PAUSE….GET IT RIGHT…MOVE ON
Review of Ghana Health Sector 2005 Programme of Work
MAIN SECTOR REVIEW REPORT
FINAL DRAFT
Ministry of Health
Government of the Republic of Ghana
April 2006
Preface
This final stage in the review of the 2005 Programme of Work followed and relied upon a
number of activities carried out in the first quarter of 2006: BMC reviews and
performance hearings; agency and partner reviews; a technical review; and key
programme and area reviews on - capital investment, tuberculosis, exemptions policy,
`common management arrangements', and the burden of disease.
The following reports were not available to the main review team, or were only available
in draft or overhead presentation form:
• BMC reviews and performance hearings (provided in PowerPoint presentation
form)
• capital investment programme
• burden of disease study (preliminary findings)
• exemptions policy (debriefing and PowerPoint presentation).
Quite crucially, the expenditure statement for the last quarter of 2005 was not available.
This meant that budget execution during a year that saw a huge increase in GoG
resources available to the sector could not be analysed properly. No criticism is implied
as the production of annual accounts within three months of year end is a formidable
challenge. Nevertheless its absence limited an important part of our work.
Although the overall picture seems to be fairly clear, what is lacking in this report is some
important detail and empirical evidence to make the findings offered rock-solid. It would
be worth testing the analysis offered here and its findings by further work and when the
data are complete.
The main review was carried out between 20 March and 7 April 2006. The team
comprised Victor Aguayo, Sam Asibuo, Mercy Bannerman, Roger Hay (team leader)
Kwadwo Mensah, and Sophie Witter. The approach adopted was as follows. Documents
and reports were reviewed and issues were discussed with key informants (listed
elsewhere). Field trips were made to Eastern and Ashanti Regions where the team met
regional staff and visited health facilities. The team leader had the opportunity to attend
two excellent de-briefings on exemption policy and common management
arrangements.
The review team wishes to acknowledge gratefully the help they received from
discussions with key staff in the Ministry of Health headquarters, the Ghana Health
Service and international agencies. Without the time these colleagues gave the team
and the insights they provided, this review would not have been possible. The team is
also grateful for the logistical and other support provided by the DANIDA HSSO.
The review team accepts collective responsibility for the findings and recommendations
offered in this report. Neither its informants nor international agency colleagues share in
this responsibility. Some of its messages are very hard but they are offered in the spirit
of collaboration and confidence that Ghana will rise to the challenges it faces in
improving the quality and volume of health services to those who need them most.
CONTENTS
Executive summary
i
1. The sector's performance
1
2.
Roles and responsibilities
4
2.1 The roles of MoH and GHS 4
2.2 Statutory bodies and regulation
6
2.3 Decentralisation and local government reforms
8
2.4 The future 9
2.5 Recommendations
9
3. Financing the sector 10
3.1
Total resources
10
3.2
The changing sources of revenue for health
11
3.3
Resource allocation and expenditure
14
3.4 Launch of the National Health Insurance Scheme
17
3.5
Meeting equity goals 21
3.6 The capital budget
26
3.7
Planning, budgeting and financial management
27
3.8 Conclusions and recommendations
29
4 Service delivery performance 33
4.1 2005 strategy and targets in the context of the 5YPOW
33
4.2 2005 Performance against targets 34
4.3 Measuring and improving efficiency 45
4.4 Performance management
46
4.5 Policy implications and recommendations
47
5 Human resource policies
50
5.1
The 2004 review and Aide Memoire
50
5.2 Current situation
52
5.3 Major issues arising in HRM 55
5.4 Policy implications for HRM and recommendations
57
6 Health management & information
59
6.1 Review of plans 59
6.2 Challenges 60
6.3 Improvements in information for management
60
6.4 Recommendations
61
7 Procurement 62
7.1
Progress toward common management & procurement harmonization
62
7.2
Capacity Development Programme (CDP)
63
7.3 Challenges 63
7.4
Evaluation of Procurement Practices at BMCs
64
7.5 Recommendations
64
8 Common management arrangements 65
8.1
History
65
8.2
A new deal
65
8.3
Recommendations
66
Annexes
1.
Terms of reference
6
7
2.
Key informants
71
3.
Review of laws of regulatory bodies
75
4. Performance indicators for PoW III
77
5.
References
79
Review of Ghana Health Sector 2005 Programme of Work
PAUSE….GET IT RIGHT…MOVE ON
Executive summary
i. This review was carried out shortly after wage negotiations resulted in a
sharp increase in the government wage bill, such that a much enhanced
government health budget has been pressed to the limit. Coincidentally some
health partners have begun to move their contributions from the health fund to
the government budget. At the same time, labour productivity responses to
the sector's increased resources have been uneven at best and a number of
problems accumulated over the years remain unresolved. On the other hand,
there have been important technical and organisational developments which
need to be consolidated.
ii.
Hence the title of the report. Its main message is that, at this moment of
crisis, a pause for some cool, strategic reflection on how the gains can be
consolidated and the risks managed will have a large future pay-off. Some
problems need to be fixed before further progress can be expected. These
primarily relate to two issues: the budget and the way it is managed; and the
workforce and the way it is motivated. The solutions are not easy and
immediate. They will require tough sustained political leadership and all the
help the international community can give.
iii.
This synopsis does not attempt to summarise all of the review findings.
Instead it sets out the immediate and key issues the Review Team
recommends that the Ministry of Health and its partners consider in leading
the sector forward.
Pause…
iv. There have been many changes in the sector and more are to come. It is
important that managers are not wearied and confused by too many changes.
There needs to be a period of consolidation, increased clarity and increased
focus on…
…getting it right
v. There are a number of sector-wide problems to be resolved.
Improving budget execution
• Despite a large increase in the health budget, it is under threat from two
sources. Another increase in the wage bill could not be financed without a
further MoFEP allocation to the sector. An already over-pressed recurrent
budget is being threatened by exuberant capital commitments without
sufficient care for their recurrent cost implications and without planned and
obligatory debt servicing and repayment schedules. Financing `the gap' in
the recurrent budget from NHIF assets risks its future financial stability.
