eResearch: the open access repository of the research output  
of Queen Margaret University, Edinburgh. 
 
 
 
This is the Publisher’s version/PDF of a book published as: 
 
 
Tashobya, C K and Ssengooba, Freddie and Oliveira Cruz, V,                         
eds. Health Systems Reforms in Uganda: processes and  
outputs. Institute of Public Health, Makerere University, Uganda.  
 
 
Accessed from:  
http://eresearch.qmu.ac.uk/762/
 
 
 
Repository Use Policy 
 
The full-text may be used and/or reproduced, and given to third parties for personal  
research or study, educational or not-for-profit purposes providing that: 
• The full-text is not changed in any way 
• A full bibliographic reference is made 
• A hyperlink is given to the original metadata page in eResearch 
 
eResearch policies on access and re-use can be viewed on our Policies page: 
http://eresearch.qmu.ac.uk/policies.html
 
 
 
 
 
 
 
 
     http://eresearch.qmu.ac.uk
Table of Contents
Foreword 5
Acknowledgements 9
Biographies 11
Chapter 1 The Ugandan health systems reforms: miracle or mirage? 15
by Rob Yates, Christine Kirunga Tashobya, Valeria Oliveira Cruz, 
Barbara McPake, Freddie Ssengooba, Grace Murindwa, Peter Lochoro, 
Juliet Bataringaya, Hanif Nazerali and Francis Omaswa
Chapter 2 Is the sector-wide approach (SWAp) improving health 29
sector performance in Uganda?
by Valeria Oliveira Cruz, Ros Cooper, Barbara McPake, Rob Yates, 
Freddie Ssengooba, Francis Omaswa, Christine Kirunga Tashobya and 
Grace Murindwa
© Institute of Public Health, Makerere University, Uganda; Chapter 3 Health sector reforms and increasing access to health 45
Health Systems Development Programme, London School services by the poor: what role has the abolition of user fees 
of Hygiene & Tropical Medicine, UK; and Ministry of played in Uganda?
Health, Uganda by Christine Kirunga Tashobya, Barbara McPake, Juliet Nabyonga 
and Rob Yates
Chapter 4 Medicines – driving demand for health services in Uganda? 61
Suggested citation: by Hanif Nazerali, Martin Olowo Oteba, Joseph Mwoga and Sam Zaramba
Health Systems Reforms in Uganda: processes and outputs.
Christine Kirunga Tashobya, Freddie Ssengooba and Chapter 5 Public-private partnership in health: working together to 83
Valeria Oliveira Cruz (editors). Health Systems improve health sector performance in Uganda
Development Programme, London School of Hygiene & by Peter Lochoro, Juliet Bataringaya, Christine Kirunga Tashobya 
Tropical Medicine, UK. and Joseph Herman Kyabaggu
Chapter 6 Meeting the challenges of decentralised health service 97
delivery in Uganda as a component of broader health sector reforms
by Grace Murindwa, Christine Kirunga Tashobya, Joseph Herman Kyabaggu, 
Eliseus Rutebemberwa and Juliet Nabyonga
Chapter 7 Have systems reforms resulted in a more efficient and 109
equitable allocation of resources in the Ugandan health sector?
by Freddie Ssengooba, Rob Yates, Valeria Oliveira Cruz and 
Christine Kirunga Tashobya
Foreword
Taking responsibility for sustainable improvements 
in the health status of Ugandans
During the 1970s and 1980s, Uganda went through a period of political and economic
upheaval, resulting in the breakdown of many services. In the health sector this was
characterized by general system failure. Funding was grossly insufficient, leading to
problems of meagre and late salaries for health workers, permanent shortages of
medicines and supplies, and dilapidated infrastructure. The National Resistance
Movement, headed by President Yoweri Museveni, took power in 1986. Initial efforts by
this government were focused on the restoration of law and order and on the re-
establishment of public systems. However, there was minimal government funding
available for social services including the health sector. Many bilateral and multilateral
donors intervened to provide post-conflict support including emergency rehabilitation
of the health infrastructure. These international agencies focused their support on
specific aspects of the health sector (usually disease programmes) in particular parts of
the country. Districts which were closer to the capital city and easily accessible attracted
many donor projects, while districts which were remote and difficult to access received
few or no projects at all. 
Following these efforts, there were some improvements in health outputs and
outcomes, which was not surprising given their initially very low levels. However, most
of these gains were seen to be stagnating or reversing by the mid-1990s. There were
growing concerns that the health sector was failing to deliver the expected levels of
improvements in outputs and outcomes, despite what was considered an appreciable
amount of funding support. In the words of one World Bank official, the health sector
had become a ‘bottomless pit’! 
In fact, the real challenge was that due to the fragmented nature of support, there
was minimal understanding of the basic features of the sector (e.g. size of the sector,
significance of the different players, quantification of requirements versus what was
available). Policy recommendations were often adopted from other country strategies
and applied to Uganda without due consideration to the local context. Donor project
activities were independent of government activities, and there were no provisions
made to allocate government resources to fund the activities following completion of
the individual projects. For example, in the mid-1990s, UNICEF was no longer able to
maintain its level of funding for the Uganda National Expanded Programme for
Immunisation. No preparations had been made for sustaining the activities using
government funding and as a result, immunisation coverage plummeted again
(following an increase during the 1980s and 1990s). 
6 Health Systems Reforms in Uganda: processes and outputs Foreword 7
In the 1990s, in light of these problems, the government initiated the process of human resources, infrastructure (the other two critical inputs to health care in addition
preparing a National Health Policy and a Health Sector Strategic Plan. At the same to medicines), and supervision and monitoring. Finally, Ssengooba et al. analyse
time, a number of bilateral and multilateral agencies had become disillusioned with the available financial data to determine the extent to which these reforms have yielded
project-based method of delivering aid and the international community thus began to improvements in efficiency and equity in the Ugandan health sector.
question the existing modalities of providing Overseas Development Assistance (ODA). I would also like to point out a few key critical higher level factors that are common
Hence the advent of the sector-wide approach for (SWAp) health development where, to each of the reforms discussed in this book.
instead of focusing on individual projects, donors would support and fund the Firstly, the importance of government stewardship cannot be overemphasized. The
implementation of a coordinated, sector-wide strategy. Since the challenges of managing Ministry of Health (in collaboration with other institutions of government) has led the
ODA were a key feature of the National Health Policy and Health Sector Strategic Plan, process with deliberate consultation with the different stakeholders to the sector. The
Uganda readily embraced this new development. majority of key decisions have been made following a consensus-building approach and
The health sector stakeholders that were involved in the development of the in cases where this has not happened (e.g. abolition of user fees), the government has
National Health Policy and Health Sector Strategic Plan (2000/01-2004/05) under the based its decisions on available information and prevailing circumstances. 
SWAp included central (Ministries of Health, Finance, Public Service, Local Second, there has been increasing use of government mechanisms at the national,
Government, Education) and local government officials, representatives of multi-lateral sectoral, and local government levels for purposes of planning, financing, procurement,
and bilateral agencies, NGOs and private providers of health services. The then and monitoring and evaluation, rather than using separate donor project systems. This
Minister of Health, Hon. Dr. Crispus Kiyonga, steered the whole process, supported by has very important implications for developing systems that are sustainable by the
a committed technical team led by Professor Francis Omaswa (Director General Health government in the medium and longer term. 
Services), Dr. Joseph Kyabaggu (Director for Planning and Development), Dr. Sam Third, it is crucial to build the trust and confidence of the stakeholders in order to
Zaramba (Director Clinical and Community Health), and Dr. Patrick Kadama ensure a successful partnership. However, this entails substantial transaction costs. For
(Commissioner Planning). Particular mention is also made of Dr. Hatib Njie (WHO example, it took more than four years to agree to the Health Sector Strategic Plan
Country Representative), Ms. Ros Cooper (DFID Uganda Country Office), Ms. Nicola 2000/01-2004/05, with innumerable meetings between stakeholders in and outside the
Brennan, (Ireland Aid), Dr. Phil Gowers (RIP), Ms. Mary Mulusa (World Bank country and several consultations between multilateral and bilateral agencies’
Headquarters, Washington) and Dr. Daniele Giusti (Uganda Catholic Medical Bureau) representatives and their headquarters. Nevertheless, this process was necessary to
for their outstanding contribution.  enable the stakeholders to agree to and own a common vision and objectives.   
The details of the different interventions in the National Health Policy and the first We are very proud of what this process has achieved. Many infections have been
Health Sector Strategic Plan (2000/01-2004/05) and what they have achieved are well prevented given the improved coverage of immunisation and other preventive and
articulated in the different chapters of this book. promotive programmes. The proportion of the population that is able to access
In the first chapter, Yates et al. provide a useful overview of the achievements of the ambulatory services almost doubled over four years (from 0.42 out-patient visits per
health sector, in terms of key outputs, since the development of the first Health Sector person per year in 1999/00 to 0.79 visits in 2003/04). The last Uganda Demographic and
Strategic Plan and the launch of the SWAp in Uganda in 2000. They identify a number Health Survey (UDHS) was carried out in 2000/01 at the time when these reforms were
of demand and supply side factors that contributed to these achievements, each of which just beginning. It is expected that the UDHS of 2005/06 will show some improvements
are taken up in the subsequent chapters and discussed in more detail to determine their in mortality figures given these improvements in health sector outputs. This is expected
respective contribution to the overall programme of health systems reforms in the despite the observed stagnation or worsening of some of the other determinants of
country. Oliveira Cruz et al. outline the mechanisms and processes of the Ugandan health including the increase in poverty (in both absolute and distributional terms) and
health SWAp, discuss its major benefits, and outline challenges for the future. Kirunga insecurity in some parts of the country. 
Tashobya et al. investigate the implications of a major demand side policy instigated in Nevertheless, there are a number of challenges that have been, and are continuing,
2001, namely, the abolition of user fees in public health facilities. They employ a range to emerge. The above improvements have occurred largely because of efficiency gains
of data sources to show that this policy led to a dramatic increase in utilisation of rather than big increases in the overall health sector resource envelope. In particular, an
primary health care services in the country, particularly by the rural poor. Nazerali et al. increasing proportion of financial support for the sector is channelled through the
provide a detailed description of the reforms in the financing and supply of medicines, government budget (i.e. government funding plus donor budget support) rather than
one of the three critical inputs to health care, which led to significant improvements in through individual donor projects, resulting in greater efficiency and equity in resource
the availability of essential medicines in local primary health care facilities. Lochoro et allocation. The interventions that have been articulated in the second Health Sector
al. discuss the role of partnership with the private sector, focusing in particular on Strategic Plan for the period 2005/06 to 2009/10 focus more strategically on reducing
progress made in partnership with the private-not-for-profit sub-sector which has child and maternal mortality and morbidity, including the provision of emergency
contributed to greater alignment and coordination in the supply of health care services obstetric care (basic and comprehensive) closer to rural women, and further
in the country. Also on the supply side, Murindwa et al. investigate the contributions of development of prevention and treatment services for malaria and HIV/AIDS. Each of
key changes in the organisation of local level primary health care services, focusing on these requires efficient utilization of a bigger resource envelope for the sector.
8 Health Systems Reforms in Uganda: processes and outputs
Further increases in the health sector resource envelope are constrained by a
number of factors. These include the pledge by the government to pursue conservative
management of the macroeconomy, with a firm cap on the fiscal deficit, articulated in
the Poverty Eradication Action Plan 2004 (the country’s Poverty Reduction Strategy Acknowledgements
Paper). Therefore, future growth in the Medium and Long Term Expenditure
Frameworks (MTEF & LTEF) is restricted, thereby curtailing future public spending
(government and donor), including health sector expenditure. Global health initiatives
(e.g. the Global Fund to Fight AIDS, TB and Malaria, and the US President’s
Emergency Plan for AIDS Relief) also pose a potential threat to the efficiency of the
health sector resource envelope. These initiatives use traditional project-based
approaches that are not consistent with the systems approach used under the SWAp. The editors are grateful to a number of individuals for their contribution to this
This has resulted in the re-emergence of disease-specific projects and the use of parallel publication. Due to space limitations we would like to especially thank Samantha Smith
systems for planning and monitoring which have implications for equity and efficiency (at the time working with the Overseas Development Institute) who provided valuable
in resource use. Moreover, funds from these less efficient funding sources are included in support to the editing of this book; as well as Nicola Lord and Tamsin Kelk (London
the health sector resource envelope outlined in the MTEF and LTEF and thus School of Hygiene & Tropical Medicine) who provided great support in relation to the
potentially displace resources allocated through the (more efficient) government budget. design, printing and dissemination of this publication. We are also indebted to the
As a final comment, sustaining the mutual trust established between government funding provided by DFID Uganda for supporting the editorial and printing costs. In
and development partners and other stakeholders is an ongoing challenge. This is addition, we are grateful to DFID funding to the Health Systems Development
affected by other decisions made at global, regional and national levels. The high Programme which allowed staff members to participate in this project. The opinions
turnover of officers in both government and other stakeholder bodies makes orientation expressed in this book are those of the authors and do not necessarily reflect the views
meetings a must. There is still room for improving partnerships for health within the of the organisations providing support to this publication.
country, especially with the private sector, with communities, households and
individuals, if these achievements are to be built on and sustained. Uganda still suffers
from a high burden of preventable diseases which can only be limited by improved
health literacy and involvement of the communities. 
On behalf of the Government of Uganda, and in particular the Ministry of Health,
we are very glad to share our experiences with other developing countries and especially
those in sub-Saharan Africa. It is not often that we have success stories to share. This
book has been put together by some of the Ministry of Health officials that have been
involved in the implementation of these reforms, together with partners from bilateral
and multilateral agencies, private healthcare providers’ representatives and researchers.
This collaboration has facilitated a factual and objective analysis of the experiences.
Dr. Sam Zaramba
Ag. Director General of Health Services 
Biographies 
Editors: 
Christine Kirunga Tashobya (Mb ChB; MA Demography; MSc Health Policy,
Planning and Financing) is a Public Health Advisor in the Ministry of Health (MoH)
in Uganda under the auspices of the Danida-funded Uganda Health Sector Programmes
Support II.  She has been involved in reforms in the Uganda MoH in various capacities
over the last decade, especially in areas of overall policy formulation, stakeholder
coordination, health financing and public-private partnership for health. Currently, she
is involved in supporting districts to provide better services in the era of decentralisation
and the sector-wide approach. Email: Christine@hsps-ug.org
Freddie Ssengooba (Mb ChB; MPH) is a member of the Health Systems
Development Programme at the London School of Hygiene & Tropical Medicine,
supported by DFID and Makerere University, Institute of Public Health. He is currently
working on studies on human resources for health, hospital service quality, and response
to performance contracts in the Ugandan health system. Email: sengooba@iph.ac.uk
Valeria Oliveira Cruz (MSc Health Policy, Planning and Financing) is a lecturer in
the Health Systems Development Programme at the London School of Hygiene &
Tropical Medicine. Her current work focuses on an investigation of the nature of the
relationship between development partners and the Government of Uganda. 
Email: valeira.oliveira-cruz@lshtm.ac.uk
Editorial Board:
Rob Yates (BA Economics; MBA) is seconded to the Ministry of Health in Uganda
by the UK Government’s Department for International Development (DFID). His work
focuses on health financing mechanisms and public sector management reforms. 
Email: yatesug@infocom.co.ug
Grace Murindwa (Mb ChB; MA Health Policy, Planning and Management) works
as Senior Health Planner in the Health Planning Department, Ministry of Health,
Uganda. He has been actively involved in the formulation and implementation of
health sector reforms in Uganda for over eight years. Currently, he is involved in
providing technical support to districts especially with regard to planning and
monitoring of health service delivery. Email: murindwag@yahoo.com
12 Health Systems Reforms in Uganda: processes and outputs Biographies         13
Barbara McPake (BA Economics; PhD Health Economics) is Professor and Director Hanif Nazerali (MRPS; MPH International Health and Development) is a Danida
of the Institute of International Health and Development at Queen Margaret University adviser attached to the Ministry of Health, Uganda through the Danish Health Sector
College, Edinburgh.  Her work focuses on the economics of health systems and has Programmes Support. He has been involved in formulating new systems for medicines
included studies on health financing, human resources, and hospital reform in sub- financing and supply and making them operational in Uganda over a period of four
Saharan Africa. Email: bmcpake@qmuc.ac.uk years. Email: hanif@hsps-ug.org
Francis Omaswa (Mb ChB; M Med FRCS) is the Special Advisor on Human
Contributors: Resources at the WHO Headquarters Geneva. Prior to this appointment he was the
Director General for Health Services in Uganda. His major interests include health
Juliet Bataringaya (BSc; Mb ChB; MSc Health Policy, Planning and Financing) systems development, stakeholder coordination and quality assurance. 
worked as consultant for the Public-Private Partnership in Health Unit in the Health
Planning Department, Ministry of Health, Uganda. She is now National Professional Martin Olowo Oteba (MSc Pharm; MIH) is the Principal Pharmacist, and overall
Officer for Health Systems Development at the WHO Country Office in Uganda. coordinator and technical advisor to the Ministry of Health, Uganda on pharmaceutical
Current work involves strengthening of district and sub-district health systems, human policy and management. His main interests include logistics, health systems, and
resources for health development, collaboration with partners and NGOs, and international health. Email: orukan33@hsps-ug.org or orukan33@hotmail.com
mainstreaming community initiatives. 
Email: bataringayaj@ug.afro.who.int Eliseus Rutebemberwa (BSc Psychology; Mb ChB; MPH) is a lecturer at the
Institute of Public Health, Makerere University, Uganda. He has worked in the
Ros Cooper (MA Sociology of Development) worked as health adviser for the Ugandan health sector for several years during which he has worked at the hospital level
Department for International Development (DFID) in Uganda from 1999 until 2004. and as a researcher. His current areas of research are in human resources for health and
Since January 2005 she has worked as a policy adviser focusing on human development hospital services. Email: ellie@iph.ac.ug
for DFID in the Democratic Republic of Congo. Email: ra-cooper@dfid.gov.uk
Sam Zaramba (Mb ChB; DLO; M Med; FAMS) is the Director of Health Services
Joseph Herman Kyabaggu (Mb ChB; DPH, M Med) is a long-serving Health responsible for clinical and community health services in Uganda. He is an experienced
Manager who has worked for more than three decades at various levels in the Ugandan health systems manager with a special interest in medicines and health supplies
health system. He has recently retired from the civil service post of Director of Health management and financing. As one of the top managers of the Ugandan health system,
Services (Planning and Development) at the Ministry of Health, Uganda. His areas of he has been deeply involved in the health sector reforms and supervises activities related
interest are policy analysis, health systems development and public-private partnership to the developments in medicines management and financing.
for health. Email: zarambasam@yahoo.co.uk
Peter Lochoro (Mb ChB; MSc Health Service Management) is Assistant Executive
Secretary at the Uganda Catholic Medical Bureau. His work focuses on improving
managerial capacity and performance assessment of the Catholic health providers in
Uganda and on promoting partnership with the Ministry of Health. 
Email: plochoro@ucmb.co.ug
Joseph Mwoga (B Pharm; MPS; MPH) is a public health specialist with a
pharmaceutical logistics background, currently working as Senior Pharmacist in the
office of the Principal Pharmacist, Ministry of Health, Uganda. He focuses on
strengthening supply systems for health commodities, and has a special interest in health
systems management. Email: jmwoga@hsps-ug.org
Juliet Nabyonga (Mb ChB; MSc Health Economics) works with the WHO Uganda
Country Office as a Health Economist. Her current work focuses on health system
performance assessment, health financing, and health and poverty. 
Email: Nabyongaj@ug.afro.who.int
1
The Ugandan health systems reforms: 
miracle or mirage?
Rob Yates, Christine Kirunga Tashobya, Valeria Oliveira Cruz, Barbara McPake, 
Freddie Ssengooba, Grace Murindwa, Peter Lochoro, Juliet Bataringaya, 
Hanif Nazerali and Francis Omaswa
Summary
Poor health indicators in the 1990s prompted the Government of Uganda and
development partners to embark, at the turn of the century, on an extensive programme
of health systems reforms to improve sector performance. With only a modest increase
in resources, these reforms have resulted in large increases in outputs for ambulatory
services. Out-patient attendances and immunisation rates have doubled. Furthermore,
the growth in consumption of these services appears to be highest for the poorest socio-
economic groups. However, statistics for key in-patient services, most noticeably
maternity services, remain virtually unchanged. This chapter attempts to assess the
significance of these changes. Is it a miracle of improved efficiency or a mirage unlikely
to lead to improved health outcomes? We try to identify the key reforms within health
and across government, which may be responsible for the changes in output
performance. A number of supply side reforms that have have increased the availability
of essential inputs are highlighted. In addition, a major demand side policy (abolishing
user fees) had a significant impact on the consumption of services. We conclude that
increased utilisation of ambulatory services does signify improvements in consumer
welfare and therefore health sector performance. However, stagnant maternity outputs
indicate that key in-patient services are still not meeting the expectations of the
population. Only reliable outcome data will resolve the debate about the significance of
the changes in output indicators. Further research will also be required to disentangle
the relative impact of the different components of the reforms.  
16 Health Systems Reforms in Uganda: processes and outputs 1 The Ugandan health systems reforms 17
1. Background – the Ugandan health sector pre-2001
Table 1: Stagnating health outcome indicators in Uganda in the 1990s 
For a country at the forefront of development reforms, the results from the 2001
Uganda Demographic Health Survey (UDHS) (based on 2000 data) were extremely Indicator 1995 2000 PEAP1 MDG2
disappointing (UBOS 2001). They showed that since 1995, infant mortality figures had Target Target
deteriorated and maternal mortality figures had hardly changed. As shown in Table 1, (2005) (2015)
the infant mortality rate increased from 81 to 88 (per 1,000 live births) over the period
Infant Mortality Rate 
1995-2000 and in 2000 the maternal mortality rate stood at 505 (per 100,000 live
(Deaths <1 year per 1000 live births) 81 88 68 41
births). These statistics were significantly off track for achieving the country’s own
Poverty Eradication Action Plan (PEAP) and Millennium Development Goal (MDG) Maternal Mortality Rate 
targets. (Deaths per 100,000 live births) 527 505 345 131
Whilst it is not the sole responsibility of the health sector to deliver the health
related MDGs, the sector clearly has an important role to play. These health status Source: MoFPED (2003). 
1
figures must therefore reflect, to some extent, poor performance of Ugandan health Poverty Eradication Action Plan.
2Millennium Development Goal.
services during the 1990s.  
It is probable that a wide range of factors contributed to this poor performance but
financing and management limitations have been particularly implicated. As well as
being chronically under-funded, the sector appears to have been using its limited 2. Time for radical reform
resources inefficiently. Using a variety of sources, an analysis of total expenditure on
public and private-not-for-profit (PNFP) health services at the start of this decade shows Recognising these failings, during the late 1990s, the Government of Uganda
that: (GoU) initiated a comprehensive programme of radical health sector reforms. This
included the decision to implement a sector-wide approach (SWAp) in order to improve
• The majority of the government budget (66 percent in 1999/00) was allocated coordination and therefore efficiency and equity in the sector. The SWAp was officially
to large hospitals (regional and national) and the central Ministry of Health launched in August 2000.    
(MoH), whose services tended to benefit the urban (and therefore better off) The blueprint for the SWAp has been the Health Sector Strategic Plan (HSSP) of
population, rather than to district level facilities providing primary health care 2000/01 to 2004/05. This document contains a clear statement of the health sector’s
services to the rural poor (MoH 2003a; Ssengooba et al., chapter 7 below). mission, which is to "reduce morbidity and mortality from major causes of ill health in Uganda
• Donor projects, with high overheads, focussed on investment goods and were and the disparities therein, as a contribution to poverty eradication and economic and social
inefficient at providing basic health care inputs (MoH 2003b). development of the people" (MoH 2000: 3). 
• User charges raised little revenue, exemption schemes did not protect The focus of the HSSP is to deliver a basic package of services, the Uganda National
vulnerable groups and user fees were a significant barrier for poor people Minimum Health Care Package (UNMHCP), as efficiently and equitably as possible. In
accessing services (MoH 2002; GoU 1999; MoFPED 2000). particular, there is a strong commitment to targeting poor and disadvantaged people in
line with the principles of the Poverty Eradication Action Plan. The UNMHCP consists
Due to these financing conditions, very little was spent on basic health care inputs of programmes which deliver both curative and preventive services ranging from the
(e.g. medicines, health workers’ salaries and health centre maintenance) in rural areas. control of communicable diseases, integrated management of childhood illness and
As a result, the coverage of services was limited, quality was poor and, combined with immunisation, to health promotion and education, and mental health services (see
the existence of patient charges, these services represented poor value for money. Not Annex 1 for full details).
surprisingly prospective health care consumers tended to stay away, choosing either to Since 2000, it has become apparent that the health sector reform programme has
self-treat or to attend alternative commercial sector providers (UBOS 2000). This was evolved rapidly, beyond improving coordination mechanisms envisaged in the SWAp,
reflected in low levels of utilisation for out-patient services in government and PNFP into a broader programme of health systems reforms. Many of these reforms have
health units. In the year 1999/00, the utilisation rate for out-patient services in originated in the health sector, such as improved budget allocations and medicines
government and PNFP units was only 0.42 visits per person (see Figure 1). Clearly the logistics. Others, however, have been due to the impact of public sector reforms across
Ugandan health system was not meeting the needs of its population. government. Examples here include the formulation of a nationwide Poverty Reduction
Strategy Paper (the PEAP), the introduction of a Poverty Action Fund to channel
resources to high priority budget areas, and processes to increase decentralisation to local
governments. The net result of all these reforms has been an improvement in all four key
functions of the health system as defined by the World Health Organisation (WHO),
18 Health Systems Reforms in Uganda: processes and outputs 1 The Ugandan health systems reforms 19
namely stewardship, service delivery, resource generation, and financing (WHO 2000). What was needed were output measures, where a logical case could be made that
Ministry of Health publications, notably a Mid Term Review of its HSSP and Annual increased consumption of the services concerned would be likely to lead to improved
Performance Reports, have attempted to document these changes (MoH 2003b; MoH health status. After much deliberation, in 2000, the stakeholders in the Ugandan SWAp
2002; MoH 2004a). However, given the breadth and simultaneous nature of the reforms, decided to concentrate on the following key outputs: out-patient utilisation rate,
it has been difficult to disentangle the relative impact of the diverse components. immunisation rates and the proportion of babies delivered in health units. It was felt that
Specific reforms which appear to have been most important in improving sector these measures covered an appropriate range of key curative and preventive services and
performance include: that measuring consumption would show whether these services were meeting the
expectations of the population. For all these indicators, the statistics covered services as
• SWAp processes (which have encouraged development partners to align their defined in the HSSP, which are provided by government health units and the large
support behind a coherent government-led strategy; see chapter 2, Oliveira PNFP sector.  
Cruz et al.); Figures 1-3 illustrate how these output measures have performed since the baseline
• Abolishing user fees in GoU units (which immediately triggered a surge in year of 1999/00. In the period from 1999/00 to 2003/04, the rate of utilisation of out-
demand and catalysed a number of supply side reforms; see chapter 3, Kirunga patient services in GoU and PNFP units increased from 0.42 to 0.79 visits per person per
Tashobya et al.); year (MoH 2004a), an increase of 88 percent.1 Similarly, immunisation rates for DPT32
• Improved management systems (especially in financing and supply of increased from 41 percent to 83 percent (a 102 percent rise). However, for deliveries in
medicines; see chapter 4, Nazerali et al.); health units the figure declined from 25 percent to 19 percent, only rising back to 24
• Public-private partnership (including US$9 million US$0.40 per capita in percent in the 2003/04 financial year.
new grants to PNFPs; see chapter 5, Lochoro et al.); Furthermore, research by the World Bank indicates that the financial burden felt by
• Decentralised service delivery (greater resources and capacity building in poor households for health services fell substantially between 1999/00 and 2002/03
district services; see chapter 6, Murindwa et al.); (Deininger and Mpuga 2004; see chapter 3, Kirunga Tashobya et al., for further
• Improved resource allocations (with a far larger share of sector resources discussion). While total household expenditure on health services has remained fairly
allocated to district primary health care services, including PNFP providers; see constant, the expenditure by the two poorest socio-economic groups fell by 13 percent
chapter 7, Ssengooba et al.); for the poorest group and 19 percent for the fourth income quintile. With 38 percent of
• Health financing (notably donors switching from project to budget support and the Ugandan population living below the poverty line, this result shows an improved
a reduction in the reliance on patient fees); distribution of the financial burden towards the non-poor population. Moreover, there
• Political leadership (from the President, the Minister of Health and the are signs that poor people are falling sick less often and are less incapacitated due to
Ministry of Finance). illness, which could be an early indication of improving levels of health status.
The purpose of this chapter is to stimulate debate amongst health policy makers and
academics as to whether the results of the Ugandan health systems reforms to date
signify a notable improvement in sector performance. Secondly, we attempt to highlight Figure 1: Utilisation Rate of New Outpatient Attendances in
areas which may prove to be the most important factors in bringing about the results Government of Uganda and Private Not for Profit Health Units
observed. 
3. Results of the health systems reforms
As the primary function of any health system is to improve the health status of its
population, the best indicators of sector performance measure changes in health
outcomes. Unfortunately, demographic and health statistics are only collected every five
years in Uganda so it is not possible to determine whether or not there have been any
improvements in outcomes since 2000. Charting the annual progress of the HSSP has
therefore required the use of intermediate measures of performance.
1 In terms of absolute numbers, total new attendances increased from 9.3 million in 1999/00 to 20.2 million in
2003/04. This represents a 117 percent increase in the absolute number of new attendances over the period,
with the bulk of the increase seen at government units. The percentage increase in the utilisation rate (88
percent) is lower than this due to the rising population. Source: Derived from MoH (2004a).
2 Third dose of Diphtheria-Pertussis-Tetanus vaccine.
20 Health Systems Reforms in Uganda: processes and outputs 1 The Ugandan health systems reforms 21
4. Discussion
Figure 2: DPT3 Immunisation Rates DPT3 for Children Under One Year
4.1 What is the significance of these results? 
According to Berman, "success in the provision of ambulatory personal health services,
i.e. providing individuals with treatment for acute illness and preventive health care on an
ambulatory basis, is the most significant contributor to the health care system’s performance in
most developing countries" (Berman 2000: 791). He also proposes three important criteria
against which to measure success: level and distribution of health outcomes, level and
distribution of financial burden, and population satisfaction.
