Commentary on: De Girolamo G, Dagani J, Purcell R, Cocchi A, McGorry P (Reference De Girolamo, Dagani, Purcell, Cocchi and McGorry2011). Age of onset of mental disorder and use of mental health services: needs, opportunities and obstacles. Epidemiology and Psychiatric Sciences. (2012).

The paper in this edition of EPS by De Girolamo et al. (2012) raises a series of issues fundamental to shape how mental health services should be delivered in future, and these are discussed in relation to understanding and minimising treatment delay from the demand side. The supply side of the equation is subject to a burgeoning literature on the ‘mental health gap (Saxena et al. Reference Saxena, Thornicroft, Knapp and Whiteford2007), including the worldwide neglect in providing mental health treatment to young people (Kieling et al. Reference Kieling, Baker-Henningham, Belfer, Conti, Ertem and Omigbodun2011).

It is now increasingly well recognised that, even in high income countries, only one-third or fewer of people with mental disorders receive treatment. The treatment rates have been found to be considerably lower than this among young people (Vanheusden et al. Reference Vanheusden, van der Ende, Mulder, van Lenthe, Verhulst and Mackenbach2008), despite the fact that half of all lifetime cases start by age 14 years and three quarters by age 24 years (Kessler et al. Reference Kessler, Berglund, Demler, Jin, Merikangas and Walters2005). Indeed treatment delay has been found to be longest for disorders which begin most early (Korczak & Goldstein, Reference Korczak and Goldstein2009). How can we understand this mismatch between such high incidence, prevalence and disability rates for mental disorders among young people (De Girolamo et al. 2012) and the low and slow rates of treatment uptake?

A series of steps can be considered to contribute to treatment delay: (1) no or poor recognition by the young person/family when the first symptoms or signs of mental disorder develop. (2) Misinterpretation of these signs and symptoms. (3) Correct interpretation of the meaning of the psychological or behavioural changes and inappropriate recourse to self-help or family/peer assistance. (4) Correct interpretation of the disorder but insufficient awareness that treatment is locally available without excessive barriers, is culturally appropriate/acceptable, is affordable, and is likely to be effective. (5) A judgment by the individual/family of whether seeking treatment is likely to lead to more harm than good.

Poor recognition of early features of mental illness has long been recognised in the literature, especially the common tendency to use moral explanation of mental illness to interpret behavioural change in young people (Thornicroft, Reference Thornicroft2006). While public education programmes can favourably change knowledge, and attitudes, less is known at present about whether they change help seeking behaviour (Reavley & Jorm, Reference Reavley and Jorm2011). Recent research has demonstrated that public knowledge about mental illness differs by disorder type, and that anxiety-related disorders are far less often correctly identified than psychotic or affective disorders (Coles & Coleman, Reference Coles and Coleman2010). This insight has potentially important implications when linked with the realisation that the mean delay in help seeking for people with social phobia in a large-scale survey in the USA was 11 years (Kessler et al. Reference Kessler, Berglund, Demler, Jin, Merikangas and Walters2005). This suggests that lay understandings, as well as specific and large-scale anti-stigma or mental health literacy programmes cannot assume that campaigns referring to ‘mental illness’ (however, phrased) will have an impact across the whole range of disorders which are understood by health professionals to constitute mental illnesses.

Where the individual/family does recognise that a mental disorder is developing (or has developed), then lay health beliefs about the aetiology of the disorder and the most effective interventions are key considerations (Prins et al. Reference Prins, Verhaak, Bensing and van der Meer2008). Interestingly it may be the case that the people who have received treatment for depression are more likely to see self-referral to a GP as appropriate in future, and to agree that depression has a bio-medical aetiology (Dahlberg et al. Reference Dahlberg, Waern and Runeson2008).

In practical terms, a correct interpretation of the disorder is often nevertheless negated by insufficient awareness that treatment is: locally available without insurmountable barriers, individually acceptable, culturally appropriate, affordable, and is likely to be effective (Thornicroft, Reference Thornicroft2008). Each of these issues can contribute to low service utilisation rates, although there is emerging evidence that where services are provided in ways that are based upon the cultural preferences of services users, that they may induce higher referral rates and lead to accelerated recovery (Afuwape et al. Reference Afuwape, Craig, Harris, Clarke, Flood, Olajide, Cole, Leese, McCrone and Thornicroft2010).

Even when it is recognised within the family that an individual has a probable mental disorder, and that local services are available, then the person concerned may still make a judgment that treatment seeking may, on balance, lead to more harm than good. Again at this stage complex factors come into play including: whether the available treatment options are consistent with preferences, where some studies on people with depression have identified a reluctance to consult with psychiatrists or psychologists (Goldney et al. Reference Goldney, Fisher, Dal and Taylor2005) in high income countries. As yet there is relatively little literature related to treatment seeking in low- and middle-income countries (Ganasen et al. Reference Ganasen, Parker, Hugo, Stein, Emsley and Seedat2008; Kermode et al. Reference Kermode, Bowen, Arole, Joag and Jorm2009; Lee et al. Reference Lee, Lytle, Yang and Lum2010). With all this information in mind, it is possible that an individual with mental illness nevertheless decides not to seek help, in the expectation that the treatment would cause more harm than good. Such trade-offs are not well researched (Flood, Reference Flood2010) in relation to the inter-relationships between self-stigma and health beliefs, although it has been reported that men are more likely to take the view that they should cope with mental illness alone without help (Vogel et al. Reference Vogel, Heimerdinger-Edwards, Hammer and Hubbard2011). Further, it appears that people with an untreated mental disorder can expect that receiving a diagnosis of a mental illness will lead to reputational damage because of stigmatising reactions from others (Brohan & Thornicroft, Reference Brohan and Thornicroft2010; Rose & Thornicroft, Reference Rose and Thornicroft2010). Vital questions therefore arise about whether self-stigma or anticipated discrimination contributes to treatment delay, and so to avoidable morbidity or mortality (Thornicroft et al. Reference Thornicroft, Brohan, Rose, Sartorius and Leese2009; Franz et al. Reference Franz, Carter, Leiner, Bergner, Thompson and Compton2010; Hovens & van der Ploeg, Reference Hovens and van der Ploeg2011). This is particularly important as there is emerging and intriguing evidence which suggests that increasing individual empowerment may reduce self-stigma (Rusch et al. Reference Rusch, Lieb, Bohus and Corrigan2006; Brohan et al. Reference Brohan, Elgie, Sartorius and Thornicroft2010, Reference Brohan, Gauci, Sartorius and Thornicroft2011; Sibitz et al. Reference Sibitz, Unger, Woppmann, Zidek and Amering2011; Hamilton et al. Reference Hamilton, Pinfold, Rose, Henderson, Lewis-Holmes and Flach2011), and also that levels of stigma and discrimination in the population are associated with self-stigma among individuals with mental illness (Evans-Lacko et al. in press). These findings give rise to hypotheses that reducing public stigma may lead to reduced self-stigma among people with mental illness, and that reduced self-stigma may lead to shorter delays in help-seeking. National level interventions such as headspace (Henderson & Thornicroft, Reference Henderson and Thornicroft2009). In Australia and Time to Change in England will offer the opportunity to test these hypotheses in future.