and understood the factsheets.
Didactic factsheets produced for the Changing Minds campaign were largely ineffective at changing stigmatised attitudes towards schizophrenia and alcoholism.
The World Health Organization and the World Psychiatric Association both recognise that the stigma and discrimination associated with mental disorders are strongly linked to suffering, disability and poverty (Corrigan & Watson, 2002). Many studies have shown that negative attitudes towards people with mental illness are widespread (Crisp et al, 2000). Discrimination seems to exist in every area of life, particularly for those with psychosis and drug dependence, and shame and secrecy associated with suffering from a mental illness may also delay seeking treatment (Byrne, 2000). For example, Docherty (1997) identified stigma as a major barrier to the management of depression. This view is echoed by the influential Safer Services report (Appleby, 1999), in which stigma is seen as a major barrier to treatment-seeking and suicide prevention. Concern about the stigma of mental illness culminated in the Royal College of Psychiatrists’ 5-year ‘Changing Minds’ campaign (1998–2003) whose aim was to promote positive images of mental illness, challenge misrepresentations and discrimination, encourage patient advocacy and educate the public about the real nature and treatability of mental disorder (Crisp et al, 2000).
The large survey by Crisp et al (2005) for the Changing Minds campaign showed that people with alcoholism and drug addiction are the most stigmatised group of all those with mental illness. Most people think they are dangerous, unpredictable and hard to talk with, and 3 out of 5 people think they are to blame for their condition - an opinion endorsed by only 6% in relation to schizophrenia. We therefore chose to study methods of reducing stigma towards the conditions that clearly evoked the most negative attitudes.
The media has generally depicted people with mental illness as violent, erratic and dangerous (Monahan 1992; Wahl, 1995; Diefenbach, 1997; Granello et al, 1999). Promoting direct interpersonal contact with people who have a severe mental illness is an effective strategy to reduce stigmatised attitudes, although how much contact is required remains unknown (Wolff et al, 1996; Corrigan & Penn, 1999; Pinfold et al, 2003). Furthermore, the positive effect of contact with people suffering from one form of mental illness (such as schizophrenia) does not generalise to other mental disorders (such as alcoholism) (Crisp et al, 2005). Unfortunately the effects of interpersonal contact and public education are restricted to those members of the public who have the time and enthusiasm to become involved in these activities. More cost-effective mass media methods must therefore be developed to address psychiatric stigma in the general population.
The objective of the current study was to assess the effectiveness of factsheets developed from the Changing Minds campaign regarding stigmatised attitudes of members of the general public towards those with two mental disorders, schizophrenia and alcoholism.
Over 1200 participants were recruited using direct mail-shots and advertisements in local newspapers. This was part of a previous study to validate the Attitude to Mental Illness Questionnaire (AMIQ; Luty et al, 2006). Factsheets were submitted to random samples from this group. Participants were paired so that each could act simultaneously as an experimental participant and a control. They were chosen using the StatsDirect statistical package random case selection procedures (http://www.statsdirect.com).
Participants were asked to read a factsheet and complete the relevant version of the AMIQ with respect to schizophrenia or alcohol dependence (see data supplement to the online version of this paper). Sufficient questionnaires were distributed to produce approximately 100 responses; this was to ensure that the results for each survey had 80% power to detect an effect size of 0.5 at the 5% significance level.
Participants were asked to read a short vignette describing an imaginary patient and then answer five questions (see data supplement). The individual questions were scored on a 5-point Likert scale (maximum score +2, minimum – 2) with blank questions, ‘neutral’ and ‘don’t know’ being scored 0. The scores for the five questions were added to give a total score for each vignette between –10 and +10. Test-retest reliability at 2–4 weeks was 0.702 (Pearson’s correlation coefficient, n=256). Construct validity (Cronbach’s α) was 0.933 (n=879). Alternate test reliability compared with Corrigan’s Attributions Questionnaire was (Spearman’s ρ) 0.704 (n=102; Corrigan et al, 2003).
A pilot study was carried out in which participants were asked to read either photocopies of a three-page factsheet produced for the Royal College of Psychiatrists’ Changing Minds campaign ‘Drugs and Alcohol – Whose Problem is it Anyway?’ or a three-page information sheet from the British Diabetes Association. Both groups were asked to complete the AMIQ with respect to two hypothetical patients: one with alcohol dependence and one with diabetes (see data supplement). Two fidelity questions were included to determine if participants had read the relevant factsheet.
Fewer than half (42%) of the participants in the pilot study chose the correct answer to the relevant fidelity question. This suggested that most had not read the factsheets. Consequently the factsheets were summarised onto a single page and four fidelity questions were included. The answer to each was presented clearly in the text.
Participants were asked to read either a single-page factsheet adapted from ‘ Drugs and Alcohol – Whose Problem is it Anyway?’ or a single-page factsheet adapted from the campaign leaflet ‘ Schizophrenia’. Both groups were asked to complete the AMIQ with respect to two hypothetical patients: one with alcohol dependence and one with schizophrenia (see data supplement).
All but two respondents to this part of the survey answered either three or all four of the fidelity questions correctly. All data were collected before the terminal illness of the footballer George Best and revelations about the alcohol problems of the politician Charles Kennedy.
