Financing and Human Resources fo by wuyunyi

VIEWS: 31 PAGES: 100

									HUMAN R ESOURCES AND F INANCING FOR THE H EALTH SECTOR
                     IN MALAWI




                   Africa Human Development
                          Africa Region

                     9/2/2010 7:03:57 AM




                              1
                                           PREFACE


1.      This report was written with contributions from the Malawi health team of the World
Bank and Malawi‟s Ministry of Health. Oscar F. Picazo, senior health economist, AFTHD, put
together the report and drafted the sections on health sector review and health financing. Tim
Martineau, human resource specialist, Liverpool Associates in Tropical Health, was funded under
a Dutch Trust Fund to draft the sections on human resources for health. Christopher Herbst,
AFTH1 intern, organized the raw data of the Malawi Medical Council and analyzed the results.
Christopher Walker, lead operations specialist, AFTHI, and Ramesh Govindaraj, senior public
health specialist and task team leader for the Malawi health team, AFTH1, provided guidance on
the overall structure, content, and processing of the report.

2.      Other people also provided critical information. Antonio Nucifora, macroeconomist,
AFTP1, provided the basic data on government expenditure trends. Khama Rogo, lead
reproductive health specialist, AFTHD, and Mungai Lenneiye, senior social protection specialist,
AFTH1, organized the brainstorming session on community health services that the Malawi
Social Action Fund wanted to provide in collaboration with the MOHP. The notes of that
meeting provided the inputs for the section dealing with the delivery of community health
services. Additional data on MOH actual expenditures were provided by Edward Kataika,
economist, MOHP‟s Planning Department.

3.       The report was peer-reviewed by Pablo Gottrett, senior economist, HD Hub; Chiyo
Kanda, senior economist, OPCIL; and Demissie Habte, consultant on human resources for health
working under the Joint Learning Initiative (JLI) for improving human resources for health, a
global initiative of the Rockefeller Foundation. Allan Whitford of the U.K. Department for
International Development (DfID) also provided comments on the economic analysis of the
Malawi Health Sector Support Project, portions of which were incorporated in this report.
Written comments were also received from Julie Mclaughlin, Son Nam Nguyen, Muhammad
Pate, and Katherin Anne Tulenko, all of AFTH1.

4.     The views expressed in this report are those of the authors and do not reflect those of the
World Bank or its officials, nor of the Government of Malawi. Please send comments to
opicazo@worldbank.org.




                                                 2
                         ABBREVIATIONS AND ACRONYMS


AIDS – Acquired Immunodeficiency Syndrome
ARI – Acute Respiratory Infection
ANC – Antenatal Care (Clinic)
BLM – Banja La Mtsogolo
BMI – Body/Mass Index
CHAM – Christian Health Association of Malawi
CWIQ – Core Welfare Indicators Questionnaire
CONGOMA – Council for Nongovernmental Organizations
CPI – Consumer Price Index
CPR – Contraceptive Prevalence Rate
C-SAFE -
DfID – the U.K.‟s Department for International Development
DHMT – District Health Manatement Team
DHO – District Health Office(r)
DHS – Demographic Health Survey
DIP – District Investment Program (District Implementation Plan?)
EPI – Expanded Program on Immunization
EU – European Union
FY – Fiscal or Financial Year
GAVI – Global Alliance for Vaccines Initiative
GDP – Gross Domestic Product
GF-ATM – Global Fund for AIDS, Tuberculosis and Malaria
GOM – Government of Malawi
HIPC – Highly Indebted Poor Countries
HIV – Human Immunodeficiency Virus
HMIS – Health Management Information System
HNP – Health, Nutrition and Population
HSA – Health Surveillance Assistant
HSSP – Health Sector Support Project
IDA – International Development Association
IEC – Information, Education and Communication
IHS – Integrated Household Survey
IMCI – Integrated Management of Childhood Illness
ITN – Insecticide Treated Net
IPT – Intermittent Presumptive Treatment (of malaria)
JICA – Japanese International Cooperation Agency
MANASO -
MASAF – Malawi Social Action Fund
MASM – Malawi Medical Aid Society
MDG – Millennium Development Goals
MEJN – Malawi Economic Justice Network
MICS – Multiple Indicators Cluster Survey
MK – Malawi Kwacha
MMR – Maternal Mortality Rate
MPRS – Malawi Poverty Reduction Strategy
MOH (MOHP) – Ministry of Health (formerly Ministry of Health and Population)



                                            3
MOF – Ministry of Finance (the Treasury)
MOLG – Ministry of Local Government
MTEF – Medium-Term Expenditure Framework
NAPHAM – National Association of People with HIV/AIDS in Malawi
NGO – Nongovernmental Organization
NHA – National Health Accounts
NHP – National Health Plan
NORAD – Norwegian Agency for Development
NSO – National Statistics Office
OAU – Organization of African Union
OECD – Organization of Economic Co-operation and Development
ORS or ORT – Oral Rehydration Salts or Oral Rehydration Therapy
ORT – Other Recurrent Transactions
PE – Personal Emoluments
PEPFAR
PHC – Primary Health Care
PMS – Poverty Monitoring Survey
PoW – Program of Work
PPE – Pro-Poor Expenditures
PRSP – Poverty Reduction Strategy Paper
PSI – Population Services International
SASE – Selective Accelerated Salary Enhancement Scheme
SIDA – Swedish International Development Agency
SWAp – Sector-Wide Approach
STI or STD – Sexually Transmitted Infections or Diseases
TA – Technical Assistance
TB – Tuberculosis
TBA – Traditional Birth Attendant
TFR – Total Fertility Rate
UNFPA – United Nations Population Fund
UNICEF – United Nations Children‟s Fund
WHO – World Health Organization




                                          4
                                                  TABLE OF CONTENTS
     P REFACE ............................................................................................................................2
     ABBREVIATIONS AND ACRONYM S........................................................................................3
     TABLE OF CONTENTS ..........................................................................................................5
     LIST OF TABLES ..................................................................................................................7
     LIST OF FIGURES.................................................................................................................8
I.      INTRODUCTION .........................................................................................................9
        A. Purpose of the Report ..................................................................................................9
        B. Organization and Limitations of the Report...................................................................9
II. HEALTH SECTOR PERFORMANCE AND ITS IMPLICATIONS ON HUMAN
RESOURCES AND FINANCING FOR HEALTH .............................................................11
        A. Review of Issues on Household Health Status, Knowledge, Behavior and Service
        Utilization.....................................................................................................................11
        B. Review of Issues on the Health Service Delivery System...............................................11
        C. Preview of Health Workforce Issues ...........................................................................11
        D. Preview of Health Financing Issues ...........................................................................12
III.       STATUS OF HUMAN RESOURCES FOR HEALTH.............................................16
        A. Dimensions of the Human Resource Crisis ..................................................................16
        B. Causes of the Human Resource Crisis.........................................................................22
IV.        OPTIONS ON HUMAN RESOURCES FOR HEALTH ..........................................27
        A. Options for Increasing the Production of Health Workers ............................................27
        B. Options for Attracting and Deploying Health Workers .................................................31
        C. Options for Retaining Health Workers ........................................................................33
        Recent HR Innovations Catering to Specific Groups ........................................................33
        Upgrading Salaries and Benefits and Improving Working Conditions ...............................35
        Improving the Career Structure of Health Workers ..........................................................37
        D. Strengthening HR Management Development Systems.................................................37
V.         STATUS OF THE FINANCING OF HEALTH SERVICES....................................41
        A. Overall Trends in Health Expenditures .......................................................................41
        B. Government Health Expenditures ...............................................................................42
        C. District Budget Planning and Implementation .............................................................48
        D. Donor Health Expenditures .......................................................................................50
        E. Private Institutional Health Expenditures ...................................................................53
        F. Household Health Expenditures and User Fees ...........................................................54
VI.  OPTIONS FOR FINANCING THE ESSENTIAL HEALTH PACKAGE AND NON-
EHP INTERVENTIONS ....................................................................................................58
        A. The Malawi Essential Health Package ........................................................................58
        B. Principles Underlying the Provision of the Essential Health Package ...........................59
        C. The Medium-Term Sector-Wide Program of Work .......................................................60
        D. Options for Reallocation of District Budgets ...............................................................66
        E. Options for Government/NGO Partnerships: Health Service Agreements......................67
        F. Options for Government/Community Partnerships.......................................................70



                                                                     5
       G. Options for Financing Non-EHP Services...................................................................71
VII.      WAY FORWARD ...................................................................................................76
  ANNEX A: A BRIEF ANALYSIS OF THE MEDICAL AND ALLIED P ROFESSIONS IN MALAWI .......83
  ANNEX B: SALARY SCALES AMONG INSTITUTIONS THAT EMPLOY HEALTH P ERSONNEL IN
  MALAWI, CIRCA 2003/04 ...................................................................................................87
  ANNEX C: DIAGRAM OF THE APPLICATION P ROCESS ..........................................................88
  ANNEX D: THE P ROPOSED SHORT -TERM SUPPLEM ENTARY STAFFING SCHEM E FOR HEALTH
  P ERSONNEL IN MALAWI ....................................................................................................89
  ANNEX E: GENDER DIM ENSIONS OF HUM AN RESOURCES FOR HEALTH IN MALAWI..............91
  ANNEX F: DRAFT TERM S OF REFERENCE FOR A STUDY ON THE INSTITUTIONAL
  ARRANGEM ENTS FOR HUM AN RESOURCE MANAGEM ENT AND P LANNING IN MALAWI ..........92
  ANNEX G: DRAFT TERM S OF REFERENCE FOR LONG-TERM TECHNICAL ASSISTANCE IN
  HUM AN RESOURCE MANAGEM ENT AND P LANNING IN MALAWI...........................................94
  ANNEX G: P REPAID P LANS AS A P ERCENTAGE OF P RIVATE HEALTH EXPENDITURES IN SUB-
  SAHARAN AFRICAN COUNTRIES, 1995 AND 2000 ................................................................96
  ANNEX H: GLOBAL ESTIM ATES OF COST -EFFECTIVE AND AFFORDABLE P UBLIC HEALTH AND
  CLINICAL SERVICES ..........................................................................................................97
  ANNEX I: ACTUAL AND EXPECTED ANNUAL DONOR COMM ITM ENTS TO THE HEALTH SECTOR
  IN MALAWI, FY94 TO FY05 ...............................................................................................98
  ANNEX J: CHAM HEALTH FACILITIES BEING CONSIDERED UNDER EHP HEALTH SERVICE
  AGREEM ENTS IN MALAWI, 2004 ........................................................................................99




                                                              6
                                                      LIST OF TABLES
TABLE 1. P ERCENT VACANCY RATES BY CADRE, MOH AND CHAM, 2003..............................17
TABLE 2. MOH STAFF LOSSES, BY CADRE AND REASON FOR LOSS, 2002 ................................18
TABLE 3. COM PARATIVE MONTHLY SALARIES OF HEALTH WORKERS IN SUB-SAHARAN AFRICA.
    ....................................................................................................................................19
TABLE 4. MONTHLY SALARY LEVELS AND ALLOWANCES OF SELECTED CADRES OF HEALTH
   WORKERS, BY CADRE AND BY SALARY AND ALLOWANCES (IN MALAWI KWACHA), 2004 .19
TABLE 5. DISTRIBUTION OF SKILLED STAFF BY CADRE, MOH AND CHAM, 2003 ....................21
TABLE 6. EXPECTED TRAINING OUTPUTS UNDER THE P ROGRAM OF WORK, 2003/04-2009/10 ..28
TABLE 7. ESTIM ATED UTILIZATION OF MOH BUDGET , BY MAJOR EXPENDITURE CATEGORIES,
   2001/02 AND 2002/03 ....................................................................................................47
TABLE 8. DISTRICT STATUS BY “O VERFUNDING” OR “UNDERFUNDING” OF ORT ALLOCATION ,
   2000/01 ........................................................................................................................48
TABLE 9. SUB-SECTORAL BREAKDOWN OF DONOR COMM ITM ENTS TO THE HEALTH SECTOR IN
   MALAWI, 1998-2000 A VERAGE ......................................................................................52
TABLE 10. TOTAL AND P ER CAPITA HOUSEHOLD HEALTH EXPENDITURES, BY DIFFERENT
   CATEGORIES, 1998/99....................................................................................................54
TABLE 11. P ERCENT OF HEALTH EXPENDITURES SPENT BY HOUSEHOLDS, BY QUINTILE AND BY
   TYPE OF HEALTH SERVICES, 2000 ..................................................................................55
TABLE 12. P ERCENT OF RESPONDENTS WHO AGREE SOM EWHAT OR STRONGLY TO HEALTH AND
   EDUCATION FEES, 1999-2000 .........................................................................................55
TABLE 13. P ERCENT OF HOUSEHOLDS DISSATISFIED WITH HEALTH SERVICES RECEIVED, AND
   THOSE CITING COST AS A REASON FOR DISSATISFACTION , 2002 ......................................56
TABLE 14. P OTENTIAL NUM BER OF ANNUAL CHILD DEATHS THAT COULD BE AVERTED IN
   MALAWI WITH CURRENT CHILD HEALTH TECHNOLOGIES, EARLY 2000 S ..........................59
TABLE 15. P ROGRAM OF WORK, 2004/05 TO 2009/10 .............................................................61
TABLE 16. ANALYSIS OF P OW BY VARIOUS CATEGORIES, 2004/05 TO 2009/10 ........................63
TABLE 17. ESTIM ATED ANNUAL RESOURCE COMMITM ENTS BY DONORS, BY P ERIOD, 1994-2006
    ....................................................................................................................................63
TABLE 18. ESTIM ATED ANNUAL RESOURCE COMMITM ENTS BY DONORS, BY P OOLED AND
   UNPOOLED FUNDING , 2003/04-2005/06 ..........................................................................64
TABLE 19. ANALYSIS OF RESOURCE REQUIREM ENTS VERSUS RESOURCE AVAILABILITY OF THE
   P OW, 2004/05 TO 2009/10 (IN MILLION U.S. DOLLARS)...................................................65
TABLE 20. CASELOAD AND COST OF THE HEALTH SERVICE AGREEM ENT BETWEEN THE
   DISTRICT HEALTH OFFICE AND CHAM HOSPITAL “X”, FY2004 ......................................69




                                                                    7
                                                LIST OF FIGURES

FIGURE 1. ATTRITION OF MOH P ERSONNEL BY CAUSE, 1990-2000 .........................................18
FIGURE 2. P ER CAPITA GOVERNM ENT HEALTH EXPENDITURES, IN MALAWI KWACHA, AT
    CURRENT AND CONSTANT 1993 P RICES, 1993-94 – 2001/02 .............................................42
FIGURE 3. P ERCENT SHARE OF GOVERNM ENT HEALTH EXPENDITURES TO TOTAL GOVERNM ENT
    EXPENDITURES, TO TOTAL NONDISCRETIONARY GOVERNM ENT EXPENDITURES, AND TO
    GDP, 1993/94 – 2001/02 ................................................................................................43
FIGURE 4. TRENDS IN ACTUAL GOVERNM ENT HEALTH EXPENDITURES, IN MILLION MALAWI
    KWACHA, AT CURRENT AND CONSTANT 1993 P RICES, 1993/94 – 2002/03.........................44
FIGURE 5. P ERCENT ALLOCATION OF MOH RECURRENT EXPENDITURES, BY ECONOM IC
    CLASSIFICATION , 1995/96 – 2005/06...............................................................................44
FIGURE 6. P ERCENT ALLOCATION OF MOH RECURRENT EXPENDITURES BY INSTITUTIONAL
    LEVEL, 1995/96 – 2005/06 .............................................................................................45
FIGURE 7. SCENARIOS FOR AVAILABILITY OF GOVERNM ENT RESOURCES FOR HEALTH, IN
    MILLION U.S. DOLLARS, 2004/05-2009/10 ......................................................................65




                                                             8
                                   I.        Introduction

A. Purpose of the Report

5.       This report, “Human Resources and Financing for the Health Sector in Malawi” is the
second part of a two-volume study analyzing the performance of the health, nutrition, and
population (HNP) sector in the country. The first report analyzes the country‟s health sector
performance in terms of outcomes, household health behavior and knowledge, utilization of
services, and the service delivery system. The implications of the findings from the first report
were used to inform the policy options reviewed in this second report. Both reports support two
objectives: (a) to provide technical inputs into the process of project preparation for the new
Malawi Health Sector Support Project (HSSP), and (b) to provide the IDA with a vehicle for
policy dialogue with the government, donors, and other development partners. The government
and IDA jointly chose the two focus areas of health financing and human resources because
issues revolving around them have emerged as the most problematic, with shortages in both
financial resources and health workers contributing to severe service deficits, reduced service
quality, and worsening access to care among the most vulnerable populations. There are equally
important areas (such as drug supply and distribution, facility planning and maintenance, the
technical soundness of health interventions, and community/household involvement in the
production of health, to mention a few), but it was deemed that these other areas are better dealt
with by other donor efforts. This report is intended to complement these other efforts.

6.      The report relied primarily on the following sets of documents: (a) Results of primary
data gathering efforts including the Core Welfare Indicators Questionnaire (CWIQ) Survey in
2002. (b) The 2002/03 licensure dataset of the Malawi Medical Council. (c) Studies conducted by
the Government of Malawi and its donor-partners in the process of preparing the sector-wide
approach (SWAp) and its related Program of Work (PoW), specifically the JICA-funded
inventory of health facilities, the national human resources plan, and the National Health Plan that
the government issued in May 1999. (c) Related studies located in Malawi and those found in the
Web. (d) In addition, the World Bank‟s Malawi health team also conducted interviews with
stakeholders during two missions, the first from December 1-19, 2003, and the second from
March 22 to April 9, 2004.

B. Organization and Limitations of the Report

7.      The succeeding chapters are organized as follows: Chapter II summarizes the key
problems in the health sector and their implications on human resources for health and financing
of the health sector. Chapter III discusses the issues on human resources for health while Chapter
IV explores options for improving the production, deployment, retention and management of
these workers. Chapter VI analyzes the current state of financing in the health sector while
Chapter VII explores options for the financing of essential health services and non-EHP services.
Chapter VIII summarizes the way forward.

8.      The following are the limitations of the report:




                                                 9
(a)   In health financing, the first National Health Accounts (NHA) exercise done in
      1997/98 is still being updated as this study was being written. There has not been
      any comprehensive assessment of user fees, nor on the private health sector.
      With respect to government expenditures, no public expenditure tracking study
      has been undertaken. No benefit-incidence analysis of health expenditures has
      been done.

(b)   In human resources, Malawi has not undertaken any head-count study, and the
      personnel data relied solely on government-reported information. This study did
      analyze the registry of the medical council, and the findings are reported in the
      chapters on human resources.




                                     10
      II.       Health Sector Performance and its Implications on Human
                           Resources and Financing for Health

9.       Volume I of this two-part study analyzed Malawi‟s health sector performance in terms of
household health status, level of knowledge, pattern of behavior, and actual use of health services,
as well as the health system‟s ability to deliver essential health services. Sections A and B of this
chapter highlights these findings and their implications on human resources and financing for
health. Sections C and D provides a preview of the specific findings on is

A. Review of Issues on Household Health Status, Knowledge, Behavior and Service
Utilization

10.         The economic and political environment.

11.      Poor health outcomes especially of the poor; impact of poverty on health, and health on
poverty.

12.    Gap in health knowledge versus practice and use; cultural impediments; supply side
impediments.

13.         Inequality and health; geographic, socioeconomic and gender issues.

14.     Increasing private utilization of health services; link to health service agreements and
contracting.

B. Review of Issues on the Health Service Delivery System

15.     Checkered performance (some positive achievements despite declining growth and
poverty); what accounts for extremes in performance?

16.      Severe constraints in the delivery of essential health services; link directly to HR and
health financing issues.

17.     Weak community health services; link to community health workers and community
financing of services; link to decentralization.

18.     Highly verticalized funding service delivery; link to sector-wide approach; challenge of
coordinating an increasing flow of donor resources; link to pooled funding.

19.         Other critical issues not covered by volume 1: central medical stores.

C. Preview of Health Workforce Issues

20.         Production issues and options.

21.         Issues and options in attracting and deploying workers.



                                                   11
22.      Issues and options in retaining health workers.

23.      Strengthening management and development systems.

24.      Improving human resource strategic planning and policymaking.

D. Preview of Health Financing Issues

25.     After years of health planning without explicitly defined health service priorities,
the Government of Malawi has formulated a plan to provide an essential health package
that supports its targets of achieving the Millennium Development Goals. The overall
financing of this package has been calculated for the next seven years, and it involves per capita
(government and donor) health spending increasing from US$7.70 to US$12.60 1 per year. These
figures are far more modest than the financing requirements estimated by the Commission on
Macroeconomics and Health (around US$34 per capita per year) and similar global estimates, for
they focus on the most immediate recurrent-cost requirements and represent a pruned-down
version of the “ideal requirements” that had a more extensive infrastructure component. To
achieve these financing targets, government and donor health spending per capita in Malawi
(US$4.27 in 2001, according to WHO) needs to double from the 2001 figure in the early years of
the PoW, and to triple in the outer years.

26.       Meeting the government’s financing targets in the PoW is feasible, but it poses a
real challenge. Although the government has demonstrated its commitment to increase its
budget allocation for the health sector, persistent macroeconomic difficulties pose a serious risk
to its ability to maintain fiscal balance. The large domestic borrowing is particularly worrisome
since it severely reduces the amount of resources that can be made available for the health sector.
Weak government discipline in adhering to its medium-term expenditure program needs to be
addressed as an urgent matter.

27.     In addition, persistent problems remain in the allocation, use, and reporting of
available government health resources . The key issues involve:

         (a)      Data consistency. In recent years, several “official” figures have cropped up
                  reporting different versions of the same “estimate” (preliminary, final, revised,
                  etc.), making it difficult to establish facts, and to use these for decision-making.
                  Moreover, budget books have a tendency to change the definition of some items
                  (especially “pro-poor” expenditures for health), making it difficult to track
                  whether or not government has met its intentions.

         (b)      Use of available resources. There continues to be poor expenditure tracking and
                  reporting. There are no reported large underspending of available resources, but
                  districts complain about unpredictable receipt of allocations, and the hurried
                  adjustments and reprioritization that they need to make for resources that are no
                  longer forthcoming during the fiscal year. Districts are also seriously
                  handicapped in the use of budget and expenditure data in priority setting.


1
  This report uses average exchange rates to the US dollar, rather than the more internationally co mparable
international dollar rate. The M OH reasons for this choice are: (a) the greater ease with wh ich
policy makers understand the implications of the figures stated in average exchange rates, and (b) the
greater ease of forecasting average exchange rates.


                                                     12
       (c)    Allocative efficiency. Severe erosion in the allocation going to personnel
              emoluments (in favor of other recurrent transactions) has hurt health workers and
              is contributing to the deepening human resource crisis; a better budgetary balance
              is needed. In the same vein, deliberate efforts have been made to allocate more
              resources to peripheral areas, but MOH Headquarters continue to receive a share
              bigger than what optimal figures would indicate. Recent initiatives, such as
              zonal health offices and the Health Commission, could increase the allocation to
              HQ further, limiting resources for rural districts.

       (d)    Geographic equity. Regional, inter-district, and intra-district inequity in the
              allocation of budgetary resources (and by implication, use of health services)
              continues, mainly because of the persistence of incremental budgeting processes
              (i.e., funding the existing infrastructure and workforce) and political sensitivity
              of drastically over-hauling the present allocation system. MOH has begun to
              consider alternative allocation formula, but these need to be thought out in the
              wider context of fiscal devolution.

       (e)    Socioeconomic equity and gender balance. The absence of benefit incidence
              analysis of health expenditures constraints public discussions of the social equity
              of government spending. Conventional wisdom suggests that most of the
              benefits of government spending flows to the poorer Malawians, but the more
              urban location of most government health facilities (compared to the more rural
              location of CHAM‟s and NGOs‟ facilities) suggests otherwise. The low
              utilization of health services especially by rural, poorly educated women, also
              needs urgent attention.

28.     To improve technical efficiency and reach wider population segments, the MOH has
initiated health service contracts and other forms of partnership with nongovernment
providers. This is a step in the right direction and needs to be supported by donors.

       (a)    The health service agreements with CHAM are moving apace. A year‟s
              experience with this scheme should be evaluated, specifically on issues dealing
              with the predictability of government funding, use/misuse of funds, actual cost of
              services, and conflict resolution.

       (b)    Extending the health service contracting scheme to the non-CHAM NGO sector,
              including social marketing programs, should be seriously considered. The recent
              passage of the NGO Law and the Procurement Law should provide the legal
              underpinning for these novel contracts.

       (c)    Private for-profit providers could also be involved in this scheme. One program
              that should be explored is the use of tertiary private facilities (such as
              Mwaimathu) to provide health services to senior civil servants currently being
              medically evacuated to other countries for treatment that could be dealt with
              locally. Contracts with these private hospitals‟ diagnostic and imaging facilities
              can also be explored. In both cases, contract prices should be lower than what
              the government currently pays for abroad.

       (d)    Formal government partnerships with communities have, so far, been limited.
              This needs to expand further if the government is serious about reaching the most
              vulnerable.


                                              13
29.     The large and increasing household spending on health is a cause of concern.

30.     Decentralization and resource flow to districts.

31.     Although the government has decided that the EHP services will be provided for
free to Malawians for equity purposes, clarifying this policy to all partners at all levels of
care is critical. Moreover, the proper instruments and incentives must be designed so that
everyone adheres to this policy. Finally, the government and its partners need to ensure that the
EHP program is fully funded so that this policy does not degenerate into “free-care-but-no-
services-or-goods-available” syndrome that has bedevilled most “free-care” regimes in sub-
Saharan Africa. The remaining sticking points in this area are the following:

        (a)     Government tertiary hospitals and nonprofit providers. Central hospitals
                currently charge fees for most patients on these EHP services; convincing them
                to forego the fees (at the same time, requiring them to have a more sustainable
                financing base under the policy of hospital autonomy) will be extremely difficult,
                unless the central government can reimburse these hospitals for the EHP services
                they provide. Under health service agreements with the government, these same
                problems also hold for CHAM and other nonprofit NGOs that currently charge
                fees for primary services. In both government tertiary hospitals and all levels of
                CHAM and NGO facilities, central government financing scheme to pay for
                these henceforth free services will need to be developed, otherwise the policy
                will be weakly enforced.

        (b)     Drug revolving funds (DRFs). Removing any form of household payment or
                community contribution for these DRFs will virtually nullify their sustainability
                objectives. The government needs to develop a strategy on how to pursue
                community pharmacies under a free-care policy.

        (c)     Social marketing programs. Little discussion has been conducted on the future
                of these programs under a free-care policy. It is possible that donors or the
                government can finance voucher programs for ITNs, oral rehydration salts, or
                contraceptives for the poorest. Indeed, there are successful experiences in these
                areas, such as the KINET ITN voucher scheme for pregnant mothers in Tanzania,
                or the reproductive health service vouchers for the poor in Nicaragua.

        (d)     Devolved health facilities. Under the Malawi Local Government Act, devolved
                health facilities will have the power to impose charges for services they provide.

        (e)     Communication program for free care. Even under existing free-care programs
                such as STI diagnosis and treatment, actual utilization has been low, implying
                that many potential c lients are not aware of the policy. An all-out
                communication program is needed to disseminate the free-care policy, and where
                people can get free-care.

32.     Donor expenditures, though increasing in recent years, has engendered issues on
efficiency, complementarity with budget financing, geographic and client targeting, and
overall sustainability.

33.     The financing non-EHP (especially hospitalization) services is not given prominence
in the PoW, but this issue needs to be dealt with as part of the overall health financing


                                                14
framework. A proposed solution to non-EHP or hospital financing is social health insurance.
Given Malawi‟s lack of experience in other health financing mechanisms, this will be a daunting
task. A task team is needed to look at the various aspects of this initiative including the
prerequisites of establishing such scheme; the analytical activities needed to underpin its
development; financial and sustainability issues; the role of private providers; possible
governance structure; and its institutional requirements (including contracting administrative
functions).




                                               15
                  III.        Status of Human Resources for Health

34.       A major reason for Malawi’s weak health sector performance is due to the crisis in
the health workforce. The country‟s health system has become so dysfunctional that unless
drastic measures to address human resource issues are implemented now, the country‟s mortality
and morbidity profile is likely to get worse, making it unable to reach its Millennium
Development Goals and its objective of reducing poverty through better health services for its
people. This chapter describes the dimensions, causes, and consequences of the health workforce
crisis, focusing on the government health system where the problem is most severe, and to a
lesser extent on CHAM. The analysis deals for the most part with nurses, since they are the most
affected. (Annex A focuses on the medical profession, which is also in extremely short supply.)
The subsequent chapter lays out the options that the country can consider in addressing the
human resource problem in the short- and medium-term, and the opportunities that should be
tapped in light of Government‟s and donors‟ belated but keen interest in dealing with this issue.
As will be shown in the next two chapters, although the crisis is due largely to the government‟s
inability to finance the production and retention of health workers, there are other policy issues as
well, including the choice of provider, weak HR systems, and the unmanaged growth of NGOs.

A. Dimensions of the Human Resource Crisis

35.     Although the human resource crisis cuts across all sectors, it is particularly
appalling in health. The vacancy rate – the number of established posts that are unfilled – at
MOH is the highest at 75 percent; for the other ministries, the respective vacancy rates are 65
percent for Housing, 56 percent for Agriculture, and 40 percent for Education (DHRMD and
UNDP, July 2003) 2 . The workforce situation began to unravel in the late 1990s. As late as
1997/98, there was only 4.8 percent vacancy rate among mid-level MOH staff (EO/TO to
CEO/CTO) and there were zero vacancy rates at the senior level (P8 to P2) and low level (CO/TA
to SCO/STA) staff. By 2002/03, the vacancy rates have reached 74.4 percent for senior level,
67.1 percent for midlevel, and 40.6 percent for low-level staff (Malawi Government, budget
books FY97/98 and FY02/03).

