UNIVERSITY OF GHANA HEALTH SECTOR DECENTRALIZATION IN GHANA: EXPLORING THE IMPLEMENTATION CHALLENGES AT THE HOSPITAL LEVEL BY FRANCIS ADANE (10245289) A THESIS SUBMITTED TO THE DEPARTMENT OF PUBLIC ADMINISTRATION AND HEALTH SERVICES MANAGEMENT, UNIVERSITY OF GHANA BUSINESS SCHOOL, UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF AN MPHIL IN HEALTH SERVICES MANAGEMENT DEGREE JUNE, 2015 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I do hereby declare that this work is the result of my own research and has not been presented by anyone for any academic award in this or any other university. All references used in the work have been fully acknowledged. Nonetheless, I bear sole responsibility for all errors and omissions inherent in the study. ………………………… ……………………… FRANCIS ADANE DATE (10245289) University of Ghana http://ugspace.ug.edu.gh ii CERTIFICATION I hereby certify that this thesis was supervised in accordance with the procedures laid down by the University of Ghana. ……………………………………... ……………………… GORDON ABEKA- NKRUMAH DATE (SUPERVISOR) University of Ghana http://ugspace.ug.edu.gh iii DEDICATION I dedicate this work solely to my mother, Veronica Abenaa Twenewaa for her vast contributions and unwavering support to my well-being in the course of my academic career. University of Ghana http://ugspace.ug.edu.gh iv ACKNOWLEDGEMENT Firstly, I thank GOD ALMIGHTY for giving me the strength, knowledge and guidance to complete this thesis. To my supervisor Dr. Gordon Abekah Nkrumah, I keenly appreciate your unflinching attention, encouragement and critical evaluation of this thesis. I would also like to acknowledge the efforts of Dr. Justice Bawole, head of department, Dr. Albert Ahenkan and all the staff of the department of Public Administration and Health Services Management for their priceless and intuitive contributions on this thesis right from the beginning. I am also appreciative of my family, my late father Mr. Simon Adane, my mother Mrs. Veronica Abenaa Twenewaa and my siblings Peter, Felicia, Monica, Mary, and Victoria for their constant support and counsel. Not forgetting Papa Kwaku Adane- Adjei and Kwarteng Adjei for their entertainment in times of stress and boredom. Finally, I say ayekoo to Maureen Ansah for her support throughout this program. To my friends Emmanuel, Acheaw, and Smith I am very grateful to you for your advice and encouragement. Finally I am indebted to all friends and family members who in diverse ways helped to make this thesis successful. University of Ghana http://ugspace.ug.edu.gh v TABLE OF CONTENTS Content Page DECLARATION .......................................................................................................................... i CERTIFICATION ....................................................................................................................... ii DEDICATION ............................................................................................................................ iii ACKNOWLEDGEMENT .......................................................................................................... iv TABLE OF CONTENTS ............................................................................................................. v LIST OF TABLES ...................................................................................................................... ix ABSTRACT ................................................................................................................................. x CHAPTER ONE .......................................................................................................................... 1 1.1 Background of the study ........................................................................................................ 1 1.2 Problem statement .................................................................................................................. 5 1.3 Objectives of the study ........................................................................................................... 7 1.4 Research questions ................................................................................................................. 8 1.5 Significance of the study ........................................................................................................ 8 1.6 Chapter Disposition ................................................................................................................ 9 CHAPTER TWO ....................................................................................................................... 11 REVIEW OF LITERATURE .................................................................................................... 11 2.1 Introduction .......................................................................................................................... 11 2.2 The Concept of Decentralization ......................................................................................... 11 2.3 The historical review of current decentralization policies ................................................... 15 2.3.1 Developed economies ....................................................................................................... 15 2.3.2 Developing economies ...................................................................................................... 16 2.4 Theoretical Framework ........................................................................................................ 17 2.4.2Operational definition of the variables ............................................................................... 20 2.4.3 Decentralization implementation process; an empirical review ....................................... 21 2.4.4 Policy process ................................................................................................................... 23 University of Ghana http://ugspace.ug.edu.gh vi 2.4.5 Agenda setting and policy formulation ............................................................................. 26 2.4.8 Empirical review of the link between Decentralization and some key reform measures for its successful implementation ....................................................................... 29 CHAPTER THREE .................................................................................................................... 57 DECENTRALISATION IN GHANA ....................................................................................... 57 3.1 History of Decentralisation in Ghana ................................................................................... 58 3.2 General Legal Regulatory Framework and revolution of Decentralisation in Ghana ......... 65 3.3 Review of the status of health sector decentralisation in Ghana .......................................... 72 3.6.1 Historical development of the hospital ............................................................................. 78 3.7 Justification of the study areas ............................................................................................. 80 3.8 Conclusions .......................................................................................................................... 80 CHAPTER FOUR ...................................................................................................................... 82 RESEARCH METHODOLOGY ............................................................................................... 82 4.1 Introduction .......................................................................................................................... 82 4.2 Research Methods ................................................................................................................ 82 4.2.1Research Paradigm ............................................................................................................. 82 4.3 Research Design ................................................................................................................... 83 4.3.1 Sources of Data ................................................................................................................. 83 4.3.2. Sampling/ Target population ............................................................................................ 84 4.3.3 Instruments for data collection .......................................................................................... 84 4.3.4 Data collection process ..................................................................................................... 85 4.3.5 Data management and analysis ......................................................................................... 85 4.3.6 Reliability and Validity ..................................................................................................... 86 4.3.7 Ethics ................................................................................................................................. 87 4.3.8 Limitations of the study .................................................................................................... 87 4.4 Summary .............................................................................................................................. 87 CHAPTER FIVE ........................................................................................................................ 88 University of Ghana http://ugspace.ug.edu.gh vii PRESENTATION OF FINDINGS ............................................................................................ 88 5.1Introduction ........................................................................................................................... 88 5.2 Description of findings ......................................................................................................... 88 5.3 Implementation challenges of decentralization at the hospital level ................................... 88 5.3.1 Funding ............................................................................................................................. 88 5.3.2 Communication barrier between key policy implementers .............................................. 90 5.4 Important contextual factors taken into consideration before decentralization at the hospital level ....................................................................................................................... 95 5.5 Implementation status of some key reform measures of decentralization at the hospital level .................................................................................................................................... 97 5.6 Summary of implementation status and major gaps of some selected policy reform measures of decentralization at the hospital level ............................................................ 105 5.7 Summary ............................................................................................................................ 106 CHAPTER SIX ........................................................................................................................ 107 DISCUSSION OF RESULTS .................................................................................................. 107 6.1 Introduction ........................................................................................................................ 107 6.2 Implementation challenges ................................................................................................ 107 6.3 Important contextual factors taken into consideration before decentralization at the hospital level. .................................................................................................................... 122 6.3.1 Socio-economic potential ................................................................................................ 124 6.3.2 Unequal endowment of local capacity ............................................................................ 127 6.3.3 Geographical factors ....................................................................................................... 129 6.3.4 Cultural factors and ethnic diversity ............................................................................... 132 6.4 Implementation status of some key decentralization reform measures ............................. 135 6.4.1 Policy issues and objectives ............................................................................................ 135 6.4.2 Organisational reform ..................................................................................................... 137 6.4.3 Planning and financing .................................................................................................... 139 6.4.4Human resource and capacity issues ................................................................................ 143 University of Ghana http://ugspace.ug.edu.gh viii 6.4.5Inter- sectoral coordination .............................................................................................. 146 CHAPTER SEVEN .................................................................................................................. 149 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ........................................... 149 7.1 Introduction ........................................................................................................................ 149 7.2Summary of Findings .......................................................................................................... 149 7.3 Conclusions ........................................................................................................................ 150 7.4 Recommendations from the study ...................................................................................... 153 7.5Contributions of the study ................................................................................................... 155 7.5.1Theory .............................................................................................................................. 155 7.5.2Policy ............................................................................................................................... 155 7.6 Suggestions for future research .......................................................................................... 156 REFERENCES ......................................................................................................................... 158 APPENDICES ......................................................................................................................... 169 Appendix A: Interview guide for key policy implementers at the hospital level .................... 169 Appendix B: Facility observation checklist ............................................................................. 170 University of Ghana http://ugspace.ug.edu.gh ix LIST OF TABLES Table 5.6 Summary of implementation status and their major gaps for some key Reform measures………………………………………………………………….. 105 University of Ghana http://ugspace.ug.edu.gh x ABSTRACT This study explored the implementation challenges of decentralization at Dormaa Presbyterian Hospital and Sampa Government Hospital. Data were collected using qualitative research approach using interviews guide. Key informants who had in-depth knowledge and play role(s) in the implementation of decentralization at the district level were purposively selected. In all, 22 key informants were interviewed from the two districts (11 in each district).In addition, extensive analysis of key policy documents for the two hospitals was made. Findings from the study showed that there are challenges that militate against the implementation of decentralization reforms at the hospital level. These included: inadequate human resource and low capacity at both national and local and district level; lack of communication among key policy implementers at the district level; financial constrains; inadequate support from government; and non-involvement of community members in planning and management of health program. It was also revealed that implementation of decentralization policy at the hospital level was not the best as many of the important policy measures were never initiated or were only partially implemented. The challenges existed at both policy design and the implementation phases. To ensure successful implementation of decentralization at the hospital level, the following recommendations were made: There is the need to restructure local governance institutions which are mandated to carry out the implementation and monitoring of decentralization. There is also the need for broader participation during policy formulation process; the role of development partners, civil society and private sectors responsible for health decentralization must be specified. Furthermore, policy makers are to carry out capacity needs assessment for policy feedback and implementation at decentralized unit. Development of an appropriate human resource and management policy to address the needs of the decentralization policy is also prudent. University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTION 1.1 Background of the study Decentralization, as defined by Crook (2003), refers to the transfer of power, authority and functions from the central to local authorities. The term decentralization is also defined by Muriisa (2008) as transfer of power and authority from the central government to regional or sub-national governments. The concept has been recognized globally as important means of improving delivery of public goods and services (Dillinger, 1994). Decentralization can be explained in general terms a ―socio-political‖ process of power sharing arrangements between central governments and local authorities in planning, management and decision making (World Bank, 2010; Smith et al., 1990). This desire is often triggered by the wish to bring politicians and policy makers closer to citizens(Powell et al., 2005; Rondinelli et al., 1989) and also to make public services efficient and effective (World Bank,2004). In recent years, many developed and developing countries have either experimented or fully decentralized the delivery of health services and/or other functions of the health system. In most cases, decentralization has been adopted to improve accountability to local population, efficiency in service provision, equity in access and resource distribution, or to increase resource mobilization (Coutolenc, 2012). Bossert (2002) argued that decentralization brings about community participation, inter- sectorial collaboration, effectiveness, efficiency and equity. In a similar vein, studies in most developing countries revealed positive outcomes of decentralization, consisting of increased access to central government resources for people who were previously neglected, especially those who were residing in the rural regions and local communities. This consequently was found to enhance participation and increased local administrative University of Ghana http://ugspace.ug.edu.gh 2 capacity to negotiate with central government organs so that they can later allocate more resources for local activities development (Bennett 1997; Cheema et al., 1983).However, the ability for local managers to handle more resources at the local level calls for more strategic measures to increase local capacity to handle additional responsibilities at the local level. Decentralization reform has a goal of enhancing equity, increase efficiency and ensuring more participation and responsiveness of government to citizens (Larson and Ribot, 2004).However, whether this reform is achieving these objectives especially in the public sector of most developing countries in managing public sector activities are not certain as results have been mixed in the decentralization literature. According to Bennett (1997), several decentralization efforts in developing countries have recorded positive results such as increased access to central government resources for people who were previously neglected. The author revealed that, those who were residing in rural regions and in local communities, there was enhanced participation and increased local administrative capacity. This ensured negotiation with central government organs so that they can later allocate more resources for local development activities. However, the benefits and challenges of decentralization show mix results, especially in the health sector. In many developing countries, these reforms may be happening at a much faster speed but with fewer resources as compared to similar reforms which have previously taken place in most western countries, thus making their implementation a considerable challenge. University of Ghana http://ugspace.ug.edu.gh 3 Amore recent study conducted by Frumence et.al(2013), showed that inadequate funding, untimely disbursement of funds from the central government, insufficient and unqualified personnel, lack of community participation in planning and political interference were the main challenges to the implementation of the decentralization policy in Uganda‘s health sector. Decentralization could create additional challenges to the national health system, thus, there is the need to develop strong organizational capacity, as an integral part of the decentralization process (Mayhew, 2003; Oyaya & Rifkin, 2003). In most developing countries, reform measures such as capacity issues, human resource, community involvement in decision making, and communication which are needed to take place in organizations before decentralization reform is implemented are often not given the necessary attention, and this might have caused the failure of this policy in most developing countries( Dhakal, 2009) Onyach-Olaa, (2003), found out that, though a lot of successes have been achieved in terms of improving governance and service delivery through democratic decentralization and community involvement, lack of capacity at sub-national levels of government was frequently cited as the challenge to the implementation of the decentralization. In most cases, capacity requirements and assessment at the local level do not exist. In cases where such capacity requirement exists, efforts to build capacity is often constrained by resources which happens to be acute at the local levels. Evidence from Tanzania and Uganda revealed that, the lower tier of government lacked the ability to manage finances and maintain proper accounting procedures. Since these were requisite for transferring money to the lower tiers, the money received was less University of Ghana http://ugspace.ug.edu.gh 4 compared to pre decentralization; it was clear as spending on primary health care fell from 33% to 16% during decentralization (Akin, 2001). In 1979, Indonesia established ―village governments‖ with locally elected village heads accountable to village council that would determine budget priorities. The study of the 48 villages showed that, since village heads chose the members of the council, accountability to the villagers was weak; only 3% of the village proposals were included in the district budgets. Those villagers who participated in decentralized government organizations were more likely to speak out at the village council meetings speaking out their worries and that of others in the village (World Bank, 2001). According to the World Bank (2005), for the past years, implementing decentralization policy in the health sector has been one of the most emphasized development issues in many developing countries. This implementation has appeared to focus on health sector reform, changes in health financing system, and human resource development (Green, and Collins, 2003), but less attention has been paid to the challenges facing the implementation of decentralization at the hospital level. As observed by the World Bank (2005), decentralization is not always helpful to strengthen the health systems in developing countries; however the limited research done in these areas especially in developing countries has still not made the evidence clear with regards to the challenges encountered in the implementation of decentralization reform at the hospital level(Larbi, 1998 ). A lot of official evaluations and researchers have investigated the factors affecting the successful implementation of health sector decentralization and reforms in Ghana (Batley and Larbi, 2004; Agyepong, 1998; Annan, 1997; Smithson et al., 1997). Though these previous researches equally studied important aspects of the reform, only few have extensively documented on the implementation challenges of decentralization at the University of Ghana http://ugspace.ug.edu.gh 5 hospital level. The question however is, will the challenges of health sector decentralization in general be the same as those at the hospital level? Given the above question, this study seeks to understand the challenges facing the implementation of decentralization at the hospital level in Sampa Government Hospital and Dormaa Presbyterian Hospital in the Brong-Ahafo region of Ghana. These hospitals were chosen because one is a private and the other is public. The researcher would like to know if the challenges will be the same for both private and public hospitals or there will be difference. If there are differences in such challenges what accounts for it. 1.2 Problem statement Many developing countries including Ghana have adopted the decentralization policy longer than twenty (20) years. About 85% of developing countries have undergone some type of decentralization reforms (World Bank, 2002; Burki et al. 1999). However, the implementations of such reforms have been constrained by many factors. In the health sector, the case is not different in spite of all the several reported achievements, there are also reported challenges facing health the sector. For instance, decentralization has failed to achieve its intended goal of increasing the power of local people, because many governments have often implemented de-concentration or delegation form of decentralization at the expense of devolution(Martinez-Vazquez et al., 2006).The lack of capacity at local units to implement and manage responsibilities for public services delivery, especially those related to public finances and maintenance of proper accounting procedures, has been reported as one of the factors that constrained the implementation of decentralized public services in Uganda and Tanzania according to Ahmad(2005). Similarly, centralized and weak management, weak legal and institutional framework, lack of implementation strategy, poor financial and human resource management system, lack University of Ghana http://ugspace.ug.edu.gh 6 of adequate preparation for managing the reform, weak capacity at all levels, and political instability were found to be the major drawback of effective implementation of decentralization in the public health sector (Dhakal et a.l,2009). These problems normally occur because the international literature on decentralization usually focuses on government and its main modalities and features. According to Coutolenc (2012),inadequate research on how decentralization operates in the health sector, the precise challenges it faces in addition to those facing the general decentralization has been problematic for it successful implementation. For a policy to achieve its intended purpose and be effective, two conditions must be met. Firstly, the given intervention should be able to produce the desired effect, that is policy design issues, and next the policy should be followed as intended and that is implementation issues according to (DPHO, 2007). In many cases, the challenge in implementation seems to lie both at policy design phase and implementation phase. Gilson (2005) argues that implementation failure can be the result of stressing policy outcomes but virtually ignoring the policy process. It is important that guidelines and working procedures should be prepared in consultation and the involvement of those who are in charge of implementation (Wattet al., 2005), but evidence in most developing countries indicates that reform efforts are in general politically motivated rather than addressing real policy needs(Jeppsson, Ostergen and, Hagstrom, 2003).This situation adversely affects the implementation process. It must be pointed out that, this problem is not different from what pertains in Ghana. Weak and fragmented management system, poor implementation strategy, lack of motivation, weak institutional framework, lack of financial and human resource, and staff attitude and behavior were found to be the dominant factors of ineffective implementation University of Ghana http://ugspace.ug.edu.gh 7 of health sector reforms in the literature according to Sakyi (2008). Though quite a number of official evaluations and researchers have investigated the factors that inhibited the smooth implementation of health sector decentralization and reforms in Ghana (Batley and Larbi, 2004; Agyepong, 1998Annan, 1997; Smithson et al., 1997), exploring the implementation challenges of this reform at the hospital level has not received much attention from the literature especially. Health sector decentralization has also been problematic because the district health management team, health workforce and stakeholders who are very instrumental in the implementation at the district level have limited information about the policy; because information is highly centralized (Sakyi, 2010). According to the author, it is an undeniable fact that a lot of benefits have been made from this reform, yet there is rich empirical and anecdotal evidence portraying the process and management of decentralization in the health sector has been faced with several problems. Most of these obstacles facing the implementation of decentralization process are still not understood as there is little evidence to understand these challenge. The major question is why the decentralization reform designed to address some of the major problems at the hospital level are not implemented as planned? This constitutes the core question of the study. The study therefore seeks to explore these problems.It will also suggest ways through which the identified problems can be solved. 1.3 Objectives of the study The overall aim of the study is to examine the challenges that militate against the implementation of decentralization at the hospital level in Ghana. University of Ghana http://ugspace.ug.edu.gh 8 The specific objectives are to; I. Find out the important contextual factors taken into consideration in the process of decentralization at the hospital level and how these factors influence the implementation of the decentralization policy at the hospital level II. Analyze the implementation status of some key reform measures of the decentralization policy at the hospital level 1.4 Research questions I. What are the challenges that militate against the implementation of decentralization at the hospital level? II. What important contextual factors were taken into consideration in the process of decentralization at the hospital level and how have these factors affected the implementation of the decentralization policy at the hospital level? III. What is the implementation status of some key reform measures of the decentralization policy at the hospital level? 1.5 Significance of the study The significance of the study can be viewed in three areas: research, practice and policy. Considering significance to research, this study extends beyond the current research on ascertaining the impact of decentralization on health sector performance, and implementation challenges of decentralization in the health sector in general, by exploring the challenges of this reform at the hospital level. Literature on the challenges of University of Ghana http://ugspace.ug.edu.gh 9 decentralization at the hospital level in Ghana is scarce; therefore the findings of this study will contribute to knowledge in this field of study. Concerning significance to practice, the study will enable local health and central managers to identify major challenges affecting successful implementation of the decentralization policy and to adopt appropriate options to address the challenges and sustain the opportunities provided by decentralization at the hospital level. With regard to its significance to policy, the study will provide information on the challenges faced at hospitals in implementing the decentralization policy. Finally, the study will enable policy makers to know how to take certain contextual factors into consideration before they decentralize, since each district has its own characteristics that need to be taken into consideration. 1.6 Chapter Disposition The work is organized in seven main chapters. The first chapter captures the introductory aspect of the study, statement of the problem that the research sought to address, the objectives and rationale for the study. The chapter also contains the relevant research questions that the study intends to answer. Important theoretical and empirical underpinnings of implementing decentralization in the health sector are discussed in chapter two. Chapter three gives an account of decentralization in Ghana and discusses the backgrounds of the two hospitals studied. Chapter four presents the methodology of the study and describes the research paradigm used for the study, design, data collection tools, sampling techniques among others. Analyses of data collected are presented in chapter five. Chapter six covers the discussions University of Ghana http://ugspace.ug.edu.gh 10 of results. The final chapter, seven includes the summary, recommendations, conclusions and gaps that need to be considered by future research. University of Ghana http://ugspace.ug.edu.gh 11 CHAPTER TWO REVIEW OF LITERATURE 2.1 Introduction This chapter examines the theoretical and empirical literature on the implementation challenges of decentralization at the hospital level. The main focus of this chapter is on the concept of Decentralization, and the evolution of decentralization in the context of Ghana. The chapter also reviews the decentralization of the Ghana Health System and explores the implementation challenges of decentralization at the hospital level. The empirical and theoretical literature explains the links between decentralization and human resource and capacity issues, planning and finance, inter-sectorial collaboration, communication, contextual factors, and community participation. 2.2The Concept of Decentralization The term decentralization according to the literature have different definitions therefore no single definition that has been fully endorsed by scholars as the concept of decentralization tends to be elusive and its explanation and application differ from one set- up to the other (Akonnor et al. 2009; Schneider, 2003; United Nations Development Program, 1999). Several conceptual approaches have been applied to the concept of decentralization involving, the public administration approach (Rondinelliand Cheema 1983), the local fiscal choice approach (Musgrave and Musgrave, 1989), the social capital approach (Putnam 1993), and the principal agent approach (Pratt and Zeckhauser 1991; Griffith 1966). Every approach contributes different elements to our understanding of the decentralization processes. For instance, the social capital approach suggests that localities with a long tradition of community organization, civic networks and solid local institutions will be more likely to be successful in a decentralization process. University of Ghana http://ugspace.ug.edu.gh 12 The principal agent approach allows examining the relationships between the center (Ministry of Health) and local governments. This explains how the former can influence the behavior of the later. The Public Administration approach provides an important and commonly used typology of modalities of decentralization based on the level and type of institutional responsibilities that are transferred to the local level. Some approaches look at the Decentralization in three angles; I. Delegation: shifts responsibility and authority to semi-autonomous agencies (e.g., a separate regulatory commission or an accreditation commission). II. Devolution: shifts responsibility and authority from the central offices of the Ministry of Health to separate administrative structures still within the public administration (e.g., local governments of provinces, states, municipalities). III. Privatization: transfers operational responsibilities and in some cases ownership to private providers. Other approaches also look at decentralization from three perspectives which include Political, Administrative, and Fiscal. Political decentralization: Political decentralization involves establishing local government structures and community participation mechanisms. From this dimension, decentralization should provide increased political accountability, transparency and representation (Pallai 2001). Administrative decentralization is usually defined in terms of the administrative structures and systems needed at the different levels of government, and where responsibilities should be vested. It implies the (re)organization and integration of administrative bodies at the local level to carry on the decentralized functions, and the responsibility of carrying out human resource functions. It thus relates to the issues of administrative capacity and accountability. Authority over staff and its management is often a source of conflict and University of Ghana http://ugspace.ug.edu.gh 13 misalignment in Decentralization processes, since execution of responsibilities may be decentralized while staff at local level may still be appointed centrally. Fiscal decentralization is the assignment of responsibility for mobilizing, managing and allocating funds to and within sub-national governments. It focuses on the main issues of who can raise revenues (fiscal autonomy) and who can spend them (financial autonomy). It relates to the issues of intergovernmental transfers, revenue mobilization at the local level, the budgeting process across government levels and fiscal/monitoring by the central government among others (Farrant and Clarke n.d.). The fiscal aspects of decentralization are quite important and tend to affect the accountability mechanisms of local governments and other dimensions of the decentralization process. In the light of this, several definitions and explanations have been put forward by different writers and scholars, Rondinelli (1981) defines decentralization as involving the transfer of responsibility for planning, management, and resource-raising and allocation from the central government to several bodies including (a) field units of central government ministries or agencies (b) subordinate units or levels of government (c) semi-autonomous public authorities or corporations; (d) area-wide regional or functional authorities; or (e) non-governmental organizations or private voluntary organizations. Ayee (2000) also defines it as ―the transfer of power and authority from the central government to sub- national units, either by political, administrative, economic and fiscal means‖. Thus decentralization generally denotes the transfer of power and responsibilities from the central government to actors at lower political or administrative tiers. As a political and administrative procedure, decentralization involves transferring decision-making capacity, University of Ghana http://ugspace.ug.edu.gh 14 resources and competencies to lower levels of the governmental ladder (Rondinelli, 1981; Ayee, 2000; Hutchinson & La Fond, 2004). Decentralization, sometimes used interchangeably with decentralized governance is viewed as a restructuring enterprise that puts in place a system of shared responsibility between the central government and lower tiers of governance based on the principle of subsidiarity. Thus, the mention of decentralization connotes a consideration of the relationships between regional, district, community, provincial and other lower tiers of government on one hand and the central government on the other hand. According to Work (2002), this relationship existing between the central government and the lower levels may be public, private or civic. Ayee (2000) cautions that decentralization should not be seen as a substitute to centralization since both are needed for the administration of the state. A desired goal or aim must be analyzed and pursued in the light of the complementary functions of both national and sub-national actors since decentralization involves the responsibilities and relationships of all societal players including government, private and civil society (United Nations Development Program, 1999; 1998). Consequently, the United Nations Development Program (UNDP) prefers the term ―decentralized governance‖ to decentralization (UNDP, 1999; 1998). Although there are different explanations to decentralization, there is a general consensus that transferring power and resources to the central government is not decentralization but decentralization connotes a shift of power and resources away from the central bureaucracy (Schneider, 2003). University of Ghana http://ugspace.ug.edu.gh 15 2.3 The historical review of current decentralization policies 2.3.1 Developed economies In developed countries, debates over either to centralize or decentralize have occurred in different environment or in different situations (Smith, 1985). Local government has historically been strong in many developed countries; indeed, central government powers have often been developed and strengthened somewhat later than those of local government. Many countries have therefore inherited local government structures that provide a wide range of services, often financed by local funds. However, central government has tended to place increasing restrictions on local government. A common theme in the expansion of powers of central government has been the need to promote greater equality of public services throughout the country by using central government policies, regulations, and specific and general grants to reallocate resources geographically. Though decentralization has remained the continuing cry in most of these countries, it is often raised in a background of strong factors promoting centralization. Recently, faced with economic downturn and eager to control public expenditure, some central governments have tried to limit local discretion further as in UK and Sweden (Greenwood, 1979). For instance, in the report of a committee of Enquiry into local government finance in 1976 in UK (Larmour & Qalo, 1985), there was a comment that ―what has been clearly visible over recent years is a growing propensity of government to determine, in increasing detail, the pace and direction in which local services should be developed, the resources which should be devoted to them and priorities between them. This has proceeded to the point where local authorities have been called the agents of central government with additional role of statutory pressure group (Eliott, 1981). Few countries were attempting to counter this trend, though many have minority political parties in favour of strengthening local democracy. University of Ghana http://ugspace.ug.edu.gh 16 2.3.2 Developing economies Decentralization trends in developing countries point to two major phases of interest in decentralization (Smith, 1985). In the 1950s and early 1960s, decentralization of local government which was promoted by the colonial administrators was seen as a vital element in the structure of a democratic state. It also served as means of establishing local responsibility for the provision of goods and services. The structures proposed and set up were usually based on models of British or French local government, though limited in their powers and functions. Independence, however, brought concerns of national unity to the fore and for a while decentralization ceased to be a major theme. In 1970s and 1980s, interest in decentralization re-emerged for different reasons. In some countries, particularly in Africa, governments felt sufficiently secured to contemplate relinquishing some of their tight control on power and decision making to local organizations. This also became more possible as corps of skilled administrators joined the administration process. In contrast to the experiences in developed countries, decentralization has been pushed by the center rather than the periphery. However, in some countries especially in the pacific area, like Australia, Japan and China, decentralization has occurred in response to pressure from local or regional groups for increased local authority (Larmour, & Qalo, 1985). The objectives of decentralization happen to be diverse. On the philosophical and ideological level, decentralization has been seen as an important political ideal, providing the means for community participation and local self-reliance, and ensuring the accountability of government officials to the population. On a pragmatic level, decentralization has been seen as a way of overcoming institutional, physical and administrative constraints on development. For instance, increased local control can result in a better response to local needs, improved management supplies and logistics and University of Ghana http://ugspace.ug.edu.gh 17 greater motivation among local officers, thus speeding up the implementation of development projects. It has also been seen as a way of transferring responsibilities for development from the center to the periphery, and in consequence a way of spreading the blame for failure to meet rural needs (Cochrane, 1983). 