Background

A combination of high prevalence, chronicity, impairment in functioning and vulnerability to premature mortality makes mental health (MH) problems among the most burdensome health conditions in the world (Demyttenaere et al., Reference Demyttenaere, Bruffaerts, Posada-Villa, Gasquet, Kovess, Lepine, Angermeyer, Bernert, de Girolamo, Morosini, Polidori, Kikkawa, Kawakami, Ono, Takeshima, Uda, Karam, Fayyad, Karam, Mneimneh, Medina-Mora, Borges, Lara, de Graaf, Ormel, Gureje, Shen, Huang, Zhang, Alonso, Haro, Vilagut, Bromet, Gluzman, Webb, Kessler, Merikangas, Anthony, Von Korff, Wang, Brugha, Aguilar-Gaxiola, Lee, Heeringa, Pennell, Zaslavsky, Ustun and Chatterji2004; Ormel et al., Reference Ormel, Petukhova, Chatterji, Aguilar-Gaxiola, Alonso, Angermeyer, Bromet, Burger, Demyttenaere, de Girolamo, Haro, Hwang, Karam, Kawakami, Lépine, Medina-Mora, Posada-Villa, Sampson, Scott, Ustün, Von Korff, Williams, Zhang and Kessler2008; Whiteford et al., Reference Whiteford, Degenhardt, Rehm, Baxter, Ferrari, Erskine, Charlson, Norman, Flaxman, Johns, Burstein, Murray and Vos2013) with estimates showing that more than 7% of the global burden of disease is due to mental and substance use disorders (Whiteford et al., Reference Whiteford, Degenhardt, Rehm, Baxter, Ferrari, Erskine, Charlson, Norman, Flaxman, Johns, Burstein, Murray and Vos2013). About three-quarters of this burden resides in low- and middle-income countries (LMIC). Even though effective treatments are available for most of the common mental disorders, affected persons often do not receive any care in most countries (Wang et al., Reference Wang, Aguilar-Gaxiola, Alonso, Angermeyer, Borges, Bromet, Bruffaerts, de Girolamo, de Graaf, Gureje, Haro, Karam, Kessler, Kovess, Lane, Lee, Levinson, Ono, Petukhova, Posada-Villa, Seedat and Wells2007). The gap between need and service is particularly striking in Sub-Saharan African (SSA) countries. For example, studies conducted in the region demonstrate that only about 21% of persons with serious mental disorders received any treatment in Nigeria in the preceding 12 months and only 26% did so in South Africa (Gureje and Lasebikan, Reference Gureje and Lasebikan2006; Saxena et al., Reference Saxena, Thornicroft, Knapp and Whiteford2007; Wang et al., Reference Wang, Aguilar-Gaxiola, Alonso, Angermeyer, Borges, Bromet, Bruffaerts, de Girolamo, de Graaf, Gureje, Haro, Karam, Kessler, Kovess, Lane, Lee, Levinson, Ono, Petukhova, Posada-Villa, Seedat and Wells2007). Among those who had received any form of treatment, only a minority had been offered what can be considered as minimally adequate care (Thornicroft et al., Reference Thornicroft, Chatterji, Evans-Lacko, Gruber, Sampson, Aguilar-Gaxiola, Al-Hamzawi, Alonso, Andrade, Borges, Bruffaerts, Bunting, de Almeida, Florescu, de Girolamo, Gureje, Haro, He, Hinkov, Karam, Kawakami, Lee, Navarro-Mateu, Piazza, Posada-Villa, de Galvis and Kessler2017).

