We have recently argued, based upon a thorough review of the literature, that in low, middle and high-income countries and settings a balance is required between investment in community-based and hospital-based mental health services (Thornicroft & Tansella, Reference Thornicroft and Tansella2009, Reference Thornicroft and Tansella2013a , Reference Thornicroft and Tansella b , Thornicroft et al. Reference Thornicroft, Ruggeri and Goldberg2013, Reference Thornicroft, Alem, Drake, Ito, Mari, McGeorge, Thara and Semrau2011a , Reference Thornicroft, Szmukler, Mueser and Drake2011b ). Is this view supported by leading mental health experts working in those low-income settings, where over three-quarters of the world's population lives? In this issue of Epidemiology and Psychiatric Sciences, two papers examine this proposition both from the perspective of clinicians and as researchers. Working in Chennai in Southern India, Thara et al. (Reference Thara, John and Chatterjee2014) set out the key elements of the context that distinguish many low-income from high-income settings. Such contextual factors include not only just the lower absolute (and often relative) levels of healthcare investment in the former, but also the greater provision of support and care by family members for those who are unwell or disabled, and less medicalised explanatory models of illness, associated with traditional patterns of consultation with traditional and religious healers (Incayawar et al. Reference Incayawar, Wintrob and Bouchard2009; Thara & Padvamati, Reference Thara and Padvamati2014). They illustrate the need to plan based upon the actual resources that are available on a sustainable basis, usually those which are state funded, and to construct teams for defined population-based areas. The mix of skills within such teams is therefore a very particular blend of disciplines, which represents historical patterns of investment in staff training, allowing for attrition by emigration abroad. This blend of the actual available staff in most low- and middle-income countries will require a form of task-sharing that expects quite different competences from nurses, for example, that those common in high-income settings, such as case identification and some medication prescribing authority (Chibanda et al. Reference Chibanda, Mesu, Kajawu, Cowan, Araya and Abas2011; Kakuma et al. Reference Kakuma, Minas, van Ginneken, Dal Poz, Desiraju, Morris, Saxena and Scheffler2011; Petersen et al. Reference Petersen, Lund, Bhana and Flisher2012), where this has been longer established in the treatment of people with tuberculosis and HIV/AIDS (Campbell & Scott, Reference Campbell and Scott2011; Gabriel & Mercado, Reference Gabriel and Mercado2011).

An extension of this concept is to think of the clinical team in new terms, namely to include new categories of staff in reconfigured roles, often called lay/community health workers (CHW). There is an emerging evidence base for community services where such health workers are the mainstay of service provision to deliver care for people, for example, with schizophrenia (Collins et al. Reference Collins, Patel, Joestl, March, Insel, Daar, Anderson, Dhansay, Phillips, Shurin, Walport, Ewart, Savill, Bordin, Costello, Durkin, Fairburn, Glass, Hall, Huang, Hyman, Jamison, Kaaya, Kapur, Kleinman, Ogunniyi, Otero-Ojeda, Poo, Ravindranath, Sahakian, Saxena, Singer and Stein2011) or depression (Chatterjee et al. Reference Chatterjee, Pillai, Jain, Cohen and Patel2009; Patel et al. Reference Patel, Weiss, Chowdhary, Naik, Pednekar, Chatterjee, Bhat, Araya, King, Simon, Verdeli and Kirkwood2011). Indeed a new randomised controlled trial assessing community mental health teams in India has found that such services, including community health workers, improved both symptoms and disability for people with psychosis (Chatterjee et al. Reference Chatterjee, Leese, Koschorke, McCrone, Naik, John, Dabholkar, Goldsmith, Balaji, Varghese, Thara, Patel, Thornicroft and Copsi2011, Reference Chatterjee, Naik, John, Dabholkar, Balaji, Koschorke, Varghese, Thara, Weiss, Williams, Patel and Thornicroft2014), indicating that the benefits of community mental health services may apply in low as well as in middle- and high-income countries.

Such CHW staff are in a position to increase the proportion of cases of people with mental illness who receive treatment (i.e. scaling-up to increase coverage) (Thornicroft, Reference Thornicroft2007; Eaton et al. Reference Eaton, McCay, Semrau, Chatterjee, Baingana, Araya, Ntulo, Thornicroft and Saxena2011), and to deliver effective evidence-based care, if they receive sufficient and ongoing training for both case finding and treatment. The possibility of providing such training has been recently revolutionised by the creation of the mhGAP Intervention Guide by the WHO (Barbui et al. Reference Barbui, Dua, van Ommeren, Yasamy, Fleischmann, Clark, Thornicroft, Hill and Saxena2010; World Health Organization, 2010; Dua et al. Reference Dua, Barbui, Clark, Fleischmann, van Ommeren, Poznyak, Yasamy, Thornicroft and Saxena2011). The treatment manual, along with practical teaching materials, are now increasingly available and have been translated into Arabic, Chinese, French, Greek, Hindi, Indonesian, Italian, Japanese, Portuguese, Spanish and Thai.

If community mental health teams in LAMICs need to be configured in quite different ways from those in high-income countries, what is the role of psychiatrists in the former? Such doctors are a scare resource: while in Western Europe there are 5.5–20.0 psychiatrists for each 100 000 population, across sub-Saharan Africa the number is <1% of this at 0.05 per 100 000 (World Health Organization, 2011).