36.      The vacancy rates are particularly high for skills that are most needed. As Table 1
shows, in MOH, vacancies are particularly acute in specialist doctors (82.0 percent), nursing
officers (77.4 percent) and nursing sisters (87.8 percent), lower-level nurses (41.6 percent),
environmental and health education officers (around 70 to 80 percent), pharmacists (68.6 percent)
and lab technicians (40.0 percent). CHAM, which receives a direct subvention from the Ministry
of Finance for the payment of staff salaries, has vacancy rates lower than MOH for most cadres,
possibly because it can offer modest salary augmentation from its user-fee revenues. Still,
CHAM‟s vacancies are particularly high for medical officers and medical assistants (higher than
40 percent), combined nursing cadres (53.2 percent), and pharmacy and laboratory specialists
(around 90.0 percent). At these rates of shortages, the Malawian health system is virtually
paralyzed.



2
  No fu rther analysis is available whether these are “real” vacancies (unfilled posts with a budget allocation
for them to be filled, i.e., the only problem being that there are no qualified applicants) or “ghost”
vacancies (unfilled posts without a budget allocation for them to be filled during the current or previous
years).


                                                      16
                 Table 1. Percent Vacancy Rates by Cadre, MOH and CHAM, 2003
                                                 MOH                     CHAM
                                                       Vacancy                   Vacancy
                 Cadre              Posts    Filled    Rate (%) Posts Filled Rate (%)
    Specialist doctor                151       27         82.0    n.a.    n.a.      n.a.
    Medical officer                   93       63         32.3     36      21       41.7
    Clinical officer                 563      425         24.5    123      79       35.8
    Medical assistant                464      285         38.6    278     154       44.6
    Reproductive Officer             258        0        100.0    n.a.    n.a.      n.a.
    Nursing officer                  883      200         77.4    n.a.    n.a.      n.a.
    Nursing sister                  2,791     341         87.8    n.a.    n.a.      n.a.
    Psychiatric nurse                118       90         23.7    n.a.    n.a.      n.a.
    Community nurse                  268      189         29.5    n.a.    n.a.      n.a.
    Enrolled nurse/midwife          1,906    1,113        41.6    n.a.    n.a.      n.a.
    Nursing (combined)              5,966    1,932        41.6   1,933    905       53.2
    Environmental health officer     483      117         75.5    n.a.    n.a.      n.a.
    Health Assistant                 475      143         70.0    n.a.    n.a.      n.a.
    Health education officer          76       16         79.0    n.a.    n.a.      n.a.
    Health surveillance assistant   4,324    4,324       100.0    n.a.    n.a.      n.a.
    Laboratory related               190       76         40.0    183      73       60.1
    Pharmacy related                 207      142         68.6     18       2       88.9
    Radiology related                149       36         24.2     86       7       91.9
    Dentistry related                 15       15        100.0    155       9       94.2
    Source: Extracted from Health Planning Services Department (2003) Tables 7 & 11; Issues and
                                  Challenges paper, p.33 for CHAM

37.     Staff losses from MOH and CHAM are mainly due to death and resignation of staff
intending to work in other areas, whether domestically or abroad. For MOH, major staff
losses are due to death (accounting for 56 percent of losses in 2002) and voluntary resignation
(which accounted for 33 percent of losses)3 (see Table 2). The recent draft human resources plan
has factored in an attrition rate of 1.25 percent due to death, and for resignations, emigration and
retirement 3.5 percent for higher-level staff and 2.5 percent for lower-level staff. Losses due to
death and resignation probably occur among the younger and most productive age groups.
Although there is currently no data available on the age profile of health workforce, the HR
planning exercise of 2002 assumed a relatively young workforce and therefore few losses from
retirement in the near future (Hornby and Oczan, 2003). A survey carried out in 1998 also
showed a relatively young age profile overall. However a significant number of nurses were 40
years or more (45 percent for enrolled nurse/midwives, and 76 percent for community health
nurses) indicating that many of these are now over five years on approaching the retirement age
of 55. Given that the enrolled nurse/midwife is the largest professional group and a key service
provider at sub-district level, this should be of some concern.



3
  These 2002 figures in the PoW are much lower than comparable figures for 2000 in the UNDP report on
the impact of HIV/AIDS on the Malawi‟s public sector (Feb. 2002). Table 26 of the UNDP report shows
the following attrit ion numbers for MOH personnel, by cause: death, 200; dis missal, 146; redundancy, 22;
resignation, 318; and retirement, 544; for a total of 371. The lower 2002 PoW figures may simp ly indicate
that there are very few M OH personnel left.


                                                    17
                Table 2. MOH Staff Losses, by Cadre and Reason for Loss, 2002
        Cadre            Death Abscondment Resignation Retirement Dismissal Total
Medical officer            0        0           4           0        0        4
Clinical officer           3        0           7           0        0       10
Medical assistant          6        1           1           0        0        8
Asst. environmental        1        0           1           0        0        2
officer
Nurse                      8                0                    9                2             0          19
Enrolled &                20                3                   13                2             0          38
community nurse
Health surveillance       42                4                   12                1             2          61
asst.
Total                     80             8            47           5                            2      142
                               Source: Program of Work, March 2004


38.     Although there are no accurate records, the excess death rates of health workers can
only be explained by HIV/AIDS. Only 8 MOH workers died in 1990; this number reached 270
in 1999 and 200 in 2000 (Figure 1). According to the UNDP study on the impact of HIV/AIDS
on Malawi‟s public sector, 59 percent of the MOH deaths in 2000 were those between the ages of
30 to 44 (UNDP, 2002), the same age range where HIV/AIDS is concentrated. Because the cause
of death is not reported in the personnel files, and in the absence of clinical diagnosis of sick
workers, it is not possible to determine the number of health workers who have died of
HIV/AIDS. However, based on NAC estimates of AIDS rates among professionals of 9.0 to 11.8
percent from 1995 to 2000, one could impute total AIDS deaths from the total number of deaths
among health workers, i.e., 109 out of 1,107 deaths in six years from 1995 to 2000. The figure,
however, appears too small, as the remainder (998) could not be reasonably explained by other
causes or adult mortality in Malawi.

                  Figure 1. Attrition of MOH Personnel by Cause, 1990-2000

                                            Source: UNDP (2002)


   450
   400
   350
   300                               107                                                     270    200
   250                                                                                198
   200
                                                                           150
   150                                                               137
   100                                          145     152
    50                     65
                  30
     0     8
         1990    1991     1992       1993       1994    1995      1996     1997       1998   1999   2000

                        Retirement     Resignation          Dismissal & redundancy      Death




                                                       18
39.      Large-scale resignations have been triggered by low and stagnant salaries and poor
working conditions in government service. Except for staff working for MOH under the
recently-introduced Performance Contract Scheme, pay levels in the government civil service are
low in comparison with neighboring countries (Valentine, 2003). Among health workers, the
median salary earner in Malawi only gets US$45-equivalent per month, by far lower than any of
the other countries in the region (Table 3). As Table 4 shows, monthly remuneration inclusive of
allowances of the most senior-level specialist doctor is only around MK25,603 (equivalent to
US$234), while an entry-level specialist doctor earns around MK18,782 (around US$170). A
senior nurse earns MK12,930 (or US$117) per month while an entry-level nurse assistant earns
MK8,109 (US$74) 4 . Low pay is a major reason for the Government‟s and CHAM‟s difficulty in
attracting and retaining staff.

        Table 3. Comparative Monthly Salaries of Health Workers in Sub-Saharan Africa.
    Country      Minimum               Median           Top      Ratio Top-to            Ratio Top-to
               Salary-Earner        Salary-Earner     Salary-      Minimum                 Median
                    US$                  US$          Earner Compensation                Compensation
                                                        US$
Botswana             127                 857           3,803           30:1                      4:1
Malawi                31                  45           3,403          110:1                      76:1
Tanzania              58                 251           1,028           18:1                      4:1
Uganda                44                 192           1,108           25:1                      7:1
Zambia                40                 145           1,357           34:1                      9:1
                                Source: Valentine, T.R. (2003), Table 1


Table 4. Monthly Salary Levels and Allowances of Selected Cadres of Health Workers, by Cadre
                  and by Salary and Allowances (in Malawi Kwacha), 2004
     Cadre & Grade          Gross Monthly          Monthly Allowances           Total
                               Salary     Medical Professional Housing Subtotal
Specialist physician,          18,652      1,500       0       15,000   16,000 34,652
top level (P2)
Specialist physician,         6,000         1,800         2,300        12,000            16,100     22,100
entry level (P8)
Nurse (TO)                    3,630         1,500         1,800        6,000             9,300      12,930
Nurse (TA)                    2,809         1,000         1,800        2,500             5,300      8,109
Medical assistant (TA)        2,239         1,000         1,800        2,500             5,300      7,539
Pharmacy technician           4,009            0          1,800        6,000             7,800      11,809
(TO)
Pharmacy technician           2,809            0          1,800        2,500             4,300          7,109
(TA)
                      Source: Ministry of Health (2004), Tables 1, 2, 4 & 6.




4
  For better analysis, these monthly earnings need to be compared with those earned by equivalent highly -
trained professions, e.g., engineers, working in the government and private sectors. Such data, however,
are unavailable. Annex B does provide informat ion on the existing salary scales among institutions that
emp loy health personnel in the country.


                                                    19
40.      Full-time equivalent health workers is even worse than the already dwindling
number of workers due to many staff absences and official time away from posts. It has
been estimated that there are 498 days lost per 100 nurses (or approximately 5 days each) through
absenteeism (MOHP POW, 2004). The actual time spent away from work due to late arrival and
early departure and for more legitimate purposes such as training or workshops is likely to be far
more significant. To supplement their meager incomes, health staff often use unofficial “coping
strategies” including attendance at workshops and training courses and entrepreneurial sidelines.
Absences from post arising from these often unnecessary appointments puts further pressure on
the staffing shortage, effectively reducing the full-time equivalent available health workers.
Other unavoidable staff absenteeism is due to illness – largely HIV/AIDS related – either of the
individual or of a dependent, while some are due to staff chasing their salaries, allowances, and
other bureaucratic tasks. In recent years, the government has also declared prolonged Christmas
and Easter weekends as implicit “savings” strategies 5 . Combined together, these reductions in
actual time spent at post translate to significant reductions in the volume of health services
provided.

41.       The mass exodus of health workers out of government service has exacerbated the
geographic maldistribution of health workers, a chronic problem in poor countries. An
analysis of urban/rural distribution of the workforce, where "urban" is defined as being based in a
hospital or the district headquarter, shows that the majority of skilled health service providers are
concentrated in urban areas. Although only 20 percent of Malawians live in urban areas, large
majorities of key health workers are in urban areas and townships: 96.6 percent among clinical
officers; 82.0 percent among nurses; 78.2 percent among enrolled nurse/midwives (Hozumi,
2003). Even among health surveillance assistants who are supposed to be posted in communities,
nearly one third (29.2 percent) of them are based in urban areas. The distribution of the CHAM
health workforce is relatively better, since most of its facilities are located in rural areas.
However, their rural location could account for their slightly higher vacancy rates for nursing
sisters, laboratory staff, and radiology-related staff.

42.       Current establishments in both MOH and CHAM are marked by major imbalances
and, thus, poor skill-mix. While skilled jobs are in severe shortage, there are some signif icant
surpluses in low and unskilled posts. The MoH staff establishment, which was revised as part of
the Functional Review in 1998, though not implemented until 2001 (Martineau, Sargent, et al,
2001), is not based on a detailed analysis of workload. Thus the comparison of posts filled
against establishment is not entirely valid. However, interviews with senior managers broadly
agree with these areas of imbalance. The shortage of Enrolled Nurses/Enrolled Nurse Midwives
(EN/ENM) means that only 152 of 585 government and non-government health facilities,
including hospitals, currently meet even the “relaxed” staffing standard of 2 nurse/midwives per
facility required to provide the Essential Health Package (EHP) (Hozumi, 2003). Sub-district
health facilities may simply be closed due to lack of staff. There are 10 districts with no MOH
doctor but with a non-government doctor, and 4 districts with none at all (Hozumi, 2003).
Surpluses against the establishment are in the lowest nonprofessional grades. There is more than
double the establishment of patient attendants. This may signal inefficiency in the distribution of
the workforce (see below); on the other hand, this may be an attempt to compensate for the lack
of more qualified staff, e.g., the shortage of nurses on the wards in hospitals.

43.      Because of severe shortages in skilled health workers, some of the technical and
clinical work are being done by less-skilled workers, with potential ill effects on quality of

5
  So-called “savings” can be generated by forcing staff to take earned leaves (or in ext reme cases not
simp ly providing leave benefits for those entitled to them), or closing offices to save on utilities.


                                                     20
care especially with weak staff supervision. The total number of skilled and unskilled health
staff in the MOH and CHAM is about 20,000 or a population per worker ratio of 587. The overall
ratio is not extreme in comparison with other countries (e.g., Uganda 1:766; Sri Lanka 1:245).
(Hornby and Oczan, 2003). However, the proportion of skilled and semi-skilled staff (i.e., with
two years or more of training) in each of the workforce areas – clinical, nursing, preventive, and
technical services – is low (Table 5). The proportion of semi-skilled or unskilled health personnel
is particularly high among nursing staff, most of whom are hospital attendants or orderlies. Many
of these semi-skilled or unskilled support staff are now providing certain direct services to
patients, in addition to support services for which they are originally hired. For instance, a study
in one health facility found that 20 out of 34 had deliveries (58.8 percent) attended by ward
attendants (Ratsma and Ostergaard, 2003). No study has been done on the adverse consequences
of these practices.

             Table 5. Distribution of Skilled Staff by Cadre, MOH and CHAM, 2003
     Cadre                        MOHP                                  CHAM
                      % of Total          Within the        % of Total          Within the
                        Workers        Cadre, % Skilled       Workers        Cadre, % Skilled
Clinical                   7.6              100.0                10.3              100.0
Nursing                   43.9               37.9                80.7               43.4
Preventive                45.5                5.6                 0.4               75.0
Technical                  3.0               n.a.                 9.0               n.a.
Total                    100.0                                  100.0
           Source: Health Planning Services Department (2003), Table 8 and Table 12

44.      Preventive services also suffer from extreme skill scarcity, worsened by a
deteriorating supervision. Only 5.6 percent of staff within the preventive-services cadre at
MOH are skilled (i.e., with two years training). Majority of these staff are health surveillance
assistants (HSA) who receive minimal training of 10 weeks. Although this is not in itself
problematic if adequate supervision is provided, the massive resignation of experienced, senior
health staff is virtually decimating MOH‟s cadre of supervisors. In the past, HSAs have been
supervised by the Environmental Health Officer (with a current approximate ratio of 1 EHO
supervisor to 41 HSAs), but it is being suggested they should be supervised by the Community
Health Nurse (with a current approximate ratio of 1 CHN supervisor to 25 HSAs). However,
most CHNs have returned to their original jobs prior to training and are, therefore, no longer in a
position to provide supervisory support to HSAs.

45.      The staff shortage is also perpetuating the health sector’s bias towards tertiary care .
Like in other poor countries, health staff have tended to gravitate around higher-level facilities,
which are typically located in urban areas where housing, educational, and cultural amenities are
better. The massive resignations have only made matters worse, since those who leave first
usually come from worse-off areas where the need for services is most acute. Today, a high
proportion of Malawian skilled staff are deployed at the tertiary level. For example, of the 103
doctors working in MOH facilities, 78.6 percent (81 doctors in all) are posted in the four central
hospitals. Some 46.0 percent of the combined MOH nursing staff also work at this level. This has
left secondary- and primary-level institutions, which are supposed to serve the majority of the
population, short of professiona l staff, resulting in severely reduced amount and quality of
services that can be provided. In turn, this scarcity of staff at peripheral facilities, encourages
rural residents to bypass the referral system and to seek services at the comparatively better-
staffed tertiary facilities.



                                                 21
46.     Curiously, the “creaming off” of health staff at MOH, and to a lesser extent CHAM,
has also left tertiary facilities without sufficient medical and nursing staff for them to
operate as referral hospitals. The surgery wing of the Lilongwe Central Hospital, which is
supposed to be a referral hospital, may close once the only surgeon retires and a replacement is
not found. In addition, insufficient number of nurses has made post-operative care infeasible to
maintain. The 970-bed facility only has 169 nurses left, and its 6 laboratory technicians are doing
the work of 38 once employed there (Burkhalter, 2004). This and other referral hospitals‟
experienced nurses and technical staff have gone to work for better-paying donor-funded NGOs
mostly providing antiretroviral drugs, or for hospitals abroad. Thus, the so-called tertiary facilities
in Malawi are, for the most part, just district hospitals located in the city.

B. Causes of the Human Resource Crisis

47.      The HR crisis arose primarily from the Government’s inability to raise the real
value of salaries of civil servants and health workers in government-subvented CHAM
facilities. MOH employs the majority of health workers in Malawi; it is estimated that a total of
13,741 workers are in MOH payroll (or 69.0 percent of total health workers). An additional
1,277 professional staff (the number of support staff is unknown) work for CHAM whose salaries
are supported directly by a subvention grant from the Treasury (O‟Carrol, 2004). Throughout
most of the 1990s and the early 2000s, the Malawi government has been in a fiscal crunch, and
because of the sheer number of workers, the Government has been unable to raise salaries across
the board. In the past, government response to this issue has been to increase staff allowances
(professional allowance, housing allowance), such that today as much as 71.9 percent of a senior
nurse‟s take-home earnings are in the form of allowances. (As late as 2002/03, 51.2 percent of
total MOH personal emoluments were in the form of allowances, and only 48.8 percent were
salaries and wages.) The government has also implemented mass promotion of certain levels of
staff. For its part, CHAM often provides salary top-ups using own-generated user-fee revenues.
While both the Government and CHAM have tried a nominal “catch-up” strategy with respect to
staff earnings, inflation simply made matters worse, with many health workers feeling the real
value of their earnings being eroded year by year.

48.      More seriously, some me thods of increasing remuneration have had a negative
impact on the wider organizational objectives. One form of allowance for higher-level staff is
the provision of a fuel allowance, which is funded from the ORT budget and therefore reduces
limited funds for running programs. Mass promotion has also been resorted to, but this
effectively removes a grade from the pay structure, thus negatively impacting on the normal
motivational aspects of the pay structure. The use of allowances is more easily adjustable,
provides a short-term tax-free benefit to the individual (though deprives the revenue authority of
much needed income), but reduces the potential benefits from pensions which are calculated on
the basis of salaries, not allowances. In any case, most pay reforms advocate the consolidation of
allowances into the salary.

49.       The human resource crisis in government and CHAM health facilities has not been
helped by outright poaching of workers by donor-funded health projects. While the
government is constrained in adjusting salaries, secular NGOs funded by donors are able to adjust
salaries in accordance with what the market can bear. With demand outstripping supply of
trained health workers, the private sector and non-mission NGOs actively gobbled up civil
servants wanting better salaries. (Annex __ provides comparison of salary scales among
institutions that employ health personnel in Malawi.) Since there is very little net production of
new health workers, the NGO/private sector has been caught up in an escalation of salary levels



                                                  22
(Martineau, Sargent, et al., 2001). In one case, as many as 88 nurses from a government hospital
moved to a single NGO in a period of 18 months. It remains to be shown whether there is a net
addition to the amount of health services provided in Malawi by these secular NGOs, since the
“revolving door” simply results in services being effectively transferred from the government to
the NGO sector. It may be that as NGO employees, these same health workers in can provide
more and better services due to their better incentive packages and better working environment.

50.     Significant losses are also due to the increasing globalization of labor market. Using
requests for validation of registration by destination countries as a proxy indicator, the Malawi
Nurses and Midwives Council reports a steady stream of losses of staff overseas: 230 nurses in
2000/01; 103 in 2000/02; and 108 in 2002/03. The total for these three years represents 3 to 4
percent of all nurses providing services across the sector; this trend is continuing and is not likely
to end soon, given the demand in Western countries. The majority of nurses go to the U.K. (83 in
2002 and 90 in 2003). The degreed nurses are recognized by the U.K.‟s Nursing and Midwifery
Council, and are therefore in greatest demand; these are also the most skilled in Malawi, which
means “creaming off” the leadership, management, and technical specialization of the country
that can ill-afford to lose them. (The less-skilled Nurse Technician qualification is not recognized
by the U.K., and is not in demand). In contrast, the retention of doctors has so far been relatively
good with 112 of the 168 graduates so far (66 percent) working in Malawi and a further 43
attending postgraduate training outside the country expected to return (Broadhead and Muula,
2002).

51.      As the MOH and CHAM facilities are losing unprecedented numbers of staff, the
government has not been able to dramatically expand the production of these workers .
There are four major reasons for this failure. Firstly, the number of school leavers sufficiently
qualified and willing to enter the health professions is falling. The throughput of the different
school levels has been erratic and is not rising significantly (secondary school leavers were only
around 30,000 in 2000) and the pass rate is low but improving (13% in 2000; 34% in 2002) (S.V.
Chandimba, November 2003). Only a small number of students pass the Malawi School Leaving
Certificate (MSLC) in biology and general sciences, especially among girls. As this is the same
qualification for entry into university, a dwindling proportion of these students are keen to join
nursing. Consequently, many courses are unable to attract enough applicants and are run well
below their full capacity. The College of Medicine was unable to attract enough sufficiently
qualified school leavers to fill its intake of 20 students in 2001. In response, it established a one-
year premedical course, taking students at MSLC level and providing them with the necessary
science teaching to enable them to enter medical course. Without this course, the College of
Medicine would not have been able to expand the number of medical students to the current level
of 60, as requested by the MOH.

52.      Secondly, the government, which is the major funder of pre -service training for
medical, nursing, and related professions, has not been able to provide an adequate and
predictable stream of resources for pre -service training. Training institutions are
intermittently subjected to closure due to lack of funds. The flow of funds for the Emergency
Six-Year Training Plan has been uneven, jeopardizing the planned increase in outputs. In recent
months, the Treasury has not provided the HIPC funds to the College of Nursing, forcing it to
remain closed when the new intake should have started in October 2003. Though the College of
Medicine has not been regularly receiving its subvention of MK17 million per month, it has only
managed to run its courses partly by temporarily diverting funds from other sources.

53.    Thirdly, the ripple effect of mass resignations in the health sector is the continuing
shortage of tutors, especially for pre -service training of nurses. To retain existing tutors, a


                                                  23
few donors have committed to fund the salary top-ups of nurse tutors in CHAM nursing schools.
About US$50 per month is provided for each tutor, in addition to the basic salary of US$60 per
month. But this additional funding will cease in June 2004. To continue the scheme for a further
three years, an estimated US$2.135 million is required (MOH, 2004).

54.     Fourthly, almost all donors have been averse to funding pre -service training,
preferring instead to devote their resources to in-service training. Donors‟ obsession with
achieving measurable health outcomes in the quickest possible way has led to health programs
that focus only on actual service delivery, and to the relegation of longer-term investments that do
not directly manifest results in terms of actual health services (e.g., formal education of health
workers). In the late 1990s, as much as 10 percent of the US$47 to US$50 million annual donor
expenditures in Malawi‟s health sector went to training, most of it for in-service training
workshops for specific disease interventions such as reproductive health, child health, and
HIV/AIDS. Most of these are patchy and poorly coordinated programs that took health workers
away from their work, often trained the same set of personnel, and did not add any new stock of
workers. Many donors claim that tertiary education is not their purview, and that there is little
assurance that professionals whose tertiary training was subsidized will remain in-country.

55.      The policy preference for the production of degree -level nurses rather than
community health nurses (auxiliary or diploma nurses) has not served Malawi well.
Advocates for this policy, who mainly came from the Nursing and Midwifery Council, couched
their support in terms of the need to provide Malawians with the best care possible, and the need
to nurture future leaders in the nursing profession. Thus, in the early 1990s the government, with
the urging of the Nursing and Midwifery Council, abolished the enrolled nursing program and
focused instead on the production of professional nurses. The higher entry requirements in the
professional nursing program meant that only a few qualified. In addition, the higher out-of-
pocket expenses incurred during the longer training period (longer by one year than the enrolled
nursing program) dissuaded many potential candidates, considering the low salaries they could
hope to earn locally. Malawi began to feel the backlash of this policy in the late 1990s as the
nursing profession became globalized, and Western countries began large-scale importation of
highly valued professiona l nurses. The professional nurses who were trained (at high government
subsidy) began to stream out of the country, while the auxiliary nurses who should have been
trained and could be retained since they are not recognized as a profession abroad, became fewer.
The Council‟s policy posture has changed in recent months, and agreement has been reached with
MOH as major employer that auxiliary nurses would be trained to work on the hospital wards. It
has also been agreed that the College of Nursing will continue to train degree-level nurses (65 per
year), but a diploma level course will be run by the Malawi College of Health Sciences (40 per
year).

56.      Nursing Auxiliaries, a new category of nursing support staff, were introduced in late
2003 (and are therefore not included in the staffing data). The first batch has been recruited
already and the POW shows plans to recruit an additional 560 Nursing Auxiliaries a year. They
will be largely trained on-the-job. While the creation of this new category is an understandable
response to severe staff shortages, it further swings the balance towards a semi-skilled on skilled
workforce. Like HSAs, Nursing Auxiliaries will need supervisory support from senior staff to be
more effective than a patient attendant. Currently there is little or no surplus capacity on the
wards to provide this support.

57.      Staff morale has been worsened by lack of complementary inputs and increasing
workload. Shortage of drugs, supplies, apparata, and equipment to work with are common, and
lack of housing and other amenities in rural areas persist. These appear to be major “push”


                                                24
factors for those already posted to rural areas, and act as a major deterrent to those who might be.
In addition, as their colleagues die, get sick, or leave, the work has to be apportioned to fewer
remaining staff. In the face of stagnant salaries, these factors conspire to lower staff morale.

58.      The administrative systems in the public sector have not been responsive to current
needs and further exacerbate staff shortages. A cross-ministry study undertaken by
DHRMD/UNDP (July 2002) found that the duration of vacancies range from a few months to
five years; majority of the technical and professional vacancies last for two years. Health and
education sectors have been exempted from the current recruitment freeze, yet serious delays in
filling vacancies are often reported (Martineau, Sargent, et al., 2001). Delays in getting new staff
on the payroll leads not only to serious demoralization but also to losses of new recruits, as these
recruits are unable to earn a salary while they are waiting. Re-engaging retirees is a lengthy
process and may lead to an unproductive gap between the date of their retirement and the issue of
a new short-term month-to-month contract. Backlogs of promotions are frequently reported.
There have been recent attempts to reduce the delays and the expectation is that the new Health
Service Commission will streamline some of the administrative processes. While it may take a
long time it is possible to replace someone who has either resigned or died, there is currently no
policy to replace someone with long-term illness who is unable to work (Kathyola, 2003).

59.      Poor government response to the crisis may be due to the highly fragmented HR
planning, management and development functions and the severe understaffing of relevant
HR units. Across the MOH, the major actors include HRMD (which covers human resource
management, development and management analysis), the Planning Directorate which until
recently housed the only HR planner, and other directors who are involved in appointment,
deployment and promotion of relevant staff within their directorates. In addition, a Human
Resource Advisory Committee, chaired by the Principal Secretary, oversees human resource
issues within the sector. The PoW also proposes to establish a separate liaison office to link the
MOH to the training institutions. HRMD is seriously understaffed (for example, there is only one
human resource development officer at PO level to look after training issues). The Planning
Directorate has only one junior human resource planner and this post recently became vacant.
The Human Resource Advisory Committee lacks any form of secretariat to support it. Some
overlaps exist in the roles of the various departments working on human resources, including the
collection and maintenance of staffing data.

60.      The important areas of employee relations and the management of change have
largely been neglected. In the midst of a tight labor market further depressed by low salaries and
high losses due to death and resignations, the health sector is facing unprecedented change with
the introduction of the EHP, hospital autonomy, devolution and a sector-wide approach. It is at
this time, more than ever, that workers‟ issues are paramount. Yet, the MOH has minimal
expertise in employee relations and change management either at headquarters or at facility
management level. For their part, the unions and professional associations are weak. This
weakness on both sides of the labor market means that communication and meaningful dialogue
between management and staff is difficult. The result of this may be manifested in low morale or
mild resistance to change, both of which will remain largely hidden. However, if the situation
worsens it may develop into industrial action, as happened in a central hospital in 2001 (Muula &
Phiri, 2003) and more frequently in neighboring countries.

61.    Staff performance monitoring has been uneven. Although staff performance
management systems are in place in the civil service, these are acknowledged to be weak. Little
evidence exists on their effective use at hospital, district and health facility levels (Valentine,
2003). The introduction of a new system across the civil service has been planned (GOM &


                                                 25
UNDP, 2002), though it has been severely delayed. The government also introduced the
Performance Contract Scheme which entitles staff at P4 level and above the option of moving to a
three-year fixed-term contract renewable subject to adequate performance in return for a
substantially higher rewards package. However, the performance benchmarks were not put in
place at the time of initiating the contracts and as less than 1 percent failed to get their contracts
renewed, this initiative is seen to have failed the “leadership by example” test (Valentine, 2003),
making it difficult subsequently to introduce a formal system for the lower ranks. Some NGO
and private health care providers do have effective performance management systems in place. A
case in point is the output-targeting per staff and results-oriented management employed by
BLM, a local NGO.

62.      Supervisory support has been weak in certain key areas. Many staff, particularly
those working in the periphery, are expected to work largely independently and often provide a
wide range of services. For this reason, they need effective supervisory support for purposes of
both quality assurance and motivation. This is particularly important for the least trained groups
of staff, especially the HSA (Hornby and Ozcan, 2003). (There are exceptions to the weak
supervisory support, including the TB Control Program and the Essential Laboratory Medical
Services, both of which are performing well.) Since the abolition of the Regional Offices in
1999, there have been problems with providing technical supervision to districts (Ager, Chinula,
et al, 2003). With the establishment of the Zonal Offices included in PoW, this gap may be
closed. Still, supervisory systems at sub-district level will need to be developed in line with the
EHP.




                                                 26
                IV.         Options on Human Resources for Health

63.      The health workforce challenges in Malawi are daunting. These include major changes
in service delivery including the EHP and programs to address HIV/AIDS; increasing the stock of
new health workers through a much-expanded training program; getting health workers back into
the system and stemming further losses; ensuring more equitable distribution of staff in a
dynamic and uncertain policy context which includes devolution and hospital autonomy; and
strengthening the technical and policy-formulation capacity within MOH to develop appropriate
human resource strategies to meet these challenges.

