Towards a framework for increasing help-seeking for social anxiety disorder

Introduction

Social anxiety disorder (SAnD) is a condition in which the individual experiences persistent, excessive fear in social and performance situations, leading to their avoidance, or intense distress and impaired role functioning, including social, educational and occupational, and routine functioning (American Psychiatric Association, 2000). Estimates of the prevalence of social phobia vary markedly (Fehm et al., 2005; Furmark, 2002; Somers et al., 2006). However, the condition is clearly common in many western countries, with nationally representative samples yielding Diagnostic and Statistical Manual of Mental Disorders (DSM) lifetime and 12-month prevalences of 8.7% (McEvoy et al., 2011) and 4.7%, respectively, in Australia (Slade et al., 2009), 7.8% and 4.8% in The Netherlands (Bijl et al., 1998) and 12.1% (Kessler et al., 2005a) and 6.8% (Kessler et al., 2005b) in the USA. SAnD is one of the most prevalent of the anxiety disorders (Bijl et al., 1998; Kessler et al., 2005b; McEvoy et al., 2011).

SAnD typically emerges in childhood or adolescence with a reported median age of onset of 13 years in Australia (McEvoy et al., 2011) and 16 years in the USA (Magee et al., 1996). The condition involves a chronic course with a mean duration of at least 20 years (Canadian Psychiatric Association, 2006) and a high rate of comorbidity with other mental disorders (Furmark, 2002). It is more prevalent in women than men (Furmark, 2002; McEvoy et al., 2011; Somers et al., 2006). SAnD exerts a profound, negative impact on quality of life, is a significant risk factor for the development of major depressive disorder, and is associated with substance misuse (Stein and Stein, 2008) and increased levels of suicidality, even in the absence of comorbid depression (Fehm et al., 2005). SAnD is also associated with significant societal costs; for example, one study reported that 22% of SAnD participants were on disability or welfare benefits compared to 10% of controls (Furmark, 2002). At an individual level, the severity of disability associated with pure SAnD is as high as that associated with pure depression (Fehm et al., 2005). For example, the work loss index for SAnD adjusted for comorbidity is the same as for affective disorder and exceeds that for significant physical illness such as diabetes and heart disease (Alonso et al., 2004). Sub-threshold social anxiety is also associated with substantial disability, leading some researchers to propose the use of a dimensional rather than a categorical approach to the assessment of the condition (Filho et al., 2010).

Despite the distressing nature of SAnD, the availability of effective psychopharmacological and psychological treatments for the condition, and its unremitting nature if left untreated, only a minority of individuals with SAnD seek professional treatment (Grant et al., 2005; Magee et al., 1996; Ormel et al., 2008; Schneier et al., 1992). Across the nine high-income countries sampled in the World Mental Health Survey, only 20.8% of individuals with SAnD reported seeking professional help (Ormel et al., 2008). Moreover, it has been reported that only 7% of people with SAnD receive ‘notionally effective treatment’ (Andrews et al., 2004). It has been calculated that with 70% and 100% coverage using optimal current treatments for SAnD, this currently very low level of disability prevention could be increased to 34% and 49%, respectively (Andrews et al., 2004). Thus, there is a clear need to promote help-seeking and access to evidence-based treatments among individuals with SAnD. However, to date, SAnD has been relegated to the role of Cinderella compared to depressive disorders, with public awareness campaigns focused almost exclusively on promoting help-seeking for depression.

The current state of knowledge about help-seeking for SAnD

Despite the critical importance of promoting evidence-based help-seeking for SAnD there is currently little empirical evidence on which to base public health initiatives to facilitate help-seeking.

A recent systematic review of help-seeking for depression, anxiety and psychological distress found no randomised controlled trials (RCTs) of a help-seeking intervention for SAnD (Gulliver et al., 2012). We have since identified an RCT of the effectiveness of a motivational enhancement technique (MET) designed to increase cognitive behaviour therapy (CBT) use among non-help-seeking individuals with SAnD (Buckner and Schmidt, 2009). MET comprises a combination of feedback about symptom levels and motivational interviewing, which aims to increase the recipient’s motivation to make a behavioural change (Miller and Rollnick, 2002). Relative to control, the MET intervention completers were more likely to seek CBT. However, only 12 participants were randomised to the MET intervention, of whom only five completed a post-intervention survey. Moreover, the participants were university students, the MET intervention involved three face-to-face sessions and the study was promoted as ‘An Interview Study of Anxiety’. The research interview concept is not suitable for translation into practice and it is difficult to envisage such a face-to-face, time and resource-intensive approach engaging a broad cross-section of non-help-seeking members of the public with social anxiety.