Given the disease burden and resource constraints faced by Uganda, it does indeed
seem appropriate to concentrate on ambulatory services when measuring health sector
performance. McPake shows that by using a Paretian perspective of consumer
sovereignty, an increase in the consumption of services can be taken to indicate a higher
level of welfare (McPake 2002). People who have switched to public and PNFP services
from alternatives (including a no service option) have benefited, since they
demonstrably judge this option to be better than their previous choice. On the basis of
this or Berman’s criteria, a doubling in the consumption of out-patient and
immunisation services would indicate a significant improvement in population
Source: Derived from MoH (2004a).
satisfaction and/or welfare levels. 
Overall therefore, it would appear that the population has been ‘voting with its feet’,
choosing to increase its utilisation of public and PNFP health services, thereby
expressing its preference for contemporary ambulatory health care services as opposed to
the services on offer at the turn of the decade. This would suggest better performance of
the services on offer. However, it should be pointed out that this happened from a very
low baseline. Furthermore, it is unclear what proportion of the increase in the use of
Figure 3: Proportion of Babies Delivered in Government and GoU/PNFP units is due to patients switching from providers in the private sector,
Private Not for Profit Health Units although the indications are that utilisation of private-for-profit health clinics has not
declined in recent years (UBOS 2003).
The picture for deliveries in health units is different. Here the stagnant output
figures indicate the population’s indifference to any changes that might have taken place
in maternity services. Why there could be this differential uptake in services is discussed
in the following section. 
4.2 Interpretation of the results
Research has shown that health care users in developing countries, like other
consumers, shop around for health services basing their choice of provider on their
perceptions of quality and price. Studies of health seeking behaviour almost always cite
quality and price as the dominant explanations of choices made and research in a variety
of settings has shown that utilisation rates respond to changes in both (Mackian 2003;
Litvack and Bodart 1993; Bitran 1995). People choose services which, for them,
represent the best value for money. If health care providers want to increase their market
Source: Derived from MoH (2004a).
share, they have two main strategies open to them: to improve quality as perceived by
the user, and/or lower the price to the consumer.  
It could be argued that for ambulatory care services in Uganda, the health reforms
have targeted both of these strategies, by addressing both demand and supply side
22 Health Systems Reforms in Uganda: processes and outputs 1 The Ugandan health systems reforms 23
constraints simultaneously. On the demand side, the President’s decision to abolish user medicines (MoFPED 2002). However, despite these shortcomings in the data, overall it
fees in all government health units (with the exception of private wings in hospitals) would appear that, for ambulatory services, the combined effect of lower patient costs
clearly reduced costs for patients. This did not result in zero costs, as patients still had to and quality changes has been sufficient to make public and PNFP services more
incur costs associated with transport, time spent at the health facility, and possibly in attractive in terms of value for money. This has led to the large increases in expressed
paying for commodities not available at the health unit. Nonetheless, this sudden policy demand as demonstrated by the higher output figures for these services. 
change stimulated a surge in demand for public services (Nabyonga et al. 2005). Given the systems reform objective of improving efficiency, it is worth noting that
Similarly, in many PNFP units, increasing financial support from the government these changes have not been associated with a large inflow of additional resources into
enabled these providers to reduce user fees and stimulate increases in utilisation as a the sector. In fact, since the 1999/00 financial year, it is estimated that the total
result. government budget  for the health sector as defined by the HSSP (GoU and PNFP
However, demand side reforms (abolition of fees) were not implemented in services) has only increased by 18 percent in real terms (MoH 2004a). However, there
isolation. They were accompanied by a more gradual programme of supply side reforms, has been a significant switch in the composition of health financing in the sector. In
whose primary focus was to improve service coverage and quality. In particular, these accordance with SWAp principles, donors are increasingly channelling their funding
supply side measures attempted to increase the availability and quality of health care support through the government budget3 rather than through individual projects (see
inputs which appeared most important to rural populations, namely: medicines, human chapter 2, Oliveira Cruz et al.). The GoU health budget (which includes GoU funds
resources and accessible infrastructure. Medicines funding increased relative to other plus donor budget support) is now the main source of funding for the health sector,
inputs, and infrastructural growth favoured primary health care services over hospitals where previously the dominant source of funding was from donor project funding (see
(Figure 4). Some of these supply side measures are already demonstrating tangible chapter 7, Ssengooba et al., for further discussion). Figure 5 illustrates how the growing
improvements in technical service quality (MoH 2004a). For example, Nabyonga et al. GoU health budget could at least partly explain the increases in outputs and efficiency
(2005) show that despite higher demand levels, medicines availability in government for ambulatory services. Although correlations may be wrongly ascribed when they
health units has increased.  depend on continuous time trends, the graph shows that increases in GoU budget
It is difficult to assess the effect these changes have had on the quality of services as expenditure since 1999/00 have been associated with increases in new out-patient
perceived by the population, as client satisfaction data is currently limited. The second attendances at government and PNFP health units. This could be due to the
Ugandan Participatory Poverty Assessment in 2002 showed that poor people were government budget mechanism demonstrating better allocative efficiency relative to
appreciating free consultations but were still unhappy with persistent stock-outs of donor projects, particularly in financing more health care inputs at the district level. 
If the Ugandan population believes that GoU and PNFP ambulatory services in
2004 represent better value for money than they did at the turn of the decade, this begs
Figure 4: Budget allocations for medicines the question as to why the same cannot be said for maternity services. This is likely to
Figure 5: GoU Budget Expenditure and Total Outpatient Attendances
Source: MoH (2004b).
Source: Derived from MoH (2003b); MoH (2001, 2002, 2003, 2004 and 2005).
3 This includes GoU plus donor budget support. See Table 1 in chapter 7, Ssengooba et al.
24 Health Systems Reforms in Uganda: processes and outputs 1 The Ugandan health systems reforms 25
be a complicated matter but if progress is to be made in improving maternal mortality in particular. This is especially needed for areas such as maternity services where
figures this question needs to be addressed and answered. On the demand side, it is consumption is inappropriately low. How can this indicator be improved? Is it a matter
possible that potential consumers do not perceive that they need medical maternity care of improving access, service quality (if so, what aspect), or helping consumers overcome
or that for cultural reasons it is not appropriate for them to leave their homes to give residual costs? Only good quality market research will answer these questions.
birth. In addition, consuming in-patient care incurs higher costs for the household in Finally, given the ongoing debate about appropriateness of user fees as a financing
terms of transport (especially if a referral is necessary), time away from home and health mechanism, how significant was their abolition to any apparent Ugandan success story?
care inputs. On the supply side, it is likely that there have not been significant changes Would this policy have had a positive impact had it happened without the concurrent
in the quality of maternity care as perceived by consumers. Access has only increased supply side reforms? In fact, it is likely that rather than being independent factors, these
gradually and many facilities still lack qualified staff, attractive accommodation, utilities processes are mutually reinforcing. For example, it could be argued that a rising health
and functioning referral systems. Available evidence indicates that the population do budget and ongoing supply side reforms encouraged the government to take the bold
not perceive overall value for money to have significantly changed for maternity services step to abolish fees. Alternatively, the sudden abolition of patient fees and the
and thus they still tend to give birth at home. This suggests that the population view immediate surge in demand may have promoted health up the government agenda and
maternity services as a very different product to simple out-patient care. in effect helped catalyse the fledgling health reforms. Given growing concern about user
In summary, it would appear that the systems reforms identified here have made fees in developing countries, any further research into the impact of this specific policy
sufficient changes to ambulatory services to trigger a rise in demand but that maternity change which could then be applied to different contexts would be most welcome. 
services are proving more difficult to improve. Greater attention must be paid to the Despite the limitations in the data available to date, what can be said with a degree
specific demand and supply constraints affecting these services if maternity outputs, and of confidence is that the Ugandan health sector has changed significantly since 2000.
therefore maternal and infant health outcomes, are to improve.  Firstly, the way in which the sector is being managed and financed has changed
It should be noted that the analysis in this chapter utilises to a large extent data from appreciably. There is now a greater reliance on government systems, especially budget
the Ministry of Health’s management information systems. As with most developing financing, and a lesser role for development projects and funding from households. More
country national systems, there are limitations to the accuracy and completeness of importantly, and due to the above, in 2004, government and PNFP health units were a
these data. Nonetheless, the output figures are consistent with those provided by other lot busier, providing twice as many ambulatory services than before. In this respect, it
independent sources, thereby providing useful validation of the MoH data (Nabyonga et appears that the most important health care product launched by the Ugandan health
al. 2005). systems reforms has been successful.
5. Conclusions
Most commentators would agree that the Ugandan health reforms can only be
deemed to be truly successful when they can demonstrate that they have contributed to
improvements in health care outcomes. Given the poor 2001 UDHS figures, there is
certainly considerable scope to do this. However, health outcomes are a multi-sectoral
responsibility and changes in other sectors’ performance will be difficult to disentangle.
On the positive side, Uganda’s largely successful implementation of a universal primary
education system is likely to benefit health outcomes in the long-term. However,
working in the opposite direction, the National Household Survey for 2002/03 has
reported rising poverty levels since 2001 (UBOS 2003). Furthermore, worsening
insecurity in the north of the country is likely to have a profound effect on health status
figures, particularly for the one million plus Ugandans currently living in camps for
internally displaced people. Given these latter constraints, it would perhaps be a
successful outcome if Uganda managed to simply maintain its overall health status
statistics while better performance might be looked for in districts unaffected by conflict.
Rather than waiting until the next UDHS, it is hoped that further research in the
interim can shed more light on the impact of the Ugandan health system reforms. For
example, have the large increases in immunisation rates reduced mortality rates from
communicable diseases in children? In addition, there is clearly a need for further
research into what factors are driving the demand for health services, and by poor people
26 Health Systems Reforms in Uganda: processes and outputs 1 The Ugandan health systems reforms 27
References Annex 1: Health Sector Strategic Plan and structure of Ugandan health sector
Berman P. 2000. Organization of ambulatory care provision: a critical determinant of health system The programmes which make up the Ugandan Minimum Health Care Package (UNMHCP)
performance in developing countries. Bulletin of the World Health Organisation, 78 (6): 791–802. consist of curative and preventive services including the following:
Bitran R. 1995. Efficiency and quality in the public and private sectors in Senegal. Health Policy and Planning, 1. Control of Communicable Diseases: Malaria; STD/HIV/AIDS; Tuberculosis
10 (3): 271–83. 2. Integrated Management of Childhood Illness
Deininger K, Mpuga P. 2004. Economic and welfare impact of the abolition of user fees: Evidence from 3. Sexual and Reproductive Health and Rights
Uganda. World Bank Working Paper 3276. The World Bank, Washington DC. 4. Immunisation
5. Environmental Health
GoU 1999. Report of the Inter-ministerial taskforce on cost sharing. Government of Uganda. 6. Health Education and Promotion
Litvack J, Bodart C. 1993. User fees plus quality equals improved access to health care: results on a field 7. School Health
experiment in Cameroon. Social Science and Medicine, 37 (3): 369–83. 8. Epidemic & Disaster Prevention, Preparedness and Response
9. Improving Nutrition
Mackian S. 2003. A Review of Health Seeking Behaviour, Problems and Prospects. Health Systems Development 10. Interventions against diseases targeted for elimination or eradication
Programme, University of Manchester, UK. 11. Strengthening Mental Health Services
McPake B. 2002. The globalisation of health sector reform policies: is "lesson drawing" part of the process? In: 12. Essential Clinical Care
K Lee, K Buse and S Fustukian (eds.) Health Policy in a Globalising World. Cambridge University Press,
Cambridge. The Health Sector Strategic Plan (HSSP) also formally establishes the ideal structure for the
Ugandan health sector which will deliver the UNMHCP (see below). It should be noted that one
MoFPED. 2000. Uganda Participatory Poverty Assessment Report: Learning from the Poor. Ministry of Finance of the important principles of the HSSP is that this overall structure should comprise public and
Planning and Economic Development, Government of Uganda. private-not-for-profit (PNFP) providers working in a close partnership.
MoFPED. 2002. Deepening the understanding of poverty. Second Participatory Poverty Assessment Report. Ministry
of Finance Planning and Economic Development, Government of Uganda. Level Health Population Services Provided
Centre (approx.)
MoFPED. 2003. Report of the Taskforce on Infant and Maternal Mortality. Ministry of Finance Planning and I Village - 1,000 Community-based preventive and promotive 
Economic Development, Government of Uganda. health services. Village Health Committee or 
MoH. 2000. Health Sector Strategic Plan 2000/01 to 2004/05. Ministry of Health, Government of Uganda. similar status.
MoH. 2002. The Uganda Health Financing Strategy. Ministry of Health, Government of Uganda. Health II Parish - 5,000 Preventive, promotive and out-patient curative 
Sub- health services, and outreach care.
MoH. 2003a. Mid Term Review Report of the Health Sector Strategic Plan 2000/0–2004/05. Ministry of Health, District
Government of Uganda. III Sub-county - Preventive, promotive, out-patient curative,
20,000 maternity and in-patient health services and 
MoH. 2003b. Annual Health Sector Performance Report 2002/2003. Ministry of Health, Government of laboratory services.
Uganda.
District IV County - 100,000 Preventive, promotive, out-patient curative, 
MoH. 2004a. Annual Health Sector Performance Report 2003/2004. Ministry of Health, Government of Uganda. maternity, in-patient health services, emergency 
MoH. 2004b. Ministry of Health, Government of surgery, blood transfusion and laboratory Health Sector Poverty Eradication Action Plan Review Paper.
Uganda. services.
V General Hospital – In addition to services offered at health centre
MoH. 2001, 2002, 2003, 2004, and 2005. District Transfers for the Financial Year. Ministry of Health,
Government of Uganda. 500,000 level IV, other general services are provided 
including in-service training, consultation and 
Nabyonga N, Desmet M, Karamagi H, Kadama P, Omaswa F, Walker O. 2005.  Abolition of cost-sharing is research for community-based health care 
pro-poor: evidence from Uganda. Health Policy and Planning, 20 (2): 100–108. programmes.
UBOS. 2000. Report of the Uganda National Household Survey for 1999/2000. Uganda Bureau of Statistics. Regional VI Regional Referral In addition to services offered at the general
Hospital - hospital, specialist services are offered, such as
UBOS. 2001. Uganda Demographic and Health Survey 1999/2000. Uganda Bureau of Statistics.
2,000,000 psychiatry, Ear, Nose and Throat (ENT), 
UBOS. 2003. Report of the Uganda National Household Survey 2002/03. Uganda Bureau of Statistics. ophthalmology, dentistry, intensive care, 
radiology, pathology, higher level surgical and 
WHO. 2000. Health Systems: Improving Performance. World Health Organisation, Geneva. medical services.
National VII National Referral These provide comprehensive specialist services 
Hospital – and are also involved in teaching and research.
24,700,000
2
Is the sector-wide approach (SWAp) improving
health sector performance in Uganda?
Valeria Oliveira Cruz, Ros Cooper, Barbara McPake, Rob Yates, Freddie Ssengooba,
Francis Omaswa, Christine Kirunga Tashobya and Grace Murindwa
Summary
International donors provide the major source of development assistance to
Uganda. Until recently, for historical reasons and because of weaknesses in national
policies and structures, development assistance had been largely organised through
stand-alone projects. This modality of aid delivery was seen as an essential approach to
allow flexibility to manoeuvre, quick response, and demonstrable results in key priority
health concerns. However, project-based support has been criticised for causing
fragmentation of the health system into several sub-systems with differing capacities,
delivering different health interventions to different beneficiaries. Concerns for equity,
efficiency, and government leadership have since led to the introduction of the sector-
wide approach (SWAp) in Uganda. The main objective of the SWAp is to improve the
overall performance of the health system and consequently the health status of Uganda’s
population. To this end, it aims to coordinate development assistance to the sector and
to reduce the administrative burden for government. This chapter analyses the
evolution of the SWAp as one of the key reforms undertaken in Uganda since 2000,
describing the structures and processes involved. Benefits of this reform include the
establishment of a platform for coalition building and policy learning among
stakeholders and improved donor coordination. However, the SWAp faces important
challenges, including the burden of persistent under-funding to the health sector and the
renewed interest in vertical approaches to health sector funding arising from the
introduction of new global health initiatives. These may adversely affect the SWAp
objectives and risk destabilising the significant progress made in the health system in
Uganda since 2000.
30 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 31
1. Introduction Essentially, a SWAp should enable governments to own and coordinate
development assistance in a given sector in the following spheres: policy design, strategic
Traditionally, in international development assistance for health, projects have been and operational management, financial pooling, resource allocation and common
seen as the preferred approach (Peters and Chao 1998) to respond quickly, allow arrangements for monitoring and evaluation (Oliveira Cruz and McPake 2004).
flexibility to manoeuvre or circumvent weak health systems, and to yield demonstrable Effective coordination of these spheres should achieve the following:
results for selected high priority concerns (Oliveira Cruz et al. 2003). Often, projects
focus on the implementation of interventions related to a particular disease or • Higher degree of government control over plans and the development agenda;
condition, such as polio or malaria (Ssengooba et al. 2004), or in a specific geographical • Substantial reduction in administrative costs in dealing with different 
area, and are characterised by a vertical or categorical mode of delivery. development partners’ (DPs) systems and requirements;
However, provision of project-based support has been criticised for causing • Flexibility in the design of strategies to accommodate local conditions and 
fragmentation of the health system into several sub-systems with differing capacities and build the capacity of the overall health system;
delivering different health interventions to different beneficiaries (Cassels and Janovsky • Greater technical and allocative efficiency of the overall resource envelope; 
1998). For example, the situation in Uganda during its post-conflict  rehabilitation in and
the 1990s is described by Macrae et al. (1996) as follows: proliferation of projects rather • Improved management and logistics systems.
than a coherent health policy, relative dominance of vertical programmes, unsustainable
escalation of recurrent costs associated with projects, poor coordination and inequitable The SWAp aims to take a holistic perspective of the whole health sector. The
distribution of aid resources, skews in service provision in favour of selected overall goal of the health SWAp in Uganda is to improve the performance of the health
interventions and urban areas, high levels of aid dependency, limited capacity building, system and consequently the health status of the population (MoH 2000a). A SWAp
and limited institutional development at both national and local levels. The problems may be interpreted in narrow terms, as a financing instrument where the government
associated with uncoordinated projects are also noted in other chapters in this book benefits from a basket of funds contributed to by different DPs. This basket would
(chapter 1, Yates et al.; chapter 7, Ssengooba et al.; and chapter 4, Nazerali et al.). represent the essence of the SWAp partnership. In Uganda however, the health SWAp
Hence, the introduction of the sector-wide approach (SWAp) in Uganda has been involves more than just a basket approach to funding. The focus is on the sector
motivated by concerns for government leadership, efficiency (Peters and Chao 1998; investment plan, entitled the Health Sector Strategic Plan (HSSP) and the supporting
MoH 2000a), and equity (MoH 2000a), with the aim of improving sector performance National Health Policy (see Annex 1 in chapter 1, Yates et al.). Hence, the SWAp is a
by means of coordinating development assistance and thereby reducing administrative platform for adopting a comprehensive approach to implement the HSSP by harnessing
costs for government. all available resources (human, financial, projects etc.). 
The SWAp is a relatively new approach, compared with decades of history of Another characteristic of the Ugandan SWAp is a strong focus on consensus
implementing projects. Therefore, stakeholders are keen to gather evidence of the building among stakeholders. Based on the consensus approach, the SWAp takes a
impacts of this strategy on the implementation of national policies and plans. Several flexible line towards channelling donor funds to the sector. Funds can be provided in the
countries are implementing SWAps e.g. Zambia, Ghana, Pakistan, Bangladesh (Cassels form of support to the budget, projects (at national and district level), and technical
and Janovsky 1998, Mozambique, Tanzania, Cambodia, Vietnam (Foster et al. 2000) or assistance. 
are considering adoption of different forms of SWAp.  The Ugandan version of the The SWAp objectives need to be analysed critically to determine the extent to
SWAp has several notable features worth documenting, particularly after the extended which they contribute to the improved health sector performance documented in
period of its implementation (since 2000). These may help inform both national and chapter 1 by Yates et al., particularly in regard to improving the harmonisation of
international debates and decision-making. This chapter intends to assess and document priorities, reducing fragmentation and duplication of efforts, increasing funding levels,
the evolution of the SWAp in Uganda and to analyse the extent to which it is and promoting sustainability of programmes.
contributing to improving the performance of the sector.
2. Definition and objectives of the SWAp 3. Uganda’s SWAp experience to date
1
"The defining characteristics of a SWAp are that all significant funding for the sector 3.1 Mechanisms and processes
supports a single sector policy and expenditure programme, under government leadership,
adopting common approaches across the sector and progressing towards relying on government The Memorandum of Understanding (MoU) between the Government of Uganda
procedures to disburse and account for all funds" (Foster et al. 2000: 1). Hence, donors’ (GoU) and DPs establishes the overall principles and mechanisms governing the SWAp
support to a SWAp is characterised by a more comprehensive approach to aid delivery partnership (MoH 2000b). The SWAp was officially launched in Uganda in August
with the funding of a coherent government-led plan of activities rather than managing 2000. While the MoU was signed by the Ministry of Health (MoH) on behalf of the
their own discrete projects. 
1 Annex I provides a detailed description of these.
32 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 33
GoU, and by various development partner agencies, a wider range of stakeholders are (2005).  For example, districts are not represented at the monthly HPAC meetings or at
actively involved in its implementation. These include the Ministry of Finance, the meetings of the SWG where important resource allocation and budgeting decisions
Planning and Economic Development (MoFPED) and the Ministry of Local are taken. Nevertheless, this situation may improve over time as both the MoH and DPs
Government, and representatives from civil society (e.g. private-not-for-profit health continue to strive to provide better support to districts in order to improve performance
organisations). This is in line with the broad definition of the SWAp in Uganda. at service delivery points (for further details see chapter 6, Murindwa et al.).
Policy advice, priority setting, decision-making, strategic management, monitoring
and evaluation (performance monitoring) are carried out on a yearly basis, through Joint 3.2 Products 
Review Missions (JRMs), and followed up on a monthly basis, through the Health Policy
Advisory Committee (HPAC). The JRM is an annual forum to which all stakeholders Possibly the main achievements of the SWAp have been to enable the
to the health sector are invited and the HPAC includes representatives from the MoH, strengthening of the budget support mechanism (i.e. donor funding directed to the GoU
other Ministries, DPs and other health partners (e.g. PNFP sub-sector). While the JRMs health budget rather than to specific projects) as well as to contribute to improved
and HPAC also have financing functions, these responsibilities are largely delegated to allocative and technical efficiency in the sector, as demonstrated in chapter 7 by
the Sector Working Group (SWG) which reviews budget priorities, agrees on broad Ssengooba et al. From the outset of the first HSSP (2000/01–2004/05), the number of
resource allocations, and decides on the project portfolio for the sector as a whole. DPs providing budget support to the government health budget increased from five (UK,
Discussions and decisions at the inclusive JRMs and HPAC meetings are informed by Ireland, Sweden, Belgium, and the World Bank) to eight (the previous five apart from
the activities of individual Working Groups (WGs). These WGs streamline the Belgium plus EU, Norway, Netherlands, and Denmark) (MoH 2003a). These donors
operation of areas such as integrated support systems (human resources, drugs and have gradually phased out their previous project activities and are now channelling their
supplies, and financing) by identifying problems, proposing action and following them funds directly to the GoU health budget. While these DPs increasingly trust the efforts
up. For technical programmes within the MoH, Interagency Coordinating Committees and commitment of the GoU, they also stress the need to improve existing
(ICCs) perform the functions of priority setting for key programme interventions, accountability mechanisms. The functioning of the above-mentioned SWAp structures
developing work plans and coordinating projects in view of national level objectives. (e.g. HPAC and JRM) seems to be providing confidence to more DPs that their
These mechanisms and processes have been playing an important role in the resources are being channelled efficiently according to the plans and agreed priorities. 
Ugandan SWAp. For example, the HPAC is seen as a crucial forum for negotiation and Another product of the SWAp is the development of a comprehensive performance
policy dialogue and agreement of plans and undertakings among stakeholders in the monitoring system which is used by all DPs (although some partners still require
health sector. It meets regularly, serves as an opportunity to "discuss and resolve issues additional elements to the overall system for their own accountability purposes). Before
openly", and has contributed to building trust between the GoU and DPs (MoH 2003a: the SWAp, the MoH and districts had to deal with a range of separate supervision and
62). ICCs are credited with facilitating programme areas to identify their core evaluation missions organised by DPs. For example, at one time alone, the MoH (not
interventions according to the sector’s overall priorities, as opposed to responding to counting districts) had about 45 different projects running with different monitoring
project or development partners’ agendas. They have also helped programme areas to and evaluation requirements. This limited efforts at national and district levels to gather,
streamline their functions with those of the health sector integrated support systems. consolidate, analyse and disseminate information regarding performance of the system
There is, however, room for improvement. For instance, the HPAC could benefit overall. In this respect, the MoH and the DPs agreed on a monitoring framework for the
from a more strategic and policy-oriented focus and concentrate less on operational HSSP (based on a selected list of performance indicators), and the strengthening of
issues (MoH 2003a), and could be used more consistently as a performance monitoring mechanisms for data collection and dissemination. Under the SWAp, emphasis was
mechanism (MoH 2001; MoH 2003b). In addition, the relationship between different placed on the preparation of Annual Health Sector Performance Reports. Although the
structures, such as that between the ICCs and HPAC, needs to be streamlined and first reports were considered to be of poor quality, with programmes reporting from an
formalised (MoH 2003a). These sorts of issues and problems are discussed, together with implementation perspective (e.g. number of workshops held), the most recent one
a general assessment of the status of partnership between the MoH and DPs, at yearly (2003/04) has been commended for providing a good synopsis of the sector’s
review meetings (the JRMs) of the SWAp. performance with regard to key outputs at the central and local levels of government. 
Finally, it is worth mentioning the linkage between the SWAp and the operational During SWAp implementation, different undertakings2 have been agreed between
level. Given the decentralised nature of service delivery in Uganda, districts are DPs and the MoH to improve integrated support from the national to the district level.
responsible for the implementation of health prevention and care activities. While More specifically, the Monitoring and Supervision WG analysed and identified priority
district representatives participate in various SWAp-related structures (e.g. JRMs and actions which have resulted in the creation of area teams and district league tables
WGs), these structures remain largely under the control of the centre, and the extent to (discussed by Murindwa et al. in chapter 6). These are considered valuable support
which the districts play an active role in these structures is questionable. This tension, structures and monitoring tools for the health sector. 
between the increased powers held at the national level through the SWAp process and
the pursuit of higher levels of devolution at district level, was highlighted at the early 2Undertakings are actions or processes in a specific area agreed during a JRM between the GoU and DPs to be
stages of the SWAp in Uganda by Jeppsson (2002) and more recently by Elsey et al. given priority during the year. Progress towards the achievement of undertakings is reviewed during the
subsequent JRM. For a number of donors, successful outcome of the JRM and achievement of the undertakings
determines the release of funds to the budget. 
34 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 35
A further product that has resulted from the SWAp partnership is the series of When DPs provide direct district support (through projects or technical assistance
tracking studies.3 These studies are agreed on during JRMs and progress relating to the for example) or at project level, the focus of attention lies on a relatively narrow range
studies’ recommendations are followed up by the HPAC. While these studies may be of issues. Although bigger problems, such as difficulties in financial flows from the
seen to have a quasi audit function, and in fact they are part of the fiduciary requirements central government to the districts and/or limited human resource capabilities at the
of some DPs, they are envisaged as a broader type of audit, answering questions such as health facilities, can still be identified in this project-specific domain, little or no action
‘why is it not working?’ and ‘where are the constraints?’ Thus these studies allow an in- can actually be taken to address such crucial systems constraints within the project
depth assessment of problems, formulate recommendations for action, and serve as framework. 
opportunities to build consensus for these actions to be carried out, instead of In addition, a better understanding of the problems affecting the sector in general,
functioning as narrow or internal types of audit. For example, disbursements of funds and an appreciation of those that lie outside the sector, helps stakeholders to improve
improved after the follow-up by MoH of recommendations from the study of financial inter-sectoral collaboration. For example, health workers not being paid on time and on
flows from central government to districts. It was possible to decrease the average time a regular basis by districts prompted stakeholders in the health sector to work with the
delay for the flow of funds from the MoFPED to service delivery points in districts from Ministry of Public Service to recentralise and computerise the payroll in 2001/02 (see
75 to less than 30 days over a two-year period (MoH 2003a). chapter 6).
Finally, the SWAp prompted the creation of the Health Development Partners’ The nature of SWAp interactions and responses has also given its stakeholders a
Group (HDPG). This forum allows for information sharing, discussion, consensus combined voice and stronger bargaining power. For example, during the budget
building and joint decision-making amongst DPs. This enables them to ‘come with a negotiation process for 2004/05, health spending was expected to suffer a real per capita
single voice’ to their negotiations with the MoH, thereby strengthening the group but reduction of 11 percent relative to the previous year’s budget (MoH 2004). Through
also helping to focus their engagement with, and support to, government. Donor combined concerted efforts, the MoH, DPs and other stakeholders successfully restored
coordination also facilitates the process of alignment of project-based support to HSSP the allocations to the health sector (MoFPED 2004a). 
priorities and makes project delivery less disruptive to the health system as the HDPG Ssengooba et al. argue in chapter 7 that the SWAp has enabled national level
encompasses both project and budget support donors. This is the case, for example, with decision-making, and a more equitable allocation of resources which accords with the
the development agencies for Italy, Japan, and the USA, who are active members of the priorities set out in the health sector plan. Sector objectives are now clearer as a result
HPAC and the JRM despite being project rather than budget support donors. The USA of developments such as the reporting, monitoring and evaluation system, which was
even chaired the HDPG for a period in 2003/04. enabled and improved through the SWAp. 
An illustrative case study of SWAp benefits is given by the reform of the medicines
3.3 Benefits  financing and logistic systems leading to improved technical efficiency in this area (see
chapter 5). Such reform was facilitated by the SWAp which provided a framework for
From the above, it seems that the SWAp provides an opportunity to assess the the required policy consensus and coordination of efforts. First, the JRMs acknowledged
system as a whole and hence to recognise constraints which would not be feasible that medicines financing was pivotal in improving health sector outputs, given the
through other, more fragmented, ways of working. The SWAp has also facilitated the investment already made in the area of infrastructure and human resources (see chapter 6)
process of coalition building and joint policy learning. Illustrations of these elements of and the increased demand for medicines after user charges were abolished (see chapter
the SWAp are the agreed commitments and decisions taken during JRMs. The JRM 3). Second, the WGs facilitated collaborative planning amongst the numerous
offers a very distinctive opportunity for a wide range of stakeholders to share their stakeholders, and provided ownership and management of incremental change using
performance, experiences, views and knowledge, thereby contributing to a more in- realistic strategies and practical measures at the operational level. Third, the Drug
depth and open understanding of the system and the reasons why certain elements are Tracking Study played an important role in both monitoring the implementation and
not working well. It also facilitates decision-making regarding which problem areas need impact of the changes introduced and providing data on the aspects needing attention
specific examination (tracking studies) and agreement on undertakings for the coming (MoH 2002; and see chapter 4).
year. In addition, given the frequency with which stakeholders meet under the SWAp
arrangement, a gradual learning process about the system has been enabled, helping to
develop further consensus and trust among stakeholders. 