In addition to low availability of MH services, reasons for the low access to service in SSA include cultural views of mental illness which often ascribe supernatural causes to mental disorders and the pervasive stigma that is commonly associated with mental illness (Gureje et al., Reference Gureje, Lasebikan, Ephraim-Oluwanuga, Olley and Kola2005; Adewuya and Oguntade, Reference Adewuya and Oguntade2007; Adewuya and Makanjuola, Reference Adewuya and Makanjuola2008). For example, studies have shown that low level of knowledge of the causes and nature of mental illness, as well as the negative attitude of the community to persons with mental illness, constitute barriers to help-seeking in formal medical care settings (Gureje et al., Reference Gureje, Olley, Ephraim-Oluwanuga and Kola2006). In view of these common beliefs about the nature of MH problems, many of those who seek care for severe MH problems do so from complementary alternative providers (CAPs), in particular traditional and faith healers (T&FH), who are close to the community and share similar views about the nature and cause of mental disorders with the lay public. The scarcity of MH specialists is a strong factor limiting access to MH care in SSA (Gureje and Alem, Reference Gureje and Alem2000). In most of the countries in SSA, the ratio of psychiatrists to the population is less than 1 per 1 000 000 (World Health Organization, 2018). Indeed, the WHO Africa region has the second lowest number of psychiatrists per capita in the world. These shortages apply not only to psychiatrists and other MH professionals but also, in most countries in the region, to general physicians as well (Mullan et al., Reference Mullan, Politzer and Davis1995; Mullan, Reference Mullan2005). In many LMIC, and certainly in SSA, the de facto MH service, or whatever is actually delivered, is therefore mainly provided by CAPs (especially T&FH) and non-physician primary health care providers.

Efforts to scale up MH services in LMIC must place a strong emphasis on improving the skills of primary health care providers and T&FH as well as design and implement programs of collaboration between the two groups of providers. In an important series of papers in the Lancet in 2007, the point was made that, in order to scale up MH services for populations in LMIC, integration of the services into primary care was essential (Saxena et al., Reference Saxena, Maulik, Sharan, Levav and Saraceno2004; Chisholm et al., Reference Chisholm, Flisher, Lund, Patel, Saxena, Thornicroft and Tomlinson2007; Saraceno et al., Reference Saraceno, Ommeren, Batniji, Cohen, Gureje, Mahoney, Sridhar and Underhill2007). An important route to achieving this integration is to facilitate collaboration between complementary and conventional primary health care services and to provide training, support and supervision to primary health care workers (Jenkins et al., Reference Jenkins, Baingana, Belkin, Borowitz, Daly, Francis, Friedman, Garrison, Kauye and Kiima2010a, Reference Jenkins, Kiima, Okonji, Njenga, Kingora and Lock2010c). This process of task sharing could provide a sustainable way for countries with low resources to use available human resources to expand MH services (World Health Organization, 2007). For such a process to succeed, several steps are necessary: fostering of mutual understanding between the providers; appropriate training in effective collaboration; and provision of support and supervision by physicians and MH specialists, wherever available.

Even though LMIC bear a greater proportion of the burden of mental illness (Prince et al., Reference Prince, Patel, Saxena, Maj, Maselko, Phillips and Rahman2007), these countries make disproportionately few contributions to the evidence base necessary for tackling the challenges of mental disorders (Saxena et al., Reference Saxena, Paraje, Sharan, Karam and Sadana2006; Patel and Kim, Reference Patel and Kim2007). For example, LMIC contributed less than 6% of the world's MH-related publications in each of the years between 1998 and 2001 (Saxena et al., Reference Saxena, Maulik, Sharan, Levav and Saraceno2004). This low level of research means that local problems are not empirically studied with a view to finding locally-relevant solutions for them. Also, research evidence produced in high-income countries is often adopted without subjecting it to examination to determine its local relevance or evaluate approaches for adaptation to take account of local contextual factors.

