Exploring decision makers’ knowledge, attitudes and practices about decentralisation and health resources transfer to local governments in Burkina Faso

ABSTRACT

In the health sector, decentralisation mainly consists of the devolution of administrative functions to local governments. Since 2009, Burkina Faso has engaged in a process to transfer health resources to local governments. This study examines the decision-makers’ knowledge, attitudes and practices (KAP) about the decentralisation and health resources transfer to local governments in Burkina Faso. We used a qualitative research method. In-depth semi-structured interviews were conducted with key decision-makers. The data collected went through a directed qualitative content analysis. Findings suggest that all respondents are aware of the rationale of the decentralisation and resources transfer to local governments. The vast majority of respondents have a positive opinion towards decentralisation and the main elements that appear to be motivating their attitude, are the expected outcomes from decentralisation. The practical experience was limited to awareness raising, training, supervision, technical assistance and resources mobilisation. Poor collaboration between health districts and local governments, the control of certain resources by the state and the health districts constrain the implementation of health resources and skills transfer policy at grassroots level. Careful attention should be given to the country’s political context and institutional design.

RÉSUMÉ

Dans le secteur de la santé, la décentralisation consiste essentiellement en une dévolution des fonctions administratives aux gouvernements locaux. Depuis 2009, le Burkina Faso a engagé un processus de transfert des ressources en santé aux communes. Cette étude examine les connaissances, attitudes et pratiques (CAP) des décideurs sur la décentralisation et le transfert des ressources en santé aux communes. Les données collectées à travers un interview semi-structuré ont été analysées en utilisant la méthode de l’analyse de contenu. Les principaux résultats indiquent que tous les participants sont informés des objectifs de la politique de décentralisation et de transfert des ressources. La majorité des participants ont une opinion positive de la décentralisation et les éléments qui motiveraient leurs attitudes sont les résultats escomptés de la décentralisation. Les expériences pratiques des participants en lien avec le transfert des ressources étaient limitées à l’information, la formation, la supervision, l’assistance technique et la mobilisation des ressources. La faible collaboration entre les districts sanitaires et les communes, le contrôle de certaines ressources par l’Etat central et les districts sanitaires entravent l’implémentation de la politique de transfert des ressources au niveau local. Une attention particulière doit être accordée au contexte politique et l’organisation institutionnelle du pays.

KEYWORDS:

• Decentralisation
• health resources transfer
• KAP
• decision-makers
• Burkina Faso

Motsclés:

• Décentralisation
• Transfert des Ressources en Santé
• CAP
• Décideurs
• Burkina Faso

Introduction

Decentralisation refers to the dispersal of power in public planning, management and decision-making from the national level to sub-national or local levels, and from higher to lower levels of government (Collins & Green, 1994; Mills, 1994; Rondinelli, 1981). Decentralisation is also about shifting authority, resources, functions and service delivery responsibility from the central government to the local level within a country. This transfer of power, functions and resources from central to local level authorities requires political and administrative reforms (Larson & Ribot, 2004).

Many Sub-Saharan African countries have undergone major decentralisations reforms over the past decades with the overall goal of improving the quality, access and equitable delivery of public services provided through or facilitated by local government authorities (Masanyiwa et al., 2012). Decentralisation is frequently advocated as a means to improve public services delivery based on the assumption that in a decentralised system services are more responsive to local needs and demands of service users because citizens can directly or indirectly influence decisions about resource allocation and service delivery (Junaid et al., 2005; Conyers, 2007; Regmi, 2014; Ribot, 2002; Saltman et al., 2007). In most countries, decentralisation reflects a broader process of political and economic reform (World Bank, 1997).

Although there has been a general consensus on the overall goal of decentralisation, its motivation and objectives may vary from country to country, from region to region and from service to service, ranging from improving service delivery in Uganda and Cote d’Ivoire (Shah et al., 2004) to responding to ethnic and regional conflicts in South Africa (Treisman, 2000). Decentralisation is also often adopted by national level elites as a strategy for mobilising and maintaining regional power bases (Crook, 2003). In the health sector, the expectation is that decentralisation provides the opportunity for health systems to attain both technical and allocative efficiency, empower local governments, increase accountability and make gains in areas like quality, cost and equity (Sumah et al., 2016).