3So far four studies have been completed, one on financial flows from the central government to district health
services, and a second on drug procurement and management. A third one on procurement of supplies and
central activities (by the MoH on behalf of districts) was presented early in 2004 and its recommendations are
being followed up. A fourth study on human resources has been completed and its results have been presented
to HPAC. A fifth study on health infrastructure is underway.
36 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 37
4. Challenges to SWAp implementation agreed policies (e.g. the way the system operates, its history, the organic nature of its
processes). If the objectives of harmonisation of priorities and reduction of
There are two sets of challenges to SWAp implementation in Uganda. The first fragmentation are to be achieved, a renewed global interest in projects represents a
refers to system-wide challenges (e.g. financing) and the second are intrinsic to this major threat to the SWAp. 
approach to aid. System-wide challenges are discussed elsewhere in the book, while this Another dilemma for the health sector is how to align the potentially competing
chapter focuses on this second set of challenges. goals of channelling more funds through the budget, with the pressing need to augment
Efforts to scale up resources available from the international community for resources for the sector in light of increasing project-based aid. In this context, high
reaching the Millennium Development Goals (MDGs) are evidenced by the profile, externally driven project aid (such as the global health initiatives) has to be
establishment of the Global Fund to Fight AIDS, TB and Malaria (GFATM) and the evaluated in terms of two country-level constraints to expanding health sector funding
US-sponsored President’s Emergency Plan for AIDS Relief (PEPFAR). However, these through the budget. The first is the imposition of budget ceilings on all sectors by the
new large-scale initiatives for the delivery of aid follow a project approach (in a disease- MoFPED, as of 2004/05. The health sector has to accommodate all project funding and
specific vertical form), with unprecedented amounts of resources channelled into the budget support under its established ceiling, as an element in the ongoing efforts to
4
country at a fast pace with strong pressure to produce results in a short period of time. ensure macroeconomic stability of the country and improved aid efficiency. 
They are thus likely to put stresses on still weak health systems. The second constraint, of a political economy nature, concerns the general desire of
Green and Collins (2003) highlight the potentialy disruptive nature of funding and the GoU to reduce the proportion of aid (budget support or project aid) to domestic
accountability arrangements developed by development partners that are not in line revenues, which is close to 47 percent (MoFPED 2004b). The motivation for pursuing
with the values of a SWAp. Project funding is usually better resourced and more focused this goal is related in part to the GoU’s assertion of its sovereignty. It is also related to
(on specific diseases, health system issues or geographical locations) than funding for the the government’s concerns regarding the (un)predictability and (un)sustainability of aid
SWAp through the government budget. Coordination structures, delivery mechanisms funds into the country. While agreements are signed over 3 to 5 year frameworks, the
and timeframes, budget cycles and objectives may be out of sync with those used by the yearly flows of aid are still unstable. In this context, project aid may be difficult to
rest of the SWAp partners. Limited resources can be diverted, such as when health incorporate into the health sector budget and is likely to displace more efficient and
workers employed by the public and private-not-for-profit sectors take better paid flexible resources channelled through the budget mechanism.
project posts, or when management time is required to cover separate coordination and The above-mentioned constraints seem to point to the conflict of technical
implementation structures, as was the case when both the GFATM and PEPFAR were objectives between the health sector, to increase its budget in order to adequately fund
originally launched. In addition, policies can be undermined. For example, the delivery the HSSP, and the government (in this case the MoFPED and Cabinet), to reduce the
of anti-retrovirals (ARVs), a key element in PEPFAR and GFATM activities, was not fiscal deficit and the proportionate volume of aid in Uganda (in spite of their
included in the priorities of the first HSSP (2000/01–2004/05) and there is evidence of international commitments to reach the MDGs). Although the MoH is part of Cabinet,
duplication and disruption as a result (MoH 2003c). its demands must compete with those of several other line ministries. The MoFPED
Further, the delivery of ARVs through projects implemented by non-governmental clearly places its major emphasis on the achievement of a more coherent national
organisations is taking place to a large extent without consideration of an equitable planning and budgeting process in the context of a medium-term expenditure
geographical distribution. Currently, anti-retroviral therapy (ART) sites are framework, taking into account all sectors in the budget allocations. However, an
concentrated in Kampala and Masaka, representing a disproportionate concentration overlooked issue appears to be the need to acknowledge the specificities of the different
compared to other parts of the country which are harder to reach and in great need, such sectors. For instance, a large proportion of health expenditure has a high import content
as the conflict ridden North (ART Coordination Committee 2004).  (e.g. drugs and equipment) where increased expenditure will have a less direct adverse
Another problem has been that while the Ugandan government’s policy is to impact on the macroeconomy relative to other sectors.
provide co-formulated, single tablet, cheap, generic triple therapy for ARVs, US projects Another significant challenge to the SWAp refers to its reliance on individual, as
adopt branded multi-tablet ARVs that are much more expensive (ART Coordination opposed to institutional, vision and commitment to its principles. The SWAp is a
Committee 2004). This could potentially create confusion for the beneficiaries of the process-based approach in which individuals have an important role to play out by Walt
different programmes, which in turn could impact on patient compliance with the et al. (1999) as pointed at all stages. This greater dependence on individuals has higher
treatment, and also means that scarce resources are not being used as efficiently as they relevance in the context of developing countries, which lack strong institutions. It is
could be. essential that individuals, part of the system, believe and act according to SWAp
Such international initiatives also have a strong political dimension. Decisions no principles without giving in easily to pressures and challenges. Uganda has so far
longer lie at the technical level (where the emphasis is on harmonisation with the
health system) but may be agreed at the highest levels of government and their agendas
carried forward (e.g. preference for service delivery through faith-based organisations).
These types of international initiatives often have their decisions taken outside of the 4The rationale underpinning this refers to concerns regarding the potential negative effects of high inflows of aid.
country, and hence lack knowledge of local realities and are not in line with locally These include the potential for appreciation of the real exchange rate and decreased exports, fiscal deficits, high
interest rates and reduced private investment impinging on productivity and growth (for further details see Adam
and Bevan 2002; Lake 2004; and MoFPED 2004c).
38 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 39
benefited from a remarkable mix of individuals on the government as well as on the Clearly this begs the question as to whether the SWAp constitutes a reform in itself
development partners side, who have shown strong character, leadership skills, vision, or if its raison d’être is to facilitate reforms. In the Ugandan context it seems that the
commitment and reform spirit as well as operational abilities to transform vision into SWAp plays both roles. It did represent a substantive shift from a fragmented system,
practical steps. But individuals come and go. Hence the replacement of individuals in where there was no coherent sector policy, strategic plan nor a consolidated planning,
the MoH or the DP group may threaten the SWAp partnership if newcomers do not budgeting or capacity building approach at national and district levels. Instead there
understand the essence of it, are not committed, or are too familiar and perhaps attached were large projects such as the Delivery of Improved Services for Health Project (DISH)
to the incentive structure of the project mode of funding. The project incentive funded by USAID or the District Health Services Pilot and Demonstration Project
structure particularly benefits the national level, emphasising project implementation funded by the World Bank. 
units and related equipment, high salaries and foreign travel, and lacks transparency The way in which the SWAp is defined and implemented in Uganda, being a holistic
(e.g. methods of recruitment). In contrast, the incentive structure of the SWAp and platform for aid coordination, funding and operation of the HSSP, suggests it is also a
budget support modes lies at the operational level (districts) with a greater element of means to an end. The main goal of the SWAp is to improve health system performance.
transparency regarding how funds are used. Within the MoH, technical groups are still We argue that the systems improvements demonstrated in the other chapters of this
divided between the incentive structures of project and SWAp modes. Increased funding book have been facilitated by the SWAp, as it provides an opportunity to work towards
through projects represents another threat to the SWAp. identifying system-wide problems and potential solutions (within and outside the
Related to the above is the problem of commitment in the political sphere. The sector). 
pursuit of different political agendas, particularly in times of political ‘stress’, may not be The SWAp, as shown in particular by the case of medicines supply reform (chapter
in line with the health sector’s priorities. This can represent a bottleneck in the process 4, Nazerali et al.), has contributed to the reduction of fragmentation and duplication of
of moving forward agreed reforms between like-minded GoU technocrats and DPs in the efforts. It also played an important role in achieving stronger harmony of priorities and
absence of political support. An illustration of this issue is the current debate in Uganda improved planning and monitoring processes at district and national levels.  By bringing
on whether or not to introduce indoor residual spraying as an intervention for malaria together all key stakeholders in the sector, building consensus on policies and processes,
control. At the technical level, given the type of malarial transmission in Uganda, the gathering better information, and working jointly towards agreed common goals, the
World Health Organisation and other experts consider this intervention not to be SWAp has given the health sector a combined voice and stronger bargaining power. 
particularly effective (EARN 2003; Root et al. 2003). However, at the political level, The SWAp has also facilitated the improvements in allocative efficiency of the
this intervention is considered to have popular appeal and is believed to deliver a quick GoU budget, which have occurred over the period of its implementation (see chapter 7,
fix to a major health problem in the country. Ssengooba et al.). 
All this suggests that the SWAp has played a key role, among other reforms, in
facilitating improved health sector performance. Because of this, attention to addressing
5. Conclusions the challenges outlined in this chapter is urgent. The objectives and essence of the
SWAp in Uganda need to be maintained and its mechanisms and processes
Since 2000, the implementation of the SWAp by means of the mechanisms and strengthened. However, continued achievements at sector level will not be possible if,
processes presented in the previous section, guided by the general principles of the for instance, initiatives such as the GFATM and PEPFAR are not harmonised with local
partnership as stated in the Memorandum of Understanding (MoH 2000b), and by the policies, systems and capacities. The way in which these projects are being implemented
vision and belief of individuals committed to this approach, points to important seems to be destabilising, even in the context of a flexible and inclusive SWAp that
products and benefits for the health sector. As highlighted by Yates et al. in chapter 1, accepts both projects and budget support.
it is very difficult to disentangle whether these products and benefits are directly The SWAp has represented a major change to the way in which the health system
attributable to the SWAp or to other reforms undertaken concomitantly in Uganda is governed, and one which it has been suggested offers the best chance of delivering
(discussed in the other chapters of this book in more depth). improved health status to the population. However, given the challenges discussed in
Hutton and Tanner (2004) note the lack of standards for assessing the tangible this chapter, the achievement of that long-term goal seems uncertain.
benefits of a SWAp. They suggest there are, even if indirectly, four key areas5 that if
strengthened could help to understand the impact of a SWAp on population health.
Within this framework, the full impact of a SWAp can be measured in terms of health
outcomes, although these can only be assessed over sustained periods of implementation
(5–10 years).6
6Moore (2003) compared 69 countries classified in terms of those that were implementing a SWAp and those
not. Among other indicators assessed, the results showed that SWAp and non-SWAp countries did not present
any significant average differences in their 1999 health outcome indicator (disability adjusted life expectancy –
DALE). However, the author recognises that the result is predictable since the implementation of SWAp by these
5 1) Country leadership and ownership, 2) institutional and management capacity, 3) flow of resources, and 4) countries was at its early stages. It is also worth noting that DALE results presented in 1999 reflect the health
monitoring and evaluation. sector situation of those countries dating back to even earlier periods than the introduction of SWAp.
40 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 41
References MoH. 2000b. Memorandum of Understanding Government of Uganda and Development Partners. Mimeo. Ministry
of Health, Government of Uganda.
Adam C, Bevan D. 2002. Uganda. Aid, Public Expenditure, and Dutch Disease. Paper commissioned by DFID. MoH. 2001. Conclusions from one day retreat MoH / DPs, Fairway Hotel. Ministry of Health, Government of
Mimeo. Accessed on 26 July 2004. Available at Uganda.
[http://www.economics.ox.ac.uk/Members/david.bevan/Papers/AidFlowDutch.pdf]
MoH. 2002. Drug Tracking Study. Final Report, August 2002. Ministry of Health, Government of Uganda.
ART Coordination Committee. 2004. Notes from ART coordination committee meeting. Kampala, Uganda.
MoH. 2003a. Mid Term Review Report of the Health Sector Strategic Plan 2000/01–2004/05. Ministry of Health,
Cassels A, Janovsky K. 1998. Better health in developing countries: are sector-wide approaches the way Government of Uganda.
forward of the future? The Lancet, 352 (9142): 1777–79.
MoH. 2003b. Outcome of SWAp Coordination Structures. Extraordinary meeting of HPAC. Ministry of Health.
De Loor R, Hutton G. 2003. Review of the Health Sector Partnership Fund in Uganda. DFID Health Systems Government of Uganda.
Resource Centre. Mimeo. Department for International Development, UK.
MoH. 2003c. Aide Memoire of the 9th Joint Review Mission. Ministry of Health, Government of Uganda.
EARN 2003. Statement regarding DDT. Mimeo, Eastern Africa RBM Network. 
MoH. 2004. Health Sector Budgetary Allocation FY 2004-05. April 2004 version. Mimeo. Ministry of Health,
Elsey H, Kilonzo N, Tolhurst R, Molyneux C. 2005. Bypassing districts? Implications of sector-wide approaches Government of Uganda. 
and decentralisation for integrating gender equity in Uganda and Kenya. Health Policy and Planning, 20
(3):150–7. Moore S. 2003. Aid coordination in the health sector: examining country participation in sector wide
approaches. Journal of Health and Population in Developing Countries, 10, July. Available at:
Foster M, Brown A, Conway T. 2000. Sector-wide approaches for health development: a review of experience. WHO [http://www.jhpdc.unc.edu].
Publications. World Health Organisation, Geneva. Available at: 
[http://www.who.int/library/database/index.en.lshtm] Oliveira Cruz V, Kurowski C, Mills A. 2003. Delivery of priority health services: searching for synergies within
the vertical versus horizontal debate. Journal of International Development, 15: 67–86.
Green A, Collins C. 2003. Health systems in developing countries: public sector managers and the
management of contradictions and change. International Journal of Health Planning and Management, 18 (Suppl. Oliveira Cruz V, McPake B. 2004. Using agency theory to analyse relationships between recipient
1): S67–78. governments and international development partners. Working Paper, Health Systems Development Programme.
London School of Hygiene & Tropical Medicine. Available at: [http://www.hsd.lshtm.ac.uk].
HDPG. 2001. Terms of Reference for the Health Development Partners Group. Mimeo. Health Development
Partners Group, Kampala, Uganda. Peters D, Chao S. 1998. The sector-wide approach in health: What is it? Where is it leading? International
Journal of Health Planning and Management, 13 (2): 177–90.
HDPG. 2002. SWAp Mechanisms and Structures. Mimeo. Health Development Partners Group, Kampala,
Uganda. Root G, Collins A, Munguti K, Sargent K. 2003. Roll Back Malaria Scoping Study.
Available at http://www.liv.ac.uk/lstm/malaria/mcproject16.htm
Hutton G, Tanner M. 2004. The sector-wide approach: a blessing for public health? Bulletin of the World Health
Organisation, 82 (12): 893. Ssengooba F, Oliveira Cruz V, Pariyo G.  2004. Capacity of Ministries of Health and Opportunities to Scale Up
Health Interventions in Low Income Countries: A Case Study of Uganda. UN Millennium Development Project.
Jeppsson A. 2002. SWAp dynamics in a decentralised context: experiences from Uganda. Social Science and
Medicine, 55 (11): 2053–60. Walt G, Pavignani E, Gilson L, Buse K. 1999. Managing external resources in the health sector: are there
lessons for SWAps. Health Policy and Planning, 14 (3): 273–84. 
Kassami C. 2004. Welcoming remarks at the Public Expenditure Review Meeting. Mimeo. Ministry of Finance,
Planning and Economic Development, Government of Uganda.
Lake S. 2004. Macroeconomics and sector background paper. GFATM tracking study. Mimeo. London School of
Hygiene & Tropical Medicine. 
Macrae J, Zwi A, Gilson L. 1996. A triple burden for health sector reform: ‘post’-conflict rehabilitation in
Uganda. Social Science and Medicine, 42 (7): 1095–108.
MoFPED. 2004a. Finalisation of estimates for recurrent and development revenues and expenditure for financial year
2004/05 and the medium term. BPD 86/107/02. Mimeo. Ministry of Finance, Planning and Economic
Development, Government of Uganda.
MoFPED. 2004b. Uganda Budget 2003/2004. A citizens Guide. Ministry of Finance, Planning and Economic
Development, Government of Uganda. Available at: [http://www.finance.go.ug].
MoFPED. 2004c. Macroeconomic issues for 2003 PEAP revision (2nd draft). Mimeo. Prepared by Ministry of
Finance, Planning and Economic Development, Government of Uganda.
MoFPED. 2004d. Draft Terms of Reference for the Health Sector Working Group. Mimeo. Ministry of Finance,
Planning and Economic Development, Government of Uganda.
MoH. 2000a. Health Sector Strategic Plan 2000/01 to 2004/05. Ministry of Health, Government of Uganda.
42 Health Systems Reforms in Uganda: processes and outputs 2 Sector-wide approach in Uganda? 43
Annex 1: Overview of SWAp-related mechanisms and processes in Uganda
Oversees the financing of the SWAp with a particular focus Monthly in the run
on maximising efficiency and equity in the annual budget up to the budget
The following tables intend to provide an overview of the SWAp-related mechanisms and Sector process. The SWG discusses and reviews the Budget and bi-monthly
processes currently in place in Uganda. Table A1 summarises the core mechanisms and processes Working Framework Paper8 as well as proposals of new health sector otherwise (but in
with a focus at the national level. Table A2 presents the connections to the implementation level. Group (SWG) projects before submission to the Development Committee practice does not
of the MoFPED. It has an important role in vetting projects happen as
Table A1: SWAp-related structures and processes in Uganda for compatibility with the HSSP and value for money. regularly as this). 
Initially created There are currently 9 working groups During JRMs and
Definition Purpose / Processes                     Frequency to prepare for which report to HPAC: throughout the
the first HSSP • Human resources for health; year as per
•Joint visits to selected districts chosen on a and are now • Drug procurement and management; programme of
rotational basis, according to performance considered to • Health infrastructure; work (e.g. on a
(low and high) based on standard terms of Working play a key role • Supervision and monitoring; more regular
reference with a view to assess progress on Groups (WGs) in translating • Basic health care package; basis during
A joint review areas such as human resources, financial HSSP outputs • Public-private partnership in health; preparations for
of sector into policies, • Research and development;flows, information and management the secondplans and • Finance and procurement; HSSP).
performance systems, and agreed technical priority areas; activities. • Health systems
by GoU and •Review of the Annual Health Sector 
Joint Review partners (i.e. Performance Report; Annual Bring together The purpose of these committees is to:
Mission (JRM) districts, •Use of the agreed PEAP indicators (health)7 (October) all implementing • Define core interventions, review 
Parliament, as the basis for progress assessment; used to be agencies and overall progress in implementation and 
NGOs, •Discussion of proposals for the Budget twice a year. donors who agree on priorities for programmes;
private sector Framework Paper / Medium Term support a • Coordinate projects and other forms of 
and donors). Expenditure Framework priorities for the particular support to a specific programme;
following financial year; programme, • Review workplans and budgets of the and other MoH programme.
•Discussion and agreement on undertakings 
Interagency departments, Examples of existing ICCs include: Quarterly
(priorities), one or two priority programmes, Coordinating NGOs and • Reproductive health;
and a tracking study for following year. Committees districts. •Expanded Programme of Immunisation (EPI);
Substitutes the previous arrangement of two JRMs per year. (ICC) • Malaria; 
Technical This meeting among stakeholders aims to review and discuss a Annual • HIV/AIDS;
Review specific technical issue agreed during the prior JRM. The 2004 • TB;(around April) • Sanitation is in the process of 
technical review meeting discussed the first draft of the organising an ICC.
second Health Sector Strategic Plan (2005/06-2009/10).
Involves broad participation from district and central level, and Government Review the general status of the SWAp
National from civil society stakeholders. The purpose of the assembly is SWAp Review and partnership and discuss specific Annual
Health Meetings problems.to act as a forum for building nationwide consensus and developmentpartners.
Assembly advocacy for the health development agenda in the country. It Annual  
(NHA) is also an opportunity to improve sector performance by Established to • Provides a forum for discussion on issues
highlighting differences in district performance. coordinate in the sector;
development • Enables partners to coordinate and 
Health Policy Established as a forum to discuss and advise the Ministry of Monthly; partners assemble joint responses;
Advisory Health and development partners on the implementation of started as Health working in the • Serves as an opportunity for members 
Committee the National Health Policy and the HSSP. weekly and Development health sector in to communicate amongst themselves Monthly
(HPAC) goal is to have Partners Uganda. and with the MoH more effectively;
it quarterly. Group (HDPG) • Functions as a space to discuss issues 
related to HPAC;
• Allows DPs to contribute more effectively
to the JRMs in the health sector.
8Budget Framework Papers are prepared by each sector ministry as part of the budget process in consultation
7 These are: utilisation of out-patient services in public and private-not-for-profit units, immunisation rates for with stakeholders (to be discussed in SWG meetings) and form the basis for the Macroeconomic Plan and
DPT3, deliveries in health units, HIV prevalence rates, proportion of posts filled by qualified staff. Indicative Budget Framework Paper, usually submitted to Parliament in April of each year (Kassami 2004).
44 Health Systems Reforms in Uganda: processes and outputs
Partnership A special bank account held by the Ministry of Health for implementation of
Fund SWAp and HSSP specific activities (e.g. the costs of the JRMs, tracking studies
Account and technical assistance). Monitoring of the account is performed by HPAC. 
Contributions of funds to the account are made by DPs, which included Ireland 3
Aid, SIDA, NORAD, DFID, Danida, and UNICEF over the period of December
1999 to July 2003. Health sector reforms and increasing access to
Sources: HDPG (2002); MoH (2003a); MoH (2003b); HDPG (2001); de Loor and Hutton (2003); Kassami health services by the poor: what role has the
(2004); MoFPED (2004d).
abolition of user fees played in Uganda?
Table A2: SWAp and Districts
Christine Kirunga Tashobya, Barbara McPake, Juliet Nabyonga and Rob Yates
Description
NHA         Representatives (political and technical) from all districts are invited to take part.
JRM and One ‘good’ and one ‘bad’ performing district from each geographical region
Technical (North, East, West and Central) are selected by the Health Planning Department
Reviews (HPD), on a rotational basis, to host the district visits and to join the JRM. In
addition, based on the key topics of the meeting, the HPD invites relevant
district representatives. 
WGs Districts participate in WG meetings generally when these take place during the
JRMs but not during their regular meetings throughout the year (unless a
district representative chairs the WG).
HPAC, SWG Districts do not participate.
District Annual meetings organised by the MoH. These meetings are aimed at sharing Summary
Directors of key information and discussing implications (e.g. budget allocations for next
Health financial year). User fees were introduced in Uganda in the late 1980s against a background of
Services
Meetings poorly funded health systems and strong international support for the role of user fees in
encouraging community participation and ownership, and for their value in generating
Planning Semi-annual meetings organised by the HPD. Regional meetings take place revenue. By the late 1990s, there were conflicting opinions about the effect of user fees
Meetings between November and February. These meetings focus on budgetary issues. on access to health services, particularly by the poor and other vulnerable groups, in
Meetings at district level, between March and June, emphasise detailed
planning issues.  Uganda and other developing countries. In March 2001, user fees were abolished in all
public health units in Uganda except for private wings in hospitals. Abolition of user
Other In line with the MoH function of providing support to districts, there are various fees is only one of a number of reforms introduced in the health sector in Uganda since
meetings / other forms of interaction and meetings between the two levels such as: the turn of the century. To assess the impact that this policy change has had on the
visits / monitoring of the primary health care grants, area team supervision visits, HMIS, health sector, this chapter draws on evidence from a number of different sources
interactions league table. 
including data from the Ministry of Health, the World Health Organisation,
District and The HPD meets with superintendents of all district and regional referral hospitals Participatory Poverty Assessment Reports and the Uganda National Household
regional to discuss planning and other general matters. Surveys. The data point to a significant and immediate increase in utilisation of health
hospitals services following the abolition of user fees, in particular by the poor. We conclude that
user fees may be a bigger barrier to health service access for the poor than was previously
envisaged in developing countries. Furthermore, in order to achieve sustained
improvements in health service utilisation, the policy of abolition of fees should be
implemented simultaneously with supply side reforms. 
46 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 47
1. Background  poor. Since 2001, several studies and reports analysing the effect this policy has had on
health services utilisation in the country have been documented. 
Following years of political and economic upheaval in the 1970s and 1980s, the poor This chapter seeks to review a number of these documents, looking at the different
state of the public systems, including health services, led the country to be heavily sources of data in order to draw conclusions on whether the abolition of user fees has
dependent on foreign aid. At that time, development policy, and as a result, health indeed had an effect on utilisation of health services in the country. Particular focus is
policy, was heavily influenced by these donor countries and aid agencies (Okuonzi and given to the impact of the policy on utilisation by vulnerable groups. Furthermore, we
Macrae 1995). attempt to assess the contribution that this specific policy change has had in the context
The rationale for the introduction of user fees in the health system, as argued by the of the overall health sector reform programme in Uganda. In chapter 1, Yates et al.
proponents of this policy, was that they had the potential to increase revenue for health report that there have been significant increases in key output indicators, particularly for
services and also to increase efficiency in the delivery of health services by giving ambulatory care, but it is not yet clear how much of this has been due to supply side
patients a stake in the system. This would lead to the use of the additional revenue for reforms and what, if anything, has been the impact of the policy to abolish user fees.  
particular needs like drugs, facility improvement and staff incentives. In addition, user
fees were said to have the potential to increase equity in health services by making it
possible to expand coverage of services for underprivileged groups (Akin et al. 1987; 2. Review of available data 
Bennett and Gilson 2001; McPake et al. 1993; Shaw and Griffen 1995). 
Close to two decades later, it should be possible to determine whether or not user Multiple sources of information are utilised for this review and include central level
fees have lived up to expectations. Internationally, there are mixed reports about what Ministry of Health (MoH) and district level data and reports; a WHO/MoH study; an
user fees have achieved. Some studies show that user fees have raised some revenue for analysis of the Uganda National Household Surveys for the years 1999/00 and 2002/03;
health services, forming a significant proportion of non-salary recurrent expenditure at and the Uganda Participatory Poverty Assessment Reports. A number of other studies
health units (Creese 1997). These resources have been used to improve service quality covering smaller geographical areas have been carried out in the country and their
by improving drug availability and boosting staff morale (Soucat et al. 1997; Audibert findings are also considered. 
and Mathonnat 2000; Kipp et al. 1999). There have also been some reports of increased
community participation in health services management, built around user fees schemes 2.1 Ministry of Health and District data 
(MoH 2001a). 
Conversely, user fees have been criticised on grounds of inequity, lack of visible The Health Management Information System (HMIS) gives data on various input,
quality improvements, and inability to raise substantial amounts of revenue (Van Der process and output variables. Data are captured at the health facility level and submitted
Geest et al. 2000; McPake 1993). Rather than improving equity in health services on a monthly basis to the district health office, where they are aggregated and sent to
delivery, user fees are said to have led to less equitable access to services. This is because the MoH.a The HMIS captures data from both the public and private-not-for-profit
exemption schemes appear to have failed to protect the poor and the vulnerable (Gilson (PNFP) health units. 
et al. 1995; Weaver 1995). The revenue generated has been limited and has not led to Reports on the data (MoH 2001b; MoH 2002a; MoH 2003a; MoH 2003b; MoH
any notable expansion of health services for improved geographical equity (Meuwissen 2004) have shown that for the whole country, the number of new out-patient (OPD)
2002). contacts in both public and PNFP health units increased significantly in absolute and
In Uganda, user fees were first introduced in the health sector in the late 1980s but per capita terms over the period 1999/00 to 2003/04.b New OPD contacts per capita did
did not spread widely until the early 1990s (Kipp et al. 1999). Initially, the fees were not change from 0.42 in 1999/00 to 2000/01, but rose to 0.56 in 2001/02, and further to
popular with health workers and with most local government administrative and 0.72 in 2002/03, and to 0.79 in 2003/04. The substantial increase in utilisation following
political leaders. However, they were unpopular with many members of the community the abolition of user fees in 2001 is notable. Note that the utilisation rate increased by
and many national level politicians (Kipp et al. 1999; MoH 1999; Hutchinson et al. 33 percent between 2000/01 and 2001/02 and by a further 29 percent in 2002/03 and 10
1999). By the late 1990s, a number of reports and studies indicated that exemptions for percent in 2003/04, compared with no increase between 1999/00 and 2000/01 (see
the poor were not working and that many people were being denied access to basic Figure 1 in chapter 1, Yates et al.). This is particularly noteworthy as these rises cannot
health services because of the fees (Kivumbi and Kintu 2000; MoFPED 2000; UBOS be attributed to other sudden demand factors such as an epidemic. Since the malaria
2000). During the run-up to the Presidential elections of June 2001, user fees for health epidemic in 1997/98, the country has not experienced an epidemic of national
services became a controversial issue, with abolition of fees included in the manifestos magnitude.c
of the major competitors to the incumbent President Yoweri Museveni. Interestingly,
the President and Parliament had never officially approved a policy on user fees in the
country.
With effect from March 2001, the Government of Uganda (GoU) abolished user
fees in all public units with the exception of the private wings of hospitals. This has aSee Chapter 1, Annex 1 for an outline of the structure of the health system.
presented health policy analysts with a rare opportunity to study the effect of the bThe Ugandan financial year runs from July to June. Thus in the financial year 2000/01 no user fees applied in
removal of user fees on utilisation of health services, and in particular, utilisation by the government units (apart from private wings in hospitals) for the period  March - June 2001.
cThe 2000 Ebola epidemic was localised in Northern Uganda. Similarly, some districts have experienced localised
cholera and malaria epidemics which would not markedly affect national OPD figures. 
