Background
A nationwide mental health literacy study reported that people with mental illness (PMI) in Singapore face a considerable amount of stigma [
1]. Mental illness stigma is a complex social phenomenon that often negatively impacts individuals experiencing mental illness. The seminal conceptualisation of stigma by Goffman is typically adopted and defined as a discrediting attribute that carries a mark of shame and greatly reduces social value [
2,
3]. One of the many consequences of stigmatising attitudes towards mental illness is the impact it has on help-seeking and recovery [
4]. For instance, compared to other types of help-seeking barriers, stigma was ranked the fourth highest in a systematic review that included both quantitative and qualitative studies [
5]. People with depression facing these stigmatising attitudes, from the public or experiencing self-stigma, could delay treatment leading to a longer duration of untreated illness; which could have severe consequences such as poorer treatment response, lower rates of remission, increased rates of chronicity and increased frequency of relapse [
6].
Psychiatric disorders are prevalent in Singapore’s population (13.9%), especially among those aged 18–34 years (21.6%) [
7]. Youths were thus identified as a vulnerable group in the Singapore population, who were at a higher risk of developing mental health issues [
8]. In line with international findings, Vaingankar et al. established that psychiatric illnesses typically present during young adulthood in Singapore [
9‐
11]. Furthermore, individuals who were tertiary educated were less likely to seek treatment for their mental disorders [
7].
A significant milestone in a young person’s life is when they matriculate to university, whereby they gain more independence from their family. During this period, youths are exposed to a plethora of psychosocial risk factors (e.g. pressures to succeed academically and loneliness) that could potentially lead to higher risks of major depressive disorder (MDD) and/or generalised anxiety disorder (GAD) [
12]. In addition to the clinical ramifications of these debilitating conditions, university students could suffer consequences in their academic performance which in turn would greatly impact their socioeconomic opportunities (i.e. employment, career, marital status) later in life as well [
4]. By targeting university students, an intervention could potentially foster an attitude of caring towards PMI among young adults who are likely to carry it throughout their adult lives. Hence, it is imperative that the gap in knowledge, and negative attitudes and behaviours towards PMI be addressed among university students, specifically, by cultivating greater understanding about mental health as well as encouraging help-seeking and mutual support among peers and community.
In an effort to alleviate mental illness stigma, short-term interventions that specifically targetted university students have been developed throughout the years, [
13]. The Advancing Research to Eliminate Mental Illness Stigma (ARTEMIS) intervention was developed in-house, by incorporating two key components that Corrigan and Penn stated in their theory of stigma reduction – education and contact [
14]. The core elements of their theory takes into consideration Thornicroft et al.’s conceptualisation of stigma, in which addressing issues related to 1) a lack of knowledge about mental health; 2) negative attitudes; 3) rejecting and avoidant behaviours towards PMI, could help alleviate stigmatising attitudes [
15]. A study by Pang et al. reported that youths in Singapore carry with them several misconceptions of mental illnesses, as well as issues of disclosure and fear of being stigmatised themselves [
16]. The education component of an anti-stigma intervention seeks to provide accurate information and enlighten individuals about the myths and stereotypes that surround depression, and this has been demonstrated as an effective method of reducing personal stigma [
17]. Treatment options and information on sources of help are essential educational content for improving attitudes towards the use of mental health services [
18]. The inclusion of direct contact with someone who has had a history of mental illness has also demonstrated effectiveness in modifying negative attitudes towards individuals with mental illnesses [
17].
Early 1990s saw the rise of advocation for greater awareness of mental illness led by the Institute of Mental Illness (IMH) and the Singapore Association for Mental Health [
19]. Roughly over the next two decades (2000s to mid-2010s), other organisations such as the Community Health Assessment Team (CHAT), Samaritans of Singapore (SOS) and the National Council of Social Service (NCSS) fostered a network of governmental and community agencies to reduce the stigma of mental illness [
19]. However, despite decades of locally executed anti-stigma programmes and campaigns to educate and raise awareness about mental illness, PMI in Singapore still face a significant amount of stigma. As commented by Kuek et al., an extensive literature search on evaluative studies on anti-stigma programmes between the years 2000 and 2019 found no peer-reviewed published articles on any anti-stigma programmes in Singapore [
19]. To the best of the authors’ knowledge, no similar intervention has been developed or evaluated in Singapore. And that the ARTEMIS intervention while similar to those conducted elsewhere such as Ahuja et al. and Friedrich et al. [
20,
21], was locally adapted to address concerns of youth in Singapore as well as give them detailed information on help-seeking in the local context. Therefore, the purpose of this study is to evaluate the effectiveness of an anti-stigma intervention towards depression on the community attitudes towards mental illness (CAMI) factors as well as to observe any sustained effects after 3-months; to locally validate the CAMI among university students; and to examine the sociodemographic correlates of the CAMI among university students in Singapore.