• Cash flows need to be more predictable. Service managers need to know
what their budgets are, what they can be used for and when to expect
Executive summary
i
Review of Ghana Health Sector 2005 Programme of Work
funds. Both the GoG service budget and the Health Fund appear to heave
have performed miserably in this respect, affecting exemptions seriously
and therefore provider incentives to look after poor people. `Getting this
right' is the most important single step towards improving service delivery
efficiencies. It also provides the best way into improving the budget
structure and allocations between priorities. The Review Team
recommends that no further moves are made towards MDBS until there is
evidence for a predictable flow of funds from MoFEP and MoH to service
levels. The results of a Public Expenditure Tracking Study, now in the
field, should provide valuable information about where variances between
budget and expenditure are arising and where delays are occurring.
Improving labour productivity
• The sector's increase in funds needs to be turned into a higher volume of
services of better quality.
• The sector now has a four-tier management system operating alongside a
growing National Health Insurance Fund with its own overhead
requirements. There is increasing functional duplication within the MoH-
GHS system and unacceptably high overheads. As a first step, the
duplications need to be stripped out. The Review Team recommends that
the Minister's Task Force resumes its work and is given expert,
independent support.
• Important recent analytical work suggests that there is great performance
variation between districts and hospitals. The best are performing
superbly. Analyses of this kind should become the heart of a simplified
information system that supports performance management. It is already
providing the basis for manager peer reviews. For the first time, this will
allow managers to be held accountable for results. It will also allow the
GHS focus on supporting failing districts and hospitals. Managers who are
unable to respond to this support will need to be changed.
… and moving on
vi. There is an exciting agenda of issues to be considered in relation to the
next PoW …
• Clarifying the relative roles of the NHIS and the GHS
• Ending the differences between government and non-government
provider arrangements
• Attending to urban health policies
• Decentralised staff budget management
• Reviewing provider payment regimes
• Modernising primary care
• Developing alternative service production models
…but not before the sector's performance is improved under current
management arrangements.
Executive summary
ii
Review of Ghana Health Sector 2005 Programme of Work
PAUSE….GET IT RIGHT…MOVE ON
1. The sector's performance
1. The table below summarises the performance indices agreed in the
Programme of Work (PoW) to track changes in health status, service
volumes, quality and inputs. As the years have passed this record has
become increasingly valuable. Despite some important gaps in the data and
technical problems with the indices, the record provides a broad view of how
the sector has performed since 1997. The data for 2002 to 2005 indicates
performance trends for the first four years of the current five year PoW,
together with the extent to which targets set for the PoW have been achieved.
The data for 1997 to 2001 provides the context against which this
performance can be judged.
2. Although the MoH has taken pains to correct previous denominator problems,
these data still need to be interpreted with some care and some apparent
trends (or lack of them) may be spurious. For example, the extension of
health services may result in an initial deterioration of performance indices as
more cases are identified or brought into curative care. In other cases,
missing data poses problems. For example it is not possible to comment from
these data on whether IMR and CMR, two crucial measures of child health,
and ones most responsive to better health services, have deteriorated or
improved. On the face of it there was deterioration until the last measurement
in 2003.
3. With these caveat in mind, there appear to have been some gains:
i. the under-five malaria mortality rate appears to have declined
ii. tuberculosis cure rates appear to have improved
iii. the proportion of supervised deliveries appears to have increased
iv. EPI coverage has increased after a decline in the early 2000's and
has now (only just) exceeded 2000 levels
v. perhaps most significantly, the number of recorded Guinea Worm
cases identified has fallen below 4,000 for the first time: if this is a
real improvement rather than a recording chimera, this is a real
achievement
vi. tracer drug availability has improved and this is borne out by the
anecdotal evidence given to the team of a decline in
pharmaceutical `stock-outs'
vii. the most striking feature of the table is the increase in the GoG
budget allocation to health, a rise of some 400% in real terms
compared with 2001, representing 14.9% of the recurrent
government budget in 2005 and well above levels that have
proven sustainable in other countries; its causes and
consequences provide one of the themes for this report.
4. Despite these achievements, the overall picture on the basis of national
averages is one of stagnation in health outcomes and service delivery
volumes. Nothing can be said about quality as its poses measurement
difficulties and the indices chosen require review. Although there has been a
Sector performance
Review of Ghana Health Sector 2005 Programme of Work
Sector performance
slight rise in OPD visits per capita, the change is probably within
measurement error. Similarly, although hospital admission rates are better
than in the previous PoW, they have not improved significantly since 2001.
Bed occupancy rates (BOR) are low, appear to have declined and may be
lower than in the previous PoW period. BOR is an incomplete measure of
hospital efficiency. It is not possible to be sure about efficiency trends without
either turnover rates or average lengths of stay (ALOS) which should be
measured in the future. However, unless patients are staying in hospital for
shorter periods of time, the evidence suggests that hospital efficiency has
declined. In any event, declining BOR's in the face of large hospital
investments suggests that the infrastructure is not being used efficiently and
that there may be over-capacity in the hospital sector.
5. The ratios of doctors to nurses to population have also declined reflecting the
high reputation of Ghana's clinicians abroad and the difficulties in retaining
them in Ghana. These issues will be difficult to resolve in the short run and
are taken up later in the report. The national averages hide important
differences in the distribution of clinical staff by district and regions.
6. Of most concern is the lack of evidence of increased activity or service quality
in response the very large increase in the financial resources made available
to the sector. In aggregate, labour productivity has declined sharply.
However, a number of factors need to be kept in mind before making hasty
judgements. Almost all of the GoG budgetary increase occurred in 2005. Any
activity response is likely to come in 2006 and beyond. If services are being
extended, some indices may deteriorate before improving.
7. Finally, and most importantly, there is strong evidence, presented later in the
report, that the national aggregates presented in this table hide great
variations between regions, between districts and between rural and urban
areas. Regional and district data were not available to the team and it is
recommended that regional performance indices are compiled for in time for
future reviews. There is evidence from other sources that some districts,
including in the northern regions are performing exceedingly well and
inducing major health improvements. However, in crude terms, for every
district performing above the national average, there is one performing below
it. The quality of management seems to be the critical factor.
8. The sector's new resources imply a huge management challenge if they are
to be converted into more and better health services for those that need them
most. One of the main conclusions of this review is that the GHS
management needs to focus more attention and support on fewer priorities in
districts where results are less than satisfactory.