2.4Theoretical Framework (The policy process) (Source; Anderson, 2006:87, 133). The study employed the J.E Anderson‘s (2006) five staged policy implementation to explain and explore the implementation challenges of decentralization reform at the hospital level. Agenda setting, formulation and adoption stages are adopted to help explain implementation of the policy issues. Implementation stage remains the focus of the study. However, the study did not cover evaluation process as described in the policy cycle. The study will give a brief notes on the agenda setting, formulation, and adoption. In-depth discussion will be done at the implementation stage which is the focus of the study. The researcher adopted this policy because, in implementing the decentralization policy at the hospital level, all the processes involved in Anderson policy process is necessary for successful implementation at the hospital as explained in many policy research literature. To ensure successful implementation of policies, it is vital to address issues concerning organizational, professional and social contexts. Well intentioned political aims are not enough to change practice, and once there is a barrier in any of these contexts, the policy Agenda setting Formulation Adoption Implementa tion Evaluation University of Ghana http://ugspace.ug.edu.gh 18 stands a chance of failure in achieving its objectives (Watt et al., 2005). There are numerous organizational issues which need to be solved for smooth implementation of a policy. Some of these issues include setting up of proper structures, planning and coordination capacity within the ministry of health, clear roles and responsibilities among the ministry of health and its sub-units, intersectoral collaboration, transfer of resources and responsibilities to the decentralized level for delivering healthcare (Paalman, 2005). There is the need for restructuring of the system to support reform initiatives in general and decentralization in particular in terms of decision making, planning, financing, organization and management. A conceptual frame work is then derived from the implementation stage as to what should go into the implementation process of the decentralization policy at the hospital level. The researcher operationalized some key measures that should take place before the implementation of the decentralization policy at the hospital level. The reform measures which include legal frame work, organizational structure, human resource and capacity issues, information and communication, participation, planning, and financing were carefully selected based on the literature, and their expected outcome. It is expected that when all these measures are successfully put in place, policy implementation will bring about decentralized management of health facilities and services, improved authority and leadership, increased participation and coordination, effective management and capacity of personnel at both national and local levels, including efficiency in health service delivery. The diagram below is a pictorial representation of the conceptual frame work adopted by the researcher. University of Ghana http://ugspace.ug.edu.gh 19 Y= a +b +c +d +e +f + se 2.4.1Conceptual frame work Expected outcome Independent variables Figure 2: Conceptual framework showing dependent and independent variables. Y = Implementation (dependent variable) a, b, c, d, e, f = reform measures for successful implementation ( independent variables) Se= error term of the equation Expected outcome= Y predictor outcomes. Legal frame work (a) Organizational structure (b) Human resource (c) Information& communication (d) Participation (e) Planning & financing (f) Decentralized management of health facilities Improved authority & leadership Increased participation & coordination Efficient management & capacity of personnel at national and local levels Improved health service delivery Implementation(Y) (Dependent variables) University of Ghana http://ugspace.ug.edu.gh 20 NB: The arrow from independent variables to dependent variables (implementation) shows direct link of the factors to policy implementation. Arrow from dependent variable (implementation) to expected output shows the expected outcome of successful implementation. Again, blue arrow from independent variables to expected outcomes shows direct relationship between the two. (Source: Modified from Dhakal, 2009:15) 2.4.2Operational definition of the variables Health sector decentralization policy: It refers to the transfer of authority and responsibility from the central level to the district health offices and local health facilities for health services planning and implementation. Policy formulation: In this study, it refers to how the decentralization policy was designed, the nature of interaction that occurred with the policy community and what is entailed in the policy. Policy adoption: It refers to the decision making process of the policy makers. Policy implementation: Concerns with the process of translating decentralization policies and plans into reality. Decentralized management of health services: This refers to the management of health facilities and services by health policy implementers at the hospital level. Organizational structure: This refers to the institutional arrangement at the local level of the Ministry of Health with redefined roles and responsibilities of concerned authorities to implement the decentralization policy at the hospital level. University of Ghana http://ugspace.ug.edu.gh 21 Implementation challenges: Implementation challenges in this study refer to the plausible difficulties and obstacles serving as hindrance for successful achievement of the policy objectives. Capacity: It refers to the competence or skills of the stakeholders in formulating, adopting and implementing the decentralization policy. Improved authority and leadership: it refers to in this study as the power assigned to different levels of the health system to put the policy into action. Decentralized planning: It refers to bottom-up planning process envisaged in the decentralization policy where relevant stakeholders are involved in planning. Participation: It refers to the active involvement and contribution in both kind and cash by health management team, local government bodies, service users, providers and civil society to translate policy objectives into reality. Coordination: It refers to the shared understanding, sharing of experience on programming and planning among government agencies at both national and local levels to maximize the output of the program. Improved service delivery: It refers to the increased access and utilization of health services. Socio-political environment: It refers to the social, political and economic environment caused by variation in geography and development priority in the case districts. 2.4.3Decentralization implementation process; an empirical review Policy does not end after passage of the law. The next important stage is the implementation which is very crucial and critical to the policy implementation process. Policy implementation is the process by which the policy is put to practice by the public University of Ghana http://ugspace.ug.edu.gh 22 and the private individuals which is key in any development strategy. Every policy implementation must pay critical attention to some measures before it can successfully be implemented. Dye expressed that policy implementation should involve a clear strategy and planned activities and some of the activities he mentioned include creation of a new organizations, or the assignment of new responsibilities to existing organizations, development of rules and regulations and bureaucratic discretion (Dye,2004). Implementation is often perceived as managerial or administrative affairs (Walt & Gilson, 1994). To determine or assess the implementation status of every policy is determined by the way its objectives are achieved, and whether a policy will be successful or otherwise depends on the implementation process. However, evidence suggests that outcomes of policies are frequently different from policy intention (Grindle & Thomas, 1991). This situation is more frequent in developing nations than the advanced countries. Leighton (1996) opined that most of the obstacles to implementing health sector reforms in Africa were as results of conflicting policy goals, political instability, weak institutional capacity, poor economic conditions, incomplete sector development and information constraints. In view of the above findings, it is obvious that decentralization policy is not just a verbal expression by policy makers but rather having a holistic view of all the necessary arrangements needed to take place before the reform is put to practice. University of Ghana http://ugspace.ug.edu.gh 23 2.4.4 Policy process An outcome of a policy does not always follow a rational process but concerns itself with the contact with actors who are influenced by social, political, economic, and historical context in which the policy is shaped and implemented. It therefore involves the combination of concepts and tools to understand its process (Sutton, 1999; Walt, 1996). Authors have different ways of defining a policy, more importantly the public policy; however, the general purpose and intention for all policies are almost the same. A policy is relatively stable, purposive course of action followed by an actor or set of actors in dealing with a problem or matters of concern (Anderson, 2006). According to Dye, a public policy is what the government decides to do or not to do. Policy may take the form of a declaration of goods, course of action, general action and authoritative decision (Sapru, 2004). Policy is also defined from the perspective of its content (Hammer and Berman, 1995) to a broad course of action (Baker, 1996). A whole lot of activities are considered in the process of policy formulation that occurs in the area of political system, and implication of this is that policy occurs in identifiable stages and that each stage can be viewed separately. Anderson has identified five stages of the policy process model. These stages include; i. Policy Agenda: This is the problem that receives the attention of public officials. The rational is getting the government to consider action on the problem. ii. Policy Formulation: It is concerned with the development of the acceptable action for public issues. This is where there is a proposal of what is to be done about the problem. University of Ghana http://ugspace.ug.edu.gh 24 iii. Policy adoption: Development of policy proposal for legitimization. It involves getting the government to accept solution to the problem. iv. Policy Implementation: Refers to the application of the policy by the government. This is where the policy of the government is applied to the problem. v. Policy Evaluation: The effort by the government to determine policy effectiveness. The government tries to find out whether the policy worked or otherwise. Dye (2004) also identified six stages of policy process model which are: i. Problem identification ii. Agenda setting iii. Policy Formulation iv. Policy legislation v. Implementation vi. Evaluation Both Dye and Anderson believe that, the policy adoption stage involves decision making process which includes bargaining, competition, and persuasion and comprise among different groups, throughout the policy making process. In such instances, the decision taken by the policy formulator tends to focus on the means rather than the end. Findings according to Walt (1996) indicate that there are technical factors including contents that are essential in understanding the policy process when judging the policy outcomes. A critical rationale for policy change could be as results of changes that have occurred earlier in the conditions set earlier in those policies. A new policy may come from an old or existing policy or overlaps with an ongoing programs. Policy change can University of Ghana http://ugspace.ug.edu.gh 25 take place within the context of policy succession and in domain between innovation and maintenance and maintenance and policy termination (Hogwood& Peters, 1983). According to, it‘s important to understand the policy analysis aspects from the perspectives of what policies the government pursues, why governments pursue such policies, and what the consequences of such policies are. Policy analysis is concerned with who gets what in politics and more importantly why and what difference does it make (Dye, 2004). Various literatures have defined and explained different analytical models in this regard (Dye, 2004; Sapru, 2004; Anderson, 2006). Dye outline eight different types of such analytical models used in political science which are institutional model, process model, incremental model ,game theory model, elite model, public choice theory model, rational model, and group model ( Dye,2004). David Easton defines policy as the authoritative allocation of values for the whole society. He is of the view that political system contains all the institutions and the process that are involved in such allocation of values, and so policy has to larger extent been regarded as a‖ black box‖ of policy making which include all the institutions of government (Easton, 1965) that converts demands into policies but whose structure is seen to be unknown and inaccessible to observation (Sapru, 2004). In the policy process, the surrounding environment or the context in which it occurs is equally worth consideration. The environment widely considers factors relating to geographical characteristics such as climate, natural resources, and topography Demographic variables such as population size, age distribution, racial composition and spatial location, political culture, social structure, the class system, and the economic system (Anderson, 2006). Policy making has been identified as one of the most complex process as it includes various stages, approaches and cycles as opined by Anderson (2006) University of Ghana http://ugspace.ug.edu.gh 26 2.4.5Agenda setting and policy formulation The way a problem actually transforms into an agenda setting is well evaluated and described by various writers (Kingdom, 1995; Sapru, 2004; Anderson, 2006). Once an issue or a problem statement has high legitimacy, high feasibility, and high support, it may become a policy agenda (Kingdom, 1995). Agenda setting could be influenced by mainly three independent streams of activities; problem stream ,policy stream politics streams which open the policy‘ window‘ permitting some matters to reach government agenda ( Kingdom, 1995). After this, agenda becomes an issue of debate and it then enters to the second phase of policy process which policy formulation. Functional activities within the policy formulation stage and blending into policy adoption include formulation of policy plan, legitimizing the course of action, and budget appropriation which lead to a problem development (Burgess, 2004). Policy formulation is concerned with an intense political process in which different groups of participants bring competing definitions of proposals. 2.4.6 Policy adoption: decision- making process Policy decisions are frequent issues encountered by policy makers and governments. They are made by public officials who give direction and content public policy action. It has to do with an action by a body or officials to accept, alter, or refuse a preferred policy option. Decision making process includes bargaining, competition, persuasion and compromise among different groups throughout the policy making process. In such situation the decision taken by the policy maker tends to focus on the means rather than the end (Dye, 2004; Anderson, 2006). Here, several approaches and forums take place at this stage. University of Ghana http://ugspace.ug.edu.gh 27 Afterwards, the panel gathers additional information and research, and models; invite comment from experts, staff and person(s) involved making the policy proposal, and secure the suitable legal and other advice as necessary to ensure the policy proposal when adopted will be in line with the state law. 2.4.7Policy Implementation (study focus) Policy making goes beyond the passage of the law. Another important stage in the policy cycle is the policy implementation which involves the process of putting the enacted policy into practice by both private and public individuals. Sapru (2004) identified that most studies in the 1960s and 1970s reported that policy designs should pay attention to capacity to implement. According to Dye, policy implementation should involve clear an planned activities, and some of the activities include policy making , for instance, the creation of new organization or the assignment of new responsibilities to existing organizations, development of rules and regulations and bureaucratic discretion(Dye, 2004). The implementation part of the policy process is often perceived as the administrative or managerial affairs (Walt& Gilson, 1994). Mostly the status of implementation is determined by the extent to which objectives are achieved (Sabatier, 1991). The process of implementation is a function of the success of the policy. However there is evidence that, outcomes are different from policy intention (Grindle& Thomas, 19991). Leighton (1996) found that obstacles to the implementation of health sector reforms include conflicting policy goals, political instability, weak institutional capacity, poor economic conditions, incomplete sector development and information constraints make implementation of reforms difficult. University of Ghana http://ugspace.ug.edu.gh 28 Legislative and judicial bodies, interest groups, community groups and political structure can influence public policy implementation (Sapru, 2004). It is obvious that, there is no blue print for effective model of policy implementation. In a nutshell, most of these studies on implementation take the form of either top-down approach or bottom-up approach. The top-down focus on the actions of top level officials whereas bottom up debate that it should rather focus on lower level officials and examine their relations with the ultimate clients as (Walt&Glisson1999; Walt &Gilson, 2006). However, for a policy to be effective, two conditions must be met. The first condition is that, the particular intervention must be able to cause the effect, and second is the policy carried as intended. The first is concerned with policy design and second involves the implementation of the policy. How a policy is implemented in an organization involved is very crucial to the overall achievement of the policy goals. Because national policies are carried out through hierarchy of bureaucratic agencies and coordination points, they are prone to implementation failure (Ratanawijitrasin et al., 2001). It is of importance that, policy evaluation process should take into consideration the policy content and implementation of the policy. Normally policy failure may lie with the policy design phase or the way it‘s implemented. The inability to identify which of these two factors leads to policy performance especially failure makes it difficult to judge if that particular policy is ineffective. For this reason looking into the process of how a policy is carried may generate lessons on policy experience as important as those to be learned from looking at policy outcomes. University of Ghana http://ugspace.ug.edu.gh 29 2.4.8Empirical review of the link between Decentralization and some key reform measures for its successful implementation There are some key links between some important reform measures and decentralization. Failure to consider these reform measures will always complicate the implementation of this policy. Empirical reviews of such links in both developing and developed nations are listed and explained below.  Decentralization and health sector reforms  Decentralization and organizational reforms  Decentralization and health system planning and financing  Decentralization, human resource and capacity issues  Decentralization and community participation  Decentralization, central –local relation and inter- sectorial coordination  Decentralization and service provision, efficiency and equity  Decentralization and communication issues  Decentralization and contextual factors 2.4.9Decentralization and health sector reforms in the world Health sector reforms are occurring frequently in both developed and developing countries and they are strongly influenced by political reform process that has become so common in the world. Decentralization of government system had become the major part of the democratic process in some countries with the objective of strengthening local government, allow greater participation of the local people and improve development across all sectors or sometimes to further their political goals (Brijlal et al., 1998). The motivation for decentralization has varied, for instance in Eastern Europe, and former Soviet union, it was part of the political and economic transformation; in Latin America University of Ghana http://ugspace.ug.edu.gh 30 the purpose was to reinforce the transition to democracy whilst in South Africa, Sri Lanka and Indonesia, it was a response to ethnic or regional conflict; and in Chile, Uganda and Cote d‘Ivoire, it was meant to improve the delivery of basic services (Shah and Thompson 2004). However, according to Gilson and Travis (1997), decentralization of the health sector occurred as a way of unifying and rectifying a fragmented and inequitable health system inherited from the apartheid era. Even when it is not explicit, improving service delivery is an implicit motivation behind most of these decentralization efforts. The reasons are twofold. First, these basic services, such as health, education, water and sanitation, all of which are the responsibility of the state, are systematically failing and especially failing poor people (World Bank 2001). The second reason why improving service delivery is behind most decentralization efforts is that, these services are consumed locally, and for that matter entrusting the management of these services into the care of the local people will better serve their needs than being managed at the central level. Historically, they were also provided locally as in the examples of Norway and Nepal where health systems and schools were run by locally- appointed health commissions until the 1930s; and communities until the 1960s respectively. Yet today the central government in these two countries (as well as most others) assumes responsibility for the delivery of these services. Many governments and their electorates associate the problems of service delivery with the centralization of these services. Further interest in decentralization was enhanced when the World Bank stressed the need for decentralization as a key reform strategy in its World Development Report 1993: Investing in Health (World Bank, 1993), as it is professed to increase the efficiency and quality of government health services (Bossert& Beauvais, 2002), decentralization of University of Ghana http://ugspace.ug.edu.gh 31 health sector is progressively recommended as an essential strategy of Health Sector Reform (HSR). It is argued that increased administrative efficiency is the prime drive for governments to decentralize (Therkildsen 2001:1; Conyers 2000:8 Increased service provision is also believed as one of the benefits of proper decentralization, as centralized government monopoly of service provision is argued to be the source of much inefficiency (Tendler 2000:118). With this argument, it is logical to introduce private firms, Non- governmental organizations (NGOs) and even local governments as providers to increase competition, thereby enhancing efficiency. However, the evidence that decentralization leads to better service provision is slim, and this is partly against the background that, the assumed causal relations are difficult to demonstrate (World Bank, 2000; Ribot 1999). Decentralization with respect to health came to light in the awake of the PHC conference at Alma-Ata in 1978 (Green, 2001). Nevertheless the implementation experiences of reform initiatives across countries were found prominently similar regardless of the socio- economic and epidemiological situations as most of these reform implementation was neither complete nor clear. The rationale behind the international community decision to push for decentralization was on the basis of its high potential in changing the centralized governance system. The advantages of decentralization in theory include; enhancing political accountability to the users of health services, greater innovation in health service delivery and local adaptation of services, improved intersectional coordination and ability to focus on developing PHC and speed up implementation of development programs (Bossert & Beauvais, 2002; Gilson& Mills, 1995). However, there are disadvantages for the above mentioned advantages (Mills, 1994). For instance local levels of government may not support University of Ghana http://ugspace.ug.edu.gh 32 national priorities, thus serving as a drawback for implementing the national policies and delivery of public goods. Decentralization is concerned mainly with intensifying health system performance or increasing health systems‘ ability to deliver quality and equitable health services that are efficient and responsive to the needs of the local people. It has been debated that, one of the best means of bridging the gap between those who have and those without is by means of decentralizing authority, and resources with effective and strict monitoring with adequate capacity measures put in place to enable local bodies to carry out people centered development programs (WHO, 2002). Conversely, it is difficult to attribute changes in health system performance to decentralization because there are other factors such as simultaneous financing reforms, political context or economic context, which may also control health system performance independent of decentralization ( Gilson & Mills, 1995). 2.7.1Organizational reforms and implementation of Decentralization in the health sector Health sector reform implementation, must take into consideration the existing organizational structures and culture of the Ministry of Health in order to enhance the chances of successful implementation (Sakyi, 2007). For any effective implementation of health sector reform policy, reorganization and restructuring of the Ministry of Health is a crucial step (Dhakal, 2009), and such processes should be based on the functional analysis of the structure to achieve the desired objectives. However, this is different in most developing countries where restructuring of the Ministry is more often than not driven by political motives rather than functional analysis of the health system. This is also partially true for the reason that most of the reorganization is influenced by the support from donors as opined by Jeppsson et al., (2003).The implementation of decentralization policies University of Ghana http://ugspace.ug.edu.gh 33 requires detailed planning and co-ordination, which in turn requires the establishment of organizational structures and procedures designed specifically to facilitate the implementation process (Matovu, 2008). Failure to do so tends to result in very slow implementation progress, disorganized decentralization, and/or the decentralization of functions without a complementary reorganization and contraction of central government activities. In all reforms, agencies will face the challenge of restructuring their resources and organizations to meet the needs of policy implementation. As new tasks are developed and procedures are created, responsibilities will shift, some divisions or departments will gain importance while others may even be abolished, and new patterns of internal resource allocation will emerge in accordance with the demands of the new policies. Restructuring also may be necessary across agencies. Some tasks may be reassigned or reallocated from one agency to another, resources will be redistributed in accordance with the new policies, some agencies will gain in importance or stature while others decline, and a greater level of coordination may be called for to ensure successful operation. The colonial structure in most developing countries encourages a centralized hierarchical administration culture with little or no involvement of district health managers and other stakeholders in planning and implementation of health services (Annan, 1997; MOH, 1996). However, these policies are often implemented in most developing countries without considering all these factors which mostly affects its implementation. According to Annan (1997), fragmented structure of health system is responsible for difficulties encountered in sectoral coordination and implementation program at national, regional and district levels; and there is a direct linkage between organizational structure and culture of the health system as indicated by Agyapong (1998). University of Ghana http://ugspace.ug.edu.gh 34 Agyapong opined that, the issue of organizational structure is closely linked to the organizational culture of the health system especially with reference to staff orientation, and this linkage was found to be constraining the practice of health decentralization at district level. As explained by Dhakal, agenda of restructuring of a system may not always be a suitable suggestion, unless it benefits the interest of an individual or the group .For instance, in the reorganization exercise in Uganda, the role of the Ministry of Health was reduced to policy formulation, planning, supervision, setting standards and inspection, management of national programs, ensuring appropriate quality, provide specialized logistical support that were not available in the district markets and even after decentralization, Ministry of health was still working the way they had in the past, that is managing vertically organized programs from the center down to local level (Jeppsson et al, 2003). Restructuring is mostly overlooked in the decentralization process, but it‘s vital to restructure and redefine the roles expected of the national and sub-national levels to prevent confusion about their respective new roles. In Nepal and Kenya, the central level was not restructured in line with decentralization reform, and roles of managers were not re-defined. This brought about the problem of vertical program, where managers dominated the central level decisions because their roles were not redefined to reflect the policy objective (Dhakal & Singh, 2006; ADB, 1999). Decentralized health systems have more administrative levels and governance structures than centralized systems and this multiple lines of responsibility can be a source of confusion for actors at all levels, with none having a clear understanding of their respective new roles ( Brijlal et al., 1998), as it was seen in the case of Zambia where the National Ministry of Health delegated all responsibilities for direct service delivery to the Central Board of Health which then contracted district health management team and hospital boards to provide services at the district and hospital levels in that order. University of Ghana http://ugspace.ug.edu.gh 35 This situation brought about confusion of roles and responsibilities and delay in timely decision making according to Bossert et al. (2003). 2.4.10Decentralization and Health System Planning and Financing Districts vary widely according to the specific needs of their population, and even more so in terms of existing interventions and available resources. Strategies, therefore, must be district-specific, not only because health needs vary, but also because people's perceptions and capacities to intervene and implement programs vary. In centrally designed plans, there is little scope for such adaptation and contextualization, hence decentralized planning becomes crucial (Gopal&, Mondal, 2007). There is also the need to evolve from a more "command and control" orientation of public health officials towards the community, to an attitude of participation, openness and accountability, recognizing the rights of the poor and the vulnerable. Decentralized planning according to the World Bank (1993) is seen as a way of coping with the changing patterns of diseases and regional disparities in health. Conyers (2000:7) provides four broad categories to outline decentralization objectives: local empowerment, administrative efficiency and effectiveness, national cohesion, and central control. However, Oyugi (2000) argues that the merits mentioned above are claims and expectations and not hard facts. He continued to posit that these outcomes depend on conditions relating to real power sharing and meaningful participation and without these, the effects may encounter the objectives. Segall (1983) is of the view that, creating conditions for the national health programming to be carried out at local level by the people should be in close contact with the local needs and conditions. Approach of this nature tries to emphasize local health programming than on detailed planning of large national health programs; and this approach is capable of University of Ghana http://ugspace.ug.edu.gh 36 reinforcing the values of planning system such as equity, need to involve communities in decision making processes; recognition of multi sectoral nature of factors affecting health and need to ensure that the interventions that are adopted are appropriate in drawing attention to promotional activities. Oyugi also asserts that when interventions are meaningful to local population and perceive their participation to be beneficial in a development situation, they are willing to contribute. A study conducted in India revealed that decentralized health planning identified the target based approach as one of the problems for implementation. The central government always imposes districts with targets without taking into account local factors that can constrain the achievements of such targets. Despite the fact that budget ceiling was provided, the large portion of the fund was used for salaries, whilst very little was spent on district specific programs. Evidence revealed that districts which put much of their efforts and focused on addressing the problems of implementation of national programs were very successful in meeting the objectives set for decentralization. In most cases, the success of decentralized planning is a function of the extent to which administrative and implementation support provided by the central ministry. Getting administrative support is far better if there is a clear understanding at the level of program planners about what can be and cannot be done at various levels and if there is confidence that plans have been developed based on rational analysis of local needs. There is no program that can be planned without understanding the situation of the district in micro level planning, and this further suggest the view that planning without identifying the viable place is likely to face challenges as opined by Collins et al, (2003).The central problem of planning is integrating local needs and aspirations into the national planning process without losing the broader development objectives that planning can serve or University of Ghana http://ugspace.ug.edu.gh 37 undermining the inclusive/participatory processes that decentralization is supposed to embody. Strategies to mobilize more resources to finance decentralization process are very vital for the success of the reform. In such situation, most hospitals will be independent and autonomous in the areas of finance which serves as a major barrier to many policy reforms. This also means that financial resource will be channeled to areas of needs to society. Many governments rely heavily on donor funding (Bossert, 1998; Walt, 1996), and this means that government is likely to have little ownership with respect to the nature and course of the reform in their country. This is because most of these funding agencies attach some conditions to the amount donated as to where the money should be invested at the expense of the needs of the local people. The imposition of user‘s fees by local government may prevent the poor people in community from accessing basic health services. User fees that were initiated in most of African countries which excluded vulnerable groups from vital services, and exemption systems proved ineffective (Dhaka, 2009). For instance, in Uganda, user fee was abolished at public facilities in 2001 in order to encourage access to healthcare and this resulted in increased utilization of ambulatory services (Kawonga, 2003). Experience from different countries show that, decentralization reform could have an influence on resource allocation mechanisms. Bossert & Beauvais (2002) examined the decentralization experiences ranging from devolution to delegation in Ghana, Uganda, Zambia and Philippines, and the results revealed that, in all countries, health expenditure increased at the local level and decreased at the central level as a result of decentralization reform. Despite this, higher spending at the local level did not result from any significant increase in revenue generation at the level but rather from increased transfers from the central government. University of Ghana http://ugspace.ug.edu.gh 38 A recent study by Cuttolenc (2012) showed that in Ghana there was a substantial delay in transfer and release of funds, both by government of Ghana and NHIS, which hampered the functioning of local governments and local facilities and programs alike. In addition, fiscal decentralization in Ghana is more apparent than real as over 50 percent of public health expenditure is allocated to the district level, but the larger part of these resources are allocated and controlled by the central government; local authorities ,whether DAs or GHS District Offices and facilities have little real decision power on resource allocation. Evidence is that continued control from the central government over salary and personnel severely limit local fiscal autonomy and hinder cost control efforts (Bossert& Beauvais, 2002). However, it is crucial to constantly monitor the local level by the central level with regards to how local financial decision are made in a decentralized system, because health may be given a low priority in relation to other sectors, and may receive a smaller financial share of limited local council grants. 