1
MID-TERM REVIEW OF BUNGOMA DISTRICT
MALARIA INITIATIVE (BDMI)
For USAID-Kenya and MOH, Kenya
Team Members
Joyce M. Olenja - Team Leader
Maria Francisco - USAID/Washington
Zablon Barake - MOH/National Malaria Control Programme
Simon Danda - DHMT, Bungoma District
June, 2000.
2
CONTENTS
LIST OF ACRONYMS
AIDS Acquired Immuno-deficiency Syndrome
AIMI The African Integrated Malaria Initiative
AMREF African Medical Research Foundation
APHIA AIDS Population and Health Integrated Assistance Project
ARI Acute Respiratory Infection
BDMI Bungoma District Malaria Initiative
BI Bamako Initiative
CAs Collaborating Agencies
CDC Centres for Disease Control and Prevention
CDD Control of Diarrhoeal Diseases
CIF Community Improvement Fund
CO Clinical Officer
CORPS Community Owned Resource Persons
CQ Chloroquine
DCO District Clinical Officer
DFID Division for International Development
DHEO District Health Education Officer
DHMB District Health Management Board
DHMT District Health Management Team
DLT District Laboratory Technologist
DMOH District Medical Officer of Health
DMRO District Medical Records Officer
DN District Nutritionist
DPHC Division of Primary Health Care
DPHN District Public Health Nurse
DPHO District Public Health Officer
EOT End of Training Assessments
EPI Expanded Programme on Immunization
3
FIF Facility Improvement Fund
HCDC Health Centre Development Committee
HIV Human Imunno-deficiency Virus
HMIS Health Management Information System
ICM Integrated Case Management
IEC Information, Education and Communication
IMCI Integrated Management of Childhood Illness
ITMs Insecticide Treated Materials
ITNs Insecticide Treated Nets
JHU John's Hopkins University
KDHS Kenya Demographic Health Survey
KEMRI Kenya Medical Research Institute
K-FPHCP Kenya Finland Primary Health Care Programme
OCG Organized Community Groups
PSI Population Services International
QAP Quality Assurance Project
SP Sulfalene/salfacholine Pyramethamine
USAID United States Agency for International Development
UNICEF United Nations Children's Fund
WHO World Health Organization
4
1. Executive Summary
2. Background
A. The Bungoma District Malaria Initiative
B. Project Objectives
C. Expected accomplishments over the Five-year project period
D. BDMI Mid-term evaluation
E. Methodology
3. Findings
3.1 Project management
A. The role of AMREF and the DHMT
B. The role of USAID/Kenya
C. Funding Mechanisms
3.2 Project Implementation
A. Objective 1 – Case management at facility level
B. Objective 2 – Improved household management of fever and
anaemia
C. Objective 3 – Improved Prevention and management of
malaria in pregnancy
D. Objective 4 – Increased household use of insecticide-treated
materials
E. Objective 5 – Effective collection and use of information for
planning, monitoring and evaluation
5
F. Operations Research: Appropriateness and Timeliness
· The Centres for Disease Control and Prevention (CDC)
· African Medical and Research Foundation (AMREF)
· Basic Support for Institutionalizing Child Survival
(BASICS)
· The Quality Assurance Project (QAP)
· District Health Management Team (DHMT)
G. Information, Education and Communication (IEC) Strategy
4. Key Issues
A. Capacity building and sustainability
B. Perspectives on IMCI implementation at the health facilities
C. Issues for expansion of the Bungoma experience
D. Linkages to national health policy and programmes
5. Conclusion
6. Recommendations
Appendix I Terms of Reference
Appendix II Workshop participants
Appendix III List of persons interviewed
6
1. EXECUTIVE SUMMARY
Introduction
The Bungoma District Malaria Initiative (BDMI) is a five-year project (1998-2002)
funded by USAID and the Kenya Government. Project objectives focus on malaria
prevention and the management and treatment of febrile illness and anaemia in children
and pregnant women. This was to be achieved through a partnership of the Ministry of
Health (MOH), the Bungoma district community and collaborating agencies (CAs) that
would provide technical assistance through training and operations research.