Review
Sect
of Ghana Health Sector 2005 Programme of Work
or performance
3
Review of Ghana Health Sector 2005 Programme of Work
2. Roles and responsibilities
1. Ghana has a constitutional duty to care for the health of its citizens, particularly
the poorest. This duty is now enshrined in various international agreements and
is increasingly monitored by international organisations with which GoG has
financial and other relations. This duty implies obtaining the best possible value
for money in terms of health outcomes by providing the best possible quality
services to those who will benefit most from them.
2. This primarily implies efficient and effective service organisation and
management, including prioritising health priorities and focussing management
attention on services that will yield the most health gains. It also implies
allocating resources and supplying services to poor people who cannot afford to
pay for them. There are two reasons for this: it is the most efficient use of
resources as the greatest health gains are available at the least cost. It is also
equitable in terms of Ghana's notions of social justice.
3. Although these duties are well articulated and understood, the ways in which the
consequent responsibilities are shared and exercised requires further clarification
and refinement. It is appropriate that at the beginning of the last year of the
current PoW, time is taken to `pause' and reflect, and to `get things right' before
`moving on' into the next PoW. The experience gained and lessons learned in
the last few years provide a solid foundation on which to base improvements.
2.1 The roles of MoH and GHS
4. Ghana has chosen an `agency' model as the management framework for its
health sector by which, in line with `New Public Management' principles, the
central agency (MoH headquarters) delegates responsibility for the delivery of
services of an appropriate volume and quality to `executive agencies'. At the time
it was introduced this was an innovation outside the major OECD reforming
countries. It is not surprising that it took some time for things to settle down and,
in particular, for ministers to learn to trust `service level agreements,' rather than
administrative dictat, as the main instrument for strategic direction, performance
assessment and control. It is therefore gratifying to find that service level
agreements are now being put in place although not universally (these need to
be made obligatory) and that performance management is beginning to take
hold.
5. It is also gratifying to note that officials of both MOH and its agencies are very
clear on their respective roles with regard to national and sub-national level
activities. Interview responses from health officials indicate that they are clear
that MOH is responsible for overall policy formulation at the national level, whilst
its agencies are in charge of the implementation of the policies in their area of
service coverage.
6. The sector has been subject to numerous reforms in recent years wearying
managers of changes and resulting in some confusion. More are inevitably on
the way with the growth in the role of the National Health Insurance Fund as a
service purchaser. The Review Team recommends that further major reforms
Roles & responsibilities
4
Review of Ghana Health Sector 2005 Programme of Work
within the MoH-GHS system should be delayed as far as possible to allow
consolidation and improvements in the present arrangements.
7. Nevertheless, collaboration has not been effective among all key players within
the health sector and between the MOH and its agencies. Officers interviewed at
both the national and regional levels spoke of the perennial challenges arising
from subjective interpretations of the roles of MOH and its agencies, duplications
and a poor management of the transition process. The report of the Ministerial
Task Team on Institutional Reform of the Health Sector and the latest draft report
on Review of the CMAs for the implementation of the Health sector five year
POW (2002-2006) are also emphatic on the issue of poor coordination.
8. In addition, the continuing decentralisation of Ghana's health services has
resulted in a four tier management system: MoH; and GHS national, regional and
district. At the same time, the NHIF is inevitably incurring its own management
costs. This inevitably leads to upwards pressures on management overheads
which can only be justified by service delivery productivity or allocative efficiency
gains. There is no evidence that these gains are being realised in Ghana; rather
the contrary. It is probable that in the next round of reforms, but not now,
consideration should be given to removing one of the GHS management tiers.
9. More immediately, there appears to be a growing duplication of activities in MoH
HQ and the GHS. This is making its presence felt by sharply rising central
overhead costs. Much of this appears to be accruing to the MoH headquarters.
These increased overheads have been partly driven by recruitment into the MoH
and partly by the granting of ADHAs to non-clinical staff. Further analysis is
required to be sure, but it would appear that management overhead costs at all
levels, as measured by the share of PE budget absorbed by managers and
administrators compared with clinicians maybe reaching nearly 50%, much
above the 20% expected in the PoW and the 15% that would be desirable. It is
difficult to retreat from this position but it is inefficient and needs to be corrected.
10. It is therefore important that the Ministerial Task Team on Institutional Reform of
the Health Sector continues its work, focussing first on `getting things right'. Its
immediate and most important tasks should be identifying and stripping out
duplicative functions and simplifying the performance management system.
Given the complex relationships involved, the Task Team's work would be aided
by an independent expert analyst and facilitator.
11. The Task Team might consider a number of points in this regard:
• MoH headquarters key roles are
o policy development
o strategy management (not operational management)
o budget management
o the management of service level agreements; supported by
o information management to observe the performance of its
agencies
• All other functions can be contracted out, including regulation
• These key functions could be performed by a small number of highly
qualified, highly motivated staff
Roles & responsibilities
5
Review of Ghana Health Sector 2005 Programme of Work
12. The Ghana Health Service is responsible for delivering the right services to the
right people of the best possible quality. It does this by engaging regional and
district agencies in service level agreements as the heart of its performance
management system. It is also a major employer of staff although, for reasons
explained later, it is not clear that this can continue for much longer.
13. There are unavoidable overlaps in the functions of the MoH headquarters and
the Ghana Health Service; indeed also between GHS national and GHS regional
and district. For example, MoH headquarters may need to analyse and advise
the Minister on HR policies that provide a framework for HR management in the
GHS. The GHS needs to assess how these policies will affect staff costs and
productivity and be prepared to debate unresolved issues. However the functions
are separable and there should not be duplication. There would be important
efficiency gains if the Minister's Task Team could work with both MoH and GHS
managers to ensure that duplications are eradicated.
14. All new systems are invariably overly complex. The experience gained to date
should allow the performance management system to be simplified and refined
so that it becomes an effective incentive for improved performance rather than an
onerous administrative task. The central MoH should set a few (no more than
three) broad strategic goals in any one year. The national GHS should then be
advised by district and regional authorities on no more than three operational
priorities at the district level, reflecting local health needs, within these strategic
guidelines. This should then be the basis for a series of service level agreements
which state what feasible progress can be made. The Ministerial Task Team on
Institutional Reform might include a review of the effectiveness of service level
agreements and the performance management system in its terms of reference.