A key factor in the low research activity in SSA is the small number of researchers in the region. In a landmark international collaborative project, it was shown that Africa had the fewest number of researchers compared with Latin America and Asia (Mental Health Research Mapping Group, 2007; Razzouk et al., Reference Razzouk, Sharan, Gallo, Gureje, Lamberte, de Jesus Mari, Mazzotti, Patel, Swartz and Olifson2010). A major limitation for the few available researchers is the relative isolation in which they often work. This isolation not only limits their capacity to be actively engaged in research but is partly responsible for the perennial brain drain of MH professionals from which the continent suffers (Gureje et al., Reference Gureje, Hollins, Botbol, Javed, Jorge, Okech, Riba, Trivedi, Sartorius and Jenkins2009; Jenkins et al., Reference Jenkins, Kydd, Mullen, Thomson, Sculley, Kuper, Carroll, Gureje, Hatcher and Brownie2010d). Another difficulty for researchers on the continent is the relative lack of requisite knowledge to pursue research ideas. In particular, basic skills in epidemiological and health services research are commonly lacking (Sharan et al., Reference Sharan, Gallo, Gureje, Lamberte, Mari, Mazzotti, Patel, Swartz, Olifson and Levav2009).

In light of the foregoing, the Partnership for Mental Health Development in SSA (PaM-D) was established as a multinational collaborative MH research hub to foster the development of research partnerships and research capacity and to conduct empirical research on task shifting approaches that are needed to help countries in SSA meet the goal of addressing the growing burden of MH conditions. PaM-D is one of a network of five Collaborative Hubs for International Research in MH in Africa, Asia and Latin America (CHIRMH) funded by the National Institute of Mental Health. As a group, the mandate of these hubs was to answer research questions (within and across the regions) relating to the global quest of improving MH outcomes for all age groups through research in task sharing for the delivery of MH services and to develop and implement research capacity building activities.

Goals, aims and objectives

The main goal of PaM-D was to create an infrastructure to develop MH research capacity in SSA and to advance global MH science by conducting innovative public health-relevant research in the region. The first aim of PaM-D was to bring together service providers, policy makers, researchers and users with expertise and experience in global MH to create a Sub-Saharan MH research hub that will establish itself as a centre supporting research capacity-building and innovative MH research in the region. The hub aimed to conduct highly relevant and impactful MH research training, research and programming, and contribute to the evolution of cutting-edge assessment, adaptation, implementation and dissemination of research. These activities are expected to lead to efficient and effective services which use existing human resources to improve MH care through task shifting. The second aim was to lay the groundwork for a broader scope of studies related to the overarching goal of addressing local and regional MH needs through the building of a sustainable MH research program in the region. The third was to develop and implement targeted programs of training and mentoring that build MH research capacity in a broad range of MH professionals with a view to creating and supporting a critical mass of experts for innovative MH research. The fourth aim was to actively collaborate with other funded global MH research hubs and other institutions interested in global MH research to conduct studies, improve services, develop policy and enhance MH science.

The partnership

PaM-D brought together partner institutions in five countries in great need for major improvements in their health systems: Nigeria, South Africa, Ghana, Kenya and Liberia with researchers from institutions in the USA and UK, in partnership with government departments and non-governmental organisations. PaM-D is a network of a multidisciplinary group of researchers from diverse areas of expertise: epidemiology, social science research (including ethnography and qualitative research), health systems research, primary MH care, statistics, ethics, mentorship, randomised controlled trials (RCTs) and policy development. The partnership leverages on the strengths of the different partners while at the same time addressing the weaknesses of traditional north-south and south-south collaborations, an approach similar to the one employed in the Emerging mental health systems in low- and middle-income countries (Emerald) project (Semrau et al., Reference Semrau, Alem, Abdulmalik, Docrat, Evans-Lacko, Gureje, Kigozi, Lempp, Lund, Petersen, Shidhaye, Thornicroft and Hanlon2018). For example, it capitalised on the relatively better resourced University of Stellenbosch to strengthen its capacity building component while providing for the institution to use the opportunities within the hub to promote research capability among disadvantaged groups in South Africa. The institutions in Nigeria, Kenya, Ghana and South Africa with their local expertise and track record of research complemented the institution in Liberia, where years of war and turmoil have led to the destruction of health infrastructures and have eroded local capacity to conduct credible research. The institution in the latter country, in turn, participated in and contributed to all capacity building activities, as well as in the process of research design, analysis and dissemination.