In June 1991, Burkina Faso engaged in democratic renewal with the adoption of a new Constitution. This laid the foundation of decentralization through the organization of the country into local governments and thus, the establishment of local governance. In 2004, the country adopted a new general code of territorial councils (Code Général des Collectivités Territoriales in french), which defined the institutional framework for the implementation of decentralisation. In 2007, a decree was issued by the central government to provide a strategic framework for implementing the decentralisation policy.

The type of decentralisation in Burkina Faso can be categorised as devolution, although devolution and de-concentration components co-exist with a strong presence of central government through the establishment of administrative units (Prefect, High Commissioner, Governors). Since 2006, elected local governments are in place in the 351 communes (municipalities), the lower administrative levels to which the central government is progressively transferring resources.

Regarding the health sector, Burkina Faso has undertaken a reform of its health system since 1992, which has led to the decentralisation of the healthcare system. This reform was led by the Alma Ata declaration and was aimed to operationalise the declaration’s recommendations to improve access to health care for all citizens. The reform was part of a wider macroeconomic policy and implementation of structural adjustment programmes, which require control of public expenditure as well as changes in public and private sector institutional structures (Gilson & Mills, 1995). This reform has resulted in the decentralisation of the health system with the establishment of health regions (13 in 2018), divided into health districts (70 in 2018) that are the operational level of the national health policy. The district health services are organised into two levels, the first of which comprises the primary health care centres (1896 in 2018) that provide communities with basic preventive and curative primary health care. The second level consists of district hospitals (45 in 2018) that are the point of referral for the primary health care centres. The type of decentralisation in the health sector can be categorised as administrative decentralisation in the form of de-concentration. The central government (Ministry of Health) has transferred some responsibilities to health regions and health districts, mainly planning and management of health services. The health policy formulation, human resources and financing remain the responsibility of the central government. The resources transfer to local governments focuses only on the primary healthcare centres.

Two main objectives underline the resources and skills transfer to local governments in Burkina Faso. The first is to boost the grass-root development by enhancing the local governments capacities and ownership, and the second is to strengthen democracy and governance. In line with the resources and skills transfer, the central government has set two principles. The first is the principle of subsidiarity, which suggests that a skill must be preferably transferred to the level where it can best be managed. The second is the principle of gradual progress in resources and skills transfer (Ministère de la Decentralization, 2014).

The resources transfer is done in consultation with a range of stakeholders involved in the decentralisation. This makes the process rather complex as it includes a large number of actors and affects the multi-level system of the state (national, regional and local) (Oluwu & Wunsch, 2004). This also presents a number of challenges for the decentralisation. Firstly, political decentralisation changes the balance of power between national, regional and local levels, disturbing the existing system. Secondly, decentralisation also tends to change the balance of power between politics, civil society and citizens at the local level because it creates new spaces for the participation of civil society and citizens in local politics (German Development Institute, 2018).

Various studies have been conducted on decentralisation in Burkina Faso. These have mostly focused on the political decentralisation context and processes (Bekale, 2011; Champagne & Ouedraogo, 2008; Guiro et al., 2011; Ky, 2010; UNDP, 1999); resources transfer process and lessons learned (Bekale, 2011; Guiro et al., 2011; Ministère de la Decentralization, 2014); the impact of decentralisation on poverty reduction (Roark et al., 2001). A SWOT analysis of health resources transfer among healthcare decision-makers was also conducted (Zon et al., 2017). None of these studies dwelled specifically on the knowledge, attitudes and practices among health resources transfer’s decision-makers. The insights from a KAP study can help to reveal misconceptions or misunderstandings that may represent obstacles to the implementation of decentralisation and health resources transfer in Burkina Faso.

Given that the health resources and skills transfer in Burkina is in a learning process and there is limited knowledge and insight on the topic, this study examines to what extent decision-makers are familiar with the resource transfer. Specifically, the study seeks to gauge their knowledge, attitudes and practices related to the transfer of health resources and skills to local governments in Burkina Faso. This qualitative exploration could be of interest to other countries as well, e.g. Sub-Saharan African countries where decentralisation in the health sector is being implemented or considered for implementation.