64.       The options presented in this chapter are largely based on HR contents of the Program of
Work. Being already in the PoW, these elements are owned by the government and will have a
higher probability of being implemented. The chapter provides further elaboration for some areas
of the PoW, but also presents additional options that the Government can consider in its menu of
available policy tools. Many of these options are not new; interviews with key stakeholders
showed that some of them, radical they may be, are being entertained by HR advocates and
decision makers. The chapter is organized as follows: options for increasing the production of
health workers are discussed in Section A; for attracting and deploying health workers in Section
B; for retaining health workers in Section C; for strengthening HRMD systems in Section D; and
finally, for improving HRMD planning and policymaking in Section E. In spite of the difficult
conditions under which health staff work and the difficulty of providing adequate support, many
of the remaining civil servants do work surprisingly diligently. This level of motivation and
commitment should be seen as an asset to build upon in considering these various options.

A. Options for Increasing the Production of Health Workers 6

65.      The Emergency Training Program needs to be managed well to get new entrants
into the labor force . In 2001, the government formulated the Emergency Training Plan covering
nursing, paramedics and lower-level staff (HSAs and now auxiliary nurses). It is expected that
over 15,000 health workers will be produced by the end of the plan period, of which nearly 2,000
are existing staff to receive upgrading courses. Except for registered nurses and laboratory and
radiography staff, the training program will bring the staffing levels up to those of the current
establishment, assuming attrition levels do not increase (Table 6). The Malawi College of Health
Sciences will start a new diploma course for training registered nurses during 2004, although the
first batch will not be available before 2007/08. The number of HSAs is set to double and,
combined with over 3,000 auxiliary nurses, this group of 12,000 providers will vastly outnumber
the higher skilled staff. Although this will potentially have a significant impact on health service
coverage, it will pose challenges to staff supervision.




6
  This section should, but does not, exp lore the possibility of private colleges or universities train ing
doctors and nurses. Explo ring this issue requires more resources and time than those allocated for this
study. Historically, nursing and medical education in s mall Southern African countries (includ ing Malawi,
Mozambique, Zamb ia, and Zimbabwe) have been the purview of the government. However, private
initiat ives in this area are sprouting in Livingstonia, Malawi and Beira, Mo zambique, with demand driven
by mission facilities‟ need for health workers that the government training institutions cannot meet.


                                                    27
       Table 6. Expected Training Outputs Under the Program of Work, 2003/04-2009/10
    Cadres          Training     Current       % of       New      Current        % of
                    Output        Stock       Current     Stock Establishment Establishment
                    by EOP       (MOH +        Stock
                                 CHAM)
Medical officers        126         84           150.1      210           129               163
Clinical officers       450        504            89.3      954           686               139
Medical
assistants              750          439         170.9      1189          742               160
Nursing
(combined)             3950          4769         82.8      8719         7899               110
Registered
nurse                   360          200         180.4      560           883               63
Nurse
technician             2690
Auxiliary nurse        3360
Pharmacy
related                 120          144          83.3      264           225               117
Lab related             240          149         161.1      389           373               104
Radiography
related                 120           43         278.7      163           235               69
Therapists               60
Environmental
Health Officers         180
Asst
environmental
health officer          120
HSAs                   4800          4324        111.0      9124         4324               211
   Note: Neither attrition during training nor in service has been included here. Attrition during
   training may be between 6 – 10% (see MOHP, 1999) and in service up to 4.5 per annum (see
                    Hornby and Oczan, 2003). Source: adapted from the POW


66.      Additional funds need to be tapped to finance the PoW’s HR component, especially
the Emergency Training Program. All in all, the HR component is expected to cost US$247.7
million over six years. Of this, US$55.2 million will be for the pre-service training of nurses. So
far, additional resources amounting to US$27 million have been secured for the first priority
section of this program, partially from HIPC funds. The government anticipates resources from
the pooled funding of donors under the SWAp to support this program. To ensure that this plan is
fully implemented, funding commitments need to be secured for the courses, tutors‟ top-ups, and
rolling the program forward for the entire horizon of the PoW. Additional funds can be tapped if
the Global Funds relaxes its expenditures-eligibility rules so that they can be used for pre-service
and in-service training, and for direct recurrent-cost support to civil servants. Indeed, the
guidelines on the use of Global Funds have been revised under Round IV specifically to allow
governments to use these funds for recurrent-cost support for civil servants. The gravity of the
HR crisis in Malawi was appreciated by senior UNAIDS and Global Fund officials who visited
Malawi recently. It remains to be seen whether indeed Malawi‟s Global Funds allocation can be
“grandfathered” to be used for civil servants. Additional financing is also needed for longer-term
needs, such as medical education, which is not included in the PoW.


                                                 28
67.      The in-service training for health workers needs to be integrated and coordinated.
A policy is being developed to integrate all in-service training courses to ensure that they are
cost-effective and support the delivery of EHP. The implementation of this policy can start
gradually by combining selected courses with a view to develop a more comprehensive system.
Over the medium-term, training programs will need to respond rapidly to any changes in staffing
policies. This may involve developing new courses, and offering training in different ways (e.g.,
distance learning). Training institutions will need to be flexible about accommodation, especially
if the gender mix of target groups change.

68.      The role of the professional councils as the regulator vis-a-vis the MOH as the
largest employer of health workers need to be clarified. The prolonged debates about the pros
and cons of pursuing degree-level versus diploma-level nurses, and whether auxiliary nurses
should be introduced has delayed improvements in service delivery at a time of crisis. Though
both issues have now been resolved, a better forum for debating such issues is needed to avert
similar policy gridlock in the future. With the current institutional arrangements, the HR
Advisory Committee seems the most appropriate forum, though this needs to have better
representation of all employers in the sector including the private sector and, in the future, local
governments. Such a forum should be taken into account in the proposed review of institutional
arrangements. During the preparation of the Health Sector Human Resources Plan in 1999, it was
recognized that the consequences of the closure of enrolled nurse training program without any
replacement had not been projected. The ability to provided fora like the HR Advisory
Committee with scenarios of different kinds of intervention is essential to support decision-
making.

69.      In the long-term, role of government as a monopoly producer/funder of trained
health workers should be reconsidered in light of the globalization of these skills. The large
public subsidies spent on the training of health workers lose their economic rationale since
immigrants tend to internalize the benefits of higher pay, with little direct benefits accruing to the
government (assuming away the foreign-exchange remittances, which may very well be
significant.) There are various options that the government can consider on this issue:

70.      Formal bonding arrangement - The government can adopt formal bonding arrangements
for persons trained under government-subsidized programs. At present, all university students
pay fees of MK25,000 per year, making it more difficult to justify bonding of doctors and degree-
level nurses. In addition, the government does not have a good track record of ensuring the
repayment of the bond if it is broken. The funding from the PoW should cover all training costs,
giving the MOH greater justification to bond trainees. In addition, there have been some
discussions about using private financial institutions to manage the bond. It is unclear how this
might work. One possible approach would be for the financial institution to provide a loan to the
student, and this loan would be called in at the end of the bond period. The MOH would pay it if
the individual had fulfilled the terms of the bond; otherwise the individual would be liable. Of
course, there would be a problem with collateral, particularly for the poorer students. Also, the
bond would have to be quite large to deter people from migrating as they would quickly be able
to pay the loan off from their higher earnings.

71.      Reduction of government subsidy – The public subsidy could be gradually reduced by
making training programs operate on a matriculation and tuition fee basis; running hostels on a
fee basis; accepting non-boarding students (this could very well work in urban areas such as
Blantyre and Lilongwe); and similar approaches. Where feasible, distance learning and similar
cost-effective models of teaching and learning should be adopted on a full-fee or partial subsidy
basis.


                                                  29
72.      Liaison with private sector – Government should encourage (and liaise with) the growing
number of private training institutions. Private education is expanding, for example, in the area of
management training. A private nursing school is being established in Livingstonia. MOH is
already “purchasing” training from government and CHAM institutions. MOH should ensure
that the departments dealing with HR have the necessary capacity to develop and manage training
contracts with the private sector. This should result in additional training output and might, by
developing some competition in the sector, lead to training provision that is more responsive to
the needs of the MOH and the sector as a whole. The capacity of the regulatory bodies may need
to be expanded in line with the growth of training providers.

73.       Mandatory government service - In some countries, there are schemes that either
encourage or compel new graduates to work in government service, often in rural areas, for a
fixed number of years dependent on the length of their training course. Incentives may be
provided which may include special financial allowances; the service may be a prerequisite for
gaining a license to practice as in Indonesia (Chomitz, et al., 1997) and Ecuador (Cavender and
Alban, 1998). The system is not always enforceable, as in Nepal (Weiner, 1989) but has been
effective in Thailand (Lexomboon, 2003) and Nigeria. South Africa recently introduced a
community service scheme, initially for doctors (Reid, 2001) and subsequently for other cadres,
though unlike the other schemes which tend to be around three years, the service period is only
one year. The advantages of the scheme include (i) the deployment of more staff in rural areas;
(ii) the frequent possibility that staff, who otherwise would not have chosen to work in these area,
decide to stay there for good upon completion of the service period; and (iii) more highly skilled
staff (doctors and dentists) get more practice in procedures that would normally be the preserve of
more senior staff. The disadvantages include: (i) the underserved areas get the least experienced
staff; (ii) at an important stage of a person‟s professional development, s/he may get insufficient
supervisory support; (iii) the potential danger that some people may leave or migrate to avoid the
compulsory service, negating the deployment objective altogether; and (iv) the required
administration places an added burden on the system, and in some cases, it may either be not
strong enough to cope with the number of staff to be monitored, or it may lack transparency, in
which case, the suspicion that not all graduates are being treated equally may lead to resentment
or even blatant non-compliance. If this were considered to be a viable option, important lessons
could be learned from the recent South African and Nigerian experiences.

74.      International instruments to manage migration - There is now more interest than ever at
an international and regional level to mitigate the damage to health systems caused by large-scale
professional migration. However, some of the instruments such as codes of practice on
international recruitment may not have a lot of impact (Willetts and Martineau, 2003; Buchan,
2003) and numbers of Malawian nurses registering in the UK has increased in spite of the NHS
code of practice on international recruitment. Given the extreme differential between Malawian
salaries and those in industrialized countries, the outflow is likely to continue. However, there is
evidence that poor working conditions, which are due to be addressed in the PoW, act as a push
factor. There may also be people who would like to work overseas for a short time, but would
like to retain their current career level and come back into the system. If this is the case, and
some research would be necessary to validate this assumption, bilateral agreements could be
sought with overseas employers for providing staff on fixed term contracts (as has been done
between South African and UK institutions) and corresponding flexibility in civil service rules to
allow re-entry at an appropriate level. This kind of “managed migration” is risky, but may satisfy
the needs of the would-be migrants and the overseas employers. There may also be benefits in
bringing back additional expertise, though only if the PoW is successful in strengthening the
health system sufficiently for this to be used. If there are no promising solutions to the migration



                                                 30
problem, the MoH will seriously have to consider what in terms of staffing structures is likely to
provide the best return on investment in training.

B. Options for Attracting and Deploying Health Workers

75.      In the immediate term, a concerted recruitment campaign should ensure that when
vacancies occur, information is rapidly transmitted to potential applicants. It will also be
necessary to ensure that potential recruits are not lost due to delays in posting, getting them on the
payroll, etc. This will require a review of current complex recruitment systems (see Annex xx
{the chart}) to identify bottlenecks and the development of appropriate solutions to speed up the
process. It is unclear what the cause of the bottlenecks is. There are a number of possibilities:
inefficient bureaucratic systems with no performance standards for how long a process should
take; high workload with few staff to process the requests; difficulty in accessing data for the
decision-making process e.g. verifying sanctioned posts; or as in some countries, the delays could
be a means of saving on staffing costs. As each application is processed by different offices (see
diagram in Annex C) the decision at each step is likely to be dated. A small survey of completed
applications should quickly reveal where the main bottlenecks are. Solutions will depend on the
type of bottleneck and the location. If it outside the Ministry of Health, which is the case for 7 of
the 11 steps in the diagram, negotiation is likely to be necessary at a level higher than that of
directorate. The advocacy message should be that removing the bottlenecks is a virtually cost-
free method of supplying much-needed staff to the health sector. The improved systems would
then need to be monitored closely – probably by HRMD with reports given to the PS or HR
Advisory Committee, and readjusted if necessary. In the medium-term, there is a need to develop
clear guidelines on recruitment schedules and practices. A major purpose of these guidelines will
be to streamline the recruitment processes for faster replacement of vacancies and to make sure
that undue delays do not lead to the loss of selected candidates. This kind of problem analysis
may be redundant when the Health Services Commission assumes all aspects of filling vacancies.

76.      MOH can attract back health workers who have left if it could offer better-than-
civil-service packages through labor agency contracts. The size of the pool of trained health
workers who are not working in the health sector is not known. However, recent estimates by the
Nursing and Midwives Council, based on numbers of nurses trained and numbers who have left
the country, put the figure at between 800 and 1,200 (MOH, 2004). This is a significant stable of
still-available health workers. It is clear that by “voting with their feet”, these workers are
dissatisfied with the status quo, and are not likely to return unless the incentives are right. Under
the current civil service rules, such incentives may not be available. To get around unwieldy and
inflexible civil rules, the MOH could consider tapping the services of these experienced health
workers through a professional recruitment agency under time-bound performance-based
contracts. The feasibility of these performance-based HR contracts for rural health centers needs
to be explored, especially as many centers do not currently meet minimum staffing needs for the
EHP. This could be based on the following principles: it is short-term; addresses areas of greatest
staffing shortage; contracts include attractive rates of pay; efficiently managed – probably by an
outside agency; scheme is stopped when adequate health workers can be recruited (and
subsequently retained) through the normal channels. Further details of a short-term staffing
supplementary staffing scheme are given in Annex D.

77.      HR recruitment and deployment strategy should take account of the devolution of
health services and the planned granting of autonomy to central hospitals . The time-frame
for the implementation of these two initiatives is still unclear. Since many countries have
experienced HR-related problems related to decentralization, a checklist for the transition is



                                                 31
needed as soon as possible. If this has been prepared by the decentralization secretariat, the MOH
should ensure that it covers the following items:

        Clear decision on which HR functions should be devolved (e.g., hiring and firing,
         promotion, payroll management); HR structures at headquarters and hospitals and
         districts to be developed accordingly

        Mechanisms in place to ensure that the current inequitable staffing situation is not
         exacerbated by the devolution, as happened in Papua New Guinea and Tanzania
         (Kolehmainen-Aitken, 1998). Attracting and even simply letting people know about
         vacancies may be challenging for local assemblies especially the poorer and remoter
         ones.

        Lessons from countries like the Philippines and Zambia have shown that in the shift of
         employment arrangements, it is essential to ensure that staff are not going to lose out by
         moving to lower pay-scales, or losing particular benefits enjoyed as health sector
         employees, or losing accumulated retirement benefits (pension and/or terminal
         benefits). It is not sufficient to state in principle that there will be no losses; these
         statements must be backed up by hard budget data.

        Under devolution, staff will be concerned about the future of their employment. It is
         essential that a communication strategy be put in place to keep staff reliably informed
         about plans and developments regarding the future of their employment. Trade unions
         an/or professional councils should be involved. The communication strategy for the
         SWAp (Harris and Nyondo, 2004) could be used as a model for adaptation.

78.      In light of the expansion of the private sector, the role of NGOs as users but not
producers of health workers need to be re -examined. Among staff, the highest proportions
working in the non-government sector are pharmacists (92 percent), lab technicians (49 percent),
enrolled nurse/midwives and nurse technicians (46 percent), and pharmacy-related staff (45
percent) (Hornby and Oczan, 2003). Enticement of civil servants especially by well-funded
secular NGOs (including research projects) should be better dealt with. Since many of these
NGOs are “off-budget” activities, their expansion cannot be predicted with any certainty, and
planning for their HR requirements is extremely difficult. These NGOs do not produce new stock
of health workers (except on the rare occasion that they bring their own service providers), and
must rely on available stock, i.e., those in government service or who have left it. Thus, secular
NGOs are for the most part free-loaders; they are also known to artificially inflate labor rates,
creating instability in the system. There are various options that the government can consider to
deal with these problems:

        (a)     Develop a standard of good NGO practice in HR recruitment, deployment and
                use. This should include making explicit the impact of the NGO project on the
                current pool of trained health professionals in initial negotiations with relevant
                authorities.

        (b)     Encourage these NGOs to bring their own workers. Although there has not been
                a formal study on the trend of expatriate health workers in missionary facilities,
                observers have noted the declining number of missionary health workers going to
                Africa in general, compared to the influx in the 1960s and 1970s.




                                                32
        (c)     Consider the possibility of NGOs reimbursing the cost of training of these civil
                servants. This principle can also be applied to countries importing Malawian
                health workers.

        (d)     Negotiate with donors funding these NGOs to set aside a proportion of their
                resources for the pre-service training of health workers specifically to meet these
                NGOs‟ HR requirements.

        (e)     Appeal to donors to extend the length of NGO grants (typically 1.5 to 3 years) so
                that they can be allowed to address longer-term HR issues. Such grants should
                include specific deliverables on human resource development, not just health
                service coverage.

79.      As a stop-gap measure, expatriates should be encouraged to fill highly-skilled
disciplines, and should be attracted and managed better. Some of the loss of national skills is
compensated for by the use of expatriates, especially in the districts where they fill a number of
the most highly-skilled posts. In 1998, 15 of the 23 of medical specialists working in Malawi (or
65 percent) were expatriates (MOHP HR Plan, 1999); the majority of heads of department in the
largest hospital, Queen Elizabeth Central Hospital, are expatriates employed by the College of
Medicine (Koot and Martineau, 2004). At a slightly lower level, an estimated 36 Peace Corps
Volunteers and four Japanese Volunteers (Anonymous, 2003) work in the health sector and
another 17 U.N. Volunteers will arrive shortly. The use of expatriates is a helpful short-term
measure, but foreign volunteers involve high staff turnover and, as strangers to the system and the
local culture, they may take longer to get sufficiently up to speed. The PoW proposes the
financing of a flexible technical assistance fund for the use of international volunteers and
contract staff to fill specialist posts at central and district hospitals; some (17) new volunteers
have already been identified, but more will need to be recruited.

C. Options for Retaining Health Workers

Recent HR Innovations Catering to Specific Groups

80.      In response to the HR crisis, the government has introduced a number of innovations
designed to retain health workers in government. These include (a) senior-level MOH
performance contracts; (b) sessional locums (defined below) as a stop-gap measure in hospitals;
(c) various innovative HR retention schemes employed by districts; and (d) outsourcing non-core
MOH functions.

81.      Performance-based contracts were introduced for senior staff (Grade S4/P4 – the
level of a Director – and above) in 1999. MOH staff who opt for this scheme must retire from
government though they maintain their accumulated pension rights and are employed on a 3-year
fixed-term contract renewable on achievement of a satisfactory level of performance. The
remuneration levels on the new scheme are substantially higher than the equivalent permanent
and pensionable levels. Monthly salary and allowances for S4/P4 not on contract are MK29,522;
for someone at the same level on contract, the total is MK77,405 with the addition of a fuel
allowance of 600 liters (MOH, 2004). Whereas the increase in compensation through promotion
from P6 to P5 is 4.6 percent, from S5/P5 to S4/P4 it is 555 percent, or, as the Valentine Report
(2003) states, "almost akin to winning the lottery." The costs of this scheme are not known for the
health sector, though there are a least more than a dozen people (16 in 1998) who are eligible.
Overall, this compensation group comprises only 0.4 percent of the civil service employment, yet


                                                33
it receives about 6.8 percent of the total Malawi Civil Service wage bill – enough to pay salaries
for 28,000 personnel in the lowest compensation group (average MK2,834 per month – a little
more that the basic salary of an Enrolled Nurse) (Valentine, 2003). The performance criteria
were not set at the beginning of the scheme, but a review is supposed to be carried out annually
by the supervisor. Of the 595 people whose contracts were due for renewal, only three were not
appointed and this is said to have been for reasons other than performance. DHRMD needs to
review the scheme, especially since it has the potential of undermining leadership. While the
scheme may not have led to demonstrably increased performance of individuals, it would be
interesting to find out whether it improved attraction and retention of staff.

82.     Following a civil-service-wide review of pay, the government has also introduced a
Medium Term Pay Policy (MTPP) focusing on mid- and lower-level staff, and the Selective
Accelerated Salary Enhancement (SASE) scheme focusing on priority groups. Instead of
boosting take home pay by increasing allowances, the SASE will increase overall salary levels
and consolidate existing allowances into the salary. Ministries may apply to join the scheme on
condition that they meet certain criteria such as availability of a strategic plan; a client charter;
cover positions key to the provision of service delivery; reform or strategic government output;
and inclusion of performance agreements for staff in SASE scheme positions. The MOH is using
the services of a health economist to assist the ministry in joining the SASE scheme. (The SASE
scheme has been in place for several years in Tanzania with unclear results). The management of
the SASE by the newly created Remuneration Board in the Office of the President and Cabinet
(and future salary increases under the Medium-Term Pay Policy) supersedes the powers granted
to the Health Services Commission (Staple, May 2004). This will enable the Health Services
Commission to concentrate on streamlining the administrative procedures of recruitment and
promotion.

83.      Hospitals are beginning to use locum arrangements on a sessional basis, either using
their own staff after they have completed their weekly allocated hours, or hiring trained
staff from the outside to address current staffing problems. Policy guidance is needed for
managers to implement this scheme more widely. The cost-effectiveness of this arrangement also
needs to be examined, and the contribution in terms of full-time equivalent (FTE) monitored
more carefully. Since the scheme is largely financed from fee revenues generated by hospitals
and relies largely on the existing staffing complement, the scheme‟s impact on relieving the
severe staffing shortage in hospitals may not be significant. Providing additional funding and
using an agency to manage the locums scheme may increase the overall FTEs available.

84.      One district creatively uses allowances to entice health workers to provide
temporary cover to currently un-staffed health centers. It was reported that one district has a
number of health centers that had been closed due to lack of staff. Unable to recruit more staff,
the management adopted a creative use of allowances to entice staff to work there temporarily.
The current rule is that if a staff stays in one place for 21 days or more, s/he does not get a night
allowance of MK9,060 per day. However, the DHO provides some allowances and nurses agree
to work for the rest of the three-month period without extra allowances. At the end of the relief
period, another volunteer takes her place. Three health centers in one district are being
successfully covered in this way, and the scheme appears to be popular amongst staff.

85.     Further outsourcing of “non-core” functions should be considered, based on
evaluation findings. In 1999, the Office of the President and Cabinet, with technical support
from the Public Sector Change Management Agency, embarked on the contracting out of “non-
core” services in central hospitals, including cleaning, transport, building and ground
maintenance, laundry, security, catering, audit, mortuary, and ambulance services. Maintenance


                                                 34
contracts could also be considered, in lieu of full-time salaried maintenance staff. Thorough
evaluation of the outsourcing experience is needed to inform whether such a scheme can be
expanded further. The evaluation should include discussion of the initial issues faced, including
the appropriate cost of contracted services; measuring the performance of contractors; and
absorption of civil servants into contractor activities.

86.       CHAM uses fee revenues to top-up health workers’ salaries. CHAM pays its staff
according to government salary rates, though it employs technical assistant (TA) level staff (e.g.
nurse technicians) at technical officer (TO) salary level, a salary difference of only MK821, but a
more substantial difference of MK3,500 in housing allowance (MoH 2004). In fact, the top-ups
vary between CHAM institutions. One institution visited in 2001 said it pays a 50 percent top-up
to staff at STO level and above and that there was a special donor-funded scheme providing a
special package for doctors (Martineau, Sargent, et al., 2001). Another CHAM institution visited
at the same time did not have any funds to pay top-ups. In some institutions, non-financial
benefits including a good working environment and access to career development is said to be an
important retention factor. From interviews with both managers and staff, it appears that CHAM
institutions have greater flexibility than MOH, and its facility-level decisionmaking makes it
better informed about the staffing situation and needs.

Upgrading Salaries and Benefits and Improving Working Conditions

87.      While the HR innovations in the previous section were an attempt to improve the
performance of specific groups of workers, what is needed in Malawi is a generalized upgrade of
salaries and benefits and improvement in the career structure of Malawian health workers.
Improved working conditions should also include explicit consideration of gender issues,
providing protection for HIV/AIDS and related occupational health hazards, and recognizing and
monitoring dual practice.

88.      Salaries and allowances must be paid on time and in full, and remuneration must be
increased. The bottlenecks in payment salaries and allowances and getting newly recruited staff
on to the payroll need to be identified and fully documented (with information on number of
occurrences, length of delays, etc). Short-term and longer-term strategies for resolving the
bottlenecks need to be formulated together with the appropriate stakeholders. The system for
implementing the improved salary disbursement system then needs to be developed and
implemented. The Health Services Commission has been established with powers to "set salaries
and conditions of service" of staff within the MOH (GOM, 2002 – Health Services Act) as a
means of improving the attraction and retention of government health workers. This requires a
study of remuneration levels of competitors within the labor market for mid- and high-level
professionals; a review of the current salary structure; and the development of proposals for
competitive salary levels. This sequence, along with the approval process, may be lengthy. In
order to stem further staff losses, it may be necessary to introduce a temporary increase in
remuneration, possibly for certain groups of staff where the risks are highest, by increasing
allowances.

89.     Financing for the salary upgrades and top-ups must be secured. Although SWAp
financing can provide significant resources for salary upgrades and top-ups, it may not be enough
to meet the needed increments for all staff. Also, as part of the government‟s MTPP, there is a
strong move away from supplementing poor basic salaries with additional allowances. Thus, the
MOH needs to look at other possible sources of support. There are indications that other bilateral
donors not keen on providing pooled funding may be able to provide support through other
means, e.g., through NGOs or through the professional councils. The government and interested


                                                35
donors should consider alternative means of coursing HR support, e.g., through institutional
grants to the medical and nursing councils who will then distribute these to their own members,
using pre-defined criteria.

90.      Universal precautions for HIV/AIDS and other occupational health hazards should
be provided. Significant work-related risks of HIV transmission exist from harmful injection
practices and poor waste disposal management. Harmful practices are due to lack of supplies,
time, or motivation rather than lack of knowledge (Aitken and Kemp, 2003). All government
departments are required to set aside 2 percent of their budget on HIV/AIDS (Aitken and Kemp,
2003) which could support important work on HIV-related occupational health activities. Some
health care is considered as relatively high-risk by health care providers, and this is likely to be a
further “push” factor contributing to attrition.
91.
        The Malawi Government is considering the use of ARV for public sector workers.
There may be an argument for targeting health workers with ARV programs, in part to ensure
there are staff available to benefit from the huge investment in ARVs by the Global Fund. There
are known problems of providing ARVs in the workplace due to employees‟ concern of being
found out. Alternative mechanisms for providing ARVs to health workers may need to be
sought. These might include special facilities run by professional associations for their members.
The problem is that the current level of membership is low and some of the organizations, such as
the nursing association, are weak. However, with some support, the associations could provide a
useful service and might through this action increase the membership.
92.
        A clear career structure for the different health cadres should be operationalized.
Career structures are often seen, particularly where salaries are very low, as a means of increasing
pay. However, clear career structures also provide a mechanism to produce sufficient staff with
the necessary levels of skill and experience to undertake the higher-level functions in the
organization as posts fall vacant. In addition, the health sector is likely to need a more flexible
workforce (e.g., to adapt to new challenges such as the continuum of care required for dealing
with the HIV/AIDS epidemic), and current career paths in government will inevitably be affected
with the structural changes associated with hospital autonomy and devolution. With this in mind,
current career structures should be mapped out and reviewed to identify potential changes that
will both meet the needs of the individual and the changing organizational requirements.

93.     Specific attention should be devoted to gender management. Gender needs to be dealt
with more explicitly in HR planning and policymaking. A survey carried out in 1998 showed that
while females were well represented in the nursing jobs as expected, their number is greatly
reduced at the senior levels (PO/CTO/ CEO and above). More recent data based on a study of
each health facility identified all staff by sex (see Annex E). Most jobs are quite clearly divided
by sex: doctor, clinical officer, environmental health officer, and lab technicians and radiologists
are largely male, while nursing staff are largely female. HSAs have the best gender mix, though
the group is still dominated by males. A small proportion of males are reported to work as
midwives and a slightly higher proportion work as general nurses.

94.     The overwhelming number of women in the nursing profession calls for particularly
suitable working arrangements so that they remain in the service. As a matter of course,
these women nurses tend to follow the posting of their husbands; or they tend to stay in localities
where they find their partners. These societal factors should be taken into account in the design
of employment arrangements for these scarce health workers. The reasons why female
representation in the ranks from PO/CTO/CEO and above diminishes rapidly should be
investigated. If it is related to access to further training (which is often a prerequisite for


                                                  36
promotion), alternative means of training such as distance learning should be investigated. One
constraint on recruiting males into nurse training was lack of suitable accommodation
arrangements for mixed-sex groups. Male hostels are now being constructed in some training
institutions, which will allow for their increased training output. However, it will be important to
monitor the impact of increased male participation in nursing on service provision. A recent
review of the College of Medicine showed that only 16 percent of the staff were female (Koot
and Martineau, 2004); this reinforces the traditional perception that medicine is a male
occupational area. The review proposed more attention to equal opportunities in the selection and
promotion processes. A basic problem is that the routine staffing data is “gender blind”
(Standing, 2000); the current facility-level report format does not include information on gender.
The gender datum is included in the PPPIS, but this information is not used for HR analyses.

95.       Dual practice should be managed better. While the private-for-profit sector employs
only 3 percent of health professionals, its share of laboratory technicians (10 percent) and doctors
(9 percent) is significant (JICA, 2002). The private sector is also staffed by people formally
employed in the government sector through dual working arrangement which has been allowed
for doctors since 1988. Dual practice is a means of retaining staff on salaries that are lower than
the market rate. Thus, the staff gains from public sector employment (which provides access to
training, job security, status and in some cases access to patients who might be diverted to their
private practice), while the employer retains staff, albeit usually on a less than full-time basis.
This system is likely to work in favor of the public sector if rules are clearly known and
regulation is effective (Maqc and van Lerberge, 2000), two factors that are absent in Malawi.
Moreover, as earnings from private practice are higher in urban areas, postings in rural areas are
often resisted. Ghana is attempting to develop „intra-mural‟ private practice which offers more
control over the time spent by practitioners on private care and may also bring some financial
benefit to the hospital. Such service is also being started in Malawi‟s Queen Elizabeth Central
Hospital. While this may be a control on private practice, it is only likely to work in urban areas
where there are larger hospitals and a greater number of fee-paying clients. The challenge in
Malawi will probably be how to reduce the amount of time spent by health workers on private
practice while salaries – even under the MTPP – will only be rising slowly compared to the gains
from private practice. And as long as there is a tight labor market, employers are likely to have
little leverage on employees who spend excessive amounts of working time on private practice.

