Strengthening mental health system governance in six low- and middle-income countries in Africa and South Asia: challenges, needs and potential strategies

Abstract

Poor governance has been identified as a barrier to effective integration of mental health care in low- and middle-income countries. Governance includes providing the necessary policy and legislative framework to promote and protect the mental health of a population, as well as health system design and quality assurance to ensure optimal policy implementation. The aim of this study was to identify key governance challenges, needs and potential strategies that could facilitate adequate integration of mental health into primary health care settings in low- and middle-income countries. Key informant qualitative interviews were held with 141 participants across six countries participating in the Emerging mental health systems in low- and middle-income countries (Emerald) research program: Ethiopia, India, Nepal, Nigeria, South Africa, and Uganda. Data were transcribed (and where necessary, translated into English) and analysed thematically using framework analysis, first at the country level, then synthesized at a cross-country level. While all the countries fared well with respect to strategic vision in the form of the development of national mental health policies, key governance strategies identified to address challenges included: strengthening capacity of managers at sub-national levels to develop and implement integrated plans; strengthening key aspects of the essential health system building blocks to promote responsiveness, efficiency and effectiveness; developing workable mechanisms for inter-sectoral collaboration, as well as community and service user engagement; and developing innovative approaches to improving mental health literacy and stigma reduction. Inadequate financing emerged as the biggest challenge for good governance. In addition to the need for overall good governance of a health care system, this study identifies a number of specific strategies to improve governance for integrated mental health care in low- and middle-income countries.

Key Messages

• Poor governance has been identified as a barrier to effective integration of mental health care in low- and middle-income countries (LMICs).

• In addition to the need for overall good governance of a health care system in LMICs, this study identifies a number of specific strategies to improve governance for integrated mental health care.

Introduction

Many low- and middle-income countries (LMICs) have to deal simultaneously with the challenges of combating communicable diseases, and maternal and perinatal morbidity and mortality, as well as the rising burden of non-communicable diseases (NCDs) (including cardiovascular disease, cancer, diabetes and mental, neurological and substance use disorders) (Mayosi et al. 2012). This rising burden of NCDs, is projected to impact negatively on economic growth, placing a large burden on health and welfare systems on the one hand, and decreasing productivity as a result of disability and out-of-pocket expenditure by families, on the other (Bloom et al. 2011). With regard to mental, neurological and substance use specifically, these disorders increased by 41% between 1990 and 2010, and account for one in every 10 lost years of health globally (Patel et al. 2016). The need to strengthen the ability of health systems of LMICs to respond to these new pressures in a cost-efficient manner is therefore pressing.

In particular, there is a need to strengthen health systems to address the large treatment gap for mental disorders, which includes insufficient medical and non-medical interventions. In low-income countries, this treatment gap is large, with between 76% and 85% of people with severe mental disorders not receiving treatment in the previous 12 months (Demyttenaere et al. 2004, Wang et al. 2007). Following the World Health Organization’s (WHO) Global Mental Health Action Plan (World Health Organization 2013), many low resourced regions and countries have responded to the need to close this treatment gap through adopting policies and plans that promote the integration of mental health into general health care. This is viewed as a more efficient mechanism to increase coverage than vertical specialist care, which historically received the lion’s share of the sparse resource allocation to mental health in LMICs (Saraceno et al. 2007). Task-sharing underpins integrated care, with mounting evidence of the efficacy and cost effectiveness of this approach (Dua et al. 2011). However, there is a paucity of knowledge on the health system requirements for effective integration of mental health care. The Emerging mental health systems in low- and middle-income countries (Emerald) program is an international research consortium in six LMICs (Ethiopia, India, Nepal, Nigeria, South Africa and Uganda) that aims to investigate these health system requirements for effective mental health care integration (Semrau et al. 2015). Selected countries provide a broad range of LMICs, from upper middle-income countries (South Africa) through to a fragile state (Nepal), thus allowing for a comparison of health system requirements across diverse country contexts.

Poor governance is one of the key system barriers to the implementation of integrated primary mental health services (Saraceno et al. 2007, Thornicroft et al. 2010), with good governance essential for ensuring accountability of governments for protecting the welfare of their citizens. Health systems governance is defined as ‘ensuring that strategic policy frameworks exist and are combined with effective oversight, coalition building, regulation, attention to system-design and accountability’ (World Health Organization 2007). Governance mechanisms can be situated at global international levels (e.g. the World Health Organization) through to regional, national and sub-national levels (e.g. districts). They also incorporate clinical governance, important for quality assurance and health system performance (Siddiqi et al. 2009), and ensuring quality of care and the best clinical outcomes for service users (Starey 2001).