2.4.11Decentralisation and human resources management and capacity issues Human resources are the most important aspect of health care system in converting available pharmaceuticals, medical technology, and preventive health information into a better use for every nation. But the absence of clear human resource policies has been one major problem encountered in the health sector reforms. Decentralization which is a popular feature of most public sector reform programs is viewed as an opportunity to improve Human resource management, but the current evidence according to Wang et al. (2002) is open to doubt. Effective delivery of health services is seriously affected by human resources constraints (Narasim-ham etal. 2004). The major concerns has been inappropriate numbers and types of staff and the way they are been distributed, as well as the performance of the staff. Dhakal ( 2009) also identified University of Ghana http://ugspace.ug.edu.gh 39 similar problems as he opined that, quantitative mismatch, Qualitative disparity, unequal distribution and lack of coordination between HRM functions and health policy objectives have been the key issues, and that decentralization reform cannot succeed if these issues are not addressed in a timely manner. Appropriate HRM policies and practices can improve human resource (HR) outcomes and consequently lead to the effectiveness of the workforce, which in the long run bring about improved organizational performance. HRM policies and practices that are aligned with appropriate health service objectives, improvement in HRM leads to improved health outcomes. Increasing number of low-skilled health workers and the fall of demand at rural health facilities in part due to the loss of their most experienced personnel was found to be the cause of the falling productivity of health workers in china (Martineau et al., 2004 Gong et al. 1997). As public sector organizations become exposed to new public management reforms, human resource activities will be required to play a greater role. Sakyi was of the view that understanding the importance of human resource management practices is therefore important for reform implementation. Four main factors were obtained as contributing to HR problems in Ghana, particularly in the Ghana health sector (Agyepong et al., 2004; Dovlo, 2003). The recruitment and selection of employees is one such challenge because of the influence of social and political connections and patronage in Ghanaian public organizations (Price, 1975). Another challenge is the wastage of public health sector workforce through misapplied skills, absenteeism, poor support and lack of supervision (Dovlo, 2005b). Furthermore, the problem of ghost workers on payroll is affecting human resource development policy University of Ghana http://ugspace.ug.edu.gh 40 making and implementation in many ways as expressed by Ackon, (1994). This aspect of the problem is not thoroughly explored. Therefore, it is important for reformers and donors to also consider seriously how to deal with these problems in the public health sector. Another related issue involves the development of human capital among Ghanaian health workers (including other public sector employees), most of these employees tend to work within the limits of their job descriptions and scarcely try new work activities. Therefore, it is important for public management reform programs to address the issue of on the job training and detailed description of tasks that public Officials are expected to perform (MOH, 2001). The issue of HR issues is very important as the quality of health services, their efficacy and accessibility depend mainly on the performance of those who deliver them (WHO, 2000). How the workforce performs is determined by the policies that define the number of staff, their qualifications, deployment and their working conditions. In view of this, it‘s necessary to deal with some of the necessary processes and dimensions that influence the performance of workers. Experience from other countries show that, there is the likelihood of serious implication of decentralization on workforce (Collinet al., 2003; Aitken, 2004). Three issues whose resolution calls for active collaboration between international and national authorities, stand out clearly among the numerous and must be given the necessary consideration. Urgent need to define the essential human resource policy, planning and management skills that national human resource managers working in decentralized units must have is the first one, decentralization purported negative impact on staffing equity due to lack of promotion and other incentive mechanisms at different level is the next, and the third is the continuous poor motivation and performance in decentralized system (Dhakal, 2009). University of Ghana http://ugspace.ug.edu.gh 41 Capacity is topical in almost all studies on new public management reform implementation, including decentralization of health services. However, this implementation has tended to focus on health sector reform, changes in the health financing system, and human resource development (Green ,& Collins, 2003 ), with less attention paid to the institutional development of health systems undergoing decentralization (Green & Collins, 2003; Gilson & Mills, 1995). Without institutional capacity, health facilities do not function well by themselves, especially at the district level, where they provide primary health care to communities (Oyoya & Rifkin, 2003). Studies indicate that a critical factor influencing the practice of decentralization is the technical and managerial as well as logistical feasibility of its implementation (Sakyi, 2007; Batley and Larbi, 2004; Agyepong, 1998; Larbi, 1998). It is important to note that, capacity can be measured in quantitative terms. But, in the context of local governance and service delivery, it is more often seen from the perspective of how decision-making processes are being organized, what quality of services is being provided, and what are the results and outcomes that are being achieved. This implies that such 'qualitative' capacities require a lot of development and contextualization for the sake of relevance, quality and acceptance. In most developing countries, the required capacity for effective decentralization to take place is lacking, and most often frequently asked questions by reform analysts is whether the required managerial capacity and capabilities are in existence in these countries. For instance, Larbi (1998) found that district health management teams failed to implement several components of decentralized management because the technical and managerial staffs are incapacitated. Studies emphasized the importance of human capacity to effective implementation (Grindle, 1997; Turner and Hulme, 1997), as health service is a labour University of Ghana http://ugspace.ug.edu.gh 42 intensive industry and human resource remains a critical component for successful service delivery (Dovlo and Nyonator, 2004; Dovlo, 2003). Capacity problems can seriously affect decentralization reform initiative, and an example was found in Kenya where the shift in the responsibility of drug purchase to the districts was reversed after one year due to the poor performance of the district resulting from poor capacity. In the same way, short of capacity was found to be major hindrance to AIDS control initiatives, and as a result The Ministry of Health was compelled to maintain a centralized system (Kawonga, 2003). Limitation in capacity can serve as a hindrance to the implementation of policies, and this can range from human resource, inappropriate skill training and lack of adequate preparedness for their new functions. Gottret & Schieber (2006) expressed that, to successfully implement decentralization policy, the leadership capacity of new managers must be strengthened, as must be the institutional capacity of new systems at the local level and capacity building must occur before and after decentralization. However, despite these positive correlation between policy implementation and capacity building, a review of 140 countries on the impact of decentralization on immunization services, showed no evidence that institutional capacity and policy implementation correlated (Gottret &Schieber ,2006). 2.4.12Decentralisation and community participation Community participation is often described as the major benefit of decentralization and mostly assumed to be the regular benefit arising from decentralization processes. It is a luxury for the poor people who are faced with problems of illiteracy, poor health, hunger, economic emaciation and poor infrastructure, among others. According to Gaventa (2001), poor people are excluded from participation in governance and state institutions are often neither responsive, nor accountable to the poor. Francis and James (2003) carried out a University of Ghana http://ugspace.ug.edu.gh 43 study in Uganda, which revealed that ―decentralization structures and processes did not constitute a genuinely participatory system of local governance‖. Community participation was initially used by the WHO as a tactic in promoting health. Community participation is often seen as a crucial aspect of health programs in developing countries (Bracht and Tsouros 1990, Rifkin 1991). As spelt out in article 4 of the Alma-Ata declaration ―the people have the right and duty to participate individually and collectively in planning and in the implementation of their health care‖ (Alma-Ata Declaration 1978). Eckerman et al. (2005) is of the view that community participation is not just a component of primary health care but a life blood philosophy, and referred to as not just the involvement of the community but an active involvement of ownership and control. Community participation is thus defined as a social process that allows people of a specific geographic location to be involved in identifying their needs (Rifkin 1986; Anderson and McFarlane 2008). The United Nation Economic and Social Council Report (UNESC 1956), defines community participation as the process by which the efforts of the people are united with those of the government authorities in order to improve the economic , social and cultural conditions of communities, with the view that such communities will be integrated thereby contributing fully to national progress. There are numerous reasons why community members should partake in the provision of health care, and one such reasons is ensuring accountability of health services to users who are regarded voters, tax payers and consumers ((Tritter and McCallum 2006). The active participation of communities and community based originations in planning and management of local health care activities is essential and can make important contributions to service quality and efficiency. Consultation with and the participation of implementers is a function of successful implementation of policies and it‘s important for two main reasons; the first is to improve University of Ghana http://ugspace.ug.edu.gh 44 the quality of change by including the views and experiences of health sector staff and users; and secondly, to develop a sense of ownership among the same group (Bossert, 2004b), and therefore implementers must be made part of the policy design process (The Kathmandu Post, 2006a). The most direct positive effect of community participation with regard to funding was found when the users of healthcare in Mexico were interviewed and lend their support for the operation of the health care unit, be it through fund raising and the awarding of economic resources to improve or expand the units, or through voluntary community at the units (Armando & Orozco, 2006). With regard to community participation, research conducted by the Maryland university revealed that , though channels for community participation was in existence but they were often ineffective, mainly because of inadequate flow of information between local government and civil society and the limited usefulness of civic structures/ local health committees ( World Bank, 2001). In the same way, despite the fact that many communities increased their participation in the health sector through local health committees following decentralization, lack of training and guidelines undermined the effectiveness of these governance structures (Hutchinson et al., 2002). 2.4.13Decentralisation, center- local relation and Inter-sectorial coordination The establishment of mutually beneficial partners between the central and the local government on one hand and non- state actors such as civil service organizations, private sector, and international agencies on one hand has become an innovative way of broadening and strengthening effective and efficient decentralization. As rightly said in 1999 by the UN Secretary General, Kofi Annan: University of Ghana http://ugspace.ug.edu.gh 45 The United Nations once dealt only with governments, but now we know that peace and prosperity cannot be achieved without partners involving governments, international organizations, the business community and civil society. In today‘s world we depend on each other. Increased communication, with other agencies will be necessary to ensure that enabling regulations or sub-policies vital to implementation of the agency‘s tasks are generated. Since successful actions by one agency may depend on the implementation of complementary actions by other agencies, there will be greater need for sharing information and resources, and more concerted coordination. Every successful implementation of decentralization policy calls for local-relation and inter-sectorial coordination (Collins et al., 2003; Green, 2001; Bossert et al., 1990; Mills et. al, 1990). Nevertheless, Gilson & Travis expressed that regardless of the form of decentralization adopted, there was no change in inter-sectorial collaboration. This could stem from the fact that, in some settings, only the health sector was decentralized and this makes it very difficult collaborating with the local level. However, in Uganda where decentralization occurred beyond the health sector, other sectors in the local government failed to recognize health as priority issue (Gilson& Travis, 1997). In most cases, choices that are made at the local levels are mostly influenced by local context, personal characteristics of decision makers and lack of incentives, than by the content of the reform. This was evident in Zambia as district did not sufficiently implement vital health care package because of the cost involved (Kawonga, 2003). 2.4.14Decentralisation and service provision: equity, efficiency and quality Strengthening health system performance to deliver high quality and the provision of equitable health services that are efficient and responsive to the needs of the local people University of Ghana http://ugspace.ug.edu.gh 46 are the target of the decentralization policy (Mills et al., 1990). Measuring health system performance in countries and establishing a causal link between decentralization and health system performance is difficult because objective measures of equity, efficiency and quality of care are unavailable due to lack of monitoring evaluation systems, especially in developing countries (Dhakal, 2009). Then again, it is difficult to boldly attribute changes in health system performance to decentralization because other factors such as simultaneous financial reforms, political context, or economic contexts may also influence health system performance independent of decentralization ( Gilson & Mills, 1995). Decentralization may have to weigh effects on quality of care provided and that is either positive or negative. One is the expansion in the range and the scope of services available resulting from decentralization may indicate better quality of services due to greater availability, and the second effect may be the situation where quality is compromised because of lack of planning and poor implementation. According to Gilson & Mills (1995), despite the expansion in scope of services, technical supervision of health care providers at lower levels of the health system declined hence undermining quality gains. In the case where there are differences in income levels among local areas, decentralization may undermine equity because areas with greater revenue generation ability will have more resources an may perform well than areas with less-resources, especially those with little generation capacity . A research by Bossert et al (2003) concluded that decentralization did not increase inequalities among districts. However, it was not clears as to whether this relative equity was attributable to decentralization as health expenditure data for the year before decentralization period was not available. Increased administrative efficiency is the prime impetus for governments to decentralize (Therkildsen 2001:1; Conyers 2000:8).Governments, donors and the private sector have University of Ghana http://ugspace.ug.edu.gh 47 supported decentralization on the grounds of efficiency for many of the standard public choice arguments: decisions are more relevant to local needs and conditions are more likely to be effective; local co-ordination facilitated and transaction costs are reduced by making decisions locally; decentralized decision making can be quicker and more flexible, therefore more efficient; local knowledge and preferences can be drawn on to make decisions more relevant and effective; local knowledge and labour can facilitate implementation, management and evaluation; and because local actors will benefit from reducing the costs of their efforts, they are likely to spend their resources more efficiently(World Bank 2000:108). Griffin (quoted in de Valk 1990:5) points out that the most recent wave of decentralization ―seems to be based more on the assumption that decentralized planning and participation can achieve effectiveness and efficiency by resolving the implementation problems of rural development planning‖. . In addition, decentralization is seen to increase the relevance and sustainability of development, as well as self-help contributions to development. The debate that decentralization will ensure more efficiency through better matching of supply and demand of local public goods should not hold for less than democratic circumstances that apply in some developing countries (Sewell 1996:147). This point is very important as many of these arguments are predicted on the belief there are mechanisms in place to hold local authorities accountable to local populations. Properly structured decentralization is capable of bringing about equity. Democratic decentralization is based on locally accountable representative bodies with powers over select local resources and decisions, and with local rights and systems of recourse. To establish such forms of local governance requires a shift in most of Africa, particularly in University of Ghana http://ugspace.ug.edu.gh 48 rural areas, away from the highly inequitable, administratively driven management of the local world. Mandani (1996) opined that rural people across Africa are managed as subjects under highly inequitable and even dictatorial circumstances. For instance, in participatory approaches to forestry in Burkina Faso, Gambia, Niger and Senegal involve only economically interested parties. This does not give equal rights to the public in determining the disposition of local forests. More generally, there is little evidence that decentralization is instituting procedures and institutions for representative, accountable and empowered forms of local governance. However, more democratic experiments appear to be unfolding in Ghana, Mali and Uganda (Ahwoi 2000). Experiences indicate that decentralization may either bring about or weaken equity, quality and access to services not because of content reforms but mainly as a result of poor implementation, limited capacity and lack of supportive legislation and policy, and hence certain mechanisms may be important to bring about quality, equity and access to health care services in a decentralized system (Gilson& Mills, 1995). Whiles decentralization is targeted at improving health performance, the ability to implement to achieve access, equity , efficiency and quality rely on functioning health systems ( Hardee & Smith, 2000), which means that transfers of authority to locals should be accompanied by mechanisms to enable effective implementation of the reform and of health services in decentralized context, including training and capacity building of local and central level managers., resource allocation guidelines, and supervision and support structures for health providers and local managers ( Dhakal, 2009). According Hanson et al (2003), distance and utilization are correlated, and it appears to drop off exponentially with distance from provider. This calls for adequate preparation, University of Ghana http://ugspace.ug.edu.gh 49 planning and the cost involved before a particular intervention is taken place. Inability to plan carefully might not be implemented as wished by policy makers. 2.4.15Policy implementation issues and challenges Since 1970s, studies of policy implementation have become abundant, in order to come out with measures spawned by the desire to explain ‗the implementation gap‘ (Dunsire 1978). The issue of policy implementation is of great importance to social scientists but seems not be the same with policy-makers who often equate proposing a policy with its effective disposal. The inability to get things done, to have ideas and decision implemented, is very common in modern organizations, be it public or private. Some researchers see it as a situation that is getting out of hand (Pfeffer 1992). The ability to get policies and decisions implemented well is becoming uncommon especially in most developing countries. Mostly, failures of policies occur as a result of bad execution or bad policy, and reasons that are less explained as the result of policy failure are inadequate information, poor reasoning or unrealistic assumptions. One common thing about policy failure is that, there can be no sharp distinction between formulating a policy and implementing, but yet there is the assumption that such distinction does exist. Studies indicate that it is more likely what happens at the implementation stage will influence the actual policy outcome in a manner that has not been anticipated. On the other hand, probability of a policy success is increased if consideration is given at the policy design stage to potential problems of implementation. This might suggest the need for a policy impact statement or audit of some kind to identify possible implementation problems or barriers to success. ‗Top- down‘ versus ‗bottom-up‘ perspectives on policy and action are at the heart of discourses on policy implementation (Barrett and Fudge 1981). At times policy failure or University of Ghana http://ugspace.ug.edu.gh 50 implementation gap may occur as a result of policy imposed from the center with no consideration as to how the policy might be received at the local level. It is not a case of bottom up approaches to policy and action being preferable to top down, but a balance between the two is necessary. A study conducted in Paraguay revealed that political, legal, financial, institutional, administrative and human resource were the challenges faced in putting reforms into practice (Zanten &Semidei, 1996). A study by Collins et al (2000) identified three major challenges for the success of decentralization in Brazil, which were the policy-making process, equity and the role of the state. Other issues often encountered while designing and implementing health sector decentralization are ;( a) mismatch between authority and responsibility,( b) tensions and conflicts among objectives, c) capacity gaps, d) tension between vertical and horizontal integration e) political and process dimensions ( Brinkerhoff &Leighton, 2002 If these are not balanced, and one is not supported by others, decentralization policy implementation moves toward failure (Adhikari, 2006). The successful implementation of any policy involve not only just a common agenda, and long term joint commitments across different agencies but a degree of transparency about the model, in this case the model of health adopted. The breadth of the inequalities agenda is therefore reflected in a wide range of decision-making processes and complex partnerships spanning different decision-making and performance management systems. If policy is to be successfully implemented, then those who have responsibility for its implementation should be involved in its design. Implementation should involve a process of interaction between organizations whose members may have different values, perspectives and priorities from each other and from those advocating the policy (Hogwood and Gunn 1984). University of Ghana http://ugspace.ug.edu.gh 51 The role of donors has also been found to play an important role in policy implementation. Evidenced gathered from other countries show that implementation could be successful if external donors are involved while developing a country‘s health policy. Discussing the whole strategic plan with donors may result in a better match between donors‘ activity and country‘s need (Walt & Gilson, 1994; ADB, 1999). 2.4.16 Decentralisation and communication issues One of the major conditions for a successful implementation of any public policy is the quantity and quality of communication about the policy. Communication brings about meaning as compared to mere message which in effect enables policy managers and implementers in understanding the goals and what they expect. Communication fulfils a lot of functions in policy making and implementation processes. It provides a means by which people in industry, business, politics and professions act and interact; exchange information and ideas, develop policies, plans and proposals; make decisions and manage people and materials (Sakyi, 2010). There is a link between communication and the implementation of the decentralization policy, and this is because communication and policy implementation are activities that involve interaction and occur in a social setting. Communication among policy managers and frontline workforces (street-level bureaucrats) about the objectives of a policy, specific programs and activities, and the methods of execution does influence the degree of implementation success. Communication of reform goals to implementers and stakeholders help maintain healthy relationship and safeguards inter- organizational harmony needed for effective reform implementation. Similarly, sharing of relevant information with communities and stakeholders is required for building peoples trust in the reform process and enhances inter-organizational collaboration and cooperation needed for reform implementation as expressed in the work University of Ghana http://ugspace.ug.edu.gh 52 of Sakyi, (2008). However, the communication between policy implementers, other stakeholders and community members is very minimal in most developing countries. In Ghana, a study conducted by Sakyi (2010) revealed that staff received little or no information on major health management decisions, and this was very common among junior staff. However, if junior workers are not informed about such decisions, there is the likelihood that policy will fail because they are the front workers in every health institution. Communication among policy managers and frontline workforces about the objectives of a policy, specific programs and activities, and the methods of execution does influence the degree of implementation success. It is when policymakers and implementers understand through effective communication, the need for a particular policy or policy reform that their interest is aroused. Goggin et al. (1990) also argued that the more legitimate and credible the message was in the eyes of implementing agency officials, the more likely implementation was to proceed promptly and effectively. For this reason, decentralization of the health sector should be accompanied by adequate information from the central to the local level and up to date information about the policy among all stake holders at the district level. Effective communication mechanisms such as workshops, seminars, rather than the traditional letter writing and so on is preferable because weak communication and information sharing mechanisms contribute to limited understanding of reforms. 2.4.17Decentralisation and contextual factors Decentralization is frequently implemented as part of a public sector reform and needs to be adapted to the requirements of the health sector. It requires broad consultation and offers a wide range of policy options for its organizational design (Bossert and Beauvais, 2002; Mills et al., 2001; Bosman, 2000; Collins, 1994). Decentralization is also University of Ghana http://ugspace.ug.edu.gh 53 contextually driven, as it can work in one district but fails in another (Atkinson et al., 2000). Districts and Municipalities are different with regards to factors such as wealth, education, natural resources, political culture, social organization, etc. These factors have major effects on the way in which health sector reforms are implemented and how the organizations functions. Unfortunately health sector decentralization has paid scant attention to these contextual factors. If the claims made for decentralized management in the health sector are to be evaluated seriously, it is critical to develop concepts and methods to evaluate not only the formal organization and the outputs of the health system, but also the contextual factors within the Districts where decentralization occurs. Though some of these contextual factors will be very difficult providing answers to them, it is inadequate to leave them under-discussed and under-researched by those in charge of designing and implementing health sector decentralization. Where decentralized management of health care has been evaluated, studies most often compare aspects of health care provision within a country or between districts decentralized. However, the unfortunate case with this kind of comparison is that those districts chosen within a region may have distinctive qualities from the other Districts. These contextual factors among different Districts have great influence on the success of the implementation of health sector decentralization and therefore it is these contextual factors that must be described in evaluating the effectiveness of heath sector reform and assessing their relation in terms quality of care delivered (Atkinson et al, 2000). These contextual factors often influence the success of implementing the health sector Decentralization at the District level; and as results of this, some writers have cautioned policymakers to be aware of these factors in designing decentralization policies. University of Ghana http://ugspace.ug.edu.gh 54 One of the potential risks with decentralization is that it can lead to greater inter- jurisdictional disparities due to the differences in socioeconomic potential and expenditure needs of subnational governments. The argument is that subnational governments with better factor endowments and potential will have a larger revenue base than other poorer subnational governments and therefore will be able to provide more local public services. This may cause disparities in economic opportunities and create gaps in income and public service delivery between various jurisdictions. Prud‘homme (1995) considers decentralization as the pure decentralization as viewed in earlier fiscal federalism literature; that is, local governments fund local public services with their own local revenues without transfers from the central government. However, evidence suggests that intergovernmental transfers, such as equalization grants, can play an important role in overcoming imbalances between revenue capacities and expenditure needs of local governments (Martinez-Vazquez and Searle 2007). Factors such as unequal endowment of local capacity and the politics of local central relations could seriously affect the successful implementation of the decentralization policy at the hospital level. However, evidence suggests that lack of resources is not necessarily the problem, but Douda (2006) using examples from South Africa and Uganda expressed an opposing view that developing political capacity for demanding accountability for existing resources is more important. There are several studies that examine the relationship between decentralization and regional disparity in both developed and developing countries (Ahmad et al., 2008; Gajwani et al., 2006; Kanbur and Zhang 2005; Bardhan and Mookherjee, 2005). These studies mostly focus on developed or developing countries of Asia and Latin America, and evidence appears to be inconclusive. However, in the context of Sub-Saharan countries, these disparities are studied very rarely despite the fact that for many countries in the University of Ghana http://ugspace.ug.edu.gh 55 region such as Ghana, it is a very practical issue as they are pursuing serious decentralization programs. However, most of these studies do not provide any analysis of the disparities between districts. Asante and Akramov (2009) concluded that Districts‘ geographical locations play a major role in shaping disparities in access to local public services between decentralized districts in Ghana, and that ethnic diversity has significant negative impact in determining access to local public services. However, most of these are factors are ignored in policy design process which in the long run affects policy implementation. In most cases socio-cultural environment facilitate or hinder a balanced distribution of professionals which has a great influence on the implementation of health sector decentralization at the hospital level. Community and local resources, conditions and opportunities can either draw or repel health professionals to or from a given area. As expressed by Van Lerberghe et al (2002), access to social, cultural, educational, and professional opportunities increase preference to settle in particular areas. This implies that policy makers must adequately explore these disparities among districts to be able to factor them in the design and implementing decentralization reform at the hospital level. Though decentralization is believed to be one of the tools to empower the local community in development projects targeted at reducing poverty and improving the distribution of socio-economic benefits in an equal way, cultural factors pertaining in a region or district can constrain or speed up these benefits of decentralization. For instance, exploring how cultural factors influence the implementation of the decentralization reform in the context of Cambodia, Rotha and Vannarith (n.d) realized that the introduction and implementation was faced with several cultural factors embedded in the Cambodian society. These factors included patron- client relationships, power distance or unequal power distribution, lack of trust between the local residents and local leaders, collectivism University of Ghana http://ugspace.ug.edu.gh 56 and gender discrimination. For these reasons, it is often of great importance to consider all these contextual factors in designing and implementing the decentralization reform at the hospital level. University of Ghana http://ugspace.ug.edu.gh 57 CHAPTER THREE DECENTRALISATION IN GHANA Ghana‘s current legal and organizational framework for the decentralization program is provided by the 1993 Local Government Act (Act 462), which replaced the Local Government Law of 1988 (Law 207). The initiative for the program was motivated by a political philosophy of ―power to the people‖ and the broader reform agenda whose principles concern the role and responsibilities of the various levels of government and private sector for economic development (Ayee 2003a, 2003b). The process of decentralization in Ghana started with the establishment of 85 districts in 10 regions of the country in 1988. By 1992, the number of districts had increased to 110.The 1992 Constitution, which marked Ghana‘s transition to a multiparty democracy, endorsed the 1988 reforms by consolidating the principles of decentralization within the overall context of a liberal democratic constitution. This constitution put down the principles of the autonomous role of local government and its downward accountability to the populace. A three-tier structure of subnational government was created at the regional, district, and sub-district levels. This includes 10 regional coordinating councils; district assemblies (DAs); urban, zonal, town, and area councils; and unit committees. However, the DA is the key local government institution. Article 241 of the constitution express that they are the highest political authority in the district with considered, legislative and executive power.‖ DAs include both elected and appointed members. Seventy percent of DA members are elected in local government elections that are held every four years. The central government appoints the other 30 percent of DA members and the district chief executive (DCE), who is an ex-officio member of the respective DA. The members of the national parliament from a district are also entitled to participate in the DA with nonvoting capacity. University of Ghana http://ugspace.ug.edu.gh 58 With respect to fiscal decentralization, districts are entitled to generate their own internal revenues. However, the potential to generate their own internal revenue varies significantly across districts: Some districts cover up to 70 percent of their total public expenditures by internally generated revenues, yet some other, poor districts cover only about 5 percent of their total public expenditures by internally collected funds. The latter districts rely mostly on transfers from the national government. 