Research component

Surveys in Nigeria and Ghana using the WHO-AIMS reveal a severe scarcity of human and material resources for MH, such that only 20% of those with schizophrenia receive any treatment (Chisholm et al., Reference Chisholm, Gureje, Saldivia, Villalón Calderón, Wickremasinghe, Mendis, Ayuso-Mateos and Saxena2008), and most of those who do receive treatment do so from T&FHs (Gureje et al., Reference Gureje, Olowosegun, Adebayo and Stein2010). Unfortunately, the treatment approaches of T&FHs are sometimes harmful or administered in forms that deny patients their basic human rights. Practices such as beating, shackling and scarifications stem directly from widespread beliefs in the magico-spiritual causation of mental illness (especially of psychosis) by the healers (Luyckx et al., Reference Luyckx, Steenkamp, Rubel and Stewart2004; Kurihara et al., Reference Kurihara, Kato, Reverger and Tirta2006). It is, therefore, imperative that while efforts to scale up MH service need to recognise the important role of T&FHs as the port of the first call for many patients with severe MH conditions, there is also a complementary need to skill up non-specialist primary health care workers to liaise with the healers in the delivery of evidence-based and humane care. The research component of PaM-D was designed to provide empirical evidence for collaboration between conventional and complementary MH providers so that the latter would be able to deliver better care for persons with severe mental disorders who use their services.

In the first phase of the research component, a series of formative activities and studies were conducted in Ghana, Kenya and Nigeria to map the services of CAPs in the three countries, document the experience of service users and their caregivers, as well as to undertake a comprehensive exploration of the views of key stakeholders to the idea of a collaborative shared care (CSC) for psychosis provided by CAPs and primary health care workers (Makanjuola et al., Reference Makanjuola, Esan, Oladeji, Kola, Appiah-Poku, Harris, Othieno, Price, Seedat and Gureje2016; van der Watt et al., Reference van der Watt, Nortje, Kola, Appiah-Poku, Othieno, Harris, Oladeji, Esan, Makanjuola and Price2017). The stakeholders included T&FHs, primary care providers, patients who had used the services of CAPs in the past and caregivers of such patients. Informed by the results of these engagement activities, formative studies as well as extensive systematic reviews of both quantitative and qualitative literature on CAPs practices (Gureje et al., Reference Gureje, Nortje, Makanjuola, Oladeji, Seedat and Jenkins2015; Nortje et al., Reference Nortje, Oladeji, Gureje and Seedat2016; van der Watt et al., Reference van der Watt, van de Water, Nortje, Oladeji, Seedat and Gureje2018), a manual was developed to guide a CSC program for the treatment of psychosis to be implemented by CAPs and primary care workers, with the latter serving as the gateway to possible higher level care delivered by general physicians and MH specialists. The manual describes the details of how the collaboration between the healers and primary health care providers is to be conducted, the dos and don'ts of the collaboration, the procedure to be followed in evaluating and treating patients with psychosis by the primary health care providers (PHCPs), how the healers can avoid the use of potentially harmful treatment practices, and what the PHCPs can do to help healers in situations in which they would otherwise use procedures such as shackling to physically restrain their patients. In regard to the provision of evidence-based care for psychosis, the manual also includes (and the PHCPs are trained in the use of) the psychosis module of the MH Gap Action Programme intervention guide (mhGAP-IG), as contextualised to the Nigerian health system (World Health Organization, 2010; Abdulmalik et al., Reference Abdulmalik, Kola, Fadahunsi, Adebayo, Yasamy, Musa and Gureje2013). The PHCPs can seek for a physician or specialist consultation or arrange a referral to a medical facility with the agreement of both the CAP and the patient. The manual was used as the basis of training workshops conducted for CAPs and PHCPs in Nigeria and Ghana, the two sites for the RCT for testing the feasibility and effectiveness of the collaboration (This manual is available on request from the authors).

