The nature of stigma

The stigma associated with mental illness contributes significantly to the burden of mental illness. Subjective accounts of people affected by mental illness testify that its effects are often perceived as more burdensome and distressing than the primary condition itself (Thornicroft, Reference Thornicroft2006). The term stigma refers to ‘a social devaluation of a person‘ (Thara & Srinivasan, Reference Thara and Srinivasan2000, p. 135), due to an ‘attribute that is deeply discrediting’ (Goffmann, Reference Goffmann1963, p. 3), and can be conceptualised as consisting of ‘problems of ignorance, prejudice and discrimination’ (Thornicroft, Reference Thornicroft2006, p. 182). Discrimination leads to disadvantages in many aspects of life, including personal relationships, education and work. As a result of the stigma which is internalised, some people with mental illness may come to accept the discrediting prejudices held against them, and so lose self-esteem, leading to feelings of shame, a sense of alienation and social withdrawal (Ritsher et al. Reference Ritsher, Otilingam and Grajales2003; Ritsher & Phelan, Reference Ritsher and Phelan2004). Therefore, people with mental illness often expect to be treated in a discriminatory way (‘anticipated discrimination’) and try to hide their illness, or to stop themselves from taking up opportunities (Ritsher et al. Reference Ritsher, Otilingam and Grajales2003; Thornicroft, Reference Thornicroft2006). A related concept is that of the Right to Health, or its violation, which is an important component of the United Nations Convention on the Rights of Persons with Disabilities (United Nations, 2006).

Although it is widely accepted that stigma and discrimination are at present universal phenomena, experiences of stigma and discrimination are often manifested in specifically local ways (Murthy, Reference Murthy2002). Yang et al. point out that ‘across cultures, the meanings, practices and outcomes of stigma differ, even when we find stigmatisation to be a powerful and often preferred response to illness, disability and difference’ (Yang et al. Reference Yang, Kleinman, Link, Phelan, Lee and Good2007, p.1528).

Although there is now a strong evidence base describing stigma, the great majority of these studies have been carried out in high-income settings (Corrigan et al. Reference Corrigan, Morris, Michaels, Rafacz and Rüsch2012; Griffiths et al. Reference Griffiths, Carron-Arthur, Parsons and Reid2014; Knaak et al. Reference Knaak, Modgill and Patten2014; Thornicroft et al. Reference Thornicroft, Mehta, Clement, Evans-Lacko, Doherty, Rose, Koschorke, Shidhaye, O'Reilly and Henderson2015). Research is clearly needed to understand which aspects of stigma and discrimination are the most common and burdensome in LAMIC settings, where about 85% of the world's popuation live, and to clarify which determinants of stigma are potentially modifiable. The aim of this paper is to contribute an understanding of stigma and dsicrimination that considers context-specific factors, and the evidence related to stigma from LAMICs that will serve to inform interventions intended to reduce stigma and discrimination. The method used is a narrative review, drawing up recent relevant primary data papers and literature reviews.

Context-specific factors relevant to stigma and discrimination in LAMICs

In LAMICs, as in any other setting, experiences of stigma and discrimination are subject to the influence of local cultural factors (Murthy, Reference Murthy2002). Key domains through which culture shapes the manifestations of stigma include: (i) notions of ‘mental illness’ and explanatory models (e.g., in many settings, psychiatric symptoms may not be seen as indicative of an ‘illness’, and sometimes attributed to supernatural forces); (ii) cultural meanings of the impairments and manifestations caused by the disorder and its stigma (e.g., the impact of stigma on a person's marital prospects may have a different impact on a person's life depending on the cultural meaning attached to marriage in a given society); and (iii) concepts of self and personhood (e.g., higher levels of family cohesion may offer more support, but also go along with a more widespread impact of stigma across family members and generations).