The aim of this study was to identify key governance issues that need to be addressed to facilitate the integration of mental health services into general health care in the six participating Emerald countries.

Methodology

Design

Given that the research was exploratory, with a view to making recommendations for strengthening mental health systems governance, the study adopted a descriptive qualitative approach, using framework analysis.

Participants

Purposive sampling was used to recruit a range of key informants across the six Emerald study countries, to ensure views were elicited on all the governance issues within the expanded framework. Key informants across the six countries included policy makers at the national level in the Department/Ministry of Health; provincial coordinators and planners in primary health care and mental health; and district-level managers of primary and mental health care services. Interviews were not conducted at the provincial/regional level in Nepal, Uganda and Ethiopia, given the lack of mental health planning/activities at this level. A total of 141 key informants were interviewed across the six countries. Some countries conducted more than the 15–20 interviews suggested, as more potential participants were available in the positions specified above. The number of participants approached and recruited in each country is shown in Table 1 below.

Table 1.

Number of participants per country

	Total	Ethiopia	India	Nepal	Nigeria	South Africa	Uganda
Number of potential interviewees approached	163	17	45	28	33	24	16
Number of interviews with national policy makers	59	7	20	17	6	4	8
Number of interviews with provincial health care planners	15	0	6	0	4	5	0
Number of interviews with district health care planners and managers	67	10	7	11	20	8	8
Total	141	17	33	28	30	17	16

Conceptual framework for data collection

Data collection was guided by a combination of the health system governance frameworks developed by Siddiqi and colleagues (Siddiqi et al. 2009) and Mikkelsen-Lopez and others (Mikkelsen-Lopez et al. 2011). The framework developed by Siddiqi and colleagues facilitates an assessment of the strengths and weaknesses of governance from the national health policy formulation level through to policy implementation at sub-national level, using ten value-driven principles. Mikkelsen-Lopez and colleagues adopt a system-wide view of governance, acknowledging the importance of good governance within the system as a whole, as well as within the WHO’s essential building blocks of a health care system (World Health Organization 2007) that include financing, medicines and technology, human resources, information systems, and service delivery. The performance of these building blocks directly affects the quality of service delivery, impacting particularly on responsiveness and effectiveness, and information, accountability and transparency. We have developed an adapted framework (see Table 2) combining these two perspectives that formed the overarching framework for the development of the cross-country interview schedule developed for this study, as well as the analysis presented. Questions covered the main governance principles in relation to policy development and implementation from national through to regional and district levels, with the effectiveness and efficiency principle expanded to incorporate governance of the health system building blocks—financing, medicines and technology, human resources, with the addition of infrastructure, given that poor infrastructure is common in LMICs.

Table 2.

Adapted governance framework principles

Principle	Domains
Rule of Law and Strategic Direction Leaders have a broad and long-term perspective on health and human development, along with a sense of strategic directions for such development, as reflected in policy frameworks. There is also an understanding of the historical, cultural and social complexities in which that perspective is grounded. Legal frameworks pertaining to health should be fair and enforced impartially, particularly the laws on human rights related to health	Long-term vision; comprehensive development strategy including health Legislative process; interpretation of legislation to regulation and policy; enforcement of laws and regulations
Effectiveness and Responsiveness: Human resources building block Processes and institutions should produce results that meet population needs and influence health outcomes while making the best use of resources. Institutions and processes should try to serve all stakeholders to ensure that the policies and programs are responsive to the health and non-health needs of its users	Quality of human resources; capacity for implementation
Effectiveness and Responsiveness: Financing, medicines and protocols, infrastructure building blocks Processes and institutions should produce results that meet population needs and influence health outcomes while making the best use of resources. Institutions and processes should try to serve all stakeholders to ensure that the policies and programs are responsive to the health and non-health needs of its users.	Quality of finance and finance mechanisms, medicine availability, infrastructure
Participation and Collaboration All men and women should have a voice in decision-making for health, either directly or through legitimate intermediate institutions that represent their interests. Such broad participation is built on freedom of association and speech, as well as capacities to participate constructively. Good governance of the health system mediates differing interests to reach a broad consensus on what is in the best interests of the group and, where possible, on health policies and procedures.	Participation in decision-making process; stakeholder identification and voice
Equity, Inclusiveness and Ethics All men and women should have opportunities to improve or maintain their health and well-being (Equity & Inclusiveness) The commonly accepted principles of health care ethics include respect for autonomy, non-maleficence, beneficence and justice. Health care ethics, which includes ethics in health research, is important to safeguard the interest and the rights of the patients (Ethics)	Equity in access to care; fair financing of health care; disparities in health Principles of bioethics; health care and research ethics
Information, Accountability and Transparency: Information system building block Intelligence and information are essential for a good understanding of the health system, without which it is not possible to provide evidence for informed decisions that influence the behavior of different interest groups that support, or at least do not conflict with, the strategic vision for health (Information) Decision-makers in government, the private sector and civil society organizations involved in health are accountable to the public, as well as to institutional stakeholders. This accountability differs depending on the organization and whether the decision is internal or external to an organization (Acccountability) Transparency is built on the free flow of information for all health matters. Processes, institutions and information should be directly accessible to those concerned with them, and enough information is provided to understand and monitor health matters (Transparency)	Information: generation, collection, analysis, dissemination Accountability: internal; accountability: external Transparency in decision-making; transparency in allocation of resources