48 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 49
Looking at data from specific districts, including hospitals (see Figures 1–5), a
marked increase in utilisation following the abolition of fees in March 2001 is immediate
and consistent across all four selected districts (Kisoro, Ntungamo, Rukungiri and Figure 1: Monthly new out-patient attendances for all the health units
Tororo; MoH selected years). For example, in Kisoro District, the consumption of out- (government and PNFP) in Kisoro District 1998–2004 
patient services immediately doubled. 
These district level data also bring out the monthly variation in utilisation, which
is quite large in some cases. 
In addition, the detailed district level data highlight the different utilisation
patterns between the public and PNFP units (who maintained user fees). This is shown
for Kisoro District where two hospitals, Kisoro Hospital (public) and Mutolere Hospital
(PNFP), are barely 4km apart. While utilisation rose at both hospitals immediately after
the abolition of user fees in the public health units, the increase in the public hospital
was substantially larger than that in the PNFP. Moreover, using the benefits of growing
financial support from the government (a specific PNFP grant), the PNFP hospital was
able to lower its user fees by a considerable amount in 2002. This was followed by a
significant increase in utilisation at the PNFP unit, and some decline at the public unit.
A similar pattern is noted for Nyakibale PNFP hospital in Rukungiri District, where
significant increases in utilisation were noted when the hospital revised its fees
downwards and made them more uniform across diagnoses and disease conditions. In
chapter 5, Lochoro et al. expand on the contribution of the PNFPs to the health sector
and the impact of reforms on this sub-sector. Source: MoH (Selected years).
In addition to out-patient rates, immunisation rates also shot up following the
abolition of fees, as measured by the proportion of the infant population that has
received the third dose of the Diphtheria-Pertussis-Tetanus (DPT3) vaccine.   This
improved from 41 percent in 1999/00 to 48 percent in 2000/01 (17 percent increase), to
63 percent in 2001/02 (31 percent increase) and to 84 percent in 2002/03 (33 percent Figure 2: Monthly new out-patient attendances in 2 hospitals 
increase), and stabilising at 83 percent in 2003/04. Again the dramatic increase in (one government, one PNFP) in Kisoro District 1999–2004
coverage after 2000/01 is noted. 
However, contrary to the above findings, the proportion of expectant mothers
delivering in these health units declined from 25 percent in 1999/00 to 23 percent in
2000/01, to 19 percent in 2001/02, with just a slight recovery to 20 percent in 2002/03
and 24 percent in 2003/04. 
2.2 WHO/MoH study on the effect of abolition of user fees
A longitudinal study carried out under the auspices of the WHO Uganda Country
Office provides both qualitative and quantitative data on the effect of the abolition of
user fees (WHO 2002; WHO 2003; Nabyonga et al. 2005a; Nabyonga et al. 2005b). Of
particular interest, the study links HMIS utilisation data for public and PNFP facilities
with demographic (i.e. age and sex) and socio-economic variables of the health users
taken from information in the community. Patients’ households were classified as
poorest, poor, well-off and richest on the basis of household assets. 
In addition to analysing OPD utilisation rates, socio-economic data were estimated
for the years 2001 and 2002. Over this period, there was a strong negative socio-
economic gradient in OPD utilisation – successively poorer groups had successively Source: MoH (Selected years).
higher utilisation rates (Figure 6). The utilisation of the poorest group was
approximately twice that of the richest group for most of the 3-year period.
50 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 51
2.3 Uganda National Household Surveys 1999/00, 2002/03 
Data on individual, household, and community characteristics are collected from a Figure 3: Monthly new out-patient attendances (in government and 
nationally representative sample in the Uganda National Household Surveys (UNHS) PNFP health centres) in Ntungamo District 1998–2004
(UBOS 2000; UBOS 2003). Between August 1999 and September 2000, when user fees
were still widespread, a sample of 10,696 households was surveyed. Between May 2002
and April 2003, after the abolition of fees, a total of 9,711 households were surveyed.
Deininger and Mpuga (2004) used these data to analyse changes in health service
utilisation and to investigate the implications of these changes for the poor. Their
analysis has shown that while the incidence of sickness in the population in the two
surveys remained at about 28 percent, an increasing proportion of those falling sick
sought health care in the period covered by the second survey relative to the first.
Interestingly, the share of households reporting failure to seek care from health services
due to cost declined from 50 percent in the first survey to 35 percent in the second
survey. This was especially noted in Northern Uganda, the poorest region of the country. 
Moreover, this increase in utilisation of services by the sick was highest for those in
the poorest two income quintiles. While average monthly expenditure by a household
with a sick person did not show any significant change between the two surveys, there
was a significant decrease in spending by the bottom two quintiles (at the 1 percent
significance level). Rationing, defined as the failure to use health services because the
cost is too high or the health facility too far, and not because the sickness is mild, was
also found to decline. Regressions showed that the degree of rationing fell between the Source: MoH (Selected years).
two surveys for both children and adults, with the decline most marked for children. In
addition, wealth and urban/rural biases in rationing were eliminated for children and
significantly reduced for adults. 
Community level data collected in the two surveys revealed that between 1999/00 Figure 4: Monthly new out-patient attendances at Nyakibale 
and 2002/03 there was a significant reduction in the proportions of communities that PNFP hospital, Rukungiri District 1996–2004 
reported paying for consultation fees (from 80 percent to only 3 percent), antimalarial
drugs (22 percent to 4 percent), and antibiotics (27 percent to 3.9 percent) in
government health facilities. 
Deininger and Mpuga (2004) also investigated whether these improvements in
utilisation of health services could be having an impact on health outcomes and
productivity. As noted above, the incidence of illness in the population remained at 28
percent over the period covered by the two surveys. In 1999/00, however, the burden of
ill-health was more heavily felt by the poorer groups in the population. By 2002/03, the
data suggest that this previous propensity for the poor to fall sick more frequently than
the rich appears to have been eliminated. 
Furthermore, regarding productivity, the two surveys show that the average per
capita number of workdays lost due to sickness declined from 8.3 days to 7 between
1999/00 and 2002/03. To calculate the economic cost of illness, Deininger and Mpuga
used a measure which incorporated the average expected length of sickness, the average
unskilled wage rate for adults (divided by four in the case of children) and the
probability of falling sick. Using this measure, the economic benefit accruing from the
improved utilisation and quality of health services observed in the second survey period
was calculated at US$9 million for adults, almost half of which accrued to the bottom Source: MoH (Selected years).
two income quintiles in the sample. 
52 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 53
2.4 Uganda Participatory Poverty Assessment reports  
Uganda has carried out two Participatory Poverty Appraisals (PPA) to date under Figure 5: Monthly new out-patient attendances in Tororo District 
the auspices of the Uganda Bureau of Statistics (UBOS) in the Ministry of Finance, (in government and PNFP health centres) 2000–2002 
Planning and Economic Development (MoFPED) as part of the process of developing
and updating the country’s poverty reduction strategy, the Poverty Eradication Action
Plan (PEAP). During these appraisals, participatory techniques are used to consult
people on their understanding of the nature and causes of poverty, and on their views of
the priority actions for poverty reduction by the government and/or the communities. 
The first PPA (PPA1) was carried out in 1997, before the abolition of user fees
(MoFPED 2000). The study covered 36 research sites in 9 districts across the country. In
all 9 study districts, there was a common view that health was an important factor in
poverty, both as cause and effect. Poor health and inadequate health services were cited
as major causes of poverty at individual, household and community levels. The factors
listed as responsible for these poor and inadequate health services included user fees,
long distances to health facilities, inadequate drug and medical supplies,
untrained/inadequate/poorly motivated health personnel and poor infrastructure. As a
result of being unable to afford the user fees, people were reported to have resorted to no
treatment, self-treatment, traditional herbalists, traditional birth attendants and
witchcraft. A respondent stated "in the 1990s you have to pay for medical care and drugs.
Many die in the villages because they cannot afford to pay the user charges" (MoFPED 2000:
62). Apart from the issue of un-affordability, it was made clear that there were other Source: MoH (Selected years).
difficulties with the user fees scheme, whereby many people did not seem to have a good
understanding of the nature of charges and the services to go with them (MoFPED
2000). 
The second PPA (PPA2) was carried out between November 2001 and May 2002 3. Discussion 
(after abolition of user fees), and covered 60 research sites in 12 districts.  Among other
issues, PPA2 consulted people on their experience of the implementation of key 3.1 Reliability of the data 
government policies in a number of sectors, including health. As in PPA1, the most
frequently cited causes of poverty were ill health and disease (MoFPED 2002). Notably, A major strength of the HMIS is that it provides national data and is therefore not
in PPA2 respondents stated that the government had been right to abolish user fees, a limited to a small geographical area or sample. Some of the possible weaknesses of the
move that was particularly appreciated by women and the poor. Across the research HMIS as a source of data derive from the fact that the system was introduced relatively
sites, both community members and health service providers reported a dramatic recently (1994) and still has teething problems with poor timeliness and completeness,
increase in utilisation of public health facilities, but also noted problems of insufficient although both have been improving over time. Therefore, some of the improvements in
drugs and inadequate numbers of qualified health workers. One sub-county chief is utilisation of ambulatory services may be due simply to improvements in the system.
reported to have said: "I can confidently say that the number of patients in these units has However, it is unlikely that this alone could explain the extent of increases observed
tripled ever since cost-sharing was scrapped. The current policy of free health services should not between 2000/01 and 2003/04. 
be tampered with since it is mainly the poor who cannot afford expensive drugs in health clinics. The study carried out under the auspices of the WHO Uganda Country Office has
Government should however provide more of these drugs" (MoFPED 2002: 111). Thus it can collaborated closely with many health sector stakeholders in Uganda. The linking of
be seen that the qualitative evidence collected through these participatory approaches HMIS and community survey data is likely to minimise the possibility of inflated health
supports the quantitative data presented earlier, and substantiates the negative facility HMIS figures. Furthermore the UNHS provides a nationally representative
implications of user fees for the poor in Uganda. household survey database. This generic database is managed by the Uganda Bureau of
A number of other studies on user fees covering smaller samples have also been Statistics and is outside of the control of any one sector. The availability of data on
carried out in the country (Burnham et al. 2004). All of these point to the increase in household characteristics and on the utilisation of health services before and after the
health service utilisation following the abolition of user fees, though they provide abolition of user fees makes this is a very important source of information. Another
differing estimates of the magnitude of the increase. perspective is provided by the PPA reports which provide the ‘voices of the poor’
themselves, rather than of their leaders or technocratic representatives. 
54 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 55
and reports outlined here report much higher increases than this. Our data suggest an 88
percent overall increase in the rate of new OPD contacts per capita per year. This
Figure 6: Outpatient utilisation rate at public lower level health units underestimation suggests a need to update the models applied by such analysts on the
by socio-economic category 2001–2003 basis of the Ugandan experience, and that unrealistic estimates of the impact of fees may
have been made in other countries where empirical tests of these estimates are not
feasible.
The WHO/MoH study and the analyses of the UNHSs and the PPAs show that the
more vulnerable sections of the community – the poor, children and women have
particularly benefited from this policy (MoH 2003b; WHO 2003). The data from the
Uganda Household Surveys provide an early indication of the benefits of the abolition
policy in terms of outcomes, with the observation that the poor now have to take fewer
days off work due to illness, thereby reducing the negative impact of illness on income.
This negative impact of illness on income is well known to communities. In the words
of one respondent: "as long as one is healthy, he thinks properly for his family, is able to travel
to towns to do business for the well-being of the family, can cultivate land, and construct shelter
and work harder" (MoFPED 2000: 62). Clearly, as a policy, the abolition of user fees is
consistent with the goals of Poverty Reduction Strategies and the internationally
endorsed Millennium Development Goals (MDGs). 
3.3 What else is involved?
Source: WHO (2002); WHO (2003); Nabyonga et al. (2005b).
One aspect of the data that requires further explanation concerns the high variation
in the health service utilisation rates recorded in public health units. The annual trend
of increasing utilisation of ambulatory services (especially OPD attendance) hides
marked variations, which cannot be explained entirely by seasonal variation in disease
3.2 How do we interpret the evidence? factors. Immediately following the abolition of user fees as of March 1st 2001, the
utilisation of public health services rose sharply, but this was soon followed by a sharp
These different studies/reports consistently show a marked increase in utilisation of drop by May and June 2001. Subsequently utilisation rates varied from month to month
ambulatory health services over the period 2000/01 to 2003/04. A number of factors and between districts. These variations appear to be more closely associated with supply
point to the abolition of user fees as a key determinant of this increase. These include factors, and in particular, with the availability of basic medicines at the health units
the close relationship in time between the abolition of fees and the observed increase in (MoH 2003b; WHO 2003). Given the inadequacy of medicine budgets, some
utilisation of services, coupled with the absence of any epidemiological reason for such irregularity in release of funds, and limited capacity for medicine management in the
an increase at this time. Another factor is the lack of a similar marked increase in PNFP districts, there are often medicine stock-outs at the health facilities. The length and
outputs in the immediate aftermath of public-fee removal. Also supporting the link is magnitude of the stock-outs, and therefore the peaks and troughs in utilisation will vary
the qualitative evidence that user fees were a hindrance to the population, particularly from district to district. For example, the beginning of the financial year is likely to be a
to the poor, in accessing public health services. challenging period for most districts as funds from the central government may be
The correlation between fee abolition and utilisation is further borne out by the delayed, and this same period usually shows poor utilisation of OPD services. In the
evidence from the PNFP sector. In more recent years, with increasing subsidies from the PNFP health units, where there is less variation in drug availability throughout the year,
government budget, many of the PNFP units have decreased their user fees, which has much less monthly variation in utilisation is observed. In more recent years-2002/03 and
also led to increased utilisation rates at these units, particularly for ambulatory services 2003/04-improvements in medicines funding, management, and financial management
(as highlighted in Figures 2 and 4 and discussed in more detail by Lochoro et al. in appear to be reflected in decreasing utilisation variation in the public units.
chapter 5). As in the public units, the timing of these shifts in output immediately Also in need of explanation is the observation that the marked increase in
follows the reduction in fees. utilisation of ambulatory services has not been matched by an equally marked increase
The increases in utilisation documented are significantly greater than were expected in the consumption of in-patient and especially maternity services. The quality of
by many analysts. For example, a World Bank Discussion Paper had predicted that services may have played an important role in this. Specific problems in quality of care
removal of user fees could only improve utilisation by the poor by 2.3 percent, and for in-patient episodes continue to be observed. A number of studies and reports,
decrease utilisation by the rich by 9 percent (Hutchinson et al. 1999). All the studies including the PPA reports, a study on Emergency Obstetric Care in Uganda, and
56 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 57
government monitoring reports, indicate that there continue to be problems associated There is, however, need for considerably more research in Uganda in this area, in
with lack of access and privacy, and insufficient supplies (e.g. cotton, gauze, protectives) particular to separate out the impacts of the user fee reforms from those of the other
which are discouraging mothers from delivering at the health units (MoFPED 2000; health service reforms. Also critical is the need for a better understanding of the costs of
WHO 2003; MoH 2003c; MoH 2003d). Where delivery at a health unit may require the achieving improvements in quality, especially with regard to in-patient services,
mother to provide gloves, a plastic sheet and other supplies, it is not difficult to see how including maternity services. In all of this work it is important that the focus continues
these ‘hidden costs’ act as a barrier to the health services in the same way as the user fees to be on the utilisation of the vulnerable groups in the population. 
used to. This is further confirmed by the fact that the modest increase in in-patient The impact of the abolition of user fees on the behaviour of health workers is also
service utilisation in public units over the period since 2000/01 is comparable to that in important to investigate. Health workers effectively suffered a loss in income following
PNFP units (MoH 2003b). the removal of user fees, and this was coupled with a dramatic increase in workload.
This brings us to the question of how much of the increase in utilisation of Though efforts were made subsequently to increase salaries, human resources are a
ambulatory services can be attributed to abolition of user fees and how much to other critical input in health services and it will be important to ensure support for the
reforms. The sudden sharp increase in the utilisation of public services that occurred reformed system is maintained.  
following the abolition of user fees can clearly be related to increased demand due to the Finally, the effect of the abolition of user fees on community participation and
fall in cost for users. However, it becomes more difficult to estimate the contribution of ownership of local health services requires attention. Through the Health Unit
the abolition of user fees to the sustained and continuing increase in utilisation of Management Committees, established to manage the allocation of user fee revenue,
services observed over time since then. Consideration must be given to other reforms community involvement was facilitated. The loss of this key responsibility may have
which have been taking place over the same period. These include the introduction of endangered the functioning of these committees and with it, local ownership.
the sector-wide approach (SWAp), the increased allocation of the government budget
to fund health services, enhanced partnership with the private sector (profit and non-
profit), and improvements in the management and delivery of decentralised health 4. Conclusions  
services (e.g. infrastructure, human resources and medicines funding and management),
as discussed in the other chapters in this book. Overall, in chapter 1, Yates et al., have discussed the effects that the many health
For example, a supplementary budget of one billion Uganda shillings (equivalent to services reforms in Uganda have had at the health output level and which are beginning
US$600,000) was provided for additional medicines immediately following the abolition to show at the outcome/poverty reduction levels. This chapter has looked at one reform
of user fees. In subsequent budgets, attempts were made to compensate the health sector in particular, that of abolishing user fees. While the other chapters in this book
for the revenue lost due to the abolition of fees, and to target most of these funds at concentrate largely on supply side reforms, the policy of abolishing user fees focuses on
primary health care inputs (especially medicines, strategic infrastructure expansion and the reduction of access barriers and the stimulation of demand. The evidence is
upgrading, and human resources). However, supply side reforms tend to take some time convincing that user fees’ removal acted as a catalyst for the initial stimulation of OPD
before they are converted into outputs. The budget for primary health care medicines at demand. Ongoing support from supply side reforms is likely to have been important in
sub-district level increased almost 3-fold from 1999/00 to 2003/04 and yet, partly as a maintaining the impetus. Further attention needs to be paid to the stimulation of in-
result of increasing utilisation of the health services, there continue to be stock-outs patient and maternity demand, and the maintenance of mechanisms of local ownership
reported at the local health units (see chapter 4, Nazerali et al.). that may have been undermined by the reform. 
Thus it seems reasonable to conclude that while the policy decision to abolish user
fees catalysed the surge in consumption services, levels have been sustained since then
by the ongoing supply side reforms.
3.4 What else needs to be done?
The Ugandan case provides strong support for the argument that it is possible to
increase the utilisation of health services in developing countries by improving financial
access, and that the provision of free services at the point of use is a good way of
targeting the poor.
Moreover, in order to achieve sustained improvements in access to health services
by the poor, this chapter, together with the others in this book, underscores the point
that several changes may need to take place simultaneously and that both demand and
supply side issues need to be tackled. 
58 Health Systems Reforms in Uganda: processes and outputs 3 Abolition of user fees and increasing health services utilisation 59
References  MoH. 2003c. Status of Emergency Obstetric Care (EmOC) in Uganda: A National Needs Assessment of EmOC
Process Indicators. Final Draft Report June 2003. Ministry of Health, Government of Uganda.
Akin J, Birdsall N, Ferranti D. 1987. Financing health services in developing countries: an agenda for reform. The MoH. 2003d. District Monitoring Reports. Ministry of Health, Government of Uganda.
World Bank, Washington DC.
MoH. 2004. Annual Health Sector Performance Report 2003/2004. Ministry of Health, Government of Uganda.
Audibert M, Mathonnat J. 2000. Cost recovery in Mauritania: initial lessons. Health Policy and Planning, 15
(1): 66–75. MoH. Selected years. District HMIS databases of Kisoro, Ntungamo, Rukungiri and Tororo. Health Management
Information System Databases, Ministry of Health, Government of Uganda.
Bennett S, Gilson L. 2001. Health financing: designing and implementing pro-poor policies. DFID Health Systems
Resource Centre. Department for International Development, UK. Nabyonga J, Desmet M,  Karamagi H, Kadama PY, Omaswa F, Walker, O. 2005a. Abolition of cost-sharing is
pro-poor – evidence from Uganda. Health Policy and Planning, 20 (2): 100–8.
Burnham GM, Pariyo G, Galiwango E, Wabwire-Mangen F. 2004. Discontinuation of cost sharing in Uganda.
Bulletin of the World Health Organisation, 82 (3): 187–95. Nabyonga J, Atuyambe L, Bagenda F, Bataringaya J, Karamagi H, Kasirye I, Omaswa F, Walker, O. 2005b. A
Three Year Report of the Longitudinal Study on Effects of Abolition of Cost-Sharing in Public Facilities. WHO
Creese A. 1997. User fees. British Medical Journal, 315 (7102): 202–3. Country Office, Uganda, SIDA and Ministry of Health, Government of Uganda.
Deininger K, Mpuga P. 2004. Economic and Welfare effects of the abolition of health user fees: evidence from Okuonzi SA, Macrae J. 1995. Whose policy anyway? International and national policies on health policy
Uganda. World Bank Policy Research Working Paper 3276. The World Bank, Washington DC. development in Uganda. Health Policy and Planning, 10 (2): 122–32.
Gilson L, Russell S, Buse K. 1995. The political economy of user fees with targeting: developing equitable Shaw RP, Griffin C. 1995. Financing Health Care in Sub-Saharan Africa through User Fees and Insurance. The
health financing Policy. Journal of International  Development, 7 (3): 369–402. World Bank, Washington DC.
Hutchinson P, Habte D, Mulusa M. 1999. Health Care in Uganda: Selected Issues. World Bank Discussion Soucat A, Gandaho T, Levy-Bruhl D, de Bethune X, Alihonou E et al. 1997. Health seeking behaviour and
Paper No. 404. The World Bank, Washington DC. household expenditures in Benin and Guinea: The Equity implications of the Bamako Initiative. International
Kipp W, Kamugisha J, Jacob P, Burnham G, Rubaale T. 1999. Cost sharing In Kabarole District, Western Journal of Health Planning and Management, 12 (Suppl. 1): S137–S163.
Uganda: communities’ and health professionals perceptions about health financing. Journal of Health and UBOS. 2000. Report of the Uganda National Household Survey for 1999/2000. Uganda Bureau of Statistics. 
Population in Developing Countries, 2 (2): 30–38.
UBOS. 2003. Uganda National Household Survey 2002/03. Report on the Socio-economic Survey.  Uganda
Kivumbi W, Kintu F. 2000. A review of exemption and waivers from health care charges in Uganda. Study report Bureau of Statistics.
submitted to UNDP/World Bank/WHO.
Van Der Geest S, Macwangi M, Kamwanga J, Mulikelela D, Mazimba A, Mwangelwa M. 2000. User fees and
McPake B. 1993. User charges for health services in developing countries: a review of the economic literature. drugs: what did the health reforms in Zambia achieve? Health Policy and Planning, 15 (1): 59–65.
Social Science and Medicine, 36 (11): 1397–405.
Weaver M. 1995. User fees and patient behaviour: Evidence from Naimey National Hospital. Health Policy and
McPake B, Hanson K, Mills A. 1993. Community financing of healthcare in Africa: an evaluation of the Planning, 10 (4): 350–61. 
Bamako initiative. Social Science and Medicine, 36 (11): 1383–95.
WHO. 2002. The effects of abolition of cast-sharing in Uganda – Report of the findings for 2001. World Health
Meuwissen LE. 2002. Problems of cost recovery implementation in district health care: a case study from Niger. Organisation, Uganda Country Office.
Health Policy and Planning, 17 (3): 304–13. 
WHO. 2003. The effects of abolition of cost-sharing in Uganda – Report of the findings for 2002. World Health
MoFPED. 2000. Uganda Participatory Poverty Assessment Study 1 Report – Learning from the poor. June 2000. Organisation, Uganda Country Office.
Ministry of Finance, Planning and Economic Development, Government of Uganda.
MoFPED. 2002. Uganda Participatory Poverty Assessment Study 2 Report – Deepening the Understanding of
Poverty. 2002. Ministry of Finance, Planning and Economic Development, Government of Uganda. 
MoH. 1999. Report on the Findings of the study commissioned to look at the current implementation of cost-sharing.
Ministry of Health, Government of Uganda.
MoH. 2001a. Rukungiri Bamako Initiative Scheme Report, January 2001. Ministry of Health, Government of
Uganda.
MoH. 2001b. Annual Health Sector Performance Report 2000/2001. Ministry of Health, Government of
Uganda.
MoH. 2002a. Annual Health Sector Performance Report 2001/2002. Ministry of Health, Government of Uganda.
MoH. 2002b. Budget Framework Paper 2002/03. Ministry of Health, Government of Uganda.
MoH. 2003a. Annual Health Sector Performance Report 2002/2003. Ministry of Health, Government of Uganda.
MoH. 2003b. Mid Term Review Report of the Health Sector Strategic Plan 2000/01–2004/05. Ministry of Health,
Government of Uganda.
4
Medicines – driving demand for health 
services in Uganda?
Hanif Nazerali, Martin Olowo Oteba, Joseph Mwoga and Sam Zaramba
Summary
Utilisation of out-patient services at government health units in Uganda showed a
significant increase following the abolition of user fees in 2001, and there is evidence
that the door was opened for those who were least able to pay. Utilisation has been
sustained at high levels since then. This chapter points to concurrent improvements in
the medicines supply system as a key factor contributing to the continued high demand.
Improved flow of medicines to the primary care level underpinned more effective
delivery of Uganda’s basic health care package including preventive services. The
exceptional results and fast pace of reforms in the medicines supply system are attributed
to good timing and an enabling environment. Chronic under-funding, fragmented
financing from multiple sources, and the related problem of uncoordinated, vertical
supply (‘push systems’) were addressed and partly resolved through the structures and
mechanisms established under the sector-wide approach. At the same time, through the
process of political and administrative decentralisation in Uganda, district and sub-
district health managers have assumed greater responsibility and control over the
medicines budget. This provided the necessary background for an accelerated transition
to an integrated ‘pull system’ of medicines supply that is more responsive to locally
determined demand and the changing requirements of a dynamic health system.
Challenges still remain however, and there is need to address problems of continuing
medicines stock-outs at some health units. A growing reliance on new, more expensive
medicines that are often funded through global initiatives with vertical programming
and parallel funding structures poses another challenge to the new systems.
62 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 63
1. Introductiona services, were used as a proxy for access to medicines. The data were collected by staff of
the District Drug Management Programme of the Ministry of Health (MoH) and/or
In Uganda, health users consider medicine availability to be a key determinant of local consultants through annual surveys and a Drug Tracking Study. Findings were
quality in health services (MoFPED 2002). It is notable therefore that there were published in the APR and in other reports widely disseminated in the health sector. 
widespread, chronic shortages of medicines at government health units throughout the The Ugandan experience was compared and contrasted with policies and reforms
era of user fees. In the context of high unmet demand, regular provision of essential internationally through a review of the published literature, including electronic
b
medicines, without fees in government health units, or at reduced fees in private-not- libraries and web-accessible resources.
for-profit (PNFP) facilities, would be expected to have a dramatic effect on health
service utilisation. Following the abolition of user fees in public health units (March
2001), utilisation increased immediately. Moreover, utilisation continued to increase Figure 1: Conceptual framework
over several consecutive years. Public opinion was favourable towards user fee abolition,
although negative about the continued poor availability of medicines at the units. This
chapter focuses on the series of reforms in the public financing and supply of essential
medicines (as part of the broader health systems reform programme) that were
instrumental in improving access to medicines, thereby contributing to the increased
and sustained demand for public out-patient services in Uganda.
2. Methods and sources
Figure 1 presents a conceptual framework showing pathways leading to increased
utilisation of out-patient services. Access to medicines is presented as the driving factor
for reforms on both ‘supply side’ and ‘demand side’. 
The situation analysis presented here is grounded in the technical reviews of the
first Health Sector Strategic Plan (HSSP) 2000/01 – 2004/05, routinely conducted as
part of the sector-wide approach (SWAp) during the period of health sector reforms in
Uganda. The technical reviews called for systematic review and synthesis of previous
consultancy reports, facility surveys and studies in the medicines sub-sector. The authors
of the reviews were directly involved in the health SWAp and this facilitated consensus
building on the systems issues to be addressed by the reforms. In particular,
interpretation of the interaction between demand for health services and access to
medicines was informed by findings from Participatory Poverty Assessments, Ugandan
National Household Surveys (UNHS), and national service delivery surveys that
measure satisfaction with government services in six sectors.
Health expenditure data were derived from the UNHS and National Health
Accounts (UBOS 2001; UBOS 2003; MoH 2004a). Analysis of the medicines budget
allocations in the public sector over a six year period pre- and post-reforms, were
conducted by the authors and updated annually for the Health Sector Annual
Performance Report (APR) and the annual budget planning process.  Survey data on
availability of medicines at the health facility level, and specifically the stock-out time
for a small number of indicator medicines that are crucial for provision of out-patient
aThe authors wish to acknowledge Khalid Mohamed (consultant attached to the District Medicines Management
Programme), who collected and reported much of the monitoring data used in the analysis, and are grateful for b
the continuing cooperation and collaboration of colleagues in the health sector, at the NMS, JMS, Uganda Documents available through the WHO Medicines Library, development information exchanges such as Eldis
Catholic and Protestant Medical Bureaux, in the districts, health sub-districts, and within the Medicines id21 and the Social Sciences Development Network; abstracts and posters from the International Conference on
Procurement and Management Working Group. Improved Use of Medicines (ICIUM) in 2004 and Strategies for Enhanced Access to Medicines (SEAM) in 2000.
64 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services in Uganda? 65
3. Medicines in health care – access, cost and management issues The following situation analysis highlights the importance of medicines access in
Uganda and the barriers related to cost and limited management capacity at the outset
3.1 Common factors among developing countries of the health sector reforms in 2000.
Access to affordable medicines is included amongst the health-related Millennium 3.2 Public perceptions and the importance of access to medicines
Development Goals. Medicines are a major health expense for poor households in most
developing countries where 50-90 percent of medicines are paid for by the patients A Participatory Poverty Assessment early in the period of reforms gave voice to
themselves, while in many developed countries, 70 percent of medicines are publicly communities’ concerns about limited access to medicines and the resulting deepening of
funded through reimbursement plans and other mechanisms (Quick et al. 2002). poverty (MoFPED 2002).  Indeed, communities in Uganda appear to be very well
Moreover, it is estimated that less than half of the population in the poorer parts of informed of the status of medicines supply at their local health facilities. Nabyonga et al.