However, differences between settings, including between HICs and LAMICs, go beyond ‘cultural’ differences (i.e., differences relating to beliefs and norms). Socio-economic factors, such as poverty and access to healthcare, have long been found to be associated with outcomes of mental illness (Lund et al. Reference Lund, De Silva, Plagerson, Cooper, Chisholm, Das, Knapp and Patel2011) and determine the context in which stigma is enacted and experienced (Switaj et al. Reference Switaj, Wciorka, Smolarska-Switaj and Grygiel2009; Thornicroft et al. Reference Thornicroft, Brohan, Rose, Sartorius, Leese and Group2009; Livingston & Boyd, Reference Livingston and Boyd2010; Evans-Lacko et al. Reference Evans-Lacko, Knapp, Mccrone, Thornicroft and Mojtabai2013). In India and other LAMICs, where most people with mental illness do not have access to social welfare benefits, the negative economic consequences of stigma, e.g., through discrimination in work, may be so severe as to threaten the economic survival of entire families (Koschorke et al. Reference Koschorke, Padmavati, Kumar, Cohen, Weiss, Chatterjee, Pereira, Naik, John, Dabholkar, Balaji, Chavan, Varghese, Thara, Thornicroft and Patel2014).

Global patterns of stigma and discrimination

There are few studies comparing the frequency of experiences of stigma and discrimination in different contexts. Recent research has sought to address this gap in the literature. International surveys of experienced and anticipated discrimination among people with schizophrenia (27 countries) and among people with depression (39 countries) found such rates to be consistently high across countries (Thornicroft et al. Reference Thornicroft, Brohan, Rose, Sartorius, Leese and Group2009; Ucok et al. Reference Ucok, Brohan, Rose, Sartorius, Leese, Yoon, Plooy, Ertekin, Mile and Thornicroft2012; Lasalvia et al. Reference Lasalvia, Zoppei, Van Bortel, Bonetto, Cristofalo, Wahlbeck, Bacle, Van Audenhove, Van Weeghel, Reneses, Germanavicius, Economou, Lanfredi, Ando, Sartorius, Lopez-Ibor and Thornicroft2013). Significant between-country variations were found for experienced discrimination, but not anticipated discrimination reported by people with schizophrenia (Thornicroft et al. Reference Thornicroft, Brohan, Rose, Sartorius, Leese and Group2009). A report on the qualitative experiences of stigma and discrimination, however, found few transnational differences (Rose et al. Reference Rose, Willis, Brohan, Sartorius, Villares, Wahlbeck and Thornicroft2011).

On the other hand, some smaller studies suggest stark differences between HIC and LAMIC settings, e.g., studies from China (Chung & Wong, Reference Chung and Wong2004) and India (Koschorke et al. Reference Koschorke, Padmavati, Kumar, Cohen, Weiss, Chatterjee, Pereira, Naik, John, Dabholkar, Balaji, Chavan, Varghese, Thara, Thornicroft and Patel2014), with rates of experienced discrimination much lower than those commonly reported from HIC studies, and qualitative differences in the meaning and appraisal of the experiences made. At first sight, this appears to support the findings of early cross-cultural research on stigma suggesting that the stigma of mental illness may be less marked in non-industrialised societies due to a more supportive environment with more social cohesion and therefore less risk of prolonged rejection, isolation, segregation and institutionalisation (Askenasy, Reference Askenasy1974; Cooper & Sartorius, Reference Cooper and Sartorius1977; El-Islam, Reference El-Islam1979; Waxler, Reference Waxler1979) cited in (Littlewood, Reference Littlewood1998) The better prognosis of schizophrenia found in international studies by the World Health Organization (WHO, 1979; Jablensky et al. Reference Jablensky, Sartorius, Ernberg, Anker, Korten, Cooper, Day and Bertelse1992; Harrison et al. Reference Harrison, Hopper, Craig, Laska, Siegel, Wanderling, Dube, Ganev, Giel, An Der, Holmberg, Janca, Lee, Leon, Malhotr, Marsella, Nakane, Sartorius, Shen, Skoda, Thara, Tsirkin, Varma, Walsh and Wiersma2001; Hopper et al. Reference Hopper, Harrison, Janca and Sartorius2007) has therefore commonly been attributed to less stigmatisation in LAMIC (Rosen, Reference Rosen2003).

