Background
The use of stakeholder analysis (SHA) as a systematic technique for gathering insights relating to a proposed action or reform is not new, and has commonly been used in business, change management, public policy, health care management and development. SHA gathers these insights by identifying, categorising and analysing individuals or groups that are likely to have a ‘stake’ (be affected by, or have an interest in) a proposed action [
1‐
3].
More recently, the utility of this approach has been reiterated amongst scholars of Health Policy and Systems Research (HPSR) [
4‐
6]. HPSR has evolved into an interdisciplinary field encompassing the policy realm, acknowledging the interconnections between policy and health systems, and highlighting the social and political nature of healthcare [
5]. SHA has been developed to better understand stakeholder power and positions around specific new policies or actions, and assess the likely implications for the acceptability of new policies or interventions. However, published research regarding its use or how to perform such analyses within the context of HPSR has been limited [
4,
7].
Health systems often fail to effectively implement evidence-based public health interventions, particularly in Low- and Middle-Income Countries (LMIC). This is due to poor knowledge translation, over-emphasising the production (supply) of research, rather than stimulating its consumption (demand) [
8]. Defined by the Canadian Institutes of Health Research, knowledge translation is a “dynamic and iterative process that includes the synthesis, dissemination, exchange and ethically sound application of knowledge to improve health, provide more effective health services and products, and strengthen the health care system [
9]”.
In order to understand the ‘needs’ of knowledge users and minimise poor knowledge translation, SHA has been applied as a technique to firstly identify the stakeholders (many of whom are likely to be knowledge users), and to gather insights into their position vis-a-vis the proposed action of a mental health research programme. Such insights will likely point to strategies for stimulating the demand for research amongst knowledge users, and hence, minimise the knowledge gap between research and policy.
Mental health is no exception to poor knowledge translation, and the failure of health systems implementing evidence-based interventions. This is supported by the assertion of global mental health scholars that mental health does not receive the policy attention expected of it globally [
10‐
13].
The primary aim of this study is to document the value of SHA as a technique for identifying and characterising support for the proposed action. In addition, the study aims to identify the level of involvement of stakeholders within the PRogramme for Improving Mental health carE (PRIME).
PRIME aims to generate health systems research on the best ways to integrate and scale up mental health into maternal and primary health care systems in LMIC, and to stimulate demand for this research. PRIME does this by advocating for evidence-based decision-making, and by promoting the uptake of its implementation research amongst knowledge users, including health and related sector policymakers and practitioners, through a dedicated ‘Research Uptake Strategy’ [
14]. PRIME is piloting the scale-up of mental health services in one District in each country: Sodo in
Ethiopia, Sehore in
India, Chitwan in
Nepal, Dr Kenneth Kaunda in
South Africa and Kamuli in
Uganda [
15].
In terms of SHA, the ‘proposed action’ identified is linked to the goal of PRIME, which is to scale-up mental health services in the districts of the five LMIC. Until the advent of this initiative, little was known about PRIME stakeholders, and their engagement with mental health policy and systems research. The SHA technique is intended to enable PRIME to identify and understand stakeholder power and positions, and assess the likely implications for the acceptability of the proposed action. Furthermore, this paper aims to add further knowledge about the method of SHA by sharing the experiences of its application in the field of mental health.
Results
Cross-country analyses by stakeholder groups
Five country teams identified specific stakeholders by group, and tabulated their characteristics in relation to the proposed action: scaling-up mental health care. Cross-country stakeholder analyses of policy makers (Additional file
1: Table S1), donors (Additional file
2: Table S2), health practitioners (Additional file
3: Table S3), persons affected by mental illness (Additional file
4: Table S4), civil society (Additional file
5: Table S5), the media (Additional file
6: Table S6) and academics (Additional file
7: Table S7) are analysed presented at country and cross-country levels (see Additional file
1: Table S1, Additional file
2: Table S2, Additional file
3: Table S3, Additional file
4: Table S4, Additional file
5: Table S5, Additional file
6: Table S6, Additional file
7: Table S7) and described below.
Policy makers
PRIME Ethiopia identified the high level of interest and support of the WHO regional office and the Federal Ministry of Health. Other sectors such as the Ministry of Social and Labour Affairs were reported to be supportive, although with lower levels of interest. Amongst other democratic institutions, Parliament was reported to be supportive given the personal interest of an influential Member of Parliament, although the team reported that greater efforts were necessary for engaging with Parliament.