In the second phase of the research component, a fully powered RCT of the collaborative intervention was implemented. The COllaborative Shared care to IMprove Psychosis Outcome (COSIMPO) is a single-blind cluster RCT being conducted in Nigeria and Ghana to compare the effectiveness of a CSC intervention program implemented by CAPs and PHCPs with care as usual (CAU) at improving the outcome of patients with psychosis (Gureje et al., Reference Gureje, Makanjuola, Kola, Yusuf, Price, Esan, Oladeji, Appiah-Poku, Haris and Othieno2017). The CSC is based on a manualised intervention package, as previously described and consists of scheduled visits as well as response to unscheduled requests by the PHCPs to CAP facilities to assist with the management of trial participants. Assistance includes administration of antipsychotic medications, management of comorbid physical conditions, assisting the CAP to avoid harmful treatment practices, and engaging with CAPs, caregivers and participants in planning discharge and rehabilitation. The PHCPs follow the specifications of the mhGAP-IG in offering treatment and initiating referral whenever indicated. The methods are described in full elsewhere (Gureje et al., Reference Gureje, Makanjuola, Kola, Yusuf, Price, Esan, Oladeji, Appiah-Poku, Haris and Othieno2017).

Future research needs to focus on exploring the barriers and facilitators to effective collaboration with T&FHs, with the aim of putting them to effective use within the public health system. Further research is also needed to implement in real-world settings and scale up CSC projects such as is being tested in COSIMPO, if found to be useful.

Capacity building component

A well-recognised link exists between health research and the provision of solutions to health problems, economic development, as well as the promotion of social equity and poverty alleviation (Gureje and Jenkins, Reference Gureje and Jenkins2007; Jenkins et al., Reference Jenkins, Baingana, Belkin, Borowitz, Daly, Francis, Friedman, Garrison, Kauye and Kiima2010a, Reference Jenkins, Heshmat, Loza, Siekkonen and Sorour2010b). Unfortunately, the generation of research-informed solutions is low in settings with the greatest health, development and equity challenges. Thus, for example, while it is commonly estimated that LMIC bear a much higher burden of the global burden of disease compared with high-income countries, the level of research activities taking place in these LMIC is disproportionately low (Razzouk et al., Reference Razzouk, Sharan, Gallo, Gureje, Lamberte, de Jesus Mari, Mazzotti, Patel, Swartz and Olifson2010). This 10/90 gap, as it has been characterised by the Global Forum for Health Research to describe the observation by the Commission on Health Research for Development that less than 10% of global spending on health research was devoted to diseases or conditions that account for 90% of the global disease burden (Global Forum for Health Research, 2000), is even more striking in the field of MH research. The reasons for this gap include low human and material resource base, brain drain and lack of supportive research infrastructure in LMIC (Patel and Sumathipala, Reference Patel and Sumathipala2001; Patel and Kim, Reference Patel and Kim2007; Gureje et al., Reference Gureje, Hollins, Botbol, Javed, Jorge, Okech, Riba, Trivedi, Sartorius and Jenkins2009; Kola et al., Reference Kola, Fadahunsi and Gureje2009). Two overarching problems are those of low research capacity and the relative isolation in which the current small pool of researchers work. These problems make the securing of competitive grant support unlikely, create disincentives for local researchers to remain engaged and carve out research oriented tenure-track positions, and retard the creation of necessary critical mass for mutual support and mentoring. There is an interconnection between low research capacity, low morale and brain drain, and poor resource generation with these factors operating in a mutually reinforcing manner (Gureje et al., Reference Gureje, Hollins, Botbol, Javed, Jorge, Okech, Riba, Trivedi, Sartorius and Jenkins2009; Kola et al., Reference Kola, Fadahunsi and Gureje2009; Razzouk et al., Reference Razzouk, Sharan, Gallo, Gureje, Lamberte, de Jesus Mari, Mazzotti, Patel, Swartz and Olifson2010; Jenkins et al., Reference Jenkins, Kydd, Mullen, Thomson, Sculley, Kuper, Carroll, Gureje, Hatcher and Brownie2010d).