Improving the Career Structure of Health Workers

96.     {Paragraph from Tim Martineau.}

D. Strengthening HR Management Development Systems

97.      The existing systems for HR planning and management are not sufficient to deal with the
current staffing challenges in the health sector. These systems, which may have worked in a
stable environment, are cumbersome and unresponsive the rapidly changing labor market and
future requirements of the health system.

98.     Complete sector-wide data are needed for HR planning and management. The
current staffing information systems (the CS-wide Payroll, Pensions and Personnel Information
System or PPPIS, and the health sector staffing returns) fail to provide accurate and timely
information to support ongoing development and reviews of HR strategies (Schenck-Yglesias,
2003). However, recent survey material, e.g., the staffing survey of 1998 and JICA study
(Hozumi, 2003) have produced useful data on which to base planning exercises. The PPPIS has


                                                 37
been reviewed and a more effective replacement is being considered. CHAM has an effective HR
information system on staff employed in its facilities. However, there is currently no information
system to capture staffing in the private and NGO sub-sectors to provide a sectorwide picture. In
this regard, the recently developed information system operated by the Nursing and Midwifery
Council and the staffing indicators included in the HMIS provide useful information. Although a
number of specific studies have been commissioned on the impact of staffing of HIV/AIDS, few
studies have been commissioned in other areas of HR. The government plans to replace the
PPPIS with a more comprehensive system, but until this future system becomes workable, it may
be necessary to conduct an annual staffing situation survey. A labor market survey should also be
undertaken to inform the magnitudes of appropriate salary adjustments and top-ups, since the
numbers produced in the POW are rather arbitrary.

99.     A comprehensive review is needed to identify all the HR functions that need to be
carried out in the health sector. The review should be made in the context of the EHP, the
forthcoming devolution and hospital autonomy and the SWAp, and should involve the MOH, the
HSC, CHAM, and other key players. The review should then propose the most appropriate
structures for carrying out functions within the MOH and the HSC given the current HR
expertise, capacity and workload of the different sections of the Ministry. The current structures,
with more clearly specified roles and channels of communication, may be the most appropriate;
or new structures may be required. The review should include a staff development plan to
support the strengthened institutional arrangements. A draft TOR for such a study is included in
Annex F.

100. Communication links between management and staff should be established. At a
time of major change in the health sector, staff become concerned about the implications of the
changes for their own jobs. At the same time, expectations have been raised about salary
increases over the past years. Poor communication and unmet expectations are well-known
“push” factors for staff already considering leaving. At present, there are no clear communication
channels between senior management levels and frontline staff. These channels need to be
developed urgently, such as biannual meetings and newsletters. The communication strategy
being developed for the SWAp stakeholders might be a useful platform to search for an
appropriate model for communicating with staff. To improve communication between MOH and
the training institutions, a “liaison office” should be created within HRMD. To monitor training
and recruitment, this "liaison office" can be created by filling the two vacant posts with
appropriate staff.

101. Staff performance management systems should be developed immediately. Staff
productivity within the government health sector is not being assessed, although it is likely that
this concept is being introduced as part of the hospital autonomy program. Workforce
productivity can also be introduced with the workload-based staff planning. The staff
performance management system should be introduced as soon as possible, perhaps initiated with
the reckoning of staff absences as a first step, and moving on to performance appraisal and
strengthening supervision structures. More radical approaches should be considered gradually,
particularly for the rapidly expanding groups of lower-level staff such as Health Surveillance
Assistants and Auxiliary Nurses, and should link to wider drivers of performance at the
organizational level, including performance targets, quality assurance mechanisms, and client
satisfaction surveys.
102.
       Employee relations need to be managed better. Ways of improving dialogue between
management and staff in all of the sub-sectors need to be developed. Employers will need to
have expertise in employee/industrial relations. Likewise, in order to effectively attract sufficient


                                                 38
numbers of members and to adequately represent them, the unions and professional associations
will need to be strengthened. This role is not a natural one for employers, so leadership may need
to come from the HR Advisory Committee and civil society organizations.

E. Improving HR Strategic Planning and Policymaking

103.     Development and implementation of longer-term HR strategy should be made a
priority. Several recent attempts at developing long-term strategies for the health sector include
the Health Sector Human Resources Plan for 1999-2004 and the Human Resources for Health
Sector Strategic Plan 2003 to 2013. While the Emergency Training Plan came out of the former,
many areas of strategy remained unimplemented. The latter plan, which included important
detailed staff projections, was never completed and approved. The plan is now out of date and
will need to be revised (MOHP, 2002), possibly for the plan period 2004-2014. At this stage, an
“ideal” plan is not needed; given the current shortage of HR capacity in the MOH, a more
achievable target for the first year might simply be an updated Emergency Training Plan based on
current staffing data available, with a view to reviewing the Emergency Training Plan annually.
Remaining HR strategies to increase staffing numbers and performance are already included in
the POW and are being translated into annual action plans. A mechanism should be established
to review and refine the strategic plan on a regular – possibly annual – basis thereafter. Part of
this process could be built into the annual review of the POW.
104.
         HR planning parameters should be updated in the context of providing EHP and
using the results of a workload assessment. The AIDS epidemic has made previous HR
planning ratios and other parameters largely irrelevant. In addition, MOH‟s focus on providing
EHP using government, CHAM/NGO, and community partners requires a new approach of
planning HR requirements. It has been suggested that a workload analysis be undertaken to
identify optimum staffing levels by type of institution and location. The current roles of different
cadres should also be reviewed to make the best use of available staff, with particular attention
given to the largest groups such as the HSAs. Findings from the workload assessment should be
used to revise the policy on setting establishments. A longer-term view should be linked to the
development of a broader HR strategy. This will necessitate developing policy options on the
type of staffing that is appropriate for Malawi in the next 10-15 years taking into account both the
type of services that will be needed and a realistic view of the future labor market. The policy
options should consider addressing the current gender biases in the staffing of various cadres to
see whether changes could address issues of attraction, retention and staff distribution.

105.     A policy forum and framework should be developed for HR across the heath sector.
As yet, there is no forum in which such HR policy could be developed. The HR Advisory
Committee is the most obvious body to lead, but it has no representation of the private- for-profit
sector, or, with the exception of CHAM, of the NGO sector. HR policies for government health
workers tend to be those used for the civil service in general. The lack of flexibility – for
example, the lack of provision for part-time working – mitigates against creative solutions to the
current staffing problems. In general, there does not appear to be a dynamic environment for HR
policy development, with the exception of policy development related to HIV/AIDS and
employment, which has already been disseminated in the higher education sector. There are
expectations that the creation of the HSC will allow for a more dynamic policy development
environment, but this remains to be seen. As an initial effort in this area, an inventory of all
personnel management policies need to be made, and each reviewed regarding its impact on staff
attraction and retention and support for the EHP. An implementation audit using checklists
should be carried out for each policy, prioritizing those that are likely to affect staff attraction and



                                                   39
retention. Changes to policies should be recommended, and a system for monitoring the
implementation of policies implemented.

106.     Long-term technical assistance in HR management and planning should be
provided. MOH‟s technical and policy-formulation capacity in HR needs to be strengthened. It
has been suggested that such expertise would be extremely difficult to find even within Malawi.
To address this, technical assistance (TA) is needed to review of institutional arrangements for
HR planning and management, and to provide long-term assistance in implementing and refining
the short- and intermediate term HR strategies in the POW as well as assisting in the development
of longer-term HR strategies. The review of institutional arrangements will identify the most
suitable location for the long-term TA. Specialized inputs may be required for the development
of certain systems, e.g., HR information system and performance appraisal system. Since many
of the HR issues relate to other parts of government, the TA should have a recognized link to the
highest levels in the MOH, i.e., that of the Principal Secretary. The TA may be provided by a
single individual. However, given the workload in terms of volume, range of tasks, and expertise
needed, and the small number of sufficiently experienced people in this field likely to be
available, the option of employing a team including short- and long-term specialists should be
considered. More detailed suggestions for the TOR and options are given in Annex G.




                                               40
               V.        Status of the Financing of Health Services

107.    This chapter looks at the past trends in health financing in Malawi, with each of the
succeeding sections analyzing the patterns by specific sources of financing including the
government, donors, private institutions, and households. Problems and issues in each of these
sub-items are also discussed. The next chapter prospectively looks at possible financing needs
and requirements.

A. Overall Trends in Health Expenditures

108.    The level of health spending per capita in Malawi is generally low, in absolute terms
and relative to other countries in the region. Malawi‟s per capita health expenditures in
1998/99 was only MK489 (equivalent to US$12.40). Although this is higher than Mozambique‟s
US$8.40, it is much less than Zambia‟s US$20.60 and Kenya‟s US$29.60 per capita health
spending for the same reference year. To be sure, per person spending for health in East and
Southern Africa is already low by world standards where it is as high as US$2,443 in OECD
countries during the period. By 2001, Malawi‟s per capita health spending only marginally
increased to US$13.00 (WHO, 2003).

109.      The low level of per capita health spending is due mostly to limited public
(government and donors) expenditures in health, especially before the onset of debt relief
and global funding initiatives for health. In 1998/99, Malawi‟s health expenditures in the
public sector (inclusive of government resources and donor resources channeled through the
government) only accounted for 8.4 percent of total government expenditures, lower than the 9.8
percent average for sub-Saharan Africa, and nearly half the OECD figure of 15.1 percent. The
government accounts for only 25.1 percent of total health expenditures while donors provide 29.8
percent of the pie. In the midst of successive fiscal difficulties and inflation, the government has
tried its best to preserve per capita government health expenditures, which rose from MK21.5
(US$4.89) in 1993/94 to MK35.3 (US$6.26) in 1998/99 and has since been in slight decline to
around US$4-5 per capita (Figure 2).

110.     With limited public spending, the private sector is taking an increasingly bigger
share in health expenditures. Overall, the private sector (inclusive of institutional and
household spending) accounted for 44.9 percent of total health expenditures in 1998/99. Out of
this share, 26.0 percent came from households and 18.9 percent from corporate employers
through self-insuring work-based health programs or through third-party health insurance. More
recent estimates of private spending are not available.

111.      Given the extremely limited development of risk-pooling mechanisms in the formal
sector as well as in communities, Malawian households are increasingly resorting to
individual out-of-pocket payments to finance health services. In 1998/99, out-of-pocket
payments reached US$3.30 per capita, mostly in the form of fees to providers; direct payments
for purchases of over-the-counter drugs and health commodities that health facilities cannot
provide; and contributions to socially marketed commodities such as bednets, oral rehydration
salts, and contraceptives. As was shown in the previous chapter, there has also been a discernible
shift in health-seeking consultations towards private providers in the wake of the human resource
crisis in the public sector, making households bear an increasing burden of household out-of-
pocket payments. Transport to a health facility is also reportedly a major cost item to families,
though it is not included in household health spending as such.


                                                 41
    Figure 2. Per Capita Government Health Expenditures, in Malawi Kwacha, at Current and
                           Constant 1993 Prices, 1993-94 – 2001/02

                                  Source of basic data: Economic Report, various issues

                                                                                                                     411.6          423
    450
    400
    350
    300
    250                                                                                                220.7
                                                                            173.5          195
    200
    150                                                       102.7
    100                             70           71.9
           21.5        43.9
     50       21.5        34.8           30.6          19.8          26.2          35.3          28           25.2          37.3          32.6
      0
             4



                         5



                                     6



                                                   7



                                                                 8



                                                                               9



                                                                                             0



                                                                                                          1



                                                                                                                        2



                                                                                                                                      3
          /9



                      /9



                                  /9



                                                /9



                                                              /9



                                                                            /9



                                                                                          /0



                                                                                                       /0



                                                                                                                     /0



                                                                                                                                   /0
        93



                    94



                                95



                                              96



                                                            97



                                                                          98



                                                                                        99



                                                                                                     00



                                                                                                                   01



                                                                                                                                 02
     19



                 19



                             19



                                           19



                                                         19



                                                                       19



                                                                                     19



                                                                                                  20



                                                                                                                20



                                                                                                                              20
                                                Current price, MK            Constant price, MK


112.     Malawi’s health sector is small but growing modestly. The sector has grown from its
6 percent share of GDP in 1995 to 8 percent in 2000 (WHO World Report). The share of
government to total health expenditures has remained fairly stagnant at around 2-3 percent of
GDP, and much of the expansion in health spending has come mainly from the growing private
sector and donor spending. Although government health spending has declined, however, the
public sector remains the major provider of health services, an important conduit of donor funds,
and its pooled resources wields considerable influence in the sector.

B. Government Health Expenditures

113.     The macroeconomic context of government health spending in Malawi has been,
and will continue to be, difficult. Since the mid-1990s, economic growth has been slow, sharply
fluctuating, and shrunk precipitously by 4.2 percent in 2001. The economy has grown on average
by less than 2 percent per year since 2000, but with the population growing by nearly 2 percent
annually, this has meant an acute fall in per capita consumption. The poor GDP performance has
been accompanied by chronic food insecurity caused by successive bouts of droughts and
flooding, an adverse external environment, the government‟s inability to reform agricultural
marketing of the key staple (maize), and overall declining purchasing power of households7 .
Poor economic management has compounded these problems. In recent years, the government
has incurred unsustainable fiscal deficits, forcing it to borrow from the domestic banking system.
In turn, interest rates have soared to as high as 35 percent in 2003/04, which crowded out the
private sector and in turn dampened potential growth. Much of the spending in excess of
available resources over the past five years have been made by Foreign Affairs, Police, Special
Activities, Office of the President and Cabinet (OPC), and State Residences which, observers
noted, do not reduce poverty (MEJN, 2004). The government‟s inability to adhere to its own
medium-term expenditure framework (MTEF) to stabilize the economy has also resulted in

7
 Malawi‟s chronic food insecurity is a highly co mplex issue, and many thoughtful reports have been
written about its causes, dimensions, and consequences. This report is not the place to analyze these issues,
but interested readers may want to see one succint analysis in Rubey (May 2003).


                                                                       42
reduction of donor inflows in the last three years. Most importantly, the government overspending
and high interest rates have fueled inflation, which has reduced the real value of government
expenditures as well as households‟ spending. Although optimistic forecasts see annual GDP
growth of 5.4 percent from 2002-06, MOH planning documents use a more sober annual growth
of 1-3 percent.

114.     Despite economic difficulties and the government’s failure to adhere to its MTEF
program, the government has shown commitment to protect health expenditures. For most
of the 1990s, annual government recurrent budget allocation for health was only around 6-8
percent of the total government budget, but this increased dramatically to 9-10 percent in the
early 2000s (Figure 3). The 2002/03 approved budget allocated 11.2 percent to health. In
nominal terms, government health expenditures reached MK4.6 billion in 2002/03, equivalent to
MK351 million in real 1993 prices (Figure 4). For 2003/04, the allocation to health increased
marginally. There would have been more resources allocated to health in the government budget
if not for the heavy burden of nondiscretionary spending which consists largely of public debt
servicing 8 . Thus, if one considers only total discretionary spending, the share of health increased
to as much as 13.5 percent in 2001/02. This is a significant commitment, although still lower
than the 15 percent recommended by African heads of state in the OAU Summit in Abuja,
Nigeria.

Figure 3. Percent Share of Government Health Expenditures to Total Government Expenditures,
     to Total Nondiscretionary Government Expenditures, and to GDP, 1993/94 – 2001/02

                   Source of basic data: Public finance tables in Economic Report, various issues


    15                                                                                                      13.5
                                   12          11.7                   12.4
          10.8                                             11.1
                                                                    9.5            9.3                    10.1
    10   7.9            8.8     8.3         8.1           8
                     6.8                                                        7.6           7.2
                                                                                           6.4
     5                    2.8         2.6                     2.2         2.6        2.2            2.1          3
               2                                  1.9

     0
         1993/94     1994/95    1995/96     1996/97      1997/98    1998/99     1999/00    2000/01        2001/02

               % Share to Total Expenditures      % Share to Discretionary Expenditures    % Share to GDP




8
  The Malawi Economic Justice Netwo rk (2004) estimates that the proposed debt servicing under the
2003/04 Supplementary Budget Statement represents a whopping 9.3 percent of GDP, co mparab le only to
Angola (11.0 percent) and the Republic o f the Congo (14.3 percent), two countries that had to borrow to
finance their war efforts. The peace-time debt servicing rat io in Malawi is seen to be unprecedented.


                                                         43
  Figure 4. Trends in Actual Government Health Expenditures, in Million Malawi Kwacha, at
                     Current and Constant 1993 Prices, 1993/94 – 2002/03

                        Source of basic data: Public finance tables, Economic Report, various issues


   5000                                                                                                                 4334       4552

   4000
   3000                                                                                                  2275
                                                                                           1975
                                                                               1714
   2000
                                                                 992
   1000                                645        679
               192           398
                 192           315       282        187                253           349          283          260           393          351
      0
              4



                            5



                                         6



                                                    7



                                                                   8



                                                                                 9



                                                                                              0



                                                                                                           1



                                                                                                                         2



                                                                                                                                      3
           /9



                         /9



                                      /9



                                                 /9



                                                                /9



                                                                              /9



                                                                                           /0



                                                                                                        /0



                                                                                                                      /0



                                                                                                                                   /0
        93



                     94



                                    95



                                                 96



                                                             97



                                                                           98



                                                                                         99



                                                                                                        00



                                                                                                                    01



                                                                                                                                 02
      19



                   19



                                  19



                                               19



                                                           19



                                                                         19



                                                                                       19



                                                                                                      20



                                                                                                                  20



                                                                                                                               20
                                                      Current Prices         Constant 1993 Prices



115.     The composition of government recurrent expenditures by inputs shows an
alarming erosion of the proportion going to personal emoluments (PE). Personal
emoluments accounted for around 40 percent of the Malawi MOH budget in the late 1990s,
peaking at 56 percent in 2002/03 (Figure 5). It has since gone down to around 30 percent since
2003. The dramatically reduced share of PE to total recurrent costs in recent years reflects the
severe staff shortages and salary freezes. To correct this situation, some donors have resorted to
implicit salary top-ups for civil servants providing services to health projects. Creative use of
non-PE expenditures such as travel allowances and attendance at workshops have also been
resorted to. There is increasing recognition that PE needs to be raised to avert further erosion in
the morale of health workers. In the 1980s, African health sectors typically used 60-70 percent of
their budgets on PE, and the complaint then was that there were not enough complementary
inputs for workers to deliver services. Today, the opposite is happening, with other recurrent
transactions (ORT) accounting for the lion‟s share (60-70 percent) of the MOH budget, somehow
indicating complementary inputs in want of staff.

   Figure 5. Percent Allocation of MOH Recurrent Expenditures, by Economic Classification,
                                     1995/96 – 2005/06

                                Sources of basic data: MOH, MOF Budget Books, various years

     100%              3.3               3.1               0                  0                   0               0                   0
                                                         58.8                43.6
                    57.1               63.1                                                   67.6               69.7              69.8
      50%
                    39.5                                 41.2                56.4
                                       36.9                                                   32.4               30.3              30.2
          0%
                  1995/96A           1998/99A         2001/02A          2002/03A           2003/04E            2004/05P        2005/06P

                                                           PE      ORT         Capital transfers




                                                                        44
116.     Trends in non-PE expenditures are difficult to analyze because of the effect of
extraneous factors that need to be taken into account. These include drug donations made by
donors that are not captured in the expenditure returns (especially the significant KfW
phamaceutical donation in 1996/97); the significant donor involvement in the financing of
projectized activities (training, supervision, commodities support, etc.) in the health sector in
which MOHP is involved; and the increasing tendency to use non-PE expenditure items to
implicitly beef-up meager salaries, e.g., through training, travel, and food provision.
Nevertheless, at their face value, the following trends are observed: The share of goods and
services has remained relatively stable at around 61-65 percent. The share of pharmaceuticals
averaged 21 percent over the period. The share of internal travel has also remained fairly stable
at around 8–10 percent, although it is recognized that central supervision of districts need to be
given more funding. Likewise, the share of maintenance has been constant at around 6-8 percent
per year, although clearly this has not been sufficient to maintain the existing health
infrastructure.

117.    District-based health services (secondary and primary care) has captured an
increasing share of the recurrent health budget as the allocation to tertiary care has
dropped. Secondary hospitals and the facilities below them increased their share from 55 percent
in 1998/99 and to 63 percent in 2003/04 while tertiary hospitals‟ share shrank from as high as 38
percent in 19990/91 to 24.2 percent in the mid 1990s and 24.1 percent in 2003/04 (Figure 6).
These figures, however, must be seen in light of the fact that tertiary hospitals continue to
perform much of the secondary-care functions in their catchment areas, and that around 30
percent of secondary hospitals‟ expenditures are actually primary care in nature. Nevertheless,
the trends show the government‟s commitment to improve peripheral health services.

 Figure 6. Percent Allocation of MOH Recurrent Expenditures by Institutional Level, 1995/96 –
                                         2005/06

                     Sources of basic data: MOH, MOF Budget Books, various years

     100%
      80%                                       45.3
              67.7       59.1       60.9                   57.9        64.1           66       65.4
      60%
      40%                                       17.1
                         24.1       24.9                   25.8        24.1
      20%     21.4                              37.6                                  24.9     24.2
              10.9       16.8       14.2                   16.3        11.7            9.1     10.4
       0%
            1995/96A 1996/97A 1997/98A 1998/99A 2001/02A 2003/04E 2004/05P                   2005/06P

                                HQ + Regional   Central Hospitals   District Health



118.     Expenditures for central and administrative functions (“headquarters costs”) are
still high, somehow limiting the resources that could go to the peripheries. Although the
central administrative expenditures of MOH have declined slightly from 21-22 percent in the
early „90s to 19.9 percent in 2003/04, it is still higher than the 15 percent standard proposed for
health administration in the World Bank‟s “Better Health in Africa”. These figures, however, do
not take account of the direct subvention of the Ministry of Finance to CHAM health facilities
located mostly in rural areas. Recent initiatives, however, could revert this trend, including the
establishment of the National Health Commission and the four zonal offices, both of which
require additional staff complements.



                                                   45
119.    Overall financial management in government has been weak across the board,
although no specific cases have been documented for MOH. This problem manifests
throughout all ministries through the accumulation of arrears, unpredictability of government
funding, and heavy government borrowing from the commercial banking sector even when it d id
not have to (Budget and Finance Committee, 2003). To maintain spending discipline, the
government has committed to the following:

        (a)     Credit ceiling allocation system. The Treasury intends to give commercial banks
                a credit ceiling for each ministry, which they draw on a daily basis, and which the
                banks will claim reimbursement from the Reserve Bank of Malawi. The
                ministries will be informed of their funding levels on a quarterly basis. This is
                expected to make ministries plan their expenditures in advance and avoid
                accumulation of arrears.

        (b)     Commitment control. Arrears of outstanding payments to various suppliers have
                been a chronic problem. To ensure that no ministry or department accumulates
                new arrears, a commitment control system is being introduced under which
                ministries will be informed on a quarterly basis of their funding allocations.
                Commitments will be based on these funding levels, rather than the budgeted
                levels.

        (c)     Integrated Financial Management Information system (IFMIS). In order to
                strengthen the management of the wage bill and human resources, the
                government is about to complete the implementation of the new system that
                integrates personnel data with the payment of salaries, pensions and staff loans.

120.      The unpredictability of government funding makes budget analysis, planning and
execution difficult. Data provided by the MOH appear to show low utilization of its budgetary
allocation. According to Table 7, only about half of the MOH budget for the two years under
review were utilized. These figures must be interpreted with extreme caution, however. Firstly,
it is not clear if the “approved budget” is indeed the accurate figure, since there is an increasing
proliferation of “official” budget figures (see paragraph below). Secondly, ensuing expenditure
cuts during budget shortfalls render the “approved budget” figure meaningless. Thirdly, delays in
the districts reporting back their actual expenditures make it unclear whether the “actual
expenditures” are the real totals of spending. In the raw data provided by MOH, a few districts
did not have complete expenditure returns. In contrast to the figures shown in Table 7, MOH
appears to have utilized its pro-poor allocations fully (see below).

121.    There is a need to clean up the budget and produce consistent format and figures on
allocation and actual spending. Various official documents show inconsistent figures on
“approved”, “estimate”, “revised” and “projected” expenditures, a problem that was raised by the
Budget and Finance Committee of the Parliament itself (2003). For instance, for the MOH, the
2002/03 revised budget is MK3,548.7 million (as it appears in the Supplementary Budget
document), and is MK3,777.7 million (as it appears in the Financial Statement), or a discrepancy
of MK228.9 million. In the Budget Document No. 4 (Approved Estimates of Recurrent
Expenditure on Recurrent and Capital Accounts), the 2002/03 approved MOH budget is
MK3,529.1 million, while the revised figure is MK3,589.2 million. Because of this lack of
consistency, it is difficult to make a proper analysis of budget allocation, performance and
execution.




                                                 46
 Table 7. Estimated Utilization of MOH Budget, by Major Expenditure Categories, 2001/02 and
                                         2002/03
  Fiscal        Expenditure Category        Approved            Actual                Percent
  Year                                       Budget           Expenditures           Utilization
2001/02         Personal emoluments          1,009.4             472.1                  46.8
                Other recurrent              1,254.4             672.5                  53.6
                transactions
                Total                       2,263.7            1,144.6                  50.6
2002/03         Personal emoluments         1,098.8             573.1                   52.2
                Other recurrent             1,112.3             442.9                   39.8
                transactions
                Total                       2,211.0            1,016.0                  46.0
                              Source: MOH Planning Department, 2004

122.     In 2002/03, the government started budgeting for “protected pro-poor
expenditures”. These protected pro-poor expenditures (PPEs) follow the Malawi Poverty
Reduction Strategy (MPRS) priorities; they are protected from ensuing expenditure cuts in the
event of budget shortfalls. They are distinctly identified in the funding instructions through the
credit ceiling authority system, thus ensuring that the ministry is aware of the portion of funding
for PPEs and does not divert them to non-priority activities. In the case of MOH, the five PPEs
are primary health care, preventive health care, health workers‟ training, secondary curative care,
and drugs. An assessment of the PPEs indicate the following:

          (a)     Although relatively high allocation has been given to priority pro-poor sectors
                  including health, some high-priority activities remained under-allocated. For
                  instance, although MOH received overall increase in the 2003/04 budget, there
                  were reductions in allocations to health worker training (less MK18.2 million
                  from the previous year), primary health care (less MK80.3 million), and
                  preventive health care (less MK8.9 million) (Budget and Finance Committee,
                  July 2003). Moreover, HIV/AIDS mainstreaming activities in the different
                  government agencies continue to be severely underfunded, with some 17 “votes”
                  not allocating any funds for these critical activities. Recommendations have been
                  made to designate HIV/AIDS mainstreaming activities as protected pro-poor
                  expenditures.

          (b)     Compared to other ministries, the MOH has been highly disciplined in its
                  spending patterns and utilization of PPEs. FY02/03 pro-poor expenditure figures
                  show MOH‟s utilization rate of 103.5 percent, compared to 119.3 percent
                  overspending for Education and 323.0 percent overspending for Agriculture
                  (MOF Budget Document No. 4, 2003).

          (c)     Changes in the terminology and expansion in the number of health PPEs make
                  monitoring difficult. In 2003/04, three other health PPEs have been added to the
                  original five, namely, infrastructure development, rehabilitation and
                  maintenance; health technical services; and clinical and population services.
                  However, “primary health care” has disappeared, but probably has been merged
                  with the other items. The reporting format and consistency of indicators should
                  be maintained as an inherent part of budget transparency and performance
                  monitoring.



                                                 47
123.      In terms of geographic equity9 , the distribution of per capita recurrent expenditures
tended to favor the sparsely populated northern region more than the densely populated
southern and central regions. The northern region as a whole had per capita health
expenditures of MK46.87 in 1998/99 compared to MK23.58 in the central and MK33.45 in the
southern regions. This could be because of the much lower population density in the north and
the need to maintain health facilities there even at higher per capita costs. However, analysis
conducted by the Malawi NHA team using an alternative composite index which provides equal
weights for district population, poverty severity, and required recurrent (ORT) funding for
infrastructure in place shows that all the districts in the northern region are actually “over-funded”
while half of the districts in the south and a few districts in the central regions are “underfunded”.
(Table 8). It must be noted, however, that these indicators only provide information on the
allocation of government ORT resources and therefore do not show a complete picture of the total
health expenditures in the districts, e.g., they do not take account of CHAM, other NGO, and
increasingly, local government health expenditures, all of which also vary considerably by
district.

    Table 8. District Status by “Overfunding” or “Underfunding” of ORT Allocation, 2000/01
Regions                Overfunded Districts                 Underfunded Districts
North        Chitipa, Karonga, Nkhatabay, Rumphi,   None
             Mzimba (n=5)
Central      Kasungu, Nkhotakota, Lilongwe,         Dowa, Ntchisi, Salima (n=3)
             Mchinji, Dedza, Ntcheu (n=6)
South        Mangochi, Machinga, Thyolo, Mulanje,   Balaka, Blantyre, Chiradzulu, Mwanza,
             Chikwawa, Nsanje (n=6)                 Phalombe, Zomba (n=6)
Total                      17 districts                            9 districts
                                  Source: Malawi NHA Report

C. District Budget Planning and Implementation

124.     Budget preparation, execution, and monitoring at the district level still leaves much
to be desired. The District Implementation Plan (DIPs) is an activity-based, resource-driven, and
costed plan spearheaded by the District Health Management Team (in consultation with
stakeholders in the district) to support national objectives for the delivery of essential health
services. A sample DIP from Chiradzulu District shows that the plan has specified objectives,
strategy, target indicators, tasks and responsible persons/units, and a district budget. In March
2004, the MOH undertook a nationwide monitoring of districts, clustered into groups, and found
serious issues in budget preparation, execution, and reporting 10 . The monitoring exercise yielded
only qualitative information on the state of the implementation of DIPs. Although there were no
apparent delays in districts‟ receipt of their budgets, there were issues of budget execution and
cash management, budget monitoring and expenditure tracking. In general, there was a highly


9
  Geographic equity has been a majo r preoccupation of Malawi p lanning and policy making, perhaps
because of the political dimension of how the reg ions are configured. What has not been given as much
attention is the socioeconomic equity and benefit incidence analysis of health expenditures. This gap in
knowledge certain ly needs to be addressed.
10
   Notes on District Imp lementation Plans (DIP), MOH. Districts were v isited by clusters of monitoring
teams. The purposes of the visits were to assess the districts‟ adherence to the current DIP in the
implementation of health activities; to identify district-specific problems that disrupt the DIP develop ment
process and its implementation; and to support the DHMTs in problem-solving and agree on way fo rward.