Given the collaboration of the PRIME Indian team with the WHO in implementing the mental health Gap Action Programme (mhGAP), the WHO regional office was reported to have high levels of interest and support. Despite the District (Madhya Pradesh) Department of Public Health and Family Welfare being a collaborator in PRIME consortium, it was reported that the interest in the issue was mixed given that some senior level policy makers may have had low levels of interest, especially considering other priority programmes such as Reproductive and Child Health, Family Planning and HIV/AIDS. Other stakeholders identified as having a medium level of interest in the issue included the Ministry of Health and Family Welfare, and National AIDS Control Organisation (national level), Department of Medical Education, the State Mental Health Authority and State AIDS Control Society in Madhya Pradesh (district level). Amongst these stakeholders, particularly those believed to have a medium impact on the issue, opportunities to mobilise them were identified. Other opportunities for mobilisation identified include stakeholders with a high influence, high impact but low interest, such as the Madhya Pradesh Legislative Assembly, and Panchayati Raj Institutions (local self-governing bodies).
The PRIME Nepal team identified the Primary Health Care revitalisation department as having a high interest in the issue, and being supportive of the issue as an active collaborator with PRIME. Other non-health sectors were also identified as having a high interest, specifically the Ministry of Women, Children and Social Welfare; and the Ministry of Peace and Reconstruction, and opportunities were identified to mobilise these Ministries. Although supportive, the WHO country office was believed to have a medium interest, whilst the Ministry of Health and Population was believed to have a low interest due to, in both cases, the prioritisation of other issues. Other democratic institutions, such as Parliament, have not been explored as yet given Nepal’s political environment.
PRIME South Africa regarded policy stakeholders, including WHO, Department of Health (national, provincial and district) and other government departments such as social development, as being supportive of the issue. However, it was believed that the stakeholders have medium levels of interest given that it is not a priority. Furthermore, the team indicated that the interest in mental health exists in relation to priority health issues such as HIV/AIDS and maternal health.
PRIME Uganda identified the WHO country office and Ministry of Health as having high levels of interest, and as being highly supportive of the issue. Despite low levels of interest, it was believed that other non-health sector departments and democratic institutions such as Parliament had medium–high levels of influence. Opportunities to mobilise these sectors were identified.
Donors
All countries believed DFID UK to be supportive of the issue, given their funding of the consortium. PRIME in Ethiopia and Nepal believed regional or country DFID offices to have high levels of interest in the issue, whilst India and South Africa believed there to be medium levels of interest amongst DFID country offices given the competing priorities of other Research Programme Consortia (RPCs). Uganda predicted a low level of interest amongst DFID country offices. Most countries, with the exception of India, report medium levels of interest amongst other donors or development agencies.
Health practitioners
Whilst Ethiopia regarded mental health specialists, particularly psychiatrists, as having high levels of interest for the issue, it regarded PHC and community health workers as having medium levels of interest. Despite the support identified, Ethiopia believed that further engagement was needed with mental health specialists and particularly amongst PHC workers, as their involvement is crucial for meeting PRIME’s objectives.
Although PRIME India did not mobilise support from mental health specialists at that stage, PHC workers (medical officers and nurses/midwives), community health workers and voluntary workers, were all regarded as having medium levels of interest in the issue. This is due to the fact that they are overburdened, without having much time to devote to things beyond their scope of work, as per the priorities set by the District Health Administration. Voluntary health workers known as Accredited Social Health Activists (ASHA) who are responsible for psychosocial education at the community level were identified as having high levels of influence on the issue, particularly from the perspective of mental health stigma reduction.
In Nepal, although health practitioners (PHC workers, community and voluntary health workers) were thought to be supportive, medium levels of interest were registered due to them being overburdened. In the case of mental health specialists, the lower interest was also explained by their presence mostly in urban areas. It was believed that the issue of scaling up mental health services will have a high impact on PHC workers (doctors, nurses and midwives), given the new roles that they will be expected to perform.
South African health practitioners were understood to have medium levels of interest, given that support at a provincial level was being nurtured at the time. Nevertheless, the level of impact was understood to be high given the political will of the government at the national level, and the fact that implementation of policy and legislation rests with the provinces and districts.
Uganda registered high levels of interest and support amongst mental health specialists, however, recognised that greater support and interest needs to be garnered from PHC and CHWs.
Persons affected by mental illness
Ethiopia believed there to be high levels of interest and influence amongst mental health service user groups, people with psychosocial disabilities, and their families or carers. Although service user groups (such as the Mental Health Society, Ethiopia) were supportive of the issue, their level of influence was regarded as low given the small number of members. Opportunities were identified for mobilising people with psychosocial disabilities and their families or carers, such as their inclusion in Community Advisory Boards (CABs). Based on previous projects, these persons affected by mental illness were expected to be supportive of the issue.