Note: Adapted from Siddiqi and others (Siddiqi et al. 2009) and Mikkelson-Lopez and colleagues (Mikkelsen-Lopez et al. 2011).

Each country adapted the questions for contextual relevance. For example, in countries where mental health policies and laws existed, the interview questions focused on the challenges in implementation; in countries where the policy was still under development, the focus was on challenges with respect to the development of policies and laws. Across the countries, the interviews were conducted by principal investigators and/or research assistants trained to conduct qualitative interviews, with the number of interviewers per country ranging from two to five. The interviews were audiotaped and transcribed verbatim. Where interviews were conducted in the local language, they were translated into English with back-translation checks applied.

Data analysis

Framework analysis provides a systematic approach to analysis and allows for predetermined themes (based on the conceptual model guiding the interview questions) to be used as the basis of the initial coding framework. Inductive themes are then also included as they emerge from the data (Gale et al. 2013). This approach thus allowed for cross-country as well as country-specific themes to be analysed. A cross-country thematic coding framework was initially developed based on the questions asked in the cross-country interview guide. The analysis focused specifically on barriers associated with each of the overarching thematic governance principles covered, as well as needs and possible strategies to overcome these. Each country used this framework to initially code their data in a country-specific matrix, adding emergent country-specific themes. These country-specific matrices were then merged into a cross-country matrix to detect commonalities and differences in the country data in order to identify key challenges and opportunities affecting mental health system governance across the six countries. Country author/s, who are all themselves experts in the mental health system strengthening in their respective countries, checked this analysis for the veracity of the themes. All countries were required to hold stakeholder meetings to feed back the findings of the interviews to the key informants involved in the interviews, to assist in the validation of the findings.

Ethics

Ethical approval was obtained from the authors’ institutes. Research staff in all the countries received training in ethical and research governance safeguards and protections. Data were de-identified, and participant-related files were subject to password-protected access in each country site.

Results

Challenges

A summary of the dominant key challenges affecting mental health system governance and policy implementation across the six countries is presented in Table 3.

Table 3.

Challenges and strategies to improve mental health governance

Governance principles	Challenges	Strategies
Rule of Law and Strategic Direction	Countries developing new/progressive mental health legislation and policies: Low public health priority of mental health Countries with new mental health legislation and policies: Poor implementation attributed to: Weak technical capacity; insufficient resources. Poor coordination and management at district level	Countries developing new/progressive mental health legislation and policies: Increase public health priority of mental health through advocacy efforts Establish a mental health unit and responsible person at the national level Technical support to Ministry of Health to develop progressive mental health policies and laws
		Countries with new mental health legislation and policies: Establish/strengthen a national directorate or unit to oversee implementation Capacitate and ensure accountability for implementation at regional and district levels Develop formal structures for multi-sectoral engagement from national through to district level
Effectiveness and Responsiveness: Human resources	Inadequate pre-service training of generalists High staff turnover Insufficient specialist capacity Biomedical/symptom orientation Psychiatric stigma	To accommodate task sharing revise: Pre-service training curricula Job descriptions In-service training in locally contextualized mhGAP tools Continuous in-service training and mentorship Improve staff retention and coverage Orientation to patient-centred care & anti-stigma workshops
Effectiveness and Responsiveness: Financing, medicines and protocols, infrastructure, information systems	Mental health services underfunded Mental health budget not ring-fenced at PHC level Inadequate and inconsistent supply of psychotropic medication Inadequate space for patients with behavioral problems/counseling	Dedicated budget for mental health from national through to district level Double funding to develop community-based services while maintaining/developing tertiary level services Disability grants for people with severe mental illness Demarcation of dedicated space for in-patient care in general hospitals Adequate counseling space Improved supply chain management of psychotropic medication
Participation and Collaboration	Poor inter-sectoral collaboration Poor service provider and service user participation in the development of policies and plans.	Role clarification of different sectors Establishment of formal inter-sectoral engagement mechanisms Supporting development of user groups Capacitation of users and managers to engage with one another
Equity and Ethics	Equity: Geographical inequities High treatment cost Institutional stigma Stigmatizing attitudes on the part of service providers and public Low mental health literacy Cultural beliefs	Equity: Public health campaigns to reduce stigma and improve mental health literacy Improve the detection of mental disorders at a community level Integration of mental health into PHC
	Ethical practice and research ethics: Poor quality control of mental health services Weak research implementation monitoring	Ethical practice: Capacitating service users to provide feedback on the quality of services Adoption of WHO QualityRights program
Information, Accountability and Transparency	Information: Mental health information at PHC level weak or not collected Poor information-based decision-making and planning, and monitoring of service implementation	Need for sufficient feasible indicators for mental health Need for capacity building for monitoring and evaluation
	Accountability and Transparency: Generally weak