Methods

This study used qualitative research methods to collect data among decision-makers involved in the health resources and skills transfer process in Burkina Faso. For this purpose, in-depth semi-structured interviews were conducted with decision-makers. These decision-makers included mayors of municipalities, health district managers and managers of health care centre as well as policy decision-makers at a regional and national level, and donors/partners.

Three dimensions were explored in this study: knowledge, attitude and practices, which are known as the KAP study framework. This framework is often used by researchers to structure qualitative studies in the field of health policy and management (Medecin du Monde, 2011; Muleme et al., 2017). For the purpose of the study, we used an operational definition based on the WHO’s KAP study framework (WHO, 2008). Knowledge refers to the interviewees’ awareness and understanding (what is known) of the study phenomenon, in this case the decentralisation as well as the related transfer of health resources and skills. Attitude refers to the interviewees’ perceptions of the health resources transfer, including any preconceived ideas (thoughts or feelings) that they may have towards it. Practices refer to the ways (what is done) in which interviewees demonstrate and apply their knowledge and attitude towards the health resources transfer through their actions or behaviours.

A non-probabilistic sampling method was used for the data collection. The participants were purposefully selected because of their involvement or participation of their organisation in the decentralisation process in Burkina Faso, as well as due to their management position. These selection criteria were applied to assure that decision-makers with relevant experience were included in the study. An invitation was sent to the selected participants to request their authorisation and explain the study objectives, scope and methods. For the interviews, we used a semi-structured interview guide. Giving that the study focus was the respondents’ knowledge, attitudes and opinions, a semi-structured interview allowed the respondents the freedom to express their views in their own terms in order to gather information and story behind each participant experiences with decentralisation and resources transfer.

The questionnaire (research tool) was developed to collect information on the following aspects: (i) knowledge – awareness and understanding of decentralisation, health resources and skills transfer, (ii) attitudes – perceptions and feelings of the rationale or intent of the health resources and skills transfer policy, (iii) attitudes – support of the policy of decentralisation and health resources and skills transfer, (iv) practices – their role in the health resources an skills transfer process, (v) practices – collaboration between health districts and local governments, vi) practices – effectiveness of the resources and skills transfer process.

To ensure the quality of the information, the research tool was pre-tested before commencing data collection. Qualified interviewers were recruited and trained on survey methodology and interview skills including protection of confidentiality, anonymity and privacy of each participant. The main researcher supervised the data collection and checked the consistency of the data collected. The interviews were transcribed verbatim. The data collection and analysis were carried out in French. The final results were translated in English.

Ethical considerations were safeguarded throughout the study process. The research protocol was submitted for ethical approval, and such approval was granted by the ethical committee of the ‘Institut de Recherche en Sciences de la Santé (IRSS)’ of Burkina Faso (clearance number A002-2018/CEIRES).

All data were analysed, reported and stored in formats that do not allow identification of the individual participants. A verbal informed consent was obtained from each individual who participated in the study, and was registered by the interviewer. This involved informing the participant about the purpose for which the information is obtained and on its use in a manner that can be understood by the participant. Participants were informed about the option to not answer questions that they felt unconformable with, and the option to stop the interview if they wanted to do so.

The data collected have undergone a directed qualitative content analysis (Bengtsson, 2016; Hsieh & Shannon, 2005; Vaismoradi et al., 2013). Data have been analysed in five main steps: reading for content, coding, displaying, data reduction and interpretation. Information that reflected the reality as expressed by participants, has been retained and reported. The key themes, namely knowledge, attitude and practices, have been used to extract relevant data form the transcripts.

Knowledge

An overall knowledge score was computed by aggregating the variables related to the knowledge that was assessed through two questions. The first question asked whether participants were aware of the health resources transfer to local government, and the second question asked whether they were informed about the health resources transfer policy/bylaw regardless the source of information. Participants who gave a positive response to both questions were considered to have ‘good knowledge’ of the decentralisation process. Participants’ knowledge was rated as ‘poor knowledge’ when one or both questions were answered negatively.

Attitudes

To assess the participants attitudes, a question was asked about the intent of the policy/decision to transfer the health resources from the central state to local governments. The participants provided an explanation of their answer in free text from their own respective points of view. The participants attitudes were considered as ‘supportive’ of the decentralisation process when their responses contain at least one relevant rationale/intend of resources transfer. Otherwise, the attitudes were considered as ‘not supportive’.