                                                     48
variable quality of DIP implementation, but also common issues, which will be discussed in the
next paragraphs.

125.    The monitoring teams identified problems related to DIP coordination, stakeholder
participation in the DIP process, activity and budget preparation, and DIP dissemination.

        (a)      Coordination: Many districts suffered from lack of coordination of the DIP
                 process within the DHMTs, or between the DHMT and program coordinators.
                 The DHO was frequently mentioned as the DIP coordinator, although in many
                 cases, there was no focal person responsible for coordinating the DIP activities.
                 As an urgent matter, a DIP focal person must be appointed to be responsible for
                 the production of monthly, quarterly and annual DIP reports.

        (b)      Stakeholder participation: The DIP process has been participatory in most, but
                 not all, the districts. In the southern districts except Blantyre, stakeholder
                 participation was minimal. In some districts, planning committees were
                 composed largely of DHMT members and not in comformity with MOH
                 guidelines. In areas where stakeholders tended to be active in the development
                 and planning phase, collaboration is minimal thereafter. DHMT roles and
                 responsibilities in the DIP need to be further clarified.

        (c)      Activity and budget preparation: Most program managers compiled some
                 activities, though in most cases, these were not comprehensive enough. Other
                 stakeholders, especially NGOs, were deemed uncooperative, and many did not
                 submit activity reports to the DHMT.

        (d)      Dissemination: In some districts, the DIP document is available to anybody
                 interested, i.e., DHMTs and program coordinators. However, dissemination of
                 the DIP was not universal; it was given only to some DHMT members. In some
                 cases, only parts of the DIP were made available to DHMT members requesting
                 them. In others, receipt of the DIP is much delayed, well into the implementation
                 period.


126.      In terms of DIP implementation and review, although the plan is developed in
almost all districts, it is scarcely used and its implementation is haphazard. Many districts
had irregular or no DIP review meetings. Where quarterly reviews were undertaken, stakeholder
participation especially of NGOs was minimal, and documentation of the review was sketchy or
unfinalized. Districts that had complete quarterly review reports typically did not share them. A
critical activity that needs to take place to increase the use of the DIP is the holding of a quarterly
review meeting to objectively assess its implementation progress, for instance, determining t he
percentage of planned activities that were completed and resources used, and analyzing the
linkage between resources received and activities undertaken.

127.     DIP budget execution is hampered by the often-unpredictable receipt of resources
by districts, and the hurried adjustments that needed to be made in case of delays or non-
receipt of planned resources. DHMTs complained that their budgets are not sufficient to fully
implement the DIP activities; that their ORT funding tended to decrease with little warning from
the Treasury; and that their district funding flows do not reflect the cash flow projections that they
made in the DIP, especially at the beginning of the financial year. The implementation of key
activities in the DIP is heavily affected by this unpredictable pattern of funding, as districts either


                                                  49
have to postpone or cancel planned activities altogether. Finally, many districts have accrued
debts of MK1.0 million or more. In the case of the NORAD ORT top-up scheme (see below),
deviation sometimes occurred from MOH guidelines on the use of these funds. A few districts in
the Southern Region had double counting and misposting of expenditures: the same activity
sometimes appeared in both the NORAD and Government ORT expenditure returns for the same
month.

128.     Re-prioritization of the DIP is not always evidence-based as most DHMTs lack a
culture of using data in decision-making. Budget and expenditure reviews were not regularly
undertaken, and in some districts especially in the Northern Region, they were never done.
Expenditure returns were often not circulated to DHMT members, who thus tended not to be
aware whether or not they had overspent or underspent. Most districts were able to link only
partially activities with expenditures. One particular district (Zomba), however, did undertake
these tasks religiously and was able to link DIP performance indicators with planned activities.
Reliable information was not available in almost all the districts. There is clearly a need to
inculcate the culture of using data in decisionmaking, although this problem is related to the more
basic hurdle of data generation.

129.      The DIP software has not been fully utilized, either due to lack of staff capacity, the
high turnover of skilled staff, the software’s unavailability, or lack of computers. Many
districts reported lack of knowledge of the functionality of the DIP software. In most cases, it
was only the DHO, the health management information system coordinator, or the staff who have
undergone training on the software that had the skills. The high turnover of skilled staff adept at
using the software is also a major problem. However, even among those trained or
knowledgeable, none had complete knowledge of how to complete all the forms in the software.
And even where the software was used fairly intensively, districts reported that it was unable to
generate the requisite outputs. The DIP software tended to be used only during the planning stage,
and minimally, if at all, during the review and reporting stage. Computer-based financial
management appears to be a major constraint, and suggestions were made that the entire DHMT
should be trained in this area. A few districts in the Northern and Central Regions reported a
more basic problem, the lack of computers.

130.     Central supervision of district health facilities has been infrequent, and DHMT
supervision of health facilities in the district has been haphazard. Budget constraints and
staff shortage has led to infrequent central HQ supervision of districts. In remote districts,
communication is also a major problem, since many district health offices do not have
telecommunications connection. DHMT supervision of district health facilities is also spotty;
although most DHMTs had supervision schedules, these were not often adhered to. Some
DHMTs did not have any supervision checklists, supervision reports were not written, and no
written feedback was sent to the supervised sites.

D. Donor Health Expenditures

131.     Donors provide nearly a third of total health expenditures in Malawi, and this will
grow even further in the future . According to the 1998/99 national health accounts exercise,
donors‟ health expenditures totaled MK1.460 billion or roughly US$ 38.8 million, using the
average exchange rate of MK37.6=US$1.00 for 1998 and 1999. This represents 29.8 percent of
total health expenditures during that year. Though large, this figure still seems underestimated
given that the actual expenditures for the health sector reported by 13 donors in a survey already
reached US$50.0 million in 1996/97 and US$47.1 million in 1997/98 (Picazo, 2002). Over the
coming years, donor commitments to the health sector are expected to grow dramatically. From


                                                50
an actual level of donor commitments of around US$ 66.5 million per year in the mid-1990s to
US$75.7 million in the late 1990s, official development assistance to the sector is anticipated to
reach US$96.1 million in FY03/04 and US$ 139.0 million in FY04/05, more than double the
figure nearly a decade earlier (see Annex __). This translates roughly US$ ___ per capita in the
mid-„90s, to US$ ___ in the late „90s, and to US$ ___ in the mid „00s.

132.      Towards the late 1990s and early 2000s, the number and activities of donors
epitomized the classic fragmentation of revenue sources and the onerous management
burden that this entailed especially for the host country. There are as many as 33 health
donors in Malawi. The 13 donors for which data were available in the late 1990s had as many as
34 projects or discrete activities, each requiring separate negotiation, monitoring and supervision.
The intensity in the administration of these of projects is such that it eroded government‟s scarce
managerial resources, thus limiting government absorptive capacity. Estimates show that on
average, the country‟s absorptive capacity for available donor allocations per year is only 60.5
percent in the late 1990s, and the trend did not show improvement over time (Picazo, 2002). In
fact, as the number of projects multiplied, the absorptive capacity was reduced from 55.8 percent
in 1994/95 to 54.7 percent in 1997/98. By 2003/04, an MOH inventory has revealed that there
were already as many as 85 projects, with a total amount of US$88 million, or around US$1.04
million per project. However, the largest 20 projects accounted for around 77 percent of the total
amount of projects, implying that the remaining 65 projects only accounted for 23 percent. It is
the proliferation of numerous but small projects (averaging in value of US$310,000) that causes
major fiscal and management problems. Yet, with all these projects, none had a positive impact
on the most onerous problem of all, which is human resources.

133.     The flipside of revenue fragmentation is the inefficiency, diffusion, and poor
targeting of donor health expenditures. Indeed, most donor health activities have the right
health service focus. In 1997/98, 87.3 percent of all donor health expenditures were oriented to
primary, preventive and promotive care, while an additional 7.4 percent were devoted to basic
curative care and district health services (Picazo, 2002). Similarly, the OECD database shows
that more than two-thirds (69.0 percent) of donor health allocations in 1998-2000 were devoted to
basic health care, infectious disease control, reproductive health, family planning, and STD
control including HIV/AIDS (OECD, 2004) (see Table 9). However, the actual input uses of
these resources left much to be desired. For instance, out of the US$47.0 million or so annual
donor health spending in the late 1990s, almost a third (34.7 percent) went to technical assistance,
management subcontracts, and other operating expenses, and very little actually went to service
delivery (only 2.9 percent were used for drugs and contraceptives, 1.8 percent for transport and
supervision, and 1.3 percent for information, education and communication). Some of the donor-
supported interventions were also too small to be brought to scale, thus diffusing their impact
while others never got off the piloting stage, a fact that many donors themselves openly admit and
have been at pains to resolve. Donor-induced inflation also characterized the costs of key services
and inputs such as technical assistance and training. For the most part, this was due to limited
local supply in the face of considerable donor demand. Partly, it was also to the lack of agreed-
upon cost standards among the multiplicity of donors that tended to negotiate individually rather
than together as a “community”. Finally, the geographic distribution of donor activities shows a
weakness of targeting, with most of these projects tending to locate in better served areas, e.g.,
Mangochi and Salima, and none in hard-to-reach areas, e.g., Nsanje and Chitipa.

134.     Poor complementarity between government and donor health expenditures reduced
the effectiveness of available external financing in the face of severe MOH fiscal difficulties.
Ideally, external assistance should provide a buffer to government spending. However, the highly
verticalized, often parallel service delivery approach of most donor projects tended to make them


                                                 51
independent undertakings. Many of these were “extra-budgetary” activities that went on with
little government involvement or supervision. The artificial, bifurcated nature of recurrent and
capital budgeting has not helped either, with all donor allocations simply being consigned to the
capital budget even if most of these expenditures are recurrent in nature. Indeed, only around
13.0 percent of donor health expenditures in the late 1990s are truly capital spending (Picazo,
2002). Thus, although 87 percent of the donor health resources were indeed recurrent in nature,
and although the continuing fiscal crisis left the government health sector starving for these
resources, the separate financing mechanisms of the government and the donors made it virtually
impossible for the latter to buffer the former. In fact, the ready availability of recurrent resources
in the donor sector probably worsened the condition in the government sector as donor-funded
NGOs, awash in resources and therefore able to offer much higher salaries, poached civil servants
and siphoned off other “tradeable” government skills for their own activities. The gutting out of
the civil service, fueled largely through ill-advised donor NGO funding, exemplifies the paradox
of the country having much donor resources but the government having so little.

Table 9. Sub-sectoral Breakdown of Donor Commitments to the Health Sector in Malawi, 1998-
                                      2000 Average
                           Sub-sectors                        Amount              % Share
                                                          (US$ million)
      Health policy & administration, management                8.3                  11.0
      Medical services, training and research                   3.0                   4.0
      Basic health care                                         9.8                  13.0
      Basic health infrastructure                               9.1                  12.0
      Infectious disease control                                6.1                   8.0
      Population policy and administration, management          3.0                   4.0
      Reproductive health                                       1.5                   2.0
      Family planning                                          22.0                  29.0
      STD control including HIV/AIDS                           12.9                  17.0
      Total                                                    75.7                 100.0
                                  Source: OECD website, 2003.

135.    Global disease initiatives could undermine rational health sector allocation and
spending, unless changes are made to make them able to provide recurrent cost support.
Malawi is a beneficiary of several large disease initiatives including the Global Fund for AIDS,
Tuberculosis and Malaria (US$ ___ million for HIV/AIDS, US$ ___ million for tuberculosis, and
US$ ___ million for malaria have been approved) and the Global Access to Vaccines Initiative
(GAVI). The guideline for Rounds I to III of Global Fund grants prohibited the use of such funds
to upgrade the salaries of civil servants, or to hire staff to fill vacant posts in the civil service,
although it did allow the same for NGOs. Recent changes in the guideline (under Round IV)
appear to allow the use of funds to support government human-resource and other recurrent cost
requirements, under the rubric of “a new public-private partnership” (Global Fund, Jan. 2004).
Examples of eligible cross-cutting activities include human capacity development, including
“recruitment, training, supervision, incentives for personnel, community workers and people
living with HIV/AIDS.” However, it is not clear whether the Malawi grants, which were
approved under the guideline of Rounds I to III can be “grandfathered” under the new Round IV
guideline.

136.   The NORAD ORT top-up program provides a good model for operationalizing the
needed complementarity between government and donor health expenditures. The



                                                 52
Government of Malawi and NORAD (and subsequently SIDA) started in the fourth quarter of
2002 a funding support program that relies entirely on the government system and supports 25
percent of government non-personnel (i.e., other recurrent transactions or ORT) expenditures to
district cost centers, conditional upon government providing the budgeted district allocations,
with NORAD providing the top-up. As of October 31, 2003, this ORT top-up program already
covered all 26 district health offices and the rural health facilities under them. Except for cases of
“double-counting” mentioned above, this program appears to be working well in terms of budget
planning, execution, monitoring, and audit. Actual expenditures as of June 30, 2003 was
MK574.6 million, of which MK499.8 million came from the Malawi Government and MK74.8
million came from NORAD (NORAD, 2003). The program, expected to run until the fourth
quarter of 2005, certainly provides a good precedent for what is envisioned in the donors‟ basket-
funding program.

137.     The move towards sector-wide approach and basket funding of donor resources is
intended to correct the problems of revenue fragmentation and we ak expenditure targeting.
The pooled funding (initially among DfID, NORAD/SIDA, and the World Bank, with possible
additional contributions from UNFPA and WHO) is a sensible approach in reducing the
management burden arising from traditional multiple donor projects; reducing the budgetary
distortion caused by off-budget expenditures; and improving allocative efficiency by focusing and
tightening the budget around the delivery of essential health package.

E. Private Institutional Health Expenditures

138.     Health insurance is in its infancy in Malawi and provides coverage only to the
better-off families, especially those employed in the formal sector. The Medical AID Society
of Malawi (MASM) is the only health risk-pooling institution of significant scale, reaching a total
membership of around 22,000. It is a voluntary scheme catering to selected parastatals, private
firms, and better-off families who can opt to have standard, executive, and VIP benefit packages.
In 1998/99, MASM reimbursed MK42.0 million to private practitioners; MK18.2 million to
private shops and pharmacies; MK3.1 million to CHAM facilities; and MK0.7 million to MOHP
tertiary hospitals. Its MK64.0 million expenditures for the year represented only 1 percent of
total health spending in Malawi. Annex G provides an indication of the extent of health insurance
payments in Malawi, relative to other sub-Saharan African countries. The slow structural
transformation of the Malawian economy, a typically small formal sector, pervasive poverty, and
inadequate capacity to design and manage sophisticated third-party payment systems have
conspired to make health insurance difficult to introduce and sustain. Although the growth of
health insurance has been slow, there are potentials for further expansion and hence for more
public/private partnership. Countries that still provide medical benefits through budget
allocations can consider converting these to contributory health insurance schemes. Insurance
coverage can be offered to the growing middle classes in cities and townships, as well as workers
in plantations and mines. Premium subsidies can be provided to target populations with limited
capacity to contribute.

139.    Corporate health schemes (in effect, self-insured programs) are growing and
provide health services to their employees. Their total health expenditures of MK868.8 million
in 1998/99 represented 18 percent of total health spending, far higher than their 7.1 percent share
in 1995/96. Households spend (through fees) an additional MK23.0 million on these corporate
programs. Donors also often work with or through these corporate programs, but their
expenditures are not available.




                                                 53
F. Household Health Expenditures and User Fees

140.     Household health spending has exploded in Malawi over the past decade, accounting
for over a quarter (26 percent) of total health spending in the late 1990s. This can be
attributed to a combination of factors: the increasing use of private providers as the underfunded
public health system steadily lost its patients due to declining service quality, the additional
expenditures that households need to make in public facilities with severe supply constraints (e.g.,
gloves, dressings, drugs, medicaments); the increasing household reliance on drug shops; and the
increasing scope of social marketing programs for health commodities that require a modicum of
household payments. Out of the total household health expenditures in 1998/99, 32.7 percent was
incurred in MOHP facilities; 28.1 percent in CHAM facilities; 16.7 percent on private
practitioners; 14.4 percent in shops and pharmacies; 1.8 percent on corporate health programs;
and 2.7 percent on traditional healers.

141.     Household health spending varies considerably across socioeconomic groups,
household residence, and gender. Analysis performed by the Malawi NHA team using the
2000 Household Health Utilization and Expenditure Survey shows that nonpoor households spent
MK117.6 more per capita than poor households; that rural households spent MK11.8 more per
capita than urban households; and that females spent MK24.3 more per capita than males (Table
10). On the one hand, the higher per capita spending by rural households and females is
particularly worrisome because of these two groups‟ common vulnerability status. On the other
hand, this may indeed show households‟ stronger commitment to finance necessary care for their
more vulnerable members.

Table 10. Total and Per Capita Household Health Expenditures, by Different Categories, 1998/99
          Categories            Total Household Health             Per Capita Household
                               Expenditures (MK million)        Health Expenditures (MK)
                               By poor/nonpoor households
Poor households                              881                             293.6
Nonpoor households                           822                             411.2
                                        By residence
Rural households                            1,374                            175.3
Urban households                             330                             163.5
                                         By gender
Male                                         788                             133.2
Female                                       915                             157.5
 Source: 2000 Household Health Utilization and Expenditure Survey, as cited in Tables 6.2, 6.4
                                 and 6.5 of NHA Report.

142.     Use of outpatient services accounts for the largest household health expenditure
across all income groups. Table 11 shows that as much as 75.8 percent of all household health
expenditures are due to outpatient attendance. Inpatient hospitalization accounts for only 15.5
percent of total household health spending, deliveries for 4.6 percent, and the remaining 4.1
percent for preventive care. It is difficult to interpret these results without the accompanying
information on volumes of services used; the conventional wisdom is that inpatient
hospitalization uses up a larger proportion of household expenditures since they are far more
costly on a per-patient basis, although caseloads are far lower. The significant level of resources
devoted by households (particularly the poorest households) to outpatient care raises concerns
about the deterrent effect that fees have on clinic attendance.



                                                 54
   Table 11. Percent of Health Expenditures Spent by Households, by Quintile and by Type of
                                    Health Services, 2000
   Health                                        Quintiles
  Services     Poorest        2nd            3rd            4th          Richest      All
Preventive       4.0          4.0            3.7            4.1            8.8         4.1
Outpatient      77.7         72.7           73.1           77.3           75.2        75.8
Delivery         4.7          5.6            5.3            3.2            4.5         4.6
Inpatient       13.5         17.6           17.8           15.5           11.4        15.5
    Source: 2000 Household Health Utilization survey, as cited in Table 6.3 of NHA Report.


143.     Among a sample of African countries, Malawi registers the greatest resistance to
user fees and the strongest endorsement of free care . Table 12 shows popular attitudes with
respect to user fees in health and education in five East and Southern African countries surveyed
by Afrobarometer (2002). Only 48 percent of Malawian respondents (the lowest proportion in the
region) think it is better to raise health standards even if it entailed fees. In contrast, 82 percent of
Tanzanians agreed to this statement. Conversely, 44 percent of the Malawians (the highest
proportion in the region) think it is better to have free health care even if the quality is low. In
contrast, only 16 percent of the Tanzanians agreed to this statement.

  Table 12. Percent of Respondents Who Agree Somewhat or Strongly to Health and Education
                                     Fees, 1999-2000


             Attitude                   Malawi Tanzania Uganda        Zambia Zimbabwe
                                       (n=1,208) (n=2,054) (n=2,271) (n=1,198) (n=1,200)
It is better to raise                     48        82        57        52        58
health/educational standards even
if we have to pay fees
It is better to have free health         44         16           37                43            35
care/schooling for our children
even if the quality is low
                                 Source: Afrobarometer, April 2002.


144.     Households cite health care cost as a major reason for dissatisfaction with health
services. Households are particularly dissatisfied with the high health care costs charged by
shops/pharmacies, mission facilities, private hospitals, and traditional healers. Yet they are also
dissatisfied with public hospitals, even if cost was not an issue (Table 13), because patients
frequently have to purchase their own drugs outside the health facility when these are not
available. Finally, utilization continues to be low for some services even if they are officially
free, e.g., STI consultations and drugs.




                                                   55
 Table 13. Percent of Households Dissatisfied with Health Services Received, and Those Citing
                          Cost as a Reason for Dissatisfaction, 2002
 Provider Where Care Was Received Dissatisfied Cost as a Reason for Dissatisfaction
Private hospitals                            31.1             52.8
Public hospitals                             35.2              0.7
Traditional healers                          21.0             23.1
Mission hospitals/ dispensaries              32.4             56.4
Shops/ pharmacies                            13.3             67.8
Mobile clinics                                6.6              7.9
Others                                       20.8             40.5
                                Source: Table 5.2, CWIQ, 2002

145.     Unlike its neighbors,11 the Malawi Government never had a national formal user fee
policy or program. Although there have been several user fee studies commissioned by MOH –
the latest was done by LATH (2001) – the government has not been able to come up with a
formal user fee policy (or a broader health financing policy, for that matter) for the public sector,
probably due to its weak popular endorsement as cited in both the Afrobarometer and CWIQ
surveys. Nevertheless, central hospitals and probably some district hospitals continued to charge
fees. As late as 1997, however, the Queen Elizabeth Central Hospital still remitted its fee
revenues to the Treasury (Mbalame, 1997), against the worldwide best practice of retaining
revenues where they are collected. Since there is no monitoring of fees, it is virtually impossible
to assess the magnitude of resources generated. The recent Program of Work of the SWAp
explicitly states that EHP services will be provided free to all Malawians, but it is less explicit
about the financing of non-EHP services.

146.    In contrast with the government, CHAM facilities all charge user fees which account
for a major source of their income. No national-level fee revenue data are available from
CHAM. However, one health facility (St. Anthony‟s Hospital, 2003) reported that out of its
2001/02 total income of MK16.5 million, 34.6 percent came from user fees; 46.8 percent from the
Treasury subvention; 16.2 from donations in cash, medicines, and other goods; and 2.4 percent
from other income. Fees are charged for all services, although the poor are often waived. Many
CHAM hospitals also report high levels of debt collection.

147.      NGO primary clinics and social marketing programs also charge user fees.
Nongovernmental clinics such as BLM charge fees for reproductive health services and use the
revenues to improve quality and expand services. A University of Southampton (2003) study
involving 16 focus group discussions in Zomba (south), Lilongwe (central), and Mzimba (north)
indicated that respondents were willing to pay for services if the quality of care improved; in
particular, if delays in treatment were removed, poor staff attitudes were corrected, and all family
planning methods were made available. Urban respondents were amenable to a modest client
contribution to services, but rural respondents felt their income was too meager to be able to
contribute. Subsidized treatment was generally accepted, but it was difficult to determine the
eligibility of clients to be subsidized. Respondents generally objected to fees in government
facilities.




11
     Kenya, Mozamb ique, Tan zania, and Zamb ia all have formal user fee policies in health.


                                                       56
57
VI.        Options for Financing the Essential Health Package and Non-
                               EHP Interventions

148.     This chapter explores likely scenarios for financing essential and non-essential health
services in Malawi. The first three sections discuss the government‟s vision of providing an
essential health package (EHP) to all Malawians, the guiding principles to accomplish such
vision, and the Program of Work (PoW) that lays out the medium-term framework for EHP
activities and resource requirements. The next sections discuss various options that the
government is considering (or can consider) to improve the financing and delivery of health
services including managing the recurrent cost implications of the PoW, reallocating budget
resources to districts in light of devolution, establishing government and NGO or community
partnerships to delivery the EHP, and financing of referral health services.

A. The Malawi Essential Health Package

149.     The Malawi EHP formally presents government’s core functions in health. MOH
has often claimed that it focuses on providing essential health services. Although this may be true
as a matter of tradition or institutional practice, for many years, such claim was not buttressed by
specific policy guideline or financing mechanism to ration scarce resources towards the provision
of these essential services. In September 2002, the MOH issued the policy document defining the
concept, rationale, initial costing, and broad strategy for implementing the EHP under an
integrated and increasingly decentralized framework (MOH Planning Department, 2002). In the
months that followed, the MOH also prepared the medium-term Program of Work (PoW) for
EHP financing, the final version of which was issued in March 2004 (MOH, 2004).

150.     The EHP consists of a cluster of cost-effective interventions delivered together in
order to reduce the total costs of the interventions by reducing the costs to patients
obtaining the services as well as the costs of providing the services. The proposed EHP
interventions were chosen “on the basis of their being able to be controlled at less than US$ 100
per disability adjusted life years (DALY) gained (MOH Planning Department, 2002). On the
basis of this cut-off, the Malawi EHP is proposed to consist of the following: vaccine-
preventable diseases (EPI); malaria; adverse maternal and neonatal outcomes, including FP,
reproductive health, and safe motherhood; tuberculosis; acute respiratory infection and acute
diarrheal diseases among children; sexually transmitted infections; schistosomiasis; nutritional
deficiencies; eye, ear and skin infections; and common injuries.

151.     The disease interventions included in the EHP support the government’s
commitments to achieve the Millennium Development Goals. Almost all the disease
interventions included in the EHP are public health problems (e.g., vaccination, management of
childhood illness, and infectious diseases such as malaria, HIV/AIDS, tuberculosis, and cholera),
or conditions the alleviation of which most societies consider meritorious (i.e., complications
arising from pregnancy and childbirth, child and maternal malnutrition). These are also the
diseases for which the Malawi Government has made global commitments to address under the
Millennium Development Goals. The “public goods” argument for government involvement in
these infective diseases are well established, i.e., because of their (negative) externalities, they
tend to be under-provided by the commercial market, which tends to cater only to individual
needs. Thus, there continues to be strong justification for government sector involvement in the
financing and provision of these public health services. However, it is clear that the government
alone cannot address all of these problems without the assistance and capacity of private partners.



                                                 58
The PoW lays out the service-delivery role of the nongovernment sector, specifically CHAM, in
the provision of EHP services.

152.     The Malawi EHP reflects global best buys . Although the EHP policy document does
not provide details on the cost-effectiveness rates of the specific interventions proposed, based on
the trends of diseases analyzed in the earlier chapters of this report and the global cost-
effectiveness rates, it is clear that the Malawi EHP has been chosen judiciously based on
universally accepted “global best-buys”12 . Annex H shows the percent of disease burden averted
based on global estimates from low- and middle-income countries. Since Malawi is at the lower
end of low-income countries, with far worse disease burden in all of these diseases, it is likely
that the percent of disease burden averted is even higher than the these figures would indicate.
For instance, Table 14 shows estimates of child deaths that could be averted annually in Malawi
with current child health technologies.

153.    Unlike traditional programs that focused only on service provision per se, the EHP
PoW also includes significant investments in human resources. In addition to “downstream”
provision of services, the EHP requires “upstream” investments in support services, including
laboratory services; drug procurement, distribution and management; information, education and
communication (IEC); supportive supervision; pre- and in-service training for health workers;
planning, budgeting and management systems; and monitoring and evaluation.

     Table 14. Potential Number of Annual Child Deaths that Could Be Averted in Malawi with
                          Current Child Health Technologies, Early 2000s
Disease or Condition         Estimated Annual Child Deaths    Estimated Deaths That Could
                                                                Be Averted with Current
                                                                      Technologies
Malaria                              28,359                               27,903
Neonatal cases                       16,392                                9,548
HIV/AIDS                             16,170                                8,287
Pneumonia                            15,537                               10,824
Diarrhea                             13,309                               12,482
Measles                                243                                  243
Others                                 708                                   0
Total                                90,718                               69,287
  Source: World Bank (2003), “Achieving Millennium Development Goals for Malawi”. Draft
               paper using assumptions based on Bellagio Child Survival Group.

B. Principles Underlying the Provision of the Essential Health Package

154.     The EHP will be provided through integrated, rather than vertical, service delivery.
Invoking the rationale of cost-effectiveness, the MOH aims to integrate more closely the health
services currently being delivered as vertical programs. This will pose a real challenge for the
MOH as well as its partners since the more successful health interventions in Malawi in terms of
coverage or cure rates have been delivered through vertical programs with their own dedicated
funds, i.e., immunization, tuberculosis, cholera, leprosy, and family planning and reproductive

12
   The new interventions proposed under the Malawi EHP, and for which there continues to be much
discussion on cost-effective delivery, are the introduction of antiretroviral (ARV) d rugs for HIV+ patients ,
prevention of mother-to-child transmission (MTCT) of HIV, and post-exposure prophylaxis for health
workers.


                                                      59
health. In the second place, most financing for these activities are currently off-budget, and
converting them into pooled funding over the short-term may be too optimistic.

155.      The EHP will be provided in partnerships among the government, CHAM and
communities. Historically, the government has had a long-standing relationship with CHAM
through its direct annual subvention grants (implicit contracts) to mission health facilities.
Through the PoW process, the subvention grants are being converted into explicit, formalized
health service agreements, a process that is moving apace and is described in more detail below.
Government partnerships with communities will be a greater challenge because Malawi does not
have a strong tradition of community health initiatives such as those in West Africa, those that
currently exist have been intermittently supported by government, or are purely “demand-driven”
with little involvement of MOH.

156.      The EHP will eventually have to be provided under a devolved health service
delivery system. Following the passage in Parliament of the Local Government Act of 1998,
Malawi embarked on a national decentralization program that envisions eventual devoluition of
primary health services to district authorities and city councils. This is the murkiest area in the
PoW and the EHP policy document, since both implicitly assume “decentralization” of financing
(i.e., central “ring-fencing” of district allocation for health) rather than fiscal devolution of health
services through direct block grants from the Treasury to local authorities.