Africa and Asia have regular access to essential medicines. This remains a major (2005) reported fluctuating monthly utilisation following the abolition of user fees at
obstacle to good health despite the many achievements in the field of essential government health units. We believe that this volatility could be explained by irregular
medicines since the Declaration of Alma Ata in 1978 (Quick 2003). Social and cultural medicines availability. A Drug Tracking Study provided evidence for this interpretation
constraints disproportionately prevent women, children, ethnic minorities, and other (MoH 2002a). Comparing out-patient caseload thirty days before and after a medicines
marginalised populations from gaining access to medicines (Ruxin et al. 2005). While
recognising that multi-sectoral action may be needed to address the socio-economic and delivery, the study observed an increase of more than 50 percent on average (n=65).
political factors underlying the access gap, Quick (2003) identifies four obstacles that Anecdotal evidence indicates that volatility in out-patient attendances was even greater
the health sector has yet to overcome: in the past when rural facilities relied on a quarterly supply of essential medicine kits.
Uganda operated a kit system from 1987 to 2002.c Throughout that period, attendance
• Unfair financing for health, including medicines; decreased when popular medicines
d were out of stock.
• High medicines prices; Moreover, health service managers in Uganda recognised that many potential users
• Unreliable delivery systems; were unwilling to pay the user fee in public health units (typically a flat fee of approx.
• Irrational use of medicines. US$0.30) in a context where non-availability of medicines was the rule rather than
exception. As illustrated in chapter 3, utilisation of government health facilities
Segall (2003) makes a strong case for increased public funding of district health care remained low as users turned to private-not-for-profit (PNFP) facilities where
services in developing countries to reverse the overall decline in the standard of public availability was better (although utilisation here remained below capacity also), private
sector health services in the 1980s and 1990s. Budget and expenditure cuts associated clinics, or local drug shopse for self-medication. Notably, the average drug shop price for
with implementation of macroeconomic stabilisation and adjustment programmes led to a course of first-line oral treatment of fever and/or malaria was usually less than the flat
shortages of medicines and deteriorating facilities. Demoralised, poorly paid health fee charged by the health units. However, prices of other medicines were likely to be
workers charged patients under-the-table fees, diverted medicines, practised privately considerably higher. Faced with non-availability at government health facilities, users
during working hours and became increasingly uncaring towards patients. The situation may have used such monetary assessments to determine their choice of health provider
in Uganda at the end of the 1990s was perhaps worse than in other countries, following in order to minimise their out-of-pocket payments, and thus medicine costs could have
15 years of instability that broke down what was once considered a model health system been a critical factor influencing access to care. These observations have led to a
in Africa, and with the additional burden that post-conflict rehabilitation did little to growing consensus of opinion that availability and affordability of medicines influence
alleviate the health crisis inherited in 1986 (Macrae et al. 1996). health care seeking behaviour in Uganda.
In a  costing study in primary facilities and district hospitals in Balochistan,
Pakistan, Green et al. (2001) showed imbalances and inefficiencies in the allocation of
public resources, with medicines often under-funded relative to salaries (and severely
under-funded overall). Furthermore, the lack of a financial skills-set in decentralised
settings results in poorly administered budgets having little relation to the health needs
of the population. Similar constraints faced the districts in Uganda during the period
when the policies of decentralisation and user fees were introduced in the latter part of
the 1990s. Official user fees had little impact on efficiency at service delivery levels, and cIn a kit system, a standard package of basic medicines and health supplies is provided to health facilities for a
equity problems were not solved in Uganda, as elsewhere in Africa (Gilson 1997; and set number of cases (e.g. 1000 out-patients). The number of kits allocated to individual facilities can be varied at
chapter 3, Kirunga Tashobya et al. 2005). Not surprisingly, even the poor in Uganda had each distribution round according to caseload. The composition of kits may be reviewed annually every two
little option but to seek private care. years.dInjectables, antimalarials, antibiotics, and analgesic/antipyretics.
eAs noted by Ssengooba et al. in chapter 7, there are concerns about poor regulation and weaknesses in service
standards and quality in many of the drug shops and private clinics in the country.
66 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 67
3.3 Cost issues Bamako Initiative scheme,h applying full cost recovery on medicines obtained at NMS
prices, medicines availability improved but with negative effects on household
Given the importance attributed to medicines availability by households in Uganda expenditure and equity (MoH et al. 2001). In Tanzania, user fees similarly failed to
and the ease of access to private sector drug shops, high levels of household out-of- match expectations of improved medicines availability, quality of care and financial
pocket medicines expenditure are not surprising. In total, it is estimated that households sustainability (REPOA 2004). 
spend in the region of US$7.00-8.70 per capita on health on an annual basis (MoH Overall, public financing of medicines and supplies showed a pattern of decline
2004a; UBOS 2003; UBOS 2001). The trend has shown a modest decline in recent throughout the 1990s, eroded by currency depreciation, population growth, and an
years, with a significant decrease in health spending by the poorest in the population, as increasing burden of disease related to high HIV infection levels. At the same time, the
detailed in Deininger and Mpuga (2004). More than 60 percent (US$4.20-5.20) of this core primary health care programmes such as treatment of childhood illnesses, malaria
out-of-pocket spending is estimated to go on medicines (MoH 2004a). This is in case management and control of sexually transmitted infections introduced a syndromic
contrast to the US$0.40-0.45 per capita spent in 2000/01 on tradable medicinesf approach which increased medicines requirements. The essential medicine kits
through the government budget (including Government of Uganda resources and donor remained relatively unchanged during the last six years of their use and were phased out
budget supportg) for all levels of health care. However, it should be noted that at the end of 2002. By this time, the kits contributed at most just one-tenth of the
government spending buys far more medicine per unit of expenditure than household medicine requirements of government primary health care services in the country, and
spending, as the government can benefit from bulk purchasing efficiencies using the encouraged parallel supply of many items. 
National Medical Stores (NMS) (Gabra and Green 2000; EuroHealth 2004). The NMS
is a semi-autonomous medical supply agency serving the public sector. Similarly, faith- 3.4 Management and logistics issues
based organisations running the many PNFP health facilities can benefit from
economies of scale through the Joint Medical Stores (JMS; Kawasaki 2001). Retail Good managers may be able to improve the quality of health services by reducing
prices in rural drug shops are much higher, and quite variable according to a national waiting times, or extending opening hours, but are often less able to solve the problem
price survey conducted by the MoH using a standard methodology established by the of irregular supply and poor availability of essential medicines. Moreover, the biggest
World Health Organisation (WHO) and Health Action International. logistics challenges for ensuring adequate and regular medicines supply are present, not
Even adjusting for these price efficiencies, government funding of medicines has at urban-based hospitals, but at the health facilities at the primary care level (e.g. rural
been very low relative to out-of-pocket expenditure. In addition, the proportion of HSDs). Yet it is these facilities that are staffed by multi-purpose cadres who have limited
medicines funding actually reaching the primary health service delivery points has been capacity for budget and stock management. The above-mentioned Drug Tracking Study
even lower, according to the Drug Tracking Study (MoH 2002a) mentioned above. The found that prolonged stock-outs were most prevalent at the lowest level of health facility
health sector in Uganda is structured around different levels of health centre, from basic (health centre IIs), where stock-out time over the one-year review period was 40 percent
out-patient services provided in a small health centre, to national level referral hospitals. on average for a range of 10 key medicines, compared with 5 percent at the level of
For a full description of these structures, see chapter 1, Yates et al., Annex 1. In district hospital (MoH 2002a). Medicines supply management is complex, specialised,
particular, district health services are divided into small health sub-districts (HSDs) and often dependent on external factors such as resource allocation, available funds,
covering health centres from level IV downwards (see chapter 6, Murindwa et al.). procurement procedures, supplier performance, and distribution logistics.
Primary health care budgets and services are managed at the HSD level. Cash and in- Another problem in ensuring constant medicines supply in the 1990s was also one
kind contributions to medicines (e.g. essential drugs kits) for the health facilities at HSD of leakage (MoH 1998). For example, a study of 10 sub-hospital facilities (i.e. HSDs) in
level were estimated at only US$0.16-0.20 per capita in 2000/01, and the actual 1995 found a median drug leakage rate of 76 percent and a high rate of informal
utilisation of such contributions is less certain. charging, at 5-10 times the formal charge (McPake et al. 1999). However, the
On the whole, user fees had negative effects. In rural districts, 25 percent of government health service was not well run at this time and many staff were induced to
households had to work for others, sell assets, or borrow to raise money for health care work outside the government health units and to sell government medicines in order to
during the era of user fees (Lucas and Nuwagaba 1999). While 30 percent of user fee survive, due to non-payment or late payment of wages, low remuneration and several
revenue at government units was supposed to be used for medicine purchases, the reality other system deficiencies (McPake et al. 1999; Danida 1998). Concerns about
was closer to 10 percent (EuroHealth 1999) and thus user fees appear to have had widespread leakage led to calls for more effective audit and control mechanisms,
limited impact on medicines supply. At most, user fees would have added US$0.02-0.03 improved supervision and surveillance, and enforcement and prosecution as needed
per capita to funding medicines supply. In practice, medicines were usually sourced (MoH 1999). This reactive approach to problem solving may have further undermined
locally by the government units at private sector retail prices, and thus the actual health worker morale. 
contribution of user fees in real terms would have been negligible. Under the pilot Other systemic problems affecting medicines supply were also well known, including
constraints facing the NMS (Crown Agents 1999), problems of fragmented financing
from multiple sources and lack of monetisation of donor-funded supplies. Other
fExcludes vaccines, contraceptives, TB drugs and other medicines supplied without charge to providers and
users on public health grounds.
gSee chapter 7, Ssengooba et al., for further explanation of funding mechanisms to the health sector. hOnly two of 56 districts had implemented the scheme before the policy change abolishing user fees in 2001. 
68 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 69
problems included poor coordination of procurement and logistics systems, inefficient expenditure of decentralised funds at US$0.15-US$0.20 per capita led to much shorter
‘push systems’, limited management capacity at the MoH and district health offices medicines stock-out periods (19 percent on average). In contrast, a poor performing
(EuroHealth 1999; Gabra and Green 2000) and lack of exit strategies on the part of district (with a percentage stock-out time of 31 percenti ) was found to be spending as
projects procuring medicines (MoH 2002a). Further difficulties were created by MoH little as US$0.03-US$0.07 per capita. Among other things, this was due to their failure
disease-specific programmes focusing narrowly on their own requirements and goals to adhere to guidelines for using a specified proportion of budget funds received from the
rather than allowing the district services to determine their own needs. Looking at the government on medicines. 
bigger picture, it can be seen that by the turn of the century, three very different The Push-Pull study found that following the decentralisation of government
modalities for medicines supply (each of which had been designed to suit a specific phase services (see chapter 6, Murindwa et al.), giving more autonomy to local governments
of Uganda’s development, summarised in Table 1), were operating concurrently. in the delivery of public services, a de facto ‘pull’ system had already been introduced to
By 2002 the health sector had reached the conclusion that existing systems lacked some extent regardless of the readiness of individual districts. Using the results from
transparency and accountability and suffered from huge inefficiencies with duplication these two studies, a task force was set up to formulate an operational strategy for a
of effort, misallocation of resources, wastage and expiry. This paved the way for coherent transition from a supply system that was traditionally based largely on allocations of
reforms in the interests of an integrated medicines supply system under the direction of essential medicines pushed down from the centre to the districts, to a demand-based
the Working Group on Medicines Procurement and Management (a structure within (‘pull’) system (MoH 2002b). 
the sector-wide approach, see chapter 2, Oliveira Cruz et al.). It was agreed that the best way forward was to combine two financing mechanisms
for medicines supply. First, the government would continue to channel government
budget resources (including donor budget support) to districts for non-wage recurrent
Table 1: Modalities for medicines supply in the health expenditures, with the guideline that 50 percent of these funds would be spent
Uganda public sector 1986-2001 on medicines. Second, there would be new earmarked budgets for each district for
medicines purchased from the NMS (or JMS for PNFPs) in the form of ‘credit lines’
backed by centrally held funds at the MoH. Essential drug kits, and a range of other in-
kind commodities previously allocated by the MoH vertical disease-specific programmes,
would no longer be supplied through a top-down approach that was known to be
inefficient, difficult to track, and prone to wastage through expiration. The district
health services would instead order according to their requirements from a pre-printed
form, within the limits of a specified budget ceiling (i.e. their individual credit line). 
To consolidate all centralised sources of funding for medicines (donor and
government), and to support these credit lines, a basket account was set up at the MoH.
This new mechanism improved the predictability of funding flows, and helped to
smooth out the irregularity of transfers associated with different financial year or project
planning cycles. Some donated commodities are now monetised and converted into
‘virtual credit’, and allocated across districts along with the other resources, through a
more systematic and equitable process. The NMS and JMS are provided with a
guaranteed business volume (backed by a Memorandum of Understanding signed with
the MoH) over a longer time horizon, aimed at improving forecasting, price stability,
value for money, and performance targets (e.g. timeliness of scheduled deliveries).
In the spirit of decentralisation, the credit lines ensure that the districts and
hospitals own and incorporate the central resources into their budgets and work plans.
4. From reforms to results Moreover, these earmarked budgets are more accessible to the technical staff at the
service delivery levels (e.g. the lower level health facilities) than the cash funds under
4.1 Key reforms in medicines financing and supply systems the control of the district government. There is also now no scope for reallocation of
these earmarked resources to other activities. Finally, the flow of medicines to individual
Two well-timed studies initiated in 2001 pointed the way towards more integrated health units can be easily tracked because computerised systems at NMS and JMS
medicines supply systems for health facilities, namely the Drug Tracking Study (MoH provide detailed supply and financial documentation and management information.
2002a) and a Push-Pull Study (MoH 2002b). The Drug Tracking Study provided
i
important feedback on the factors determining good and bad medicines supply at health Stock cards for ‘tracer medicines’ were reviewed at 8 health units in each district over a 12-month period. The
facilities. In a well performing district, adherence to guidelines on procurement, and number of stock-out days in the review period was recorded. The percentage stock-out time represents thenumber of stock-out days divided by the number of days in the review period. The figures of 19 percent and 31
percent represent the average stock-out time for 10 tracer items in the well-performing district and poor
performing district, respectively.
70 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 71
It is interesting to note that the health sector in Uganda planned for this nationwide
transition from ‘push to pull’ with only 6 months’ preparation. This was made possible
by allowing the HSDs several options within the new systems, and by setting the pace Figure 2: Increase in the medicines budget in real terms (1999-2004)
of change according to local capacity. The HSD headquarters could use pre-printed
forms to make a consolidated order for all the facilities in the sub-district, or it could
send a batch of orders from individual facilities. An order could be for a number of
(updated) pre-assembled kits, or for variable quantities by item, or a combination of the
two. The new kit was empirically determined using a quantification method based
primarily on prescription demand. This used actual prescribing practice rather than ideal
practice, although still very much focused on a prioritised list of essential medicines for
a basic package of primary health care services (Nazerali et al. 2004). 
The transition was implemented, with minor delays, in early 2003, in close
collaboration with the NMS and JMS. Importantly, progress made so far with this new
hybrid of system modalities helped set the stage for the mobilisation of critically needed
additional resources for medicines. The possibility for further integration of other project
funds including those from new global initiatives, such as the Global Alliance for
Vaccines and Immunisation (GAVI) and the Global Fund to Fight AIDS, TB and
Malaria (GFATM), is also on the agenda. 
4.2 Increased budget allocations
There has been a clear prioritisation in health sector resource allocations towards
primary health care inputs. In real per capita terms, while the hospital medicines budget
increased from US$0.11 to US$0.21 over the period 1999/00-2003/04, the medicines Sources: MoH (Selected years); Own MoH/Danida Database.
budget for the HSDs increased three-fold over the same period, from US$0.11 to
US$0.36 (see Figure 2). In fact, the medicines budget for hospitals was stagnant for many
years and only recently showed an increase when in 2003/04, the credit lines were 4.3 Increased budget expenditure and medicines consumption
extended to cover basic requirements in the hospital referral system. The government
also provides a grant specific to the PNFP health units which is estimated to add It is one thing to demonstrate a growing budget for medicines and another to show
US$0.17 per capita to PNFP medicines expenditure (assuming usual percentage that this is translated into actual expenditures on medicines. Data collected from the
expenditure of recurrent budget on medicines).  Table 2 shows that increases were even NMS, the JMS and surveys in a sample of districts show dramatic increases in the value
greater in nominal terms and in local currency. of medicines purchased by mid-2004. The increase is particularly noticeable following
The total funding available to government and PNFP health services for medicines the introduction of the credit lines in 2003 and the subsequent incorporation of some
at all levels almost doubled from US$0.88 to US$1.65 per capita over a four-year period donated commodities into the credit lines (monetisation). Annual NMS sales alone
(2000/01 to 2003/04). This includes centrally procured, non-tradable supplies such as increased from a baseline of US$4.0-4.5 million prior to these reforms, to US$9.7
vaccines, contraceptives, anti-TB drugs, HIV tests, blood products, and other million in 2003/04, with a similar trend observed at the JMS (from US$6 million to
commodities supplied through donor project funding for specialised disease control. US$10.8 million). 
Excluded from this figure are new interventions additional to the minimum health care Based on NMS and JMS sales data, budget performance has differed across the two
package outlined in the HSSP, such as the pentavalent vaccine (US$0.60 per capita) different funding sources. The credit lines have performed relatively better than the
and anti-retrovirals (reaching US$0.40 per capita by 2006) that are funded by global government cash funds released directly to the districts, particularly at the health sub-
initiatives (i.e. GAVI and GFATM). district level (see Figure 3). Survey data indicate that in some cases, sources outside the
In summary, the medicines budget is now not only larger, it is also allocated NMS and JMS were used, or that part of the indicative cash budget for medicines was
equitably, and is utilised more efficiently. An increasing proportion of the medicines utilised for other purposes.
budget is now controlled by the service delivery providers themselves, who are better In part, this reflects service level limitations at NMS, where on average, only 66-75
informed of their requirements, thereby minimising waste and duplication.  percent of order value was served. Performance against the cash medicines budget has
been used as a proxy indicator of district management capacity and a score for this is
incorporated into a district league table that is now drawn up in the Health Sector
Annual Performance Report (see chapter 6, Murindwa et al.). This strategy will help to
72 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 73
5. Enabling framework for medicines sector reforms
Figure 3: Performance against the medicines budget 
(cash and credit lines) 2003/04 financial year 5.1 Role of the sector-wide approach
The HSSP recognised the importance of establishing an integrated medicines supply
system and of streamlined funding (MoH 2000). However, it was the sector-wide
approach (SWAp), initiated in 2000, that provided a framework for the required policy
consensus among partners, coordinated reform of medicines financing and logistic
systems, and the decision to re-institute a ‘public good’ role for the NMS. 
In chapter 2, Oliveira Cruz et al. provide a full account of the SWAp structures and
processes in Uganda. In particular, the annual stakeholder conferences - the Health
Sector Joint Review Missions (JRM) - provided for continuity, joint learning and
decision-making on several issues including reforms of the medicines supply system. For
example, the first JRM recognised the importance of MoH stewardship for this reform
and the need for an upgraded status for the MoH unit responsible for medicines and
health supplies, policy and regulation (although this is yet to be realised in full). The
2001 JRM acknowledged that medicines financing was pivotal in improving health
sector outputs. Consensus was reached on the budgeting requirements for medicines in
the sector, estimated at US$3.50 per capita, and realistic targets for the short to medium
term were set. In health sector stakeholder discussions on the budget process, priority
was given to medicines funding and there was strong support for the creation of the
Source: MoH (2004d) central basket account, the Essential Drugs Account.
The Drug Tracking Study (MoH 2002) was an undertaking by the JRM
focus support for the poor performers, and motivate improvements in utilisation of the stakeholders, and unlike previous studies and consultancies in this area, there was
medicines budget using the national supply agencies. widespread ownership of the findings and support for the recommendations, giving rise
As a critical demonstration of the impact of the increased budget allocations and to the Push-Pull operational strategy. The new funding and logistics system was fully
the improved management and allocation systems, ongoing quantification studies at the endorsed by a wide group of stakeholders at the National Health Assembly (NHA) in
district level are beginning to show increased consumption of priority medicines per 2003, after less than one year of operation. The NHA is attended by district political and
1,000 out-patients (MoH 2004b). These data indicate a better overall match of supply administrative leaders and parliamentarians as well as technical staff from the district
and demand resulting from the new pull systems, and quality improvement at a technical health departments. The same forum recognised the challenges facing the NMS that are,
level. The methodology also identifies where consumption exceeds prescription demand in part, related to the successes of the new systems and the associated increased volume
for a given medicine (suggestive of leakage), and so far this is occurring in less than 1 in of business, and agreed on a further undertaking for a technical review of the NMS. An
every 10 centres.  external team conducted the review, and the recommendations were accepted by the
Table 3 shows a trend of reducing stock-out time at the health centre (HC) level for MoH and health development partners, including government assurance against
k
several of the indicator medicines that are routinely monitored nationally.j These data privatisation for 5 years (EuroHealth 2004). As a result, Danida, the lead agency in the
are from periodic surveys because the Health Management Information System does not pharmaceutical sub-sector, allocated substantial new financial and technical support for
yet reliably capture information on medicines availability.  There has been substantive NMS. Again, much was achieved through a consensus approach, within the relatively
improvement in availability at the lowest level of health unit (health centre II) in the short period of one year. 
health sub-district and reduced stock-outs of treatments for malaria and pneumonia, As explained in chapter 2 by Oliveira Cruz et al., Working Groups support the
which together account for two-thirds of the disease burden. Poor availability of oral functioning of the SWAp and by 2003 the formal Working Group on Medicines
rehydration was attributable to poor delivery performance from the local manufacturer. Management had been fully established to further guide reform in this area. The mid-
Nevertheless, stock-outs remain surprisingly high. term review of the HSSP in 2003 encouragingly noted that "reforms in policy and
management of essential drugs to increase efficiency are already being put in place" (MoH
jThe ‘percentage of health units reporting a stock-out of any of six indicator medicines’ is one of 20 national 2003a: 76).
indicators for the Health Sector Strategic Plan (MoH 2000). Stock-out time for each of the medicines, as
presented here, is actually more useful at this point, when stock-outs are still more the rule than the exception
(60 percent of health centres on average experienced a store-room stock-out each month according to survey
estimates). Data are not compared for injectable contraceptives (not routinely available at health centre level II)
or measles vaccine (unreliable due to missing data). kDanish International Development Assistance, part of the Danish Ministry of Foreign Affairs. 
74 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 75
5.2 Role of decentralisation
The decentralisation process in Uganda has moved at a much faster pace than in
other countries in the region. The whole government has been decentralised, with a
wide range of powers and resources transferred to the district level (Jeppsson and
Okuonzi 2000). Of particular focus here, however, is how the decentralisation process
posed both opportunities and threats for the medicines reforms.
One strength of the decentralisation process is the greater ownership and control
over resources that has been provided to districts, thereby enabling district health
managers to plan and allocate resources to match local needs. This is particularly
important to focus limited resources on priority interventions and the main burden of
disease, while avoiding wastage from drug supplies that have expired due to non-
utilisation. This greater accountability at local level is also useful for medicines, which
are tradable goods and, as was shown in the past, at risk of leakage. Moreover, the credit
lines offered greater transparency and clear entitlements that did not apply in the top-
down push systems.
In contrast, one known weakness has been the limited district capacity to manage
and coordinate the medicines logistics cycle, resulting in variable performance between
districts and even among health sub-districts. A recent human resource tracking study
in Uganda (MoH and AMREF 2005) identified unfilled posts for pharmaceutical cadres
in the public sector as a critical area needing improvement. Thus the push-pull
transition was formulated such that users could graduate from simple ordering options to
more complex ones, at their own pace. 
Guidelines from the MoH require that districts procure medicines from the two
national supply agencies, NMS and JMS (for PNFPs), as these agencies meet the
required technical standards as well as value for money. However, procurement decisions
by local government (LG) may be biased in favour of entrepreneurial private sector
suppliers who offer commission or other incentives not provided by NMS and JMS. LG
regulations are often cited in support of these alternative suppliers. The reliance on such
private-for-profit suppliers means higher prices and mark-ups, and could undermine the
performance and future viability of the NMS and JMS. 
6. Discussion and conclusions
6.1 Progress
In a relatively short period of time, the health sector has taken a leap forward in the
strategic area of organising medicines supply for improved quality of care and health
services utilisation. There is a better balance of resource inputs and dramatic
improvement in medicines funding overall. Available resources have been increasingly
prioritised towards the sub-district level and more efficiently used to match the burden
of disease prevented or managed at the primary care level.
This chapter presents the case that a combination of pro-poor reforms abolishing
user fees and new systems to address medicines financing and improved logistics
management promoted consumer preference for public primary health care facilities
over the poorly regulated private sector retailers operating in their communities. With
Table 2: Medicines budget for health sub-districts and hospitals
(Financial Year 1999/00-2004/05)
Financial Year
HEALTH SUB-DISTRICT 1999- 2000- 2001- 2002- 2003- 2004-
2000 2001 2002 2003 2004 2005
HSD cash budget for medicines (Uganda Shillings Billions) 1.0 2.3 5.0 7.2 9.5 9.5
ED kits and/or credit lines (Uganda Shillings Billions) 2.4 2.4 2.4 4.0 7.2 7.2
Total HSD medicines budget (Uganda Shillings Billions) 3.4 4.7 7.4 11.2 16.7 16.7
US$ per capita drug budget $0.11 $0.12 $0.17 $0.28 $0.36 $0.36
HOSPITAL
Hospital cash budget for medicines (Uganda Shillings Billions) 3.5 3.4 4.4 4.8 5.2 5.2
Hospital credit lines (Uganda Shillings Billions) 0.0 0.0 0.0 0.0 5.4 5.4
Total Hospital medicines budget (Uganda Shillings Billions) 3.5 3.4 4.4 4.8 10.6 10.6
US$ per capita drug budget $0.11 $0.09 $0.10 $0.12 $0.23 $0.23
Exchange rate (1US$:Uganda Shilling) 1500 1800 2000 1750 1850 1750
Population (millions) 20.47 21.19 21.93 22.70 25.40 26.29
Sources: MoH (Selected years); Own MoH/Danida Database.
76 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 77
respect to quality of care at public facilities, Nabyonga et al. (2005) found there were Uganda provides a useful empirical case study of policy change and reforms with
fewer medicine stock-outs in 2002 compared with 2000 and 2001. This was the first sign specific focus on medicines, while placed within overall health sector and government
of improvement in the supply system. Ultimately, reduced household health spending reforms. This enriches the somewhat limited literature on evaluations of health sector
may be better proof of improved access to medicines through primary health care reforms in sub-Saharan Africa reviewed by Gilson and Mills (1995). Recent country
facilities. While the national household survey of mid-2003 did not show a dramatic case studies focussed on areas such as health care financing reform (South Africa and
decline in out-of-pocket expenditure compared with 2001, it may have been too early to Zambia; Gilson et al. 2003) but without specific consideration of medicines financing.
judge the effect of these changes, and there had already been significant reductions in Much of the published country experiences examine medicines policies at the
health spending by the poor. The next survey will be more appropriately timed to assess community level in the context of user charges or cost recovery and/or revolving drug
the medicines sector reforms in relation to both access and equity, answering the funds following the Bamako Initiative, and are reviewed by Gilson (1997). The Uganda
question ‘Did the sub-sector reforms benefit the poorest of the poor?’ case is unusual, but perhaps illustrative of a new trend towards abolition of user fees and
Since the 1990s, health sector reforms in a number of developing countries have greater emphasis on public financing of essential medicines.
been evaluated within an analytic framework presented by Walt and Gilson (1994). We Country experience with the supply of essential medicines using ‘kit systems’ was
have used this framework to interpret the medicines sub-sector reforms in Uganda, reviewed by Haak and Hogerzeil (1995) but there is little documentation on the
recounting how reforms were implemented, and the important roles played by both transition to demand driven (‘pull’) systems based on end-user budgets. We found a
technical and political actors through the annual joint review missions and national single study from one region of Tunisia, reporting successful order-based, cost-conscious
health assemblies. This included consideration of factors of process and context such as management of the medicines budget by facility medical staff (Garraoui et al. 1999).
the participatory and inclusive Ugandan health SWAp, the broad model of Guidance on ‘what works’ and how to institutionalise sustainable medicines supply
decentralisation, institutional reform and support for the autonomous central medicines systems is needed from WHO and Danida based on their accumulated experience of
supply agencies (NMS and JMS), constrained pharmaceutical human resources and supporting government and/or autonomous central supply agencies. WHO has an
poorly enforced regulatory systems at district and sub-district level. ongoing multi-country study of five alternative medicine supply systemsl (WHO 1998)
and a multi-country project supported by the Gates Foundation may also be expected to
Table 3:  Stock-out time for indicator medicines at health units (HU) by level synthesise country experience once completed.
In general, the focus of publication appears to have moved from the local to the
Percentage Stock-out Time global, addressing issues of trade and pharmaceutical pricing in the new context of
A B C patent enforcement through the Trade-Related Aspects of Intellectual Property Rights
2003-04       2002-03        2001 (TRIPS) agreement, the Doha declaration that aims to safeguard access to effective
6 districts    6 districts    4 districts medicines for priority diseases (malaria, TB, HIV/AIDS), and research and development
strategies that aim to address neglected diseases in developing countries.
ALL LEVELS (HC II - HC IV) 36 HU 36 HU 32 HU
1 Chloroquine tab                                                                 12%                   6%                 7% 6.2 Challenges
2 Sulfadoxine Pyrimethamine tab (S-P) 9% 23% 41%
3 Cotrimoxazole 480mg tab 22% 36% 35% Overall, despite the considerable progress made in improving the management and
4 Oral Rehydration Salts (sachet) 27% 18% 37% funding of medicines supply, particularly for primary health care activities in Uganda, a
Average 18% 21% 30% number of challenges remain to be tackled. 
LOWEST LEVEL (HC II) 10 HU 10 HU 12 HU Shortage of medicines remains the main complaint in the recent user satisfaction
1 Chloroquine tab 12% 10% 6% survey with regard to government health services (UBOS 2005), with less than half of
2 Sulfadoxine Pyrimethamine tab (S-P) 15% 47% 59% respondents grading this category of service as either ‘fair’ or ‘good’ while more than 70-
3 Cotrimoxazole 480mg tab                                                  29% 47% 54% 80 percent graded responsiveness of staff and cleanliness of the facility as fair/good. The
4 Oral Rehydration Salts (sachet) 22% 25% 87% actual flow and volumes of medicines reaching the lowest level of health facility are still
Average 20% 32% 52% less than optimal and percentage stock-out times for key primary care medicines are
unacceptably high. There is need both to understand why these stock-outs persist and to
NOTE: Data were missing for part of the review period due to incomplete stock records. Data completeness formulate effective strategies and interventions to remedy them. Among other reasons
varied by item: >90% for items 1-3 and 50-66% for items 4. Review period was 6/12 months in surveys A and
B, and 12/12 months in survey C. The number of stock-out days in the review period was obtained from the for continuing stock-outs (e.g. paradoxically, increased out-patient utilisation induced
item stock card. The stock card may not have been complete for the whole review period (some missing data). by increased supply), in a setting where health workers aim to maximise client
The percentage stock-out time was calculated based on the actual number of days reviewed. satisfaction, high patient demand for injectable medicines or antibiotics may be served
Sources: MoH (2003b); MoH (2004c); MoH (2002a; column C)
lCentral medical stores, autonomous supply agency, direct delivery system, primary distributor system, and fully
private supply.