Rule of law and strategic direction

Two middle-income countries (India and South Africa) had a mental health bill or law, although problems with implementation were reported by both. In India, the process of translating the bill into law was reportedly protracted. In South Africa, insufficient training and resources to enable uniform implementation across the country emerged. Two countries (Nigeria and Uganda) had obsolete legislation which was largely ignored, while Ethiopia and Nepal, both low income, did not have any mental health legislation. However, Uganda, Nigeria and Nepal were in the process of establishing new mental health laws.

In relation to Strategic Direction, although all countries either had a new or a draft mental health policy, or were in the process of developing one, mental health still emerged as being a low priority in practice. Poor implementation of existing policy was highlighted in all countries. This was attributed to weak technical capacity, with all sites indicating that implementation was limited by poor coordination and management of mental health activities on the service frontline. Institutional stigma also emerged as being responsible for poor policy implementation in India. According to an Indian national level respondent:

For mental health, hesitation … is also responsible for uninformed planning, thereby limiting integration at facilities.

Effectiveness and responsiveness: human resources

In relation to human resource challenges to effecting integration, all countries except Nigeria reported inadequate pre-service training of generalists in mental health, and high staff turnover which requires constant in-service training. Insufficient specialist capacity to provide training and support to generalist health care providers was reported by all countries.

The following barriers to identification of mental disorders were reported: i) Mental health stigma (all countries); ii) The biomedical symptom orientation (Nepal, India, South Africa and Uganda); iii) The perception that mental health care was additional work, and time consuming (Ethiopia, Nepal and South Africa).

The lack of positions for mental health workers and over-burdening of general community health workers, as well as concerns about relying on volunteers, were identified as challenges associated with task sharing of mental health care in Ethiopia, Nepal and Uganda. As indicated by a health worker in Nepal

Other patients requires 2–3 minutes but the psychiatric patient needs minimum of 30 minutes to one hour. It is difficult to manage time for them—even for the other patient also—and besides this we have other work also—so we are having work burden.

Effectiveness and responsiveness: financing, infrastructure, medicines and protocols

With the exception of India, funds for mental health were purportedly inadequate across the other five countries. Four countries (Nepal, Nigeria, South Africa, Uganda) also indicated that the lack of a ring-fenced budget at PHC level was a problem. The supply of psychotropic medication was identified as being sub-optimal in all six countries. This was due to a variety of reasons including low procurement because of low demand; procurement being centrally established, with psychotropic medication not prioritized; and communication breakdowns between clinics and distribution agencies. Concern about the lack of suitable available space to manage patients with behavioral disturbances as well as for counseling was identified as a problem in Ethiopia, Nepal, Uganda and South Africa. In South Africa, this was partly linked to the low priority given to mental illness compared with other conditions, with counselors sometimes having to use ‘a corner in the waiting room’ (district representative). Poor coverage of guidelines, including the WHO Mental Health Gap Action Program (mhGAP) intervention guideline, was reported as a challenge, particularly in Nepal and South Africa.

Participation and collaboration

Poor inter-sectoral collaboration at a national level, that filters down to regional and district levels, was identified by all six countries as resulting in a lack of clarity of mandates and roles, particularly for population- and community-based interventions. In terms of the development of policies and plans, a lack of participation by staff at the district and facility implementation levels was identified as a problem by India, Uganda, South Africa and Nepal. While not directly mentioned in responses from all these countries, lack of participation is likely to result in a lack of buy-in at the implementation levels as suggested by this quotation by a district representative in South Africa

the process was a talk-down … we were told exactly what to do and we were not allowed to innovate …”

With the exception of India, a lack of service user participation was identified across all six countries, with the general uncertainty on how to engage service users. Low mental health literacy, lack of empowerment of service users, and stigma were reported to contribute to this problem. A district representative from Nigeria indicated that,

[S]ervice users hardly get the opportunity to participate in decision-making, as they are deemed to be incapable. They only participate indirectly via the involvement of caregivers.