2.2 Statutory bodies and regulation
15. As the providers of healthcare in Ghana and its funding become more diverse,
the regulation of service price, volume and quality is becoming both more
important and more challenging. The most cost-effective way of regulating
healthcare is a matter of substantial international debate. It is given added
importance by the rapid growth of However, a consensus is beginning to develop
and new literature is emerging.1 Some of the key points characterising this
consensus are as follows:
1. the benefits of regulation need to be greater than its costs for
society to gain
2. a legal framework is a necessary but insufficient basis for
regulation
3. administrative and bureaucratic controls tend to be high in
enforcement costs and relatively ineffective
4. market-based incentives and sanctions appear to be both
cheaper and more effective
5. self-regulation by professional bodies is helpful but relatively
ineffective as, until reputation is at stake, professional bodies act
as clubs to protect the interests of their members
1
This literature has been reviewed recently by Tim Ensor and colleagues. The review is available on the
Oxford Policy Management website at www. opml. co.uk.
Roles & responsibilities
6
Review of Ghana Health Sector 2005 Programme of Work
6. government and non-government providers should be governed
by the same regulatory arrangements.
16. Ghana still relies too much on a complex set of laws governing a range of issues
from staff competencies and facility registration to drug safety. These are under
review with the objective of simplifying and updating them. Progress to date is
summarised a table in Annex 3. However, legislation cannot govern the really
important attributes of health services: their quality, volume and price. This
requires a different set of incentives to encourage good practice and sanctions to
deter bad practice. However, what the law can do is to define and protect the
rights of both healthcare providers and consumers. This is important.
17. The proliferation of Statutory Bodies similarly reflects Ghana's undue reliance on
administrative regulatory instruments and self-regulation by professional bodies.
During the review, it was observed that functions of most of the regulatory bodies
overlapped. Many receive subventions from the health budget. This can only be
justified to the extent that they provide public good-type services. Finally, there is
protection asymmetry in that providers' interests are protected more than
consumers'.
18. On the other hand, the review team notes that an independent regulator is to be
established for the NHIF. It is to be hoped that its scope and powers will take into
account the desirable attributes of regulation summarised in the next paragraph.
19. The team also notes two important practical steps that have or could be taken:
1. A consultative forum for Regulatory Bodies to discuss important
regulatory problems has just been created should be sustained.
2. Annual Reports from Regulatory Bodies should be
acknowledged and feedback given.
20. Although the review team is not recommending further immediate reforms, it
does recommend that the Minister establishes an expert task force to review the
application of regulatory arrangements, particularly in the light of new sources of
funding and provision. Its terms of reference might include the following:
1. review of roles and functions
2. consolidation of statutory bodies where appropriate
3. consider ways of extending the influence, if not the presence, of
the regulator to the regions and districts
4. review of funding arrangements to ensure that only public good-
type services are funded from the health budget
5. consideration of the creation of an independent health
commission, possibly based on the NHIF regulator, to
1. promulgate information about where the most cost
effective services are available;
2. mediate disputes between providers and consumers
3. provide support to both providers and consumers in
seeking legal redress where disputes cannot be resolved
and there is evidence for harm.
Roles & responsibilities
7
Review of Ghana Health Sector 2005 Programme of Work
2.3 Decentralisation and local government reforms
21. Decentralisation has been and still is one of the strategic policies of successive
Ghanaian Governments including the current one. Ghana's vision for
decentralisation can be summarized as promoting responsive and accountable
governance at the local levels that allows effective participation, equity in
resource allocation and the effective delivery of services especially for the poor.
22. The MOH has made substantial progress in decentralising management
responsibilities to BMCs:
a. Service planning and non-staff budgeting has been decentralised to
BMCs. The annual planning process is bottom-up. Inputs from regional
and district levels are aggregated at regional and national levels to form
the basis of the annual budget. However, local priorities are still not
reflected in regional and district service level agreements; budget ceilings
are not set ahead of time and there is great uncertainty about what GoG
funds can be expected (see section 3). Another major weakness is that
District Assemblies and other stakeholders are not yet involved in
deciding health priorities and drawing up plans. In particular, District
Health Planning is not integrated into District Assembly plans and
budgets.
b. The management of non-staff budgets is also decentralised to the district
level but not necessarily felt at the sub-district level. The sub-district
managers complained to the Review Team that they are not involved in
the implementation of the sub-district budget on grounds of inadequate
financial management capacity. They argue that while they have
managed their IGFs well without any auditors' query, they do not
understand why the districts continue to manage their GOG and Donor
Pooled Funds (DPF).Sub-district managers should be involved in the
implementation of their budget.
c. A fundamental objective of the current POW was to decentralise
procurement to MOH agencies and providers. This was to be achieved
through development of procurement capacity at all levels. That capacity
has been developed and all BMCs plan and execute procurement
according to agreed thresholds and methods. However, BMCs are
advised that drugs and other items should be purchased from the
central/regional medical stores. This should allow economies of scale in
international procurement to bring prices below those offered by private
wholesalers. Although there is recent evidence of a fall in Central Medical
Store prices, some claim that they are still not competitive.
d. The decentralisation of human resource management is also proceeding
but somewhat slowly. The review team was told at Kwabre District
(Ashanti Region) for instance that the DDHS is empowered to post health
personnel in the district to any part of that district especially to ensure
equity in the distribution of personnel. This amounts to about 80%
decentralisation in their estimation.
e. Some in-service training is also decentralized to the district level, at least
in the Ashanti Region.
Roles & responsibilities
8
Review of Ghana Health Sector 2005 Programme of Work
2.4 The future
23. The next `natural' steps in the decentralisation sequence are follows:
1. giving authority to regions/districts to manage the staff budget
(item 1) and to hire and fire staff;
2. funding regions, then, districts with block grants that take into
account population, health needs and poverty
3. giving authority to regions (maybe some districts) to procure
large ticket capital items unless there are economies of scale in
project management at the national level that need to be
captured.
24. The review team is not recommending further radical reforms immediately.
Moreover, the next steps are large ones and require careful preparation. They
should probably only be implemented after the MoH and senior GHS
management is confident that there are accredited HR management skills in
place, that financial resource flows are predictable and that there are reliable
ways of monitoring performance and of intervening when required.
25. However, until staff budget management is decentralised, it will remain difficult to
allocate staff according to health needs. This is because, at the moment, budgets
largely follow staff, rather than following health needs. If regional managers were
given PE budgets to manage and allowed to recruit into approved posts, the
pattern of financial resource allocation would begin to reflect health needs. Block
funding would further increase the incentives managers had to hire the staff they
need to give them the best results at least cost but this step is some way off.