Discussion
This paper examined the effectiveness of the ARTEMIS intervention on the 3-factor model of the community attitudes to mental illness (CAMI) scale towards depression. Significant shifts in attitudes towards depression was observed across the timepoints (pre-intervention, post-intervention, and 3-months follow-up). When comparing baseline and post-intervention scores, favourable attitudes (i.e. lower scores on authoritarianism, and higher scores on benevolence and community mental health ideology) were attained in all 3 factors of the CAMI. It is established that the lack of mental health literacy (i.e. knowledge) is closely associated with greater stigmatising attitudes towards mental illness [
28‐
30]. This is in line with findings of previous studies that incorporate education as a component of their respective anti-stigma intervention to address the lack of knowledge [
17,
20,
31]. For instance, the education component of the ARTEMIS intervention incorporates a biopsychosocial approach of depression for university students in Singapore. In a study by Han et al., they found that biological attribution of depression saw greater improvements in help-seeking intentions, whereas de-stigmatisation information helped reduced negative appraisals of people with depression [
32]. Aside from the education component, social contact via the sharing session by a person with lived experience could have contributed to the shifts in attitudes as well [
33]. According to the intergroup contact theory, as part of the four optimal conditions, equal status needs to be established to reduce prejudice between groups [
34]. Therefore, the social contact component of the intervention could have helped breach the status gap between groups, which could have led to alleviated levels of anxieties surrounding depression as well as an increased sense of empathy towards PMI [
33,
34]. Wood and Wahl recruited undergraduate students for a mental health education programme known as In Our Own Voice (IOOV) [
31]. When Wood and Wahl compared their participants who attended IOOV against their control group, they found significant improvements in attitudes towards PMI.
On the other hand, significant findings were found between post-intervention and at 3-months follow-up as well; although there is a reversal of the results which had previously shifted. The analysis indicated that authoritarianism scores increased, whereas benevolence and CMHI scores decreased when comparing post-intervention and 3-months follow-up timepoint. However, when comparing mean scores of the 3 factors between baseline and at 3-months follow-up, it seems to suggest that scores at the follow-up reverted to baseline scores. The strength of one’s attitude could be a possible explanation for the reverted scores. Attitude strength is a notion that describes how strongly one feels about a specific person, object or concept [
35]. Petty et al. had established a correlation between attitude strength and the likeliness of being persuaded – where the stronger the attitude the individual holds, the more resistant it is to change. It could be reasoned that participants possessed strong attitudes towards people with depression prior to the intervention. Supplementing the fact that the study took a convenience sampling approach to recruitment, a potential for self-selection bias might be in play, thus the participating university students might already have vested interest in the topic. Although being resistant to change due to the strength of one’s attitude does not account for the immediate shift in favourable attitudes from baseline to post-intervention. This trend indicates that the ARTEMIS intervention was able to elicit a short-term favourable shift in attitudes towards depression but was not able to achieve any lasting effects. As reported by Thornicroft et al., although short-term improvements in knowledge and attitudes could be observed immediately after an anti-stigma intervention, the effects would weaken over time [
36]. To resist changes in attitudes across time, McGuire proposed the inoculation theory, two types of bolstering efforts which would maintain the intervention effects: 1) providing information for active recollection of the session, and 2) subsequent exposure to less complex follow-up sessions [
35]. Therefore, future anti-stigma interventions should be mindful of temporal influences and incorporate McGuire’s proposed efforts to maintain lasting and favourable shifts in attitudes following the intervention.