78 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 79
at the expense of technical standards for rational medicines use (Jitta et al. 2003). compulsory licensing to protect public health, may be so beset by regulations that there
Changing behaviours for more appropriate use of medicines will require long-term is reasonable doubt that "the solution can be effectively implemented" (Messerlin 2005:
efforts. 1199).
The situation is complicated since, in the Ugandan context where the private sector The price of these new medicines and related diagnostic tests may be dictated by
is not well regulated, gaps in the supply at government health units will benefit local agreements between global agencies and the pharmaceutical industry (e.g. WHO/Roll
drug shops or private providers, who can at the same time be government health Back Malaria and Novartis in the case of the new ACT medicine branded "co-artem").
workers. We need to better understand vested interests and their influence on provision Such products are increasingly funded and supplied through global initiatives often with
and consumption of health services. A future Drug Tracking Study should use qualitative vertical programming and parallel funding structures. Unless this new funding is
research methods, and include a community component to assess the possible factors incorporated into existing channels, the recent efficiency gains in Uganda may be
behind these stock-outs. It will be increasingly important that the regulatory framework reversed, and the different strands woven together by the reforms could once again be
keeps pace with the public sector reforms in Uganda, to ensure effective regulation and unravelled. Undermining of the role of national agencies and integration mechanisms,
enforcement of the private sector in the areas of medicines use and professional practice. and concerns about the financial sustainability of such initiatives remain unresolved
Strategies for coping with expansion of medicine as a business, and greater consumerism issues that will need to be addressed by the health sector in Uganda and elsewhere if the
in health care, were recently identified after a mapping of the regulatory system in gap in access to medicines is to be closed (Segall 2003; Quick 2003). 
Thailand (Teerawattananon et al. 2003).
Resource constraints (both financial and human) may soon limit increases in out-
patient utilisation. The budgets for 2004/05 and 2005/06 financial years show no growth
for medicines in real terms. A study of demand for health services and quality of care in
Mali suggests that structural attributes of quality, such as medicines availability, are
necessary but not sufficient conditions for demand (Mariko 2003). It is important to
remember that the reforms to medicines supply investigated in this chapter are only one
of a series of supply and demand side reforms that are needed to achieve continued high
utilisation of primary health care services in the country if health outcomes are to
improve. The medicines supply reforms thus need to be seen in the context of the overall
reforms outlined in this book. For example, improvements in the area of human resource
development may facilitate placement of pharmaceutical personnel at the HSD level, or
improve the remuneration of health workers, and these could be critical factors for
improved service coverage. 
Improved public expenditure and medicines supply are expected to translate into
higher utilisation of health services overall. However, with increased autonomy
provided in the spirit of decentralisation, differences in utilisation rates between health
facilities are likely to continue. This is likely to be related to variable performance at the
management level (district and/or sub-district level) and facility level, leading to
differences in medicines spending, availability, access, or other quality factors affecting
user satisfaction.  
Finally, recent trends in health care technology in combination with trade and
intellectual property agreements, may translate into substantially higher medicines
prices in developing countries. New medicines and health products, such as artemisinin-
based combination therapy (ACT) for malaria, anti-retroviral medicines for HIV/AIDS,
and the pentavalent vaccine for universal immunisation, are driving an exponential
increase in the essential medicines and vaccines budget for a national minimum health
care package (Nazerali and Oteba 2005). These products tend to come at fixed high
prices. Patent protection for these new products and restrictions or royalties on
domestically manufactured generics under the TRIPS came into force in 2005 in middle-
income countries such as India, South Africa and Brazil. These are significant producers
and exporters of generics. While enforcement in the poorest countries will not take
place until 2016, importation from countries that use the TRIPS flexibilities, such as
80 Health Systems Reforms in Uganda: processes and outputs 4 Medicines – driving demand for health services 81
References MoFPED. 2002. Uganda Participatory Poverty Assessment Study 2 Report – Deepening the Understanding of
Poverty. 2002: 104, 115. Ministry of Finance, Planning and Economic Development, Government of Uganda. 
Crown Agents 1999. Procurement and Logistics Review, 2/11/98-27/11/98, January 1999. Report to Danish Red MoH. 1998. Drug Leakage Meeting. Report dated April 1998. Ministry of Health, Government of Uganda.
Cross/Danida. National Medical Stores, Entebbe, Uganda.
MoH. 1999. Way Forward on Drug Leakage. Report dated February 1999. Ministry of Health, Government of
Danida 1998. Mid-Term Review of the Uganda Essential Drugs Support Programme, 23 Nov-12 Dec 1997.  Final Uganda. 
Report June 1998: 40-41. Ministry of Foreign Affairs, Denmark, Danida. 
MoH. 2000. Health Sector Strategic Plan 2000/01 to 2004/05: 74-76. Ministry of Health, Government of
Deininger K, Mpuga P. 2004. Economic and Welfare effects of the abolition of health user fees: evidence from Uganda.
Uganda. World Bank Policy Research Working Paper 3276. The World Bank, Washington DC.
MoH, WHO, AMREF, Rukungiri District Health Team. 2001. Rukungiri Bamako Initiative Scheme Evaluation
EuroHealth. 1999. Consultancy Report on the Harmonization of the Pharmaceutical Supply System in Uganda, Final Report. Ministry of Health, Government of Uganda, World Health Organisation, Uganda Office, AMREF
Draft April 1999: 17. EuroHealth Group. For Danish Red Cross. Uganda, Rukungiri District Health Team, Uganda.
EuroHealth. 2004. Technical Review of the National Medical Stores. Final Report July 2004. EuroHealth Group. MoH. 2002a. Drug Tracking Study. Final Report, August 2002. Ministry of Health, Government of Uganda.
For Ministry of Health, Government of Uganda and Health Development Partners Group.
MoH. 2002b. Push-Pull Study and Push Pull Transition - Way Forward. Report, June 2002. Ministry of Health,
Gabra M, Green T. 2000. Review of Drug Procurement and Management in Uganda. Rational Pharmaceutical Government of Uganda. 
Management Project, Management Sciences for Health Arlington, VA for USAID, World Bank and Ministry
of Health. MoH. 2003a. Mid Term Review Report of the Health Sector Strategic Plan 2000/01–2004/05. Ministry of Health,
Government of Uganda.
Garraoui A, Le Fevre P, Ledoux M. 1999. Introducing management principles into the supply and distribution
of medicines in Tunisia. Bulletin of the World Health Organisation, 77 (6): 525–9. MoH. 2003b. District Medicines Management Programme Annual Survey 2003. Ministry of Health, Government
of Uganda.
Gilson L. 1997. The lesson of user fee experience in Africa. Health Policy and Planning, 12 (4): 273–85. 
MoH. 2004a. Financing Health Services in Uganda 1998/1999 – 2000/2001. National Health Accounts. June
Gilson L, Mills A. 1995. Health sector reforms in sub-Saharan Africa: lessons of the last 10 years. Health Policy, 2004: 25, 42. Ministry of Health, Government of Uganda.
32 (1-3): 215–43.
MoH. 2004b. Debriefing on the Drug Quantification Exercise in Seven Districts. Presentation, Kampala, Uganda,
Gilson L, Doherty J, Lake S, McIntyre D, Mwikisa C, Thomas S. 2003. The SAZA study: implementing health March 2004. Ministry of Health, Government of Uganda.
financing reform in South Africa and Zambia. Health Policy and Planning, 18 (1): 31–46. 
MoH. 2004c. District Medicines Management Programme Annual Survey 2004. Ministry of Health, Government
Green A, Ali B, Naeem A, Vassall A. 2001. Using costing as a district planning and management tool in
of Uganda.
Balochistan, Pakistan. Health Policy and Planning, 16 (2): 180–6. 
MoH. 2004d. Annual Health Sector Performance Report 2003/2004. Ministry of Health, Government of
Haak H, Hogerzeil HV. 1995. Essential Drugs for ration kits in developing countries. Health Policy and
Uganda.
Planning, 10 (1): 40–9. 
MoH. Selected years. District Transfers for Health Services (Selected years). Ministry of Health, Government
Jeppsson A, Okuonzi SA. 2000. Vertical or holistic decentralization of the health sector? Experiences from
of Uganda.
Zambia and Uganda. International Journal of Health Planning and Management, 15 (4): 273–89.
MoH, AMREF. 2005. Tracking Human Resource and Wage Bill Management in the Health Sector. Report February
Jitta J, Reynolds Whyte S, Nshakira N. 2003. The availability of drugs: what does it mean in Ugandan primary
2005. Ministry of Health, Government of Uganda and AMREF.
care? Health Policy, 65 (2): 167–79.
Nabyonga J, Desmet M, Karamagi H, Kadama PY. Omaswa FG, Walker O. 2005. Abolition of cost-sharing is
Kawasaki E. 2001. Drug Supply Systems of Missionary Organizations. Identifying Factors Affecting Expansion
pro-poor: evidence from Uganda. Health Policy and Planning, 20 (2): 100–8. 
and Efficiency: Case Studies from Uganda and Kenya. MPH Dissertation. Boston University School of Public
Health. Nazerali H, Mohamed K, Oteba M, Mwoga J. 2004. Managing the dynamics of drug supply, demand and health
services utilization – a new multipurpose tool for forecasting drug requirements and resource management.
Lucas H, Nuwagaba A. 1999. Household coping strategies in response to the introduction of user charges for
Paper presented at the Second International Conference on Improving Use of Medicines (ICIUM), Chiang
social services: a case study on health in Uganda. IDS Working Paper 86. The Institute of Development Studies,
Mai, Thailand, March 30 – April 2, 2004. Unpublished document. 
Sussex, UK.
Nazerali H, Oteba MO. 2005. Escalating medicines prices and the Uganda Minimum Health Care Package:
Macrae J, Zwi AB, Gilson L. 1996. A triple burden for health sector reform: ‘post’-conflict rehabilitation in
For Health or for Profit? Paper presented at the International Health Economics Association (IHEA 2005)
Uganda. Social Science and Medicine, 42 (7): 1095–108.
Conference, Barcelona, Spain, July 10-13, 2005. Unpublished document. 
Mariko M.  2003. Quality of care and the demand for health services in Bamako, Mali: the specific roles of
Quick JD. 2003. Essential medicines twenty-five years on: closing the access gap. Health Policy and Planning,
structural, process, and outcome components. Social Science and Medicine, 56 (6): 1183–96.
18 (1): 1–3. 
McPake B, Asiimwe D, Mwesigye F, Ofumbi M, Ortenblad L et al. 1999. Informal economic activities of public
Quick JD, Hogerzeil HV, Velasquez G, Rago L. 2002. Twenty-five years of essential medicines. Bulletin of the
health workers in Uganda: implications for quality and accessibility of care. Social Science and Medicine, 49 (7):
World Health Organisation, 80 (11): 913–4.
849–65. 
REPOA 2004. Equity implications of user fees in Tanzania. Report published by ETC Crystal, Netherlands.
Messerlin PA. 2005. Trade, drugs, and health-care services. The Lancet, 365 (9465): 1198–200.
Research for Poverty Alleviation (REPOA).
82 Health Systems Reforms in Uganda: processes and outputs
Ruxin J, Paluzzi JE, Wilson PA, Tozan Y, Kruk M, Teklehaimanot A. 2005. Emerging consensus in HIV-AIDS,
malaria, tuberculosis, and access to essential medicines. The Lancet, 365 (9459): 618–21. 5
Segall M. 2003. District health systems in a neoliberal world: a review of five key policy areas. International
Journal of Health Planning and Management, 18 (): S5-S26. Public-private partnership in health: working
Teerawattananon Y, Tangcharoensathien V, Tantivess S, Mills A. 2003. Health sector regulation in Thailand:
recent progress and the future agenda. Health Policy, 63 (3): 323–38. together to improve health sector 
UBOS. 2001. Uganda National Household Survey 2000/01. Report on the socio-economic survey. Uganda
Bureau of Statistics. performance in Uganda
UBOS. 2003. Uganda National Household Survey 2002/03. Report on the Socio-economic Survey. Uganda
Bureau of Statistics.
Peter Lochoro, Juliet Bataringaya, Christine Kirunga Tashobya and 
UBOS. 2005. National Service Delivery Survey, April 2005. Uganda Bureau of Statistics.
Joseph Herman Kyabaggu
Walt G, Gilson L. 1994. Reforming the health sector in developing countries: the central role of policy
analysis. Health Policy and Planning, 9 (4): 353–70. 
WHO. 1998. Managing Drug Supply. Essential Drugs Monitor No. 25 & 26, Editorial (WHO/DAP-
WHO/EDM). World Health Organisation.
Summary
One of the major reforms in the health sector of Uganda has been in the area of
public-private partnership in health. Partnerships in health are increasingly seen as a
valuable mechanism for moving faster towards the attainment of national and
international health goals. For example, recent initiatives such as the Global Fund to
Fight AIDS, TB and Malaria and the Poverty Reduction Strategy process contain an
emphasis on public-private partnership. Partnership in health in Uganda is a key feature
of the National Health Policy and the Health Sector Strategic Plan, and in the sector-
wide approach processes and structures. Although efforts at improving public-private
partnership date back to the pre-independence days, it is only in recent years that real
progress has been made with the development of a specific partnership policy for the
sector. In particular, partnership with the private-not-for-profit sub-sector has advanced
well and its representatives are involved in a number of policies and reviews within the
health sector at the national level. The government is also providing increasing
financial support to this sub-sector and structures have been established to facilitate
dialogue (e.g. a partnership office). There is evidence that the current partnership has
brought dividends to the whole health sector in terms of increased access, quality, equity
and efficiency, although as yet these rewards pertain mainly to partnership with the
private-not-for-profit sector and progress is yet to be seen with the other private partners
in the sector (for-profit, and the traditional and complementary medicine sub-sectors).
Existing constraints will have to be overcome to sustain the rewards of the partnership.
84 Health Systems Reforms in Uganda: processes and outputs 5 PPPH - improving health sector performance 85
1. Introduction developed capacity for health training and skills development that can be capitalised
upon by government. PNFP hospitals filled a critical gap in health service provision
Since the 1990s, there has been growing international interest in promoting public- during the 1970s and 1980s when the government health service had deteriorated
private partnerships in the delivery of health services (Cassels 1995; Green et al. 2002). considerably.
Partnerships are increasingly seen as a valuable mechanism for moving faster towards the However, by the early 1990s the health sector had undergone a time of increased
attainment of national and international health goals. Consider the emphasis on public- polarisation between public and private health providers, with little or no sharing of
private partnership in new initiatives such as the Global Fund to Fight AIDS, TB and information and no coordination in planning or development. A significant trigger
Malaria, the Poverty Reduction Strategy process (PRSP), the Global AIDS Vaccine point for the recent reform occurred after the mid-1990s when the PNFP sub-sector was
Initiative and many others. The objectives of these various health partnerships range on the brink of collapse. An emergency situation was declared by the PNFP hospitals
from developing a product (e.g. an AIDS vaccine) or distributing subsidised products, to (MoH 1996), whereby as a result of decreased capacity to mobilise external resources,
strengthening health systems and educating the public (Widdus 2001). the rising cost of health services delivery, the need to keep user fees affordable, and
In Uganda, public-private partnership in the health sector has been pursued with having exhausted efficiency gains, many PNFP hospitals had reached a point where
varying degrees of effort and success since pre-independence days.1 At that time, shutting down was the only option. 
voluntary health providers collaborated with government and received funding which In response to their critical condition, and in recognition of their contribution to
continued into the post-independence period. Owing to economic collapse and the health service delivery in the country, the government decided to provide financial
shrinking of the government budget in the 1970s and 1980s, this funding declined until subsidies to the PNFPs, representing a major step forward in the development of public-
it stopped altogether. private partnership in the sector. On the part of the PNFPs, partnership with the
However, despite the long history of interest in public-private partnership in health government was agreeable, once autonomy of management was preserved, as it provided
in Uganda, it is only in recent years that real progress has been made, in particular since a critical contribution to funding.
the early 2000s in tandem with the other key health systems reforms that are discussed
elsewhere in this book. In chapter 1, Yates et al. describe the impact that these reforms Table 1: Health facilities by ownership
have had on improving health sector outputs. The purpose of this chapter therefore is to
identify the contribution of public-private partnership in health to the improved Government PNFP PHP Total
performance of the sector. Also, on an international level, it is useful to provide a Hospitals 55 42 4 101
specific country example to inform the debate on public-private partnership issues in
public service delivery. Health Centre Level IV 
The private health sector in Uganda includes the private-not-for-profit (PNFP) or (Higher services with doctor and theatre) 151 12 2 165
voluntary providers (mainly facility-based health providers), the for-profit private health Health Centre Level III 
practitioners (PHP), and the traditional and complementary medicine practitioners (Providing in and out-patient service) 718 164 22 904
(TCMP) (MoH 2003a). Together, these private providers account for a sizeable
proportion of the health sector in Uganda (see Table 1). In particular, 42 percent of all Health Centre Level II 
hospitals are PNFP hospitals. In total, the PNFP and PHP sectors account for 42.5 (Basic out-patient facility) 1,055 388 830 2,273
percent of the health facilities in the country (not including the TCMP sub-sector that Total Health Units 1,979 606 858 3,443
is very diverse and constantly changing).
Source: MoH (2002a). 
2. Rationale for public-private partnership in health
3. Structures of the partnership
The rationale for entering into partnership with private health providers in Uganda is
straightforward. Given the sizeable contribution made by the private providers to the The events that followed the re-instatement of financial subsidisation opened the
delivery of health services in the country, it is argued that rather than creating duplicate way for a more structured partnership between the government and the private health
government structures and thereby wasting limited resources, it makes sense to sector. At the policy level, the need for public-private partnership in the delivery of
coordinate efforts and avoid competition. Moreover, the private sector has access to health services is identified in the National Health Policy of Uganda and in the Health
additional non-government sources of resources (e.g. PHP profits, PNFP donor Sector Strategic Plan (HSSP) (see chapter 1, Yates et al., Annex 1). Since then, a
contributions) that can be mobilised to support the common goal of improving the specific policy laying out the principles for public-private partnership in health service
health services available to the population. Over time, the private sector has also delivery has been developed (MoH 2003b). 
1Uganda achieved independence in 1962.
86 Health Systems Reforms in Uganda: processes and outputs 5 PPPH - improving health sector performance 87
This policy defines partnership as "The formal relationship between two or more Local Government
partners who have agreed to work together in a harmonious and systematic fashion and being Level
mutually supportive towards common goals, including agreeing to combine or share their
resources and/or skills for the purpose of achieving these common goals" (MoH 2003b: 5). The District Health To prepare annual plans, provide technical assistance to the health
To achieve the level of partnership described in the policy, the areas in which public Management Team facilities, inform, consult and coordinate with other district stakeholders,
and private health providers are to collaborate are wide ranging. In particular they monitor and evaluate progress, and propose improvements.
include partnership in the following areas: The District PPPH - To facilitate information flow between district authorities and 
Desk Officer PNFP/PHP representatives;
- Policy development; - To facilitate understanding and coordination of implementation 
- Coordination and planning; arrangements;
- Financial resource mobilisation and allocation; - To carry out assignments given by the technical head of health 
- Human resources for health management; services from within the existing district structure, in consultation 
- Capacity building and management; with the PNFP/PHP coordination committees.
- Community empowerment and involvement;
District To represent the PNFP/PHP facilities and programmes of different- Service delivery (management and implementation).
PNFP/District PHP ownership, to coordinate their facilities and programmes, to harmonise
Coordination approaches, define common issues, propose solutions, and ensureTable 2: Structures of partnership
Committees information exchange.
Structure of partnership Roles - To prepare plans and assist the health facilities and community 
Central Level The Health Sub- groups to deliver the Uganda National Minimum Health Care 
District (HSD) Package (UNMHCP) within the HSD;
Joint Review Mission A joint review of sector performance by GoU and Partners (i.e. Management - To inform, consult and coordinate with other sub-district stakeholders, 
districts, Parliament, NGOs, private sector and donors) on an annual
Committee monitor and evaluate progress, and propose improvements.
basis.
To advise and supervise hospital management on key operational
Health Policy Advisory A forum to discuss and advise the Ministry of Health and Hospital Board issues including finance and human resources.
Committee (HPAC) development partners on the implementation of the National Health
Policy and the HSSP. - To direct and supervise implementation and quality of UNMHCP 
Health Unit services;
Working Group on To facilitate dialogue between government and private health Management - To monitor and evaluate progress;
PPPH providers in preparation of guidelines and policy proposals, and to Committee - To supervise management of the health facility;
facilitate coordination of Ministries. The group is represented on - To liaise between management, community and stakeholders.
HPAC.
To prepare plans, reports and budgets to be presented to the Sub-
Ministry of Health PPPH - To advocate and facilitate partnership at different levels; Sub-county Health County Council (Local Sub-county council), and to the HSD
Desk - To facilitate operations research into specific PPPH issues; Committee management team.
- To collect and disseminate information;
- To liaise with private sector partners; - To collect and analyse data, identify the community’s health needs, 
- Secretariat to the PPPH Working Group and sub-working groups. Parish Development prioritise and implement appropriate responses;
Committee (PDC) - Where such committees exist, the private sector partners can provide
Inter-ministerial Standing To set the priorities, coordinate the stakeholders, monitor progress technical assistance and support to train PDC members, to assist in 
Coordinating Committee and refine the policy and strategic Human Resources development data analysis and generally advise the PDC during its deliberations. 
– Education and Health plan.
for training health - To identify the community’s health needs and priorities and develop 
workers. Village Health Team plans to make appropriate responses;
- To mobilise additional resources and monitor use of all resources for 
The Umbrella - To represent and coordinate the different health providers of the their health programmes including performance of health centres;
Organisations PNFP sub-sector; - To mobilise communities and to select and oversee Community 
- To provide support services and accredit the member facilities; Resource Persons;
- To coordinate and promote professional development and ethics; - To maintain registers of households and their health status, and to 
- For the PHP: to represent each professional category and promote serve as the primary link between the community and health service 
Partnership initiatives. providers.
Source: MoH (2003b).
88 Health Systems Reforms in Uganda: processes and outputs 5 PPPH - improving health sector performance 89
While the policy refers to partnership with the whole private health sector, in support is also provided in the form of salaries for government doctors who are posted to
practice, partnership has developed to varying degrees with the three different private PNFP hospitals and in the form of a credit line for medicines from the national level
sub-sectors. The PHP sub-sector has come on board recently and some progress has been pooled medicines fund (see chapter 4, Nazerali et al., for further description).
made in more clearly identifying partnership activities. In contrast, minimal progress has Since 1997/98, when financial support to the PNFPs was re-introduced, in the form
been made with regard to the TCMP sub-sector. While the need for partnership with of the conditional grant, the funding has grown significantly and has also been extended
this group is widely acknowledged, there are difficulties in identifying all the to PNFP health training institutions (see Figure 1). Total funding to the PNFP sub-
practitioners falling into this category, whose areas of focus range from traditional sector amounted to just 0.5 percent of the total health sector budget in 1997/98, and this
herbalists to what might be classed as sorcerers and witchdoctors. A Bill is currently had grown to 7 percent by the year 2002/03 (see chapter 7, Ssengooba et al.).
being drafted, the aim of which will be to facilitate definition and regulation of the This funding made a considerable contribution to the financial sustainability of the
TCMP and prepare the sub-sector for a better-structured partnership. PNFP health units. For example, in the year 2001/02, government funding from the
Partnership with the PNFP sub-sector has been the most successful to date. In PNFP conditional grant constituted nearly 30 percent of the budgetary requirements of
particular, the leaders in this process have been the religious-based institutions, the PNFP health units (Bataringaya and Lochoro 2002). Thus, despite accounting for a
represented by their umbrella organisations: the Uganda Catholic Medical Bureau small proportion of the total health budget, this funding is extremely valuable to the
(UCMB), the Uganda Protestant Medical Bureau (UPMB), and the Uganda Muslim PNFP health units. 
Medical Bureau (UMMB). Thus the remainder of this chapter will focus on progress in While government funding is restricted to the PNFP sub-sector at present, the way
partnership with the PNFP sub-sector. may be opened for specific publicly funded activities in the PHP sub-sector in the future,
In terms of partnership structures, a coordination office and partnership mechanisms with the aim of expanding access to services in under-served areas in the country. 
have been established both at national and more gradually at district level (see Table 2).
Through the sector-wide approach (SWAp) to health sector management, partnership
at the national level has been facilitated. As explored in more detail in chapter 2 Figure 1: Government funding1 to PNFP health institutions
(Oliveira Cruz et al.), the health SWAp itself is a partnership arrangement between the
Government of Uganda, development partners and other stakeholders to the sector.
Within these SWAp structures, the participation of the PNFP sub-sector (through its
umbrella organisations) was able to take on a formal status in national level health
policy discussions. The PNFP umbrella organisations are represented on the SWAp
Health Policy Advisory Committee and at the annual stakeholder review missions and
the budget allocation discussions for the sector. Such involvement has also opened the
way for the interaction of the PNFP umbrella organisations at policy levels even beyond
the Ministry of Health (MoH), in particular with the Ministry of Finance, Planning and
Economic Development (MoFPED).
4. Government efforts towards the partnership
The development of the partnership policy and the practical arrangements that
have been made for planning and coordination with the PNFP sub-sector have been led
by the government, with strong support from the PNFP umbrella organisations.
However, the most critical contribution by government to date with regard to the
partnership has been the provision of increasing financial support. A specific PNFP
conditional grant is channelled to PNFP health units via the district government health
office (see Annex 1 in chapter 1, Yates et al., for a description of the health service
structure in Uganda). The funds are provided for recurrent expenditure and are Source: MoH (Selected Years). 
1
conditional on the preparation and approval of a comprehensive work plan and budget The amounts exclude a wage subvention that is approx. Ushs 0.7 billion annually 
and the signing of a Memorandum of Understanding with the local government. The (exchange rate: 1 US$ = 1750 Ushs at July 2004).
decision to route the funds through the district government facilitates partnership
between the PNFP health units and the district government health system. Additional
90 Health Systems Reforms in Uganda: processes and outputs 5 PPPH - improving health sector performance 91
5. Private sector efforts towards the partnership examples of this increased utilisation following the reduction in user fees at specific
health facilities are presented in chapter 3, by Kirunga Tashobya et al.
In addition to the government grants, the sub-sector continues to mobilise funds Second, with regard to efficiency, the monetary cost of providing services in PNFP
from other sources including user fees, donations from abroad and other income- health units has been increasing on a yearly basis, in line with the rest of the health
generating activities. Extrapolating from UCMB data, it is estimated that in the year sector, which may indicate rising quality of care across the sector (see below). However,
2002/03, user fees accounted for 45-50 percent of the PNFP budget while the other in 2002/03, for the first time, UCMB hospitals demonstrated a fall in the cost per unit
sources (mainly donations and income-generating activities) funded approximately 25 of output compared to the previous year 2001/02. There has also been an observed
percent. increase in output per unit of staff (Giusti et al. 2004). The latter indicator in particular,
Driven by the decline in other sources of funding (in particular, foreign donations), given a consensus that PNFP facilities have been underutilised on the whole, suggests
user fee levels in PNFPs had gradually increased over time. A central element of the that efficiency improvements have been achieved by the PNFP health facilities in recent
PNFP mission is to support the health of the population, with particular focus on the years.
poor and most vulnerable, and yet this rising trend in user fees had led to a decline in Third, in terms of quality, improved performance has been found in at least one
utilisation by the poor (see chapter 3, Kirunga Tashobya et al.). Over the last two years, indicator of quality, the proportion of qualified staff at the health facility. This
growing financial support from government grants and improvements in management proportion has increased as a result of improved pay conditions for staff at the PNFP
efficiency have enabled many of the PNFP health facilities to adjust their user fees facilities (Giusti et al. 2002). An improvement in quality of care in PNFP facilities due
downwards. At least 81 percent of the PNFP hospitals and a number of lower level to an increased proportion of qualified staff is therefore possible. 
health units have been able to lower fees, leading to increases in utilisation, especially Although not explicitly demonstrated, an improvement in equity may also be
for child in-patient stays and general out-patient consultations (Odaga and Maniple suggested by the results. First, it is known that 86 percent of the PNFP health units are
2003). based in rural areas (MOH 2001a) where 85 percent of the poor reside (MoFPED 2004).
Under the guidance of the umbrella organisations, PNFP facilities were encouraged Thus it is likely that the majority of the increased utilisation observed above was
to lower their fees for selected services, focusing on those that benefit the vulnerable in accounted for by the rural poor. Moreover, with regard to equity, the largest jump in
particular (e.g. maternity services, child health services). In addition to fee reduction, a utilisation was observed for child health services, especially for in-patient utilisation.
policy of flattening the fee structure was pursued, making the fee structures much more These results indicate that prior to the user fee adjustments in these PNFPs, very sick
predictable and transparent. Training, technical assistance and planning tools for price- children needing hospitalisation were being excluded from health care. Also, the user
setting were also provided to the participating health facilities. fee reductions were focused on services for women and children, identified as the most
Figure 2:
6. The rewards of the partnership
To assess the rewards of the partnership, in particular those of the partnership with
the PNFP sub-sector, available analysis has looked at performance in four areas, namely
access (utilisation), efficiency, quality and equity. A UCMB initiative has collected and
analysed data across a range of PNFP facilities under its umbrella, both as an evaluative
exercise and as a tool for facility level management (Giusti et al. 2002; Giusti et al.
2004). UCMB health facilities represent the major share of this sub-sector, accounting
for 64 percent of PNFP hospitals and 46 percent of PNFP lower level units. As a general
assessment, it is noted that the current partnership has enabled the PNFP sub-sector to
make a substantial contribution to improvements in health sector performance (Giusti
et al. 2002) and has helped services to increase access for the poor (Giusti et al. 2004). 