3.1 History of Decentralization in Ghana According to Ayi (2000), the history of decentralization is dated back to the introduction of the indirect rule by the British colonial authorities in 1878, lasting until 1951. During this period the colonial administration ruled indirectly through the native political institution (i.e. the chiefs), by constituting the chief and elders in a given district as the local authority, with powers ―to establish treasuries, appoint staff and perform local government functions (Nkrumah, 2000). Nkrumah also made a vital recognition that, under indirect rule, downward accountability of chiefs to the people was replaced by upward accountability to the colonial authorities. He also posits that, democratic ideals underlying chieftaincy in Ghana, which made chiefs accountable to their peoples, began to suffer as the recognition by the central government was more crucial to the chief than the support of his people”. This situation has been a major carry forward problem to the current decentralization policy. Key policy implementers at the district level in charge of implementing the policy often downplay the effort of the local people and recognize the central level policy makers as the main source of achieving the objectives of the decentralization reform. There are similarities here as well as differences, compared to contemporary period between central and local government in Ghana, driving away any lingering notion of a University of Ghana http://ugspace.ug.edu.gh 59 necessary association between decentralization and democracy, thereby assuring how decentralization can be used as a political mechanism by ruling political elites to reinforce their control. According to Tordoff (1997), in the post-independence period from 1957 onwards, local government was generally weak and subject to the centralization of power that was typical of the post-colonial state in Africa. Efforts with decentralization reforms were initiated at different times, for instance in 1974 under the military regime of Lt. Col. Acheampong, generally characterized by deconcentration, and aimed at strengthening central government control at the local level (Nkrumah 2000). A historical aspect of this was the decentralization policy which was introduced in the early days of Rawlings‘ military rule (1981-92). The announcement of the policy of Administrative decentralization of the central government ministries, alongside the creation of People‘s Defense Committees (PDCs) in each village and town by Rawlings‘ PNDC took place in 1983. The PDCs, made up of local PNDC activists as self-identified defenders of the ‗revolution‘, effectively took over local government responsibilities, though often limited to mobilizing the implementation of local self-help projects (Nkrumah 2000), while the deconcentrated ministries played a more significant role. Aryee (2000) was of the view that despite the PNDC‘s popular rhetoric, its interest in decentralization reflected the previous regimes, which is an interest in Administrative decentralization of central government and not the devolution of political authority to the local level. Then again, Aryee (2000) perceives a key feature of local governance in the pre-1988 period as a dual hierarchical structure in which central and local government institutions ―operated in parallel‖, but with encroachment at times by better-resourced central government on the roles and responsibilities of under-resourced local government. University of Ghana http://ugspace.ug.edu.gh 60 3.2 The current status of decentralization in Ghana- the reality The process of Decentralization in Ghana involves a mixture of political devolution enshrined in the constitution; as well as administrative and technical de-concentration of key service delivery institutions which are in part backed by law and in the main reflecting conventional practices. Though the institutional and legal framework for decentralization implementation have modest progress since 1992 (such as the establishment of District Assemblies, operation of the District Assembly Common Fund), Government support to deepen and institutionalize the decentralization efforts have been incoherent, in part due to several un- coordinated projects and divergent approaches used to implement these projects ( NDAP 2003-2005). Reality of decentralization practices Decentralization is underpinned by Article 35(60) (d) of the 1992 constitution which provides for making democracy a reality by decentralizing the administrative and financial machinery of government to the regions and districts; meanwhile the reality has been a growing concentration of power and resources in key sector Ministries, Departments and Agencies (MDAs) that plan, implement, monitor and evaluate essential services to communities. The drive of Ghana‘s decentralization policy is devolution, and not deconcentration or delegation. Devolution relates to constitutionally or legislatively assigned role, responsibilities and accountabilities to all levels of government. However, the reality of Ghana‘s process of decentralization gives a different picture. University of Ghana http://ugspace.ug.edu.gh 61 The following gives an explanation of the empirical reality of current situation of the decentralization policy in Ghana; Policy and Institutional Arrangement for Decentralization The main motivation for reconstructing the process of decentralization is the faraway accomplishment of the decentralization policy that promises of a transformation of government from centralized administrative and political structure to one in which the powers and authority to plan and implement development are devolved to local institutions and citizens. The three major aspects of this problem in relation to policy development and coordination of decentralization are analyzed below; Level of leadership for decentralization policy management: Political commitment and leadership for decentralization has so far been generally weak. Key stakeholders in the process have been competing and duplicating their roles rather than cooperating and consequently the capacity to manage decentralization is no more a priority and as such diminished drastically at all levels. The extent of policy development and management at the level of The Ministry of Local Government and Rural Development (ML GRD), under civil service law and local Government Act is responsible for managing the decentralization process. It is the lead agency coordinating the overall implementation of the process with the following specific functions; monitoring the decentralization process and effectiveness of Local Government (LA), advising the government on all local government issues, promoting and administering local government training institutions; advising on the review of LG administrative boundaries; coordinating the implementation of donor- support programs on decentralization and local government. University of Ghana http://ugspace.ug.edu.gh 62 To date, the Ministry has not been very effective in the areas of policy development, coordination, performance monitoring and evaluation. The Ministry‘s performance is constrained by limited coordination and cooperation with the sector MDAs, MAs, inadequate budget allocation, inadequate qualified staff, inadequate logistics and equipment, and lack of phased practiced implementation strategy. District Development Funding Facility Resource transfer to districts for the financing of district development plans is addressed in the 1992 constitution. Section 240(2) c states….. ―there shall be established for each local government unit a sound financial base with adequate and reliable sources of revenue‖. Section 252(1) further provides for the ―creation of District Assemblies Common Fund (DACF) and commits government to allocating to this fund, and not less than 5% of the total revenues of Ghana to the District Assemblies for development…‖ Though there exists a constitutionally ingrained policy and mechanism for transfer of resources, from the District Assemblies (DAs), implementation has been problematic. In the recent past, government has not been able to fulfill the promise of making timely and reliable transfer of the District Assembly‘s Common Fund. In addition, forecasts of total district revenues of Ghana to the district revenues tend to be too optimistic while cash released over the physical year regularly fall short of the actual financial requirements of the district plans and operations. For most district plans, actual resource inflows permit only the partial implementation of the District Development Plans. The effect of this on the performance of DAs in the execution of this mandate is deep. University of Ghana http://ugspace.ug.edu.gh 63 Capacity Building and Human Resources Development for Decentralization Capacity building may be defined to include training, human resource development, organization and institutional development, availability of logistics, financial resources and change management (NDAP, 2005). According to NDAP, 17 out of 23 projects in support of decentralization may be classified as capacity building or local government strengthening activities, as reflected in the new programming proposed in the Ghana Poverty Reduction Strategy (GPRS) and Government‘s Decentralization Action Plan. The major concern of Government and development partners is that the current capacity building or Local Government Strengthening Projects are not closely or deliberately aligned with government‘s larger development agenda or vision for decentralization. Rather, attentive and well intentional capacity building activities, developed in collaboration with previous administration, appear to be largely supply driven, based on development partner rather than governments‘ programming interests and priorities. The approach accommodates avoidably high levels of duplication, places heavy administrative and reporting demands on Government at both central and local levels, and makes the mainstreaming of best practices and lessons learned difficult. It also draws government‘s attention away from developing and implementing in systematic fashion. Partnership and participation The government is charged by the provisions in Act 462 to consult with Traditional Authorities in the appointment of 30% District Assembly Members. The lack of specificity in the nature of consultations and the inherent weakness in the institutional involvement of traditional authorities inevitably constrains the achievement of the laudable objectives in this clause. A quick assessment of the current situation confirmed that there is no University of Ghana http://ugspace.ug.edu.gh 64 structured and formalized arrangement that seeks to foster partnership and participation of civil society, traditional institutions and private sector in local governance. Relationship between the District Assemblies and Traditional Authorities are generally restricted to consultation on the release of land and participation in ceremonial functions. The private sector is effectively marginalized in the local government system. Its potential as an important helper in policy/ plan formulation, implementation, monitoring and evaluation is often overlooked by local government. The interaction with the sector is limited to the contractual arrangement to deliver goods and services. Non-governmental Organizations (NGOs) operating within districts also serve mere as donor resourced service providers rather than active partners and advocates in the development discourse. At the national level, the vast potential of ―think tanks‖, academic research institutions, as well as the media are not tapped systematically to promote inputs into the decentralization policy formulation and implementation process (NDAP, 2003- 2005). University of Ghana http://ugspace.ug.edu.gh 65 3.3 General Legal Regulatory Framework and revolution of Decentralization in Ghana The move to decentralization took place in several waves of different intensities, usually linked to a particular government or regimes (Ahwoi, 2010); and the main legislative and regulatory pieces of each wave are listed below; First wave: 1957–1981—Initial steps Second wave: 1982–87—Reflection and preparation Third wave: 1988–1999—Legal basis for current policy Fourth wave: 2000–present—Implementation First wave- (1972–1974) The first wave of decentralization reform begun with the independence and was characterized with conflicting movements of back and forth. Its major drive was the set of legal documents produced in 1972–74. It initially concentrated on regionalization (Constitutions of 1957 and 1960).The Local Government Act 54, of 1961 maintained the duality between central ministries acting through their local offices, and local governments with expanded roles but still limited decision powers. The functions of the District Councils (DCs) included: environment inspection, water and sanitation; protection and security; social services provision (education, health and welfare); and road infrastructure. However, the Act contributed, along with the weak financial and administrative capacity of local agencies, to duplication of responsibilities and services, regional disparities, and an overall bad image of local governments. In face of this rather confusing background, the military Government of the National Liberation Council (1966–1971) commissioned three reports in 1967–68, which pointed to the excessive centralization of authority and University of Ghana http://ugspace.ug.edu.gh 66 recommended a move toward devolution of central authority to the local level. Some of the Commissions‘ recommendations were incorporated into the 1969 Constitution and the Local Administration Act of 1971, and62 DCs were established. But the Act was criticized for attempting to balance a system of quasi-autonomous elected councils and administration by agencies of central government (Ahwoi 2010). For instance, the District Administrative Officer was a representative of the central government. Furthermore, the initiative at implementing that policy was faced with a lot of challenges. The Local Administration Act (359), passed in 1971, was only implemented in 1974, due to a change of government in 1972 and after significant modifications. That reform sought to unify central government branches and local government agencies under the authority of the DCs, and most government functions were to be decentralized, including health; and district departments were to be created for that purpose. The regions were strengthened to provide support to local administrations. However, the 1974 reform, enacted mostly through the Local Administration Amendment Decree (NRCD 258) never succeeded, because of several reasons, among which are: I. lack of political legitimacy and structure at the local level, since two-thirds of lack of political legitimacy and structure at the local level, since two-thirds of council members were to be appointed by the central government and one-third by local chiefs; II. centralization of decision-making in the regions (which was supposed to be an intermediate stage but was never transferred to the districts); III. no strengthening or transferring of capacity and competence at the local level; IV. policy contradictions, with the issuance of decrees which actually recentralized activities that had been under local responsibility University of Ghana http://ugspace.ug.edu.gh 67 V. lack of staffing and recruitment policies for DCs, which ended up hiring large numbers of staff , including many relatives of council members; VI. insufficient efforts at addressing the administrative and logistics needs of DCs; VII. enactment of a Financial Administration Decree (1979) which centralized all fiscal controls in Accra, including payment of local civil servants. In fact the, the 1974 reforms were characterized to some extent by a movement of recentralization of functions previously decentralized, to central ministries or newly created national entities, such as the Ghana Education Service (1976), and removal of local election of council members. The third Republic Constitution of 1979 reinstated elections for local governments, empowered District Councils, and established a Development Fund to support district development through grants-in-aid, and established Regional Councils with responsibilities of regional coordination, planning, supervision and regional development. It was followed in 1981 by a Local Government Amendment (Act 403), which amended parts of the 1971 Act and changed the composition and functions of the Regional Councils. The second wave of reforms: (1982–87) was actually a moment of reflection and preparation for a more solid basis for decentralization (Ahwoi 2010). The PNDC government that took over in 1982 dissolved the recently elected councils and issued Policy Guidelines and a new Local Government Law (PNDCL 14), seeking to restructure the Public Administration System and promoting a ―fully decentralized government system‖. Other policy proposals were made under the Public Administration Restructuring and Decentralization Implementation Committee (PARDIC), translating first into the 1987 Blue Book on ―District Political Authority and Modalities for District Level Elections‖, and then into the Local Government Law (PNDCL 207) of 1988. Three main obstacles University of Ghana http://ugspace.ug.edu.gh 68 were identified in the Blue Book to previous efforts at decentralization: literacy in English (required to be elected to the councils, while 50% of the population was not literate in that language); poverty (50 percent of population lived on less than 1 US$ a day, and could not commit resources to be elected); and capture of local representation by urban elites or political parties. The Law (and related policies) sought to remedy these obstacles, and restrict central ministries‘ role to policy planning coordination, monitoring and evaluation. It envisioned to ―transfer functions, powers, means and competence to the district assemblies (DAs)‖ and to a lesser extent to regional entities. Twenty-two departments from central ministries were to be transferred and established as DA departments. DAs increased in number to 110, and were supposed to exercise state power deliberative, legislative, and executive and administrative as the people‘s local government (Ahwoi, 2010). The law also promoted community participation in administration, facilitated joint ventures among districts, established a district composite budget, and provided a system of checks and balances between local authority and central power. This legal document provided the most comprehensive and clear vision of decentralization to date. Following the district level elections of 1988–89, a National Commission for Decentralization (NCD) was established to strengthen decentralization and evolve a true democracy. Regarding decentralization, the NCD Report acknowledged the weaknesses of local government systems post-independence and the incompleteness of decentralization: ―low development capacity, weak resource and revenue base, poor financial administration and corruption, lack of technical expertise, poor quality staff, poor remuneration, unclear delineation of functions, dysfunctional effect of partisan politics and gerrymandering with local government boundaries‖. The reforms passed during this period resulted in the 1992 Constitution and the Local Government Act of 1993 (Act 462), the University of Ghana http://ugspace.ug.edu.gh 69 first major legal basis for implementation of decentralization. With the 1992 Constitution and its Article 240 begins the third wave of decentralization. The Constitution provides the key principle of the current model of decentralization in Ghana: ―local government and administration should, as far as practicable, be decentralized‖, to ensure ―functions, powers, responsibilities and resources are at all times transferred from the Central Government to local government units in a coordinated manner.‖ The Local Government Act of 1993 (Act 462), section 12 mandated the transfer of functions and responsibilities to the DAs, and is to date the most comprehensive legal document governing decentralization. The act defined the political government entities at the local level to be districts, municipalities and metropolitan areas, and establishes the district (or municipal or metropolitan) Assembly as its highest political and administrative authority, with deliberative, legislative and executive functions. 70 percent of DA members shall be elected by the local population and 30 percent appointed by the president. The DA exercises its administrative and executive functions through an Executive Committee, chaired by a District Chief Executive appointed by the president. The Executive Committee shall have several sub-committees, including, Development Planning, Finance and Administration, Justice and Security, Works, and Social Services. The latter is responsible for social sectors planning, information, needs assessment, and integration with other district sectors and areas (Ahwoi 2010 and Act 462). A specific subcommittee may be formed for health, including at times environment and sanitation. The DA and district-level activities under its authority are to be funded through its own sources of revenue mainly from property rates, levies, fees, and licenses. In addition to transfers from the central government, and has authority to borrow. The act also establishes in each region a Regional Coordinating Council (RCC) and a Regional Planning Coordinating Unit (RPCU) and defines their composition and functions. It University of Ghana http://ugspace.ug.edu.gh 70 defines that existing local branches of central ministries and departments should be transferred, together with their staff, to the DAs, and form part of a Local Government Service (LGS). The District Assemblies‘ Common Fund Act (Act 455) defined the financing mechanisms for DAs. Finally, the National Development Planning (System) Act (Act 480 of 1994) instituted a decentralized planning system in which the DAs were key players of the planning process. The 1992 Constitution and these three legislative pieces form the legal basis for the current decentralization framework in the country. However, they only provide the general legal basis for the establishment, structure and functions of LGs. Actual establishment of DAs requires a specific legal instrument (LI) issued by the MLG for each district created. Therefore, in the late 1980s a burst of LIs were issued for the establishment of 110 districts. Others were issued in the mid -2000s for the 60 newly created districts. Among other things, these instruments provided a detailed list of 86 functions to be performed by the DAs. These functions may be as specific as possible but in other areas they are very vague. Overall, these functions are not structured along functional or strategic areas, and focus on specific activities to be performed rather than general responsibilities, objectives or results. Legislation and regulations in the early 2000s have been more incremental and instrumental. The new legislation focused on complementing the basic framework given by the Constitution of 1992 and the Local Government Act of 1993, by detailing regulations for supporting public sector operation under decentralization (financial management, auditing, local government service) or on providing guidance for the implementation of reforms. The main legislation pieces produced in that period include the Local Government Service Act, the Internal Audit Act and the Institute of Local Government Studies Act, the Public Procurement Act (all from 2003), and the Operational Guidelines for DPCUs and RPCUs and the Creation of Districts Instrument of 2003). They later created 19 new districts on top of the 110 defined University of Ghana http://ugspace.ug.edu.gh 71 by the Constitution. An important policy document was produced by the MLGRD in 2003: the National Decentralization Action Plan. It recognized the slow pace of progress in the institutional and legal framework, and the inconsistency and weak coordination of initiatives aimed at deepening the process, and proposed a more incremental and realistic approach to decentralization, with a strong emphasis on establishing systems and mechanisms for supporting decentralization, building consensus, convergence of the currently competing approaches, strengthen institutions especially at the local level, harmonization and coordination of decentralization efforts and initiatives. The Plan aimed to ―promote convergence of the decentralization efforts, consolidation of the processes of resource allocation and management, building capacities for poverty-targeted development and governance at the local level and promotion of partnership and participation between local government, civil society, the private sector and traditional authorities.‖ Within that broad objective, the document proposed to work along four strategic program areas: (i) policy and institutional arrangements for decentralization implementation, (ii) a district development funding facility, (iii) capacity building and human resources development, (iv) partnership and participation for accountable local governance. Additionally, it strengthened inter- ministerial coordination on decentralization through cabinet-level structures (secretariat and committees) and processes. The objective was reinforced by merging local deconcentrated departments and agencies often subsisting along strong hierarchical lines into one administrative unit under the DA‘s authority. This was also a key element in the Local Government Service Act of 2003 (Act 656), which established a separate service for local government civil servants. As a result, it was expected that 33,000 civil servants (78 percent of the total) would be redeployed to local governments starting January 1st 2007. The Act established a single LGS separate from the Civil Service, covering employees of the RCCs, DAs, and sub-district structures. University of Ghana http://ugspace.ug.edu.gh 72 Rather than decentralizing personnel management to LGs, the Act has centralized it in a new bureaucracy, responsible through its district offices for staff appointment and promotion. The staffs of the decentralized departments listed are assumed to be transferred to the DAs‘ authority and, thus, become staff of the LGS. Both education and health staff, however, were left out of the Act and remained as separate services (GES and GHS). Act 656 therefore conflicts with Act 462, which had included health and education as decentralized departments. Finance of the DAs was re-centralized into a general Consolidated Fund According to the Act. 3.4 Review of the status of health sector decentralization in Ghana There have been big disagreements between government intentions for decentralization, including the intentions as expressed in the constitution and the actual practices in the country. This divergence is supported by different types of laws which give power to the divergent practices especially in the choices made by sectors that tend to define decentralization as‖ deconcentration‖. The divergence can be partially explained as a result of lack of clarity of the local government Act itself since the Act fails to adequately deal with range of issues further debated in the report including the assignment of functions to different levels of government and the overall government structures and their relationships. The Act pays very limited attention to sub-district levels just as there is substantial un-clarity regarding the extent to which the regional level should be considered a fully- fledged local government. Though considerable progress has been made in the establishment of political and administrative institutions to implement decentralization, progress in terms of Finance for the past 10 years such as the overall legal framework for MMDA finance, especially concerning financial management within the treasury systems, the transfer systems, and various capacity building initiatives such as initiative to University of Ghana http://ugspace.ug.edu.gh 73 strengthen the budgets towards the MTEF and preparations for composite budgeting. In addition, various studies of the transfer schemes, financial management performance and revenue mobilization practices of the MMDAs have been completed, which will provide valuable input to the coming reforms, and in areas of Human resource issues. However, the status of the implementation policy has not reached the level assumed to be at this period especially in the health sector. The decentralization of the Ghana health service started in 1988 with the World Health Organization- supported strengthening District Health system projects, which led to passing of the 1996 comprehensive program of administrative deconcentration and delegation. While the early years of this program focused primarily on the institutional strengthening of Regional and district health administrations within the Ministry of Health, more recent proposals have gone yet further toward increasing the range of decision-space of these regional and local institutions. Notable among the proposed reforms is a new system of budget decentralization to regional and district level offices of the MOH, implemented through the 1997 Program of Work (POW) in the form of a hierarchy of semi- autonomous budget management centers (BMCs). The program of fiscal decentralization with the MOH already had significant results in allocating health resources toward district based – services and primary care. Although the 1988 reforms have achieved some political mobilization at the local level and some limited fiscal decentralization, they have had little visible impact on the health sector or public service delivery, with the possible exception of road and infrastructure development. Overall, Ghana has established several building blocks over the years needed for successful decentralization, but this efforts lack cohesion and unity of purpose (Couttolenc, 2012). University of Ghana http://ugspace.ug.edu.gh 74 Most of the basic elements of administrative decentralization are already existing in the form of the district political and administrative units, their DA with a significant, though still weak, management structure, and GHS District Administration offices. Though a number of useful information systems and management tolls are put in place and implemented, including planning and budgeting systems, reporting as well as information systems, performance measurement, financial transfer mechanism to local government and many others, however their effectiveness is limited by the many overlaps and duplications, the fragmentation among systems, and especially their inability so far to produce reliable information (Couttolenc, 2012). In recent times, the MOH has been able to come out with a consistent policy document addressing specific issues of decentralization in the health sector. Despite this effort by the MOH, a comprehensive and clear policy framework to guide implementation in health is still lacking. The framework in place concerning health is confusing and contradictory. That is existing laws and regulations offer changing and conflicting views of what decentralization should look like in the health sector and are quite unclear as to which functions are to be devolved. In addition the financing system of local governments is complex and confusing as different funding sources specialize in financing specific line items or programs, and the DAs allocation of resources to the sectors is not transparent. According to Couttolenc(2012), fiscal decentralization in Ghana is more apparent than real: over 50 percent of public health expenditure is allocated to the district level, but the larger part of these resources are allocated and controlled by the central government; local authorities whether DAs or GHS District Offices and facilities have little real decision power on resource allocation. There is a contradiction in general legislation that relates to management of staff. A broader legislation calls for ― full‖ devolution to local governments‖, but management of University of Ghana http://ugspace.ug.edu.gh 75 local staff has been centralized in parallel LGS, which in effect withdraws from LG authority over the major resource they need for managing local services. The issue of resource transfers to Districts for financing of District Development plans is taken care of in the 1992 constitution. Section 240(2) c. states ―there shall be established for each local government unit a sound financial base with adequate and reliable sources of revenue.‖ Section 252 (1) goes on to provide for the creation of District Assemblies Common Fund (DACF) and commits government to allocating to this fund‖….not less than five percent of total revenues of Ghana to the District Assemblies for development...‖ (NDAP, 2002- 2005). Though there is a constitutionally entrenched policy and mechanism for the transfer of resources from and to the District Assemblies, implementation has been problematic. In recent past, government has not been able to make DACF transfer of resources to DAs in a timely and reliable fashion. The political commitment to, and leadership for decentralization has so far been weak. Key stakeholders in the process have been competing rather than cooperating and consequently, the capacity to manage decentralization has been diminished. Capacity needed for successful implementation and management of the decentralization policy in the health sector is low, not only at the district level, but across all levels of government, although the weakness may be different at each level, and one of the main problems is the lack of reliable information for decision making, monitoring an evaluation. Participatory mechanisms in a form of local councils and committees appear to have quite variable effectiveness; in many areas, they have not been functioning as an effective channel of participation in decision making and planning, and this poses a major obstacle in the implementation of the decentralization policy in the health sector. This is because; University of Ghana http://ugspace.ug.edu.gh 76 many stakeholders have limited understanding of the process objectives, requisites and implications. This is especially true in the health sector, as the survey of regional and district officers clearly showed, and makes it difficult to build consensus and support for the process (Cuttolenc, 2012). 3.5 Study areas 3.5.1 Sampa Government Hospital Sampa Government Hospital is the public facility that was used for the study. It was started in the colonial days as an immunization unit for the control of smallpox. During the period of 1957- 1966, the Hospital benefited from the health center projects and was upgraded to that status. During the PNDC/ NDC era, the Health center was given a hospital status but without infrastructural enhancement. Sampa is a town in the Brong Ahafo Region of Ghana, on the border with Cot D‘Ivoire. It is the capital of Jaman North District, and was formerly the site for a slave market. It is the biggest border town in Ghana with a population of 14,974 (GSS, 2010). It is the principal town of the Nafana ethnic group. The geographical coordinates of Sampa are 70 57` 0‘‘ North, 20 42` 0‖ West. Agriculture is the dominant economic activity, and employs 70% of the total population within the labour force. Major sectors of agriculture in the district are crop farming and livestock rearing. The vegetation in Sampa is characterized by woodland consisting of widely dispersed short trees and grasses/ shrubs. This part of land is suitable for the cultivation of cashew, yam, cassava, rice, beans and groundnut. In 2007, a philanthropist residing in the United States of America, who previously donated an ambulance, reference books, computer and assorted medical equipment to the Hospital, in collaboration with a philanthropist group based in the United States of America, built University of Ghana http://ugspace.ug.edu.gh 77 the Maternity and the General Wards which necessitated the movement of the Hospital to its new site in July, 2010 and currently operating with 71beds. The hospital is the only Primary Ghana Health Service facility in the Jaman North District of the BrongAhafo Region operating on not- for-profit basis. The Hospital serves as a Referral Hospital for all sub- District health facilities with an average daily attendance hovering 135 patients. The Hospital serves as the main health facility contracted for the provision of health services to clients of the Jaman North District Health Insurance Scheme (JNDHIS) of the National Health Insurance Authority (NHIA). The Hospital has both Mission and Vision statements which are in line with Ghana Health Service (GHS). 3.5.2 Services provided The hospital provides services in the form of General Outpatient Consultation, General Inpatient Care, Emergency medical care, Basic Diagnostic Services (Laboratory and Ultrasound, Surgical interventions, Antenatal, Deliveries and Postnatal, Eye Care, Basic ENT services, HIV Testing and Counseling (HTC)/ Prevention of mother to Child Transmission (PMTCT) services, Anti- Retroviral Therapy( ART), Pharmaceutical services, General administrative support services and Health education University of Ghana http://ugspace.ug.edu.gh 78 3.6Dormaa Presbyterian Hospital 3.6.1 Historical development of the hospital With communal labour, the construction of Dormaa Presbyterian Hospital was started in 1953 and completed in 1954. The official opening of the facility took place at the end of 1954. In 1955, the management of the Hospital was transferred to the Basel mission with assistance from the Government of Ghana. Dormaa Ahenkro is the capital of Dormaa Municipal, of the Brong Ahafo Region, South of Ghana. It is located about 15 kilometers from the Ivorian border. It lies in the geographical coordinates of 70 17‘ 0‖ N, 20 53` 0‖ W. It has a population of about 22,000(Dormaa Municipal Assembly, 2012). The Bonos dominate all the ethnic groups in the town with about 97% of the total population. The town is well endowed with natural resources, particularly tourist‘s attraction sites, mineral deposits, forest and timber species, rich soil and good climate conditions. The mainstay of the town‘s economy is agriculture. It currently employs about 60% of the economically active labour force. Farming is largely carried out on small scale basis. At the moment, the poultry industry specifically, eggs production is operating on large scales. Livestock such as cattle, sheep, goats and grass cutters are also reared. The upgrading of the facility to a District Hospital in 1976 required some infrastructural improvement. In September, 1976, at the required of and in consultation with the regional medical officer and the hospital board, Dr. Van Es started work on an application for the expansion of the hospital. In 1978, the project was approved by one of the Church‘s overseer partners; inter Church Organization for Department Co-operation (ICCO) and the Dutch Government. By the end of the in project in 1984, new maternity, children and isolation wards, a store and some staff houses had been provided in the Dormaa Hospital. University of Ghana http://ugspace.ug.edu.gh 79 3.6.2 Major activities of the hospital Brong –Ahafo Presbyterian Health service aims in three core areas clinical service, research and teaching which are shown below. 3.6.3 Clinical service The hospital provides efficient and effective service affordable to commissioners and desirable to patients and referring clinicians. It also provide additional services for the local population and support the principle of local access wherever possible and to deliver service to the highest and standard to the patients. 3.6.4 Research and Development They develop collaborative and consultative research partnership with patients, cover and public and to support research of national and international excellence and innovation. They develop the Trust‘s research portfolio in line with its service strategy. 3.6.5 Teaching and Learning The Dormaa Presbyterian hospital pursues teaching and learning partnership with educational providers and other stakeholders. They embrace personnel and organizational development, and also encourage a culture of innovation and enterprise and also to maximize recruitment and retention by meeting the development needs of the current and prospective staff. Apart from this, the Hospital has priorities of improving their corporate image, boosting staff morale, improving client satisfaction and ensuring respect for the core values of owners and promoting greater accountability to all stakeholders. University of Ghana http://ugspace.ug.edu.gh 80 3.7 Justification of the study areas Sampa Government Hospital in Jaman North District and Dormaa Presbyterian Hospital in the Dormaa Municipality in Brong- Ahafo were the case hospitals. We chose these hospitals because, one is a mission hospital and the other is a public hospital. For this reason, the study of these facilities will facilitate comparism of the findings from the two hospitals, whether they have the same implementation challenges of decentralization. If not, what account for such differences? Again, Sampa District has a low level of economic activities, and has a lot of challenges in terms of human resource, finance, and social amenities and so a study of such facility is likely to reflect most of the challenges of implementation of decentralization in other District hospitals with similar problems. Dormaa Municipalities on the other hand, is a little better than Jaman North District in terms of socio- economic factors, social amenities, favorable geographical factors like vast forest land to support most cash crops in Ghana, proximity to the regional capita (Sunyani), adequate rainfall and many more. For this reason, the study will be able to know whether these factors play a role in the implementation of the decentralization policy or otherwise, and if so then what measures can be adopted to balance these disparities. 3.8 Conclusions This chapter considered extant literature on the implementation challenges of health sector decentralization at the hospital level in the context of both developed and developing countries. The chapter starts with introduction which explains how the researcher went about it. The concept of decentralization and its historical background in both developing and developed world are discussed. The revolution of the concept and its legal University of Ghana http://ugspace.ug.edu.gh 81 underpinnings in Ghana are reviewed as well. The background of hospitals and the rational for their choices were also considered in this chapter. Though the empirical literature from both developed and developing countries on decentralization as a reform in the health sector has been extensively evaluated, there is different success rate across the world. However, evidence on the implementation challenges at the hospital level is scarce in both developed and developing economies. A review of empirical literature is Ghanaian context often concentrated on the implementation challenges of health sector decentralization in general with little focus on the challenges of implementing decentralization reform at the hospital level. The methodological stance of the study which elucidates the approaches or methods to be employed in achieving the set objectives of the study follows this chapter. University of Ghana http://ugspace.ug.edu.gh 82 CHAPTER FOUR RESEARCH METHODOLOGY 4.1 Introduction This section presents the methods that were used. It explains the choice of qualitative research methodology used in the study; data collection and analysis. It also outlines some of the strengths of qualitative research approach that was used in the study. 