Practices

Three questions assessed participants’ practices. The first question that asked the collaboration between health districts and local governments was assessed on 5-points scale (0 = not at all satisfied, 5 = very satisfied). In the second, the participants were asked to indicate their contribution to the resources transfer process. The participants were considered to have contributed to the decentralisation process when they declared to have been involved in any action or activity related to health resources transfer (awareness raising, training, supervision, documents development, technical assistance, resources mobilisation). The third question was asked in order to gauge the effectiveness of resources transfer. The responses were assessed on 5-points scale (0 = not at all effective, 5 = very effective) and participants were invited to provide information or explanation to support their position.

The key results are presented narratively and supported by quotations from the interviews.

Results

Table 1 presents results on the profile of the study respondents. In total, 20 in-depth interviews were carried out with health care and decentralisation decision-makers. Over half of the respondents (55%) live in urban settings that is the habitual place of residence of national policy makers and country donors.

Table 1. Profile of study respondents.

	Residence		Total
Profile of respondents	Urban	Rural
National policy- and decision-makers	2	-	2
Regional council President	-	1	1
Health regional Directors	1	2	3
Health district team Managers	2	1	3
Mayor of municipalities	2	2	4
Primary healthcare centre Managers	2	3	3
Country Donors	2	-	2
Total	11 (55%)	9 (45%)	20

The results of the data analysis are presented below following the KAP study framework.

Knowledge – awareness of the resources and skills transfer process

As indicated by the respondents, one of the prerequisites for the success of a policy reform is the support of the different stakeholders, which heavily depends on their awareness and knowledge about the objective of the reform. All respondents (20/20) declared that they were informed about the health resources and skills transfer policy through three main channels: information notes or letters (12/20), meetings (5/20) or short trainings (3/20).

According to a health care centre manager:

We were informed about the health resources transfer to local governments. The health district management team shared with us the documents explaining the objective of this new policy led by the central government.

Another interviewee, a member of a local government’s council, added this:

I have participated in several meetings on resources transfer at local and central level. These meetings were planned to explain the reform and the role of the different stakeholders involved.

The interviewees were asked whether they have any documents (letters, notes, decree) related to health resources and skills transfer. More than half (12/20) of the respondents said that they had a document at their disposal about resources and skills transfer. Among those who did not have any document (8/20), half (4/8) were health care centre managers. On this matter, a head of a health care centre said:

When I got posted here, I haven’t found in the archives a document on health resources transfer.

Attitude – perceptions of rationale of resources and skills transfer

The interviewees had different perceptions regarding the reasons or rationale for health resources and skills transfer from the central state to local governments, including:

• to increase local stakeholders’ involvement and ownership (6/20 respondents)

• to improve local governance and community participation (5/20 respondents)

• to enhance efficiency and effectiveness of resources management (4/20 respondents)

• to bring services closer to beneficiaries (4/20 respondents)

• to implement the decentralisation policy (3/20 respondents)

• to provide local governments with adequate resources to boost local development (2/20 respondents)

• to implement the principle of subsidiarity (1/20 respondents)

• to increase local resources mobilisation (1/20 respondents)

• to accelerate the decision-making process at local level (1/20 respondents)

On the rationale of resources and skills transfer, a mayor observed the following:

From my understanding, the decision to transfer health resources to local governments is part of the decentralization policy. As such, the resource transfer should be seen as a logical continuation of the decentralization policy implementation.

From the perspective of some interviewees, namely policy-makers, the health resources and skills transfer to local government seeks to bring health services to communities through greater empowerment of local authorities. With relation to that, a policy-maker explained:

The elected local governments remain the closest to the population and on that account, the better informed about the citizens’ needs. Hence, the resources transfer will strengthen their capacities in order to respond more adequately to population health needs.