Equity, inclusiveness and ethics

In relation to equity, on the supply side, geographical barriers to accessing care were identified by Nepal, Uganda and South Africa. Cost of treatment was a barrier in Ethiopia, Nepal, Nigeria and Uganda, all of whom have limited social insurance. A district representative from Nigeria had the following to say about this issue:

There has never been [an] adequate supply of medications, and so it is only patients who can afford to buy from outside (private pharmacies) that receive them.

Stigma was identified as a barrier to accessing care across all the participating countries. On the provider side, stigmatizing attitudes by service providers were identified as a barrier to access. On the demand side, stigma and low mental health literacy by community members were equally reported as barriers to accessing care. Cultural beliefs were reported as an additional barrier to public sector care in Ethiopia, Nepal, Uganda and Nigeria. A district psychiatrist from Uganda had this to say:

There is low demand for mental health services because the public has a different orientation [traditional beliefs] … even when it is severe, they seek alternative methods of healing.

In relation to ethical practice, all countries except Nigeria indicated poor quality assurance practices for mental health services, leaving open the possibility of unethical practice. South Africa does have mental health review boards that are meant to provide an oversight function, but implementation is not consistent due to staff shortages. Concerning ethics for mental health research, all countries have ethical review boards for research. However, monitoring of research implementation was reported to not always be optimal across all the countries.

Information systems, accountability and transparency

Mental health information systems were generally weak across all the country sites, hampering information-based decision-making, planning and monitoring of service implementation. Mental health indicators were not routinely collected (Ethiopia, Nepal and Nigeria). Where they were being collected, they were deemed to be inadequate and/or not optimally used, due to a lack of resources and capacity for the collection and management of the data (India, South Africa and Uganda). In South Africa, a provincial representative reported: ‘nurses are inundated with collection of data which is just collected for its own sake and nothing is done with it’.

Weak implementation of systems to ensure accountability was identified in India, Ethiopia and Uganda, with lack of transparency of decision-making also emerging as a problem in Ethiopia and Nepal.

Needs and suggested strategies to promote good governance

A summary of the key needs and strategies identified by each country to improve mental health systems governance are outlined in Table 3.

Rule of law and strategic direction

In countries still in the process of developing new mental health policies and laws, identified strategies focused on developing and formally adopting these laws and policies. They included i) capacity-building for ministries of health to develop progressive mental health policies and laws; ii) strengthening advocacy efforts to increase the public health priority of mental health; and iii) the establishment of a mental health unit at the national level with an individual responsible for spearheading the process. In respect of those countries with new mental health policies or laws, strengthening implementation emerged as important. Strategies to achieve this included i) establishing or strengthening a national directorate or unit to oversee implementation; and ii) developing mechanisms to capacitate and ensure accountability for implementation at regional and district levels. In order to engage other sectors in the implementation of mental health policies and laws, the need for the development of formal structures for multi-sector engagement from national through to district level was also highlighted.

Effectiveness and responsiveness: human resources

Several strategies were identified to improve effectiveness and responsiveness of human resources and promote good clinical governance for the delivery of integrated mental health care by all country sites. First, in line with the task-sharing, revisions to pre-service training curricula of the different health care providers involved in mental health care at PHC level was emphasized across all country sites. Revisions identified included an orientation to patient-centered care for both specialists and generalists; additional skills to identify and treat mental illness for generalists; and additional skills for specialists to enable them to provide training, supervision and mentoring. Second, the need for job descriptions to be in line with diversified roles and functions was highlighted. Third, mechanisms to ensure continuous in-service training and mentoring in locally contextualized mhGAP tools, was identified. Change management, including workshops to orientate and provide clinical communication skills for patient-centered care, was an additional need identified in South Africa.

Given that mental health services are being integrated into existing primary health care service delivery platforms, optimal staffing of these platforms was identified as desirable. Strategies to improve staffing of these general platforms included (i) improving staff retention at facilities; (ii) ensuring that vacant posts are timeously filled; and (iii) promoting more equitable distribution of staffing across geographic regions, particularly between urban and rural areas.

Effectiveness and responsiveness: financing, infrastructure, medicines and protocols

The need for additional funding and a dedicated budget for mental health from national through to district level (based on costing of mental health care plans) was identified as central to responsiveness across all the countries. The need for social insurance in the form of disability grants for people with severe mental illness was also deemed important in those countries that do not have such a safety net.

With regard to infrastructure, across all countries, dedicated space for in-patient care and for patients with behavioral problems at general hospitals was identified as a need, as was the need for counseling space at PHC facilities. In relation to medicines and protocols, improving supply chain management of psychotropic drugs at the district hospital and PHC facility levels emerged as being required across the board, as was ensuring sufficient mental health protocols and guidelines.