26. There are also two sets of issues that need to be resolved. The first is the
growing role of the NHIF as the purchaser of clinical services in relation to BMCs
who may become the purchasers of promotive and preventive services. The
second is the growing role of District Assemblies and whether or not their role as
service purchasers will expand. The review team understands that these issues
are being considered and recommends that they receive intensified attention.
2.5 Recommendations
1. It is important that the Ministerial Task Team on Institutional Reform of the Health
Sector continues its work. Its most immediate and important tasks should be
identifying and stripping out duplicative management functions between MoH and
the GHS and proposing strengthening and simplifications for the performance
management system. It is recommended that it is assisted by an independent,
organisational expert and facilitator.
2. It is recommended that the Minister establishes an expert task force to review the
application of regulatory arrangements in order to simply them and make them
more effective in the face of increasing diversity of provision and financing.
3. It is recommended that the next steps in decentralisation be planned and the
future roles of BMCs be clarified taking into account the growing role of the NHIF
as a service purchaser and the possibly increasing roles of District Assemblies
as fund managers.
Roles & responsibilities
9
Review of Ghana Health Sector 2005 Programme of Work
3 Financing the sector
3.1
Total resources
1. There has been a substantial increase of nearly 400% in the financial resources
available for health care since the beginning of the current PoW (table 3.1) This
is attributable to a number of factors. The first is the generally positive
macroeconomic climate, including real GDP growth of 5-6% per annum over
2003-5; inflation rates declining, although they are still around 15% for 2005; and
tax revenue increasing from 17% of GDP in 2001 to 22% in 2005 (Government of
Ghana 2005). The second important factor is the proportion of government
expenditure which has been allocated to health, which was increased
dramatically in 2005, such that health now has the third largest share of the
government budget for 2006 (after education and economic affairs). This shift
was in line with regional and international commitments. However, there is
indirect evidence that an important driver was the implacable demands of a
unionised health workforce for better remuneration.
Table 3.1 Sector-wide indicators for health financing
Finance
Indicators
2001 2002 2003 2004 2005
2006
budget
% GOG budget on health
8.7
9.3
9.1
8..2
15
17
% GOG recurrent spending on health
10.2 11.5 11.2 11.9 14.5 15
% GOG recurrent on health on non-
salary items (2+3)
8.1 5.9 6.9 5.4 6.6
(budget)
NA
% spending on districts and below,
items 2+3
NA
40.9
35.4
37.9
48
43
% Earmarked / total DP
62.3
32.8 39.5 26.3 40 40
% IGF from pre-payment schemes
3
NA NA NA NA 20
% Recurrent funds from GOG+HF
allocated to CSOs
1.2 NA NA NA 3.1 2
% Recurrent funds. on exemptions
3.6 NA NA NA 8 8
Per capita expenditure on health (USD) 6.3
8.1 10.5 13.5 23.9 25.2
Source: (External review team 2005;Ministry of Health 2006)
2. The GoG budgets for 2006 include projections for donor funding and internally
generated revenue (IGF, or user fees). As table 3.2 shows, health does relatively
well from all sources: its share of GoG spending is second only to education
(19% of the total), and its share of HIPC spending is also relatively high (15%),
while it is able to generate or attract 14% of the total IGF and donor resources.
3. Overall the health budget has grown beyond the expectations of the 'high-case
scenario' in the Five Year Programme of Work (Ministry of Health 2002) -
exceeding that estimate of ¢ 1,002,460 million by more than fourfold and going
well beyond the Abuja Declaration target of 15% of government expenditure
devoted to health.
Financing the sector
10
Review of Ghana Health Sector 2005 Programme of Work
Table 3.2 Shares of total GoG budget for 2006 (¢ million)
GoG
Share
(GoG)
IGF +
donors
Share
(IGF/donors)
HIPC
Share
(HIPC)
Share
(total)
Public
services 1,460,546.7 8% 401,943.7 5% 179,700 13% 7%
Defence 879,079.5 5% 0.0 0% 0 0% 3%
Public order 1,043,100.6 6% 75,420.8 1% 25000 2% 4%
Economic
Affairs
2,422,444.0 14% 4,928,729.8 57% 405,000 30% 28%
Environment 66,392.4 0% 36,026.1 0% 0 0% 0%
Housing 1,478,825.8 8% 1,305,065.9 15% 197,000 14% 11%
Health
3,465,062.3 19% 1,202,979.3 14%
200,000 15% 17%
Recreation 151,194.4 1% 59,559.1 1% 25,000 2% 1%
Education 6,812,333.9 38% 659,570.0 8% 300,000 22% 28%
Social
protection
112,191.3 1% 5,706.7 0% 35,000 3% 1%
Total
17,891,170.9 100% 8,675,001.4 100% 1,366,700 100% 100%
Source: (Government of Ghana 2006)
3.2
The changing sources of revenue for health
4. The sources of revenue for the health sector are changing and are likely to
continue to change as the National Health Insurance Scheme picks up
momentum and health partners shift their funds to budget support. The trends
shown in table 3.3 should be interpreted with care as the entries for 2005 are not
yet complete (they are based on data up to the end of the third quarter, with
some categories missing), while the data for 2006 represent budgets and may
not reflect either actual budget allocations or expenditure. The main change
foreseen in 2005-6 is contribution of the new National Health Insurance Scheme,
which has brought in some additional funds (from a 2.5% levy on VAT), but which
is assumed to have also displaced other sources, especially IGF. The extent to
which the projected 23% of budget to come from the NHIS in 2006 is realistic and
whether the NHIS will augment total revenue is discussed below.