Attitude – opinions on resources and skills transfer

The interviewees were asked about their opinions about resources and skills transfer to local governments. The vast majority of respondents (17/20) had a positive view of the resources and skills transfer policy and the reasons justifying their opinions were based on the expected outcomes of the resources and skills transfer that can be arranged into five broad areas:

• better management of health facilities (8/17);

• improved capacities of local governments to provide services of better quality to communities (5/17);

• better management of resources transferred (3/17;

• increased population’s access to health services (2/17);

• increased local governments’ accountability towards communities and beneficiaries (1/17).

To support his view, a health district manager said:

I strongly believe that resource transfer to local government is a good decision. It will enhance local decision-making and strengthen the local governments’ capacities to offer services adapted to the needs of individuals and communities.

A few interviewees (2/20) expressed contrasting assessment about the resources and skills transfer policy. To illustrate his disagreement, a mayor said:

I have the feeling that the central government wants one thing and also its opposite. It has transferred one part of the resources and retains one part, making the resources transfer incomplete.

In the same way, a former mayor added:

My negative opinion is reinforced by the fact that the resource transfer process was not properly planned. Some prerequisites have been not met and some people have the feeling that the country has been put under pressure by country’s partners and donors.

One (1/20) respondent reported a neutral opinion on the resources transfer policy because of its political exploitation.

Practices – contribution to the resources and skills transfer process

To assess their contribution, the respondents were asked about the role they have played in the resources and skills transfer process. Vast majority of respondents (14/20) indicated that they had not directly participated in the design of the decentralisation process, either because they were not employed in their current position (5/14), or they were not invited or asked to participate in that policy phase (9/14). Among those who declared that they have participated in the process (6/20), the actions undertaken were awareness raising and training of the different actors (2/6), development of documents (manuals, guidelines, standards) for decentralisation and resources transfer (2/6), technical assistance and resources mobilisation (1/6), supervision (1/6).

To expand on his contribution to the process, a respondent, a health district manager declared:

I have participated in the development of a manual to guide the management of health resources transferred to local governments. This was an important step because this manual is key to ensure that all the local governments own a harmonized understanding and approach to resources management.

A country donor expressed support to the process in this way:

We strongly support the country to fully implement the decentralization policy, including the resources and skills transfer process. For this matter, we provide a technical assistance and funding to design the policy documents and guidelines, as well for capacity building.

Similarly, a mayor added:

t was necessary to widely inform and raise awareness of the population on the resources transfer process. For this purpose, we have organized awareness raising sessions and meetings with communities-based organizations to explain the rationale and the process of resources transfer to local governments.

Practices – collaboration between local governments and health districts

The study sought to gauge the relationship between the local key actors in the resources and skills transfer, namely the local governments’ councils and health district management teams. For this purpose, 7 respondents (4 mayors and 3 health district managers) were asked about their mutual relationship. The findings indicated that the collaboration is not satisfactory according to the majority of respondents (5/7). The main issue raised by both groups is the lack of formal coordination framework between local governments and health districts.

A health district manager illustrated it in this way:

I think we need to work closely to ensure a proper implementation of this new and important policy. I have the feeling that local governments have a misunderstanding of their role in the resources and skills transfer and which makes difficult our collaboration.

On the collaboration matter, a mayor expressed his concern as follows:

There is no formal coordination mechanism between health districts and local governments. This gives the impression that everyone is working in isolation.

Practices – effectiveness of the resources and skills transfer process

The interviewees were asked about the current level of resources and skills transfer from the central to local governments. From their response, it can be noted that the vast majority (17/20) found that as of today the resources (financial, human, equipment) were transferred to local governments according to the decree issued by the central government. Three out of 20 respondents deem that there is a delay in the resources and skills transfer.

A question was asked to the interviewees about the extent to which the resources and skills transferred, and to what extend the local governments had taken over the management of these resources. The responses showed that nearly half (9/20) of the interviewees felt that the health resource management rested with the health district management team and the central state.

To illustrate this point, a mayor said:

The ministry of health and the health districts continue to manage entirely the human resources. For instance, the health personnel are appointed by the health district management teams without involving the local governments whose role is limited to issue authorizations for annual leave.

In relation to this matter, another respondent, a local government council’s member, reiterated by saying that:

Compared to the education’s sector where all the resources were transferred and handled by the local governments, the health sector drag their feet.