Participation and collaboration

In relation to improving service user participation in advocacy and policy development, supporting the development of service user groups and providing capacity building to enable greater engagement in advocacy and policy development were identified as important strategies across all country sites. Likewise, building the capacity of planners and managers to engage with user groups was also highlighted. With regard to multi-sectoral engagement, clarification of roles and responsibilities of other sectors and the establishment of formal inter-sectoral mechanisms to facilitate engagement from national through to district level emerged as important strategies for all countries.

Equity, inclusiveness and ethics

In order to create more equitable access to mental health services, all countries suggested public health campaigns to reduce stigma and improve mental health literacy, as well as to improve the detection of mental disorders at a community level. A strategy that could assist in this regard is the integration of mental health into PHC.

With respect to ethical practice, building the capacity of service users to provide feedback on the quality of services emerged as important for promoting quality of care. In relation to the protection of people with mental disorders, adoption of the WHO Quality Rights project (World Health Organization 2012) was suggested as a strategy that could improve quality of care by all countries. Concerning research ethics, there were no suggestions for how to improve monitoring of research implementation.

Information, accountability and transparency

In addition to the need to ensure sufficient mental health indicators for monitoring and evaluation of integrated mental health services, all countries suggested research demonstrating the validity and feasibility of mental health indicators that could be used as an advocacy tool, as well as capacity-building to improve the quality of existing monitoring and evaluation processes.

Discussion

The expanded health system governance framework described in the introduction was found to be a useful framework for identifying challenges, needs and potential strategies for strengthening health system governance across the six participating Emerald countries. With regard to the first governance principle of Rule of Law and Strategic Direction, progressive mental health laws and policies that promote the decentralization and integration of mental health care into primary health care are an important first step in closing the treatment gap and promoting protection of people with mental disorders in LMICs. The latest data from the WHO Mental Health Atlas report indicates that 55% and 60% of responding countries in the African and South-East Asia region, respectively, have a stand-alone mental health law (World Health Organization 2014). While the study reported on here found a lower rate, three out of the four countries without updated legislation (Ethiopia being the exception) were moving in a positive direction with respect to rule of law. The slow pace of these efforts, as well as the potential for poor implementation in the absence of sufficient resources and capacity (as reported in countries that did have recent mental health legislation) is concerning, as it opens the possibility for malpractice and human rights abuses of people with mental disorders (Drew et al. 2011).

Technical support and capacitation of ministries of health to develop progressive mental health laws, within a human rights framework, were identified by countries as strategies that could assist to move this agenda forward. As part of their Mental Health Policy and Service Guidance Package, the WHO provides technical packages and assistance to countries embarking on the development of mental health laws (World Health Organization 2003). In addition, the WHO QualityRights Project provides a toolkit for helping countries to assess and ensure that the standard of care in both inpatient and outpatient mental health and social care facilities meets that of the International Convention on the Rights of Persons with Disabilities (World Health Organization 2012). Making use of such tools and technical support to develop progressive mental health laws and regulations that are in line with human rights standards can directly promote good practice and better standards of care (Petersen et al. 2015).

The situation with regard to strategic direction was better, with all countries having new or draft mental health policies. However, good governance requires the translation of mental health policies into implementable plans, while most countries reported implementation to be a challenge. To this end, capacity-building of management in public mental health at regional and district levels—to develop and implement mental health plans in line with policy directives—emerged as particularly important. This should include capacitation of district managers in implementation science methods to promote evidence-based scale-up of plans (Evans-Lacko et al. 2010, Glasgow and Chambers 2012).

In respect of financing, with the exception of India, where underspending of the mental health budget was a challenge, in the other countries, effectiveness and responsiveness were hampered by budget allocations for mental health that are inadequate to properly implement mental health policies and plans. Robust financing is required to ensure that there is sufficient financing of posts, medicines, protocols and suitable infrastructure to provide ‘good enough’ care. Spending on mental health in LMICs is still marginal, with the WHO Mental Health Atlas reporting that median government/public mental health expenditure is less than $2 per capita in middle-income countries able to report this, with expenditure in low-income countries even less (World Health Organization 2014). In the Emerald countries, India, Ethiopia and Nepal spend <2% of their general health budget on mental health. Uganda and South Africa spend more, ranging from 3% to 5% of the general health budget (Hanlon et al. 2014), with the bulk of these mental health resources historically being dedicated to tertiary care. In LMIC contexts, where there is still underservicing of patients requiring specialist hospital-based ervices, shifting resources from tertiary care to secondary and primary care (identified as a mechanism for financing of decentralized mental health care in high-income countries) may not be feasible without the injection of additional funding for mental health care (Thornicroft et al. 2010). Further, in order for task-sharing not to become task-dumping, a critical number of specialist posts is required to provide the necessary supervision and referral pathways for more complex and treatment-resistant problems (Ventevogal 2014).

