Stigma related to mental health issues – a study among adolescents in Sri Lanka

One definition of stigma is that it is a “collection of negative attitudes, beliefs, thoughts, and behaviours that influence the individual or the general public to fear, reject, avoid, be prejudiced, and discriminate against people” (1). Stigma impedes treatment seeking, erodes self-esteem and limits the individual’s social network. Although stigma was described as a unitary characteristic in earlier studies, currently it is understood as being more complex and having many dimensions and attributes (2-4). Culture also influences the way stigma relating to mental illness is presented (5).

Stigma related to mental health issues -a study among adolescents in Sri Lanka outcomes (13). Thus, reducing stigma and promoting early help seeking is essential in improving treatment results.
Although stigma among adolescents has been studied amongst western populations, there are very few studies conducted amongst Asian populations (3,(14)(15)(16). According to the most recent population and housing census In Sri Lanka, adolescents (aged between 10-19 years, both male and female) account for approximately 16.2% of the population (17). While there are no major studies on this subject among adolescents, a study among undergraduates in early adulthood concluded that creating greater awareness about mental health issues improved help-seeking (18). Therefore, the objective of this study is to describe several attributes of stigma related to a selection of mental health issues amongst school going adolescents.

Methods
This cross-sectional study was conducted in the Sri Jayewardenepura educational zone in the Colombo District. Those in grades 9 and 10 were chosen for the study. The minimum sample size was calculated to be 844. Details of sample size calculation can be found in a previously published study (19). A multistage cluster sampling method was used as this education zone included several categories of schools, with different resource levels and different streams of study, and a total of 46 schools were selected for the study (19).
Questions to assess stigma related to mental health issues, were adapted from the Attribution Questionnaire-C-8 (AQ-C-8), with the author's permission (20). This version of the questionnaire has been formulated for children and applied on adolescents between 10-18 years to assess for public stigma (20). The attribution questionnaire uses a collection of attributes about people with mental health issues (such as pity, shame, guilt, level of perceived danger, likelihood of helping or staying away from them). The selected attributes were self-scored by participants on a pre-coded 9 response Likert scale, with a higher score indicating that the given attributes were being more strongly endorsed by the respondent (i.e., score of 9 -strongly agree and a score of 1 -strongly disagree).
The participants were asked to score these attributes in relation to four case vignettes. These vignettes were modelled on the Australian National Survey on Mental Health Literacy and Stigma 2011 (15). The four case vignettes used in the current study were: depression with suicidal ideation, social phobia, psychosis and diabetes mellitus. A vignette on a physical health condition (diabetes mellitus) was included for comparison. The content and face validity of the questionnaire was established using the Delphi method. A pilot study was also conducted to establish acceptance, comprehension and ease of administration (19). The data was analysed for frequency distribution, crosstabulation and median comparisons. A p value of 0.05 was used as the significance level.

Ethics
Consent and approvals have been obtained from the respective education offices of each zone and the principals of the selected schools. Ethical clearance for the study was obtained from the Ethics Review Committee (ERC) of the University of Colombo, Sri Lanka. The data collector visited the schools and obtained prior consent (informed and written) from the parents of students selected for the study. Information sheets about the study were also shared with the consent forms. Only those adolescents with parental consent were included in the study. Prior assent from participating adolescents was also obtained.

Results
Of the 1,500 students approached, 1,002 responded (response rate of 67%). There were 590 (58%) males in the sample with a mean age of 14 years (SD ± 0.94). An average monthly family income of more than LKR 20,000 was seen among 59.4% of the sample. Of the sample, 9.6% (n=97) responded positively to knowing a relative with mental health issues (family history), while 5.2% (n=52) responded positively to being treated for a mental health issue themselves (personal history). The central tendencies (mean and median scores), for the different stigma related attributes are listed in Table 1. A comparison of the median scores of each attribute between the different vignettes showed that there was a significant difference only for the attributes of blame, anger, willingness to help and preference to stay away (p<0.05) ( Table 2). The overall scores for anger and staying away were significantly higher (p<0.05) for psychosis compared to depression or social phobia ( Table 2).
Comparison of median values in relation to family income (categorised as below and above LKR 20,000), the education level of the mother and father (categorised as above and below the General Certificate in Education -Ordinary Level exam) did not show significant difference in relation to any of the attributes.
A family history or a personal history of being treated for a mental health issue in the past, was associated with a significantly higher median score for the attributes of avoidance, anger, and danger across all three mental health vignettes (p< 0.05) ( Table 3). A personal history was also significantly associated with lower median scores for pity and willingness to help (p< 0.05) ( Table 3).