This feeling of weak ownership of health resources and skills management by local governments was greater among health services providers at grassroots level, as illustrated by the point of view of a health care centre manager:

We still don’t see the contribution and the accountability of local governments in the management of health facilities. All the issues and concerns that we have reported to the Mayor are not yet resolved. The hierarchical and reporting relationship is unclear.

Among the respondents who stated that local governments have taken over the management of resources as of today (11/20), some believed that the blame lies with local governments as pointed out by a respondent, a health district manager:

From my point of view, some stakeholders fail to perform their duties. For instance, some local governments continue to think that health districts management teams must handle certain issues related to health facilities.

Another respondent, a policy-maker supported this view by noting that:

The management of resources transferred by the central state is not still well-controlled by local governments. For instance, it was reported that some local governments don’t even know that funds were transferred by the central government in their bank account … . while they claim a delay in resources transfer

Discussion

The study has assessed the decision-makers’ knowledge, attitudes and practices related to health resources and skills transfer to local governments in the context of decentralisation in Burkina Faso. All respondents in our study expressed awareness and understanding of resources and skills transfer policy to local government, which is crucial for the policy implementation and may ensure its success. In particular, Sakyi et al. (2011) argued that the implementation is likely to suffer whenever there is misunderstanding and uncertainty over the policy objectives due to a communication gap as well as a failure on the part of the central government staff to share information.

According to our results, based on the responses of the decision-makers we interviewed, several factors have led to the decision of the central state to transfer resources to local governments. The main targets include: (i) to increase the local stakeholders involvement and ownership, (ii) to improve local governance and community participation, (iii) to enhance efficiency and effectiveness in resources management, (iv) to bring services closer to beneficiaries, (v) to implement decentralisation policy, (vi) to provide local governments with adequate resources to boost local development, (vii) to implement the principle of subsidiarity, (viii) to increase local resources mobilisation and (ix) to boost of the decision-making process at local level. These reasons, stated by the respondents in our study, correspond to evidence reported elsewhere on the rationale for decentralisation (Conyers, 2000; Inman & Rubinfeld, 1997; Kundishora, 2009; Litvack et al., 1998; Musgrave, 1983; Oates, 1972; Ribot, 2002; Saltman et al., 2007; Sumah et al., 2016).

According to Conyers (2000) four broad categories outline decentralisation objectives: local empowerment, administrative efficiency and effectiveness, national cohesion and central control. In the same way, Ribot (2002) suggests that the underlying developmentalist logic of decentralisation is that local institutions can better discern, and are more likely to respond to, local needs and aspirations. Thus, the central state of Burkina must make every effort to ensure an effective implementation of decentralisation given that the main objective of a resources and skills transfer policy is to boost the grassroots development through the leadership of local governments. This underpins the principle of subsidiarity adopted by the central state. It suggests that the decentralised government, which is closer to the people, is better able to respond to the preferences of its citizens (Ministère de la Décentralisation, 2014).

Regarding the decision-makers’ perceptions, our study found that the majority of the respondents have positive attitude towards decentralisation and transfer of resources and skills, and there is only little criticism. The main elements that appear to be motivating their attitude are the expected outcomes from decentralisation, including: (i) better management of health facilities, (ii) improved capacities of local governments to provide services of better quality to communities, (iii) better management of resources transferred, (iv) increased population’s access to health services and (v) increased local governments’ accountability towards communities and beneficiaries.

The opinions expressed by respondents in our study, are likely to be supportive to the decentralisation and health resources transfer with respect to the expected outcomes and their awareness and understanding of the resources transfer policy. It is recognised in previous research that the attitudes of beneficiaries or stakeholders to a given policy is key in the success of its implementation. In that regard Makinde (2005) argues that where the implementers of a policy are perceived to be favourable to the policy, their attitude to the implementation is highly positive. The expectations from the decentralisation raised by the respondents, have been identified by previous studies on decentralisation as well. According to Smoke (2003), the most purported benefits of decentralisation are the existence of democratic mechanisms that allow local governments to discern the needs and preferences of their constituents, and the way for these constituents to hold local governments accountable to them. In a review on decentralisation in African, Ribot (2002) has highlighted the various expected outcomes from decentralisation in selected countries. These are mainly: (i) promoting grassroots democracy and local development, (ii) achieving national solidarity oriented towards development, (iii) sharing of power between central and local levels and (iv) learning process in democracy and people’s participation in development.