The expected outputs of the BDMI were:
· Health workers trained in correct diagnosis and appropriate treatment of febrile
illness, anaemia and maternal malaria infection.
· Capacity for early recognition of fever and anaemia at the household level is
improved.
· The distribution and use of Insecticide-treated Materials (ITMs) at the community
level is increased.
· The delivery of malaria drugs through antenatal clinics for prophylaxis and
presumptive treatment during pregnancy and the use of ITMs by pregnant mothers
promoted.
In combination, these outputs were expected to contribute to a reduction in numbers of
deaths and cases of malaria through improved access to and use of services at both the
facility and community levels. This would be enhanced by a health system that has health
providers with improved capability in diagnosis, treatment, counselling and interpersonal
skills. At the national and district levels the project is expected to disseminate
information on how to design and implement facility and community based approaches
for the delivery of effective and sustainable control of malaria.
7
The purpose of the mid-term review is to assess the progress towards the achievement of
project objectives; the current and potential contribution to child survival in Kenya;
challenges and lessons learned with a view to refining the approach and strengthening
future project implementation.
Main Findings
The project has been in operation for two and half years. By design, the operations
research dominated the first two years of the project activities, delaying the
implementation of interventions. However, the training component was initiated in the
first year of the project and progress has been made towards improving the management
of fever and anaemia principally among children, at the facility level.
Project management has been undertaken by various actors. It is understood that AMREF
administers funds for the implementation of project activities while the DHMT is the
project implementing agency. The CAs make recommendations based on their studies but
the decision on what to implement is jointly made by the DHMT, USAID and CAs
during the annual workplan meetings. Given that not all recommendations from all
studies can be implemented through USAID funding, the DHMT is expected to make a
decision on priority areas with technical advice from AMREF. AMREF may not have
performed this role satisfactorily by asserting itself as the co-ordinator of the project. This
may partly explain why implementation of activities was, at times, ad hoc. This is
exacerbated by a limited oversight role of USAID/ Nairobi. The Mission through its
senior staff has not been proactive in providing programmatic support to the project.
One of the key areas of concern is the capacity building of DHMT members to be able to
carry out activities in the future. This will depend on the extent to which they have been
prepared for the task. Training so far has equipped them with knowledge and skills on
management of illness, and they are capable of imparting the same to others. However,
there has been an omission in equipping the DHMT members with the necessary skills
for data analysis report writing and appropriate use of information gathered.
8
Generally activities undertaken by the District Health Management Team (DHMT) are in
line with the stated objectives, although some are weighted differently in terms of
emphasis. Facility-based interventions have been given priority over community based
ones with training and supervision having a major thrust so far.
In general, dissemination of project activities, including findings from research conducted
has been inadequate and slow. Even within the project, communication between the CAs
was reported to be limited. Nevertheless, the findings so far have great potential for
policy and programmatic change.
Management skills were often cited as lacking among the DHMT and the in-charges of
peripheral health facilities. The need for training in management was therefore presented
as one of the priority needs. This is in readiness for the decentralisation process that is
underway and the financial and human resource responsibilities that are expected to come
with it.
It was evident that Bungoma district under the BDMI experiences constraints that are
largely logistical and relate to resource availability, both financial and human. For
instance even though supervision is scheduled to be carried out weekly by the DHMT,
some IMCI trained providers had been supervised at most once or twice a year since they
received the training. A major obstacle is the lack of transport facilities noting that there
is only one reliable vehicle at the district headquarters that is assigned supervisory
activities.
Sustainability of the programme was also a source of concern especially with regard to
supervision. Transport at the district level is problematic. Presently the project is
providing for transport and lunch allowances. It is anticipated that through cost-sharing,
funds would be set aside for supervision. Already the DHMT have decided to integrate
the IMCI supervision with the general supervision to maximize on the use of available
resources.