Table 3.3 Health sector revenue shares by source: 2000-2006
Year GOG Financial
Credits
NHIS
IGF
Health
Fund
Earmarked
Funds
HIPC Total
2000 55% 5% 12% 8% 20% 0% 100%
2001 49% 2% 14% 13% 22% 0% 100%
2002 50% 11% 14% 18% 8% 0% 100%
2003 49% 10% 14% 15% 10% 3% 100%
2004 43% 11% 13% 21% 6% 5% 100%
20052
56% 15% 24% TBC
20063
45% 0% 23% 3% 13% 13% 3% 100%
Sources: Ministry of Health financial reports for 2000, 2001, 2002, 2003, 2004; financial
report up to the quarter ending September 30, 2005; budget for 2006
5. The other main changes affecting resource inflows are the shift to multi-donor
budget support (MDBS) by some of the main donors to the health sector and the
2
Up to the end of the third quarter: full year statistics were not available at the time of the report
3
Budgeted, not actual
Financing the sector
11
Review of Ghana Health Sector 2005 Programme of Work
relative growth of earmarked funding. In 2005, the EU shifted its support to
MDBS. The World Bank, DFID and the RNE are currently debating the modalities
of shifting their support for 2007. These are the largest contributors to the health
fund. According to the September 2005 financial statement, the World Bank
provided 45.7% of all donor assistance for the year; DFID 14.5%; the Royal
Netherlands Embassy 13.2% and DANIDA 10.6%. For 2006, the main funders of
the HF are the RNE (18.7% of total aid to the sector); DFID (13%); and DANIDA
(10%) (POW 2006). The Danish government has set an upper limit of 25% of its
aid which can be channelled through budget support, but it faces the prospect of
being the only donor to the Health Fund in 2007.
6. The shift to MDBS, which is consistent with the Paris Declaration on Aid
Effectiveness of 2005 and the Ghana Partnership Strategy (GoG & Health
partners 2005), poses a potential threat to the health sector budget, which must
be managed carefully. There are three main areas of concern: first, whether GoG
resources to health will fall if partners shift support to MDBS; second, whether
the MoFEP disbursement procedures will be more cumbersome or less flexible
and predictable than the current Health Fund arrangements; and finally, whether
there will be a loss of sectoral dialogue with donors, which could feed into lost
profile within MDBS and potential reductions from current health partners.
7. All three of these are serious, but manageable concerns. Ring-fencing or
sequestration of donor funds within MDBS is acceptable to the MoFEP and has
precedents in other sectors, though the issue of fungibility remains. More
important is the necessary precondition of improving coordination and dialogue
between MoFEP, MoH and donors by establishing a forum where all health
partners, whatever their disbursement mechanism, can be involved in health
policy development. In particular, MoH needs to develop proactive resource
mobilisation strategies with respect to MoFEP and needs the weight of its health
partner advocacy for these to be effective. Most important of all is the
requirement that BMCs receive predictable disbursements from MoFEP. There is
considerable evidence to suggest that BMCs, at least until recently, have relied
on the predictability of health fund disbursements more than of MoFEP
disbursements.
8. The current policy matrix contains two health-specific objectives with explicit
targets: increased coverage of supervised deliveries (57% for 2006; 60% for
2007; 63% for 2008); and increasing coverage by the NHIS (35% of the
population in 2006; 45% in 2007; 60% in 2008; 65% in 2009 (Ministry of Finance
2006) but their role and content is still under debate.
9. The shift to MDBS and the growth in earmarked funding is likely to change the
relationship between the MoH and donors of earmarked funds (of which USAID
is the largest in 2006, providing 15.8% of total aid to the sector, followed by the
Global Fund with 12.8%). Whereas previously, the main dialogue was with
partners in the SWAP, the MoH will need to focus more efforts on linking
earmarked funds to its priority areas. At present, earmarked funding focuses
more on investment and less on recurrent service costs, compared with HF
support (see table 4), but this balance should be revisited. Evidence from districts
suggests that some vertical programmes generate additional overhead costs for
monitoring and management, for which they do not budget, and that they
Financing the sector
12
Review of Ghana Health Sector 2005 Programme of Work
therefore drain, rather than augment, the meagre flexible resources in the
system.
10. HIPC funding for the health sector was withdrawn in 2005, because of
accumulating positive balances in the NHIF account from public sector
deductions. This was based on over-optimism about the speed with which the
national health insurance system could be developed to provide protection to
vulnerable groups (which was one of the uses for the HIPC funds). It also
reflected a lack of clarity about the two roles of the NHIS: the insurance against
the costs of health risks and an instrument for funding exempted services and
persons. In recognition of this, presumably, the budget for exemptions from HIPC
has been reinstated for 2006.
11. The proportion of IGF in relation to overall funds fluctuates but has not changed
dramatically since the beginning of this Five Year PoW, when it was 15%
(External review team 2002). IGF continues to provide a significant proportion of
flexible funding4. The proportion of 'flexible' (i.e. non-drug) IGF to total IGF
ranges from 30% at sub-district level to 77% for the teaching hospitals (MoH
2005).
12. There are a number of areas of inflows and expenditure on health care that are
currently un- or under-reported. These include household payments other than
officially reported IGF and in particular private sector expenditure. In addition,
there is unreported aid (particularly from NGOs). A national health accounts
exercise is now underway which should give a broader picture of health financing
flows.
Is there a funding gap?
13. There is concern that the health budget faces a funding gap and the review team
was asked to clarify this issue. Discussions of funding gaps are confusing, as the
term is used in a different ways. In 2004, a needs-based system of budgeting
was introduced, in order to produce an annual estimate of the resource needs of
the sector. Invariably, there is gap between the resource requirements and
resources available. That is the first type of gap. The second is the gap between
budget and the actual disbursements which arrive. However, the term gap is also
used to describe the funds needed to cover the debts carried forward from the
last FY, when commitments such as DAIA have not been met5. In addition, the
PPME sometimes adds in the funds which they know will be needed to fund pay
increases during the year (but which could not be included in the budget under
MOFEP rules)6. The 'gap' for the 2005 PoW was stated to be USD 283.49 million
(Ministry of Health 2005), while for 2006, it is stated to be USD138.48 million
(Ministry of Health 2006).
4
For Brong Ahafo in 2005, for example, IGF provided half of the non-wage recurrent funding,
according to the annual regional report. At facility level the proportion is often higher. However,
these figures include payments for drugs, which do not provide flexible funding for other activities.
5
For example, the MoH starts this year owing: c 50 billion for exemptions; c 42 billion for DAIA; c
50 billion for capital investment; c 75 billion for ADHA. This amounts to about USD 24 million in
total.
6
Estimated at c 600 billion for 2006 (about USD 66 million).
Financing the sector
13
Review of Ghana Health Sector 2005 Programme of Work
Will the health budget continue to grow?