First, in terms of access, the reversal of the decline in utilisation of health services
was immediate, occurring just after the introduction of the subsidies. Figures 2 and 3
show the upward trend in utilisation of composite units of output2 and this rise in
utilisation has continued at an even steeper pace into the present day. This is attributed
to the effect of the government subsidy in replacing user fees and allowing the charges
to be gradually pushed downwards, as shown in Figure 4 (Giusti et al. 2004). Other
Source: UCMB (2004).
2A measure which adds in-patients, out-patients, deliveries, immunisations and antenatal attendances all
weighted for their cost relative to the out-patient cost.
92 Health Systems Reforms in Uganda: processes and outputs 5 PPPH - improving health sector performance 93
vulnerable in the population and thus another indication that the fee adjustments had The administrative partnership arrangements, for example the government grant
a positive impact on equity. and the medicines credit line, have used government planning, budgeting, accounting
All in all, we conclude that improved performance in the four areas of access, and reporting procedures for both public and PNFP providers. This is seen by many in
efficiency, quality and equity can be attributed to better funding of the health units, the PNFP sub-sector as weakening their autonomy and introducing a degree of
improved focus and capacity of the institutions (better planning, better qualified staff), inflexibility that may lead to loss in value for money. They consider it necessary to
and support and supervision from the umbrella organisations made possible by the introduce contractual arrangements between the government and the private health
growing public-private partnership in health. providers to smooth over administrative differences and to maintain the autonomy of
the private sector. 
Figure 3: Despite improving remuneration, the PNFPs are still unable to competitively recruit
and maintain health workers. Most of the losses of staff from the PNFP are to
government employment posts (MoH 2005), due to both pull (higher government
salaries) and push factors (e.g. more stringent management practices, fewer
opportunities for professional development in the PNFP sub-sector (MoH 2001b)). The
human resource constraint is probably the most difficult test for partnership currently.
The PNFP units have been working to reduce the salary gap between public and PNFP
staff, only to be met with further increases in public wages and increasing staff loss. This
is compounded by stagnation in the size of the PNFP grant against a growing health
sector budget. Furthermore, in successive government budget allocation forecasts, the
size of this grant has been flattening which gives a worrying indication of a decline in
emphasis on the partnership on the part of the government (UCMB 2005). For example,
in the budget allocation of 2002, the PNFP grant was projected to rise quite steeply over
the medium term, while in the latest medium-term budget allocations, the size of the
grant is projected to grow much more slowly (see Figure 5). Maintenance of the positive
achievements of the partnership will hinge on finding a solution to these constraints.
Figure 4: Median fees charged per Standard Unit of Output OP in 
UCMB Hospitals (Not adjusted for time preference)
Source: UCMB (2004). 
Note: The degree of completeness and reliability of data from lower level units prior to 2000/01 
were poor.
7. Constraints
There remain a number of constraints that still need to be tackled in the area of
public-private partnership (MOH 2003a). The TCMP, the PHP and the non-facility
based PNFPs have poorly developed structures for adequate representation and dialogue
and help is needed to improve their internal organisation.
While the partnership concept has been well received at the national level, there is
much less appreciation at the district level, and both public and PNFP district partners
have been slow to engage with the partnership process. The main reason behind this
divergence possibly lies in the fact that the impetus for partnership originated at the
national level and has only slowly been gaining the acceptance of the districts. In many
cases, at district level, there remains mutual suspicion between the public and private
health facilities with each complaining that the other has secret, non-disclosed sources
of funding. The dominant motivation has often appeared to be one of competition (for
resources) rather than of coordination (in service delivery). 
Source: Giusti et al. (2004).
94 Health Systems Reforms in Uganda: processes and outputs 5 PPPH - improving health sector performance 95
are purchased from the private sector (Verhallen 2001). In contrast, the Ugandan case
Figure 5: Projected budget allocations to PNFPs over time is seen as a more extensive partnership, with partnership areas ranging from policy
development to service delivery. The overall goals and objectives of the government and
private health providers, in particular the PNFP providers, overlap significantly, leaving
little room for exploitation of either party, and consequently the need for legal structures
is limited. Moreover, support of the private sector by the government has been shown to
yield very positive results in health sector performance. Financial support to the PNFP
sub-sector, equivalent to 30 percent of their budget, led to an increase in outputs of over
50 percent in a set of UCMB hospitals between the years of 1997/98 and 2002/03. Such
a significant output improvement over time was achieved despite the growth in the
government grant to this sub-sector to only 7 percent of the total health budget by
2002/03. 
Finally, the value of the partnership with the PNFP sub-sector goes beyond that of
achieving health output improvements, as it has helped this sector to refocus itself on its
original mission – to assist the poor and vulnerable in the country in the hope of
alleviating poverty and improving equity in the population. While much of this chapter
has focused on the PNFP sub-sector, it is possible that similar benefits in terms of
improved health sector performance can be achieved as greater partnership at the
district level is encouraged, and that with the PHP and TCMP sub-sectors is further
developed.
Sources: MOH (2002b); MoH (2003c); MoH (2004); MoFPED (2003); MoFPED (2005).
BFP: Budget Framework Paper.
MTEF: Medium Term Expenditure Framework.
8. Conclusions
The public-private partnership in the health sector has advanced ahead of many
other developing countries and international initiatives. It is clear that in promoting the
partnership, the government appreciates the valuable role played by private health
providers in the delivery of health services to the population. In particular, there is a
clear understanding that the partnership seeks to utilise as much as possible the different
capacities of all the partners from policy development through to service delivery.
The current level of partnership has been facilitated by a number of factors. The
installation of the new government in 1986 paved the way for wide reforms in all sectors
including health. The sector-wide approach, initiated in the year 2000 (see chapter 2),
was one of the key reforms that provided an enabling environment for partnership. Due
to sustained economic growth and continued financial support from donors, resources
for government services have increased over time (see chapter 7), and have thus allowed
some contribution towards private sector partners, in particular the PNFP health
providers. 
In many countries in sub-Saharan Africa, partnership with the private sector is
largely understood in terms of a funding contract whereby mainly non-clinical services
Uganda Shillings Billions
96 Health Systems Reforms in Uganda: processes and outputs
References
Bataringaya J, Lochoro P. 2002. The Framework of Public-Private Partnership. Health Uganda Health Bulletin,
8, July–December: 294–301. 6
Cassels A. 1995. Health Sector Reform: Key issues in less developed countries. Journal of International
Development. Special Issue on Health Policies in Developing Countries, 7 (3): 329–438. Meeting the challenges of decentralised health
Giusti D, Lochoro P, Mandelli A. 2002. Public-Private Partnership in Health: What is its Effect on the
Performance of the Health Sector? Uganda Health Bulletin, 8, January – March: 22–26. service delivery in Uganda as a component of
Giusti D, Lochoro P, Odaga J, Maniple E. 2004. Pro-Poor Health Services - The Catholic Health Network in
Uganda. Development Outreach, 6 (1) March: 18–21. The World Bank Institute. broader health sector reforms
Green A, Shaw J, Dimmock F, Conn C. 2002. A shared mission? Changing relationships between government
and church health services in Africa. International Journal of Health Planning and Management, 17 (4): 333–53.
Grace Murindwa, Christine Kirunga Tashobya, Joseph Herman Kyabaggu, 
MoFPED. 2003. Medium Term Expenditure Framework 2004/05-2006/07. Ministry of Finance, Planning and
Economic Development, Government of Uganda. Eliseus Rutebemberwa and Juliet Nabyonga
MoFPED. 2004. Background to the Budget for Financial Year 2004/05. Ministry of Finance, Planning and
Economic Development, Government of Uganda.
MoFPED. 2005. National Budget Framework Paper 2005/06-2007/08. Ministry of Finance, Planning and
Economic Development, Government of Uganda.
MoH. 1996. Memorandum of the PNFP hospitals to the Ministry of Health. Ministry of Health, Government of Uganda.
MoH. 2001a. Public-Private Partnership in Health. A quantitative Survey of the PNFP Health Facilities in Uganda.
Ministry of Health, Government of Uganda.
MoH. 2001b. Public Private Partnership in Health, The Effects of Government Recruitment of Staff from PNFP
Health Facilities. A study report. Ministry of Health, Government of Uganda.
MoH. 2002a. Health Facilities Inventory October 2002. Ministry of Health, Government of Uganda. Summary
MoH. 2002b. Budget Framework Paper 2002/03. Ministry of Health, Government of Uganda.
Decentralisation in its present form was introduced in Uganda in the 1990s. Uganda
MoH. 2003a. Mid Term Review Report of the Health Sector Strategic Plan 2000/01–2004/05. Ministry of Health,
Government of Uganda. has a devolved form of government where local governments have extensive political
and administrative powers. The health sector has been markedly influenced by this
MoH. 2003b. The National Policy on Public-Private Partnership in Health – Final Editorial Draft, October 2003.
Ministry of Health, Government of Uganda. decentralisation. In the National Health Policy and the Health Sector Strategic Plan
(2000/01-2004/05), the sector has sought to decentralise health services even further by
MoH. 2003c. Budget Framework Paper 2003/04. Ministry of Health, Government of Uganda. creating lower level management structures at a sub-district or county level. At the same
MoH. 2004. Budget Framework Paper 2004/05. Ministry of Health, Government of Uganda. time, the health sector undertook a number of other reforms to improve the delivery of
MoH. 2005. Human Resource Constraints of the PNFP. A Report of the PNFP Human Resource Task Force to primary health care services in the districts. These included reforms to human resource
HPAC May 2005. Ministry of Health, Government of Uganda. management and physical infrastructures, and the establishment of appropriate
MoH. Selected years. District Transfers for Health Services for the Financial Years 1997/98 to 2003/04. Health structures and institutions for support supervision and performance monitoring. This
Planning Department, Ministry of Health, Government of Uganda. chapter investigates each of these reforms in turn, noting where progress has been made
and where challenges still remain. Particular attention is paid to the impact of the
Odaga J, Maniple E. 2003. Faithfulness to the Mission, Effect of Reducing User Fees on Access to PNFP
Health Services. Unpublished report. reforms on efficiency, equity and community participation in the health sector. We
conclude that the Ugandan health sector has performed quite well against a number of
UCMB. 2004. UCMB Database. Uganda Catholic Medical Bureau. these objectives, but that performance is varied across and within districts. In particular,
UCMB. 2005. PPP for Health: Looking at Trends. UCMB Position Paper to HPAC, April 2005. Uganda Catholic remote areas suffer from a number of drawbacks where, for example, staff are unwilling
Medical Bureau, Kampala, Uganda. to be stationed in hard-to-reach unattractive locations. Thus, under the new Health
Verhallen M. 2001. WHO Inter-country Meeting: Lessons from Health Sector Experiences in Contracting in Sector Strategic Plan for the period 2004/05-2008/09, a number of challenges still need
Contracting in Africa, Addis Ababa November 2001. Report to MMI. to be addressed in order to further improve the decentralised delivery of health services
Widdus R. 2001. Public-Private Partnerships for health: their main targets, their diversity, and their future in the country. 
directions. Bulletin of the World Health Organisation, 79 (8): 713–20.
98 Health Systems Reforms in Uganda: processes and outputs 6 Decentralised health services management 99
1. Introduction 2. Ugandan primary health care delivery system in the 1990s
Decentralisation can be defined in many ways. This chapter draws on the work of Following decentralisation in the 1990s, District Health Management Teams, under
Møgedal et al. (1995), defining decentralisation as a "dynamic process of changing the leadership of District Medical Officers (DMO), were responsible for the planning,
relationships between the centre and the periphery with the local level taking on more and more management and provision of health services in the districts. Thus, the district teams
authority and responsibility" (p.355). Moreover, within this definition, different forms of were now expected to handle both the day-to-day issues of service delivery (e.g.
decentralisation can be described, including deconcentration, devolution, delegation managing the logistics of medicines delivery to the health units) and the more strategic
and privatisation. Using these definitions, the form of decentralisation that was functions of planning, coordination and resource mobilisation. However, these teams
introduced in Uganda is best described as devolution, where the main emphasis is on the were seriously constrained in terms of human, financial and logistical resources. Not
creation or strengthening of local government. Decentralisation in its present form was surprisingly, therefore, the capacity for management of the decentralised district health
introduced in Uganda in 1992 and by 1997 all districts in the country had established system was generally low (MoH 1998). 
local governments with full administrative and political structures in place (MoLG Compounding these basic capacity problems, district health services suffered from a
1994).a However, it is important to bear in mind that when these definitions are applied number of other constraints in the 1990s. First, geographical access to basic health
to real situations, a mix of different forms of decentralisation is likely to be observed services was low, where only 49 percent of the population was estimated to be living
within the same country/system and Uganda is no exception (Rondinelli et al. 1993; within a 5km radius of a health facility. By 2000, this had increased to 57 percent (MoH
Møgedal et al. 1995).  1991; MoH 2000a). Districts in Uganda have an average population of 500,000,
Decentralisation in Uganda occurred as part of the broad civil service and public although land surface area and terrain vary from district to district, affecting ease of
sector reforms which were introduced over the late 1980s and 1990s in the context of a access to services. Thus there were wide variations in access to basic health services
country emerging from over a decade of civil conflict with many economic challenges. across districts and within districts, with some of the districts reporting less than 10
These reforms were aimed at improving the accountability of public resources and percent of the population living within a 5km radius of a health facility. Looking at
making spending more efficient. specific elements of basic health care, there was even wider variation in access across
Prior to the 1990s, management of public services, including health, was centralised the country, where, for example, some districts were not able to provide emergency
and districts were responsible for carrying out functions delegated by the central surgical and obstetric services. 
government. In the health sector, districts were delegated the responsibility for the Second, there were problems with human resources, both in terms of physical
management of the smaller health units below the level of the hospitals. The Ministry manpower and management issues. As part of the civil service reforms in the 1990s, the
of Local Government (MoLG) was responsible for supervision of the districts, while government imposed a ban on recruitment of civil servants, including health workers.
management of general and regional hospitals was under the responsibility of the A study carried out by the MoH in 1999 showed that only 34 percent of the established
Ministry of Health (MoH). Following the reforms of the 1990s, the districts became positions in health centres were filled by qualified health workers (MoH 2000b). The
responsible for the management of all health services within their jurisdiction with the remaining positions were either vacant or filled by untrained nursing aides. Moreover,
exception of regional and national referral hospitals (Hutchinson et al. 1999). After the distribution of the few available qualified health workers was skewed towards
almost 10 years of decentralisation, the health sector stakeholders realised that the hospitals and urban areas. Local governments were responsible for the management of
decentralised districts were still big both in terms of size and population. In 1999, human resources for district health services and many problems were experienced,
therefore, the health sector further decentralised the delivery and management of health particularly with regard to payment of salaries. Local governments were provided with
services to a level below the districts, thus creating a Health Sub–District level of unconditional block grants from central government from which they were expected to
management. pay, among other things, salaries of civil servants including health workers. Yet in many
This book discusses many health sector reforms that have taken place in Uganda districts, salaries for health workers were not being paid on a regular basis. Some local
since 2000. The focus of this chapter is on decentralisation, looking specifically at the governments continued to recruit health workers even though they could not access the
implications of the extended decentralisation undertaken by the health sector, and on formal payroll. These workers were supposed to be paid either through user fees
other infrastructural, human resource and supervision reforms in the sector. Given the collected at the health centre or through sub-county local revenues, but in fact many
focus on decentralisation, and hence on district health systems where primary health never received any salary at all (MoH 2000c). 
care services are delivered, we mainly address primary health care services, which is in Third, there was weak supervision and monitoring of health services at all levels.
line with the central priority outlined in the Health Sector Strategic Plan (HSSP) With a limited number of skilled supervisors and inadequate logistical support, neither
(MoH 2000b) and the National Health Policy (NHP) (MoH 1999). the MoH nor the district teams were able to carry out regular and appropriate
supervision and monitoring (MoH 2003a). At the district level, planning for activities
such as supervision was usually undertaken without appropriate reference to available
resources (financial and human), and it was often simply assumed that the required
aBy 2000 Uganda had a population of 21 million, spread over 45 districts. The population is now inputs would somehow materialise. 
close to 25 million and some districts have been sub-divided. For much of the period under review
in this book, the total number of districts was 56; this has now increased further to 69. 
100 Health Systems Reforms in Uganda: processes and outputs 6 Decentralised health services management 101
Clearly, decentralised district health services suffered considerable drawbacks in the (PNFP) sector, thus demonstrating one of the ways in which partnership with the PNFP
1990s. With the development of the NHP (MoH 1999) and the first HSSP 2000/01- sector has developed in Uganda, as discussed in more detail by Lochoro et al. in chapter 5. 
2004/05 (MoH 2000b), the health sector aimed to address these problems in order to However, while the concept was introduced in the late 1990s, the HSD strategy was
improve the delivery of primary health services in the country. implemented at a varied pace across the country. This was due both to variations in
human resource and infrastructural capacity across the regions and also to differing levels
of interest by the political and administrative leaders in the districts. For example, in
3. Efforts to improve decentralised health service delivery some areas, the DDHSs readily embraced the strategy, whereas in others, the change was
considered to represent a loss of power and influence and was therefore not encouraged.
3.1 The Health Sub-District strategy There have been particular problems encouraging hospital staff at HSD headquarter
hospitals to consider the management of the wider HSD rather than just that of the
As mentioned above, the health sector extended the reach of decentralisation hospital. By now, all districts have operational HSDs in place, although with continued
through the creation of a health sub-district level of management. This shift was varying levels of functionality (MoH 2004a). 
provided for in the NHP and outlined in detail in the Health Sub-District Strategy It is possible, however, to assess the performance of the HSD strategy in light of the
Paper in 1998 (MoH 1998). A health sub-district (HSD) is a functional sub-division of objectives outlined above. In terms of improved management of health services, progress
the District Health System with the main objective of improving the management and to date has been mixed. In some districts the HSD headquarters provide effective
delivery of health services at the local level. It is therefore located within a district, management of resources, in particular of medicines and financial resources (and to a
covers a population of approximately 100,000 and is equivalent to a county/constituency lesser extent, human resources). Good use is made of information available through the
(see chapter 1, Yates et al., Annex 1). Health Management Information System (HMIS) and other data sources in order to
prepare realistic and sensible HSD work plans and to make decisions for the
The main objectives of further decentralisation to HSD level were to: improvement of services. In the districts where such HSDs are operating, the DDHS and
other members of the District Health Team are released from having to deal with day-
• improve the quality and management of routine health service delivery; to-day operational issues and are able to concentrate on the more strategic roles of
• increase equity of access to essential health services; district planning, supervision, and coordination with the central government and
• foster community involvement in the planning, management and delivery of development partners (MoH 2004a). Moreover, the results of such improvements in
basic health care. management at district and sub-district level can be seen in terms of service delivery
performance. Districts that perform well in a league table prepared by the MoH
These three are key principles of the HSSP, and as such the HSD strategy was put (discussed further below) are more likely to be those that have made progress on the
in place to ensure they could be achieved. management side (MoH 2003c). 
Following the creation of the HSD, new roles and responsibilities were assigned to In contrast, in other districts progress has been much slower, with some of the roles
the district level. The district level health offices, now headed by the District Directors of the HSD still being carried out by the District Health Team. Limited capacity is a
of Health Servicesb (DDHS), became responsible for overall leadership, strategic major factor in this as many of the health workers at HSD level do not have the requisite
planning, supervision, monitoring and coordination of district health services. The HSD skills for planning, management or supervision. To address these problems, a programme
headquarters were made responsible for operational planning and management of health of training the HSD core management teams in planning, organisation and management
care delivery within their individual catchment area (i.e. health service delivery, was initiated by the MoH, although progress has been slow and to date about 50 percent
monitoring and supervision of all the basic health services in the HSD). As part of the of the HSD teams have been covered (MoH 2004d). Another major challenge has been
broader strategy to bring basic services closer to the people, each HSD was required to the high turn-over of staff at the HSDs, which is disruptive to attempts to establish
c smooth long-term planning and management processes. have a hospital or a high level health centre (level IV) to provide emergency surgical
Looking at the second objective of HSD decentralisation, increased equity of access,
and obstetric services. This health facility forms the headquarters of the HSD, and is some progress has been observed, particularly in terms of resource allocation processes.
referred to as the HSD Referral facility. It provides technical and management Each HSD currently receives a budget from the central government, and the formula for
leadership for the rest of the health units (i.e. health centres at levels II and III) in the allocating these funds across districts and HSDs has improved over the years. The
designated area. The facility is managed by a team of senior health professionals headed allocation has moved from being a fixed amount for each HSD, to one that is based on
by a medical officer or, in his/her absence, a clinical officer or public health nurse. It is population and other demand side variables (e.g. poverty, morbidity), as described in
also noted that this health facility can be from either the public or private-not-for-profit more detail by Ssengooba et al. in chapter 7. Also, as a result of these changes, a far
greater proportion of resources is now spent on service delivery inputs at the HSD level
bThe title of the technical head of the District Health Team was changed from District Medical Officer to District rather than on management functions at the district level, thereby improving the
Director of Health Services in reflection of the adjustment of the post from one of direct service provision to one availability of basic health services for the population. 
of health systems management. 
cSee Chapter 1, Annex 1, by Yates et al. for a description of the structure of the health system.
102 Health Systems Reforms in Uganda: processes and outputs 6 Decentralised health services management 103
On the other hand, some inequities may have been exacerbated since the upgraded. From 2000 to 2003, over 400 new health centres (level IIs) providing basic
implementation of the HSD strategy. Some districts are still unable to recruit medical out-patient services have been constructed and a further 180 have been upgraded to
officers to head the HSDs, possibly because the districts are too remote and hard-to- level IIIs to provide in-patient, laboratory and maternity services. In addition, a total of
reach or because of poor human resource management practices by the districts in 150 new operating theatres and doctors’ houses have been constructed at health centre
question. While this inequity might not be caused by the HSD strategy per se, it does level IV (MoH 2003d). A considerable amount of new equipment has also been
indicate that there are challenges to fully implementing the strategy universally across purchased to improve the functionality of the health facilities as well as a number of
the country. vehicles to facilitate outreach services, deliveries (medicines, vaccines, gas etc.), and
According to the third objective, the HSD strategy was expected to accelerate the referrals (MoH 2004c). 
process of community involvement in the management of health services (and in the As a result of these efforts, population access to health facilities has increased and
management of their own health) by bringing health service management structures the proportion of the population living within a 5km radius of a health centre had
closer to the population. Local level managers have a greater understanding of local reached 72 percent by the end of 2004 (MoH 2004b). 
requirements and needs, and are thus better placed to take these needs into account in One of the downsides of these construction activities is that the rate of expansion,
decision-making processes. In the health sector, such community participation is particularly of the lower level health units, has been much faster than was planned due
facilitated through the establishment of Health Committees and Health Unit to a high level of interest by the population and political leaders and the multiplicity of
Management Committees (HUMCs). In the HSD strategy, several tools highlight the financial resources earmarked for capital development (e.g. specific project funding
need for community participation. For example, guidelines for the development of work etc.). Consequently, many of the new units lay unused for a year or more due to a lack
plans identify the need for such participation and for the integration of community of complementary resources such as staff and equipment (MoH 2003c; MoH 2004a). 
health-related priorities and activities with those of the health facilities. Another new
initiative has been the development of Village Health Teams (VHT), the goals of which 3.3 Human resource management
are to encourage communities to take responsibility for their own health and well-being,
and to participate actively in the management of their local health services. As outlined earlier, numerous human resource problems existed in the health sector
In practice, progress in encouraging community participation has been mixed. in the 1990s and concerted efforts were needed to address these problems affecting
There has been some improvement in the degree to which HSDs involve communities probably the most critical input to health service delivery. Health sector stakeholders
in decision-making processes. There are also varying levels of co-operation between the successfully negotiated with appropriate institutions in government (Ministries of
political, administrative and technical leaders at this level. Often, the higher academic Finance and Public Service), which led to lifting the ban on recruitment of health
qualifications of the health workers at the lower levels of government intimidate the workers. Thereafter, provision was made in the health sector budget for the recruitment
other leaders, hindering co-operation. However, in many cases, local government of over 3,000 new primary heath care workers, and by 2003 85 percent had been
committees have been involved in decisions to allocate local level resources to the recruited. As a result, the proportion of established posts filled by qualified health
health sector, particularly for construction activities, whenever these have been workers increased from 34 percent in 1999 to 53 percent in 2003 (MoH 2003c; MoH
available (e.g. general non-sector specific development funds received from central 2003d).
government or locally collected tax revenue). The HUMCs were much more active at To improve payment conditions, in 2001/02 the health workers’ payroll was
the time when public facilities were collecting user fees as they participated in the recentralised (i.e. moved back under the control of central government) in order to
management of these resources. However, since the abolition of these fees (see chapter ensure all the properly appointed health workers were receiving their salaries regularly,
3, Kirunga Tashobya et al.) many of these HUMCs are no longer in place or are not and that arrears were provided for. Furthermore, particular effort has been made by the
functioning. The VHTs have been established in only a few districts and their government to increase health workers’ salaries, although the level of remuneration
operationalisation requires a considerable amount of resources for training, equipment, remains a thorny issue and is certainly not high enough to stem the flow of health
and logistics, and as yet these resources are not available in current budgets (MoH workers to developed countries. 
2003c). To further address human resource capacity constraints, over 4,000 nursing aides
have undergone a 3-month in-service training programme to become nursing assistants,
3.2 Improving physical access to primary health care services and similar in-service training programmes for other cadres of health workers (e.g.
anaesthetic assistants, theatre attendants) have been provided. A number of other
In addition to the HSD strategy and the associated efforts to improve the trained health workers have undertaken further training to improve their technical
management of local level health services, the health sector has also focused on competence. On the management front, nearly all DDHSs have received postgraduate
improving physical access to health services within the decentralised context. Over the training in Public Health/Health Services Management, resulting in improved capacity
period of the health reforms, with the support of various stakeholders (e.g. central and for the management of district level health services (MoH 2003c; MoH 2003d). 
local governments, development partners, communities, external funding agencies), Despite these improvements, there remains a critical problem of variability in
many new health facilities have been constructed and other existing ones have been human resource availability across districts. The capacity to recruit qualified health
104 Health Systems Reforms in Uganda: processes and outputs 6 Decentralised health services management 105
workers varies from district to district with some districts able to attract a higher number monitoring of key output performance indicators of out-patient utilisation and
of qualified health workers than others. Not surprisingly, health workers are attracted to immunisation since 2000. A new innovation since 2002/03 has been to rank districts in
urban areas, and more easily accessible and attractive districts, resulting in a skewed terms of performance against a selection of 12 indicators (input, process and output
distribution of human resources across the country. The health sector is under increasing indicators) to create a District League Table. The major objective of this league table is
pressure to resolve this issue and to come up with appropriate incentives to attract staff to compare performance among districts, recognise good performers in order to learn
to the more remote areas.d In addition, the public sector comes under pressure of from their success, and to identify the poor performers and develop ways in which to
competition from the more lucrative private-for-profit sector, particularly where supply support them (MoH 2003c). As mentioned earlier, the league table helped to identify
is limited (e.g. pharmaceutical and diagnostic staff). As a further constraint, high entry management factors as important contributors to good performance in district health
requirements for nurse training programmes (set by the Professional Council for Nurses) services. However, while collection and monitoring of performance data has improved
have seriously jeopardised the selection of students into the training schools, especially over time, utilisation of these data to inform planning and decision-making at district
in the hard-to-reach areas (which tend to be more poverty-stricken and hence less level continues to be poor.   
academically successful) where the need is greatest. Finally, it is important to note that decentralisation both supports and challenges
the new roles and responsibilities of the different levels of the health sector. With the
3.4 Support supervision and performance monitoring continued progress in the decentralisation of public services in Uganda, local
governments have a high level of autonomy in political, administrative, and
Under the framework of decentralisation, the roles and responsibilities of the increasingly, financial management. This poses a particular challenge to the MoH in
central government, in this case the MoH, change from service management and fulfilling its roles and responsibilities, as in this context, the MoH may struggle to ensure
delivery to those of policy formulation, quality assurance and standard-setting, capacity that nationally agreed policies and standards are adhered to, and that resources are used
development, technical support, and monitoring and evaluation of overall sector in the most efficient and equitable manner.
performance. As already outlined, the roles of the district health services are to plan,
coordinate, supervise and monitor, while the HSD is responsible for actual service 4. Discussion
delivery. In particular, a key role for each level of management in the health system
(central, district and sub-district) is to provide support and supervision to the levels that The focus of the reforms outlined above has been to improve the delivery of
fall under their responsibility. This creates a structure of mutually reinforcing and decentralised primary health care services in Uganda. Since the turn of the century,
coordinated supervision from the top right through to the lowest level health unit. particular efforts have been made to improve management at the lower levels of the
The MoH has made good progress in fulfilling some of these key roles and health system and to support the provision of the key health inputs of human resources,
responsibilities. Support supervision from the centre is now better coordinated with health infrastructure and medicines (the latter are addressed by Nazerali et al. in chapter
different MoH programmes conducting integrated district visits, and annual meetings 4). This is in recognition of the fact that in addition to demand side reforms such as the
are held to help districts develop individual work plans. As a result, many districts are abolition of user fees and supply side reforms such as improved allocation of public
now able to produce credible work plans and provide appropriate health services for the resources, management reforms are equally important. Management issues at the health
populations they serve (MoH 2003c; MoH 2003e; MoH 2004a). Drawing on guidance service delivery level are often not given due consideration, and instead health sector
received from the centre, the district is then expected to be able to support the HSD in reforms are equated with high-profile policies such as the abolition of user fees or high-
the preparation of their work plans, and the HSD subsequently helps the lower level level organisation management in terms of restructuring of Ministries of Health (Frenk
health units, although success in this continues to vary from one HSD to another. 1994; Cassels 1995; Møgedal et al. 1995). It is important therefore to consider the
However, the support offered to districts from the MoH is still far from optimal, with impact that these local level management reforms have had on the delivery of primary
complaints that supervision visits continue to be poorly coordinated, planned and even health care services in the country in the context of the other demand and supply side
implemented. Thus a recent initiative has been to implement an Area Team Strategy, reforms that have taken place concurrently and are discussed elsewhere in this book. 
whereby groups of officers from technical programmes within the MoH (e.g. health As already highlighted by Yates et al. in chapter 1, in combination, the health sector
planning, child health, reproductive health, quality assurance etc.) are designated to reforms have yielded observable improvements in key health outputs. While it is
provide continuous support to a small number of districts each (i.e. 3-8 districts per difficult to isolate the contributions of the different reforms to these outputs, it is
group). This set-up is envisaged to initiate a continuous, effective and sustainable possible to assess the individual reforms against useful criteria of efficiency, quality,
mentoring relationship, and to provide appropriate follow-up to problems raised.  acceptability, relevance and equity, as suggested by Møgedal et al. (1995). 