Kigozi and Ssebunnya address the specific role of psychiatrists in advancing treatment and care, not just within their host organisation, but also in terms of the wider public health, drawing upon their experience in Uganda (Kigozi & Ssebunnya, Reference Kigozi and Ssebunnya2014). They do not flinch from setting out the very considerable barriers facing those who intend to improve the provision of mental health care, locally or nationally (Saraceno et al. Reference Saraceno, Van, Batniji, Cohen, Gureje, Mahoney, Sridhar and Underhill2007). They conclude that it is undeniable, simply on the grounds of the epidemiological frequency of mental disorders worldwide (Wang et al. Reference Wang, Angermeyer, Borges, Bruffaerts, Tat, de Girloamo, Fayyad, Gureje, Haro, Huang, Kessler, Kovess, Levinson, Nakane, Oakley Brown, Ormel, Posada-Villa, Aguilar-Gaxiola, Alonso, Lee, Heeringa, Pennell, Chatterji and Ustun2007), that the only part of the healthcare system that has the potential to deliver care in proportion to the scale of the need is primary healthcare (Tansella & Thornicroft, Reference Tansella and Thornicroft1999).

The challenge that follows is: how can psychiatrists most effectively support primary care staff to more often and more effectively detect and treated people with mental illness (Prince et al. Reference Prince, Patel, Saxena, Maj, Maselko, Phillips and Rahman2007; Saxena et al. Reference Saxena, Thornicroft, Knapp and Whiteford2007)? This reconceptualisation of the role of the psychiatrist requires first of all a new training curriculum for trainee and for qualified psychiatrists, one that emphasises the public health need for psychiatrists to work both directly in secondary and tertiary services, and also to act as multipliers by potentiating the capacity of primary care staff to detect and treat people with mental illness. It has been suggested (Von Korff & Goldberg, Reference Von Korff and Goldberg2001) that in high-income countries this capacity (particularly in the treatment of people with major depressive disorder) may well be enhanced by changes in the organisation and function of healthcare teams, such as those already being used to improve outcomes in other chronic diseases. Responsibility for active follow up should be given to a case manager (for example, a practice nurse); adherence to treatment and patient outcomes should be regularly monitored; treatment plans should be frequently adjusted when patients do not improve; and the case manager and primary care physician should have the possibility to consult and refer to a psychiatrist when necessary. Flexible and accessible working relationships between the primary care doctor, the case manager and a mental health specialist are considered essential to allow most patients with mental disorders to access more effective treatment in primary care, as well as the minority needing ongoing specialist care to be identified and referred. The adaptation of the ideas behind this model to LAMICs is still to be investigated.

This is a very considerable challenge because in many countries there is no tradition of primary care being engaged in recognising and treating people with mental illness, or treating the co-morbid physical illnesses of people with mental illness. But even going beyond this, there is an active aversion to people with mental illness by some staff. In general, health professionals tend to hold negative attitudes towards individuals with mental illness (Corrigan et al. Reference Corrigan, Williams, McCracken, Kommana, Edwards and Brunner1998; Chaplin, Reference Chaplin2000; Hugo, Reference Hugo2001; Cooper et al. Reference Cooper, Corrigan and Watson2003; Corrigan, Reference Corrigan2004; Lundberg et al. Reference Lundberg, Hansson, Wentz and Bjorkman2008; Horsfall et al. Reference Horsfall, Cleary and Hunt2010). Practitioners, including general practitioners, report more negative ratings of individuals with a mental illness than the general public (Jorm et al. Reference Jorm, Korten, Jacomb, Christensen and Henderson1999; Nordt et al. Reference Nordt, Rossler and Lauber2006) and also expressed a greater desire for social distance (Reynolds et al. Reference Reynolds, Pitts-Brown and Thornicroft1996; Arvaniti et al. Reference Arvaniti, Samakouri, Kalamara, Bochtsou, Bikos and Livaditis2009; Feret et al. Reference Feret, Conway and Austin2011; Hori et al. Reference Hori, Richards, Kawamoto and Kunugi2011; Nguyen et al. Reference Nguyen, Chen and O'Reilly2012; O'Reilly et al. Reference O'Reilly, Bell, Kelly and Chen2013).

There may be powerful and adverse consequences for people with mental illness from the reluctance of primary care staff to treat them. The term ‘diagnostic overshadowing’ has been defined as the process by which people with mental illness receive poorer physical health care because staff mis-attribute physical symptoms to mental illness, and so under-investigate and treat physical disorders (Desai et al. Reference Desai, Rosenheck, Druss and Perlin2002; Jones et al. Reference Jones, Howard and Thornicroft2008). There is also strong emerging evidence, although from high-income countries, that life expectancy is reduced by 15–20 years among people with mental disorders, which appears to be related at least in part to under-treatment of physical disorders among people with mental illness (Wahlbeck et al. Reference Wahlbeck, Westman, Nordentoft, Gissler and Laursen2011; Thornicroft, Reference Thornicroft2011, Reference Thornicroft2013; Lawrence et al. Reference Lawrence, Hancock and Kisely2013).

Epidemiology and Psychiatric Sciences therefore welcomes an intensifying debate of these issues and especially how to developing a growing evidence-base that informs global effects to provide more and more effective treatment and care to people with mental illness in low- and middle-income settings.