9
At the community level, a positive response has also been harnessed through the Health
Centre Development Committees (HCDCs). The funds generated under Facility
Improvement Fund (FIF) and Community Improvement Fund (CIF) are partly used to
purchase basic supplies particularly essential drugs and equipment for Oral Rehydration
Therapy (ORT) corners. In one Community, Kaptenai, the health provider indicated that
community members were beginning to notice fewer deaths in the community among
children and they attributed this to the new approach at the health facility; reflecting the
practice of IMCI.
In NGO facilities, ANC mothers are not given presumptive treatment with SP; if at all
they are given, this is selective for those mothers who can afford to pay for it. This also
applies to folic acid and ferrous sulphate tablets. In all the government health facilities
visited, presumptive treatment for pregnant mothers attending ANC was provided though
not always consistently and correctly. In addition, when the supply of SP is irregular at
the health facility level, there is a tendency to use it for general treatment rather than
prophylaxis for ANC mothers. Moreover, in facilities with Facility Improvement Funds
(FIFs) that now charge for services, the provision of SP for ANC mothers is problematic
since the current policy states that maternal and child health services are to be provided
free of charge.
Patient referral is a major problem affecting project performance. This is not limited to
Bungoma district alone but is a nationwide problem, affecting access to appropriate and
quality care. Often patients referred failed to comply for a variety of reasons that included
lack of transport and the fear that they may not get adequate attention at the next level.
At the stakeholders workshop it was reported that laboratory equipment in the district
hospital and the health centres was obsolete. A presentation at the workshop based on
assessment of microscopic diagnosis of malaria in the health facilities raised issues of
concern with regard to the accuracy of the results noting that the negatives were too high
with a predictive value of 93.
10
Community members recognise the value of nets as prevention against malaria. However
the issue of availability and particularly affordability may prohibit the actual use of ITNs.
The creation of demand for bednets through identified community outlets and willingness
to pay for them as a means of protection against malaria will be a major achievement
towards the project goals.
The issue of poverty resonated in all the discussion groups. A visit to the health facilities
and a discussion with the ITN distributors within the community revealed that there is
willingness among community members to use bednets as protective mechanism against
malaria. However, the issue of cost is real and it is not quite clear how behaviour change
can be achieved at community level in the absence of uplifted standard of living among
the population to acceptable levels to enable them participate fully in the development
processes and health provision in particular.
Overall, the IEC strategy as developed by BASICS was very clear on the activities to be
implemented and the mechanisms for doing this. Initially the implementation of the IEC
component was by different CAs who developed and disseminated activities without
coordinating and harmonizing messages, resulting in overlap and overload to the
community. Currently AMREF is responsible for IEC and it is hoped that coordination
and harmonization of IEC messages will be one of the tasks AMREF undertakes to avoid
overlap and duplication of information while maximizing on available resources.
In terms of a strategic approach, the BDMI blends very well with the IMCI. The latter has
been embraced by health providers who value the systematic management applied to a
sick child. This approach is also appreciated by caretakers as being comprehensive. It is
evident that the two components of IMCI - improving case management skills for health
workers and improving the health system to deliver IMCI - are being addressed within
this project.
11
However, to realize the full potential of BDMI in addressing child survival, an aspect of
the strategy aimed at the community will be crucial in the remaining part of the project.
This was emphasized at all levels including CAs, policy makers, service providers and
other stakeholders working on child survival.
Within the project there are data that have been collected on behavioural and preventive
aspects of health. The Centres for Diseases Prevention and Control (CDC) research on
malaria in pregnancy indicates that in terms of home care and care-seeking practices,
caretakers are principle providers of anti-malarial treatment to children with fever. This
provides a strong evidence base for undertaking community–based approaches, which
would complement the activities currently being implemented at the facility level. By
focussing on caretakers as a means to achieve the project goals and objectives one would
simultaneously address improving compliance to appropriate treatment which remains a
major challenge to the success of IMCI and consequently child survival.