However, contradicting this, there is now a considerable body of evidence documenting that in many LAMIC settings, experiences of stigma, discrimination and human rights abuses due to mental illness are common and severe (Phillips et al. Reference Phillips, Pearson, Li, Xu and Yang2002; Thara et al. Reference Thara, Kamath and Kumar2003; Murthy, Reference Murthy, Okasha and Stefanis2005; Lee et al. Reference Lee, Lee, Chiu and Kleinman2005; Botha et al. Reference Botha, Koen and Niehaus2006; Lee et al. Reference Lee, Chiu, Tsang, Chui and Kleinman2006; Lauber & Rossler, Reference Lauber and Rossler2007; Alonso et al. Reference Alonso, Buron, Rojas-Farreras, De, Haro, De, Bruffaerts, Kovess, Matschinger and Vilagut2009; Barke et al. Reference Barke, Nyarko and Klecha2011; Drew et al. Reference Drew, Funk, Tang, Lamichhane, Chavez, Katontoka, Pathare, Lewis, Gostin and Saraceno2011; Sorsdahl et al. Reference Sorsdahl, Kakuma, Wilson and Stein2012; Lasalvia et al. Reference Lasalvia, Zoppei, Van Bortel, Bonetto, Cristofalo, Wahlbeck, Bacle, Van Audenhove, Van Weeghel, Reneses, Germanavicius, Economou, Lanfredi, Ando, Sartorius, Lopez-Ibor and Thornicroft2013). One international study using population-wide data from 16 countries found even higher rates of reported stigma among people with mental disorders in developing (31.2%) than in developed (20%) countries (Alonso et al. Reference Alonso, Buron, Bruffaert, He, Posada-Villa, Lepine, Angermeyer, Levinson, De, Tachimori, Mneimneh, Medina-Mora, Ormel, Scott, Gureje, Haro, Gluzman, Lee, Vilagut, Kessler and Von2008).

In sum, our understanding of global patterns of stigma and discrimination is still rather limited, and further high-quality cross-cultural research is needed to throw light on the forces that drive inter-cultural differences in the manifestation of stigma. Given that there is emerging evidence that not all aspects of stigma and discrimination may be equally subject to cross-cultural variation (Koschorke et al. Reference Koschorke, Padmavati, Kumar, Cohen, Weiss, Chatterjee, Pereira, Naik, John, Dabholkar, Balaji, Chavan, Varghese, Thara, Thornicroft and Patel2014), such research should take into account a range of forms of stigma and discrimination and their context-specific burden and meanings, in order to inform the development of context-specific anti-stigma interventions.

Measurement of stigma and discrimination

The creation and validation of instruments to measure stigma and discrimination against people with mental illness have been underway since the 1960s. Although early scales such as the Opinions About Mental Illness Scale (Cohen & Struening, Reference Cohen and Struening1962; Link et al. Reference Link, Yang, Phelan and Collins2004) and the Community Attitudes Towards the Mentally Ill (Taylor & Dear, Reference Taylor and Dear1981) are still used in some studies, there have been many developments in both the breadth and quantity of measures to assess stigma, reflecting the growing interest in the field (Evans-Lacko et al. Reference Evans-Lacko, Fiorillo, Knapp, Luciano, Park, Brunn, Byford, Chevreul, Forsman, Gulacsi, Haro, Kennelly, Knappe, Lai, Lasalvia, Miret, O'sullivan, Obradors-Tarragó, Rüsch, Sartorius, Švab, Van Weeghel, Van Audenhove, Wahlbeck, Zlati, Mcdaid and Thornicroft2014), incorporation of a wider range of perspectives (especially that of service users and carers (Lee et al. Reference Lee, Lee, Chiu and Kleinman2005; Henderson et al. Reference Henderson, Corker, Hamilton, Williams, Pinfold, Rose, Webber, Evans-Lacko and Thornicroft2014b), and the changing aims and targets of anti-stigma interventions. Nevertheless, studies which include measures developed or validated in LAMICs and/or non-Western European cultures are still rare and only a few include a component focused on stigma which was developed specifically in a LAMIC country and/or non-Western European cultural setting (Thornicroft et al., in press). A recent systematic review (Yang et al. Reference Yang, Thornicroft, Alvarado, Vega and Link2014) assessed studies of stigma in non-Western European cultural groups and found that 77% identified studies assessed stigma using an adaptation of an existing measure which was developed in Western European countries. Moreover, only 2% studies used stigma measures which were derived within a non-Western European cultural group. Even among identified research studies which used a qualitative research methodology, 82% applied generic qualitative approaches which tended to be inductive and which did not incorporate a theoretical framework of stigma.