India recorded medium levels of interest in the issue amongst service users, people with psychosocial disabilities, and their families or carers. High levels of influence, and impact of the issue on these sub-groups were recognised by India, as have opportunities to mobilise them.
Despite believing a lower level of influence, Nepal recorded high levels of interest and support amongst service user groups, people with psychosocial disabilities, and their families or carers in relation to the issue.
Although not yet mobilised, medium levels of interest were predicted by South Africa, pointing to the fact that some affected persons may be uninterested due to apathy, discrimination and stigma and a lack of awareness and education about mental health and their right to health care.
In Uganda, families or carers of people with psychosocial disabilities were understood to have low levels of interest, followed by medium interest from people with psychosocial disabilities, followed by high levels of interest from service user groups, which are also believed to be supportive of the issue. Opportunities to mobilise people with psychosocial disabilities, and their families have been recognised.
Civil society
In Ethiopia, international NGOs and FBOs were believed to be supportive of the issue, with medium levels of interest. Some in these groups are represented in PRIME’s Community Advisory Board (CAB). The medium–high influence of FBOs, and their power to raise awareness through anti-stigma campaigns have been recognised. Due to the high potential for influencing communities, opportunities to mobilise CBOs were recorded.
In India, low levels of interest in the issue were recorded amongst international NGOs, however, medium to high levels of interest were recorded amongst national NGOs, many of whom were regarded as being supportive. Opportunities to mobilise CBOs, FBOs and international NGOs were documented.
Nepal recorded high levels of interest and support amongst national NGOs and CBOs, regarding the issue as having a high impact on these groups given their ability to provide technical support to PHC staff, and advocate for people to seek mental health services. Lower levels of interest and support were recorded amongst FBOs and traditional healers given the fact that this group has been providing a service for persons affected by mental illness. Although supportive, international NGOs were also regarded as having lower levels of interest given the competing priorities.
South Africa identified some specific international NGOs (Basic Needs) and national NGOs (South African Federation for Mental Health) as having high levels of interest and support, with opportunities to mobilise CBOs and FBOs.
Ugandan civil society was recorded as having low-medium levels of interest in the issue. International NGOs were regarded as being supportive of the issue, whilst the opportunity to mobilise CBOs and FBOs has been recognised.
The Ethiopian media was regarded as having medium levels of interest and being supportive of the issue at a national level. Variable levels of interest amongst District media, and opportunities to engage this sector have been recorded.
Although regarded as highly influential, the media in India were expected to have low levels of interest in the issue. Opportunities to mobilise them further were recognised.
Nepalese media were believed to have medium–high interest in the issue, and to be highly supportive of the issue, particularly from the perspective of sensitization and awareness.
Whilst South African media had not yet been mobilised regarding the issue, medium levels of interest, and high levels of influence were expected, particularly in terms of placing the issue higher on the policy and implementation agenda.
The Ugandan national media was believed to have a medium interest in the issue, and was identified as being supportive, with high levels of influence, and impact on the actor. Lower levels of interest were recorded amongst District level media, and opportunities to mobilise media at this level were recorded.
Academics
Although academics and researchers in Ethiopia (such as high level officials in universities) were regarded as being supportive of the issue, the level of interest was believed to be low-medium given that mental health programmes were not prioritised. However, the potential influence on the issue was noted as high, as mental health issues could become mainstreamed into education and training at the universities.
In India, universities were regarded as having high levels of interest in order to enhance the academic pool of resources, however, other research institutes were believed to be less interested in the issue given their objectives, priorities and links with government.
Nepal regarded universities and other research communities as having high levels of interest, and support for the issue due to the potential impact that the scaling-up of mental health services has on their development agenda.
South Africa regarded universities and research communities as having medium levels of interest, with some academics having specific interests in mental health, whilst others, more general interests in public health. The potential influence of universities and research communities regarding the issue was recognised, as more research was being published about scaling-up mental health care, which was believed to have a greater impact.
Uganda believed universities to be supportive, with a medium interest in the issue. Lower interest was anticipated amongst other research communities, and the opportunity to mobilise them has been noted.
Cross-country force field analysis map: summarising stakeholder engagement opportunities
Content analysis of the stakeholder analysis tables identified a range of opportunities for increasing the evidence-based scale-up of mental health care across countries. However, not all of these may succeed given varying degree of support and power of stakeholders.