Regarding the decision-makers’ practices regarding resources and skills transfer, the findings revealed that many of the respondents did not directly participated in the process, because they were not employed in their current position at that time or they were not entitled to participate in that policy phase. The role or contribution of those who have participated in the decentralisation process, was limited to awareness raising, training, supervision, design of manuals and guidelines, technical assistance and resources mobilisation. The findings showed that those who have not directly participated in the process, have more positive perceptions and opinions towards resources and skills transfer. From our findings, it could be said that the direct contribution of stakeholders or beneficiaries in the implementation process of a policy, would not make a condition for their support to that policy. It emerged from the results that stakeholders’ knowledge and awareness of the expected outcomes from decentralisation and health resources transfer policy made their perceptions and opinions supportive.

The study revealed a poor collaboration between health districts and local governments. Two main factors could contribute to this stormy collaboration between health districts and local governments. The first lies in the fact that health manager’s expectations about the issue of transfer of resources are mixed probably driven by their fear or feeling of losing some of their prerogatives of authority over the primary healthcare centres placed under the responsibility of local governments. In relation to that, a decentralisation policy review in Burkina Faso (Champagne & Ouedraogo, 2008) suggested that any social change creates resistance particularly among the state’s officials. Often, the lack of understanding of decentralisation policies is used as an excuse to justify the officials’ refusal to collaborate with the decentralised bodies and councils. Such hostilities are related to a more general fear of losing their power, in particular financial control, and therefore their influence on people. The second factor has to do with the discrepancy in the decentralisation stage between the health sector and the political sector. The progress in the implementation of health resources could be hampered by the fact that the decentralisation in political sector is in a fairly advanced stage (devolution) compared to the health sector (de-concentration). One the key consequences of this discrepancy is the mismatch between health and administrative territorial breakdown, thus creating confusion in the planning, the implementation of health services delivery and oversight among health managers and local governments.

Challenges in collaboration between health sector and decentralisation sector were reported by several studies (Guiro et al., 2011; Ministère de la Décentralisation, 2014) conducted in Burkina Faso in which the lack of formal coordination mechanisms between health centres and local governments, impede the provision of health care to communities. On this matter, a study in Ghana (Viglo, 2014) found that the reasons given for the failure of decentralisation often included the unclear definitions of the distribution of functions between central government agencies and local authorities. Smoke (2003) also asserted that one of the greatest deficiencies in decentralisation efforts is the lack of coordination of the actors involved and the consequent failure to build linkages among the components of decentralisation at the national, intergovernmental and local levels.

Another dimension of the decision-makers’ practices was related to the effectiveness of resources and skills transfer. The findings suggested two main positions among respondents: a first group who argued that the resources and skills were entirely transferred and local governments have taken over it, the second group that comprises largely health managers expressed an uncertain outcome to health resources transfer. The main issue raised by this group is the weak ownership of local government that failed to perform their duties, which is a reason for this group to question the added value of health resources transfer. A further analysis of health manager’s position may conceal the reality of a feeling of losing their prerogative over a field exclusively reserved for them. This could undermine their support to the process and explain the current challenges to collaborating with local governments revealed by the study, which can adversely affect the implementation of health resources transfer.

The issue related to the effectiveness of resources transfer in the context of decentralisation was raised in many studies in Burkina and others Sub-Saharan African countries (Bekale, 2011; Guiro et al., 2011; ROCARE, 2011). A systematic review on local governments experiences in health services provision in Sub-Saharan African countries (Zon et al., 2017) highlighted a partial or incomplete transfer of resources and skills due to the intensive state intervention and political interference. The key functions remain under the control of the central or regional government (decision-making of financial resource allocation, resources transfer, management of human resources).

Overall, our results show that the key decisions-makers interviewed have a fair knowledge and understanding, as well as a positive attitude towards the resources transfer in Burkina Faso. However, they seem to be divided regarding the effectiveness of health resources transfer. This appeared to be confirmed by different reports and studies that have highlighted the partiality in the health resources transfer either in Burkina Faso or others West African countries (Rocare, 2011). The implementation process of decentralisation and health resources transfer, likewise any public policy, is heavily determined by the political will and the adequate financial resources. According to Rondinelli et al. (1983), in nearly all countries where governments have attempted to decentralise, they have faced serious problems of implementation that arose from insufficient central political and bureaucratic support, and from ingrained centrist attitudes and behaviour on the part of political and administrative leaders. Decentralisation was weakened by the failure to transfer sufficient financial resources to those organisations to which responsibilities were shifted.