We know from a growing body of research that stigma and discrimination against people with mental illness is an issue which persists across countries and cultures. Variations in the manifestation of stigma and hence, the ‘cultural validity’ of stigma indicators suggest that measurement of stigma-related constructs also requires local adaptation (Weiss et al. Reference Weiss, Jadhav, Raguram, Vounatsou and Littlewood2001). Although applying a proper translation and incorporating appropriate language or relevant idioms are important for comprehensibility and understanding of questions, forward and back translation of an instrument is not sufficient to address the ‘cultural validity’ of stigma concepts. There may be important differences in: the nature of stigma, what is stigmatised and how it is stigmatised (Weiss et al. Reference Weiss, Jadhav, Raguram, Vounatsou and Littlewood2001) according to different population subgroups. Thus, local cultural adaptation of an instrument may require consideration of a variety of factors including, for instance, geographic region, race, ethnicity, nationality, social class, or other factors which might influence language, beliefs and experiences. All of these are critical considerations for measurement as they also relate to the consequences of stigma and potential targets of anti-stigma interventions.

Currently, there are two approaches which are mainly used when developing culture-specific measures (Yang et al. Reference Yang, Thornicroft, Alvarado, Vega and Link2014). In the first instance, an instrument which was developed in a HIC is translated and then psychometrically validated in a new subpopulation, potentially with some slight adaptations. The second involves the development of a ‘composite measure’ which incorporates experiences which were assessed across a range of contexts and/or cultures. As outlined above, Yang et al. have recently proposed a third approach, that is incorporating a ‘what matters most’ perspective (Yang et al. Reference Yang, Valencia, Alvarado, Link, Huynh, Nguyen, Morita, Saavedra, Wong, Galea and Susser2013; Yang et al. Reference Yang, Thornicroft, Alvarado, Vega and Link2014). This approach would require investigating and operationalising everyday activities which are significant in participants’ lives and which play a role in shaping stigma and/or have consequences for the stigmatised person.

In addition to the issue around cultural adaptation of measurement tools, there are other gaps in what can be assessed using available measures. These include a lack of indicators for structural stigma and which measure stigmatising behaviour, targeted or tailored measures for specific subgroups and use of unvalidated measures. Future efforts should continue to address these gaps, with a particular emphasis in LAMIC countries where literature is sparse, as measurement and evaluation are critical to understand the underlying mechanisms and effectiveness of anti-stigma interventions.

Evidence of interventions to reduce stigma and discrimination

Evidence from qualitative studies and the grey literature

Most studies analysing the outcomes of interventions addressing mental health stigma are quantitative (Thornicroft et al. in press). Within the limited qualitative literature, even fewer studies are carried out in LAMICs (Thornicroft et al. in press). A recent systematic review appraising the global evidence for effective interventions to change mental health stigma revealed a single randomised controlled trial (RCT), with qualitative outcomes, carried out over several countries, including some LAMICs (Wasserman et al. Reference Wasserman, Hoven, Wasserman, Carli, Sarchiapone, Al-Halabi, Apter, Balazs, Bobe, Cosman, Farkas, Feldman, Fischer, Graber, Haring, Herta, Iosue, Kahn, Keeley, Klug, Mccarthy, Tubiana-Potiez, Varnik, Varnik, Ziberna and Postuvan2012). ‘Regarding interventions related to knowledge rather than attitudes or behaviour/discrimination’ (Thornicroft et al. Reference Thornicroft, Rose and Kassam2007), this project was an educational awareness programme and it was developed to meet the mental health needs of adolescents using a mixture of learning modalities. Coordinators within their respective sites answered a semi-structured questionnaire, the responses of which were later analysed and grouped into themes by two independent assessors. Although the intervention was successful in promoting discussion of certain mental health-related topics and encouraging social support networks between students, the weaknesses of the programme were due to the lack of flexibility of the RCT design and the burden on schools delivering the programme.