14. The MTEF forecasts a fall of some 2% in the health budget for 2007 and 2008 in
nominal terms and a greater reduction in real terms, depending on the inflation
rates, as a result of decreases in both donor and GoG contributions (Government
of Ghana 2006). It is also probably the case that the health budget share, already
high by international standards, may not be sustainable in the face of pressures
from other sectors. Budget compression and its consequent pressure on
expenditure makes the subjects discussed next - the allocation of resources, the
efficiency of their use and the development of the NHIS - particularly important.
3.3
Resource allocation and expenditure
15. Table 3.4 shows the allocation of resources by source and line item. As
expenditure data for 2005 are not yet available, we have compared the budget
for 2006 with the expenditure for 2003 to give an indication of trends. The capital
investment has grown (to 19%), as have services (to 36%). This in part reflects
the growth of earmarked funding, as well as a high expectation from the NHIS7.
Actual expenditure for 2006 may vary from these budget estimates. PE and
capital expenditure tend to overspend their budgets, resulting in crowding out
items 2 and 3 (see below).
Table 3. 4
Allocation by line item and source: 2006 budget (¢ Mn unless %)
GOG
Health
Fund
IGF
Ear-
marked
funds
NHIS HIPC
Total
% of
Total
(2006
budget)
% of Total
expenditure
(2003)
PE
1,702,890
10,000
1,712,890 40% 45%
Admin
166,500
60,000
226,500 5% 9%
Services
409,730
243,940 128,710 726,400 50,000 1,558,780 36% 29%
Capital
185,800
421,000 100,000 100,000
806,800
19% 17%
Total
1,920,000 545,000 143,000 549,710 996,000 150,000 4,303,710 100% 100%
% by
Source
45 13 3 13 23 3 100
Personal emoluments
16. The trend of growing expenditure on Item 1 (PE)8 has continued and is unlikely to
be reversed in the near future. The wage bill currently accounts for 97% of GoG
recurrent health expenditure and 67% of combined government and donor health
7
It is not entirely clear why the NHIS is expected to contribute to categories such as PE, admin
and capital expenditure. The purpose of an insurance fund is to pay for services on behalf of its
members and seems erroneous to treat it as part of the MoH budget.
8
In addition to the salaries and allowances which are accounted for under item 1, there are
numerous personnel costs attributed to items 2 and 3, such as out of station allowances, per
diems, consultancy fees, and travel allowances, some of which are effectively increasing staff
income. Item 1 therefore understates the total personnel costs.
Financing the sector
14
Review of Ghana Health Sector 2005 Programme of Work
expenditure (World Bank, MoH, & GHS 2006). This is in large part due to
ADHAs, which have grown from ¢ 17 billion when they were first introduced in
1999 to ¢ 720 billion in 2005, and projected to rise to ¢ 936 billion by the end of
2006 (Ministry of Health, Ghana Health Services, & Cedar Care Trust
International 2005). This represents an average annual real growth rate of 178%.
17. There is general consensus that ADHAs have not been an effective driver of
improved performance, largely because they have become a fixed addition to
income rather than being performance-related. They have now also been
extended to the whole range of staff, rather than simply rewarding those grades
which are either in short supply or are genuinely working longer hours.
18. Health worker pay and allowances in Ghana are generous relative to other
countries, amounting to 8 times GDP per capita (Policy and Health Systems
Programme 2005). However, more importantly, Ghana's reputation for producing
high quality doctors and nurses make them attractive in OECD labour markets.
As a result, Ghana's domestic health markets face international demand for
workers at internationally competitive prices.
19. The dilemma is that, if the government is going to continue to employ clinicians
directly, the efficiency and sustainability of this budget structure depends heavily
on continued and growing inflows of non-GoG funds to finance essential
pharmaceuticals and medical supplies. In particular, there is are strong
budgetary risks associated with possibly optimistic forecasts of NHIS revenue
growth. Worse, the growth of the PE category is already crowding out the other
items of expenditure. In 2005, for example, the PE allocation from the GoG
source was overspent by 58%, leading to shortfalls in all of the other line items
(see table 3.5)9.
Table 3.5
2005 GoG budget and expenditure (¢ million)
Budgeted
Received Variance (%) Variance (%) Debts
P.E.
1,328,980 2,098,800
769,820
58% 75,700
Admin
95,716 87,716
-8,000
-8%
Services
73,659 57,037
-16,622
-23%
Investment
53,577 32,827
-20,750
-39% 50,000
Total
1,551,932 2,276,380
724,448
47% 125,700
Source: (PPME 2006a)
Pay reforms
20. An in-depth study in 2005 recommended that ADHAs be abolished and replaced
with a 12-tier new pay scale (MoH, GHS, & Cedar Care Trust 2005). The
recommendations also include recruitment on merit, pay by post rather than
grade and no automatic increments. While these measures would improve
payroll efficiency and simplify procedures, they are unlikely to reduce the wage
bill, for a number of reasons. The study's efficiency comparator was staff costs if
ADHAs had been allowed to continue to grow. In order to get professional 'buy
9
There has been an attempt to 'ring fence' important public health activities from in-year budget
cuts. This has been a partial success, in terms of central funds for vaccine procurement etc.
However, funding to the regions and districts is regularly cut by an across-the-board factor.
Financing the sector
15
Review of Ghana Health Sector 2005 Programme of Work
in', the report recommends that no health worker will be worse off in terms of
take home pay. The report foresees an increase in clinical staff without much
change in the workforce structure. The study report makes provision for overtime
pay, at least for the bottom 7 grades. The wage bill is likely to increase rather
than diminish.
Capital expenditure
21. The report of the in-depth review of the capital investment programme was not
available to the review team. However, an analysis of budget trends suggests
that capital expenditure has been growing disproportionately, particularly in
comparison with the target set in the 5 year PoW (10%) and considering the
average national bed occupancy rates of just 62.7% in 2004 (Ministry of Health
2006).
22. Capital investment is particularly prone to grow, as commitments whose costs fall
in the future can easily be entered into now. Some past investment decisions are
now wreaking havoc on the budget - the cuts falling, once again, on the service
line. Already this year, ¢ 12 billion was lost from the service budget, being vired
by the MoFEP to pay for equipment purchased from Spain five years ago.
Another similar amount may be taken during the second quarter. In the districts,
no service funds have yet been received for 2006.
23. A hospital costing study carried out recently found that most capital expenditure
by facilities was misclassified under maintenance (item 2) - presumably because
there is more flexible access to these funds - which means that item 4 is
understating the true amount spend on capital items (GHS 2005).