PaM-D sought to address research capacity building in four different ways: (1) developing research capacity and appreciation of the value of research in a broad range of stakeholders: academics, clinicians (both physicians and non-physicians) and policymakers, with an emphasis on inculcating and developing MH leadership skills; (2)developing training opportunities that focus on specific research skills with a view to meeting previously identified needs of potential trainees; (3) encouraging inter-disciplinary research that builds mutual support and promotes partnership; and (4) initiating sustainable mentoring relationships.

To achieve these goals, the programme of work was designed to, among other things, promote face-to-face engagement of, and interaction between, trainees from the five PaM-D hub countries thereby nurturing potentially useful collegial and collaborative relationships. MH professionals and senior health policy makers from each of the hub participating countries were sponsored to attend the annual 2-week training workshop on MH leadership in Ibadan (Abdulmalik et al., Reference Abdulmalik, Fadahunsi, Kola, Nwefoh, Minas, Eaton and Gureje2014). This was aimed at building their expertise in the process of using evidence for policy formulation, critical review of evidence, consultative process in decision making, evaluating the quality of policy against available evidence and operational issues such as workforce needs assessment, and providing operational know-how on generating and supporting the workforce, and other related issues. We took the view that the basic unit and effective medium for knowledge translation that has the potential for improving policy development and health planning is the systematic review (Grimshaw et al., Reference Grimshaw, Eccles, Lavis, Hill and Squires2012). Consequently, between 2013 and 2016, three early career investigators from the participating countries were supported to undergo mentored research fellowships to acquire expertise in systematic review research approaches under the tutelage of senior members of the hub and their partners who also served as research career development mentors. One of these fellowships was specifically to gain expertise in systematic review research methodology which is expected to be applied to the published literature on the estimated prevalence, risk factors and treatment of anxiety and depressive disorders (especially in primary care) in Africa more generally, and in regions of Africa insofar as there is sufficient evidence for region-based estimates. The second of the fellowships focused on the topics of collaborative care interventions for psychoses in LMIC while the third was on estimated prevalence, risk factors and treatment of HIV-related psychiatric disturbances in Africa.

As part of the collaboration with the other hubs funded by the National Institute of Mental Health (Pilowsky et al., Reference Pilowsky, Rojas, Price, Appiah-Poku, Razzaque, Sharma, Schneider, Seedat, Bonini and Gureje2016), two early career investigators were supported by PaM-D to attend a writing and biostatistics workshop organised by the AFrica Focus on Intervention Research for Mental Health (AFFIRM) hub (Lund et al., Reference Lund, Alem, Schneider, Hanlon, Ahrens, Bandawe, Bass, Bhana, Burns, Chibanda, Cowan, Davies, Dewey, Fekadu, Freeman, Honikman, Joska, Kagee, Mayston, Medhin, Musisi, Myer, Ntulo, Nyatsanza, Ofori-Atta, Petersen, Phakathi, Prince, Shibre, Stein, Swartz, Thornicroft, Tomlinson, Wissow and Susser2015). As shown in Table 1, PaM-D also organised a number of bespoke workshops on biostatistics, quality research methodology and scientific writing. As a sequel to the writing workshops, four of the early career attendees were awarded seed grants for research and are being supported by the hub as research fellows to receive the mentoring and other support required to help them prepare grant applications as independent investigators. Capacity building activities were monitored with a set of indicators, some of which were commonly employed as indicators across all the funded hubs. For example, workshop courses were monitored through pre- and post- evaluation surveys and their longer-term impact, while fellows, mentees and awardees were tracked in terms of their career trajectories over the lifespan of the hub and their uptake of higher degrees, grant writing, publications, other research and awards (Pilowsky et al., Reference Pilowsky, Rojas, Price, Appiah-Poku, Razzaque, Sharma, Schneider, Seedat, Bonini and Gureje2016; Schneider et al., Reference Schneider, van de Water, Araya, Bonini, Pilowsky, Pratt, Price, Rojas, Seedat and Sharma2016). At the end of the capacity building activities of PaM-D, a total of 60 articles had been published while 55 grants had been applied for, 21 of which were successful (Table 2). The published work was a product of data from PaM-D activities as well as manuscripts that the trainees produced from their own data after attending manuscript writing courses organised by the hub, with mentorship from the senior members of the hub. Early career and mid-level researchers from the five Sub-Saharan countries were first authors on 21 of the 60 published papers. The grants were applied for by the trainees in the SSA country members of the hub in collaboration with other early career and midlevel researchers, with mentorship and support from the more senior members of the hub. The projects funded by these grants were a mix of those related to the overall research agenda of the hub and other unrelated projects. Expectedly, the number of publications and grants are under-reported, as the monitoring and evaluation response rate from the 35 participants in the capacity building activities ranged from 20 to 75% at various evaluation points in the life of the project.