4.2Research Methods 4.2.1Research Paradigm In order to find out the implementation challenges of decentralization at the hospital level, there is the need to understand the experiences and the roles of the key policy implementers of health decentralization at the hospital level. The qualitative research paradigm is used due to its suitability with regards to getting to know peoples experiences and current insights about the implementation challenges of decentralization at Sampa and Dormaa hospitals. Different stakeholders including key policy implementers, decentralized health institutions, civil society organizations, NGOs among others are involved in the service delivery under the decentralized system. For this reason, the qualitative research method or paradigm offers the best platform in situations where events can be interpreted, described and explained by different actors with different functions in a given context. In view of the fact that qualitative research technique dwells more on the interpretations, it provides the opportunity for individuals with intricate views and experiences on issues to bring them out and hence aids in conducting detailed search about a given phenomenon. University of Ghana http://ugspace.ug.edu.gh 83 4.3 Research Design The qualitative research design is employed for the study, and the case study approach was adopted for this study as the researcher focuses on the implementation challenges of decentralization at the District hospitals in the Dormaa Municipality and Jaman –North District in the Brong - Ahafo Region of Ghana. The case study according to Yin (1989), is an empirical inquiry that investigates a contemporary phenomenon within its real life context using multiple source of evidence. The case study therefore turns to focus on a particular issue, feature or unit of analysis. The purpose is not for generalizability but to fully understand a given phenomenon. The case study method is seen suitable because it ensures that a phenomena under study is not explored through one lens, but rather a variety of angles which allows for multiple facets of which a phenomena can be revealed and understood. The case study approach is also adopted because, it offers the researcher the opportunity to learn about issues that are very complex and perhaps needs a broad assessment of the issues which can be obtained through extensive explanation and analysis. 4.3.1Sources of Data Primary and secondary data were used for the study.Primary data constituted interviews held with key informants who are in-charge of implementation of the decentralization policies at the district hospital levels. Secondary data composed of review of annual reports, service delivery report before and after decentralization and the District health report of the two districts. University of Ghana http://ugspace.ug.edu.gh 84 4.3.2. Sampling/ Target population The purposive sampling technique was used. The purposive sampling refers to the selecting of participants to give an information about a given problem under study because of the characteristics these people possess. This technique was found suitable because it enabled the study to target only individuals responsible for the implementation of the health sector decentralization at the district hospital. Different groups of key health policy implementers were targeted in the study in order to explore the implementation challenges of the health sector decentralization. The health sector decentralisation at the district hospital level brings together different groups of people in different organisations to ensure their successful implementation. In view of this, the target population included some key staff members of the Assembly, core management team members of the hospitals and the director of health in the district. The key units of the assembly include finance, planning, coordinating officer and the office of the MCE/DCE. The heads of these units were purposively selected and interviewed since they are the main people responsible for policy design and implementation in the assembly. In all, four (4) each including the MCE/DCE, Planning officer, coordinating officer and the finance officer were interviewed in the two assemblies. 4.3.3 Instruments for data collection Data were obtained from respondents with the aid of an interview guide that was designed to solicit the views of interviewees. Questions in the interview guide gathered information relating to health sector decentralization policy implementation issues, and the challenges of implementing this reform at the hospital level (see appendix B). Interviewees responded to the same questions in both facilities. Prior to the interview, the consent of the interviewees were sought to record the interview using a tape recorder. In addition to the University of Ghana http://ugspace.ug.edu.gh 85 recordings, notes were taken with the help of a research assistant and the researcher to ensure that all important details are captured. Each interview lasted for forty five minutes. On few occasions did interview lasted for fifty minutes due to the interference by colleague workers and patients. 4.3.4 Data collection process Key groups of people interviewed included the District Assembly‘s staff, the District Director of Health, and the core Management Members of the District Hospitals. Total of 22 respondents were interviewed from the two hospitals (11 from each hospital). The respondents included the MCE, District Planning officer, District finance officer, coordinating director, and District director of health. The other respondents included the Health administrator, Matron, Medical director, Head of pharmacy, Finance officer, and Head of records of the two hospitals. The research was in two phases. To ensure reliability of the responses, the researcher went back to some key informants after the first phase to ask them the same questions that they have answered already. This informed the researcher as to whether the information given already was the same as the one given in the second phase. This also gave the researcher the chance to clarify some issues that were not clear during the first phase of the interview. 4.3.5 Data management and analysis The recorded data were transcribed whilst field notes were arranged and analyzed descriptively using thematic analysis technique. This involved reviewing and grouping the data, making notes and sorting the data into categories based on the objectives of the study. It allows the researcher to move the analysis from a broad reading of data towards University of Ghana http://ugspace.ug.edu.gh 86 discovering patterns and developing themes. Finally, the summary of the data analyzed was then provided. 4.3.6 Reliability and Validity Reliability and validity are two important concerns of every study since they may undermine the strength of a set of data in which final conclusions are made. The researcher ensured the validity and reliability of findings by making sure that pilot study was conducted prior to the main study. Some selected respondents were also interviewed the second time after some gaps were found in the initial findings. During the second part of the data collection process, the researcher contacted some respondents who were interviewed initially and asked questions that had been asked earlier (first phase of the interview). Despite the fact that respondents were proving difficult in answering questions they had answered earlier, the researcher was able to convince them by explaining the rationale of asking informants the same questions that have been asked during the first interview. This approach helped the researcher to verify whether there could be some differences in the responses of the same respondent at different time periods. In addition, because the data or the responses were obtained from different group of people in charge of implementing health sector decentralization, data reliability was assured as responses from different health policy implementers were compared for similarities and differences and were all factored into the analysis. Again, because the note taking were done by two people, in addition to recordings, all detailed information were captured. Also, the researcher used two different facilities that were private and public hospital. To ensure reliability the researcher ensured that the views of the respondents from the two University of Ghana http://ugspace.ug.edu.gh 87 hospitals are sought and analyzed which reflect the totality of challenges of implementing the decentralization policy in the mission and the government hospital. 4.3.7 Ethics The research was conducted in accordance with the principles of the University of Ghana research ethical rules. Respondents did participate in the study voluntarily because their consent and permission were obtained before involving them in the study. They were also free to decline from the study as and when they wished. The rationale or the purpose of the study was explained to the respondents before they decided to partake in the study. Permission was sought from participants before responses were recorded and notes taken. Essence of recording responses was explained to the participants. Interviews were conducted under a strict confidentiality and privacy at the respondent own fixed time. 4.3.8 Limitations of the study The major and only problem encountered by the researcher was financial constraint. The study was constrained by limited financial resources. Collecting data involved extensive travel around the two districts as well as paying research assistant who helped in the data collection process among other costs. This made it difficult to include more hospitals in the study. 4.4 Summary This chapter dwelt on the methods and techniques of data collection for the study. It begun with the research approach and design. It also described the methods of data collection and what informed the choice of such methods. Besides, data management and analysis, measures to ensure reliability as well as ethical issues were all taken care of. University of Ghana http://ugspace.ug.edu.gh 88 CHAPTER FIVE PRESENTATION OF FINDINGS 5.1Introduction The focus of this chapter is to present and analyze the data gathered from the interviews on the implementation challenges of decentralization at a public and private hospital at Sampa and Dormaa respectively. 5.2 Description of findings In order to provide explanation of the implementation challenges at the hospital level, the core management team at the hospital, the district directors and some key members in the selected Assemblies were asked questions about:  The implementation challenges of decentralization at the Sampa public and Dormaa private hospitals.  The important contextual factors taken into consideration in the process of decentralization at the two hospitals.  The implementation status of some key reform measures of decentralization policy at the Sampa public hospital and Dormaa private hospital. 5.3 Implementation challenges of decentralization at the hospital level 5.3.1 Funding It was revealed through the interview that, funding has been the major obstacle for the successful implementation of the decentralization policy at the hospital level. For the success of the decentralization reform, hospitals must be financially sound to be able to carry out their operations effectively. It was recognized through the interview that, the University of Ghana http://ugspace.ug.edu.gh 89 operations of hospitals were solely based on their Internally Generated Fund (IGF), and about 95% of this fund comes from the Health Insurance, and the remaining coming from mortuary charges, cash and carry system and donors. Unfortunately, there are delays in reimbursing the hospitals by the National Health Insurance Authority. This problem makes it difficult for them to be independent in carrying out their activities. The Medical Director of Sampa stated that: In my view, funding has been the major challenge of decentralization at the hospital level, because the payment of the health insurance does not come as expected, most of our activities are not followed as scheduled. As we sit here, the health insurance Authority owes us from December last year up to date and we are almost half way into the year and we are supposed to implement the plans we have for 2014. Meanwhile we have not been paid. We are still receiving monies for 2013, and if we have planned for 2014, the first quarter is gone, but yet we have not received anything. How do we implement our planned activities for the first quarter? But we still have to run with consumables. We have to take items on credit from suppliers, so when the money comes we need to pay suppliers first before you can think of other things. The planning officer of the Dormaa Municipality also added that: For Administrative decentralization, I think we are doing well but financially, we are not close to it at all. You prepare your budget, submit and go and defend it, but somebody in Accra tells you, Dormaa you need this rather than what you have submitted, and this may prevent you from achieving your planned activities. University of Ghana http://ugspace.ug.edu.gh 90 These financial irregularities have been a major challenge for hospitals to run independently in terms of planning and other decisions such as ensuring quality of health care delivery and accessibility mostly by rural dwellers as package for decentralization. The issue of financial autonomy was problematic in both hospitals. Though government had improved on other forms of decentralization like devolution, delegation and privatization, financial decentralization was not in operation as it is highly centralized. This limits the operations of the hospitals as approval to spend must come from the top hierarchy. 5.3.2 Communication barrier between key policy implementers For every successful policy implementation, there is the need for effective communication, but lack of communication among the key implementers of decentralization policy at the hospital level was found to be one of the major drawbacks of this policy. There were no clear communication mechanisms between the members of the Assembly, the Health Directors of the district and the hospital. This problem was more pronounced at the Mission Hospital as compared to the Government Hospital, because they consider themselves as a Christians Health Association of Ghana (CHAG) facility. For this reason they have a different channel of reporting to Ghana Health Service as compared to the Government Hospitals in the district. They do not report to the Municipal Director of Health, and she is not aware of their activities and plans for the year. She only gets to know some of their activities, because she is a board member of the hospital not because of her position as Municipal health Director. University of Ghana http://ugspace.ug.edu.gh 91 The planning officer of Dormaa Municipality pointed out that: Rubbing shoulders with the district director in terms of their activities is missing and because of this the MDH has decided to set up a clinic due to some of these reasons, and in the process, this facility can grow to become a public hospital. Even among the core management members of the hospital, most of them were not aware of the decentralization and how it works, as some of them claim there has not been enough communication from the center to Districts in terms of individual‘s role, and what constitute the policy in general. Some respondents complained that they need to be updated on what the policy is about because they hardly get the whole concept of decentralization. Because of this problem, the accountant of Sampa Hospital stated: Even we were deciding to put up a nursing school, but the assembly is not clear as to who is responsible for the project, whether the health directorate, or the hospital. Everybody uses his discretion because we have little communication and collaboration among the assembly, health directorate and the hospital. The Municipal Health Director of Dormaa also added that: Because this is a Christian Health Association of Ghana (CHAG) facility, they can decide not to tell you anything about their activities. Initially, everything passed through the districts with regards to the activities of these facilities, but now the situation is worsened and it‘s very difficult to comment with regard to their staffing and other activities because they can decide not to tell you. University of Ghana http://ugspace.ug.edu.gh 92 Lack of communication among the key policy implementers at the hospital level is creating a whole lot of problems because there is little communication and collaboration among the key implementers of the policy at the hospital level. 5.3.3 Human resource and capacity issues Human resources play a major role in the implementation of every policy especially decentralization. This is because to empower local people to be independent, you also need competent people to work in the various departments to ensure that quality health care service is delivered. However, throughout the interview, human resource was mentioned as one of the major challenges to implementing decentralization at the hospital level. The hospitals frequently rely on casual workers because the hospital is in short supply of skilled staff. The District Director of Health at Sampa opined that: There is only one medical officer in the whole hospital who also serves as the medical director of the facility. He is a human being and in case he falls sick, or travels what happens to our patients? The illness that could have been managed here now has to go to the next hospital with the dangers involved. The Medical Director of Sampa expressed that: Human resource in terms of quality and quantity has been a major challenge for effective decentralization reform, because if we want to run twenty-four (24) hour emergency or pharmacy, we do not have the adequate and quality staff to render all these services. In Sampa, the health management team had assisted in recruiting forty (40) casual workers which has helped to strengthen management and service delivery, but this also University of Ghana http://ugspace.ug.edu.gh 93 comes with its own disadvantages. This is because, according to the health administrator of the hospital, ―you cannot strictly criticize them on their output since they are not professionals‖. 5.3.4 Inadequate support from government The respondents indicated that the policy has not received much attention from the government in terms of infrastructure, funding, and even the power given to the local people to take some basic decisions as an independent institution. This was more serious in the areas of human resource functions such as recruitment, transfer and so on. The health administrator at Sampa Government Hospital indicated: Of all the buildings here, the only one put up by the government is the emergency unit structure over there and it‘s even yet to be completed, all the rest are from donors and you want decentralization to work under this condition. He further added that: Even if somebody wants to build for the hospital, the Central level must be informed before. We just purchased this Nissan over there and we were asked to get clearance from the regional level before we are able to buy it, even with our own money. We don‘t even have the authority to create our own website for people to see the condition of the hospital so they can come to our aid. Hospitals operate basically on the internally generated fund mainly from health insurance. However, government mostly delays the release of this fund, but emergency drugs must be University of Ghana http://ugspace.ug.edu.gh 94 bought irrespective of whether you have money or not and this makes their operations very difficult and making it impossible to be self- reliant. 5.3.5 Non- involvement of the local community by health workers in planning and managing health service programs. Findings from the district level showed that some health workers still think that planning and managing health service activities are the sole responsibilities of health workers in the hospital. In other words, the health workers have a perception that, representatives of communities cannot take part during the planning process as well as in the management of the health service programs. When asked about how the planning of their activities takes place, almost half of all the respondents made the same statements about how their planning take place without community involvement. One of the respondents said: We have what is called Program of Work (POW), which is done at the beginning of every year and this details what is to be carried out from the beginning of the year to the end. When approaching December every year, a memo is sent to the various units to ascertain what their needs for the year are, and we then take them into consideration and cost all the items and see whether it‘s within our budget. If there is the need to cut it down afterwards, we do so after meeting with the core management team. Almost all the key informants from the core management team repeated the same thing which was a clear indication that the community representatives are not part at the initial planning stage, but only come in when board is taking decision concerning other matters. University of Ghana http://ugspace.ug.edu.gh 95 This makes it difficult for the needs of the community members to be captured in the planning process. 5.4 Important contextual factors taken into consideration before decentralization at the hospital level This objective seeks to get ideas from respondents as to whether they were aware of some contextual factors that were considered before designing and implementing the decentralization policy at the hospital level. This is because Districts and Municipalities are different with regards to factors such as wealth, education, natural resources, political culture, and social organization. These factors have major effects on the way in which health sector reforms are implemented and how the organization functions. However, whether the government considered these contextual factors in the process of decentralization, a whole lot of varied responses were given by the informants. Majority of them were not aware whether some factors were considered, and how these factors have affected the implementation of the decentralization. For instance, the Director of health of Jaman North District expressed that: I cannot tell because the decision was taken from the top before coming down and once they accept, it is brought to us but not aware whether they were considered”. Some of the respondents were of the view that population factor was one major consideration. They claimed that, before you obtain the status of a district, there is the need to get a given number of population. However, the respondents were not certain about the required number. The Administrator at Dormaa Hospital added that: ―I think they considered factors concerning whether they will be able to work with little supervision, but the challenge is getting the qualified staff‖. University of Ghana http://ugspace.ug.edu.gh 96 Some of the responses obtained from the informant showed that, districts were different and management of hospitals were also different and so cannot carry what worked for one hospital or district to the other and the outcome cannot be generalized. In view of this, they think some contextual factors were taken into consideration. Though they were not clear about the factors taken into consideration, respondents were able to come with how socio- economic factors affect the implementation of decentralization at the hospital level. One respondent revealed that: Unemployment restricts people from accessing health care, as well as population growth and its effect on health facilities, and all these can negatively affect the purpose of decentralization and its implementation. A pharmacist at Sampa Government Hospital commented that: I think a lot of such factors might have been considered because of diversity, and I hope not every single factor was given consideration but I believe a lot of them were considered though majority may also have been borrowed from foreign countries. The respondents expressed that these factors might not have been considered fully, but we have some districts which are very poor and think they may be given more money, since they cannot generate enough income locally compared to others. The head of pharmacy at Dormaa Presbyterian hospital opined: University of Ghana http://ugspace.ug.edu.gh 97 I think some contextual factors were taken into consideration, because even in Dormaa hospital, we have our own decentralization policy concerning siting of CHIP compounds under this hospital. For us to site a clinic at any place, we take a lot of things into consideration, and I‘m sure those factors were taken into consideration at the top. Our Primary Health care unit is responsible for identifying areas where healthcare services are needed and then comes to consult. When they are given the go ahead, they build a facility over there. Distance and referral systems are some of the factors considered. But as stated earlier, most of the respondents were not aware of such factors and those who think the factors were considered were not sure of these factors. 5.5 Implementation status of some key reform measures of decentralization at the hospital level This objective sought information from respondents on the implementation status of some key policy reform measures and the gaps in the implementation of health sector decentralization. The findings from the key informant responsible for the implementation of health sector decentralization at the hospital level are presented below. 5.5.1Policy issues and objectives It was recognized that majority of the respondents were not consulted in the policy formulation process. The respondents from both hospitals were of the view that health specific policy goals was in conflict with each other. They were lacking clarity and no clear strategy for implementation. “Spreading of information to the various stakeholders in the district was lacking which brought about misunderstanding and delay in the University of Ghana http://ugspace.ug.edu.gh 98 implementation of the decentralization policy” as stated by one of the respondents put by one of the respondents. In the same way, the policy was unable to take into consideration intra- district variations such as communication issues, utilization, geographical difficulties, and language issues during the implementation of the policy. 5.5.2 Organizational reforms The participants were of the view that there have been some slight changes but were not aware whether such changes were as a result of the decentralization reform. However, there is no right structure to coordinate decentralization at the district level, and achieving the objectives of the health sector decentralization with the current situation will be difficult were the comments of some respondents. Some informants were even doubtful if the capacity to coordinate all functions related to decentralization is in place. The respondents were of the view that; “they have not experienced any new structure and that they are still working with the old one which is making the implementation very difficult”. Again, the respondents criticized the central government for not taking proactive measures in restructuring the organizations involved. They opined the need for proper structuring at the district level as well as the capacity for implementation. University of Ghana http://ugspace.ug.edu.gh 99 5.5.3Planning, financing and monitoring Annually prepared plans by the districts are not recognized by the central authority (MOH). Mostly, the plans of the districts are based on the targets of the central plans. Some respondents expressed that: You plan your budget and someone in Accra tells you what you need instead of the things you have planned for. Sometimes they even cut down your budget and tell you where to put more of your budget without knowing the needs of the community in question. Respondents explained that central planning teams lack the required skills and the capacity to guide the district team in preparing a need- based district plan. Community members were not aware of the planning practices in the two facilities. Support from the local people only lies in the areas of community mobilization activities, or contributions to support the hospital financially. They are not consulted in regular programs of the health facility. One respondent expressed that, ―even in decentralized health facilities, the centralized planning and budgeting system still continues”. Elements of essential healthcare services carried out in the hospitals and the peripheral health facilities including administrative cost were mainly financed by the health insurance. The rest is compensated through the ―cash and carry‖ system and donor agencies. However, the health insurance which constitutes about 95% is sometimes unavailable and/or it is not released at the right time needed. This makes the operations of the hospitals very difficult. The Medical Director at Sampa Government Hospital opined that: University of Ghana http://ugspace.ug.edu.gh 100 We are almost half way of the year, but we have not been paid by the Health insurance Authority for the year. How do you expect us to operate and run the hospital .The government does not give us any money as we rely solely on our IGF‖ The health sector budget does not reflect district needs as expressed by the district plans. The rules and regulations governing financial issues were very rigid and complex making it very difficult for local decisions on internal transfers as per local needs. Central level mostly fails to put into practice the feedback from the district level with regards to financial matters. An example given was that, approved funds were not timely released to increase efficiency and timely achievements of set targets. 5.5.4 Human resource and Capacity issues Human resource functions were highly centralized, leaving very little decision for districts to be taken on these functions. One respondent expressed that: Some decisions regarding human resource management should be taken at the district level. This is because, we can only control the output of the workers if basic decisions concerning firing, hiring, sanctions, and so on are taken at the district level. A worker puts up irresponsible behaviour and says you cannot do anything to me because you did not employ me. Human resource was in short supply in both facilities but this was more severe in the Government Hospital (Sampa Government Hospital) than the Mission hospital (Dormaa Mission Hospital). Unavailability of technical and management staff was reported as one of the major problems. This problem was because most of the technical staffs were in high University of Ghana http://ugspace.ug.edu.gh 101 demand by the big hospitals. The poor and less developed district hospitals could not put proper strategies in place to maintain their staffs. Respondents stated that long absence of health workers in health facilities caused by frequent transfers and failure to accept postings to the hospital affected the continuity of service provision especially in the remote districts. In both hospitals, the hospital management team had assisted in recruiting support staff which in a way has helped improve eservice delivery in the facilities. However, this comes with its own negative effect as you cannot strictly hold them accountable because they are not highly qualified. In Sampa Government Hospital, there was only one medical doctor, and this made it difficult to run a 24hour emergency service at the hospital. This problem was not entirely different from other categories of health workers at the same hospital. Capacity needs assessments were not carried out before the implementation of the decentralization policy at the district hospital level. Key policy implementers, health workers and core management team of the hospitals were not prepared nor capacitated for their new roles under the health sector decentralization as explained by some interviewees. Reform measures on staffing proposed by the decentralization policy were not implemented partly because of technical capacities and low priority of the MOH. Regular capacity needs assessment is lacking at both the central and the local level. Most of the respondents were aware of the rationale of the decentralization policy. Some of the benefits outlined were: It encourages bottom- up planning, ensure community participation in decision making process and planning, effective and efficient delivery of health services, makes healthcare accessible to the local people in the village and, but the respondents were of the view that the policy is yet to achieve majority of its objectives. University of Ghana http://ugspace.ug.edu.gh 102 Decentralized management of health facilities and involvement of local bodies in planning is still not in place. Bottom-up planning is not in place, and there is lack of new initiatives as per the policy. The planning officer of Dormaa Municipality expressed that: You prepare your budget based on the needs of the local people and somebody in Accra determines what you need, but needs are different from one person to another and you cannot be in Accra and decide the needs of the people in Dormaa, so I think the autonomy that decentralization pledged is missing. In Dormaa Presbyterian Hospital, there was no clear means that was used to ensure that the community members were fully involved in their planning activities, because when asked, means used to engage their various stakeholders in their planning and decisions were not clear. Composition of hospital advisory committee was not fairly represented in the community such as the religious groups, the traditional leaders, and children. The medical Administrator at the Sampa Government Hospital was of the view that: Recruitment of health workers should be decentralized to the hospitals so that we can sack and punish. This is because, when we appoint our own people, we can control and check their output since the power to sack and hire rest on us. Because we cannot hire and fire, most of them fails to put up their best and when you complain, they tell you that, you did not appoint them and you cannot sack them neither. Apart from internal query, no other sanctions can be meted out to them. I hope this is the reason why permanent workers are not giving their best because you cannot control anything about them. This is creating a big problem for us. University of Ghana http://ugspace.ug.edu.gh 103 Proper and full decentralization should give power to local management teams to be in charge of human resource functions but these functions are highly centralized. One respondent added that: ―There are some times where we have to go all the way to Accra to lobby for staff‖. 5.5.5 Inter- Sectoral Coordination Inter- Sectorial coordination was totally absent from the hospital level. To ensure effectiveness of decentralization, there is the need for the various departments such as Agriculture, Finance, and Education, to collaborate to ensure effective health care delivery. Most of the respondents did not know the need for them to collaborate with other sectors .It was clear that there is no effective collaboration mechanism from the top to the district. The planning officer of Jaman-North Assembly stated that: They don‘t include us in their activities and they only consult us when they need our help, meanwhile I think there is the need for education, health and the Assembly to collaborate on a lot of issues. It is mandatory for the hospital to give us their quarterly report, but they don‘t and when you ask them, they will tell you we don‘t have it because they were not able to go round because there is no vehicle to go round . In Dormaa Municipality, though they had the District Health Management Team, as part of the decentralization process, when asked whether they frequently collaborate in planning and decisions, it was not clear even the number of times they meet to discuss issues of common interest in a year. The finance officer at Sampa hospital expressed that: University of Ghana http://ugspace.ug.edu.gh 104 ―Anything goes to the district director, to the Regional and to the National level. We don‘t have anything to do with the Assembly, unless the issue is political‖ The Planning Officer of Dormaa Municipality stated that: What we are practicing now is called de-concentration. De-concentration at the district level is a weak form of decentralization. Every decision is taken from headquarters and if you take any action, it has to be sanctioned from the top. Therefore, I see it as a decentralized system being super- imposed by a centralized system. University of Ghana http://ugspace.ug.edu.gh 105 5.6 Summary of implementation status and major gaps of some selected policy reform measures of decentralization at the hospital level Reform actions Current status Major gaps Human resource and capacity issues Human resource functions are not fully decentralized. Only casual or temporary staffs are recruited at the hospital level. There is an unclear HR policy issues. Lack of comprehensive Human resource (HR) policy to address the needs of decentralization policy objectives, inadequate technical and managerial staff in Health facilities (HFs). Planning and financing Planning of the hospital activities take place annually to endorse central targets to the neglect of local targets. Bottom-up planning and monitoring developed fails to address the needs of decentralized planning. Lack of bottom- up need- based planning practice comprising tools for planning, poor HRM policy, lack of authority on funds as per the needs and finance for local plans. Organizational structure The current structure is highly not sufficient to address the needs of decentralization. Lack of timely restructuring for decentralization lead to unclear roles and responsibilities, and poor decision making practice. Decentralization reform with clear objective directives and implementation Policy lack clear strategy and objectives. Lack of clear policy on health sector decentralization and lack of strategy for implementation Coordination mechanism at central and district levels Coordination between policy implementers at the district level was weak, central and local coordination was not strong as well. Communications were through letters and telephone. Weak intra and inter- sectoral coordination at national and local level created duplication of resources. University of Ghana http://ugspace.ug.edu.gh 106 5.7 Summary: This chapter focused on the analysis of responses from interviews with key informants from the district assemblies, district health directorate and members of the hospital core management team. The chapter brings out the important themes that are in line with the study objectives, and uses verbatim statements from respondents to explain them. The next chapter discusses the results and findings. University of Ghana http://ugspace.ug.edu.gh 107 CHAPTER SIX DISCUSSION OF RESULTS 6.1 Introduction This chapter discusses the findings of the study elucidated in the previous chapter. The purpose is to explain the findings in line with existing literature work on implementation challenges of decentralization at the hospital level. The chapter will further compare the similarities and differences in the implementation challenges of health sector decentralization in the two hospitals and examine how these challenges have constrained or promoted the implementation of this reform in these two hospitals. Important contextual factors that were considered in each district in the process of health sector decentralization, and how these factors have contributed to the challenges discussed in each hospital will also be captured in this chapter. Lastly, the views of respondents on the status of key policy reform measures or some important factors that are responsible for successful implementation of health sector decentralization such as Clear policy objectives and guidelines, Organizational structure, Planning and Financing, Human resource and Capacity issues, Information and communication issues, and Participation will be discussed. 6.2 Implementation challenges The first objective was to identify the implementation challenges of the decentralization policy at the two hospitals and how these challenges affect the implementation of the decentralization policy at the hospital level. The study discovered that there are a lot of challenges to the successful implementation of the decentralization reform at the hospital level which include human resource and capacity issues, lack of communication among University of Ghana http://ugspace.ug.edu.gh 108 key policy implementers, financial constraints, inadequate support from government, and lack of community participation in planning and management of health programs. These challenges are explained below. 