Chisholm and others have estimated additional funding required for scaling up integrated mental health services in five of the Emerald countries (India, Nepal, South Africa, Ethiopia and Uganda) for three to four priority conditions. Over a 10-year scale-up period, they estimate that in all of the country settings, less than $0.10 per capita needs to be invested each year to reach specified target coverage levels (Chisholm et al. 2016a). While there are existing cost-effectiveness studies showing the benefits of integration for mental health outcomes in LMICs (Levin and Chisholm 2015), given other health priorities confronting LMICs, persuading finance ministries to make the necessary investment in mental health may require evidence of the cost benefits of mental health integration for health priorities such as HIV/AIDS, maternal and child health, and the rising burden of NCDs.

In relation to Human Resources, in addition to ensuring there are sufficient posts for integrated care—both specialist and non-specialist, as discussed above—there is also a need to ensure that staff are adequately equipped to take on the diversified roles of task-sharing. This requires in-service training as well as revisions to the curricula of pre-service training. This finding is not unique to the six Emerald countries (Patel et al. 2010). The WHO’s mhGAP Intervention Guide (World Health Organization 2010), provides evidence-based guidelines for the identification and management of mental disorders by non-specialists in LMICs and is well suited to in-service training programs (Barbui et al. 2010). With respect to pre-service training, the Institute of Medicine has spearheaded an important initiative in sub-Saharan Africa to identify and strengthen core competencies for integrated mental health care, and over the past few years, has co-hosted meetings in sub-Saharan Africa to advance this agenda (Institute of Medicine 2013, Academy of Science South Africa (ASSAf) 2014). Further, to address the biomedical/symptom orientation of service providers that promotes mind/body dualism, an orientation towards patient-centered care and building training in clinical communication skills were identified by at least one of the Emerald countries as being important to assist in the identification of less visible common mental disorders which often co-exist with other conditions.

In relation to the organization of health care delivery systems, integrated mental health care requires that health systems are able to accommodate patients in both recurring episodic acute phases as well as stabilized phases. Acute care requires adequate infrastructure, medication and a critical number of specialists—found to be inadequate in all the six Emerald countries, with a recent review indicating that management of psychiatric emergencies in non-specialist health settings in LMICs is generally poor (Nadkarni et al. 2015). An example of how acute care has been addressed in South Africa is through the new Mental Health Care Act 17 of 2002 (Department of Health 2004), which requires that service users undergo a 72-h emergency management and observation period for involuntary admissions to designated general district and regional hospitals across the country, before they can be referred to specialist psychiatric hospitals. However, once again, implementation remains a challenge, with infrastructure and specialist staff being found to be inadequate for the 72-h emergency management and observation service, despite findings that 75.6% of psychiatric admissions were involuntary or assisted (Petersen and Lund 2011).

The long-term health care needs of chronic mental disorders are similar to the needs of other chronic conditions. A key issue in long-term care is the need for adequate supply chain management of psychotropic medication to manage stabilized patients at the PHC level. Strengthening human resource capacity to promote patient self-management; establishing robust information systems that can identify non-adherent patients; as well as the formation of linkages with community supports to assist in tracing and returning loss-to-follow-up patients to care, to promote continuity of care and adherence, are all desirable. The rising burden of chronic communicable and NCDs in LMICs is likely to lead to the diversification and re-design of PHC delivery systems to accommodate long-term health needs, and may provide potential opportunities for integrated mental health care to leverage these chronic care platform delivery systems (Patel 2009).

With regard to Participation and Collaboration, weak inter-sectoral collaboration is probably a function of the historical focus on institutionalized care for severe mental disorders, which traditionally fell under health. Population and community platforms are, however, important for the promotion of mental health; mental health literacy and anti-stigma interventions; primary prevention and identification of mental disorders; as well as long-term community care and rehabilitation (Petersen et al. 2016). While there was general consensus that ministries of health should spearhead the development of mental health care plans, implementation was understood as needing to be multi-sectoral to accommodate the population- and community-level platforms. This remains a challenge across LMICs (Jenkins et al. 2011). Strategies identified by the six Emerald countries included the establishment of formal inter-sectoral mechanisms to facilitate engagement from national through to district level. Skeen and others take this further by suggesting that there is a need for legislation of inter-sectoral forums for mental health from the national to the local levels (Skeen et al. 2010).