Table 1. Capacity building activities undertaken by PaM-D

Table 2. Outcomes of the capacity building activities undertaken by PaM-D

From our experience and the shared experiences of our sister hubs, it is our recommendation that future MH capacity building activities on the continent adopt a multidisciplinary approach; and that capacity building should be linked to the overall MH research agenda of the region. We also recommend that the vehicle for delivering these capacity building opportunities be a mix of formal academic degrees or fellowships as well as the regular filling of gaps in knowledge through the organising of bespoke workshops to address expressed areas of need in LMIC.

Collaboration with other National Institute of Mental Health hubs

PaM-D collaborated with the other National Institute of Mental Health-funded hubs to conduct research that sought to explore the facilitators and barriers to task sharing approaches to delivering MH services in LMICs. In the course of their work, junior members of the different hubs participated in capacity building opportunities as they were available in sister hubs. Some of these activities have resulted in joint publications by the five hubs (Pilowsky et al., Reference Pilowsky, Rojas, Price, Appiah-Poku, Razzaque, Sharma, Schneider, Seedat, Bonini and Gureje2016). Of particular importance is PaM-D's collaboration with the other Sub-Saharan hub, AFFIRM (Lund et al., Reference Lund, Alem, Schneider, Hanlon, Ahrens, Bandawe, Bass, Bhana, Burns, Chibanda, Cowan, Davies, Dewey, Fekadu, Freeman, Honikman, Joska, Kagee, Mayston, Medhin, Musisi, Myer, Ntulo, Nyatsanza, Ofori-Atta, Petersen, Phakathi, Prince, Shibre, Stein, Swartz, Thornicroft, Tomlinson, Wissow and Susser2015). Scholars from each of the two hubs attended capacity building activities of the other. The fact that AFFIRM and PaM-D are based on the same continent provides an important opportunity for future joint activities aimed at addressing the peculiar needs of the region with the possibility of expanding those activities to include other countries in the region.

Involvement of service users in the activities of PaM-D

Service users were involved in the activities of the hub from its inception. First, before the commencement of the main trial, there were several formative engagement meetings and key informant interviews with service user organisations to gain their perspective on the proposed work. Second, service users were among the participants in the preparatory Theory of Change workshop from which a ToC Map was developed to guide the activities of the hub. Third, service users were supported by the hub to participate at the annual 2-week MH Leadership and Advocacy Training Programme at the University of Ibadan, Nigeria. Furthermore, two representatives of service user groups from Nigeria and Ghana served on the PaM-D Oversight Advisory Committee, which was set up to provide ethical oversight for the hub's program of work and to provide guidance on how PaM-D activities can have policy relevance to the participating countries as well as to the region.