157.      For reasons of equity and expanding access, the government has adopted a policy
that EHP services will be provided free to all Malawians, and to be funded by the
government budget and external assistance. In effect, this means that most of the services
provided by government district hospitals and facilities below them will be free. The case of
CHAM facilities providing EHP services remains to be dealt with, since these facilities have a
tradition of charging primary health services for a modest fee. Referral care and non-EHP
services are likely to be funded from partial fees. Government policymakers have also expressed
interest in a social health insurance program for catastrophic care. The implication of this policy
is that the EHP will have to be funded only through the government budget and external support,
as specified in the PoW13 . A pooled funding arrangement among an initial group of donors is
being worked out for this purpose; the government has also expressed bringing vertical donor
financing closer into the fold of the EHP.

C. The Medium-Term Sector-Wide Program of Work

158.     The current PoW is a “prioritized” version of the “ideal requirements” PoW that
had a much higher price tag. The original “ideal” PoW was costed at US$1,525.7 million; the
“prioritized” version of US$735.7 million (shown in Table 15) represents 48.2 percent of the
“ideal” requirements. The prioritization exercise, done over a period of two years, took account
of factors including the government‟s own absorptive capacity, likely availability of donor
resources, and most urgent needs. The “prioritized” PoW rises from an annual level of US$90.3
million in the first year, to US$147.9 million in the last year. Although these annual levels look

13
   This is the expressed policy, although in reality a number of key health interventions and goods
considered as “essential” (contraceptives, ORS, ITNs, some drugs) are being provided by NGOs and social
market ing programs funded by donors under full cost or subsidized fee arrangements. In addition, CHAM
facilit ies charge fees for services, includ ing some that would fall under EHP. Finally, if and when health
services do get devolved to local authorities, the Local Govern ment Act clearly specifies that such
authorities can set their own local resource-generation mechanisms, including health fees. These issues
continue to be debated and unresolved.


                                                     60
large, it translates into very modest per-capita government + donor health expenditures, rising
from a very modest level of US$7.70 in the first year to US$12.60 in the sixth year. These
figures are just a third of the estimated cost of essential health service of US$32.00 calculated by
the Commission on Macroeconomics and Health (2002).

                         Table 15. Program of Work, 2004/05 to 2009/10
    Purposes                                 Million US Dollars                                % of
                    2004/05 2005/06       2006/07 2007/08 2008/09          2009/10 Total       Total
                                                                                               PoW
1. Training of        28.2       34.1       39.1        44.6      48.6       53.1     247.7     34.0
skilled personnel
2. Provision of       15.4       29.7       19.9        22.2      24.9       27.9     140.0     19.0
pharmaceuticals
and medical and
lab supplies
3. Provision of        6.2        7.6        8.1         8.6       9.0        9.4      48.8     6.6
essential medical
equipment
4. Facility            7.0        6.7        9.0        10.0      10.5       11.0      55.0     7.4
development
5. Routine            24.0       27.3       30.1        30.0      31.7       33.3     176.0     23.7
operations
6. Institutional       9.4       10.4       10.9        12.0      12.6       13.2      68.6     9.3
support to svc
delivery and
policy dev‟t
Total                 90.3      115.8      117.0        127.5    137.2       147.9    735.7    100.0
Per capita            7.70       9.90      10.00        10.80    11.70       12.60      -        -
equivalent
                             Source: Program of Work, March 2004.

159.     The detailed, activity-based unit-costing exercise employed in the PoW provides a
reasonable basis for calculating overall resource requirements, but these need to be updated
periodically. From most indications, the assumptions used in the costing exercise are sound.
However, costs are, by their nature, endogenous and cannot be divorced from the way these costs
will be financed. In the same vein, costs will be affected by the way services will be delivered,
e.g., whether in a vertical or integrated fashion, whether in a centralized or decentralized manner,
or whether they will be delivered by the funder (government) or contracted out to CHAM under
health service agreements, or to community-based distributors. Because of these factors, it may
be necessary to review the realism of the resource requirements on a periodic basis.

160.     Several key issues need to be kept in mind in the future update of the cost figures,
some of which were recognized by the EHP policy document itself: (a) Data gaps with respect
to the incidence of diseases and existing coverage of interventions, i.e., service statistics
frequently understate the true burden of disease since many people do not go to health facilities
for care even if sick. (b) Lack of practice guidelines for some disease interventions especially new
or soon-to-be-introduced programs such as ARV treatment, emergency obstetric care, neonatal
care, etc. (c) The PoW assumptions about human resource salary and benefit adjustments are
rather arbitrary; alternative assumptions need to be underpinned by a thorough assessment of
labor market conditions and remuneration of those doing similar work in the NGO and for-profit


                                                   61
sector. (d) The pre-service training costs only refer to various nursing cadres that are in severe
shortage; they exclude the pre-service training costs of physicians, clinical officers, and medical
assistants who are also in short supply. (e) Pharmaceutical costs were derived from a presumably
unreformed, currently inefficient drug procurement and distribution system. Efficiency savings
from a much improved drug distribution system can yield lower costs. (f) Some overhead costs
were deliberately omitted or were not adequately captured, e.g., general central level
administrative, management, and supervision costs. (g) There are potential efficiencies to be
gained in delivering an integrated EHP rather than separate, vertical health programs. Cost
reductions can be realized in integrated rather than vertical/parallel management, monitoring and
supervision, use of vehicles, procurement and logistics, and training. (h) The foreign exchange
costs and conversion factor need to be explicitly taken into account. The PoW does not
distinguish local and foreign exchange cost requirements, although there are significant items that
will be imported, e.g., medical equipment, transport vehicles, pharmaceuticals and technical
assistance. It will be necessary to do this as part of the procurement plan. (i) The derived PoW
costs are average, not marginal, costs. In theory, if utilization is low (as they are for many health
services in Malawi), the average cost is higher than the marginal cost of services. However,
applying the same average costs to expand high-utilization services such as EPI may under-
estimate true resource requirements, e.g., it gets more costly to cover more remote areas. (j)
Devolution of health services will have profound impact on the unit cost of services. In some
countries, devolved health services incurred costs higher than they were before devolution. (k)
The budget execution or procurement process itself can engender individual and institutional
behavior that may skew actual costs, e.g., delayed payment for a procured drug entails penalties.

161.      In terms of resource allocation, the PoW emphasizes making the existing health
network functional rather than expanding the system. The PoW calls for investments totaling
US$735.7 million over a period of six years, or an average of US$122.6 million per year (Table
16). By component, around 43.4 percent of total PoW resources will be devoted to direct inputs
used in the delivery of the EHP, 37.5 percent for human resources, and 19.1 for systems
strengthening. Most (80.9 percent) of the resources will be devoted to districts and lower-level
facilities; only 19.1 percent will be for central institutions (headquarters, central hospitals‟
support to district services, and a small amount to maintain zonal offices). In addition, districts‟
share of total PoW resources will gradually increase from 79.6 percent at the start of the PoW to
81.1 percent towards the end of the program. A large proportion of PoW resources will be used
to support recurrent costs (92.7 percent); capital expenditures will account for a modest 7.3
percent, to be used mainly to equip existing facilities, to get them connected to basic utilities
(electricity, communications, and water), and to provide transport vehicles for referral. Capital
expenditures are also anticipated to decline dramatically from 10.3 percent at the beginning of the
PoW to 4.6 percent towards the end of the program. Thus, the PoW clearly differs markedly from
previous planning documents in that it emphasizes the recurrent cost needs to make the existing
health facilities functional, rather than further expanding the network.

162.     In marked contrast to previous health plans, the PoW takes a cautious approach to
public investments and focuses on the recurrent costs needed to make health facilities
functional. In the past, because of the multiplicity of non-transparent off-budget, donor-funded
activities, Malawi‟s health sector has tended to expand, and underspend on recurrent costs. As a
result, many health facilities are under-staffed and ill-supplied, and infrastructure and vehicles are
ill-maintained. The PoW seeks to revert this trend, allocating only US$53.9 million for capital
investments, including medical equipment, utility systems and facility upgrading, and transport
equipment. Thus, only 7.3 percent of total expenditures are capital costs, and pooled-funding
donors have committed that no infrastructure construction and rehabilitation will be undertaken in



                                                 62
the first three years of the PoW. Moreover, the PoW itself already budgeted for the recurrent cost
(maintenance) requirements of capital investments of US$88.4 million over the six-year period.

             Table 16. Analysis of PoW by Various Categories, 2004/05 to 2009/10
              Items                04/05 05/06 06/07 07/08              08/09 09/10        Total
Total in US$ million               90.30 115.80 117.00 127.50           137.20 147.90      735.70
By components (% of annual total)
Essential health package            45.4    46.8      44.3    42.2        41.9     41.4      43.4
Human resources                     34.2    32.9      37.7    39.2        39.5     39.7      37.5
Systems strengthening & referral 20.4       20.3      18.0    18.6        18.6     18.9      19.1
By recurrent and capital expenditures (% of annual total)
Recurrent                           89.7    91.5      89.9    93.8        94.0     95.4      92.7
Capital                             10.3     8.5      10.1     6.2         6.0      4.6       7.3
By level (% of annual total)
HQ, central & zonal                 20.4    20.3      18.0    18.6        18.6     18.9      19.1
Districts & below                   79.6    79.7      82.0    81.4        81.4     81.1      80.9
By expected source of funding (% of annual total)
Domestic resources                  37.0    31.0      31.0    28.0        26.0     24.0      29.0
Ext. dev‟t assistance               63.0    69.0      69.0    72.0        74.0     76.0      71.0
                             Source of basic data: March 2004 PoW.

163.     Donors’ future resource commitments to Malawi’s health sector are large, but the
size of the pooled funding is still small relative to what is required by the PoW. The PoW
envisages a jump in government and donor (including off-budget) health spending from around
US$80 million at present, to US$ 90.3 million when the program starts in its first fiscal year
(2004/05) and to US$147.9 million in 2009/10 (see previous table). Compared to these resource
requirements, Tables 17 and 18 show the annual resource commitments of donors to Malawi‟s
health sector, which are estimated to almost double from US$66.5 million in the mid-1990s to
US$124.3 million in the mid-2000s. However, less than a third (29.2 percent) of the donor
commitments in 2005/06 will go to pooled funding that will jump-start the EHP activities. Thus,
although total donor commitments exceed the PoW requirements, much of them (70.8 percent)
are tied to vertical health projects. To be sure, some of these projects also support EHP “in-kind”
(goods, services, technical assistance), though not in cash. What remains to be estimated is how
much “financial input” does the PoW actually require for it to work, and to map specific vertical
activities against the PoW activities and check them as being effectively funded.

     Table 17. Estimated Annual Resource Commitments by Donors, by Period, 1994-2006
               Period          No. of Donors    Estimated Annual Commitments
                                                          (US$ Million)
          1994-97 Actual            13                         66.5
          1998-00 Actual             9                         75.7
         2003-04 Forecast           19                         96.1
         2004-05 Forecast           19                        139.0
         2004-06 Forecast           19                        124.3
  Sources: 1994-97 data are from Picazo (2001); 1998-00 data are from the OECD website; and
                         2003-06 data are from MOHP/LATH (2003).




                                                63
     Table 18. Estimated Annual Resource Commitments by Donors, by Pooled and Unpooled
                                  Funding, 2003/04-2005/06
Funding Agency                         Expected Commitments
                    In Own Currency   2003/04             2004/05                              2005/06
                                      US$ Mn              US$ Mn                               US$ Mn
DfID                40 Mn Br pds        14.5                14.5                                 28.7
NORAD               50 Mn Kr             1.0                 1.2                                  3.9
IDA                 US$ 15 Mn            7.5                 3.5                                  3.7
Total – pooled funding                  23.1                19.3                                 36.3
Total – unpooled funding                73.0               119.7                                 88.0
Grand Total                             96.1               139.0                                124.3
                Source: MOHP/LATH (November 2003); interviews with donors.

164.      Conservative economic growth and MOH budget-share scenarios indicate that the
government is able to meet its financial commitments under the PoW, subject to the caveat
that the heavy debt-servicing problem is eased. Figure 7 shows the estimated annual resource
commitment of the Government, using six scenarios of alternative GDP growth rates (1 percent, 2
percent, and 3 percent) and assumed share of the health sector in total government budget
allocation (13.5 percent and 15.0 percent).14 Table 19 shows the resource requirements versus
resource availability for both the Government and external development assistance over the
period of the PoW. The resource availability row for Government reports the results of two very
conservative scenarios constructed by the MOH for the likely availability of government fundin g,
i.e., (a) 1 percent annual GDP growth and no change in the 2001 MOH share to total government
expenditures, and (b) 1 percent annual GDP growth and 13.5 percent MOH share to total
government expenditures15 . (No assumption is made about user fees and other extra-budgetary
financing since EHP services will be provided free.) The dramatic increase in MOH share to total
government spending starting in 2001/02 means that if this trend continues, and if the economy
grows at a modest rate of 1 percent annually, there is likely to be enough resources for the
government to meet its required contribution in the PoW. In fact, the government‟s contribution is
comfortably within the GOM required resource envelop throughout the forecast period. What
this implies, though, is that the PoW underestimated the expected contribution from the
government.




14
  Specifically, the following scenarios were constructed: A – 1% annual gro wth in GDP and 11% share of
the government budget going to health sector (most pessimistic); B – 1% GDP growth and 13.5 percent
share of health; C – 1% GDP growth and 15% share of health; D – 2 percent GDP growth and 11% share of
health; E – 3% GDP gro wth and 13.5 percent share of health; and F – 3 percent GDP growth and 15 percent
share of health (most optimistic).


15
   While the assumption about a 13.5 percent MOH share may have been unrealistic in the late 1990s until
2001 when the actual share hovered around 6-8 percent, the Govern ment dramat ically increased the budget
for MOH so that by 2001/ 02, it has reached 10.1 percent of total govern ment expenditures and is estimated
to have increased to 11.2 percent in 2002/ 03. Thus, the 13.5 percent share assumption is feasible, subject to
an important caveat about the reduction of government domestic borrowing that is squeezing all
discretionary government expenditures in Malawi. (Th is issue is dealt with in g reater detail below.)




                                                     64
   Figure 7. Scenarios for Availability of Government Resources for Health, in Million U.S.
                                  Dollars, 2004/05-2009/10

                                     Source: MOH Program of Work

                    41.3
                                        41.9
      2004/05                                                                         42.7

                    41.7
                                         42.9
      2005/06                                                                              44.6

                    42.2
                                          43.9
      2006/07                                                                                46.5

                    42.6
                                             44.9
      2007/08                                                                                     48.6

                    43
                                               46
      2008/09                                                                                          50.7

                    43.4
                                                 47
      2009/10                                                                                             52.9



                0          10           20                     30              40                 50             60

                                             A        B        C    D      E   F


Table 19. Analysis of Resource Requirements Versus Resource Availability of the PoW, 2004/05
                             to 2009/10 (in Million U.S. Dollars)
         Items             2004/05      2005/06                2006/07         2007/08            2008/09        2009/10
Government of Malawi
PoW expected                33.41        35.90                     36.27           35.70            35.67         35.50
contribution (a)
Resource availability     41.30 to     41.70 to    42.20 to   42.60 to    43.00 to                               43.40 to
forecast (b)                41.90        42.90       43.90      44.90       46.00                                 47.00
Difference (b-a)           7.89 to      5.80 to     5.93 to    6.90 to     7.33 to                                7.9 to
                             8.49         7.00        7.63      9.20        10.33                                 11.50
External development assistance
PoW expected                56.89        79.90       80.73      91.80      101.53                                112.40
contribution (w)
Resource availability       23.10        19.30       36.30       n.a.        n.a.                                     n.a.
forecast, pooled funds
(x)
Resource availability       73.00       119.70       88.00       n.a.        n.a.                                     n.a.
forecast, unpooled
funds (y)
Difference (x-w)           (33.79)      (60.60)     (44.43)      n.a.        n.a.                                     n.a.
Difference (x+y-w)          39.21        59.10       43.57       n.a.        n.a.                                     n.a.
                    Source of basic data: Program of Work and Table 4 above.



                                                          65
165.     The sustainability of MOH expenditures under the PoW is at risk unless the
government adopts fiscal prudence and eases the domestic debt burden. The sustainability of
PoW expenditures and, more directly, of MOH expenditures, can best be analyzed in the context
of the sustainability of overall government spending. As has been discussed, the Government
succeeded in protecting expenditures in health over the past six years (FY97/98 to FY02/03), with
a dramatic rise in the share of MOH to total government spending occurring in 2001. MOH
expenditures were also protected from erosion due to inflation and population growth. Thus, real
per capita MOH spending (at constant 1995 prices) rose from MK60 in FY97/98 to MK85 in
FY01/02. (In current terms, these are equivalent to MK103 and MK423, respectively.) Although
the MOH‟s share to government spending has been increasing, it still falls short relative to need,
and relative to the 15 percent allocation to health that African Governments committed
themselves to meet in the OAU Summit in April 2001 in Abuja, Nigeria. The 15 percent target
for adequate and sustainable government financing for health would be within Malawi‟s ability to
meet if not for the heavy load of “nondiscretionary” or “unallocable” expenditures that the
government needs to set aside, mainly to pay off its large domestic debt. Nondiscretionary
expenditures have ballooned in recent years, accounting for around 23.5 percent in FY01/02, or
an average share of 20.6 percent from FY97/98 to FY01/02. Because of this burden, further
expansion in government a llocation to the MOH and other social sectors has been constrained.
Thus, if the GOM adopts a strategy of bringing domestic borrowings down, significantly more
domestic resources could be made available for health. On the other hand, if it fails in this area,
government discretionary expenditures (including health) could be further squeezed, reducing the
sustainability of the PoW.

D. Options for Reallocation of District Budgets

166.    Two impelling factors call for a review of the budgetary allocation to the districts. One is
the perceived inequity in the geographic allocation of government health funds and the increasing
focus given by the GOM for equity and poverty alleviation. The impending devolution of health
services also requires re-examining the current allocation criteria. Several formulae are being
proposed.

167.     Option 1: Reallocate using MOHP formula under a deconcentrated financing
scenario. The MOHP formula being considered is a weighted scheme based on a number of
allocation factors and their respective weights: OPD utilization level as a measure of actual usage
(5 percent); poverty severity, i.e., higher poverty results in higher demand for health care (10
percent); existing stock of health care facilities, i.e., districts with more and higher-level facilities
require greater resources (20 percent); previous allocations to be adjusted slowly to a new
allocation scheme (60 percent); a “safety net” provision to prevent any district from limiting
losses vis-à-vis the previous year (5 percent). This proposed MOHP formula is intended to
evolve over time.

168.     However, some observers (Boex, et al, 2001) noted that it has some inherent weaknesses
that need to be corrected. First, the formula is too data intensive, and the required data are not
always available in good quality in the country. Second, utilization rate reflects both supply and
demand; its use masks potential significant hidden demand, especially in poor areas currently
without facilities and therefore with low or zero utilization. Thir d, the stock of health facilities is
not necessarily reflective of need; some of these could have been constructed in response to
political imperatives. (To be sure, some of these need to be maintained and cannot just be shut
down.) Fourth, the formula does not take account of the role of CHAM facilities, which also




                                                   66
receive government funding through subventions. Finally, the formula provides districts with
adverse incentives to exaggerate their utilization rates so they can obtain more funding.

169.    This proposed MOHP formula also reflects the Ministry‟s cautious approach to
decentralization. The intention is to devolve the clinics first, while the rest of the district health
care system remains retained by MOHP. While this approach seems reasonable given Malawi‟s
capacity constraints to manage more sophisticated facilities at the local level, the artificial
separation could disrupt service delivery (especially referral) and the fragmented responsibilities
over local clinics and MOHP facilities could engender administrative problems, as has been
shown in other countries.

170.     Option 2: Reallocate using a new formula to be developed under a devolved
financing scenario, with health subsumed within the overall allocation to the district. This is
the option preferred by those pushing for a dramatic reallocation of resources based on need. As
an interim measure, reallocation could be done using available demographic and socioeconomic
data. Total district population is a simple, “clean”, objective, readily available, and easily
verifiable (therefore transparent) indicator that can be used to calculate required per capita health
spending. Indeed, this is used in Bangladesh (?), and is increasingly being used in other low-
income countries.

171.      The PoW is silent on how the expenditures can be sustained at the district level when
primary health services are eventually devolved to local authorities. (Such services have been
devolved to city authorities effective January 1, 2004.) The challenges during this period of
transition, and the proposed approach for dealing with them, are: (a) To ensure that local
authorities devote as much (if not more) resources to priority health interventions as the PoW
currently does. One approach that some countries with devolved health services have employed to
solve this problem is to establish categorical or conditional grants that specify (priority) activities
to be funded. This has its pros and cons, but it ensures that in the short-term, PoW activities can
be adequately funded. (b) To ensure that fiscal transfers to districts are made on an equitable
basis, i.e., that no district will be made worse off than it was before the devolution. This requires
that the existing district allocation formula be reformulated to take account of the resource
requirements of assets and personnel actually transferred, and health needs that were previously
not captured in the “incremental budgeting” practice.

172.    Option 3: Use allocation formula that also takes account of HIV/AIDS. This has not
been dealt with, possibly because much of the current funding (Global Funds, other official
development assistance, private and public flows through NGOs) is outside the MOH. Moreover,
a sizeable part of these flows are not strictly “health expenditures” (e.g., funds for community
mobilization, orphan care, home-based care and social protection programs for people living with
AIDS, inter-sectoral activities). Finally, NAC‟s funding portfolio so far has been largely
“demand-driven” based on the types of proposals it receives from project proponents. Because of
these factors, little effort has been done to track these flows and find out how much is going to
each district. There are positive indications that the involved agencies (NAC, MOH, NGOs, and
local governments) are beginning to coordinate their work. They should also share client
coverage, financial, and other information and use these to develop more comprehensive
resource-allocation formulae that takes account of intra-district equity considerations.

E. Options for Government/NGO Partnerships: Health Service Agreements

173.    Although the government and CHAM have had a long-standing relationship in the
health sector, the Health Service Agreement (HSA) is the first attempt to formalize this


                                                  67
relationship. Traditional input-based subvention grants between the Treasury and CHAM were
implicit contracts in that they did not specify outputs to be delivered and penalties for non-
delivery. Although subvention grants will continue to be provided to CHAM for salaries of
health workers, in 2004 MOH launched Health Service Agreements (HSAs) for more specific
services that it wants CHAM to deliver. HSAs formally define the roles and responsibilities of
the District Health Office as funder/regulator and the CHAM facility as a provider of EHP
services. The annual agreement is signed by the district health officer, the MOH Permanent
Secretary, the hospital manager/director, and the hospital owner (e.g., the bishop of the diocese,
in the case of a Catholic facility). Each HSA is specified in terms of the health services to be
delivered (output), and the time frame of the agreement; the process of patient identification and
referral; the estimated or targeted volume of patients; the cost of services covered under the
agreement; the excluded cost of services including start-up obligations; the payment process; the
reporting requirements; and arbitration procedures. So far, the scale-up process involves 11
CHAM hospitals and 29 health centers (see Annex J).

174.    One example of an HSA is that between the government and CHAM Hospital “X”
for the provision of maternal health services for a period of one year. The draft agreement is
impelled by the government‟s desire to provide maternal services to the catchment population in
the Nkhotakhota District, which it currently cannot provide through its district hospital. The draft
agreement has the following features:

        (a)     Health services to be provided: These include normal delivery for the immediate
                catchment population, post-partum hemorrhage, eclampsia, obstructed labor,
                severe anemia, sepsis, newborn complications, abortion complications, post-
                partum care for all deliveries, and any other complications as indicated. St.
                Anne‟s will provide these services for free through an exemption system
                described below. All clinical costs (labor, drug charges, equipment use) and
                hotel costs (ward charges, food) are included in the HSA and are to be provided
                free to eligible patients.

        (b)     Process of patient identification and referral: The pregnant woman must attend
                the antenatal clinic at the District Hospital. At the third and final antenantal visit,
                the pregnant woman is given an Exemption Pass to be presented at St. Anne‟s on
                the date of delivery. St. Anne‟s staff collect the Exemption Passes and returns
                them to the District Hospital once a month.

        (c)     Estimated case-load of patients: Based on incidence data from the district‟s
                health information system, 102 normal deliveries and 28 complicated cases are
                expected to be seen at St. Anne‟s (see Table 20).

        (d)     Cost of services covered under the agreement: The estimated costs per case are
                shown in Table 43. These include all clinical costs (personnel salaries and
                benefits, drug charges, use of equipment) and hotel costs (ward charges, food,
                etc.) but exclude the following: (1) Cost of drugs to be provided in kind by MOH.
                (2) Cost of referral transport, which will be provided by the DHO. (3) Cost of
                maintenance of buildings and equipment, utilities, and blood supply and
                equipment, which are all on account of St. Anne‟s. (4) Cost of staff salaries and
                benefits covered under CHAM‟s subvention grant from the Ministry of Finance.
                (5) Costs arising from utilization of ineligible patients. (6) Cost of program start-
                up, including repainting of delivery suites, maternal ward, and neonatal ward;
                provision of mosquito nets in maternal and neonatal wards; and provision of


                                                 68
               anesthetic services. (7) Provision of nursing services. (The MOH is looking at the
               possibility of seconding nurses from the district hospital.)

   Table 20. Caseload and Cost of the Health Service Agreement Between the District Health
                          Office and CHAM Hospital “X”, FY2004
 Maternal Condition       Estimated Caseload         Cost Per Case               Total Cost
                              per Month             Excluding Drugs                (MK)
                                                         (MK)
Normal delivery                  102.00                   581                      59,262
Obstructed labor                 18.76                   3,797                     71,232
Post-partum                       3.92                   3,607                     14,139
hemorrhage
Abortion complications            1.96                   3,117                     6,109
Newborn complications             1.68                   4,494                     7,548
Severe anemia                     1.12                  13,842                     15,503
Eclampsia                         0.56                   2,502                     1,401
Sepsis                            0.00                   5,508                        0
Post-partum care                 28.00                   44.28                     1,240
                               Total                                              176,434
                  Source: Based on actual draft HSA with a CHAM Hospital.

       (e)     Payment process and adjustments: As soon as the district receives its monthly
               allocation from the Treasury, the DHO shall make a monthly transfer or check to
               CHAM Hospital “X” equivalent to MK 176,434 as per agreement. If the actual
               caseload exceeds (or falls below) the target caseload for each of the maternal
               conditions, the DHO shall increase (decrease) the amount of reimbursement
               equivalent to the number of cases and the agreed-upon costs per case. These
               adjustments will be made on a quarterly basis. Should the actual utilization
               pattern systematically be higher or lower than the target caseload, either party to
               the agreement may request an appropriate revision to the target caseload. (A
               similar provision is needed in costing, but has not been agreed upon at the time of
               writing.)

       (f)     Reporting and monitoring requirements and arbitration procedures: The
               contractor hospital is expected to send monthly activity reports to the DHO,
               including the number of cases seen, use of the transferred money, and issues and
               constraints in the implementation of the agreement. In addition, quarterly
               meetings will be held by the Steering Committee of the HSA involving
               representatives of the district health office, the hospital management, and civil
               society. The MOH and DHO may also carry out spot checks.

       (g)     Arbitration procedures: The agreement specifies the procedure for arbitrating
               conflicts. The HSA is currently “non-binding” under Malawi law, i.e., the
               government/MOH has no legal right of redress if the standard of service provided
               by the CHAM facility as “contractor” is deemed unsatisfactory. MOH‟s only
               course is through the termination or non-renewal of the agreement.

175.   Although HSAs have been adopted as a national strategy to provide EHP services in
areas where government does not have a health facility, it currently does not cover non-
CHAM NGOs and other private providers . The same HSA principles and practices can be


                                               69
applied to other nongovernment providers, including stand-alone NGO clinics such as those of
BLM, employer-based health programs, social marketing programs, and perhaps even private
clinics. An increasing number of studies (e.g., Loevinsohn and Harding, 2004) have documented
the attractiveness of engaging NGOs as active “contractors” for the provision of public health
services, including: (i) ensuring a greater focus on measurable results, particularly if contracts
define objectively verifiable outputs and outcomes; (ii) overcoming “absorptive capacity”
constraints that often plague government health care systems and prevent them from effectively
utilizing the resources made available to them; (iii) utilizing the private sector‟s greater flexibility
and generally better morale to improve services; (iv) increasing managerial autonomy to
managers on the ground; and (5) using competition to increase effectiveness and efficiency.

           (a)      NGOs – The rapid emergence of local NGOs and NGO coalitions and the recent
                    enactment of the NGO Law that formally regulates NGO activities 16 strengthens
                    the case for exploring government HSA arrangements with these health
                    providers.

           (b)      Employer-based health programs – The successful experience of Project HOPE‟s
                    maternal and child health programs that it negotiated with 39 tea estates in
                    Malawi in the 1990s (Franco, et al, n.d.) demonstrates the large-scale feasibility
                    of this arrangement. Estates are particularly well-suited since they are located in
                    rural areas where there may be no government health facility.

           (c)      Social marketing programs – It should be possible for government or donors to
                    purchase critical health goods (ITNs, ORS, contraceptives) through these agents
                    as an alternative to unreformed Central Medical Stores or other monopolies.

176.     Malawi’s experience with contracting out “non-core” health services in the late
1990s need to be documented and evaluated. As part of the overall rationalization of
government functions, the government embarked on service agreements with private providers in
cleaning, transport, building and ground maintenance, laundry, security, catering, mortuary, audit,
and IEC (drama and band) services, especially in central hospitals. Malawi‟s lack of experience
in this area, notably the absence of government procurement rules and procedures dealing with
health services and ancillary services, hindered progress in the 1990s and early 2000s. The recent
passage of the Procurement Act should provide new motivation to re-examine this program
initiative.

F. Options for Government/Community Partnerships

177.     There is an increasing recognition that much of the EHP information, goods and
services can actually be delivered at the community level. Arrangements are being made with
community-based initiatives, i.e., through the Malawi Social Action Fund (MASAF III), to make
these community EHP services be provided in communities on partnership with MOH and
eventually with local governments. Key activities have been identified as falling under
“community EHP”: prevention and early treatment of common illnesses, promotion of maternal
health including intermittent presumptive treatment of malaria among pregnant women,
distribution of family planning commodities, and growth monitoring of under-five children. In
addition, the community is mandated to be responsible for the rehabilitation of health centers and
the expansion of health posts into centers, and staffing of these health posts/centers with persons
they identify to be trained.