Hence, a cross-country force field analysis (Table
1) indicating the extent of stakeholder support (across) and their perceived power (down) has been applied. The block on the top right indicates those stakeholders that are the most influential and supportive of scaling-up mental health care, whilst the bottom left block indicates the least influential, least supportive stakeholders.
Table 1
Cross-country stakeholder forcefield analysis map
High | | Parliament (IN, UG) | MH Specialists, PHC Workers, CHW (SA) | WHO, MoH (ET, NP, UG) |
PHC Workers (UG) | Parliament (ET) |
Volunteer Workers (IN) | DFID UK (ET, NP, UG) |
Persons with mental illness, Families (ET, IN) | DFID local (ET) |
Service User groups (IN) | Other donors (NP) |
CBOs (ET) | MH Specialists (UG) |
I-Media (SA) | N-Media (NP, UG) |
R-Media (ET, SA) | State/District Media (NP) |
N-Media (IN, SA) | Universities (ET) |
State/District Media (ET, IN, SA) |
Medium–high | | | | Non-Health Ministries (ET) |
MH Specialists (ET) |
Service User groups (NP) |
INGOs, NGOs, CBOs (NP) |
FBOs (ET) |
N-Media (ET) |
Medium | | Non-Health Ministries (NP, UG) | | WHO, MoH (IN, SA) |
CHW (UG) | | Non-Health Ministries, Parliament (SA) |
Persons with mental illness (UG) | | DFID UK (IN, SA) |
CBOs, FBOs (IN) | | DFID local (IN, NP, SA) |
I-Media (IN) | | Other donors (ET, SA) |
State/District Media (UG) | | CHW (ET) |
Universities (IN, SA) | | Service User groups (UG) |
| | INGOs (UG) |
NGOs (IN, SA) |
CBOs, FBOs (SA) |
Universities (UG) |
Low–medium | | MH Specialists, PHC Workers, CHW (IN) | | MH Specialists (NP) |
| | PHC Workers (ET) |
Persons with mental illness, Families (NP) |
Low | | Non-Health Ministries (IN) | FBOs (NP) | PHC in MoH (NP) |
DFID local (UG) | | PHC Workers, CHW, Volunteer Workers (NP) |
Other donors (IN, UG) | | Service User groups (ET) |
Persons with mental illness, Families (SA) | | INGOs (ET) |
Families of persons with mental illness (UG) | | I-Media (NP) |
INGOs (IN) | | Universities, Research Institutes (NP) |
CBOs, FBOs (UG), Research Institutes (UG) | | |
In terms of priority setting in the context of limited resources, it can be deduced that success amongst the mobilised will be most likely amongst highly influential (powerful) stakeholders, who are most supportive. Should further capacity exist, medium-highly influential stakeholders that are supportive can be targeted next, followed by the supportive medium influential stakeholders, or the highly influential non-mobilised stakeholders. Although it was early in the programme and many stakeholders were not mobilised at the time of this research, it is noteworthy that none of the key informants recorded any opposition or resistance to scaling-up mental health care, indicating that there is no direct opposition to the issue from a wide range of stakeholders.
In Ethiopia, the policy engagement opportunities with the WHO regional office, Federal Ministry of Health and Parliament could be prioritised. In the case of other stakeholders, the donor sector (DFID UK and local offices) and universities could also be prioritised.
In the case of India, policy engagement opportunities could be prioritised with the WHO regional office; Ministry of Health and Family Welfare and AIDS control organisation (national level); Department of Medical Education, state mental health authority and state AIDS control society in Madhya Pradesh (district level). Other opportunities identified that could be prioritised included the donor sector (DFID UK and local offices) and national NGOs.
Nepal could prioritise its policy engagement activities with the WHO and the Ministry of Health, including PHC revitalisation department. Other engagements could be prioritised with the donor sector (DFID UK and other donors), national and district media, service user groups, international and national NGOs and CBOs.
Policy engagement priorities in South Africa could be with the local WHO, Department of Health (national), non-health sector departments (Department of Social Development) and Parliament. Other stakeholders that could be prioritised by South Africa include health practitioners (mental health specialists, PHC workers and CHWs), donors (DFID UK and other donors) and civil society (NGOs, CBOs and FBOs).
Uganda’s policy engagement priorities could be with the local WHO and Ministry of Health. Other stakeholder engagement priorities could be with donors, mental health specialists, national media (high power); followed by service user groups, international NGOs and universities (medium power).