To our knowledge there have been no other studies of factors which facilitate help-seeking among individuals with SAnD. There have, however, been two retrospective, self-report studies of the individual factors which impede such help-seeking (Chartier-Otis et al., 2010; Olfson et al., 2000). Among the barriers reported were the individual’s lack of knowledge about where to seek help (Olfson et al., 2000), their belief that help would not be effective (Chartier-Otis et al., 2010; Olfson et al., 2000) or that they did not have an anxiety disorder (Olfson et al., 2000), the belief that they could cope with the situation themselves (Chartier-Otis et al., 2010), their fear of taking medications, and their perceived and internalised stigma (Chartier-Otis et al., 2010; Olfson et al., 2000). Other reported barriers were financial constraints (Chartier-Otis et al., 2010; Olfson et al., 2000), an inability to ‘get through to the healthcare professional’s office by telephone’, lengthy waiting times for treatment, lengthy travel times between home or work and the health provider’s office, and work or child care commitments that prevented the individual from accessing help (Chartier-Otis et al., 2010). In addition, avoidance behaviours inherent in SAnD may be important barriers to help-seeking (Chartier-Otis et al., 2010). Other research suggests that the detection of social anxiety is very poor among primary care providers (Wagner et al., 2006). Thus, overall, the results suggest that barriers to help-seeking for social anxiety include poor mental health literacy, stigma, a belief in self-reliance, low levels of accessibility and inadequate provider recognition of the problem.

These findings are consistent with community-based research that has identified low levels of mental health literacy and high levels of stigma concerning SAnD among the public. In particular, a recent Australian study reported that only 3% of young people aged 15–25 years (Reavley and Jorm, 2011c) and 9.2% of adults (Reavley and Jorm, 2011a) were able to correctly label social phobia. This is substantially lower than the levels of identification of depression in these groups (75%; Reavley and Jorm, 2011a, 2011c). Further, two studies have reported that members of the public are more likely to attribute SAnD than other anxiety conditions or depression to personal weakness (Coles and Coleman, 2010; Reavley and Jorm, 2011b). Over one-third of a sample of US university students attributed social phobia to personal weakness compared to 4.3–6.1% for other anxiety disorders and depression (Coles and Coleman, 2010). A representative national survey of Australian adults found that 41.7% of respondents believed that people with SAnD were unpredictable, 20% that a person with SAnD should ‘snap out of it’ and 15.5% that they were dangerous (Reavley and Jorm, 2011b). Further, 29.5% of the sample indicated that they would not tell anyone if they had SAnD, a significantly higher figure than for depression (22.7%) (Reavley and Jorm, 2011b). Moreover, 55.8% anticipated that a person with social phobia would be ‘discriminated against’ and more than half believed that other people would think that a person with SAnD was weak and that they should ‘snap out of it’.

Framework for increasing help-seeking

As described above, little is known about the means by which help-seeking for SAnD might be increased among members of the community. Many health behaviour change theories are relevant to the problem, as is a four-step framework for mental health help-seeking developed by Rickwood and colleagues (2005). However, these models lack the detail required to inform the specific content of an intervention for increasing help-seeking for mental illness, and social phobia in particular. Accordingly, here we present a preliminary guiding framework for the development of an intervention to increase help-seeking for SAnD. It draws from and integrates into a single schema existing work in the field, including data on self-reported barriers to help-seeking for SAnD, the quantitative evidence of poor SAnD literacy and high levels of stigma towards SAnD in the community, together with converging evidence from studies of the barriers to and effective interventions for promoting help-seeking for other anxiety and mental disorders (Gulliver et al., 2010, 2012).

On the basis of this evidence, it can be postulated that a number of factors might increase the likelihood of help-seeking for SAnD. These are shown in Figure 1. It should be noted that the processes which link these elements are unlikely to be linear or invariant across individuals. Nor is the framework intended to imply that it is necessary for all elements to be satisfied for help-seeking to take place. Furthermore, the framework in Figure 1 is consistent with existing behaviour change theories such as Ajzen’s Theory of Planned Behavior (Ajzen, 1991), which posits that positive beliefs about a behaviour, salient referents (social pressure) and perceived behavioural control predict intentions which in turn predict behaviour.

Figure 1.

SAnD help-seeking behaviour framework.

Figure 1 is based on the assumption that the likelihood of help-seeking for SAnD will be a function of knowledge, beliefs, attitudes, the availability of help, and the extent to which the illness itself impedes help-seeking and intentions to seek help. It follows that intervening to improve knowledge (A), beliefs (B) and attitudes (C), and to increase the accessibility of help (D) may increase help-seeking for SAnD, as may interventions designed to increase willingness to seek formal help (F).