Conclusion
BDMI as an integrated child survival project is doing more than IMCI. Although the
programmatic focus of BDMI has been on malaria, the introduction of IMCI as a
mechanism for accelerating its objectives reintroduces a broader child health perspective
to the project. It provides a logical platform for cross- sectional collaboration and
integration of child health strategies in immunisation, nutrition, HIV/AIDS, among others
to occur. Given the point that it includes a substantial component of maternal health, it is
simultaneously addressing the issue of safe motherhood and child survival, which is a key
component in the National Reproductive Health strategy.
Project activities are on course and research findings are available even though these
have not been systematically disseminated. This arises partly from the fact that AMREF
as the co-ordinator of the project has not asserted itself enough to provide technical
leadership and a framework within which the CAs were to operate.
12
This is exacerbated by a limited oversight role of USAID/ Nairobi. The Mission through
its senior staff has not been proactive in providing programmatic support to the project.
Furthermore communication and co-ordination in general were said to be problematic
This is particularly in reference to internal sharing of reports among the CAs. Whereas it
had been agreed that quarterly reports would be circulated to all CAs to date this has not
happened. Similarly at the November 1999 planning meeting it was agreed that an
electronic database of BDMI documents would be created to which all CAs would
contribute. However only CDC is reported to have undertaken this.
 The DHMT raised their lack of management skills. Equipping the DHMT with these
skills and their participation would allow for a sense of ownership as well as
demonstrated partnership and capacity building as stipulated in the project goal and
objectives. In addition it was felt that the roles and responsibilities of the project deputy
manager were not clear and formalised. Both concerns are critical for capacity building.
Bungoma district has just embarked on a range of activities that simultaneously touch on
all the five project objectives. A community-based component is already evolving, and
drawing from experiences of other donors and collaborating agencies, it is likely to
contribute to achieving the project outputs and consequently its goals. This is likely to
result in a BDMI working model that could be replicated nationally with appropriate
modifications through the planned national adaptation of IMCI.
There is potential for community participation particularly in drug supply although the
issue of gazettement will need to be clarified by the MOH.
Patient referral remains problematic and is a major challenge to practice and success of
IMCI. To address the problems presented by patient referral will require harnessing both
facility-based and community-based resources.
13
The study intervention namely, problem-solving teams may not have been adequately
understood or defined prior to the start of the study as evidenced by the difficult start-up.
An inexperienced trainer was used to introduce the approach and the subsequent need
was to re-train the teams as well as provide substantial 'coaching' or external support to
apply the problem-solving steps. It is acknowledged that 10 senior health workers, 5 of
whom are DHMT members were trained in Quality Assurance concepts and
methodology. Yet from the discussions with the DHMT, the activity seemed to lack
adequate ownership at this level and the expected improvements in IMCI as a result of
this approach were not realised.
To monitor trends in malaria morbidity and service utilisation rates, a basic tool is record
keeping at the health facilities. From the workshop presentation on Health Information
System (HIS), follow-up discussions with relevant officers, health providers and
observations at health facilities, it is clear that this component of BDMI is the weakest.
At the time of the review there was an exercise to pilot data collection forms in selected
health facilities and at the community level. However, at the moment these are project
specific and there is a danger that they may not be applicable nationally given that the
MOH headquarters has not been involved in their development for standardisation.
On evaluation and observation, health providers still have difficulty in recognising severe
illness among children. In one facility a pregnant woman was given SPs in the ninth
month. This is a pointer to the need to revisit the training strategy and supervisory
mechanisms that would facilitate retention of skills through practice and assessments.
The results from the study on assessment of microscopic diagnosis of malaria raised the
issue of quality control. This calls for training of laboratory personnel given the
importance of malaria microscopy at the higher referral levels.