With regard to weak participation across the countries, this aligns with the WHO Mental Health Atlas report that indicates that only 41% and 60% of responding countries in the African and South East Asia regions, respectively, indicated participation of persons with mental disorders and family members in ministry of health planning, policy, service development or evaluation (World Health Organization 2014). With the shift to decentralized integrated care, community engagement becomes more important as this approach requires community-based interventions and self-management beyond the formal health sector (Samb et al. 2010). This engagement should include the empowerment of service user groups as well as efforts to collaborate with local resources and supports within communities such as spiritual leaders and traditional healing services, to ensure the best quality care for patients. Kleintjes and others found that self-help organizations in Africa are helpful for promoting recovery, de-stigmatizing mental health issues through advocacy work, as well as promoting the protection of users’ rights and improving access to health care. To achieve this, support from ministries, non-governmental organizations and development agencies was identified as important for capacitating and providing support for sustainability (Kleintjes et al. 2013).

With regard to Equity and Ethics, poor mental health literacy and widespread psychiatric stigma were identified as barriers to equity across all the Emerald countries. There is growing evidence from high-income countries of the effectiveness of community-level interventions to improve mental health literacy (Jorm et al. 2005), as well as reduce stigma (Evans-Lacko et al. 2012), with a recent review suggesting that interventions incorporating education, protest, or contact are most effective (Corrigan et al. 2012). Delivering these interventions on a mass scale through community-based campaigns, which emerged as the preferred method in this study, remains a challenge. Recent evidence of the effectiveness of virtual contact via electronic media (Clement et al. 2013), however, presents a possible leapfrogging opportunity to tackle mental health stigma and low mental health literacy in LMICs with widespread access to mobile technology.

In relation to ethical practice, as suggested, the WHO Quality Rights project (World Health Organization 2012) could assist to improve quality of care across the participating countries. It provides a practical toolkit to assist in the monitoring of quality of care, with the WHO providing capacity-building and training in the associated toolkit. Concerning ethics in research, the need to strengthen the monitoring of research implementation by the local ethical review boards was noted as a concern.

Health information systems should include indicators for mental health to adequately monitor and evaluate implementation of integrated mental health care plans and promote accountability. Mental health indicators were collected in only half the Emerald countries and, where collected, were reportedly inadequate, with challenges in the collection and management of information as well. A Delphi study conducted among the Emerald country partners of mental health indicators that could be collected in LMICs resulted in a list of 15 indicators covering different domains of measuring mental health treatment coverage (Jordans et al. 2016). In addition to the incorporation of these indicators into the health information system, there is a critical need to build monitoring and evaluation capacity, particularly at district and regional levels, to implement and use these indicators for quality improvement and evidence-based implementation.

Limitations

Limitations of this study include that it was purely a qualitative study. Bias may have been introduced as a result of purposive sampling, as well as the fact that the study used data from only six selected LMICs. The results are therefore not generalizable to all LMICs. Further, the lack of a civic society voice is a limitation, given that this is a governance study that seeks to consider the principle related to inclusion and participation.

Conclusion

While overall good governance of a health system is necessary for integrated mental health care, a number of specific strategies to improve governance for integrated mental health care in the participating LMICs were identified. Beyond the need for the development of appropriate mental health policies and laws, this study highlights the challenges of implementation of these in contexts where the basic building blocks of the health system are weak. Inadequate financing of integrated mental health care remains the biggest challenge for good governance, with many of the other strategies identified being contingent on adequate funding. A silver lining may be that the rising burden of chronic conditions in LMICs may provide impetus for investment in strengthening of chronic care delivery systems that should provide a more enabling platform for integrated mental health care. Notwithstanding this, advocacy efforts highlighting the need for greater investment in the prevention and treatment of mental, neurological and substance use disorders in LMICs remain urgent. Inaction will result in greater health expenditure and impact negatively on economic growth, with the global cost as a result of lost production due to CMDs alone being estimated to be greater than $1 trillion a year up to 2030, if ‘business as usual’ continues (Chisholm et al. 2016b).

Ethics

Ethical approvals for the study were obtained from the authors’ institutes (King’s College London, Psychiatry, Nursing and Midwifery research ethics subcommittee; Nepal Health Research Council Ethical Committee; University of KwaZulu-Natal, Biomedical Research Ethics Committee; World Health Organization, Research Ethics Review Committee; Public Health Foundation of India, Institutional Ethics Committee; Addis Ababa University, College of Health Sciences IRB; Makerere University, College of Health Sciences School of Medicine Research Ethics Committee; University of Cape Town, Faculty of Health Sciences Human Research Ethics Committee; University of Ibadan Ethics Committee).