In comparison with other anti-stigma interventions, the ARTEMIS intervention seems to report similar trends. For instance, in Ahuja et al., the combined utilisation of education and social contact in their anti-stigma intervention for college students (18–21 years) in India observed favourable shifts in CAMI scores immediately after the intervention [
20]. Their follow-up period of 1-week also saw sustained favourable attitudes towards PMI, although it would be important to note that compared to the ARTEMIS intervention their follow-up period was much shorter. Friedrich et al. also reported similar results, where medical students (mean age = 23.5) underwent the Education Not Discrimination (END) component of the Time to Change anti-stigma programme to reduce mental illness stigma among medical professionals and trainees [
21]. Immediately after the intervention that included both education and social contact components, favourable shifts in stigma-related attitudes were observed. However, unlike Ahuja et al. but more similar to the current study, they saw no sustained effects of the intervention except for a single item on the CAMI “There is something about people with mental illness that makes it easy to tell them from normal people”. This study, thus, evinced that the combined components of education and social contact could result in immediate favourable shifts in attitudes among university students in Singapore.
The study found that gender, nationality, having experience in a mental health field, as well as close social contact (i.e. friends and family members with mental illness) could influence the CAMI factors. Male students were less supportive of the community mental health ideology dimension towards people with depression when compared to female students. In a study in Spain which was conducted among participants aged 14–18 years, similar gender differences were detected; and Vila-Badia et al.’s findings align with the current paper which found that male students had lower community mental health ideology scores than female students [
37]. Savrun et al. postulated that female students possessed more “optimistic values” towards treatment for PMI than male students [
38]. Individuals who were on student visas were less supportive of community mental health ideology. As attempts at deinstitutionalisation occur gradually, it is possible that traditional psychiatric mental health care (i.e. asylums and institutionalisation) is still pervasive among the country of origin which the students with visas are from; and thus more familiar to them [
39,
40]. Regarding having experience in the mental health field, the results indicated a favourable shift across the CAMI factors. This supports findings from other studies that looked at healthcare professionals and volunteers, which found lower stigma scores after exposure and experience in a mental healthcare environment [
34,
41,
42].
Across the 3 CAMI factors, having a close social contact with PMI resulted in more favourable mean scores than those without. The current findings are consistent with previous studies that suggest individuals with a close social contact with PMI reported lower levels of stigmatising attitudes [
43,
44]. In contrast, the linear effect observed a smaller decrease in authoritarianism scores among university students with close social contact than those without. Perhaps they are more aware of the realities of caring or befriending a PMI, and as such were less influenced by the intervention [
35]. Quadratic effects were observed for both benevolence and CMHI, indicating a greater decrease in scores from post-intervention and at 3-months follow-up. External influences such as recent negative interaction with PMI during the 3-months intermission could provide a plausible explanation for this observed shift for benevolence. Nevertheless, it would be important to highlight that participants with close social contact continue to show more favourable attitude mean scores across the 3 CAMI factors than their counterparts (Fig.
1-
3).
Limitations
A limitation of this study is that the participants were university students recruited from a single local university, and thus would not be considered representative of all university students as there could be unique university cultural influences at play that were not accounted for. As the university students were voluntary participants, baseline scores could be skewed towards more favourable attitudes because of motivation and interest towards the topic as mentioned above. Moreover, seeing as the students come from a diverse background of disciplines and courses, the concept of mental health and stigma might have been a topic covered in certain courses which was not accounted for in this study. Furthermore, not reporting the response rate is a limitation of the convenience sampling method adopted in this study.
Conclusion
To conclude, the anti-stigma intervention towards depression observed promising short-term favourable shifts across the CAMI factors. After adjusting for sociodemographic correlates, short-term effects of the anti-stigma intervention were significant. However, the intervention did not result in any sustained effects, possibly due to cognitive biases, attitude strength, and the lack of additional sessions of the intervention. Sociodemographic correlates such as gender, nationality, and experience in a mental health field as well as having a close friend/family member with mental illness were identified as significant correlates. Future research exploring whether additional and less complicated follow-up sessions would have any sustained long-term effect on stigmatising attitudes should be considered. Perhaps future interventions could target areas of mental health which is most prevalent in the community or relevant to the environment and slowly integrate more topics opposed to a general “mental illness” approach. Additionally, stigma is highly influenced by the culture of the community, conducting qualitative studies into the specific culture of the university and of the students’ respective community could evoke nuanced insights of the stigma faced. Other possible avenues to explore include the comparisons of anti-stigma interventions focusing on 1) different mental illnesses; 2) other local universities; and 3) other populations.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.