In addition to providing support supervision, more formal monitoring of As already noted, the specific reforms investigated in this chapter have shown
performance also takes place. The HSSP includes a comprehensive monitoring variable performance. The efforts made to further decentralise the management of
framework for the sector, with indicators for assessing performance at national and health services have already yielded positive results in some districts, in terms of
district levels. As outlined in chapter 1, considerable progress has been achieved in the increased capacity for planning and management of the lower level health units and
dAn incentive scheme has been proposed but is yet to be approved by the Ministry of Public Service.
106 Health Systems Reforms in Uganda: processes and outputs 6 Decentralised health services management 107
more equitable distribution of health resources across the country. This would suggest As discussed in more detail by Oliveira Cruz et al. in chapter 2, there are some emerging
improvements in quality and equity. However, the implementation of the HSD strategy factors that threaten to destabilise this supportive environment, such as the resurgence
has not yet delivered any significant improvements in community participation, which of disease-specific global initiatives which tend to be dominated by central level vertical
would suggest a more modest score on the criteria of acceptability and relevance. programmes and are not as conducive to a decentralised context. 
Similarly mixed performance has been achieved on the part of the human resource,
health infrastructure, supervision and monitoring reforms. For example, the increase in
the numbers of filled qualified staff positions, the continued on-the-job training, and the
efforts to improve staff motivation (e.g. payroll improvements) all point to improved
quality and efficiency in service delivery. Clearly however, serious gaps in efficiency and
quality still exist given the proportion of health worker positions that remain unfilled. 
In addition, while there have been marked improvements in physical access to
health services with clear attempts to redress equity concerns, the mismatch between
infrastructure development and availability of complementary inputs implies
considerable room for improvement in resource-use efficiency. Similarly, the criteria of
quality and acceptability are not being fully met either. Finally, the supervision and
monitoring framework still faces a number of challenges. In particular, all levels of the
health sector need to internalise the importance of appropriate performance assessment
in order to guide decision-making and achieve more efficient use of resources to improve
service delivery. 
Thus, rather than seeking to isolate out the specific impact that these reforms to
decentralised primary health care services have had on the improvements in overall
health output indicators, this discussion has shown where these reforms have succeeded
and where further progress is needed. Certainly, it is possible to see how these reforms
have contributed to the improved health sector performance. As argued by Cassels
(1995), health sector reform is not a sequential process but rather that improvements in
organisation management will often occur both in parallel and sometimes in response to
other aspects of reform (Cassels 1995; Møgedal et al. 1995). Thus it is important to
remember that the reforms in Uganda were undertaken in parallel and that some of the
improvement is due not to the individual incremental changes, but to the way in which
these reforms interacted and responded to each other.
5. Conclusions – lessons for the future 
Overall, as shown in this chapter, the reforms that focused on improving the
decentralised system of health service delivery have led to improvements in efficiency,
quality, acceptability, relevance, and equity in the primary health care sector. These can
in turn be assumed to have made a significant contribution to the notable improvements
in health sector outputs in the country since 2000. 
However, as also noted here, a number of challenges still remain and progress on
some of these reforms has been variable, particularly across districts and sub-districts.
The next phase of reforms will take place under the guidance of the second HSSP for
the period 2005/06 to 2009/10. In preparing the second HSSP, a crucial part of the
process was to incorporate into the plan strategies and actions to address these
challenges. It will be equally important to ensure that the new HSSP operates in an
environment that is as supportive as the one in which the first HSSP was implemented.
108 Health Systems Reforms in Uganda: processes and outputs
References
Cassels A. 1995. Health sector reform: key issues in less developed countries. Journal for International 7
Development, 7 (3): 329–47. 
Frenk J. 1994. Dimensions of health system reform. Health Policy, 27 (1): 19–34.    Have systems reforms resulted in a more efficient
Fumihito S. 2001. Decentralization theories revisited. Lessons from Uganda. Faculty of Intercultural
Communication, Ryukoku University, Japan.    and equitable allocation of resources in the
Hutchinson P, Habte D, Mulusa M. 1999. Health Care in Uganda: Selected Issues. World Bank Discussion Ugandan health sector?
Paper No. 404. The World Bank, Washington DC.
Møgedal S, Steel HS, Mpelumbe G. 1995. Health sector reform and organisational issues at the local level:
lessons from selected African countries. Journal of International Development, 7 (3): 349–67.
Freddie Ssengooba, Rob Yates, Valeria Oliveira Cruz and Christine Kirunga Tashobya
MoH. 1991. Health Facilities Mapping. Ministry of Health, Government of Uganda.
MoH. 1998. The Health sub-district Strategy. Ministry of Health, Government of Uganda.
MoH. 1999. The National Health Policy. Ministry of Health, Government of Uganda.
MoH. 2000a. Health Facilities Inventory. Ministry of Health, Government of Uganda.
MoH. 2000b. Health Sector Strategic Plan 2000/01 to 2004/05. Ministry of Health, Government of Uganda.
MoH. 2000c. Primary Health Care Conditional Grant Monitoring Reports. Ministry of Health, Government of
Uganda.
MoH. 2003a. Support to Districts in the era of Decentralisation and SWAp: Area Team Strategy, September 2003.
Ministry of Health, Government of Uganda. Summary
MoH. 2003b. Aide Memoire of the 9th Joint Review Mission. Ministry of Health, Government of Uganda.
MoH. 2003c. Annual Health Sector Performance Report 2002/2003. Ministry of Health, Government of One of the claims of the sector-wide approach in the health sector is that improved
Uganda. coordination mechanisms should lead to a better allocation of resources within the
MoH. 2003d. Mid Term Review Report of the Health Sector Strategic Plan 2000/01–2004/05. Ministry of Health, sector. In particular, it is argued that the pooling of financial resources, in order to fund
Government of Uganda. a coherent sector-wide plan, should lead to improved efficiency. This chapter explores
whether or not there is any evidence to support this assertion in the case of the Ugandan
MoH. 2003e. Primary Health Care Conditional Grant Monitoring Reports. Ministry of Health, Government of
Uganda. health systems reforms since 2000. It also asks if the reforms are addressing another vital
performance measure, that of improving the equitable allocation of health care
MoH. 2004a. Area Team Monitoring Report March-April 2004. Ministry of Health, Government of Uganda. resources. Due to the complex nature of health sector financial flows, providing an
MoH. 2004b. Health Infrastructure Inventory Update. Ministry of Health, Government of Uganda. accurate analysis of the breakdown of expenditure is problematic. However, an
assessment of the largest financing mechanisms, the government budget and donor
MoH. 2004c. Health Policy Statement 2004/05. June 2004. Ministry of Health, Government of Uganda.
projects, seems to indicate that overall, allocative and operational efficiency have
MoH. 2004d. Health sub-District Teams Training Reports. Health Planning Department, Ministry of Health, improved. This has largely been driven by improvements in the performance of the
Government of Uganda. government budget, which increased allocations to district primary health care services
MoLG. 1994. Decentralization in Uganda: The Policy and implications. Vol. 2. Decentralization Secretariat, by a considerable proportion in only four years. This in turn led to an increase in the
Ministry of Local Government, Government of Uganda. provision of services available for the rural population and for addressing the major
Rondinelli et al. 1993. Decentralisation in Developing Countries. Staff Working Paper 581. The World Bank, causes of ill health. Donor projects appear to be less efficient at allocating resources to
Washington DC. these services, so it may be appropriate that this mechanism has been overtaken by the
budget as the largest contributor to sector funding. Increased funding for rural services
also indicates a more equitable allocation of resources towards high need groups.
However, the battle has not yet been won and a number of issues still need to be
addressed, in particular the overall shortage of resources for the sector. 
110 Health Systems Reforms in Uganda: processes and outputs 7 Resource allocation in the Ugandan health sector 111
1. Introduction rather than the more dispersed and distant hospitals. As a result, the health plan sought
to increase the proportion of funds allocated to primary care services as opposed to
The health sector in Uganda is under-funded to such an extent that current secondary and tertiary care in hospitals – where health benefits per shilling were likely
available resources are estimated to fund just 32 percent of the amount required to fully to be lower. Allocative efficiency in this context meant that the proportion of resources
implement the country’s Health Sector Strategic Plan (HSSP)1 (MoH 2000). allocated to district health services (primary care) should have increased in comparison
Nevertheless, there are indications that efforts to increase allocative efficiency within with hospital services (MoH 2003a).
existing resources could result in appreciable improvements in health system In terms of operational efficiency, at the beginning of Uganda’s SWAp in 2000,
performance. One of the claims of the sector-wide approach (SWAp) is that improved stakeholders perceived that there was an inappropriate concentration of resources in
coordination mechanisms should lead to a better allocation of resources within the relatively large hospitals and in centralised services, in particular project administration
sector (see chapter 2, Oliveira Cruz et al.). In particular, it is argued that the pooling of activities (Macrae et al. 1996; Matthauer 2001; Hanson et al. 2002). As a result, there
financial resources, in order to fund a coherent sector-wide plan, should lead to was a relative under-investment in basic inputs such as drugs and staffing in health
improved efficiency. This chapter explores whether or not there is any evidence to centres, and yet such inputs are central to the goal of stimulating demand for services
support this assertion in the case of the Ugandan health systems reforms since 2000. It from the population (MoFPED 2002). In this context, operational efficiency (outputs
also asks if the reforms are addressing another vital performance measure, that of per resource input), needed to improve by investing in inputs with better ‘value for
improving the equitable allocation of health care resources. money’, as and when more funding became available. For example, the large-scale but
In order to assess whether or not there have been significant changes in efficiency short-term orientation training of nursing assistants (on-the-job apprentices) was given
and equity in the Ugandan health sector, it is necessary to determine and monitor priority as an interim strategy to expand access to many critical elements of the
indicators for these performance measures. Economic and political theory suggests that minimum package (see chapter 6, Murindwa et al.). Training more comprehensive
both allocative and operational efficiency (McPake et al. 2002), and equity (McIntyre nurses and other medical cadres to fill the personnel gaps continues as a longer-term
and Gilson 2002) are important social goals for resource allocation. strategy. In addition, better management of the medicines supply system ensures greater
availability of essential drugs through pooling of previously uncoordinated sources of
funding (chapter 4, Nazerali et al.), and expansion of cheaper health centre
2. Efficiency and equity issues infrastructure, rather than building and equipping new hospitals, brings health servicescloser to the rural poor (chapter 6, Murindwa et al.). These were explicit policies set out
in the first HSSP to address operational efficiency in resource allocation (MoH 2000).
2.1 Efficiency
2.2 Equity
Health resources can be allocated across a number of different uses including disease
control programmes, reproductive health, mental health and curative care. Allocative The concept of equity in relation to the distribution of health care benefits is more
efficiency is concerned with allocating resources across these different uses in such a way complicated and controversial than that of efficiency. This is because assessment of
as to maximise health benefits. In this regard, allocative efficiency would pursue the whether resources are distributed fairly or justly is a concept based on normative value
financing of health programmes whose benefits exceed their costs or seek to judgements (Peter and Evans 2001). In addition, when setting an equity objective, there
expand/contract programmes up to the point where marginal benefits equal marginal is not always consensus as to whether the health system is trying to secure equal health
costs. On the other hand, the goal of operational efficiency is to minimise costs for any care for people with equal needs (horizontal equity) or providing more health care for
given health output using an optimal mix of inputs within a particular health people with greater needs (vertical equity), or indeed trying to achieve equity in access,
programme (Donaldson and Gerard 1993). utilisation or outcome. Monitoring performance in equity objectives is therefore
If health systems reforms are to improve allocative efficiency, resources for health complex. Diderichsen (1995) proposes that one should "look at equity both in terms of
care interventions need to be directed towards the health programmes that tackle the equal access to health care and in terms of its contribution to equality in health and the social
major contributors to a country’s burden of disease, thereby providing opportunity for consequences of illness" (p.144). The value judgements implicit in the social
maximum health benefits. In the Ugandan context, stakeholders preparing the first consequences of illness assign a high premium to context, specific equity measures.
HSSP (for the period 2000/01 to 2004/05) recognised that this meant increasing funding Therefore, an equity measure that may be agreed in Uganda may not be appropriate
for a minimum health care package that would tackle over 60 percent of the preventable elsewhere, even in a neighbouring country.
disease burden (MoH 2003a). This package focused on providing preventive and basic In producing the Ugandan HSSP, the stakeholders did not clearly define equity
curative services, with a particular emphasis on the major causes of the disease burden indicators. However, the goal of vertical equity is implied in the central objective of the
such as communicable diseases and child and maternal health problems (see Annex 1 of health sector, which seeks "to reduce morbidity and mortality from major causes of ill health
chapter 1, Yates et al.). Furthermore, in order to maximise uptake of these services, it in Uganda and the disparities therein" (MoH 2000: 14). From the perspective of vertical
was realised that they should be provided from health centres close to the population, equity, the HSSP does refer to preferential access for those with ‘high needs’, for example
1See Annex 1, Yates et al. (this book) for a description of the HSSP.
112 Health Systems Reforms in Uganda: processes and outputs 7 Resource allocation in the Ugandan health sector 113
those afflicted by the major causes of ill health such as malaria and childhood illnesses. 3.1 Donor projects
Information on access to health services in Uganda indicated poor service coverage in
rural communities, and thus improvements in equity implied the need to incorporate According to the Ministry of Finance, Planning and Economic Development
pro-rural targeting of services into health sector allocations so as to achieve (MoFPED), donor projects made the largest (45 percent) contribution towards the
disproportionately higher consumption of services by rural populations relative to urban health sector resource envelope3 in 1999/00 (see Figure 1). However, due to the large
ones in order to redress the imbalance (UBOS 2001). numbers of projects, it is extremely difficult to determine whether or not they were being
Therefore, when assessing the performance of the system in improving efficiency allocated to inputs necessary for implementation of HSSP priority activities. What is
and equity in the Ugandan policy context, we looked for the following as evidence: certain is that there were concerns at the time that uncoordinated projects were not
allocating resources efficiently (Macrae et al. 1996). In terms of operational efficiency,
Efficiency Allocative A greater proportion of resources allocated to donor projects concentrated expenditures on areas such as capital development, training
Efficiency district primary health care services relative to and capacity building (training workshops), and spent less on recurrent inputs such as
hospital services. drugs and supplies, and human resources. Large proportions of resources channelled via
Operational Relatively more expenditure on inputs that projects were spent on project administrative costs where project headquarters tended to
Efficiency represent good value for money in generating be located in large urban towns, especially in the capital city of Kampala.
additional health sector outputs (e.g. drugs and Trying to analyse the beneficiaries of project expenditure for equity purposes is also
personnel). problematic but there is no evidence to suggest that projects at the time were allocating
resources equitably. On the contrary, project managers wanting to demonstrate a
Equity The extent of access or coverage of rural 
tangible impact often focused on districts with higher absorptive capacity and these
populations with primary health care services.
tended to be the more affluent ones (Matthauer 2001). In addition, some project donors
who were concerned about the safety of their staff set restrictions on operating in
conflict areas where displaced persons face acute health care needs.
3. Resource allocations 
Figure 1: Trends in funding sources contributing to the Health Sector 
Before one can determine the impact of the reforms since 2000 on efficiency and Strategic Plan 1997/98 to 2002/03
equity in the Ugandan health sector it is necessary to obtain a picture of how the sector
was performing against these measures in the late 1990s. Given the paucity of financial
data for this period, this is not an easy task, but an analysis of the main financing
mechanisms provides some insight. 
In 1999/00, the financial year preceding the official launch of the first HSSP, the
total resource envelope available for the health sector was estimated at 310 billion
Uganda Shillings2 (MoH 2000). There are two main sources of funding to the health
sector in Uganda. The first, channelled through the government budget, includes both
Government of Uganda (GoU) domestic resources and donor budget support (general
or earmarked to the health sector). The second includes all donor project funding which
may be directed to projects handled by the central Ministry of Health (MoH), by
individual districts or by non-governmental organisations (NGOs). In terms of
allocating resources in line with efficiency and equity objectives, the first channel
provides the greatest scope as activities for donor project resources are often pre-
determined. The performance of these modalities against efficiency and equity criteria is
addressed below.
Added to this resource envelope is out-of-pocket household expenditure. Although
out-of-pocket household payment is estimated to contribute about 45 percent of total
health expenditure (UBOS 2001), a large proportion of which is spent in the private-
for-profit and private-not-for-profit sectors, there is limited scope for policy influence on
how these funds are allocated by households. Source: Derived from MoH (2003b).
2In 2003/04 prices.  Exchange rate US$ 1 = Uganda Shillings 1850. 3I.e. Government and donor resources only, excluding out-of-pocket household expenditure.
114 Health Systems Reforms in Uganda: processes and outputs 7 Resource allocation in the Ugandan health sector 115
3.2 Government of Uganda budget improved operational efficiency) is illustrated in chapter 4 by Nazerali et al., in terms of
medicines supply, and in chapter 6 by Murindwa et al., in terms of decentralised delivery
Better data are available on the breakdown of the GoU budget in 1999/00 and show of primary health services. The former shows significant improvements in the
baseline performance of the HSSP against the efficiency and equity measures. Table 1 availability of essential drugs for primary health care activities, and the latter, higher
outlines the trends in resource allocation of the GoU budget from 1999/00 to 2002/03. investment in human resources and infrastructure inputs.
Although these trends are set in a context of overall resource insufficiency for health With regard to equity concerns in allocation, some progress was made in increasing
care, an increase of 18 percent (inclusive of donor budget support) occurred in the resource allocations to rural populations, as reflected above, but more was needed to
period under review (MoH 2002; MoH 2003a). target particularly under-privileged districts. Since 2002/03, adjustments to the
The baseline allocation level in 1999/00 shows that secondary and tertiary hospital allocation criteria have tried to address this problem. For example, conditional grants to
care (mostly located in affluent urban centres) captured the largest share of the budget districts for primary level care are now allocated across districts by the MoH according
(44 percent), whereas the primary care level accounted for less than 25 percent of the to a weighted capitation formula, taking into account need factors such as: poverty,
budget. Given the high proportion of tertiary services provided by large hospitals, their morbidity, prevalence of refugees and internally displaced people, and health facility
high cost structures and their location in urban areas, the allocation in 1999/00 does not coverage. Thus, as an example, in the 2003/04 budget, the district with the highest need
score well against either efficiency or equity measures for targeting high-need client (i.e. in the conflict ridden northern Uganda) received 25 percent more primary health
groups. However, by 2002/03 about 43 percent of the health sector budget was allocated care funding per capita than the relatively privileged district capital, Kampala (MoH
to the primary care level. This reallocation, combined with an increase of 18 percent in 2003b). In the same vein, the government grant to support private-not-for-profit
real terms in the health budget, has meant that the level of financing for district health (PNFP) health facilities (mostly located in high need areas) was significantly increased
services has increased approximately 3.8 times. Evidence that this change in allocation (see chapter 5, Lochoro et al.).
has resulted in increased levels of cost-efficient inputs for district health activities (i.e. If the GoU health budget (GoU funds plus donor budget support) performs better
in terms of efficiency and equity than the project-channelled mechanism, an increase in
Table 1: Percentage allocation of GoU budget by health sector level the former vis-à-vis the latter should result in improvements in overall sector
1999/00 – 2002/03 performance as measured by sector outputs, particularly for targeted high-need groups or
rural populations. There have in fact been big changes in the composition of financing
Level of   1999/00 2000/01   2001/02    2002/03    Variance such that the GoU budget now makes up the largest contribution, discussed further
Care 1999-2003 below.
Health Sector Budget1 Table 2 indicates the type of medical attention sought by representative populations
(Billions of Uganda 79.9 114.2           170.1           196.0 116.15 in national surveys for the years 1999/00 and 2002/03. Although the survey definition
Shillings) of provider types may mis-classify some public dispensaries and overestimate the
District  PHC2 Primary 19% 33%              36% 36% 17 Table 2: Source of medical care four weeks before National Survey, 
Services PNFP3 4% 6% 8% 7% 3 1999/00 and 2002/03 (percentage)
District Secondary              8% 6% 4% 5% -3
Hospitals Level of Care Source of Care if   1999/00 2002/03
Sick Last Month Rural Urban Total Rural Urban Total
Regional Hospitals 14% 10% 11% 8% -6
Household Level None 8 4 8 7 6 7
Central Hospitals Tertiary   22% 13% 14% 12% -10
Home Self Medication        23 20 23 12 10 11
MoH Headquarter4 Sector-Wide 30% 30% 26% 28% -2 Primary Care Traditional Healers 1 1 1 1 1 1
Other MoH Agencies 2% 2% 2% 2% 0 Level Drug Shops 10 9 10 14 8 13
Total 100% 100% 100%            100% Pharmacy 0 1            0 0 4 1
Private Clinic 27 44 29 34 52 36
Source: Muhwezi (2003). Dispensary 8 2 7 7 2 6
1Includes GoU funding and donor budget support. Health Centre1 3 1 3 12 4 11
2This item also included salaries of all district services and operational funds for primary health care Secondary and Out-patient in Hospital       18 17          18 9 11            9     
(PHC).
3 Tertiary Care  In-patient Hospital 2 1 2 2 3 2Private-not-for-profit. Level
4 Includes national service delivery programmes, in particular bulk procurement, 40 percent of which is 
attributable to district services in 2002/03. Source: UBOS (2003). 
5 Increase in resources over the period is 18 percent in real terms. 1Includes government health units and private-not-for-profit units
116 Health Systems Reforms in Uganda: processes and outputs 7 Resource allocation in the Ugandan health sector 117
contribution of private clinics, it is a useful source for establishing baseline figures and
trends in service utilisation across the rural and urban settings and across alternative
providers (see also Figure 5 in chapter 1, Yates et al.). Figure 2: Sample of project funding to the health sector, 2003 
The survey indicates that the proportion of people in the rural areas utilising
curative services at the primary care level increased from 49 percent to 68 percent
between 1999/00 and 2002/03. There has been a fourfold increase in the utilisation by
rural populations of services at public and PNFP health centres, while the proportion of
those that do not seek any care when sick has declined from 31 percent to 18 percent.
This represents a promising trend in moving the delivery of the minimum package of
primary health services to lower levels of care that are closer to the groups perceived to
be most in need of health services. 
Of concern however, is the increase in the percentage of the population using the
private clinics. The percentage increase in the use of the private clinics and drug shops
in the survey period was higher for rural (poor) populations compared to their urban
counterparts. This indicates that there is still a gap in coverage and/or quality of care
provided through the public and PNFP sectors for which the private sector is a preferred
substitute. Drug shops and the majority of private clinics are poorly regulated and have
weaknesses in service standards and quality (Birungi 1998). 
4. Trends in health sector funding composition 
The rise in the health budget in recent years has partly been due to the government
allocating additional domestic revenues to health (courtesy of the HIPC4 Debt Relief
initiative), but is mostly due to the fact that many development partners have switched Source: Derived from MoH (2003b). 
their assistance from project funding to direct GoU budget support (see chapter 2, Note: "Other Inputs" refer to project overheads, technical assistance, and investment goods not 
envisaged in the Health Sector Strategic Plan
Oliveira Cruz et al.). 
With a number of key development partners (DPs) switching to channel their
support through the GoU budget in line with the principle of the SWAp, the proportion
of expenditure directly disbursed by DPs for health projects now stands at just 34 percent
of the total resource envelope, down from 45 percent in 1999/00. An analysis of this 5. Conclusions
residual project portfolio undertaken by the MoH in 2003 shows that this shift in Due to the complexity of health sector financing, it is difficult to obtain an accurate
resource flows has been potentially beneficial (MoH 2003b). However, this is bound to picture of the level and distribution of health expenditures in any country. This is
change as more funds from the Global Fund to Fight AIDS, TB and Malaria and the US particularly true in developing countries where large proportions of supposed health
President’s Emergency Plan for AIDS Relief are earmarked for the project support mode.  expenditure are not channelled through the public budget or through social insurance
Figure 2 shows the resource allocation pattern for a representative sample of five systems. In turn, the poor quality of health expenditure data, especially on out-of-pocket
health projects under the donor-disbursed, project support mode (in contrast to GoU expenditures, makes it difficult to monitor related performance indicators such as
budget support). Only 32 percent of donor project expenditure in 2003 can be attributed efficiency and equity.  
to inputs envisaged to contribute directly to the objectives of the health sector plan. The evidence presented here seems to suggest that the government budget is being
Instead, allocation to technical assistance, project administration costs, and other allocated more efficiently at present than at the beginning of the reform period in 2000.
complementary inputs were preferred areas of expenditure. Particularly noticeable in The increase in service uptake at primary care levels and health centres in particular
this analysis was the observation that projects had allocated very little (just 3 percent) seems to suggest that the shift in the allocation of resources from hospitals towards
to Ugandan health workers employed in local health facilities. primary health care services at district level has improved the overall efficiency of the
health system. Furthermore, the ring-fencing of resources for essential drugs and
personnel recruitment suggests improved operational efficiency in the sector (see
chapter 4, Nazerali et al.; chapter 6, Murindwa et al.). However, one emerging problem
4Heavily Indebted Poor Countries Initiative.
118 Health Systems Reforms in Uganda: processes and outputs 7 Resource allocation in the Ugandan health sector 119
for resource allocation is the poor uptake of services such as maternity services, where References
utilisation levels have stagnated for the last 15 years (UBOS 2001; Ssengooba et al.
2003). Birungi H. 1998. Injections and self-help: risk and trust in Ugandan health care. Social Science and Medicine,47 (10): 1455–62.
As illustrated in this book, the Ugandan health sector has been undertaking a range
of comprehensive reforms since the turn of the century. The radical approach taken was Burnham G, Pariyo G, Galiwango E, Wabwire-Mangen F. 2004. Discontinuation of cost sharing in Uganda.
facilitated by a number of factors both technical and political. Health sector reforms Bulletin of the World Health Organisation, 82 (3): 187–95.
were clearly on the international agenda, and different experiences had been Diderichsen F. 1995. Market reforms in health care and sustainability of the welfare state – lessons from
documented from which Uganda could learn. Politically, the country was recovering Sweden. Health Policy, 32 (1–3): 141–53.
from a prolonged period of civil and economic unrest, and thus the government was able Donaldson C, Gerard K. 1993. Economics of health care financing: the visible hand. Economic issues in Health
to capitalise on the positive attitude of the population to spearhead fundamental Care. MacMillan, London.
changes to the public sector. Similar experiences have been witnessed in other Frenk J. 1994. Dimensions of health system reform. Health Policy, 27 (1): 19–34.
countries. For example, Mozambique (following a long period of civil war) and Zambia
(following a change of leadership and type of democracy) have implemented fairly Hanson K, Atuyambe L, Kamwanga J, McPake B, Mungule O, Ssengooba F. 2002. Towards improving hospital
radical reforms since the 1990s, whereas changes have been much more incremental in performance in Uganda and Zambia: Reflections and Opportunities for Autonomy. Health Policy, 61 (1):73–94.
the more politically stable systems of Botswana and Tanzania (Frenk 1994; Møgedal et
al. 1995). Macrae J, Zwi A, Gilson L. 1996. A triple burden for health sector reform: 'post'- conflict rehabilitation in
Uganda. Social Science and Medicine, 42 (7): 1095–108.
The chapters in this book describe a series of policy measures and progress towards
health sector goals that have taken place within an overall setting of resource McIntyre D, Gilson L. 2002. Putting equity in health back on the social policy agenda: experiences from South
insufficiency for the health sector. In spite of the resource problems, we suggest that as Africa. Social Science and Medicine, 54 (11): 1637–56.
the reforms have progressed, there has been an improvement in both allocative and McPake B, Kumaranayake L, Normand C. 2002. Health Economics: An International Perspective. London,
operational efficiency, and also in equity. Routledge.
There is, however, still a long way to go and it would be dangerous to become Matthauer I. 2001. Institutional Pluralism and Interorganizational Relations in Local Health Care Provision
complacent. The poor uptake of maternity services needs to be urgently addressed if we in Uganda: Institutionalised Pathologies or Healing Organizations? Research Thesis, University of London.
are to see an improvement in maternity health outcomes. More research is needed to
MoFPED. 2002. Uganda Participatory Poverty Assessment Study 2 Report – Deepening the Understanding of
understand what is driving demand for services in both the public and private health Poverty. 2002. Ministry of Finance, Planning and Economic Development, Government of Uganda. 
sectors and how these services are perceived in terms of quality. It is hoped that the
anticipated increase in project expenditure by new global health initiatives will be more Møgedal S, Steel HS, Mpelumbe G. 1995. Health Sector Reform and Organisational issues at the local level:
lessons from selected African countries. Journal of International Development, 7 (3): 349–67.
transparent, aligned with the needs of the health sector plan, and channelled through
the government budget system (see chapter 2, Oliveira Cruz et al.). More research is MoH. 2000. Health Sector Strategic Plan 2000/01 to 2004/05. Ministry of Health, Government of Uganda.
required to track and assess the performance of recent resource flows from new global MoH. 2002. The Uganda Health Financing Strategy. Ministry of Health, Government of Uganda.
health initiatives in terms of efficiency, equity, coordination with overall system
development, and sustainability. Finally, there are upper limits to the levels of efficiency MoH. 2003a. Financial Year 2003/04 District Transfers for Health Services. Ministry of Health, Governmentof Uganda.
and equity gains attainable without enlarging the resource envelope for the health
sector. The current resource envelope amounts to approximately US$7-115 per capita MoH. 2003b. Annual Health Sector Performance Report 2002/2003. Ministry of Health, Government of
Uganda.
compared with the target of US$28 required to afford every Ugandan a minimum
package of health services. Muhwezi J. 2003. Policy statement to parliament, Ministry of health expenditure and budgets for 2003/2004.
Ministry of Health, Government of Uganda.
Peter F, Evans T. 2001. Ethical dimensions of health equity. In: T Evans, M Whitehead, F Diderichsen, A
Bhuiya and M Wirth (eds.) 2001. Challenging Inequities in Health: from Ethics to Action. Oxford University
Press, Oxford
Ssengooba F et al. 2003. Maternal situation analysis in Uganda. Health Policy Unit, London School of Hygiene
& Tropical Medicine.
UBOS. 2001. Uganda Demographic and Health Survey 1999/2000. Uganda Bureau of Statistics.
UBOS. 2003. Report of the Uganda National Household Survey 2002/03. Uganda Bureau of Statistics.
5GoU plus donor resources.