16
     These are documented in the first report of this two-volu me study.


                                                       70
178.    For the “community EHP” to be realized, the following issues need to be addressed:

        (a)     Refining the community EHP service package, including the logistics needed to
                keep supplies flowing: The MOH still needs to finalize what this package would
                be, which includes training, commodities, and monitoring and supervision. The
                commodities involve, among others, birth kits, chlorine for making water safe,
                contraceptive commodities (condoms, pills), essential drugs included in the drug
                revolving funds, first-aid kits, insecticide-treated nets, micronutrients (Vitamin
                A, iron folate), oral rehydration salts, and vaccines. It is not clear how these
                commodities will be sourced, and what the logistics would be for them to be
                distributed to each and every community.

        (b)     Human resource requirements: The ratio of 1 HSA per 2,500 persons is deemed
                too low for the effective delivery of community EHP. No alternative ratio has
                been proposed. One option is to train and hire more HSAs, but since HSAs are
                civil servants, a dramatic increase in their number has significant budgetary
                implications. Alternatively, other community workers (traditional birth
                attendants, community-based distributors, growth monitoring volunteers, etc.)
                could be deployed in larger numbers. Gearing up the community EHP would also
                involve significant increase in supervision by higher-level staff. The government
                has to determine who will supervise the HSAs, the additional full-time equivalent
                staff needed to do this supervisory task, and its budgetary implications.

        (c)     Standardization of staff incentives: There is a wide variety of community health
                workers in Malawi, some are paid, while others are not. Both paid and voluntary
                community health workers also rely on different monetary and non-monetary
                incentive packages. There is a need to standardize these packages, including
                orientation and refresher training, certification, uniform and ID card, and bicycle
                ownership and maintenance requirements. This is a tall order, as there are many
                donors and NGOs involved in community health, and each has a different set of
                incentives.

        (d)     Community structures: Strengthening village health committees and health center
                committees in the context of eventual devolution of health services to district
                assemblies is a major challenge, and for which the PoW has not made adequate
                funds provision.

G. Options for Financing Non-EHP Services

179.     The financing of non-EHP services (inpatient hospitalization) is not dealt with in the
PoW. However, discussions with MOH policymakers revealed: (a) the recent issuance of a
policy making all inpatient hospitalization and outpatient consultations at central hospitals subject
to fees; (b) the vision of converting the central hospitals into autonomous bodies; and (c) the
pursuit of social health insurance to finance catastrophic and referral health services for formal-
sector employees.

180.    MOH is considering the setting up of “amenity wards” for paying patients. To
decongest queues and overcrowded wards, MOH is exploring the “voluntary payment” of fees for
improved hotel services in paying wards (called “amenity wards” in other countries) and OPD1
services. The proposed policy is targeted to those with ability to pay. LATH (1999) has
conducted a study estimating the likely revenue to charging MK60 for each OPD1 visit, and


                                                 71
MK75 for an inpatient stay at the private wing of district hospital, with MASM patients being
charged full-cost. The revenue yield of this scheme was estimated to be MK1.4 million per
district, or a total of MK33.6 million for the then-24 districts (equivalent to US$780,000 at the
then-prevailing exchange rate of US$1=MK43). Using FY98/99 total recurrent budget for
districts (MK341 million), the cost-recovery rate of this scheme is about 10 percent, comparable
to the rates obtaining in other African countries with cost sharing programs. Formalizing this
revenue-generation scheme, however, necessitates policies of revenue-retention at the point of
service and the “additionality” of these revenues to the annual budget allocation (resource-
augmenting, rather than “cost-shifting” 17 ), otherwise the incentive effects of the fee revenues will
not translate into improved services.

181.     As part of an overall goal of granting autonomy to the central hospitals, MOH also
intends to charge fees to inpatient admissions and outpatient consultations in these facilities.
Discussions about the proposed granting of hospital autonomy to the central hospitals has been
going on since the mid-1990s, with little advancement. An assessment made in 1999 (Shehata
and Cripps, Sept. 2000) found that the Queen Elizabeth Central Hospital (QECH) and Lilongwe
Central Hospital (LCH) have limited control over their operations and have insufficient financial
and human resources. This assessment proposed that decisionmaking authority be transferred
from MOH to these hospitals for five key functions: strategic management, procurement,
financial management, human resources management, and administration. After three years of
inaction, the hospital autonomy initiative was revived in 2002, with the production of a draft
briefing in November, the holding of a workshop in December, the appointment of a Hospital
Autonomy Technical Assistant (HATA), the initiation of a Joint Implementation Plan (JIP), and
the drafting of a hospital autonomy bill in mid-2003 (Faulkner, July 2003), which is presumably
being refined. Achieving the autonomy objective will be a complicated task; the minutes of the
workshops and meetings on this issue attest to the familiarity of the participants to the issues,
although actually addressing them will be far more daunting.

         (a)      Hospitals’ legal structure and personality. MOH policymakers appear to favor
                  an “incremental” rather than “radical” approach to hospital autonomy, preferring
                  “minimal adjustment to the current legal structure” rather than completely
                  severing the two hospitals from MOH as independent entities. The draft hospital
                  autonomy bill calls for the creation of a hospital board for each of the hospitals,
                  with some members appointed by the Minister of Health and others acting on an
                  ex-officio basis.

         (b)      Referral policy regarding use of tertiary hospitals. The intention is to require a
                  tertiary hospital client to present a referral letter for a lower-level facility, or
                  failing to do so, pay a by-pass fee which is designed to encourage care-seeking at
                  the appropriate level. The absence of district hospitals in Blantyre and Lilongwe
                  complicates this problem, since both QECH and LCH also act as district
                  hospitals. Thus, any patient in these catchment areas will, by default, move from
                  a health center to the tertiary hospital.

         (c)      Financial support from the government. The central government is expected to
                  finance the EHP services that the autonomous hospitals will render, possibly

17
   “Cost-shifting” is the term widely used in Anglophone Africa for a health facility‟s tendency to rely
increasingly on fees to support recurrent expenditures when the central government withholds an increasing
amount of the facility‟s annual budget allocation as fee revenues increase. The term is a derision of “cost -
sharing” which is a co mmon euphemism for any fee-based revenue program.


                                                     72
                through a memorandum of agreement, similar to the health service agreement
                that CHAM facilities sign with the government. This is a tricky issue because
                tertiary hospitals have higher per-unit costs, and in the interest of efficiency, as
                many EHP patients as possible should be dealt with at the district hospital level
                with lower per-unit costs. Reimbursing these tertiary hospitals at their tertiary-
                level rates provides them incentives to get as many EHP patients as possible,
                while reimbursing them at district-hospital rates causes them to incur deficits.

        (d)     User fees and other extra-budgetary sources of revenue. The draft policies call
                for revenue retention in autonomous hospitals and fee-setting that gradually
                increases to market rates as quality improves. There is also a realistic realization
                that these hospitals can never be expected to be fully able to finance their
                financial operations through fees. Public solicitations are being encouraged as
                additional sources of revenues, but whether the tertiary hospitals can indeed
                receive and manage grants directly from donors without the intermediation of
                MOH may depend on their degree of autonomy. In any case, there is a need for a
                careful financial analysis and projections of the various revenue options for these
                hospitals, far more detailed than what is currently available.

        (e)     Staff transition from civil servants to employees of autonomous agencies. This
                complex issue should be given more attention as it has significant staff-incentive
                (job security, performance), financial (pensions, salaries), and management
                implications.

182.    MOH foresees establishing a social health insurance (SHI) program for formal-
sector employees. This proposal is at a very early stage of discussion and no documents have
been prepared to initiate the process. Nevertheless, critical questions that need to be dealt with

        (a)     Mandatory or voluntary program. By definition, an SHIs is a mandatory
                program. However, in practice, this is often difficult to impose unless there is a
                reliable system where compliance can be monitored, especially among private
                enterprises. In addition, many employees in Malawi (from the government and
                private sector alike) are already purchasing insurance coverage under the private
                Medical Aid Scheme of Malawi, and the imposition of a mandatory SHI may be
                deemed duplicative.

        (b)     Options for covered population. Most SHIs start with civil servants of central
                ministries, then branch out to include employees of local governments and
                parastatal corporations, then to large private corporations, and eventually to all
                workers in the formal sector. The SHI could also initially cover only the
                employee, and then the spouse, and then the dependents. Given the small size of
                the Malawian civil service – among ten ministries/departments included in the
                July 2003 report of DHRMD and UNDP, there were 68,292 established posts, of
                which 39,717 are filled – the SHI could be initiated quickly.

        (c)     Options for alternative benefit packages. Assuming the EHP is indeed provided
                free, the SHI program will only cover admission to tertiary hospitals for chronic
                or catastrophic care. This implies that the providers are, for the most part, tertiary
                public and private hospitals and a few district hospitals that could be upgraded to
                provide tertiary care. Limiting the benefit package to catastrophic and chronic
                care may dampen the enthusiasm of potential members who want a more


                                                 73
                 comprehensive package but wish to obtain services not from government and
                 CHAM facilities – where EHP is presumably free – but in the growing private
                 sector. Benefit-package analysis also needs to consider whether HIV/AIDS will
                 be covered, given its magnitude and implications on financial sustainability and
                 equity. Finally, medical referrals abroad – currently treated as a separate line
                 item in the budget and, according to informants, “rationed” on a rather ad-hoc
                 basis among senior civil servants – should be folded over into the SHI. For
                 equity and efficiency reasons, these referrals are better treated within an overall
                 health insurance framework.

        (d)      Premium or contribution rates. As a hallmark of social solidarity promoted under
                 SHI, a uniform contribution rate is usually imposed (typically 2-5 percent of a
                 member‟s salary, but sometimes with a defined ceiling), providing for a uniform
                 benefit package for all members and/or their eligible dependents. It is possible to
                 offer a differentiated benefit package with different contribution rates, but this is
                 unwieldy as it requires asking the member on a periodic basis what his/her
                 preferred package is; this arrangement is also inequitable. A more basic issue is
                 whether the government can pay its share of the SHI contribution as an employer.
                 Using MASM‟s Standard Scheme costing MK1,200/member/year, and assuming
                 40,000 civil servants, and assuming 50/50 sharing between worker and employer,
                 the expected government contribution for a typical Malawi civil servant‟s family
                 with 6 eligible members and dependents is MK144 million (the other MK144
                 million to be contributed by civil servants). Alternatively, if one uses MASM‟s
                 Executive Scheme of MK4,800/member/year, the government contribution under
                 the same assumptions as the previous case is MK576 million 18 .

        (e)      Options for providers and provider payment system. The SHI program will have
                 to decide whether to use Government central hospitals only, or tertiary CHAM
                 hospitals, or private doctors and hospitals (in which case, it needs to undertake an
                 inventory). Additionally, a decision has to be made on the type of payment
                 system for individual/professional and institutional/facility providers. Given the
                 lack of documented utilization experience and costing data, this task is going to
                 be challenging. The decision also has to depend on the sophistication of the
                 information system available.

        (f)      Alternative institutional arrangements. The conventional SHI strategy in
                 developing countries has been to establish a separate, fully-staffed insurance
                 organization doing the entire range of health insurance functions. Given the
                 extremely limited insurance capacity in Malawi, it may be more advisable to tap
                 an existing organization, such as MASM, to provide defined services to the SHI
                 fund under an “administrative services only” (ASO) arrangement. An ASO is a
                 contract between an insurance company (in this case, MASM) and a self-funded
                 plan (the SHI fund) where the insurance company performs administrative
                 services only and does not assume any risk. These services usually include
                 claims processing, but may include other services such as actuarial analysis,
                 utilization review, information systems, and so forth.


18
  These examp les are used only to provide orders of magnitude of what the government expects to shell
out under an SHI. The SHI scheme may very well have lower contribution rates that the for-profit MASM
scheme.


                                                  74
183.    Dedicated technical assistance and agreed-upon process are required to launch the
above initiatives on financing non-EHP. Further analysis and synthesis, design, stakeholder
consultation, and policy development (including legal reform) are required to translate the above
ideas – user fees for amenity wards, hospital autonomy, and social health insurance – into action.
Although many of these ideas have been percolating in Malawi since the 1990s, little has been
achieved, despite an intermittent trickle of consultants, due to a combination of factors including
weak technical capacity especially at the Planning Department, continuing disconnect between
the technical agreements reached within MOH and the political support needed at Parliament for
key reforms to be made, and lack of donor consensus on the importance of these initiatives.




                                                75
                                VII.        Way Forward


184.    In the area of human resources for health, this study proposes the following:

        (a)     Strengthen the basis for HR planning by conducting actual headcount of MOH
                and subvented CHAM health workers, an assessment of health workers‟
                workload, and a study on the labor market for health workers.

        (b)     Table for wider-scale discussion the proposed short-term supplementary staffing
                scheme for health personnel in Malawi. The illustrative terms of reference for
                this scheme is found in Annex D.

        (c)     Conduct a study on the institutional arrangements for improved human resource
                management. The illustrative terms of reference for this study is found in Annex
                E.

        (d)     Procure long-term technical assistance services on human resource management.
                The illustrative terms of reference for this TA is found in Annex F.

185.    In the area of health financing, this study supports government efforts to accomplish the
following:

        (a)     Produce more accurate and consistent health budget allocation and expenditure
                data. This can be done within the framework of an expenditure tracking study.

        (b)     Plan and coordinate the financing and delivery of community health services.

        (c)     Scale-up, monitor, and fine-tune the health service agreements with CHAM and
                consider extending this same performance-based contracting scheme to other
                NGOs.

        (d)     Formulate a realistic medium-term plan for fiscal decentralization, including the
                reform of district allocation formula in line with eventual devolution.

        (e)     Include a greater representation of private sector providers in the policy
                development process.

        (f)     Conduct analytical work and feasibility assessment of health insurance for civil
                servants or (wider) formal sector employees.

        (g)     Develop and implement policy on hospital autonomy for central hospitals.

        (h)     Conduct analysis of benefit-incidence of government and donor health
                expenditures.

        (i)     As part of the SWAp, consider developing a “transition plan” for current vertical
                financing so that more of these can be folded over into the pooled funding, or at
                least identified in the development budget. This should include global health
                initiatives.



                                                76
                                      References
Afrobarometer (April 2002). Key Findings About Public Opinion in Africa. Briefing Paper
No. 1. www.afrobarometer.org

Ager, A., A. Chinula, et al (2003). Review of the Organizational Structure of the Central
MOHP in the Light of the Transition to a Developed Health System and Related Reforms .
Final report submitted to the European Union HSRD Programme. Liverpool Associates in
Tropical Health.

Aitken, J.M. and J. Kemp (2003). HIV/AIDS, Equity and Health Sector Personnel in
Southern Africa. EQUINET Discussion Paper No. 12. Regional Network for Equity in Health
in Southern Africa in Cooperation with Oxfam GB.

Anonymous (March 1995). Community Financing: Experiences and Options .

Anonymous (January 1998). Malawi: Cost Recovery and Private Sector Participation in the
Health Sector.

Anonymous (2003). Report on the Identification of Critical Areas for Capacity
Replenishment in the Public Service. Study submitted to the Malawi Department for Human
Resources Management and Development. Appendix 5.

Boex, Jamie, Randson Mwadiwa, and Reckford Kampanje (April 2001). Malawi
Intergovernmental Fiscal Trasnfers Study. Government of Malawi, United Nations Capital
Development Fund/ United Nations Development Program, and Georgia State University.

Broadhead, R. L. and A.S. Muula (2002). Creating a Medical School in Malawi: Problems and
Achievements. British Medical Journal 325 (7360): 384-7.

Buchan, J. (2003). Here to Stay? International Nurses in the UK. London, RCN.

Budget and Finance Committee of Parliament (July 2003). Report on the 2003/04 Budget.
Malawi National Assembly.

Burns, Monica, Khoti Gausi, Emma Jefferys, Takondwa Mwase, and Claudio Politi (May 1999).
Development of a Health Care Financing Strategy. Liverpool, U.K.: Liverpool Associates in
Tropical Health. Study funded by DfID.

Castro-Leal, Florencia, Julia Dayton, Lionel Demery, and Kalpana Mehra (2000). Public
Spending on Health Care in Africa: Do the Poor Benefit? Bulletin of the World Health
Organization 78 (1): 66-74.

Claeson, Mariam, Tazim Mawji, and Christopher Walker (December 2000). Investing in Best
Buys: A Review of the Health, Nutrition and Population Portfolio, FY 1993-99. HNP
Anchor. The World Bank.

Dussault, Gilles and Khama Rogo (June 2003). Back to Office Report, World Bank Mission to
Malawi.




                                             77
ESA NHA Network (2000). National Health Accounts in Eastern and Southern Africa: A
Comparative Analysis.

FAHAMU Learning for Change (n.d.). Profile of Health and Training Infrastructure in
Malawi.

Franco, Ciro, John C. Quinley, Bettina Schwethelm, Timothy E. Kachule, and Barton R.
Burkhalter (n.d.). Employer-Based Programs in Maternal and Child Health: Project
HOPE’s Strategy for Attaining Long-Term Sustainability of Health Promotion. BASICS
Project.

Global Fund (January 2004). Guidelines for Proposals: Fourth Call for Proposals . Geneva.

Government of Malawi (August 1997). Human Resources Situational Analysis. Ministry of
Health and Population.

Government of Malawi (November 2002). SWAp Implementation Plan: SWAps Today,
Tomorrow, Together. Final Report, Vols. 1 and 2.

Government of Malawi (November 1999). Malawi National Health Plan 1999-2004. Volume 3
– Human Resources Plan. Ministry of Health and Population.

Government of Malawi (2002). Health Service Commission Regulations.

Government of Malawi (November 2002). SWAp Implementation Plan: SWAps Today,
Tomorrow, Together. Final Report, Vols. 1 and 2.

Government of Malawi (2003/04). Approved Estimates of Expenditure on Recurrent and
Capital Budget for Financial Year 2003/04. Detailed Estimates. Volume 12 (Vote 310).
Zomba, Malawi: Government Printer.

Government of Malawi and UNDP (2002). Public Sector Management Program 2002-06.
Programme Support Document.

Gwatkin, Davidson R., Shea Rustein, Kiersten Johnson, Rohini Pande, and Adam Wagstaff (May
2000). Malawi Socio-economic Differences in Health, Nutrition and Population. HNP
Poverty Thematic Group. The World Bank.

Harris, D. and J. Nyondo (2004). Communication Strategy for Health Sector Reform.
Ministry of Health.

Hornby, Peter and S. Ozcan (January 2003). Human Resources for Health, Sector Strategic
Plan 2003 to 2013. Report submitted to the Ministry of Health and Population.

Hozumi, Dai (October 2003). Status of Health Services Facilities in Malawi. Findings from
Preliminary Analysis of the Malawi Health Facility Survey 2002. Study submitted to the
Ministry of Health and Population and funded by JICA.

JHPIEGO (no date). Strengthening Capacity to Monitor Training and Licensure of Nurses
and Midwives: the Malawi Experience. (rmagarick@jhpiego.net).



                                            78
Kaluwa, Ben and Ron Managani (n.d.). Pro-Poor Budgeting in the Social Sectors, Efficiency
or Uninformed Patronage? Some Answers from Malawi.
www.poverty.worldbank.org/library/view/13389.

Kamuzu College of Nursing (2003). KCN’s Priority Requirements for Capacity Building.

Kadzandiwara, John M. and Wycliffe R. Chilowa (December 2001). The Role of Health
Surveillance Assistants (HSAs) in the Delivery of Health Services and Immunization in
Malawi. University of Malawi, Centre for Social Research. Study funded by UNICEF.

Kathyola, Janet (March 2003). The Status of Human Capacity Development for an Effective
Response to the HIV/AIDS Epidemic in Malawi.

Kenwrick, Robert J. and Warwick C. Temple (May 1999). Job Grading and Salary
Restructuring Project. Volume 2. Draft final report submitted to the Malawi Civil Service,
Department of Human Resource Management and Development. Bryher Partnership.

Kolehmainen-Aitken R-L (1998). "Decentralization and Human Resources : Implications and
Impact." Human Resource Development Journal 2(1): 1-16.

Koot, J. and T. Martineau (2004). Institutional Appraisal. Draft report for the College of
Medicine, University of Malawi.

Lexomboon, D. (2003). Recruitment and Retention of Human Resources for Health in Rural
Areas: A Case Study of Dentists in Thailand. PhD thesis. Liverpool University.

Maqc, J. and W. van Lerberge (2000). Managing Health Services in Developing Countries:
Moonlighting to Serve the Public? In Providing Health Care Under Adverse Conditions:
Studies in Health Services Organization and Policy. P. Ferrinho and W. van Lerberge.
Antwerp, ITG Press.

Martineau, Tim, K. Sargent, et al (2001). Institutional and Human Resources Apprais al of the
Sexual and Reproductive Health Programme . Mission report submitted to the Ministry of
Health and Population. LATH and JSI/U.K.

Makuti, M., A. Gonani, A. Macheso, B. Ndyabangi, and S. Kinoti (November 2003). Impact of
HIV/AIDS on the Health Workforce in Malawi. Slide presentation during the Regional Health
Ministers‟ Conference. Livingstone, Zambia.

Malawi Economic Justice Network (MEJN) (2004). Civil Society Contribution to the
2003/2004 Supplementary Appropriation Bill. Lilongwe, Malawi. www.mejn.org

Malawi Economic Justice Network (MEJN) (2003). How Are Services Delivered to the People
in Malawi? Results of the Service Delivery Satisfaction Survey. Lilongwe, Malawi.

Malawi Government (1997). Approved Estimates of Expendiutre on Recurrent and Capital
Accounts for the Fiscal Year 1997/98. Budget Document No. 4. Zomba, Malawi: The
Government Printer.




                                               79
Malawi Government (2003). Approved Estimates of Expenditure on Recurrent and Capital
Accounts for the Fiscal Year 2003/2004. Budget Document No. 4. Zomba, Malawi : The
Government Printer.

Mbalame (1997). Hospital autonomy.

Medical Aid Society of Malawi (MASM). www.masm.malawi.net.

Medical Council of Malawi (2002/03). List of Registered Health Professionals . To be
published in The Malawi Government Gazette for 2002/03.

Ministry of Finance (2003). Budget allocation and expenditure tables.

Ministry of Finance (2003). Budget Document No. 4.

Ministry of Health and Population (1999). Human Resource Plan.

Ministry of Health and Population (1999). Malawi National Health Plan, Volume 3: Health
Sector Human Resources Plan. Lilongwe.

Ministry of Health and Population (May 1999). Malawi National Health Plan 1999-2004.
Lilongwe, Malawi.

Ministry of Health and Population (March 2000). Project Financing Proposal for Human
Resources Development in the Health Sector.

Ministry of Health and Population (August 2001). Malawi National Health Accounts: A
Broader Perspective of the Malawian Health Sector. Planning Department.

Ministry of Health and Population (May 2002). The Malawi Essential Health Package, Annex
2: Additional Detail on the Costing of the EHP. EHP Working Group.

Ministry of Health and Population (September 2002). Towards Institutionalizing the Essential
Health Package [EHP]. Planning Department.

Ministry of Health and Population (June 2003). Situation Analysis on Human Resource in the
Light of the EHP Implementation. Planning Department.

Ministry of Health and Population (October 2003). Status of Health Services Facilities in
Malawi: Findings from Preliminary Analysis of the Malawi Health Facility Survey 2002.
Study funded by JICA.

Ministry of Health and Population (November 2003). NORAD Top-Up Support to Districts.
Progress Financial Report for Period Ended on 31 st October 2003.

Ministry of Health (2004). Human Resources in the Health Sector: Issues and Challenges.
(April 2004)

Ministry of Health and Population (MOHP) and Liverpool Associates in Tropical Health (LATH)
(November 2003). Forecasting of Resource Envelop for the Health Sector SWAp
Programme of Work. Study commissioned by the Ministry of Health.


                                              80
Ministry of Health (March 2004). A Joint Programme of Work for a Health Sector Wide
Approach (SWAp), 2004-2010. Department of Planning.

Muula, A. and A. Phiri (2003). Reflections on the Health Worker’s Strike at Malawi’s Major
Tertiary Hospital, the QECH, Blantyre, 2001: A Case Study. Nursing Ethics 10 (2): 208-214.

National AIDS Commission (March 2003). Human Capacity Development for an Effective
Response to HIV/AIDS in Malawi. Summary of Discussions Held on 27th and 28th March,
Lilongwe Hotel, Malawi.

National Economic Council (June 2001). The Determinants of Poverty in Malawi 1998: An
Analysis of the Malawi Integrated Household Survey, 1997-98.

National Statistics Office (2000). Malawi Demographic and Health Survey 2000. Measure
DHS+ and ORC Macro.

National Statistics Office (2003). Malawi Core Welfare Indicators Questionnaire (CWIQ)
Survey. Zomba, Malawi.

NORAD Mission (2003). Report from a Visit to Malawi on Different Issues Related to
NORAD and SIDA ORT Top-Up.

OECD (2003). DCR Statistical Annex 2003. www.oecd.org.

Office of the President and Cabinet (n.d.). Human Resource Capacity Replenishment in the
Public Sector in the Wake of the HIV/AIDS Crisis: A Concept Paper. Department of Human
Resource Management and Development.

Office of the President and Cabinet (October 2003). National HIV/AIDS Policy: A Call for
Renewed Action. National AIDS Commission.

Picazo, Oscar F. (2001). Better Health Outcomes from Limited Resources: Focusing on
Priority Services in Malawi. Africa Region Human Development Working Papers Series. The
World Bank.

Planning Department (2002). A Six-Year Emergency Pre -Service Training Plan. Lilongwe.
MOHP.

Republic of Malawi (1993). Report of a Workshop on the Introduction of Cost Sharing
Scheme in Government Health Facilities. Mangochi, June 20-23.

Republic of Malawi (August 2001). Interim Poverty Reduction and Growth Strategy Paper:
A Roadmap.

Republic of Malawi (November 2003). A Joint Programme of Work, 2004-2010. Ministry of
Health and Population, Department of Health Planning Services.

Schenck-Yglesias, C. (2003). Malawi Health Human Resource Information Systems:
Supporting the Development and Monitoring of the Human Resource Deployment and
Training Policies and Plans.


                                             81
Standing, H. (2000). Gender - a Missing Dimension in Human Resource Policy And Planning for
Health Reforms. Human Resource Development Journal 4(1): 27-42.

St. Joseph Hospital and School of Nursing (2002). Annual Report 2002. P.O. Box 5505.
Limbe, Malawi.

United Nations Development Program (February 2002). The Impact of HIV/AIDS on Human
Resources in the Malawi Public Sector. Final Report.

University of Southampton (2003). Appropriateness of User Fees for Family Planning in
Malawi: Findings from Focus Group Discussions. Fact Sheet No. 7. Reproductive Health
Research. Study funded by DfID.

Valentine, T.R. (2003). Towards the Development of a Medium-Term Pay Policy for the
Malawi Civil Service. Final Report submitted to the Department for Human Resources
Management and Development.

Weiner, S. J. (1989). 'Source Force' and the Nepal Medical Profession. Social Science and
Medicine 29(5): 669-75.

World Health Organization (2003). World Health Report. Geneva: WHO.

World Bank (September 2001). Malawi Public Expenditures: Issues and Options. Africa
Region Report No. 22440 MAI.

World Bank (2003). Africa Development Indicators . Washington, D.C.




                                              82
ANNEX A: A BRIEF ANALYSIS OF THE MEDICAL AND ALLIED PROFESSIONS
                           IN MALAWI


         This Annex analyzes the contents of the Medical Council of Malawi‟s “List of Registered
Health Professionals” published in The Malawi Government Gazette in 2002/2003. The database
derived contains the names of all the registered medical and allied professionals in the country,
their profession including specialization, their sector employment (private or public), the date
they registered, their degree including where and when they were obtained, and their postal
address by district. Caution must be exercised in interpreting the district allocation of health
professionals, since this relied solely on postal address which does not necessarily reflect the
professional‟s location of work. Moreover, since the publication of the List, Malawi has
incorporated three more districts. The analysis does not take these new districts into account and
instead uses the initial 24 districts.

         Malawi suffers from severe scarcity of medical professionals. In 2002/2003, 1,111
health professionals were registered with the Medical Council of Malawi, of which only 194 (17.
5 percent) were physicians/medical practitioners including five interns, and only 16 dentists for
the entire country (Table A-1). For a country of around 11 million, Malawi has only one clinical
psychologist, two occupational therapists, eight physiotherapists, five doctors of osteopathy, and
four opticians. Much of the remainder consists of technical staff and support personnel.

             Table A-1. Total Number of Medical Professionals in Malawi, 2002/03

                      Professions                            Number           Population Per
                                                                               Professionals
Physicians/ medical practioners (incl. interns)                199
Doctors of osteopathy                                           5
Dentists & dental tech.                                        18
Opticians                                                       4
Clinical officers (incl. anesthetic and orthopedic COs &       275
interns)
Medical assistants                                             404
Occup‟l, therapists, physiotherapists, rehab‟n                 11
technicians
Radiography technicians & assts                                  6
Lab and medical lab technologists, technicians, assts.          87
Enrironmental health officers and assts                         48
Others                                                          54
Total                                                         1,111

          Spatial maldistribution of medical professionals is severe. Blantyre and Lilongwe
districts have the largest percentage (42.4 percent) of the country‟s medical professionals (Table
A-2). Together, the two districts have 76.8 percent of all physicians/medical practitioners and
81.3 percent of all dentists. The districts with the lowest proportions of medical professionals are
Chitipa (0.9 percent), Ntchisi (0.6 percent), Nkhata Bay (0.5 percent) and Nsanje (0.4 percent).
On the whole, the inequitable distribution of health personnel is so severe that nine out of the 24



                                                 83
districts do not have a single physician/medical professional, and 19 districts do not a single
dentist.

                Table A-2. Number of Medical Professionals by District, 2002/03

                        District      No.      %         District       No.    %
                     Blantyre         255     23.0 Mulanje              36     3.2
                     Chikwawa         31       2.8    Mwanza            15     1.4
                     Chiradzulu       15       1.4    Mzimba            84     7.6
                     Chitipa          10       0.9    Ntcheu            22     2.0
                     Dedza            28       2.5    Nkhata Bay         5     0.5
                     Dowa             46       4.1    Nkhotakota        19     1.7
                     Karonga          18       1.6    Nsanje             4     0.4
                     Kasungu          40       3.6    Ntchisi            7     0.6
                     Lilongwe         216     19.4 Rumphi               42     3.8
                     Machinga         34       3.1    Salima            31     2.8
                     Mangochi         24       2.2    Thyolo            32     2.9
                     Mchinji          19       1.7    Zomba             64     5.8
              Note: 14 professionals (or 1.3 percent of total) had unknown districts.