The use of mass media as a portal for raising awareness has shown positive effects in a number of HICs, including Australia, Canada, New Zealand and the UK (Sartorius & Schulze, Reference Sartorius and Schulze2005). From considering the grey literature (see Table 1), within LAMICs various forms of media have been utilised. In India, as part of the anti-stigma campaign, a group of artists performed a ‘Street play’ in various locations to increase mental health awareness among the public (Table 1). In Egypt, a study by Khairy et al. (Reference Khairy, Hamdi, Sidrak, Sabry, Nasreldin, Khoweiled and Loza2012) evaluating the potential benefits of a mass media campaign concluded that it was somewhat successful in changing attitudes and behaviours of participants that had been exposed to the televised educational clips. However, when changes in attitudes and behaviours were compared between those that had been exposed to the campaign and those that had not, no statistically significant differences emerged (Khairy et al. Reference Khairy, Hamdi, Sidrak, Sabry, Nasreldin, Khoweiled and Loza2012).

Table 1. LAMIC stigma studies identified by search of grey literature

Key: country income categories are: LI, low income; LMI, lower middle income; UMI, upper middle income.

Note: Other countries searched, but without any relevant findings, include the following: (English language) – Ethiopia, Algeria, Sudan, Mozambique, Madagascar, Ivory Coast, Cameroon, Burkina Faso, Niger, Mali, Belize, Indonesia, Pakistan, Bangladesh, Philippines, Vietnam, Iran, Thailand, Myanmar, Iraq, Malaysia, Uzbekistan, Saudi Arabia, Nepal, Afghanistan, Yemen, Sri Lanka. (Spanish and English language) – Colombia, Peru, Venezuela, Chile, Ecuador, Guatemala, Cuba, Bolivia, Dominican Republic, Honduras, Paraguay, El Salvador, Nicaragua, Costa Rica, Panama, Uruguay. (French language) – Benin, Burundi, Central African Republic, Cameroon, Chad, Comoros, Democratic Republic of Congo, Guinea, Haiti, Madagascar, Mali, Niger, Togo, Congo, Côte d'Ivoire, Djibouti, Morocco, Sao Tome & Principe, Senegal, Vanuatu, Algeria, Gabon, Lebanon, Mauritius, Seychelles, Tunisia, Rwanda, Mauritania.

Such interventions in LAMICs have had varying degrees of geographical and governmental involvement. Most interventions, such as those of the Mental Health Foundation in Nigeria and the Kintampo project in Ghana are at a national level. Others, such as the Minds Foundation in India, work regionally to improve mental health literacy through educational workshops within rural communities. Some countries i.e., Liberia and Morroco have demonstrated anti-stigma policies within national policy papers. In Liberia the anti-stigma campaigns aim to promote the rights and needs of those with mental illness through advocacy.

Although many interventions are descriptive, a selected few systematically evaluated the effectiveness using project outcomes (Thornicroft et al. in press). Additionally, a limited number are published in peer reviewed journals or reports (Kakuma et al. Reference Kakuma, Kleintje, Lund, Drew, Green and Flisher2010). Only one study in Jamaica found positive changes in attitudes following the deinstitutionalisation and integration of mental health into primary healthcare (Table 1). The Minds Foundation in India, lacking formal evaluation, however does state that questionnaires have shown an improvement in attitudes towards those with mental health disorders. In overview, there are few stigma-related intervention studies in LAMICs, they tend to be small, they are methodologically rather diverse and do not allow combined interpretation, and there are no cost-effectiveness analyses of these interventions.