6.2.2Human resource and capacity issues Human resource and capacity issues were discovered as a matter of concern throughout the interview. Human resource is the most important aspect of converting available pharmaceuticals, medical technology and preventive health information into a better use for every nation, but the absence of clear human resource policies is one of the major problems identified by the informants. Lack of a decentralized human resource functions is one of the critical issues in the overall management of human resources in Ghana‘s public health system. In the two facilities, human resource constrain was identified by some of the respondents as one of the major setbacks of effective implementation of decentralization at the district hospital level. However, the situation was far better in Dormaa Hospital as compared to Sampa Government Hospital. At Dormaa Presbyterian hospital, there were encouraging numbers of personnel not because of the size of the hospital but due to pragmatic measures put in place to attract and retain health workers. Dormaa private hospital have put up a new Nurses and Midwifery training school where scholarships are awarded to students in areas the hospital has short supply of workforce, especially in midwifery to directly employ those who benefits from their scholarship scheme to solve some of the problems faced in the area of human resource. Casual workers are trained and understudy professional workers so that in the event of schooling, leave, or even death of a professional worker, the hospital can easily replace the person before official postings are done or till the person comes back. This strategy hardly makes them run short of health workforce. Frequent workshops, training and capacity building were organized than their counterparts in the public hospital. Dormaa private hospital turn University of Ghana http://ugspace.ug.edu.gh 109 not to rely too much on government for their workers but rather have incentive packages that help them to retain and attract more workers than Sampa Government Hospital. The improvement in the area of human resource by the private hospital than the public one is because, the process of obtaining workforce and integrating health professionals who have returned from other countries to work in Ghana is easy to be integrated in the private hospitals than the public facilities. This motivates health workers to work in the private hospitals thereby helping solve some of it human resource problems. Dormaa Hospital had three specialists and well trained and competent workers heading their various departments and CHIP compounds. This has made the implementation of the policy somehow smooth as compared to Sampa hospital. There is frequent internal monitoring and evaluation, and people at their CHIP compounds had very good record in terms of health care delivery per the records inspected. This made the main hospital less crowded because there were equally good and competent workers in charge at the sub- district level, and this prevented people from coming from their villages to the main hospital. This also increased accessibility in the rural areas thereby making the decentralization reform efficient at hospital level. Sampa Hospital on the other hand has a serious crisis on human resource. There were little strategic measures put in place to deal with the issues of human resource. The hospital rely solely on government postings for their staff, and occasionally employ few casual workers. Unavailability of technical and management staff due to long absence of health workers in health facilities. This hampered service provision in Sampa hospital. The procurement unit for instance relies heavily on the assistance of experts from Sunyani to help them in their procurement issues. There is only one doctor in the whole facility. Lack of incentives and slow economic activities in Sampa has made most of their staff moved University of Ghana http://ugspace.ug.edu.gh 110 from the hospital in search for better opportunities elsewhere, and others refusing postings to the facility. The problem is more pronounced in the public facility because; management in Sampa hospital has little influence on issues relating to human resources as most of the functions of human resource are centralized. This problem is making the implementation of the policy very difficult because Sampa hospital do not have adequate human resource at the various departments to efficiently manage their own affairs. Due to the scarcity of human resource, few CHIP compounds are established and this is making accessibility very difficult. Crowding at the district hospital has been a major routine. The concern of human resource constraints has been indicated by Narisim-ham et al (2004), that effective delivery of health services is affected by human resource constraints. Inappropriate numbers and the way they are distributed, the type of staff and their performance has been a major issue for most health facilities. An increasing percentage of the workers in the Sampa hospital are not skilled, as there has been continuous desire to recruit casual workers to augment the low numbers of the skilled workers, but this also comes with its negative effects. This finding confirms the work of Dhakal (2009) as he identified qualitative disparity and quantitative mismatch and unequal distribution of health personnel as major problem facing health sector decentralization in Nepal. Comparing the numbers and quality of health workforce in the two hospitals, it was obvious that there is high disparity in terms of quality and quantity in favour of Dormaa Presbyterian Hospital. Whilst Dormaa Presbyterian Hospital can boast of four hundred and fifty (450) clinical staffs with three specialists, Sampa Government Hospital has two hundred (200) clinical staffs with only one physician. Internal migration and failure to accept postings to less endowed districts in terms of social amenities and wealth were two major causes for these disparities as this has been a major problem in most developing University of Ghana http://ugspace.ug.edu.gh 111 countries. This finding is also in line with a research conducted in China, which showed that increasing numbers of low skilled health workers and reduced demand at rural health facilities in part due to the loss of their most experienced personnel was found to be the cause of the falling productivity of health workers in China (Gong & Tang, 2004; Gong et al., 1997; Martineau,). Inadequate and loss of health personnel due to the factors already mentioned above have made both hospitals to train casual workers to augment the few qualified ones. There were 40 casual health workers in Sampa hospital which in a way affected the quality of health care delivered, and this finding is common with what Sakyi (2011) found in Nkwanta North district of Ghana. He concluded that inadequate personnel were among the factors militating against health sector decentralization. Decentralization which is a popular feature of most public sector reform programs is viewed as an opportunity to improve Human resource management, but the current evidence (Tang & Martineau, 2002; Kirkpatrick, Davis& Oliver, 1992; Wang, Collins,) is open to doubt Most of the respondents at Sampa Government Hospital expressed that, measures to ensure capacity building were lacking in the facility. However, all the two facilities realized the importance of the capacity building in ensuring successful implementation of the decentralization policy. This finding confirms the work of Larbi (1998) when he found that district health management teams failed to implement several components of decentralized management because the technical and managerial staffs are incapacitated. In both hospitals, most of the key people in charge of implementing the policy did not have the requisite capacity to effectively carry out their roles well. For instance, some departments had to seek assistance from the regional level before they can carry out their duties well. Inadequate funding to organize capacity building workshops, poor capacity needs assessment at both local and national level were main reasons for the poor capacity at the facilities studied. University of Ghana http://ugspace.ug.edu.gh 112 This result is also consistent with the study of Kawonga, (2003) who showed that capacity problems can seriously affect decentralization reform initiative, and an example was found in Kenya where the shift in the responsibility of drug purchase to the districts was reversed after one year due to the poor performance of the district resulting from poor capacity. In the same way, short of capacity was found to be a major hindrance to AIDS control initiatives, and as a result The Ministry of Health was compelled to maintain a centralized system. However, this finding is uncommon with the study of Gottret &Schieber (2006) as there was no evidence that institutional capacity and policy implementation are correlated in his study of 140 countries on the impact of decentralization on immunization services. This may be so because, those countries may have made capacity building as part of their training in school and so without further institutional capacity, target of reforms can still be achieved unlike the context in which this study took place, capacity building is often enhanced on the job. 6.2.2. Lack of communication among key policy implementers A major situation for successful implementation of every public policy is a function of the quantity and quality of communication. Communication fulfils a lot of functions in policymaking and implementation processes. Not only do policymakers and policy implementing agents like to receive information about a policy from the central level (MOH) and, they also want to communicate what the policy is expected to do and what implementers have to focus their attention on. However, the study revealed that communication issues formed a major hindrance to the implementation of the decentralization in the two hospitals. According to the findings, communication between University of Ghana http://ugspace.ug.edu.gh 113 the key policy implementers in the Districts; that is the assembly, the health directorate, and the hospital in general was problematic. This problem was more pronounced in Dormaa hospital as compared to Sampa Government hospital, because Dormaa hospital considers itself as a CHAG institution, and because of that, do not report to the Municipal Director of Health. In the same vein, the communication between the Hospital and the Assembly was also limited because the assembly is of the view that the hospital is a mission hospital, and for that matter has little to do with government agencies especially those in charge of health related policies in the district. Communication between the Central level (MOH) and the district was also missing. The communication between the central and the district was occasionally based on telephone calls, letters, memos, etc. which was ineffective. This made it difficult for the implementers of the health decentralization at the hospital level to know their respective roles to ensure the successful implementation of the policy because there was always communication gap among the implementers. Hospitals and policy makers should organize community durbars where community members come to meet the hospital authority, and ask questions about their activities .Hospital advisory committee should involve the representatives of the various groups in the community such as the women, various religious groups, and traditional authority among others to ensure easily transmission of information to their groups. Again, policy makers should frequently meet to review the problems, achievements and the way forward of their activities as this will ensure efficient information transmission. District implementers were also not clear as to what the policy entails and how to approach the whole policy implementation. This makes it difficult to achieve smooth implementation because implementers themselves were confused due to unclear policy guidelines and poor communication from the central level. University of Ghana http://ugspace.ug.edu.gh 114 Sampa hospital on the other hand, though the communication between the policy implementers was problematic, it was however better as compared to Dormaa Presbyterian Hospital. There was frequent communication between the district health director and the hospital as he often visits the facility to find out what their needs are, and how possible he could help them. Notwithstanding this, there was unhealthy communication between the hospital management team and the assembly. The effective and constant communication between the district directors of health enables them to work as a team and solve together some of their problems in terms of human resource, and gave them the opportunity to plan together which facilitates the successful implementation of the decentralization policy in the health sector. The finding supports the results of Sakyi (2010), who expressed that staff awareness about health services reforms decisions was very little and even in some cases nothing was received at all on major health management decisions and this was very common among junior staff. Lack of information sharing and collaboration between the key policy implementers and the community members has also been justified by Sakyi (2008) as he stated that, ―communication between policy implementers and other stake holders and community members is very minimal in most developing countries‖. Goggin et al (1990) also confirm the results of this finding when he realized the role of communication in the successful implementation of the decentralization policy by making the assertion that, the more legitimate and credible the message was in the eyes of implementing agency officials, the more likely implementation was to proceed promptly and effectively. Communication in the two hospitals was problematic because, support coming from the assembly and health directorate was minimal. This makes them feel they are independent University of Ghana http://ugspace.ug.edu.gh 115 and that there is no need communicating their activities to them which has created a gap among the key decentralization implementers at the hospital. Again, the means of communication among the policy implementers at the district was very weak. They normally rely on verbal communication and text messages which are weaker means of communication and this affects the seriousness attached to such communication. 6.2.3 Financial constraints Funding constraints was found to be a major challenge of the implementation of the decentralization reform. For the success of this reform at the hospital level, hospitals must be financially sound to be able to carry out effectively their daily activities. In the two facilities, funding was identified as one of the major problems encountered under the implementation of the decentralization policy. Both facilities relied heavily on their internally generated fund (IGF) of which about 95% comes from the health insurance. It was clear from the study that it took so much time for government to reimburse them and this affected their annual and day to day planning of their activities. The two facilities seem not to have any clear means of generating fund internally apart from the health insurance. Donor funds were not forth coming and government hardly gives them money for their activities. Sampa Government Hospital had come out with a plan to expand their mortuary in order to get something small to support their daily operations. The government used to give them some amount to support their activities, but this is no more in place. This problem has affected the hospitals ability to effectively decentralize in some ways; first of all, provision of quality healthcare is affected because there is inadequate fund to purchase quality drugs for treatment. Most of the drugs in these facilities are the drugs for the first line treatment. Health sector decentralization has among University of Ghana http://ugspace.ug.edu.gh 116 its objectives the provision of quality health care to community members, but the problem of funding is making it difficult to achieve this objective in the two facilities. Secondly, effective monitoring and evaluation is also affected due to the problem of inadequate funding which seriously affects the implementation of health sector decentralization at the facility, district and the Central level. This is because inadequate funds make it difficult for management team to organize frequent checks within the facility and sub- facilities to monitor and regulate their activities which was frequently done by a team from the central level. Since district health policy implementers also cry of the same problem, the ability for them to rigorously monitor the health facilities under them is limited owing to weak capacity to generate funds to assist these health facilities especially at the Assembly level. Due to the distance and the busy schedules of health workers at the central level, frequent monitoring is also problematic. This problem has therefore limited the successful implementation of health sector decentralization especially at the hospital level bas people seem to do whatever they like because there is no effective monitoring and evaluation system in place. This in the long run affects effective health care delivery. There was an instance in Sampa Government Hospital, where annual health records were not obtained because there was no vehicle to go round to take data from their sub- district health facilities. The problem of funding also affected the facilities in organizing effective training for their staff, and this has negative influence on performance of health workers. However, ensuring effective health sector decentralization means having well trained and competent staff to handle their own affairs. The issue of funding is in line with the assertions of Couttolenc (2012); Ahwoi (2010), and Sakyi ( 2010) who were in agreement that; in Ghana and in most of the developing countries, inadequate transfers and general insufficiency of funds continue to be a University of Ghana http://ugspace.ug.edu.gh 117 major challenge in implementing policies under the decentralized systems. It was found that the hospitals operated solely on the Internally Generated Fund (IGF), and about 95% of this amount comes from the health insurance with the remaining percentage accruing from mortuary charges, out of pocket payments and donors. Unfortunately, there is a delay in reimbursing the hospitals by the government making it difficult for them to be independent and bringing to a stop most of their planned activities. One respondent was of the view that funding to him has been the major challenge for the implementation of the decentralization policy because the payments of the health insurance do not come as expected. This problem confirms the findings of Couttolenc (2012) as he expressed that there was a substantial delay in transfer and release of fund both by government of Ghana and the NHIs, which affects the functioning of local government and local facilities as well as programs. Fiscal decentralization was found to be lacking as compared to the other forms of decentralization, and Couttolenc (2012) again expressed worry that fiscal decentralization was more apparent than real as over 50% of public health expenditure is allocated to the district level, but the larger part of these resources are allocated and controlled by the central government whether DAs or GHS district Offices and facilities have little real decision power on resource allocation providing services through increased efficiency. The problem of funding is severe in the hospitals studied because apart from the NHIS, there is no strategic measures put in place to generate money both externally and internally to support the running of the hospitals in the event of delay in reimbursement of the NHIS by government. Though Sampa Government Hospital has expanded their mortuary to generate enough revenue, this is not a reliable means of revenue generation as the number of deaths cannot be predicted. The context of this study is not different from the cited studies as most of the studies were carried out in districts of most developing countries University of Ghana http://ugspace.ug.edu.gh 118 especially in Ghana, which is an indication that funding has been a major challenge to the implementation of decentralization policy at the hospital level in most developing countries. However, Scott (2009) holds a contradictory view that funding should not be a problem if decentralization is managed well, since decentralization could open up new frontiers of tax revenue while minimizing the cost of providing services through increased efficiency. This also means that, there is the need for capacity building at the hospital level so that competent people can generate and manage effectively the revenue generated. 6.2.4. Inadequate support from government For every policy to achieve the highest rate of success, government‘s commitment in terms of finance, human resource and capacity issues, infrastructure, clear policy guideline, etc. are very vital. Despite the high recognition of government‘s commitment to public policies in achieving expected outcomes, he has not done enough to support the implementation of the decentralization policy at the hospital level. The policy has not received enough attention in the areas of fiscal decentralization, infrastructure and even the power to take some basic decisions especially in the area of human resource, and these worry were common to all respondents from the two hospitals. There was no clear policy guideline as to what the policy entails and what is expected of them. This has led to frustrations concerning the confusing nature of the whole policy. Even one of the health directors stated that he has not seen any well documented material on the decentralization of the health sector, and all these serve as a barrier to the implementation because one can University of Ghana http://ugspace.ug.edu.gh 119 work and implement a policy well if only he is are aware of what the policy is about. In Sampa Government Hospital for instance, of all the structures put up, only the emergency unit, which was even yet to be completed, was put up by the government. The hospital does not receive any money from the government, but only rely on their IGF. They rely greatly on the health insurance, but reimbursement is not timely which makes their operations very difficult. Health sector decentralization can only be implemented well if there is a clear policy guideline, provision of human resource, capacity building at all levels, finance, etc. provided by government so that decentralized units can be effectively managed independently. However, , all the above mentioned tools for engineering the implementation of health sector decentralization is yet to be experienced, and therefore making the implementation of this reform a major challenge. This finding supports a review by Zenten& Semidei (1996), as they unearth that political, legal, financial, institutional and administrative as well as human resources are the main challenges putting back administrative including decentralization into practice. In Sampa Government Hospital especially, there is inequitable distribution of health workforce as only few personnel are posted to the facility by government as compared to other hospitals. There is only one building put up by government (the emergency unit) which is yet to be complemented. Amount released by government to the facility is ceased. This affects the smooth implementation of the decentralization policy at the hospital level. The issue accounting for this in the study indicates that enough was not done in terms of preparation before the policy was rolled out. Most of the government representatives at the district level are not up to the task assigned to them. Again, structures needed to be in place before decentralization by government are lacking, and this complicates the University of Ghana http://ugspace.ug.edu.gh 120 operations of the decentralized units because resources needed to be in place for effective implementation of the decentralization policy at the hospital level are not in existence. Collins et al (2000) also supported this claim by identifying the role of the state as one of the major hindrance to the success of the decentralization policy in Brazil. Indeed the findings of the study strengthen the evidence that, government support is needed at all levels for any public policy to achieve its objectives. 6.2.5 Community participation in planning and management of heath programs. Community participation in planning and management of health programs was found to be a factor militating against the successful implementation of decentralization at the hospital level. Every successful implementation of policy depends largely on the nature of participation of service beneficiaries during policy formulation and implementation process. Community participation is often described as the major benefit of the decentralization and often assumed to be the regular benefit arising from the decentralization process. Despite this perceived benefit, findings from the research showed that, some of the core members of the hospital management team think that planning and management of health activities are the sole responsibility of health experts, and thus most community representatives are excluded from the planning and management of health programs. Not involving community members in the planning of the hospital activities prevents the views and the needs of the local people to be captured in the planning of the hospital activities, and this may have serious consequences on health care delivery. In the two facilities studied, both of them seem not to include the community members in the planning and management of health problems. They do not have any clear means of University of Ghana http://ugspace.ug.edu.gh 121 engaging community members in their planning and management of health programs. Involvement of non- state actors which include traditional authorities and civil society organizations in decision making is important in strengthening Ghana‘s decentralized democratic governance, but it was realized that there was little space in the formal local government structure for traditional authorities and civil society organizations to engage with MMDAs and hospital management team in decision making and planning of the hospital activities in the two facilities. This problem has constrained the hospitals ability to implement the decentralization policy in two ways; the quality of change expected by decentralizing the health sector is undermined. This is because; the views and experiences of health care beneficiaries or users are not incorporated in the planning and decision making process. This violates the major principle of health decentralization; that is community participation. The interest of community members are undermined by this problem. Secondly, the sense of ownership among service users, health workers and even policy implementers in the district is affected. This is because implementers at the district level are mostly not involved in decision design process. Since, they were not involved in the policy design process; there is some sort of apathy among health workers and key policy implementers in the districts. They see the policy as belonging to the government alone. These problems constrained the successful implementation of the health sector decentralization at all levels. The community participation issue has been stressed by James and Francis (2003) when their study on ―Balancing Rural Poverty Reduction‖ and citizen participation under Uganda‘s decentralization program revealed that decentralization process and structures failed to constitute genuine participation of local government which affected the policy implementation. However, Hutchinson et al (2002) expressed that despite the fact that many communities increased their participation in the health sector through local health University of Ghana http://ugspace.ug.edu.gh 122 committees following decentralization, lack of training and guidelines undermined the effectiveness of these governance structures. The findings of the study show that health sector planning and decision making process should actively involve community members as well as key policy implementers in order to feel some sort of ownership of the whole program as expressed in existing literature. Involving community members in health planning and health services management is problematic according to the findings of the study, because people are not well informed about the activities and programs of the hospital. Due to this, community members fail to actively involve themselves in the planning and managerial issues of the hospital. Community members see the hospital as belonging to the workers at the hospital and this has as a serious implication for implementing the decentralization policy at the hospital level. In addition, the attitude of policy implementers is also an issue of concern. Most of the policy implementers are of the view that issues concerning health are the key responsibilities of the core health management team and that community members have little to do in these areas. As a result, they are not consulted to consider their inputs in the planning and decision making process. In view of this, there is often conflict between the needs of community members and that of policy implementers at the hospital level. 6.3 Important contextual factors taken into consideration before decentralization at the hospital level. There are serious and somewhat neglected issues about whether, and how, national policy can be effectively implemented locally and what needs to be in place for this to occur. Contextual factors are very influential in determining the outcome of decentralization. Health sector decentralization policy and the ever-changing social, geographical, economic and cultural context often generate problems and challenges with regard to University of Ghana http://ugspace.ug.edu.gh 123 capacity, equity, participation, poverty reduction etc. Decentralization is frequently implemented as part of a public sector reform and needs to be adapted to the requirements of the health sector, which requires a broad range of consultation and policy options for its design and implementation. According to Atkinson et al, (2000), decentralization is contextually driven and that what works in one District can fail in another district. This assertion implies that decentralization of the health sector cannot lift what worked in one district to another district without considering some of these factors which are unique to some districts. These contextual factors can either inhibit or promote the success of the decentralization reform at the hospital level. Decentralization is a complex process which involves a wide range of changes, many of which are not easily reversible should they prove wrong. It can also generate a wide range of contradictions and can lead to inequity, political manipulation, higher administrative costs, poor staff management, health system fragmentation and a weakening of health program effectiveness (Lakshminarayanan, 2003; Crook, 2003; Bossert and Beauvais, 2002; Fielder and Suazo, 2002; Kaufman and Jing, 2002; Kivumbiand Kintu, 2002; Patton, 2002; Hardee and Smith, 2000; Mayhew, 2000; Kolehmainen-Aitken, 1999). In short, it is not a policy for those lacking weight in policy analysis, and that its design and implementation calls for thorough consideration of all these contextual factors. Through the study, the following contextual factors were identified in the two district hospitals and these factors either ensured or inhibited the successful implementation of the decentralization reform at the two facilities based on the characteristics of the district in question. The following contextual factors were found to have impact on the implementation of the decentralization policy at the hospital level in the two districts. University of Ghana http://ugspace.ug.edu.gh 124 6.3.1 Socio-economic potential Socio- economic factors can create disparities in economic opportunities and create gaps in income and public service delivery between various jurisdictions. The study revealed that the District of Dormaa, with Dormaa- Ahenkro as its capital is a forest zone and the fertility of the soil ensures the growth of many cash crops grown in Ghana. This has helped reduce the unemployment rate as majority of the people are into agriculture. Most of the people are economically sound, and have invested in many areas in the district. The district is also known to be the leading producer of poultry in the region and hence majority of the population are employed in these areas. This has made the district very endowed in the area of finances since a lot of economic activities take place in the town. Because of this, they are able to provide more of public local services. They are able to generate and support health sector activities internally which ensures the smooth running of the district hospital. Furthermore, because of the economic and busy nature of the District, a whole lot of facilities are available to the population and this serves as an incentive to workers who are posted to the district to work in the hospital. This was evident when three (3) specialists were found in the facility and other cadres of staffs were also at post. The issue of personnel absenteeism or refusal to accept posting was minimized as most of the social amenities found in the big cities were also present in the district. Accommodation problem which is found to be one of the major challenges facing most of the hospitals was not found to be a major issue as some of the local people even provided rooms for the health workers. This has helped the rate of implementation, as community members supported most activities of the hospital and more social services were provided. Human resource which is very critical to the implementation of the decentralization policy was not much of a problem as compared to Sampa government hospital, and this has helped in the implementation of this reform process. University of Ghana http://ugspace.ug.edu.gh 125 Jaman-North with Sampa as its capital on the other hand, is transitional savanna vegetation. The major cash crop in the district is cashew. Economic activities here are very slow as few people are absorbed in this area of farming. There are no major economic activities apart from small scale trading. This renders majority of the people poor. Hardly do people contribute to the development of the hospital, and majority of the populace seldom report to hospital because of poverty which makes equity and accessibility a problem in the district. Secondly, most health workers refused to work in the district because of the poor nature and lack of essential social amenities in the District. This has made the scarcity of health workers the norm in the hospital. For instance, in the whole facility, there was only one medical doctor, and facilities are inadequate. The few ones do not meet the standard of a normal hospital as compared to the Dormaa Presbyterian Hospital. This problem has indeed slowed the process of implementing the decentralization reform at the hospital comparatively. This finding is in line with the assertion of Akramov & Asante(2009), who were of the view that, One of the potential risks with decentralization is that it can lead to greater inter-jurisdictional disparities due to the differences in socioeconomic potential and expenditure needs of sub-national governments. The argument is that sub-national governments with better factor endowments and potential will have a larger revenue base than the other poorer sub-national governments and therefore will be able to provide more local public services. This may cause disparities in economic opportunities and create gaps in income and public service delivery between various jurisdictions. The district of Dormaa is endowed with natural resources and the fertility of the land supports varieties of crops. Most of the people are employed and this has made it possible to raise revenue University of Ghana http://ugspace.ug.edu.gh 126 locally to support the activities of the hospital in terms of human resource, facilities, purchase of equipment, sponsoring of workshops, capacity building and many more, which are all necessary for the successful implementation of the reform. This economic advantage has made the implementation in Dormaa Presbyterian Hospital better as compared to Sampa Government hospital. On the other hand, Sampa Governmental Hospital is located in a district less endowed in natural resources with little social amenities. The district is poor compared to Dormaa district, and in view of this, the few health personnel in this facilities are lost to the hospitals in the wealthy districts. Little revenue is generated internally to educate, build capacity, motivate and retain workers. Decentralization works well when well experienced workers are recruited to man the decentralized units. However, the situation is different in Sampa hospital and this has negatively slowed the pace of implementation of this policy at the hospital. In supporting this findings, Prud‘homme (1995) considers decentralization as the pure decentralization as viewed in earlier fiscal federalism literature; that is, local governments fund local public services with their own local revenues without transfers from the central government. However, evidence suggests that intergovernmental transfers, such as equalization grants, can play an important role in overcoming imbalances between revenue capacities and expenditure needs of local governments (Martinez-Vazquez and Searle 2007).Conversely, some authors have questioned the effectiveness of such equalization transfers. On one hand, equalization transfers could give disadvantaged districts the scope they need for investments in public infrastructure and services. On the other hand, it is not clear whether they use these transfers effectively (Feld and Dede 2005). It is also possible University of Ghana http://ugspace.ug.edu.gh 127 that instead of investing in public services, these transfers are used for consumption and rent-seeking activities. Nonetheless, competent personnel and capacity building may be needed before these intergovernmental transfers are put to good use, though there was no evidence of such equalization plan in place to balance these inter-jurisdictional disparities. 