         Most medical and allied professionals are in government service, but private sector
employment is becoming significant. Majority (87 percent) of all registered medical and allied
professionals work in the public sector, with only 13 percent of health staff active in the private
sector. However, a significant proportion of general practitioners (19.8 percent), medical
specialists (17.5 percent) and dentists (37.5 percent) are in the private sector (Table A-3).

 Table A-3. Public/Private Distribution of Selected Medical and Allied Professionals in Malawi,
                                            2002/03

             Profession                   Total        % in Public Sector     % in Private Sector
General practitioners (excl. interns)      137                80.2                   19.8
Medical specialists                        57                 82.5                   17.5
Doctors of osteopathy                       5                100.0                    0.0
Dentists                                   16                 62.5                   37.5
Dental therapists                          11                 90.9                    9.1
Occupational and physio therapists         10                 90.0                   10.0
Others                                      6                100.0                    0.0

        Majority of all doctors in Malawi are foreign educated. Among the 220 registered
doctors (including general practitioners, specialists, interns, doctors of osteopathy, and dentists),
74 percent have at least one degree from a country other than Malawi. Some 130 out of a total
162 foreign-educated doctors in Malawi reside in the Blantyre (79) and Lilongwe (51) districts
(Table A-5). A higher percentage of private sector doctors are foreign educated (e.g. 90.7 percent)
than public sector doctors (69 percent).




                                                  84
Table A-4: District Distribution of Medical Specialists by Education and Sector of Employment,
2002/03

                 District   Total     Education              Sector Employed
                                    Foreign Local      Public Sector Private Sector
                 Blantyre   29      96.6    3.4        75.9          24.1
                 Chitipa    1       100.0   0.0        100.0         0.0
                 Lilongwe   19      89.5    10.5       84.2          15.8
                 Mangochi   1       100.0   0.0        100.0         0.0
                 Mulanje    2       100.0   0.0        100.0         0.0
                 Thyolo     3       100.0   0.0        100.0         0.0
                 Zomba      2       100.0   0.0        100.0         0.0
                 Total      57      94.7    5.3        82.5          17.5


         The number of specialists is very limited, and most of them were trained abroad.
More than two-thirds (71 percent) of all 194 medical practitioners in Malawi are general
practitioners, with only less than a third (29 percent) as specialists. The majority of these
specialists are surgeons (26 percent), followed by specialist physicians (19 percent),
obstetricians/gynecologists (18 percent), pediatricians (9 percent) and others. Almost all (94
percent) of all registered specialists are educated abroad, of whom 17.5 percent are employed in
the private sector (see Table A-4 above). The district distribution of specialists in Malawi is also
highly uneven, with the Blantyre and Lilongwe districts together claiming 48 (84 percent) of all
57 specialists in the country.

          There has been a fairly steady increase in the number of medical practitioners
registering throughout much of the 1990s and early 2000s. The Medical Council of Malawi
lists registration figures commencing from 1988, with no previous registration years listed. This
possibly explains the relatively high number of medical practitioners (28) and dentists (4) shown
to have registered in 1988, followed by much lower numbers registering until 1991. Whatever the
reason, what is clear is that the registration figures for medical practitioners generally increased
between 1991-2003, with a relatively steady and steep incline particularly evident after 1997
(Table A-5). A significant and fairly consistent proportion of those registering are foreign
educated. There is no indication of an increase in the proportion of medical practitioners working
for the private sector. As for dentists, the low proportion of them registering between 1988-2003
means that not real trend can be established.

Table A-5. Number of Medical Practitioners (Including Specialists and Interns) and Dentists
Registering in Malawi between 1988 and 2002

Year                  Medical Practitioners                                    Dentists
          Total        Education       Sector Employed        Total        Education    Sector Employed
                    Foreign Local Public Private                        Foreign Local Public Private
 1988       28        27        1        8        20            4          4        0     2         2
 1989        8         8        0        6         2            0          0        0     0         0
 1990        0         0        0        0         0            1          1        0     1         1
 1991        5         5        0        3         2            0          0        0     0         0
 1992        5         5        0        5         0            1          1        0     0         0
 1993        1         1        0        0         1            0          0        0     0         0



                                                 85
 1994      10         8         2         9        1         1          1        0         1       1
 1995       7         2         5         6        1         0          0        0         0       0
 1996       7         5         2         6        1         0          0        0         0       0
 1997       9         5         4         6        3         0          0        0         0       0
 1998      14         7         7        13        1         1          1        0         1       1
 1999      14        11         3        11        3         2          2        0         1       1
 2000      23        12        11        22        1         0          0        0         0       0
 2001      31        17        14        31        0         2          2        0         0       0
 2002      37        27        10        36        1         4          4        0         0       0

        Most of the registered physicians/ medical professionals are young. Just under half
(45.9 percent) of all registered doctors have between 1-12 years of medical experience. As Table
A-6 shows, 14.1 percent of all doctors graduated between 1990-1995; 25.9 percent between 1995-
2000; and 5.9 percent between 2000-2003. Although there are few very old doctors - only 18.3%
of doctors graduated before 1975 – a significant proportion (33.3 percent) graduated between
1975-1990. Assuming that graduation immediately led to work, at the end of 2002, roughly 30
percent of all doctors working in Malawi had 1-7 years of medical experience; roughly 20 percent
had 8-12 years of experience; roughly 30 percent had 13-27 years of experience; and roughly 20
had 28-52 years of experience.

                    Table A-6. Periods that Doctors and Dentists Graduated

                   Period Graduated       % of All Doctors       % of All Dentists
                   1950-54                      1.4                    0.0
                   1955-60                      0.5                    0.0
                   1960-64                      1.4                    0.0
                   1965-70                      6.4                    0.0
                   1970-74                      8.6                   12.5
                   1975-80                     10.5                    6.3
                   1980-84                     11.4                   18.8
                   1985-90                     11.4                   12.5
                   1990-94                     14.1                   18.8
                   1995-00                     25.9                   25.0
                   2001-2003                    5.9                    6.3
                   Unknown                      2.7                    0.0




                                              86
      ANNEX B: SALARY SCALES AMONG INSTITUTIONS THAT EMPLOY HEALTH
                    PERSONNEL IN MALAWI, CIRCA 2003/04
     Institution       Salary Scale Per          Allowances                Part-time Fees Per Day
                            Month
Medical Rescue       K26,000 – K33,000      Transport               Are planning to adjust increase the
Services (MARS)                             Housing                 Salary K40,000 in March part-
Lilongwe                                    Night duty              time/night duty allowance: K200.00
UNC Research         K27,000 – K40,000      Transport               Orientation course with allowances
Project                                     Housing                 3 years of service attracts 3 month‟
Lilongwe                                    Tax free                salary as bonus.
                                                                    International short courses and MSC
                                                                    Night duty of K1,200.
Malaria Research     K14,008 – K35,531      40% Housing             None
Project                                     Allowance
                                            Duty allowances
                                            2-10% annual
                                            increment
City-Centre Clinic   K7,000 – 25,000        50% Housing             Part-time Nurses are paid:
                                            Allowance               K200 per day shift
                                            Professional            K500 per night shift
                                            Allowance
                                            Pension Scheme (6%
                                            employee, 13%
Light house One      Government salary                              Is planning to employ their own by
Ward                 plus                   Field allowances.       March. And proposed salary K27,000
                     K5,000 top-up          Workshops.              – K40,000
BLM                  K15,000,00-            K10,000,00-             K300.00 – K500.00 per day
                     K40,000.000            K17,000.00
                                            Housing allowance
                                            Medical Scheme
                                            Gratuity (17.5% of
                                            Basic Salary)
CHAM                  Government Salary     TAs are employed as     None
TA – CTO              plus top-up K1,669 – TOs.
                      K4,932                Government
                                            approved allowance
                                            Top-up (amount not
                                            disclosed)
Ministry of Health    K1,669              – Housing Allowance       No official part-time policy
TA – P5               K9,8033.month         (K1,500 – K12,000)      No night duty allowance
                                            Special       Medical   No transport allowance
                                            Allowance K1,000-
                                            K1,800
                                            Professional
                                            Allowances K1,800-
                                            K2,300
      Source: Ministry of Health, 2004




                                                  87
ANNEX C: DIAGRAM OF THE APPLICATION PROCESS




                    88
   ANNEX D: THE PROPOSED SHORT-TERM SUPPLEMENTARY STAFFING
            SCHEME FOR HEALTH PERSONNEL IN MALAWI

         Background. This scheme provides staffing of health system in a state of collapse
primarily due to staff shortages. While traditional longer-term solutions must be sought,
including the strengthening of planning, recruitment and retention systems, more radical solutions
are needed in the short term to plug the staffing gap. With Malawi‟s current labor market
situation, there are limited options. An important segment of the labor market has been
identified, namely those staff who have been trained as health workers, but who are no longer
working in the health sector, or are not working at all. It has been suggested that there might be
8-1,200 nurses in this category (MoH, 2004).

        The purpose of the short-term supplementary staffing scheme (SSSS) is to tap into this
segment of the labor market and to use these resources as flexibly and efficiently as possible. The
scheme is essentially short-term and is based on the assumption that the longer-term strategies
will reduce the problems of attraction and retention over the next 5 years which would effectively
render the scheme redundant.

         The Scheme will start with one target group of the workforce – possibly ENMs – who are
currently not working in the health sector. Before starting the scheme a survey would be needed
to identify the approximate number of ENMs who might move to the government health sector,
and to establish the competitive compensation packages that they might receive. The survey
would also have to test the assumption that increased pay would be sufficient to attract the health
workers back into government service.

        DHOs would be invited to request ENMs to fill vacant posts at remote health centres.
ENMs would be recruited, employed on contract by an independent agency contracted by the
Ministry of Health, and posted to the health centres. Assistance with housing (and possibly some
other facilities) would be provided, as well as an attractive fee.

         Performance (mainly attendance) would be closely monitored (though mainly only
attendance initially) jointly by the DHO and the agency. Contracts would be terminated if
performance falls below an agreed standard. Contract renewal would be on the basis of a
continued need and a more rigorous performance review of the post holder.
A communication strategy would be needed: to advertise the scheme to employers and potential
recruits; and to explain the scheme to existing staff who may well feel disgruntled as working in
the same organization as these more highly paid staff.

         Following a successful trial of the scheme with one cadre, the scheme could be extended
to others – either large groups with severe shortages e.g. clinical officers, or small groups with a
key function e.g. pharmacists or lab technicians.

         As conditions of service improve leading to improvement of attraction and retention of
the target groups of health personnel, the scheme would be phased out for these staff. Attrition
rates from the public sector are an important indicator of improvement of conditions of service
and would need to be monitored closely.

        The scheme would be funded by donors on an ongoing basis. An agency would be
contracted to run the scheme and would have a representative based in the Ministry of Health.



                                                 89
     System Requirements. The agency would be required to carry out the following functions:
(a) responding to DHO requests, (b) recruitment of candidates, (c) placement of candidates, (d)
payment of fees, (e) supervision of candidates (with the DHO).

     The Ministry of Health would be required to: (a) recruit and contract the recruitment agency;
(b) provide funds for agency staff fees; develop and implement communication strategies; (c)
oversee the scheme; (d) monitor staffing data to identify new candidates groups for the scheme or
to terminate existing schemes.

     Benefits and Risks. As a radical solution to the staffing crisis, there are bound to be risks, but
at the same time there will also be benefits to the provision of health services.

     The scheme will be: Rapid (and much faster than training new staff – especially mid and high
level staff). Responsive: it can be stopped if not longer needed; targeted groups can be changed
according to demand/shortage – either by cadre or specific health programme (e.g. ART). It can
be implemented incrementally starting with one or two groups with severe shortages, minimizing
risks of untargeted investments. It doesn‟t interfere with existing systems (except possibly
existing staff motivation, see below). It can be made competitive – so only quality staff are
selected. It is demand driven on the basis of the DHOs‟ requests. It can be funded in- and outside
the SWAp; any donor can provide funds. The contracts are performance based, so although the
cost is higher this should provide value for money. Having a significant number of staff on
performance contracts may have a positive influence on the wider organizational culture

    There will be risks, some of which may be mitigated as part of the scheme: (a) Such a
segment of the labor pool available and willing to work in the government health service, is very
small or non-existent. An initial survey will be needed for each group of staff being considered.
(b) Suitable recruitment agencies are not available. An initial survey will be needed. (c) MoH
unable to manage the recruitment agency effectively. TA could be provided to the MoH initially.
(d) Staff currently in government employment wish to leave to join the service. Applications
could be restricted to those who have left the Ministry more than 2 years earlier. (e) Staff
currently in government employment resentful of pay differential and may not cooperate with the
contract staff. A communication strategy, involving the unions, could be used to explain the need
for using contract staff. (f) Work and social environment (including housing) remains a “push”
factor. The POW including improvements to the work environment.??? (f) Government salaries
do not improve fast enough, so SSSS becomes a long-term scheme and is second-best solution

    Monitoring. Both process and output indicators would be needed for monitoring the
program.

   Process: Initial surveys conducted; Agency in place; Communication strategy implemented;
Governance structure in place; Review system in place (see output indicators).

    Output: Number of requests met; Number of contracts completed; Number of sub-schemes in
place/used; Scheme closed within 5 years of start up




                                                  90
 ANNEX E: GENDER DIMENSIONS OF HUMAN RESOURCES FOR HEALTH IN
                          MALAWI
               Table E-1: Distribution of MOHP Personnel by Grade and Sex, 1998

    Grade            Types of Job and Examples of Job Titles                   M    F     Total
P4 and above Senior Management, Specialists                                    14   2      16

P8 – P5 / S8 –   Mid-level Management: Principal Nutritionist, Sr. Nursing     56    26    82
S5               Officer
PO/CTO/          Medical Officer, Environmental Health Officer, Health        135    68   203
CEO              Education Officer, Nursing Officer, Chief Clinical Officer
STO/SEO          Sr. Nursing Sister, Sr. Clinical Officer, Sr. Lab Tech, Sr.   63   119   182
                 Pharmacy Tech., Sr. Dental Tech., Sr. Assistant
                 Accountant, Sr. Assistant Disease Control Officer
EO/TO            Nurse Technician, Clinical Officer, Laboratory Tech,         513 338     851
                 Dental Tech, Pharmacy Tech, Assistant Statistician,
                 Assistant Environmental Health Off
SCO/STA          Sr. Enrolled Nurse, Sr. Accts Assistant, Sr. Laboratory      241 783 1024
                 Assistant, Sr. Dental Assistant, TB Officer
CO/TA            Enrolled Nurse, Medical Assistant, Laboratory Assistant,     665 775 1440
                 Health Assistant., Laundry Assistant
D1-D8            Secretarial Staff                                             63   144   207
SC1-SC4          Health Surveillance Assistant, Support Staff                4,067 2,108 6175
Total                                                                        5,817 4,363 10,180
Percentage                                                                   57% 43%
                                 Source: HRD survey, MOHP (1998)

             Table E-2: Distribution of Public and Private Health Cadres by Sex, 2002
   Cadres                     Male                              Female                  Total
                       No.              %              No.                %
Doctor                 132              84              26                16             158
CO                     409              96              15                 4             424
EHO                    228              90              25                10             253
Matron                   0               0             108               100             108
Nurse/midwife           57               2            2,622               98            2,679
Midwife                  5               6              74                94              79
Nurse                   29              17             139                83             168
(Nurse                                   3
combined)               91                            2,943              97             3,034
Lab techs              134              89              16               11              150
Radiologists            39              95               2                5               41
HSA                   2,961             67            1,460              33             4,421
                                     Source: JICA study, 2002




                                                91
      ANNEX F: DRAFT TERMS OF REFERENCE FOR A STUDY ON THE
 INSTITUTIONAL ARRANGEMENTS FOR HUMAN RESOURCE MANAGEMENT
                    AND PLANNING IN MALAWI


         Introduction. The functions related to the planning, management and development of
health human resources (HHR) are carried out by a number of different government organizations
(Department of Human Resource Management and Development of OPC, the Ministry of
Finance and Economic Development, the Ministry of Education, Science and Technology, the
Ministry of Health and the new Health Services Commission). Within the Ministry of Health
itself, HR is covered by a number of departments: the Directorate of Planning, the Directorate of
Human Resource Management and Development, and the technical directorates of Clinical
Services, Nursing Services, Preventive Health Services, and Technical Support Services – all of
which perform functions related to recruitment, posting and promotion of their relevant cadres.
With the addition of the Health Services Commission and the more established Human Resources
Advisory Committee, the total number of groups working on HR in the MoH is quite large.

        The HHR related functions of the MoH need to be adjusted to conform with changes both
within the sector and outside (e.g. the Essential Health Package, hospital autonomy, devolution to
local government and the creation of the new central government Remuneration Board).

         In order to take full advantage of the effort put in by these groups and to ensure
efficiency in this effort, it is important to ensure that there are neither unnecessary overlaps nor
gaps in the functions being covered and those required in future. In addition, it is important to
ensure that the appropriate skills mix is available to cover these functions adequately.

        The MoH would like to use long-term technical assistance to support the development of
the HHR functions in the Ministry. However, it has not yet been decided where the TA should be
located and how it might best be used for capacity building.

      The MoH has therefore decided to conduct a review of the institutional arrangements for
the HHR-related functions within the Ministry and including the Health Services Commission.

        Objectives.

    1. Map out the distribution of all current HHR-related functions in all government agencies,
       and for individual departments within the MoH. Identify the information flows between
       and within agencies and identify current or potential barriers to effective implementation
       of the HHR-related functions.

    2. Identify which of these, due to changes both within and outside the health sector, are
       likely to change and move location.

    3. In agreement with senior MoH managers, identify the future HHR-related functions that
       will be managed by the Ministry.

    4. Map out the current HHR-related skills within the MoH.

    5. Based on 1 – 4, identify – if considered necessary – 2 options for restructuring the HHR
       functions within the Ministry, stating the costs and advantages and disadvantages of each.


                                                  92
    6. Identify staffing patterns for each option. Should skill gaps have been identified, identify
       how these should be closed – either through skills development initiatives or through a
       change in staff skills mix, assuming appropriately skilled staff are available in the job
       market.

    7. Propose costed implementation plans for each of the two options.

    8. Identify the most appropriate location and role for long-term TA in the development of
       the MoH‟s capacity to carry out its HHR-related functions.

        Outputs. (a) Presentation of findings and provision recommendations to a stakeholder
group at the end of the consultancy period. (b) Report of not more that 20 pages (excluding
annexes) stating the key findings of the study providing justified recommendations – which might
include restructuring – for strengthening the HHR functions in the MoH.

    Inputs. A team consisting of 1 HR consultant with experience of strategic work at sectoral
level in developing countries and 1 Institutional Development consultant with experience of
restructuring government organizations. The team will require 2 – 3 weeks for the field work plus
preparation and report writing time.

   Reporting Relationship. The consultant should report to the Director of Planning, but
because of the implications for reorganization should also closely liaise with the Principal
Secretary.




                                                93
  ANNEX G: DRAFT TERMS OF REFERENCE FOR LONG-TERM TECHNICAL
   ASSISTANCE IN HUMAN RESOURCE MANAGEMENT AND PLANNING IN
                           MALAWI

        Introduction. Since 1999 the area of health human resources has been recognized as one
of a major challenge by the Ministry of Health and its partners in the health sector. The current
climate both in relation to the GoM‟s recognition that significant improvements are needed in the
conditions of service of all public sector workers and the increasing amount of external assistance
being made available signal opportunities for tackling some of the major HR obstacles.

         While there are a number of sections within the Ministry of Health handling various
aspects of health human resource (HHR), the approach is not fully coordinated and overall level
of expertise in HR planning, management and development is weak. The main aim of the
technical assistance is to support the MoH to implement current short-term staffing strategies
effectively, and to provide assistance in the further development and implementation of longer-
term HHR strategies contained in the Programme of Work for 2004 – 2010. A key function of
the consultancy will be to build local capacity in effective HR planning, management and
development both within the MoH and where possible, amongst other partners within the sector
including the Health Services Commission.

        The TA will follow an institutional review of HHR-related functions within the MoH.
This review will further advise on the role of this TA and on the most location within the MoH.
Hence these TOR are only a first draft.

        Objectives. The objectives of the TA are likely to include some or all of the following
support to the MoH:

    (a)      implementation of current short-term HHR strategies included in the Programme of
          Work 2004 – 2010 and related annual plans

    (b)        development longer-term staffing projections and incorporate into a wider HR
          strategy for the sector.

    (c)       implementation of programs to improve pay and conditions of service

    (d)       improvement of HR management systems (regular planning and review; the HR
          information systems; effective and timely recruitment and posting; performance
          management; continuous professional development; etc)

    (e)      assistance in planning and management of transition of HR functions and staff
          employment to districts (devolution) and to autonomous hospitals

         Outputs. Outputs will depend on the nature of the prevailing requirements, but are likely
to include:

    1.    Advice to senior managers of the MoH and the HR Advisory Committee
    2.    Contributions to policy and strategy papers
    3.    Studies on specific topic areas as identify in the HR strategy
    4.    Assistance in the design and implementation of HR systems, as required.


                                                 94
        Inputs. Given that a more detailed TOR will be developed following the Institutional
Review, and the fact that globally there is a limited number of advisors available with suitable
experience for this TA, two options for the inputs are provided:

        Option 1: A Technical Assistance „package‟ which includes:
       A lead consultant/backstop to establish and oversee programme – long-term and broad
        experience in HR planning management and development at sector level and in the
        context of reforms;
       Long-term TA, with broad but less HR experience, located in the MoH;
       Short-term consultants (National/international) as required for specific areas of systems
        development (HR planning (including workload analysis), performance management,
        HRIS, employee relations, training, personnel administration) -- recruited by TA
        company, coordinated by long-term TA

       Option 2: A single consultant with long-term and broad experience in HR planning
    management and development at sector level and in the context of reforms

        Reporting Relationship. To be determined by the Institutional Review. However, there
should be a link to a sufficiently senior member of the MoH to facilitate good contacts with other
government and non-government partners dealing with HHR.




                                                95
    ANNEX G: PREPAID PLANS AS A PERCENTAGE OF PRIVATE HEALTH
   EXPENDITURES IN SUB -SAHARAN AFRICAN COUNTRIES, 1995 AND 2000



                      0                   10           20            30         40          50   60   70           80

         S. Africa                                                                                         71.7
                                                                                                                  76.6

        Zimbabwe                                              23.4
                                                                                        46.5

         Botswana                                                     29.3
                                                            21.6

      Cote d'Ivoire                                    18.6
                                                12.9

      Madagascar                           9.6
                                            10.3

           Senegal                        8.5
                                          8.7

            Kenya                   4.5
                                    4.5

         Tanzania         0
                                4.2

           Malawi             2
                              1.8

           Uganda         0.5
                          0.5

          Rwanda          0.4
                          0.3

                                                                             1995    2000

Source: World Health Report 2002




                                                                              96
 ANNEX H: GLOBAL ESTIMATES OF COST-EFFECTIVE AND AFFORDABLE
            PUBLIC HEALTH AND CLINICAL SERVICES
   Health Services       Percent of Global    Cost Per      Estimated         Included in
                          Disease Burden       DALY       Annual Cost Per       Malawi
                             Averted          (Global         Capita            EHP?
                                               Ave.)       (Global Ave.)
Integrated management            14.0          40.00           1.60                Yes
of childhood illness
Immunization (EPI                 6.0            14.50           0.50              Yes
Plus)
Prenatal and delivery             4.0            40.00           3.80              Yes
care
Family planning                   3.0            25.00           0.90              Yes
AIDS prevention                   2.0             4.00           1.70              Yes
program
Treatment of sexually             1.0             4.00           0.60              Yes
transmitted diseases
School health program             0.1            22.50           0.30               No
Tobacco and alcohol               0.1            42.50           0.30               No
program
         Sources: Claeson, Mawji, and Walker (2000); Malawi Program of Work (2004).




                                             97
  ANNEX I: ACTUAL AND EXPECTED ANNUAL DONOR COMMITMENTS TO
           THE HEALTH SECTOR IN MALAWI, FY94 TO FY05

    Donors        Actual Annual Ave.      Actual Annual Ave.             Expected
                    Commitments             Commitments             Annual Commitments
                       FY94-97                 FY98-00                (Donor Survey 2)
                       (Donor             (OECD Database)        FY03/04 FY04/05 FY05/06
                      Survey 1)
ADB &                    n.a.                     4.4                4.0        0.7        2.5
BADEA
CIDA                       2.8                    1.1                1.0        0.7        0.7
CIM                       n.a.                   n.a.                1.0        0.9        0.9
DfID                       4.2                   31.9               20.0       45.5       42.3
EU                        18.8                   10.9               10.0        1.3       n.a.
GTZ                        3.2                    5.8                2.2        4.8        3.4
IceIDA                    n.a.                   n.a.                0.6        0.6        0.6
IDA/World                 12.7                    1.7                8.3        3.8        3.8
Bank
JICA                       3.3                   n.a.                2.5        2.5        2.5
KfW                       n.a.                   n.a.                2.0        1.9        1.7
National AIDS             n.a.                   n.a.                4.8       28.2       28.2
Comission
Netherlands               n.a.                    3.7                1.7        0.3        0.3
NORAD &                   n.a.                    2.2                7.3        9.7        8.3
SIDA
OPEC Fund                 n.a.                   n.a.                4.0        4.0        1.4
UNAIDS                    Negl.                  n.a.               n.a.       n.a.       n.a.
UNDP                       0.2                   n.a.               n.a.       n.a.       n.a.
UNFPA                      2.8                   n.a.                2.1        1.1        1.1
UN Global                 n.a.                   n.a.                5.0        9.0       10.0
Fund for ATM
UNICEF                     4.1                   n.a.               n.a.       n.a.       n.a.
USAID                     10.3                   12.7               17.2       21.6       21.6
WFP                        3.2                   n.a.                1.5        1.5        1.5
WHO                        0.9                   n.a.                0.9        0.9        0.5
Total                     66.5                   75.7               96.1      139.0      131.3
    Sources: Donor Survey 1; Donor Survey 2; OECD Databank (hhtp://www1.oecd.org/dac).
Note: The total in Column 3 (OECD data) is higher than the details because of “other donors” not
                    shown, which had commitments totaling US$1.3 million.




                                               98
 ANNEX J: CHAM HEALTH FACILITIES BEING CONSIDERED UNDER EHP
          HEALTH SERVICE AGREEMENTS IN MALAWI, 2004
                Table J-1: CHAM Hospitals by Key Characteristics, 2004
    District         Hospitals           Catchment           Nearest            Distance
                                         Population        Government            (km)
                                           (2004)         Health Facility
Hospitals
Chikwawa        Montfort Hosp              66,364         Chikwawa DH               45
Dedza           Mua Hosp                   19,129         Dedza DH                  50
                                                          Salima DH                 65
Mzimba          Embangweni Hosp           44,214          Mzimba DH                 50
Phalombe        Holy Family Hosp            n.a.          Mulanje DH                60
                                                          Phalombe HC               10
Rumphi           Livingstonia DGM           n.a.          Rumphi DH                 60
Mzimba           St. John of God MH   Mental hospital for northern & central regions
Community Hospitals
Chitipa          Kaseye CH                 10,133         Chitipa DH               30
Dowa             Mtengowantenga CH         56,704         Lumbadzi HC              12
                                                          Lilongwe HC              41
Mchinji         Kapiri CH                  50,000         Kochirila RH             20
Mzimba          Mzambazi CH                16,130         Euthini HC               12
Likoma          St. Peter‟s CH              6,809         Nkhata Bay DH            50
                                   Source: CHAM




                                         99
                Table J-2: CHAM Health Centers by Key Characteristics, 2004

    District         Health Centers        Catchment           Nearest        Distance
                                           Population        Government        (km)
                                             (2004)         Health Facility
Health Centers with Maternity Service
Chitipa            Chambo HC                  6,365        Chitipa DH           40
                                                           Nthalire HC          30
Dedza              Mtendere HC                 46,272      Dedza DH             20
Likoma             St. Mary‟s HC                3,002      Nkhata Bay DH        20
Lilongwe           Dzenza HC                   24,192      Area 25 HC           15
Machinga           Kankao HC                   17,565      Balaka DH            20
Mangochi           Makanjira HC                22,145      Mongochi DH          85
Mangochi           Malembo HC                  17,194      Monkey Bay HC        40
Mangochi           Nankhawali HC                7,178      Monkey Bay HC        40
Mzimba             Kalikumbi HC                10,517      Mzimba DH            80
Nkhata Bay         Chilambwe HC                 7,000      Chintheche RH        15
Nkhotakhota        Chididi HC                  11,254      Nkhotakhota DH       30
Ntcheu             Mlanda HC                    7,050      Ntcheu DH            50
Ntcheu             Tsangano HC                 16,704      Ntcheu DH            70
Ntcheu             Chigodi HC                   9,205      Ntcheu DH            65
Ntchisi            Chinthembwe HC               8,797      Ntchisi DH           25
Phalombe           Mwanga HC                     n.a.      Phalombe HC          30
Phalombe           Sukasanje HC                 6,422      Phalombe HC          60
Thyolo             Chingazi HC                 18,150      Thyolo DH            30
Thyolo             Makapwa HC                   9,658      Thyolo DH            35
Thyolo             Chipho HC                   12,327      Thyolo DH            80
Thyolo             Thomas HC                   28,246      Thyolo DH            45
Health Centers Without Maternity Services
Dedza              Kanyama HC                  12,903      Dedza DH             30
Karonga            Sangilo HC                   7,142      Chitimba HC          20
Machinga           Nthlowa HC                    n.a.      Balaka DH            58
Nkhotakhota        Alinafe Rehab‟n Ctr           n.a.      Benga HC              8
Phalombe           Chiringa HC                 37,150      Phalombe HC          25
Rumphi             Mlowe HC                     8,414      Rumphi DH            75
Rumphi             Tcharo HC                    2,890      Rumphi DH            85
Zomba              H. Parker HC                 7,519      Domasi RH            12
                                       Source: CHAM




                                           100

								
To top