Where required approval was obtained from the local research ethics committee.
Non-parametric (Mann–Whitney) tests were performed using StatsDirect version 2.4.3 (http://www.statsdirect.com) to compare subgroups.
There were 200 questionnaires distributed (100 for each of the substance misuse and schizophrenia factsheets). There were 80 out of the 100 questionnaires on substance misuse completed and 78 out of 100 for the schizophrenia version (overall response rate 79%); results are shown in Table 1. The mean age of respondents was 47.2 years (s.e.=1.34); 35% were male; 46% in paid employment. There was no significant difference in age or other demographic factors between respondents and the original validation sample for the AMIQ.
The AMIQ scores for the alcoholism vignette did not differ between experimental and control groups (–0.71; s.e.=0.38 and –0.99; s.e.=0.38 respectively; P=0.46), and neither did those for the schizophrenia vignette (–2.02; s.e.=0.39 and –1.76 s.e.=0.38 respectively; P=0.65). All but two respondents answered either three or all four of the fidelity questions correctly.
Comparative observations on stigma
In a further study we presented a further sample of 100 participants with a modified vignette giving a description of an individual who had recovered from heroin addiction and who had returned to work. This produced a 40% positive change in the AMIQ stigma score compared with the description of an individual who was currently addicted to heroin (effect size 1.6). These results show that the AMIQ was sensitive to change.
Attitudes to Mental Illness Questionnaire
The AMIQ (Luty et al, 2006) was used in this project as it is convenient to use and has been well-validated. Other instruments are available but tend to be much longer, involve interviews or tend to address the experience of stigma by people with mental illness (for example the Internalised Stigma of Mental Illness scale; Ritsher et al, 2003). Other instruments have been used to measure stigmatised attitudes within the general population, although few have been fully validated. Pinfold et al (2003) reported the use of a questionnaire including four factual and five attitudinal statements that can be scored using a Likert response in secondary school children. Each made reference specifically to ‘people with mental illness’. This could not be used to compare attitudes towards those with mental illness and other groups such as people with substance use disorder who may or may not be regarded as having a mental illness. Crisp et al (2000) reported a survey of 1737 adults throughout the UK, and although attitudes were scored using a 5-point Likert scale, the survey used interviews that are too costly and cumbersome for routine use. Some researchers have used questionnaires and interviews with a ‘most people would say...’ approach (Link et al, 1989). However, we chose to develop a short questionnaire designed primarily to assess what an individual believes themselves (rather than asking individuals to estimate the views of the larger society).
Methods to reduce stigma
Action on Mental Health – A Guide to Promoting Social Inclusion (Office of the Deputy Prime Minister, 2004) provides 12 factsheets designed to encourage best practice to reduce the stigma and social exclusion experienced by people with mental illness. This supplements the efforts of the College’s Changing Minds campaign. Both campaigns give practical advice to health agencies, employers and a variety of stakeholders on how to tackle stigma. Providing factual information that addresses misconceptions about people with mental illness has been reported to reduce stigma whether this is provided as brief factsheets or more extensive interventions such as educational courses on mental illness (Mayville & Penn, 1998; Penn & Martin 1998; Corrigan & Penn, 1999). Factsheets have been used by Penn et al (1994), Thornton & Wahl (1996) and Penn et al (1999). Unfortunately the responses tend to be small, especially if the negative consequences of a mental illness are also disseminated.
A survey of the general public’s attitude to people with mental illness was performed as part of the Changing Minds campaign and was repeated at the conclusion of the 5-year anti-stigma campaign (Crisp et al, 2005). Despite the enthusiasm of those involved in the national campaign, the changes in public attitudes were small and likely to be insignificant. There were no measurable effects on public attitudes towards people with substance use disorders. These results are comparable to the previous ‘Defeat Depression’ campaign (1992–1996) which had a ‘marginal effect on public opinion’ (Paykel et al, 1998). As far as we are aware there has been no demonstration of the effectiveness of the measures advocated in Action on Mental Health – A Guide to Promoting Social Inclusion (Office of the Deputy Prime Minister, 2004).
Previous studies also showed disappointing results when participants were presented with technical factsheets. For example, Penn et al (1999) reported a study involving 182 US undergraduates who read one of three single-page factsheets that gave technical descriptions repeating the relatively weak association between violence and schizophrenia. The factsheets had no significant effects on the perceived dangerousness of a hypothetical patient with schizophrenia. It is often reported that contact with people who suffer with a mental illness leads to a more positive attitude towards those with mental illness than factsheets or positive descriptions of hypothetical patients (Penn et al, 1994).
Practical steps to change stigmatised attitudes have been suggested. Knox et al (2003) showed that training members of the US armed forces in the recognition and treatment of mental illness significantly reduced the suicide rate (Knox et al, 2003). A central component of this programme was addressing the stigmatised attitudes to mental illness. However, although the intervention (mandatory training) was carried out on an enormous scale (4 million members of the US armed forces), the reported outcome was suicide rate among military personnel rather than stigmatised attitudes.