Potential policy impact

An important desire for researchers is for research findings to serve the goal of improved patient care. This goal can only be achieved when research findings influence policy development. Unfortunately, the distance between researchers and policymakers is traditionally large with the latter often unaware of what the former is doing or unable to make sense out of the plethora of research papers on any given topic. In PaM-D, a conscious effort was made to attempt to bridge this divide in order to improve the chance of the findings of the research efforts influencing policy in the region. Specifically, representatives of each of the governments of the participating countries served on the hub's Oversight Advisory Committee which met once a year and functioned to advise the hub on how its activities could be relevant to the policies of the member countries. The hub also produced policy briefs from time to time. We took the view that the science had to be ‘context-sensitive’, a concept coined by Gibbons (Reference Gibbons2000) that refers to the development of knowledge within the context in which it is intended for use, using trans-disciplinary collaboration and stakeholder involvement early in the process and throughout the project.

At a landmark policy roundtable held in Abuja, Nigeria in October 2009 and attended by Ministers of Health or their representatives from several African countries (Burundi, Nigeria, Ghana, Sierra Leone, Liberia, Cote d'Ivoire, Niger, Ethiopia and Kenya) (World Psychiatric Association, 2016), participants had discussed the issues of how best to integrate the service of the large numbers of T&FHs who provide care for individuals with mental disorders into the formal MH sector in the participating countries and how to make primary care service a more effective level for MH care. In particular, a wish was expressed for T&FHs to deliver service in which harmful forms of treatment and abuse of human rights were avoided and for primary care service practitioners to be trained and supervised to provide and monitor effective MH promotion, prevention, treatment and rehabilitation services at the local level. The research component of PaM-D addressed these concerns by exploring the feasibility of collaboration between conventional and complementary MH providers the essential features of which include task-shifting from specialist MH service to primary care providers as well as to T&FHs.

Challenges and responses

There are challenges inherent in implementing a transnational partnership such as PaM-D; in carrying out intervention research in human subjects with severe mental disorders, who are to be cared for within a newly developed partnership involving non-physician, non-specialist actors in both formal and informal sectors of the health system and in settings prone to human right abuses; in forging partnerships among diverse stakeholders; as well as in carrying out sustainable capacity building activities among scholars and other stakeholders from diverse regions on the continent. These challenges were anticipated and carefully prepared for through the rigorously-conducted formative studies, the results of which were reflected in the protocols and were useful in the course of project implementation.

First, the physical distance between the researchers was bridged through both a participatory management approach, with researchers in the various sites serving on different committees, as well as through the setting up of communication portals, online meetings and face-to-face meetings. Second, the research component had the inherent challenge of taking consent from persons with a severe mental disorder (psychosis) and of explaining to patients that they may be randomised to receive care that was judged by researchers not to be the gold standard. These were addressed through careful attention to them in the study protocols (Gureje et al., Reference Gureje, Makanjuola, Kola, Yusuf, Price, Esan, Oladeji, Appiah-Poku, Haris and Othieno2017) as well as well-planned training in partnership building by those involved in the field work. For example, several informal meetings were organised between PHCPs and T&FHs during training and debriefing activities to encourage rapport building and mutual trust. Third, we minimised human right abuses by training the providers in both the intervention and control arms prior to the commencement of the trial. It is pertinent to note that these challenges are not unique to PaM-D; other hubs as well as similar projects have reported being faced with such challenges and had confronted them by employing similar strategies (Lund et al., Reference Lund, Alem, Schneider, Hanlon, Ahrens, Bandawe, Bass, Bhana, Burns, Chibanda, Cowan, Davies, Dewey, Fekadu, Freeman, Honikman, Joska, Kagee, Mayston, Medhin, Musisi, Myer, Ntulo, Nyatsanza, Ofori-Atta, Petersen, Phakathi, Prince, Shibre, Stein, Swartz, Thornicroft, Tomlinson, Wissow and Susser2015; Nyström et al., Reference Nyström, Karltun, Keller and Andersson Gäre2018).