6.3.2 Unequal endowment of local capacity The capacity to implement decentralization policies continues to be under scrutiny. This is due to the realization that, despite perceived huge investments, local government performance in many African countries continues to be a matter of concern. Capacity means having the aptitudes, resources, relationships and facilitating conditions that are necessary to act effectively to achieve some intended purpose. In operational terms therefore, capacity building is the process through which the abilities to achieve set goals are obtained, strengthened, adopted and maintained over time. Comparing the two Districts in terms of local capacity, Jaman-North District is less endowed with respect to local capacity than Dormaa district. Some of the policy implementers at Dormaa hospital have very good educational background as about five of them have their master‘s degree and quite a good number of them also pursuing different postgraduate degrees in their areas of work. This has made financial, administrative and managerial capacity at the hospital more efficient than Sampa hospital. Furthermore, the research revealed that, majority of the population have at least basic education and this makes it easier for the local people to read, write and understand the purpose of any health policy. This in the long run enhances community support and participation as they are aware and understand the purpose of the policy. Infrastructure is not of a major problem though not adequate, partly because of the competent nature of University of Ghana http://ugspace.ug.edu.gh 128 their management team. They are able to come out with appealing proposals to get support from donors. For instance, the hospital is putting up an ultra- modern eye center which is solely financed by various foreign donors. They have also established a nurses and midwifery training school to help solve some of their human resource problems at the hospital without necessarily waiting on the government to provide all these facilities. Most of the effort to put up the training school was initiated by the community itself because they had the financial capacity to do it. Dormaa District is also known to be the home of most scholars from the Brong – Ahafo region. This has made the district to have a fair share of representatives in government, and in terms of lobbying for public goods and services; they are able to get it quickly. This has in a way facilitated the implementation of the decentralization policy in the district. On the contrary, because the Jaman-North District is one of the newly created districts, majority of the policy implementers are not experienced. Most of them are fresh graduates from tertiary institutions especially at the district assembly. As a result of this, financial, administrative and managerial capacity needed for effective implementation of the Decentralization at the hospital level is low compared to Dormaa hospital. The educational level of most of the people in Jaman-North District is low, and for that matter getting the commitment and support of the community is very difficult. Majority of the population are not aware of their roles in implementing the decentralization policy. They think management of the hospital is the sole responsibility of the health workers, but improved education will help get rid of some of these prejudiced thinking. This will consequently bring about community participation and support. It was realized that about twenty percent (20%) of health workers posted to Sampa hospital refuse posting because of poor nature of the district in terms social amenities and other factors. University of Ghana http://ugspace.ug.edu.gh 129 The issue of unequal endowment of local capacity has been discussed in the literature by Collier (2008) and Crook (2003) as a factor militating against the successful implementation of the Decentralization policy. They expressed that factors such as unequal endowment of local capacity and the politics of local central relations could seriously affect the successful implementation of the decentralization policy at the hospital level. Government accountability and allocative efficiency may not be achieved with decentralization when managerial capacity is more problematic at the lower levels of government. However, evidence suggests that lack of resources is not necessarily the problem. Douda (2006) using examples from South Africa and Uganda expressed that developing political capacity for demanding accountability for existing resources is more important. The issue of capacity building is problematic in the study areas because policy implementers fail to do proper capacity needs assessment. Failure to do proper capacity needs assessment means capacity building is bound to fail because they are not able to identify where they fall short and where to build capacity. The financially sound districts are able to train and develop the capacity of their workers. Some even sponsor some of their workers to higher levels of education to gain more experience. These experienced workers and their organization find it difficult to release these workers to share their experience with new hospitals, thereby creating the gap of unequal local capacity endowment in terms human resources and finance. 6.3.3 Geographical factors Geographic locations play a major role in shaping disparities in access to local public services in decentralized districts in Ghana. Geography is considered as a deeper factor which affects both local services and population density. Geography relates to the advantages and disadvantages created by districts‘ physical locations and agro-ecological University of Ghana http://ugspace.ug.edu.gh 130 conditions. Geography might also directly influence the demand for and supply of local public services. Different types of countryside might have an advantage for different types of local public goods. For example, supply of drinking water is highly dependent on the availability of surface and ground water sources in a given geographic area. The evidence also suggests that areas closer to large metropolitan areas tend to have higher population density due to agglomeration effects (World Bank 2008).Further, geography may influence local public services through other factors such as institutions. As Eichengreen (1998), Engerman, Haber, and Sokoloff (2000), and Sokoloff and Engerman (2000) have shown, the initial factor endowments to a large extent explain income, human capital, and political power inequalities; these inequalities in turn explain the structure and functioning of institutions that insure the persistence of inequalities. These findings were not different from the findings of this study. In terms of proximity, Dormaa Municipality is closer to Sunyani which is the regional capital. In view of this, the district tends to benefit in terms of access to local public services as compared to Jaman – North which is farther away from the regional capital. Similarly, the proximity of the District to the regional capital has facilitated access to reliable information. That is, because the District is closer to Sunyani, information with regards to health and other health policies are better accessed by the hospital than Sampa hospital. Due to this, the hospital is abreast with current information with regards to their operations and current health policies which is very vital for smooth operation of the decentralization policy. On the contrary, the employment rate in Jaman- North is very low, and because the soil does not support most of the major cash crops grown in Ghana, there is less incentive for people to move to these areas hence less population density and unemployment. This leads University of Ghana http://ugspace.ug.edu.gh 131 to reduced internal revenue generation which seriously affects the decentralization policy. The district is also far from the regional capital. This at times limits their access to local public services which serve as major tool for the operation of the decentralization reform. These geographical factors have rendered the district disadvantaged in terms of implementing the reform at the hospital level, especially in the area of human resource where people are not willing to accept postings to areas with less social amenities and employment opportunities. According to Bilodeau & Leduc (2003), one major factor affecting the retention of health personnel in rural and remote areas is contextual or environmental factors including community amenities, quality of life, Population‘s educational level, etc. All these factors shape the success rates of the decentralization policy in different district. One major cause of this problem at the Sampa hospital is because, the district is poor in social amenities, and standard of living is very low. These had served as disincentive for health workforces who are very instrumental in the decentralization process to work in the Sampa hospital. Municipality is a forest zone and specializes in industrial crop such as cocoa, which serves as the source of income and livelihood for about 25% of Ghana‘s population. This attracts a lot of people to the district thereby resulting in higher population density. A district with higher population density will benefit from more local public services. The higher population also means that the district will be able to generate more internal revenue to provide most of their needs. Once there is adequate fiscal capacity, the district can support its health facility without relying solely on the government for the provision of goods and services. Being self- reliant ensures less influence from the central government, thereby ensuring that the needs of the local people are taken into consideration. This ensures effective and efficient operation of the University of Ghana http://ugspace.ug.edu.gh 132 decentralization policy. This advantage of Dormaa Municipality has really made the implementation of the decentralization more effective than Sampa Government Hospital. The issue of how geographical factors shape disparities in decentralized Districts is evident in the work of Akramov and Asante (2009). The study showed that there is a strong correlation between geographical factors and access to local public goods and services, and that Districts closer to regional capitals have greater access to local public goods and services. Also, districts located in coastal and forest zones were likely to have better access to local public services. Again they opined that, the districts‘ geographic locations are strongly correlated with district characteristics such as average literacy and poverty rate. They further expressed that, geography has significant correlations with the districts‘ population density, access to information, and fiscal capacity. This findings support the evidence of this study as due to the distance from the regional capital, access to public goods and services that cannot be obtained from the district are difficult to get from the regional capital because of distance involved, and all these affect the quality of life in the Jaman-North District. However, the opposite is holds for Dormaa Municipality. 6.3.4Cultural factors and ethnic diversity In recent years, there has been increasing interest in the impact of ethnic diversity on the preferences for public services. The literature suggests that preferences for public services and ethnic origins are strongly correlated. Different ethnic groups tend to have different tastes and preferences for local public services, and thus, heterogeneity of preferences across ethnic groups in a jurisdiction is likely to influence the amount and type of public goods the jurisdiction provides (Vigdor 2004; Alesinaet al., 1999).Furthermore, literature also suggests that participation in social activities tends to be significantly lower in more fragmented localities, leading to lower social capital and weak local institutions. As a University of Ghana http://ugspace.ug.edu.gh 133 result, the incapability of weak local institutions to impose social sanctions in diverse communities leads to collective action failures. Diverse communities thus face higher coordination costs in provision of local public services (Miguel and Gugerty 2005). Though evidence suggests that there is a link between ethnic diversity and community participation as well as preference for public services, the study did not find how this factor facilitated or inhibited the implementation of the decentralization reform at the two hospitals studied. Despite the fact that the two districts were homogeneous, there was no evidence indicating how it has speed up or slowed the decentralization process at the two hospitals. Although people migrated to the two districts, especially in Dormaa, it was mainly within that same region and hence the impact of ethnic diversity on implementing decentralization at the hospital level was not felt. Decentralization is seen as a means of ensuring community participation and empowerment to bring about development and fair distribution of limited resources in our various Districts to reduce poverty. However, despite these laudable objectives of the reform, cultural issues pertaining to a particular district can affect the implementation of the policy at the hospital level which can either enhance or prohibit these objectives. In the Jaman- North District, it was realized that apathy was very high among the local people. They see the running and everything about the hospital as belonging to the health workers. They normally feel that the management of the hospital is the sole responsibility of the health workers. Frequent participation in the activities of the hospital was very poor as most of them were not aware of how their efforts were crucial in achieving a better health system. In addition, because majority of the populace are Muslims, women participation especially in areas of policy discussions, and other issues relating to health was low. This seriously affects the implementation of the decentralization at this hospital, since women who play an active role in issues relating to health are normally relegated to University of Ghana http://ugspace.ug.edu.gh 134 the background in this community. It also inhibits the consideration of local needs when planning. Community participation is also limited because people see the health facility as belonging to the health workers only. However, the key instrument for ensuring effective and efficient Decentralization policy at the hospital level is community participation at all levels of the reform. Contrary, cultural factors were not found to play any major role in implementing the reform at the Dormaa Presbyterian Hospital. Though community participation was better as compared to Jaman North, it was not predictable. This was not found to be as a result of cultural factor(s) within the district. The role of cultural factors in the implementation of the decentralization policy was confirmed in the work of Rotha and Vannarith [n.d] when they found that, the introduction and implementation of reform was faced with several cultural factors embedded in the Cambodian society. These factors included patron- client relationships, power distance or unequal power distribution, lack of trust between the local residents and local leaders, collectivism and gender discrimination. The findings of the study contradicts the assertion of Rotha and Vannarith as such cultural factors were not found to have any influence on the implementation of the decentralization policy at the hospital level. This contradiction may be as a result of Cambodian society having different ethnic groups as compared to the context of this study where there were few ethnic groups hence making its impact on the policy insignificant. Again, the culture and the way of life of Cambodians may be entirely different from that of the districts studied in Ghana; this can bring about differences in findings as cultural factors embedded in one community may vary from another. University of Ghana http://ugspace.ug.edu.gh 135 6.4Implementation status of some key decentralization reform measures The third objective of the study was to ascertain from key policy implementers of health sector Decentralization the implementation status of some key policy measures necessary for the successful implementation of the Decentralization policy at the hospital level. The analysis of the key reform actions for Decentralization indicated that majority of them were having incomplete implementation or were either never initiated. The data gathered from key informants and document analysis identified gaps that were mostly related to areas of policy issues and objectives, organizational reform, planning, financing and, human resource and capacity issues, and inter- sectorial coordination. 6.4.1 Policy issues and objectives The ability to get policies and decisions implemented well is becoming uncommon especially in most developing countries. This is mostly as a result of a policy imposed from the center with no consideration as to how the policy might be received at the local level. Most of the key policy implementers at the District level were not consulted in the policy formulation process, and because of this most of the respondents were confused about the whole objectives of the reform. Some of the key informants opined that health specific goals were conflicting with each other. The policy is also lacking clarity on content and there is no clear strategy available to implementers on how to go about the implementation process. Dissemination of information to the District health policy implementers and other stakeholders was lacking and this resulted in a form of confusion and delay in implementation. Policy was also weak in addressing intra- District variations such as communication, geographical difficulties, language, etc. during implementation. With regard to policy issues and objectives, most of the key implementers at Sampa hospital had limited knowledge on the content and overall objectives of the reform. Some University of Ghana http://ugspace.ug.edu.gh 136 of the responds asked the intention of the Decentralization policy and others expressed worry that, the policy was not clear to them. There was no clear policy document detailing what the policy entails. The policy is a top- down issue where everything is planned centrally and carried to the local implementers without necessarily educating them about the policy. This has slowed the implementation process in that; the people in charge of the implementation know less about the whole policy. Community support is also limited because implementers are not able to explain the objectives and benefits of the policy to their people. Though the problem was not entirely different from Dormaa hospital, policy objectives were known to most of the policy implementers, but respondents were not clear about whether they or their organization was consulted because they could not recall what actually happened. Content of the policy and how they engage community members and other stakeholders in information sharing was considerably better than Sampa hospital. This has in a way help the implementation at the hospital because respondents were aware of what the policy is about and what is required of them to implement the policy. However, information about the policy was generally low in the two facilities. This study is line with the study of Brinkerhoff &Leighton (2002) when they observed that tensions and conflicts among objectives often resulted in the failure of most Decentralization reforms. The major cause of this problem at the areas of the study was lack of knowledge and the intention of the overall health sector decentralization. The key policy implementers were not aware of the objectives and the rationale of the policy. Policy implementers can implement a policy well if they fully understand the objectives and purpose of the policy in question. Respective roles of individuals heading the decentralized units were found to be confusing among workers. There was no documented University of Ghana http://ugspace.ug.edu.gh 137 material to serve as a road map for implementing the policy at the district level and this has hampered the implementation of the policy among the policy implementers in the district. This means that policy is designed centrally and imposed on the local people without knowing the purpose, objectives and benefits of the reform. This finding equally explains what Barret and Fudge (1981) revealed by explaining that, at times policy failure or implementation gap may occur as a result of policy imposed from the center with no consideration as to how the policy might be received at the local level. It is not a case of bottom up approaches to policy and action being preferable to top down; however, a balance between the two is necessary. This has been a major characteristic in most developing countries as information about policies to both implementers and people at the community level is very low thereby leading to major implementation problems. 6.4.2Organisational reform There were some slight changes in the organizational structure of the health facilities, but respondents were unable tell whether these changes were as a result of the Decentralization reform. At Sampa hospital, some respondents explained that, they did not have Health administrator as well as purchasing and supply officer previously, but currently the hospital has these personnel. However, it was unclear as to whether the change was as a result of the Decentralization reform. The position of the Administrator especially has relieved some of the burdens of the Medical director as he can now concentrate more on the clinical aspect whilst the Administrator takes care of the non- clinical duties of the hospital. This in a way has speed up the implementation of the Decentralization as roles are even further decentralized within the facility. University of Ghana http://ugspace.ug.edu.gh 138 In a nut shell, there was no right structure to coordinate the activities at both the central and the District level (see table 5.6), and achieving the objectives of health sector decentralization with this current situation will be much difficult. Respondents were confused with the regional level structure as they were not clear about which issues to report to the regional level and which ones should go to the central level. The two facilities were almost working with the old system with little modification. There is no timely restructuring of the organizations involved by government at the lower level as well as capacity for implementation. Organizational reform as a means of ensuring effective implementation of health sector Decentralization has been expressed by many scholars in the literature. For instance Sakyi ( 2007) opined that Health sector reform implementation, must take into consideration the existing organizational structures and culture of the Ministry of Health in order to enhance the chances of successful implementation. Dhakal (2009) also supported this claim by saying that for any effective implementation of health sector reform policy, reorganization and restructuring of the Ministry of Health is a crucial step and such processes should be based on the functional analysis of the structure to achieve the desired objectives. Lacks of restructuring and reorganization have complicated the implementation of the policy in the study areas, as the introduction of the decentralization policy is not accompanied by effective restructuring and reorganization of the health system. The rather confusing organizational structure is still in place and functions supposed to perform at the district level are still at the central level. Facilities and human resources needed to engineer the reform measures are lacking, which is making the implementation difficult at the decentralized units. In most developing countries restructuring of the Ministry is more often than not driven by political motives rather than functional analysis of the health system. This is also partially true for the reason that most of the reorganization is influenced by the support from donors (Jeppsson University of Ghana http://ugspace.ug.edu.gh 139 et al., 2003), and they determine how to organize and reform our health system at the expense of our intended objectives and motives of the reform. This has been a major setback for successful implementation of the decentralization policy at the hospital level. 6.4.3Planning and financing Districts vary widely according to the needs of their population, and even more so in terms of existing interventions and available resources. Strategies, therefore, must be district- specific, not only because health needs vary, but also because people's perceptions and capacities to intervene and implement programs vary. However in most cases, centrally designed plans have little scope for such adaptation and contextualization, hence decentralized planning becomes a very crucial element in the Decentralization process (Gopal&, Mondal, 2007).Periodic plans prepared by the Districts are not acknowledged by the center. Most of the plans of the Districts were based on the targets set by the Central level. Despite the crucial role played by planning in the Decentralization of the health sector, majority of the community members and key stakeholders were not involved in the planning process. What was even worse was the issue of the central level determining what should be the needs of a given District, forgetting that community needs may be different from what central planners may think of. Most of the planned activities at the Districts level were subject to Centralized planning to the neglect of the preference of the local people. This often resulted in misplaced priorities at the District level. The situation was worse at Dormaa hospital among the local planners. This was due to the fact that, the hospital is CHAG facility and often sees itself as separate from the Assembly and the District health directorate. Some of the policy implementers at the hospital level explained University of Ghana http://ugspace.ug.edu.gh 140 that, they were not consulted in the annual planning activities of the hospital. This issue has created a lot of problems as most of the plans did not reflect the needs of the community at the District level. In the same way, the annually planned activities at the District were not followed as the Central planners dictate what should be planned for the year. This makes it difficult for community needs to be pursued, and therefore not ensuring independent planning at the District level which makes the practice of Decentralization very difficult. Though the problem was not different at Samapa hospital, planning at the District level was better coordinated than at the Dormaa Hospital. This was so because the hospital is a government hospital and coordinating with the Assembly and the health directorate was easily done. This in a way ensured effective planning as key policy implementers from different sectors were involved in the planning process. However, one key factor for the central influence on the planning was lack of capacity to plan efficiently at the District level. This finding is in line with the work of Conyers (1990) when he indicated some typical reasons for decentralization as a planning and development tool: ―increase popular participation in planning and development; make plans more relevant to local needs, facilitate coordinated or integrated planning; increase speed and flexibility of decision making; generate additional resources; and encourage more efficient use of existing resources. However, Oyugi (2000) argues that the merits mentioned above are‖ claims and expectations and not hard facts‖. He continued to posit that these outcomes depend on conditions relating to real power sharing and meaningful participation and without these, the effects may encounter the objectives. In most cases, the success of decentralized planning is a function of the extent to which administrative and implementation support was provided by the central ministry. According to (Dhakal, 2006), lack of capacity of University of Ghana http://ugspace.ug.edu.gh 141 central planning team may create mistrust and planning processes at the district level. For this reason, capacity building must be a necessary tool in ensuring Decentralized planning and the frequent influence of the Central level. However, getting competent people to effectively plan to reflect the needs of the community was problematic, and this resulted in the frequent interference of the central level who are not well informed about the needs of the community. A Conflict between planned activities of the district and the central level is a major problem that needs attention if decentralization policy at the hospital is to be implemented and work well. On the other hand, finance which serves as a life blood for the survival and implementation of policies was found to be a major obstacle for the implementation of the Decentralization policy at the hospital level. For hospitals to effectively carry out their duties, they should be financially sound and sometimes financially independent. Strategies to mobilize more resources to finance decentralization process are very vital for the success of the reform. In this case, most hospitals will be independent and autonomous in their finances which serve as a major barrier to many policy reforms. This also means that financial resource will be channeled to areas of needs to society. However, this was not the case at the facilities studied. Funds released from government to these facilities were not forthcoming. They rely solely on their Internally Generated Funds (IGF) which was generally generated from the National Health Insurance Scheme (NHIS). The release of funds or reimbursement of the hospitals was problematic. It took a lengthy time for the government to pay or reimburse the hospitals which slowed their operations. Lack of adequate funds was one of the issues affecting the implementation of local plans. Implementations of planned activities are delayed for almost a year because of untimely reimbursement of the facilities and this affects the smooth running of the hospitals. Though some respondents partly agreed that financial transparency at the local level has University of Ghana http://ugspace.ug.edu.gh 142 improved, strategies to generate money locally apart from the health insurance was lacking. More often than not, monies received from donors were invested in areas that are of interest to them at the expense of community needs. Untimely release of funds to the hospitals was a common problem for the two hospitals. This has affected the implementation of the reform at the two hospitals, in that the amount needed to respond to the health needs of the people in real time is not available. Essential drugs are not available due to lack of fund. Monitoring and proper supervision are mostly not in place both at the local and the central level as inadequate financial resources are always cited as a major problem for monitoring and evaluation. This has seriously affected the implementation of the Decentralization at the hospital level since personnel to supervise, monitor and evaluate the reform to know the challenges and benefits are not available. The financial challenge to the implementation of health sector decentralization has been expressed by Cuttolenc (2012), as he explained that in Ghana there is a substantial delay in transfer and release of funds, both by government of Ghana and NHIS. This has hampered the functioning of local governments and local facilities and programs alike. He also realized that the frame work for financing the local government is very complex and confusing ; that is different sources of funding specialized in financing a specific items or programs, and the DAs resource allocation to sectors is not transparent. In addition, fiscal decentralization in Ghana is more apparent than real: over 50 percent of public health expenditure is allocated to the district level, but the larger part of these resources are allocated and controlled by the central government; local authorities ,whether DAs or GHS District Offices and facilities have little real decision power on resource allocation. This problem is due to the fact that capacity at District level is low and that central level managers do not have the trust in the local managers to efficiently and University of Ghana http://ugspace.ug.edu.gh 143 effectively manage the financial resources independently. This problem has also been highlighted in the work of Bossert & Beauvais (2002) when they examined the decentralization experiences ranging from devolution to delegation in Ghana, Uganda, Zambia and Philippines. In all countries, health expenditure increased at the local level and decreased at the central level as a result of decentralization reform. Despite this higher spending at the local level, there was no significant increase in revenue generation at the local level but rather from increased transfers from the central government. This confirms the fact that capacity to generate revenue locally is problematic in most of the Districts and this has been a major problem for the hospitals studied as they rely solely on the NHIS, and in a situation where they are not released on time, there is no revenue to lay hands on to ensure the operation of the hospital. This problem has forced most hospitals to reject the NHIS in their facilities and this was not different in Dormaa hospital. 6.4.4 Human resource and capacity issues Unavailability of technical and management staff was found to be a major problem posing a lot of challenges to the implementation of the decentralization reform at the hospital level. Human resource issues which were believed to be solved by Decentralization are open to doubt. Respondents commented that human resource is run under a centralized mechanism as shown in table 5.6. The reform measures that were proposed by Decentralization and Human resource department at the central level concerning staffing are not implemented due to lack of local capacity at the District level. Human resource functions are highly centralized and very little decision space is left to the Districts. Reform measures needed to ensure personnel management were not in place in the Districts. University of Ghana http://ugspace.ug.edu.gh 144 Capacity needs assessment was not adequately put in place before moving to the implementation of the reform. Some of the staff are lacking both managerial and technical skills and some of the weak capacity areas included need- based planning, management and accounting. Due to inadequate health personnel in the facilities caused mainly by non- acceptance of postings and other factors hampered continuity of service especially in the less developed Districts. In the two facilities studied, they often recruited support staff which helped in strengthening management and service delivery. These casual workers constituted about 30% and 20% at Sampa and Dormaa hospitals respectively. According to the respondents, most functions of the Human resource are maintained at the central level and this makes it difficult for local managers to control their workers within their facilities since they have little control over them. Respondents complained that there are no clear policies on human resource issues, and as such were confused as to which functions are decentralized to the Districts and those that are maintained at the central level. Inadequate health personnel have seriously affected the implementation of decentralization at the hospital level. This is so because you need people to effectively carry out day-to-day activities to ensure the continuity of service delivery. However, in Sampa hospital for instance there is only one medical doctor who is also the medical director. In the case of absence due to ill-health or when he travels, there is no doctor to take care of patients, and this is preventing people to get the necessary health care they are supposed to get. This condition applies to all class of health workers ranging from nurses, lab technicians, etc. Though the condition is not entirely different from Dormaa Presbyterian hospital, the problem caused by inadequate appropriate measures taken on human resource issues is not as intense as Sampa hospital. In terms of numbers, the numerical strength far outweighs University of Ghana http://ugspace.ug.edu.gh 145 that of Sampa hospital because they are able to sponsor some of the students in their nursing training school and recruit them afterwards. This has enabled them to have personnel to fill most of the needed positions, thereby enhancing quality and continuity of health care. This has enhanced the implementation of the reform as most of the crucial personnel needed for effective running of the facility are fairly represented. However, issues regarding limited control on staff, unclear policies concerning human resource issues were not different from Sampa hospital. The relationship between health workforce and health care delivery has been established by Narasim-ham etal. ( 2004). They opined that effective delivery of health services is seriously affected by human resources constraints .The major concerns has been inappropriate numbers and types of staff and the way they are been distributed, as well as the performance of the staff. Dhakal ( 2009) also identified similar problems as he expressed that, quantitative mismatch, Qualitative disparity, unequal distribution and lack of coordination between HRM functions and health policy objectives have been the key issues, and that decentralization reform cannot succeed if these issues are not addressed in a timely manner. The findings of the research is also in line with the evidence of Gong et al., (1997); Martineau, Gong & Tang (2004) which revealed that increasing number of low-skilled health workers and the fall of demand at rural health facilities in part was found to be the cause of the falling productivity of health workers in China. Appropriate HRM policies and practices can improve human resource (HR) outcomes and consequently lead to the effectiveness of the workforce, which in the long run brings about improved organizational performance. HRM policies and practices that are aligned with appropriate health service objectives, improvement in HRM would likely improve health outcomes. University of Ghana http://ugspace.ug.edu.gh 146 The literature supports the finding of this study as the small numbers of health personnel in the two hospitals; especially Sampa hospital has affected health care delivery. Lack of personnel at Sampa hospital has made it impossible for the hospital to run a twenty-four (24) hour emergency service. This is because the few workers who come during the day find it difficult to come during the night as well. In the case of any emergency at night, the person must be referred to a different hospital. Low- skilled health workers are often found in these facilities because the most experienced ones are difficult to maintain as there are no attractive incentives to maintain them. The over reliance on casual workers to supplement the efforts of few health workers is gradually reducing the quality of care offered in these facilities. Health workers are inequitably distributed to the disadvantage of the less endowed hospitals. This makes the implementation of the decentralization at the hospital difficult especially in the less endowed facilities. Financial constraints have inhibited the hospitals ability to embark on effective capacity building for the few workers available. The issue of human resource has limited most developing countries ability to effectively implement health sector decentralization in Africa. 6.4.5 Inter- sectoral coordination Governance was previously seen as collaboration with only inter-governmental agencies, but involving NGOs, citizens‘ movements, trans-national corporations, academia, and the mass media is very crucial. Ensuring collaboration between central and the local government, non- state actors and international organizations has become an effective means of achieving the goals of Decentralization. This also broadens and strengthens the reform. Increasing communication with other agencies is necessary to ensure enabling regulations or sub-policies critical to implementation of the agency‘s tasks. Since University of Ghana http://ugspace.ug.edu.gh 147 successful actions by one agency may be a function of the implementation of complementary actions by other agencies, there will be greater need for sharing information and resources, and more concerted coordination. However, the findings indicated that there is not enough communication between inter- governmental agencies and other non- state institutions to effectively carry out the Decentralization policy at the District level. Collaboration between local government agencies in the District such as Agriculture, Education, etc., is very poor. Even communication among the key policy implementers was very poor. Some respondents at the hospital explained that key policy makers at other sectors often refuse to accept their invitation to some of their programs as well as health related issues. This at times happens because they give low priority to health related issues and concentrate more on issues within their area of work. This finding supports the work of Gilson & Travis (1997), when they realized that in Uganda, when decentralization occurred beyond the health sector, other sectors in the local government failed to recognize health as priority issue. Even some of the respondents at the District assemblies complained about lack of collaboration among agencies both at the District and the central level. This is impeding the progress of health sector decentralization since the efforts and roles of other sectors are instrumental in achieving health sector Decentralization. The role that local- relation and inter-sectorial collaboration play for successful implementation of decentralization has been expressed by many scholars like Mills et al., (1990); Bossert (1990); Green (2001); and Collins et al.( 2003). University of Ghana http://ugspace.ug.edu.gh 148 Lack of inter-sectorial collaboration among other agencies was more intense at Dormaa hospital. This comes as a result of the fact that, the facility belongs to CHAG, and so have little to do with other agencies especially government institutions. Lack of support from the other organizations in the District in matters relating to health has been the outcome of this problem. This has made the implementation difficult as successful implementation of health sector policies and plans involve the active participation of other sectors. Though the problem is common at both facilities, the collaboration between Sampa hospital and other government and non- government institutions were somehow better. In some of their programs, they even involved the education and the agriculture units to participate and suggest means of helping them achieve their targets. This was explained to have positive impact on most of their programs especially in education when the focus is on school children. Though the situation was better at the public facility, the overall collaboration of the facilities with other agencies was one major problem facing the implementation of the policy at the two hospitals. 