Wolff et al (1996) reported a controlled study of the effect of a public education campaign on community attitudes towards people with mental illness following the closure of Tooting Bec Hospital in London. Surveys of attitudes were performed before supported housing was opened for people with mental illness in two areas of London. One district was subject to an educational campaign involving distribution of information packs containing a video, social events and discussion forums. Follow-up involved 51% of the original 215 respondents. Public attitudes in the experimental area improved and patients’ social integration was enhanced. However, the increased contact with patients in the experimental area was probably responsible for the improved attitudes rather than the dissemination of the information. The problems for this campaign included the time-consuming nature of the campaign for staff. The public education campaign used by Wolff et al (1996) produced effects sizes for fear and exclusion of 1.23; social control, 1.22; and goodwill, 0.66. This was from follow-up interviews using the Community Attitudes to the Mentally Ill Inventory conducted 1–2 years after initial interview. However, two earlier studies showed a public education campaign to be ineffective (Cummings & Cummings, 1957; Gatherer & Reid, 1963).
Pinfold et al (2003) reported a project in which 472 English secondary school children attended two 1-hour mental health awareness workshops. Stigma scores were measured on a 5-item scale ranging from –25 to +25. Scores rose from 1.2 to 2.8 (16% improvement compared with baseline) at 1 week and 2.3 (11% improvement) at 6-month follow-up. ‘ Overall there was a small but positive shift in students understanding of mental illness.’ Greater responses were observed for females and those who had contact with people suffering from mental illness.
Penn et al (2003) reported a study of 163 US undergraduates who were assigned randomly to four groups to watch documentaries about schizophrenia, polar bears, being overweight or a ‘no video’ control group (films ranged from 43 to 70 min). Attitudes were assessed using well-validated instruments after these documentaries had been watched. The film used in this study (‘I’m still here’) was based on the personal experience of two of the authors and represented a realistic image of schizophrenia. Some of those in the film were employed, others had disabilities, were homeless and had acute psychosis. The schizophrenia documentary did not change attitudes towards people with schizophrenia (for example perceived dangerousness) or intentions to interact with individuals with schizophrenia.
Penn et al (2003) discussed the fact that depicting negative consequences of schizophrenia in their film may be realistic but might not be the best way to reduce stigma. Depicting a ‘success’ story may be more effective although participants may then classify the depicted individual as an exception to the rule (Corrigan & Penn, 1999). The Changing Minds campaign factsheets describe the negative consequences of a mental illness. The substance misuse factsheet contains an image of a destitute person which may reinforce the negative impression.
Marketing strategies for commercial products invariably associate the product with positive images (such as popular celebrities or attractive models) regardless of whether these are relevant to the actual items being advertised (Atkinson et al, 1996). Similarly, marketing campaigns avoid associating products with any negative images and will avoid disseminating information on potential problems. For example, Jureidini et al (2004) reviewed seven drug trials, most of which were sponsored by pharmaceutical companies, concerning the efficacy and safety of antidepressants in children and adolescents, and concluded that these trials were likely to underreport adverse reactions.
We suggest that a better strategy to reduce the stigma associated with mental disorders would be to present positive images. For example, by reporting success stories of recovery and presenting images of sufferers returning to work rather than people in florid states of dishevellment or in-patients who are acutely disturbed.
Strengths and limitations
Some limitations should be acknowledged; the survey involved subgroups of participants recruited for a previous trial to validate the AMIQ. The age and employment status of the participants was reasonably matched to those from the UK census surveys, although there was an excess of female respondents. Although the sample appears to be a reasonable cross section of the British public, it is clearly self-selecting and may not generalise across the whole population. There were no differences in demographic factors between the composition of the random samples of 100 participants compared with that of the parent population.
The survey was relatively small (n=200 participants). However, power studies showed that the results had an 80% power to detect an effect size of 0.5 at the 5% significance level. Although 0.5 is a large effect size for a single intervention of this kind, this survey was designed to detect a 10% overall change in AMIQ score (2-unit change in the 20-point scale). We reasoned that a smaller change than 10% is not likely to be of any potential benefit.
The factsheets used in the survey were single-page, black and white summaries of the Changing Minds campaign factsheets. It is possible that the enlarged colour versions would be more effective at changing stigmatised attitudes. However, the three-page factsheets that were used in the pilot study had no effect on stigmatised attitudes to a hypothetical individual with alcoholism and it rapidly became apparent that participants had not read them. In reality, it seems unlikely that members of the general public would read and absorb long information sheets unless they had a specific personal interest.
The study presented a vignette of an imaginary person with mental illness which might not be an accurate representation of real-life experience. Clearly it is seldom possible to measure stigmatised behaviour towards real people who are mentally ill. Social desirability also may affect the results. Furthermore, the written views and expressed attitudes may not translate into any actual activity or behaviour (such as readiness to employ a person with mental illness).
The research was supported by grant British Academy grant no. SG-35479. The Changing Minds campaign factsheets are now only available from the College website (http://www.rcpsych.ac.uk/campaigns/changingminds/mentaldisorders/alcoholanddrugmisuse.aspx).
- © 2007 Royal College of Psychiatrists