From the study results, an in-depth analysis of different points of view in the interviews, highlights a number of issues. According to Derick and Leighton (2001), these challenges are encountered in implementing a policy of resources or power transfer from central state to local governments:

• Political and process dimension: It is well known that decentralisation is a highly political because it changes the balance of power between national, regional and local levels. While there may be strong technical arguments in favour of health sector decentralisation, without attention to the politics of decentralisation, reforms may fail to yield the expected outcomes. The process dimension of decentralisation highlights the importance of stakeholder participation, effective communication and political will.

• Tensions and conflicts among objectives: In health care decentralisation, while the objective of responsiveness to local preferences may be met, from the standpoint of the national health system, the result is a sub-optimal allocation of resources. This could be explained by the underfinancing of the vast majority of local governments that heavily depend on central government subsidies and transfers of resources. These transfers of resources (functions and responsibilities) are not always accompanied by corresponding measures of financial, material resources and particularly human capacities to adequately manage these resources.

• Roles and responsibilities: A misunderstanding over the roles and responsibilities of different stakeholders in the decentralisation process can heavily impede a policy implementation. A clear definition of stakeholders’ roles in addition to a formal collaboration or consultation mechanism among stakeholders, could improve the coordination towards policy implementation.

• Space for participation: Attention should be paid to stakeholders’ awareness raising as a crucial step in the design and implementation of a given policy. The focus should be on the policy rationale, expected outcomes in order to enhance stakeholders’ support and participation to policy implementation.

• Delay in policy implementation: The different components or phases of a policy should be implemented as planned or agreed with the different stakeholders. This could help to avoid frustrations among stakeholders and ensure their commitment and ownership towards decentralisation’s and health resources transfer’s outcomes.

Limitations of the study

This is a qualitative study that provides an insight into the health resources in Burkina Faso. The sample was limited to 20 participants from the targeted organisations and institutions that are not representative of all organisations and institutions involved in the decentralisation process. Therefore, the findings should be seen as indications of the selected key decision-makers’ KAP towards health resources and skills transfer, rather than providing a representative picture that can be generalised to the entire country. Future studies are needed to thoroughly evaluate the decentralisation process in Burkina Faso.

Conclusions

This study sought to assess the knowledge, attitudes and practices (KAP) of decision-makers about health resources and skills transfer to local governments in Burkina Faso. The findings indicate that the key factors that determine the decision-makers positive attitudes and support to decentralisation are their good knowledge and understanding of the health resources transfer policy. However, the health managers’ support to decentralisation could drop if they feel they are losing their power due to the resources transfer process, exacerbating the antagonism between health districts and local governments. The poor collaboration between the different stakeholders, the control of certain resources by the central state and the health districts could constrain the implementation of health resources transfer policy at a grassroots level.

From the results, it can be concluded that the success of failure of resources transfer in the context of decentralisation depends on the country politic and institutional context, as well others associated factors. Decentralisation is neither good or bad strategy but it rather depends on the implementation of the decentralisation and the country-specific institutional design (Daflon & Madiès, 2012). The study recommends to Burkina Faso and others countries that have initiated decentralisation reforms, to pay attention to the key factors that influence the success of decentralisation, among others the country political context and institutional design, the stakeholders’ awareness and understanding in order to enhance their accession and support, the establishment of a clear framework for an effective collaboration, the transfer of sufficient financial resources. The process must be preceded by an in-depth analysis and a careful planning and organisation to ensure a successful implementation.

Acknowledgements

We are grateful to the Ministry of Health and the Ministry of Decentralization officials, the local government officials, district health authorities, heads of health centres, country donors and other stakeholders who participated in our study. The views expressed in this paper are those of the authors and do not necessarily represent those of their institutions.

Disclosure statement

No potential conflict of interest was reported by the author(s).