Allocation by level
24. Table 3.6 compares expenditure in 2001 with the 2006 budget to show the
changing allocation of GoG and Health Fund resources between different
agencies and levels of the health system. Provided the 2006 budget is a reliable
indicator of expenditure, the single largest winner will be the MoH, which will
have captured nearly 98% of the increase in health budget allocations over the
period and will be funded beyond its target in the second Programme of Work10.
The GHS and teaching hospitals will also have made small gains (around 5%
each) and district health services will have benefited slightly from the increase,
but both are well below their targeted share in the PoW II.11 The subvented
organisations will have seen losses of more than 5%, as will have the regional
health services (9%). While there may be some justification for the reduction in
regional services' share, given the increasing decentralisation to district level, it is
not clear why funds for subvented organisations have been reduced. One factor
may be the growing ADHA payments, which have been calculated differently in
10
This figure is based on budget, though, and the budget for MoH includes items procured on
behalf of the regions and districts, such as vaccines and contraceptives. When the expenditure
figures are available, the HQ level share can be reassessed: expenditure figures give a more
accurate picture of where resources are used.
11
The Review of CMA II suggests that allocations to regions and districts may be understated on
account of procurement by MoH on their behalf (Aduonum-Darko et al 2006).
Financing the sector
16
Review of Ghana Health Sector 2005 Programme of Work
GHS facilities, compared with CHAG ones, such that CHAG staff are reported to
receive only an average of one-third to one-fifth as much in ADHAs.
Table 3.6
2001 expenditure & 2006 budget by agency (GoG & HF)
2001
2006
(¢ Bn)
2006 prices
(¢ Bn)
Increase
on 2001
Shares
2006 (%)
Share of
increase since
2001 (%)
Targeted
share
(%)
(PoW II)
MOH
42.95 1,157.44 1,114.49 47% 98.52 5
Teaching
Hospitals
236.28 291.65 55.37 12% 4.89 16
GHS
61.50 125.02 63.52 5% 5.62 7
Psychiatric
Hospitals
48.25 44.38 -3.87 2% -0.34 7
Regional
Health
Service
250.89 144.17 -106.72 6% -9.43 23
District
Health
Service
580.15 651.09 70.94 26% 6.27 42
Subventions 113.67 51.16 -62.51 2% -5.53 n/a
Total
1,333.69 2,464.91 1,131.22 100%
100
100
Source: (External review team 2004); (PPME 2006b); (Ministry of Health 2002)
25. Given that the Christian Health Association of Ghana (CHAG) manages 152
facilities and is estimated to provide 36% of the public health services in the
country (Nyonator & et al. 2006), its allocation of 2% of the total resource
envelope in 2006 is inadequate12. In addition to receiving lower ADHA rates, it is
often disadvantaged in terms of allocation of staff and finds it hard to get new
staff accepted onto the government payroll (80% of CHAG's staff are paid by the
GoG, but its overall staffing numbers are lower than for public facilities). It does
now receive some assistance with administrative and service costs, but these are
small and unpredictable amounts. Similarly, it has been allocated a budget from
the government for investment, but in 2005 did not receive any of the amount
budgeted. As its overseas support base is dwindling, and it is increasingly being
integrated with the public health system, the issue of having a 'level playing field'
should be reviewed. The MOU with the GHS which is currently under discussion
is a good forum for ensuring that resources are allocated fairly between the
public and mission sectors.
3.4 Launch of the National Health Insurance Scheme
26. One of the most main features of 2005 was the launch of the National Health
Insurance Scheme, as mandated by Act 650 in 2003. Its aim, in conformity with
the targets set in the PoW II, was to reduce financial barriers and increase
12
CHAG offers services from primary care to specialist secondary care. It focuses on poor and
remote areas. At the district level, CHAG's hospitals provide 51% of the total number of beds,
53% of the patient days, 45% of admissions and 47% of outpatient visits (CHAG 2004).
Financing the sector
17
Review of Ghana Health Sector 2005 Programme of Work
access to healthcare. A number of preparatory studies were undertaken and a
detailed framework was published in 2004 (Ministry of Health 2004).
Coverage
27. According to the most recent report (NHIC 2006a), 120 district mutual health
insurance schemes (DMHIS) are operating, with an overall coverage of 22% of
the population: that is, the proportion of the population registered, which is not
the same as having joined or paid up. Table 3.7 shows the number of members
who have paid or are exempt. The membership varies significantly by region,
with the highest proportion by far being in Brong Ahafo (nearly 27% coverage),
where there were many independent schemes before the establishment of the
NHIS. Overall, the average is just under 16% nationally. However, these
members are not entitled to benefit from services until 6 months after they have
paid their premia. The number of ID card-holders (those who are entitled to
receive free services) is much smaller: 6.8% nationally. The prediction by the
NHIC that coverage will reach 50% in 2006 (NHIC 2006b) seems unrealistic, in
light of progress to date..
Table 3.7 NHIS coverage, by region, 2006
Region
Estimated
population Membership
% of pop.
members
ID card
holders
% of
pop. ID
holders
UE
963,448 67,9957.06%34,159 3.55%
UW
561,866 52,8709.41%21,564 3.84%
Northern
1,790,417 270,451
15.11%
82,244 4.59%
BA
1,968,205 525,252
26.69%
432,075 21.95%
Ashanti
3,924,925 592,449
15.09%
201,840 5.14%
Western
2,042,753 284,863
13.95%
74,711 3.66%
Central
1,687,311 234,449
13.89%
47,597 2.82%
GAR
3,576,312 597,768
16.71%
106,803 2.99%
Eastern
2,274,453 385,577
16.95%
318,706 14.01%
Volta
1,636,462 211,680
12.94%
68,963 4.21%
TOTAL 20,426,152 3,223,354
15.78%
1,388,662 6.80%
Source: (NHIC 2006a)
Exempt categories
28. Moreover, a large proportion of members fall within exempt categories (table 3.8)
poses a major challenge in terms of financial sustainability. Only 12% of current
members are formal sector workers (the ones contributing the highest amount),
and a further 16% are informal sector workers. Thus a full 72% of members do
not pay for the services which they receive (the largest proportion being children,
though only children of two paid up parents are eligible for this 'exemption').
There is also anecdotal evidence that premium collectors, who are paid a 10%
commission to enrol members, may also be taking bribes to register people as
indigents (NHIC 2006b). This will increas