Commentary on ‘The Relationship Between Public Causal Beliefs and Social Distance Toward Mentally Ill People’

Significance of topic

Dietrich et al.'s paper addresses a topic of critical importance. There is evidence that stigma is the leading issue of concern among people with mental illness [1], [2]. Stigma is thought to be an important factor in low levels of help-seeking by people with a mental illness [3]. There is also evidence that it leads to lower self-esteem [4], increased psychological distress [5] and discrimination in employment and accommodation [6].

There is a clear public health need for interventions that reduce the stigma associated with mental illness. Public destigmatization programs have been developed. For example, a community awareness program was carried out in Australia under the National Mental Health Strategy at a reported cost of $8 million [7]. However, both from the point of the individual affected by mental illness and that of the tax payer who funds them, antistigma programs should be founded on a sound evidence base, including an understanding of aetiology [7].

Issues raised by the study

Dietrich et al.'s study has the potential to contribute to our understanding of the aetiology of stigma and to the evidence-base for the development of optimal interventions for reducing this stigma and discrimination.

However, two questions are important. First, how replicable, valid and generalisable are the findings of the current study? Second, what lessons are to be learned from this study about future directions for research and how to modify current stigma awareness campaigns?

One potential problem in interpreting the study's findings relates to the extent to which the descriptors used in describing biological causal attributes reflect the terms most commonly used in Western health promotion programs. In particular, the authors use the term ‘brain disease’ as a casual attribute, whereas most public education material use phrases such as ‘biochemical imbalance in the brain’. At least for depression, the connotations of brain disease may be quite different from that of biochemical imbalance, both in terms of the degree of control an individual may exercise over their behaviour (greater in chemical imbalance) and the extent to which the condition might be treatable. Interestingly, Martin et al. [8], who did not find that biological attributions led to greater social distance, used the descriptor ‘a chemical imbalance in the brain’ in their study of the relationship between social distance and causal attributions of mental illness.

The authors themselves raise other potential complications in interpreting their results. They note for example, that the results might be due to the indirect effect of a third factor that affects both attribution type and social distance, for example certain personality characteristics. The authors also note the relatively low level of variance in social distance explained by causal attributes. This is important if the negative effect of these attributions is outweighed by the positive effects in terms of appropriate help-seeking actions and positive mental health outcomes. For example, although a biological explanation may be associated with greater social distance, it may propel the person with depression to seek appropriate and effective treatment with a ‘biological intervention’ such as antidepressant medication. The paper does not provide clear detail of the relative contribution of each of the causal attributes studied and this would be welcome.

Another issue of concern is the high degree of overlap in endorsement of different causal attributes. The fact that for schizophrenia, 72–87% of respondents endorsed life events, and 70–71% brain disease (for depression 81–85% endorsed life events, and 38–59% brain disease) suggests that a significant percentage of respondents endorsed both biological and environmental causes. If these causal attributes are highly correlated, we need to interpret the study findings cautiously. There is no information about the percentage of people who endorsed both brain disease and other causal attributes. However, if a majority of people did this, the regression analysis might have had the effect of excluding them and focusing on the minority of people who held unitary views about brain disease and mental illness. Regardless of whether it affected the overall findings, the nature and degree of overlap in causal attributions is of interest in its own right. It has been argued by some that public health messages should emphasize both biological and environmental influences and their interactive effects [9]. It would be of theoretical and practical interest to know whether the desired social distance of people endorsing multicausal views differs from that of people who hold a unitary view of causation.

Finally, the study focuses on the participants' views about causes of depression and schizophrenia in the community. However, the paper provides no evidence concerning the effect of causal attributions on selfstigma. This term refers to the stigmatizing views a person with mental illness holds about their own illness. Self-stigma is an important issue because it may affect self-esteem [10] and help-seeking. We do not know if causal attributions operate the same way for self-stigma as they do for stigma directed toward other people. It is even conceivable that the mechanisms operate in different (opposite) directions. On the basis of a qualitative study of 49 women with depression, Schreiber and Hartrick [11] suggested that internalization of the biological explanatory model decreases internal stigma, allowing the person with depression to conclude that it is not their fault and thereby to reduce their sense of guilt and shame. Women in the study were also reported to attempt to use the model to ease the stigma directed toward them by others, but with little success. More particularly, biological explanations had little effect on the perspectives of significant others, in that they continued to insist that the depressed person simply ‘pull up their socks’. Interestingly, one study has reported that people with schizophrenia who attribute their condition to biological/physical causes rather than ‘mental illness’ show a trend toward enjoying a greater quality of life and better social relations than their counterparts [12].

Areas for new inquiry and research

There is evidence that some educational interventions are effective in reducing stigma [see [13],[14]. However, little is known about what specific components of such programs are most effective and in particular the extent to which endorsement of biological causes are likely to be helpful (or harmful).

Arguably, we need well-controlled intervention studies that explore this question. Such studies would need to consider multiple outcomes (help-seeking and attitudes, as well as social distance) and focus on self-stigma among consumers as well as stigma and discrimination in the general community. The design might involve a randomised controlled trial in which participants are provided with either a biological or a social explanation of depression or an explanation involving both. Of course, it is quite possible that effects of information about causal attributes could be modified by other information. For example, data on the prevalence of violence among people with depression might reduce stigma associated with a biological explanation by addressing concerns about loss of control. Information about why non-biological treatments might be effective for a ‘biologically determined’ illness could influence the credibility of explanations. Developing an evidencebased intervention that optimizes these interactive effects could be a considerable undertaking. Moreover, ensuring the ecological validity of an intervention may be problematic given the complexity of inputs in a community setting.

Conclusion

Further research is needed to investigate the issues raised by Dietrich et al. Meanwhile, their paper makes an important contribution by pointing to the serious gap in the evidence base that underlies public awareness campaigns. In doing so they remind us that such interventions should, like clinical treatments, be based on evidence of effectiveness.