More specifically, individuals with SAnD need knowledge about the symptoms of SAnD so that they can recognise their condition, be aware of effective treatments for SAnD and where they can be obtained, and understand the time course of SAnD if it remains untreated (A). This information should increase the probability that individuals believe that a SAnD diagnosis is applicable to them, that there is a need to treat this condition, that there are a number of effective treatments and, in particular, that medications (which are often seen as a barrier to treatment) are not their sole treatment option, that medications are not addictive and that the benefits of treatment exceed the costs (B). This in turn may improve attitudes (C) and increase willingness to seek help from formal sources (F). Not all individuals will self-identify their social anxiety condition from the symptom lists. They will require feedback about their symptom level and diagnosis (B). This information is also likely to increase individuals’ capacity to articulate the nature of their problem when seeking help (G)

Stigma may adversely impact on help-seeking willingness in individuals with a mental illness (Barney et al., 2006). As detailed above, there is evidence that there are high levels of stigma attached to SAnD. It follows that an intervention designed to reduce the stigma associated with social anxiety and to promote skills for coping with this stigma (C) might increase help-seeking intentions from formal sources. Interventions targeted at improving mental health literacy (Corrigan et al., 2001), reducing myths about mental illness (Gulliver et al., 2013) and providing contact with consumers (Corrigan et al., 2001) have been effective in reducing the stigma associated with other mental illnesses. They may also prove effective for SAnD.

Help-seeking may also be increased by providing a means for increasing the geographical availability, affordability and convenience of treatment (D) and finding a means to remove the impediment to help-seeking associated with avoidance due to the illness itself (E). Further, once an individual has reached a point where they are willing to seek help there is a need to ensure that this intention is translated to help-seeking behaviour (F), that they are able to articulate the nature of their problem if they seek face-to-face help (Rickwood et al., 2005) and that they commence and adhere to treatment (G).

There are a number of ways in which points D to G can be addressed. Policy changes and additional funding could increase the availability and affordability of face-to-face services and training might increase the recognition by health practitioners of SAnD and the importance of treating it (D). As described in more detail below, the Internet also has the potential to remove structural (D) and illness-related barriers (E) and to promote help-seeking behaviours (F). The framework suggests that interventions should not only be targeted to the individual with SAnD but also to the general public with the aim of equipping family, friends and colleagues with the skills and confidence to promote help-seeking in others (see D: Instrumental facilitation). Further, Motivational Interviewing may prove a useful technique for increasing an individual’s willingness to seek help for SAnD (F) and to adhere to the course of treatment (G) (Miller and Rollnick, 2002). Thus, for example, individuals with SAnD might be encouraged to consider their goals and how treating or not treating SAnD might be expected to impact on the achievement of these (Buckner and Schmidt, 2009). Face-to-face Motivational Interviewing is unlikely to be a feasible, scaleable approach to intervening to increase willingness to seek help for SAnD at a point prior to face-to-face presentation (F). However, such techniques have been successfully adapted to distal and, in particular, online delivery (Kay-Lambkin et al., 2009) and could be incorporated into an online intervention for promoting help-seeking for SAnD.

The Internet: a potential delivery mechanism for increasing help-seeking for SAnD

The Internet provides an unparalleled opportunity to increase the accessibility of educational interventions designed to promote help-seeking for health conditions such as SAnD. Potentially, such interventions could be conveniently provided in school and university settings and workplaces; for example, as part of a student or work induction program.

However, the Internet also has the potential to increase the accessibility of evidence-based treatments for decreasing the symptoms of SAnD. To date, a number of trials have demonstrated the efficacy of Internet-based social phobia treatments (Andrews et al., 2010; Griffiths et al., 2010). Such treatments can be delivered at low cost and at a time and place that is convenient to the user (D) and in a fashion (anonymously) that can minimise the barriers associated with stigma (C). However, the Internet may offer an additional, more distinct and critical advantage to individuals with SAnD. Conventional treatment requires face-to-face social interactions which individuals with SAnD may avoid as a consequence of their illness. An online application requires no such commitment. Indeed, an automated intervention requires no interactions of any kind with another individual. Thus, the Internet may particularly be associated with a lowered level of illness-related avoidance (E).

In summary, universally available online interventions for promoting help-seeking for SAnD may reduce the treatment gap for SAnD as may the availability of evidence-based automated online SAnD therapy programs. However, it is unlikely that either program alone will be sufficient to address the treatment gap. Certainly, there is no evidence that the existing online SAnD therapy programs have made significant inroads into increasing help-seeking for SAnD at a population level. Rather, the key to tackling the gap may well be a seamless connection between the online educational intervention to promote help-seeking and the online therapy program to provide instantly accessible treatment for SAnD. It may be this seamless connection which ensures that positive help-seeking attitudes and intentions induced by the help-seeking intervention are translated into appropriate (evidence-based) help-seeking behaviour. The medium of the Internet is ideally suited to such a combined intervention approach since it can be effected by means of a single hyperlink to the SAnD self-help program from the online help-seeking program after the presentation of the knowledge, symptom feedback, destigmatisation and motivational interviewing components of the intervention.

Conclusion

Social anxiety disorder is a common, chronic and highly debilitating condition which interferes substantially with an individual’s life and which is largely untreated in the community. There is a need to develop evidence-based programs for improving help-seeking for SAnD. The preliminary framework presented in this paper provides broad guidance for developing such a tool to reduce the prevalence of unmet needs for evidence-based social phobia treatment in the community. The author is currently developing an Internet version of such a tool as part of an online mental health workplace induction program.