Stigma in relation to the different case vignettes
In relation to all mental health case vignettes, there were higher scores for pity and helping the affected person and lower scores for danger, anger, and avoidance of the person, compared to the diabetes health vignette. Overall, this indicated that mental health issues were perceived as requiring more help, compared to those with diabetes. Interestingly, the responses also indicated that those with mental health issues were more likely to be offered help than those with diabetes, which was a good sign in relation to mental health issues.
The responses indicate that this cohort is more likely to avoid those with psychosis than those with depression or social phobia. Although one would expect the attribution of danger to be associated with avoidance, this was not indicated in the current study. Previous studies have shown that those with psychotic disorders and/or substance misuse are more stigmatized compared to those with other mental health issues (21). In another study on stigma among adolescents, relative to depression, psychosis was associated with higher scores on the dimensions of 'dangerous/unpredictable'and 'social distance' (14). In the 2011 National Mental Health Literacy survey in Australia (youth component), perceptions of danger, unpredictability and a desire for social distance were generally higher for psychosis/ schizophrenia, than for other disorders (15). However, in the current study the scores indicate that there was also significantly more anger in relation to psychosis than other mental health vignettes. Therefore, it is possible that in this study the reported avoidance of those with psychosis may be more related to anger than danger.

Stigma expressed by those who know a family member with illness, or with a personal history of mental health issues
In the current study, knowing a relative with a mental health issue, or having a personal history of being treated for a mental health issue was associated with a significant increase in the scores for the attributes of anger, danger and avoidance across all three mental health case vignettes. A personal history of being treated for mental health was associated with significantly lower scores for pity and the likelihood of helping. Although previous evidence has reported that knowledge of mental health issues and exposure to them has been found to lower stigma, some research suggests that knowledge and contact may also increase certain attributes of stigma (7,20). The findings of the current study appear to support this latter view. The findings of this study suggest that it may not be mere exposure to mental health patients that reduce stigma. Factors such as the length of exposure, nature of experience (positive or negative experience), perception of treatment (beneficial and acceptable), and whether they recovered from their condition successfully, may all be important factors influencing stigma. A previous study reported that stigma was lower for an individual who suffered from major depression and recovered successfully after treatment, compared to an individual who was not treated (22). In another study undergraduates with a history of mental health treatment were assigned as room-mates to those with no mental health history. The stigma in the latter group stigma towards the former often increased, highlighting that naturalistic contact alone, if not structured appropriately, might be more harmful than helpful (23).
Thus, it might be that in this study sample, those with family and personal history of mental health issues did not have positive experiences/ outcomes, which may have influenced the degree and nature of stigma they have towards those with mental health issues.

Limitations
The responses of the participants may not necessarily indicate how they would respond in reality. This has been described as an issue in vignette based studies (24). As some of the data was retrospective, recall bias as well as courtesy bias may have influenced the responses. It is also likely that there are more attributes and dimensions of stigma than the ones used in this study. As various studies have used multiple tools with different dimensions and attributes, comparisons between studies is also difficult.

Conclusions
The findings of this study suggest that adolescents in this cohort are more likely to feel pity and help in the cases of mental health issues, compared to the case of the given physical health issue. They were also more likely to avoid those with psychosis, than those with other mental health issues. Thus, interventions aimed at reducing stigma could make use of this sense of pity and tendency to help, to promote directing those with mental health issues towards appropriate services. However, strategies will also have to consider ways of changing perceptions about psychosis related disorders, as avoidance by peers may lead to poorer help seeking where early intervention is most needed.
As seen in other studies, knowledge and exposure does not automatically reduce stigma towards those with mental health issues. In contrast, this study showed that merely knowing a person with mental health issues or having been treated for mental health issues can have a negative effect on stigma related attributes. As such, stigma mitigating programs may have to include aspects that go beyond mere exposure to those with mental health issues. A positive experience of the acceptability and benefits of treatment helping in recovery might be aspects to consider.