6.5 Conclusion The chapter discussed the findings of the study in chapter four. The chapter compared the findings with existing literature and made comparisons and contrasts in the implementation challenges of Decentralization at the hospital level in the health literature. The study revealed that, to solve the challenges of Decentralization at the hospital level, human resource and capacity needs must be addressed, financial, inter-sectorial collaboration, adequate support from government, and active participation of community members are vital for successful implementation of the Decentralization reform at the hospital level. University of Ghana http://ugspace.ug.edu.gh 149 CHAPTER SEVEN SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 7.1 Introduction This chapter summarizes the findings of the study based on the data collected and presents the research conclusion as well as recommendationsfor policy and future studies. 7.2Summary of Findings The overall objective of this research was to explore the implementation challenges of decentralization policy at the hospital level. The study also sought to examine the important contextual factors that were considered in the implementation of health sector decentralization, and finally assessed the implementation status of some key reform measures needed to be in place for successful implementation of the policy at the hospital level. Data were collected using the qualitative research technique employing various research tools such as key informants interviews, observation and documentation of analysis. Interview guides were used to collect data from key informants in order to get an in-depth understanding of the implementation challenges of health sector decentralization at the hospital level. Data were analyzed thematically, which involved reviewing and grouping the data, making notes and sorting the data into categories based on the objectives of the study. It allowed the researcher to move the analysis from a broad reading of data towards discovering patterns and developing themes. Finally, the summary of the data analyzed was then provided. Findings from the study indicate that inadequate human resources and capacity issues, lack of communication among key policy implementers, financial constraints, inadequate support from government, and non- involvement of community members in planning and management of health programs were the major issues serving as barriers to the University of Ghana http://ugspace.ug.edu.gh 150 implementation of the health sector decentralization at the hospital level. Though the problems identified cut across the hospitals studied, most of them were more intense in the government hospital in Sampa compared to the mission hospital in Dormaa. The findings also revealed that contextual factors or differences in Districts in terms of socio-economic potential, unequal endowment of local capacity, geographical factors, ethnic diversity and cultural factors were not taken into consideration before implementation of the health sector decentralization policy. Majority of the respondents explained that these factors were not considered and that everything was designed centrally and brought to the district level, and this has brought about a large disparity of resource distribution in the districts. Finally, the study showed that reform measures needed to be fully implemented to ensure the success of the decentralization policy at the hospital level are partially implemented and some not even implemented at all. Some key reform measures that were not fully implemented or not implemented at all were in the areas of planning and financial, human resource and capacity issues, organizational reforms, inter- sectorial coordination, and policy issues and objectives. Human resource and financial issues were still under national government control. 7.3 Conclusions Despite all the numerous reported achievements of health sector decentralization, there are major challenges facing the implementation of health sector decentralization across the world. This is partly true because most governments have often implemented de- concentration or delegation at the expense of devolution as expressed by Martinez- Vazquez et al., (2006). In Ghana, though quite a number of official evaluations and researchers have investigated the factors that have acted as a stop to the smooth implementation of health sector University of Ghana http://ugspace.ug.edu.gh 151 decentralization and reform (Batley a&Larbi, 2004; Agyapong, 1998; Annan, 1997; Smithson et al., 1997) exploring the implementation challenges of this reform at the hospital level has not received much attention from the literature especially in Ghana. This paper therefore investigates the implementation challenges of decentralization at the hospital level. Two hospitals; Sampa Government and Dormaa Presbyterian Hospitals were used as the case hospitals for the study. The results showed that though, government of Ghana has taken some commendable measures in the health sector by transferring some level of authority and responsibilities to local government through deconcentration and devolution approaches, there is the need for government to allow more decision space to local mangers especially in the areas of human resource, planning and financing. Decentralization policy has not achieved its intended purpose because of the numerous challenges facing its implementation at the hospital level and this has even affected the quality and sustainability of health care delivery in the health facilities studied. The gains in equity are also not full under decentralization since there is still large disparity of resource distribution especially human resources which is a key resource in the health facilities. Numerous factors were found to be the main reasons for the unsuccessful implementation of the reform at the hospital level. These factors included among others: Inadequate and skilled health workforce and capacity issues. The study revealed that human resource policies were unclear as well as over centralized human resource functions. The requisite capacities needed at the district level to enable local managers and policy implementers to efficiently manage decentralized units were not taken care of. Communication among key policy implementers was not smooth as some activities and plans of the hospitals were not known to some key policy implementers at the district assemblies and the health directorate. Again financial constraints limit the hospitals ability University of Ghana http://ugspace.ug.edu.gh 152 to stick to their planned activities in a timely manner as reimbursement of the NHIS was not paid on time. Inadequate support from government in terms of infrastructure, capacity building, training of more personnel, unclear policy objectives, and non- involvement of community members in planning and management of health programs poses major challenges to the implementation of the policy at the hospital level. Contextual factors have the potential of influencing the rate at which the decentralization policy is implemented in each district and that there is the need to take them into consideration. The study also investigated the important contextual factors that were taken into consideration before decentralizing the health sector and how they have influenced the challenges of implementation at the hospital level. The findings indicated that socio- economic factors, unequal endowments of local capacity, geographical factors and cultural and ethnic diversity were some of the variables that either facilitate or inhibit the implementation of the policy at the hospital level. The implementation status of some key reform measures which should serve as prerequisite for successful implementation was also investigated in this study. It was recognized that benefits of decentralization are not realized in the health sector because the reform measures needed to be implemented to ensure full achievements of this policy are not fully implemented. Lack of clear objectives and policy issues, lack of the existence of new organizational structure to accompany the new policy, over centralized and non- involvement of community members in planning and financial constraints, human resource and capacity issues and lack of inter-sectoral coordination have combined to pose a major challenge to the implementation of the decentralization policy in Sampa Government Hospital and Dormaa Presbyterian Hospitals. University of Ghana http://ugspace.ug.edu.gh 153 7.4 Recommendations from the study The findings of the study holds many implications for policy makers on the implementation challenges of decentralization reform at Sampa and Dormaa hospitals, both at the policy formulation process and implementation level. To ensure successful implementation of the policy, restructuring of the local government system is a first step. That is restructuring of the local government system and other stake holder institutions which are mandated to implement and monitor the decentralization at the hospital level to effectively carry out their roles well. Functional structure with adequate financial autonomy, more decision space, capacity for policy formulation, monitoring and enforcement of regulations at sector ministry level such as the MOH, and the Ministry of Local Government must be established. Furthermore, contextual factors such as geographical variations, socio- economic, and cultural issues, in each district must be taken into consideration before the implementation. This is because some districts have unique characteristics that either put them ahead of other districts in terms of wealth, education, population, and natural resource. The ability for policy makers to identify these differences and put measures in place to correct these disparities before the implementation will reduce the problem of equity in health care delivery and other services among the districts University of Ghana http://ugspace.ug.edu.gh 154 In addition, the policy should clearly define vision and objectives of the health sector decentralization with clear implementation strategy. There is the need for broader participation during policy formulation process, and role of development partners, civil society and private sectors responsible for health decentralization must be specified. Furthermore, policy makers have to carry out capacity needs assessment for policy feedback and implementation at decentralized level. The needs assessment must be done at both national and the local level before starting with the implementation. In terms of human resource, it is prudent to develop an appropriate human resource development and management policy to address the needs of the decentralization policy and curb the current human resource crisis in the public health system. Another point that is worth considering is the issue of highly centralized Human Resource Management functions. It is counterproductive to decentralization implementation. In view of this, there is the need to reform the national Human Resource Department and management policy to make the health workforce accountable to local government bodies whilst maintaining their permanent posting at the national level. However, the policy must plan for capacity building measures at all levels to ensure that these functions are well managed at all levels. In addition, the local policy makers with technical support from the MOH should come out with fully costed district plans that reflect the needs of the local community. The district, health management committees and other stakeholders must be part of the annual health planning activities, and this planning must consider factors like disease pattern in the district, geographical variations, availability of alternative health providers and the purchasing power of the local people. University of Ghana http://ugspace.ug.edu.gh 155 Then again, a detailed needs assessment taking the views of all the stakeholders into consideration must be carried out. This must be followed by capacity building measures. Good mix of both managerial and technical competencies needed for the entire management of health services is recommended. 7.5 Contributions of the study 7.5.1Theory This study attempts to provide input to the theory of existing knowledge on implementation as it explores the process of policy making and identified implementation challenges as well as some contextual factors such as geography, socio-economic, and cultural issues which are sometimes unique to one district but not others, need to be considered before decentralization at the hospital level because these factors have the potential to influence the rate of implementation of decentralization at the hospital level. The study also provides input to the improvement of the existing decentralization policy process and contributes to knowledge by creating an innovative policy frame work at the local level needed to be followed for successful implementation at the hospital level. 7.5.2 Policy The findings of the study would enable MOH, local government bodies, donors and collaborating institutions to initiate or come out with the necessary reforms from their respective institutions for successful design and implementation of health sector decentralization in Ghana. Secondly, the findings will inform policy makers about how contextual factors within districts affect policy implementation and equalization measures that can be taken to University of Ghana http://ugspace.ug.edu.gh 156 balance these inequalities that have the potential of creating more gaps among districts. This can also ensure equal rate of policy benefits in different jurisdictions. 7.6 Suggestions for future research Contextual factors have major roles to play in shaping the disparities in the districts of Ghana, and its corresponding effect on the success or otherwise of implementing the health sector decentralization in Ghana. However, there are limited studies on how these factors affect the implementation of the decentralization policy at hospital level and how it has shaped disparities in resource allocation in general. This research only considered some of the factors including geographical, unequal socio- economic potential, unequal endowment of local capacity, cultural factors and ethnic diversity. Future research should consider how civil society organization, and participation affect the implementation of health sector decentralization and disparities among the districts in a whole. The way society is organized in terms of information flow, clear lines of authority, respect for authority and members active participation in community activities also influence the implementation of important policies not only at the hospital but other important sectors. For instance, these factors can influence policies in education, agriculture, tourism among others and therefore the need for future studies to explore it effects on policy implementation in other sectors. Measures put in place at the central planning level to ensure equity and successful implementation of this reform given these contextual factors must also be addressed by future studies. University of Ghana http://ugspace.ug.edu.gh 157 Secondly, there are no Human resource personnel in most of the District Hospitals, and these functions are handled by the already burdened Health Services Administrators in such facilities. Future research should consider how effectively the health services administrators combine their roles as administrators with these human resource functions to ensure effective and efficient health care delivery in these hospitals. University of Ghana http://ugspace.ug.edu.gh 158 REFERENCES Adhikari, D. (2006). Towards local democracy in Nepal: Power and participation in District development planning (spring research series no. 47). Kathmandu: Spring Center, Universitat Dortmund. Agyepong, I. (1998). Reforming health service delivery at district level in Ghana: Perspective of a district medical officer. Health policy plan 14(1): 59-69 Agyepong, I. A., Anafi, P., Asiamah, E., Ansah, E. K., Ashon, D. A., &Narh‐Dometey, C. (2004).Health worker (internal customer) satisfaction and motivation in the public sector in Ghana. The International journal of health planning and management, 19(4), 319-336. Anderson J.E (2006).Public policy making: an introduction (6th ed.). Boston: Houghton Mifflin Anderson, E. T., & McFarlane, J. M. (2010).Community as partner: Theory and practice in nursing. Lippincott Williams & Wilkins. Ahmad, E., Brosio, G., &Tanzi, V, (2008), ―Local service provision in selected countries: do decentralized operations work better,‖ in G. Ingram and Y-H. Hong, Fiscal Decentralization and Land Policies, Lincoln Institute. Ahwoi K. 2010: Local Government and Decentralization in Ghana. Accra: Unimax Macmillan Ltd. Ahwoi, K. 2000. ―Trends in Local Government Reforms and Decentralization in Ghana.‖ In A Decade of Decentralization and Local Government Reform in Ghana: Retrospect's and Prospects, eds. W. Thomi, P.W.K. Yankson and S.Y.M. Zanu, 15-24. Accra: EPA Research/Ministry of Local Government and Rural Development. Annan J. (1997). Health sector review: Ghana. Report prepared for the Royal Danish Embassy, Accra, J.S.A. Consultants Ltd Akramov, K. T., and F. Asante. 2009. Decentralization and Local Services in Ghana: Do Geography and Ethnic Diversity Matter? Discussion Paper 872. Washington, DC: International Food Policy Research Institute Aitkinson, S., Medeiros, R., Oliveira, P. & de Almeida, R.D. (2000), ―Going down to the local: incorporating social organization and political culture into assessments of decentralized health care‖, Social Science and Medicine, .51, 619-36. University of Ghana http://ugspace.ug.edu.gh 159 Aitken, R.L.K. (2004).Decentralization impact on health workforce: perspectives of managers, workers and national leaders. Human resource for health.2, (5), 1-11. Armando, A., & Orozco, E. (2006). ―Effects of health decentralization, financing and governance in Mexico". Rev. Saudi publica. 40, 1, 152-160 Asian Development Bank (ADB). (1999). Health reform in Asian and the pacific: option for developing countries. Manila: Asian Development Bank Ayee, J.R.A. (2000). The global context of decentralization. In W. Thomi, P.W.K. Yankson, & S.Y. Azanu, (eds.) A decade of decentralization and local government reform in Ghana: Retrospect and prospects (pp. 3-11). Accra: EPAD Research Project and Ministry of Local Government and Rural Development, Ghana Ayee, J. R. A. 2003a. Local government, decentralization and state capacity in Ghana. In W. Tettey, K. P. Puplampu, and B. Berman (eds.), Critical perspectives on politics and Socio- economic development in Ghana, 45–81. Leiden, Netherlands: Brill. Ayee, J. R. A. (2003b). Towards effective and accountable local government in Ghana, critical Perspectives, No. 13(1–27). Ghana Centre for Democratic Development, Accra Ayee, J.R.A. 2003. ―Decentralization and Local Governance: The Ghanaian Experience. ―In Local Government in Ghana: Grassroots Participation in the 2002 Local Government Elections, eds. Nicolas Amponsah and Kwame Boafo-Arthur, 19- 47. Accra: Uniflow Publishing Ltd. Baker, C. (1996). The health care policy process. Sage publications, London. Beaglehole, R., Sanders, D., Dal Poz, M. (2003).The public health workforce in sub Saharan Africa: Challenges and opportunities‖. Ethn Dis. 2003, 13(2 suppl2): 24-30. Bossert T. Decentralization. In: Jamovsky K. (ed) Health policy and systems development: An agenda for research. WHO/SHS/96.1, Geneva; WHO, 1996. Bossert,T., & Thomas J. 1998. Analyzing the Decentralization of Health Systems in Developing Countries: Decision Space, Innovation, and Performance. Social Science and Medicine 47 (10): 1513–27 Bossert, Thomas J., & Joel C. Beauvais. 2002. Decentralization of Health Systems in Ghana, Zambia, Uganda, and the Philippines: A Comparative Analysis of Decision Space. Health Policy and Planning 17 (1): 14–31. University of Ghana http://ugspace.ug.edu.gh 160 Bosser, T. J., Chitah, M.B., & Bowser, D. (2003). Decentralization in Zambia: resource allocation and district performance. Health policy and planning.18, 4,357-369. Bossert, T. J, (2004b).Institutional capacity issues. Boston: The World Bank. (Unpublished training module). Bracht, N., and Tsouros, A. (1990). Principles and strategies of effective community participation. Health promotion international, 5(3), 199-208. Breton, A. (2002). An introduction to decentralization failures. In E. Ahmad and V. Tanzi (eds.), Managing Fiscal Decentralization, 31–45. London: Routledge State University. Brijlal, V., Gilson L., McIntyre, D., and Thomas, S. (1998). Health sector reform and financing: Key issues in decentralization. Cape Town: University of Cape Town & University of the Witwatersrand. Buchan J. (2000): Health sector reform and human resource: Lessons from the United Kingdom. Health policy plan, 15; 319-325 Burgess, S. M. (2003). Within-country diversity: Is it the key to South Africa‘s prosperity in a changing world? International Journal of Advertising, 22(2): 157–182. Cheema, G. S., & Rondinelli, D. A. (1983).Decentralization and development: policy implementation in developing countries. Sage Publications. Chen, L. C. (2010). Striking the right balance: health workforce retention in remote and rural areas. Bulletin of the World Health Organization, 88(5), 323-323. Cochrane, G. (1983).Policies for strengthening local government in developing countries. Washington, D C. World Bank, (Staff working paper, 582) Conyers, D., 2000, ‗Decentralization: A Conceptual Analysis (Parts 1 and 2)‘, Local Government Perspectives: News and Views on Local Government in Sub-Saharan Africa, 7, Nos.3/.7–9, 13, 18–24. Crook, R. C. (2003). Decentralization and poverty reduction in Africa: The politics of local-center relations. Public Administration and Development, 23:77–88. Collier, P. (2008). Oil, growth and governance in Nigeria. In P. Collier, C. Pattillo, and C. Soludo (eds.), Economic Policies for a Prosperous Nigeria, 45–60. New York: Palgrave. University of Ghana http://ugspace.ug.edu.gh 161 Collins, C., Omar, M., Adhikari, D., Emmel, N., Dhakal, R., and Dhakal, M. R., (2003). Developing health decentralization in Nepal: Collaborative policy development. Kathmandu: British council. Couttolenc, B.F. (2012). Decentralization and governance in the Ghana health sector. World Bank, Washington D. C. Dauda, C. L. (2006). Democracy and decentralization: Local politics, marginalization and political accountability in Uganda and South Africa. Public Administration and Development 26, 291–302. Dhakal, R., & Singh, A. (2006). Health sector Decentralization in Nepal. Prospects and challenges. Journal Nepal Public Health Association, 3, 1, 26-34. Dhakal, R. (2009). Analysis of the implementation challenges of health sector decentralization policy in Nepal Dormaa Municipal Assembly (2012). Municipal Medium Term Development Plan (MMTDP) Dovlo, D. (1998). Health Sector Reform and Deployment, Training, and Motivation of Human Resources towards Equity in Health Care: Issues and Concerns in Ghana .Ministry of Health, Accra. http://www.hrhresourcecenter.Org/node/181. Dovlo, D. (2005b). Wastage in the health workforce: some perspectives from African countries. Human Resources for Health, 3(1), 6. Dye, T. R. (2004).Understanding public policy. New Delhi. Pearson Education Inc. Dunsire, M. (1978), Implementation in a Bureaucracy, Martin Robertson, Oxford Dussault, G., and Dubois, C.A. (2004).Human resource for health policies: a critical component in health policies. (HNP discussion paper no.1). Washington DC. The World Bank. Easton, D. (1965). A framework for political analysis. New Jersey: Prentice Hall Eliott, M. J. (1981). The role of law in central- local relations. London, Social science Research Council. Fielden, R., and Ole Frank N. (2001). Immunization and health reform: Making reforms work for immunization. A reference guide. Geneva: World Health Organization, Department of Vaccines and Biologicals. Available at http://www.who- Int/vaccines-documents. Accessed 11th May, 2014 Francis, P., & James, R. (2003).Balancing rural poverty reduction and citizen participation: The contradictions of Uganda‘s decentralization program. World University of Ghana http://ugspace.ug.edu.gh 162 Development, 31(2), 325-337. Gaventa, J. (2002). Towards participatory local governance: Six propositions for discussion. Currents, 28, 29-35. Fudge, C., &Barret, S. (1998). Reconstructing the field of analysis. In Policy and Action. Essays on the implementation of policy. (Edited by Barret, S and Fudge C.), 71- 82. Methuen, London. Ghana, Republic of, Decentralization Secretariat, Ministry of Local Government and Rural Development (MLGRD) (2003) National Decentralization Action Plan (NDAP): Towards a Sector-Wide Approach for Decentralization Implementation in Ghana (2003-2005). MLGRD, Accra (unpublished). Ghana Statistical Service (2012), 2010 Population and Housing Census: Summary Report of Final Results, Accra Ministry of Health, Ghana (2007.) Health Sector 5 Year Program of Work 2002–2006. Independent Review ofPOW-2006. Accra, Ghana. Ministry of Local Government and Rural Development, 2010, National Urban Policy, 2010, Ministry of Local Government and Rural Development, Accra. Gilson, L., and Travis P. (1996) Health system decentralization in Africa: An overview of experiences in eight countries. A background document prepared for the Regional meeting on Decentralization in the context of health sector Reform in Africa. World Health Organization Geneva. Gilson, R.J, and Mills, A. (1995). Health sector reform in Sub- Saharan: lessons of the past 10years. Health planning.32, 1,215-243. Gopal KM, Mondal S: District health planning. (2007) an overview. In District Health Planning Edited by: Sundararaman T, and Prasad V. (2007):1-12. Public Health Resource Network, Book 10] experiences in 8 countries. Geneva: World Health Organization. Gottret, P. E., & Schieber, G. (2006). Health financing revisited: a practitioner‘s guide. Washington DC: World Bank, 2006. In Scott Burris and Leo Beletsky, OSI Seminar on Global Governance of Health (Conference Report, Open Society Institute: Schloss Arenberg. Greenwood, R. (1997). Relation between central and local government in Sweden: the control of local government expenditure. Public administration, 57,457-470 Green, A. (2001). An introduction to health planning in developing countries. New University of Ghana http://ugspace.ug.edu.gh 163 York: Oxford University press. Griffith, J. A. G. (1966) Central Departments and Local Authorities. University of Toronto Press, Toronto, Ontario. Grindle, M., and Thomas, J. (1991). Public choice and policy change: the political economy of reform in developing countries. Baltimore. The Johns Hopkins University. Hammer, J. S., and Bermah, P. (1995). Ends and means in public health policy in developing countries. Health policy. 32, 3, 29-45. Hanson, K., Ranson, M.K, .Oliveria, V, and Mills, A. (2003). Constraints to expanding accessing to health interventions: an empirical analysis and Country typology. Journal of International Development, 15(1), 15-38. Hogwood, B.W., and Gun, L, (1984).Policy analysis for the real world. Oxford: Oxford University press. Hunter, D. (1997). The challenges of healthcare restructuring. Nurse Times, 93(39) 67-70. Jeppsson, A., Orstegen, P. O., and Hagstrom, B. (2003). Restructuring a ministry of health- an issue of structure and process. A case study from Uganda. Health policy and planning. 18(1), 68-73. Kamiljon, T. A., and Asante, F. (2009).Decentralization and Local Public Services in Ghana: Do Geography and Ethnic Diversity Matter? Development Strategy and Governance Division. Kanji, N., Kanji, N., &Manji, F. (1991). From development to sustained crisis: structural adjustment, equity and health. Social science & medicine, 33(9), 985-993. Kanbur, Ravi, Zhang, Xiaobo, (2005). Fifty years of regional inequality in China: a journey through central planning, reform and openness. Review of Development Economics, 9 (1), 87–106. Kawonga, M. (2003).Decentralization- Africa Regional paper. Women‘s Health Project, Johannesburg (unpublished). 48 accessed at http://www.accessmylibrary.com/coms2/summary(accessed20th May, 2014) Kingdom, J.W. (1995).Agenda, alternatives, and public policies. New York: Longman. Human resource for Health 2003, 1:1. Available from: http://www.human- resources- University of Ghana http://ugspace.ug.edu.gh 164 Kolehmainen-Aitken R-L. The impact of decentralization in health workforce development in Papua New Guinea. Public Administration and Development, 1992, 12:175-191 Larmour, P and Qalo R; ed. (1985) Decentralization in South pacific. Suva, university of South pacific. Leighton, C. (1996). Strategies for achieving health- financing reform in Africa. World development. 24, 9, 1511- 1525. Litvack J, Ahmad J, and Bird R, 1998. Rethinking decentralization in developing countries. Washington DC: The World Bank. Macrae J, Zwi AB, and Gilson L.1996. Triple burden for health sector reform: post – conflict Rehabilitation in Uganda. Social science and medicine 42: 1095-1108 Mamdani, M. (1996). Indirect rule, civil society, and ethnicity: The African dilemma. Social Justice, 145-150. Martinez-Vazquez, J., and B. Searle (eds.). 2007. Fiscal equalization: Challenges in the design of intergovernmental transfers. New York: Springer. Martineau T, Gong Y, Tang S. Changing medical doctor productivity and its affecting factors in rural China. International Journal Health Plan Manage 2004; 19: 101-11. Mubyazi, G., Kamugisha, M., Mushi, A., and Blas, E. (2004).Implication of decentralization for the control of tropical diseases in Tanzania: a case study of four districts. International Journal of Health planning and management, 19, (S1)167S 185. Muriisa, R. K. (2008). Decentralization in Uganda: prospects for improved service delivery. Africa Development, 33(4). Murthy, N. (1998). Decentralized health planning: lessons learned from two districts in India. Journal of health and population in developing countries, 1(2), 1- 10. Musgrave, R., and Musgrave, P. (1989). Public Finance in Theory and Practice. New York, McGraw-Hill International. Oyayo, C., and Rifkin, S. (2003). Health sector reforms in Kenya: An examination of district level planning. Health Policy, 64, 113-127. Narasimhan V, Brown H, Pablos-Mendez A, et al. Responding to the global human resources crisis. Lancet 2004; 363:1469–72. University of Ghana http://ugspace.ug.edu.gh 165 National Decentralization Action Plan (NDAP). (2003- 2005). Towards a Sector- wide Approach for Decentralization Implementation in Ghana. Nkrumah, S. A. (2000). Decentralization for good governance and development: The Ghanaian experience. Regional Development Dialogue, 21 (1), 53-67. Pan American Health organization (PAHO) Development and strengthening of Human Resources management in the health Sector (2008). Washington Dc, 128thsession of the Executive Committee 2001. Pfeffer, J. (1992). Managing with power: Politics and influence in organizations, 524.Harvard Business Press. Price, J. L. (1975). The measurement of turnover. Industrial Relations Journal, 6, 33-46. Prud‘homme, R. (1995). The Dangers of Decentralization. World Bank Research Observer 10(2), 201-220. Putnam, R. (1993). The prosperous community: social capital and public life. The american prospect, 13: 35.42. Radulescu, S, .and Vladescu, C, (2005). Recent experience in the Romania health sector. Ratanawijitrasin, S., Soumerai, S. and Weerasurya, K. (2001). Do national medicinal drugs policies and essential drug programs improve drug use? A review of experiences in developing countries. Social science and medicine, 53(7), 831844. Ribot, J. C. (1999). Decentralization, participation and accountability in Sahelian forestry: Legal instruments of political – administrative control. Africa, 69 (2), 23-65. Rotha, C, and Vannarith, C., [n.d]. Decentralization process in Cambodia. Sakyi, E.K. (2008). Implementing decentralized management in Ghana: The experience of the Sekyere West District health administration. Leadership in Health Services, 21(4), 307- 319. Sakyi, E.K. (2010). Communication challenges in implementing health sector decentralization at district level in Ghana: A study of health workforce and stake holder opinions from three district health administrations. Leadership in Health Services, 23(2), 156-179. Sapru, R. K (2004). Public policy: formulation, implementation and evaluation. New Delhi: Sterling. Saltman, Richard B., Vaida B, and Karsten, V. eds. (2007). Decentralization in Health Care: Strategies and Outcomes. Maidenhead, U.K., and New York: Open University Press. University of Ghana http://ugspace.ug.edu.gh 166 Schneider, A. (2003). Decentralization: conceptualization and measurement. Studies in Comparative International Development, 38(3), 32-56. Segal, H. (1983). Planning and politics of resource allocations for primary health care: Promotion of meaningful national policy. Social science and medicine.17, 24, 1947-1960. Shah, A., & Thompson, T. (2004). Implementing decentralized local governance: a treacherous road with potholes, detours, and road closures (3353). World Bank Publications Shah, A., Thompson, T., &Zou, H. F. (2004). The impact of decentralization on service delivery, corruption, fiscal management and growth in developing and market economies: A synthesis of empirical evidence. CESifo DICE Report, 1(2004), 10-14. Smith, B C. (1985) Decentralization. The territorial dimension of the state. London. George and Uwin publications. Suma, D. F. (2007).The external debt crisis and its impact on economic growth and investment in Sub-Saharan Africa. A regional econometric approach of ECOWAS countries (Doctoral dissertation, WU Vienna University of Economics and Business) Sutton, R. (1999). The policy process. An overview (ODI working paper no. 188).143- 150. Tendler, J. (2000). Why are social funds so popular? Local Dynamics in an Era of Globalization: 21st Century Catalysts for Development, 114. Therkildsen, O. (2000): Demokratisk Desentralisering. Den Nye Verden, No. 4. Center for Utviklingsforskning, København. Tordoff, W. (1997). Government and Politics in Africa United States Agency for International Development (USAID) (2003), ‗Decentralization in Ghana: an Assessment‘ (mimeo). United Nations Development Program (1998). Decentralized Governance Monograph: A global Sampling of experience. Management Development and Governance Division, Bureau for Policy development. United Nations Development Program (1999). Decentralization: A sampling of definitions. Working Paper for Joint UNDP-Government of Germany Evaluation of the UNDP Role in Decentralization and Local Governance. UNDP. University of Ghana http://ugspace.ug.edu.gh 167 United Nations Development Program (1999). Decentralization: A sampling of definitions. Working Paper for Joint UNDP-Government of Germany Evaluation of the UNDP Role in Decentralization and Local Governance. UNDP. Van Lerberghe, W., Adams, O. and Ferrinho, P. (2002a) ‗Human resources impact assessment‘, Bulletin of the World Health Organization, 80(7), pp. 525. Walt, G. and Gilson, L. (1994). Reforming health sector in developing countries: The central role of policy analysis. Health policy and planning. 9, 4, 353-360. Walt, G. (1996). Health policy. An introduction to process and power. London. Zed Books. Walker, L. & Gilson, L. (2004).We are bitter but we are satisfied: nurses as street level bureaucrats in South Africa. Social science and medicine.59, 6, 1251-`1261. Wang, Y., Collins, C., Tang, S., and Martineau, T. (2002). Health systems decentralization and human resources management in low and middle income countries. Public Administration and Development, 22(5), 439-453. WHO. Report of Working Group 5 of the Commission on Macroeconomics and Health. Geneva: World Health organization, 2002. World Health organization, Regional Office for Africa. The African regional health report, (2006): The health of the people. Available from URL:http://whqlibdoc.who.int/afro/2006/929023033-rev-eng.pdf or http://www.who.int/bulletin/africanHealth/en/index.html (Accessed20 September, 2013) Work, R. (2002).Overview of decentralization worldwide: A stepping stone to improved governance and human development. Paper Presented at the 2nd International Conference on Decentralized Federalism. Manilla, Phillipine. The World Bank (1993). World Development Report 1993: Investing in Health. New York: Oxford University press. World Bank (2000).Decentralization and Healthcare. Decentralization Thematic Group, World Bank, Washington, D. C. World Bank (2001). Decentralization and governance: does decentralization improve public service delivery. (World Bank PREM Notes, no. 55). World Health organization (WHO). (2006). The World Health Report 2006: Working together for health. Geneva: WHO. World Health organization, Regional Office for Africa. The African regional health Health/en/index.html (Accessed 20 September, 2013). University of Ghana http://ugspace.ug.edu.gh 168 World Health organization, Regional Office for Africa. The African regional health. Health/en/index.html (Accessed 20th September, 2013). Yin, R., (1989). Case Study Research in Education, California, pp: 4-25 Zanten, T. V and Samedei, C. (1996). Assessment of health sector decentralization in Paraguay (Technical report no.1). Bethesda, M.D: Partnership for health reform project, Abt. Associate Inc. University of Ghana http://ugspace.ug.edu.gh 169 APPENDICES Appendix A: Interview guide for key policy implementers at the hospital level SN Interview check list Coding I Date and place of interview Interview Ii ID of respondent ID Iii Organization and duration(years) ID A Formulation of health sector decentralization policy 1. Why do you think that the government decided to decentralize the health sector to local bodies? Context, respondents knowledge, rational of decentralization 2. How would you describe the features of decentralization policy? Objectives, characteristics. 3. Were you or your organization consulted in developing health sector decentralization? If yes, could you please share your experiences? Policy formulation 4. Does the policy addresses the health service organization and management needs of the District hospitals? Policy characteristics 5. What important factors concerning the District and hospital were taken into consideration before Decentralization?, If any, how have these factors positively or negatively affected the success of the policy Important contextual factors. B Policy implementation (reform in organizational set- up& service delivery mechanisms). 6. What changes were made in organizational structure to implement decentralization policy at the hospital level? Organizational structure and issues 7. How does the hospital level planning takes place? Planning practices and participation 8. Was the required financial resource ensured during planning? If not, what alternative mechanisms were adopted to finance the planned activities? Financing of the plan 9. What changes are made in financing and expenditure system? Financing, decision space before and after decentralization 10. What changes were made in human resource management in the district and hospital level before and after decentralization? Local government‘s mandate and practice in human resource management 11. What capacity is needed at the district and hospital level for decentralization? Capacity development University of Ghana http://ugspace.ug.edu.gh 170 12. What is the communication mechanism between center, District, and hospital level? Information and communication. 13. In your opinion, has decentralization policy received sufficient attention from political and bureaucratic level? What mechanism exists to seek participation of relevant stakeholders at the hospital level? Political commitment, participation. 14. How is the supervision and monitoring of programs at the district and hospital level? Supervision and monitoring system 15. What effects did the socio- economic and arm conflict made in the decentralization policy implementation at the hospital level? Socio-economic and political environment. 16. Could you briefly share with us on the implementation status of the planned activities at the district and hospital level? Implementation status 17. What are the major problems encountered during policy implementation at the hospital level? Issues of policy implementation. 18. In your opinion, what are the important factors for successful implementation of decentralization policy in your area of work? Policy reform proposal. 19. Do you have anything more to add? Please have your comment. Opinion 20. Researcher‘s own observation and comments on the overall interview process. Thank you very much for your valuable time. Appendix B: Facility observation checklist 1. Total number of staff in health facility (appointment from central government, by the local management committee, others and their length of stay in this health facility). 2. Type and condition of facility (concrete, mud, rented and general condition). 3. Number of room different purposes (consultation, dispensing, meeting, etc. 4. Record keeping system (well maintained, poor). 5. Financing; i.e., account keeping system, sources of funding, cost centers? 6. Types of services provided by the health facilities including referral mechanisms 7. Composition of the management committee( Male, Female, ethnic representation) 8. Meeting organization and decision making practices (if possible researcher himself will participate in some meetings to observe interactions of the participants and decision making practices. 9. Checking the minutes of the meeting (number of meetings held, decision made, participants (male/ female). University of Ghana http://ugspace.ug.edu.gh