Training and supervision are the most expensive aspects of the IMCI. For cost￾effectiveness of the IMCI strategy pre-service training is an option. This is already being
implemented at the Universities of Addis Ababa, Ethiopia and Dares-Salaam, Tanzania.
14
The review team learned that some of the components are already being taught to
students at the Medical School, University of Nairobi, Kenya. There are plans to extend
to medical training colleges and nursing schools. Medical officers posted to the field and
who are already acquainted with IMCI would be informed supervisors who would steer
child survival activities within the district.
At the beginning of the initiative, there may have been some misgivings at the national
level, principally because the project approaches and activities may not have been very
clear to all parties, but with time this has changed. Recently different MOH personnel
have participated at various levels as part of the project implementation including
training, provision of malaria guidelines for training and participation in the project
quarterly meetings. From the discussions with MOH senior staff, it was evident that
BDMI is expected to provide insights into the implementation of IMCI nationally.
Recommendations
Consolidation of the IMCI activities in Bungoma rather than expansion of the same to
other parts of Kenya should be a priority for the project. Expansion could still be
achieved by providing a working BDMI model, which could be replicated nationally,
with appropriate modifications, through the planned national adaptation of IMCI. To this
end, USAID can contribute at the policy level by participating more actively in the
review of the adaptation process and facilitating the dissemination of findings of the
project so far.
USAID senior health staff could play a more active role in representing the achievements
of BDMI at the national level by participating in relevant task forces/technical
committees. This would serve to bridge this pilot USAID-funded activity with national
malaria control, IMCI, RBM and other related strategies currently under development as
well as help leverage the resources of other development partners.
15
In project management USAID/Kenya will need to strengthen its oversight role such that
if the implementation and monitoring needs of the Mission are not being met by the
current reporting process the issues should be raised promptly with AMREF
management. In the same vein, there is need for greater AMREF/Nairobi senior staff
involvement in the project to provide the needed technical backup to the project manager
in Bungoma. This would ensure that priority areas agreed upon at the planning meetings
are addressed in a timely manner and that the CAs adhere to these requirements.
Additionally, AMREF as the co-ordinator should be more proactive in facilitating
communication between the CAs. This is especially so with regard to distribution of
reports among the CAs to promote awareness of each other's activities and how they fit
together in the overall objectives and goal of the project.
USAID and implementing partners should review the current monitoring and evaluation
strategy. This should distinguish between the need for programmatic (activity-level) data
that would allow USAID and project managers to make interim adjustments/corrections
in specific project areas, and the need for more rigorous measures of district-level
performance and impact.
The working relationships of the CAs and accountability to the DHMT were reported to
have greatly improved. However, USAID should consider formalizing the position of the
DHMT deputy manager in the contract agreement in order to facilitate the administration
of the project, particularly in the absence of the AMREF coordinator.
To facilitate acquisition of management skills for DHMT members and health facility in￾charges, the project will need to identify and organise for short-term management courses
that are available locally.
16
In the second half of the project it will be timely to forge a true linkage between
organisations that work in child survival activities, the Division of Primary Health Care
(DPHC) as well as the BDMI. Presently Siaya district under Care-Kenya offers a
Community-based model for IMCI and may provide insights for replication in Bungoma
district.
There is good information already collected on behavioral and preventive aspects of
health which provides a strong evidence base for undertaking community–based
approaches. Should such an approach be undertaken, they should build on the existing
community structures and activities of CBOs and NGOs working to address the key
IMCI household and community behaviors, and forge multi-sectoral linkages across
Ministry departments and among organizations working in child survival.
The need for alternatives to the current modes of supervision should be explored given
that transport and allowances are currently provided by BDMI. The expectation is that
the DHMT will somehow absorb these costs once the project ends, but to sustain the
intensity and frequency of current visits will require some considerable planning and
mobilisation of additional resources, including at community level.
Based on the premise that routine points of intervention work better; building on them
rather than creating new structures is a cost-effective approach to improving maternal and
child health. In this respect, AMREF should harmonise the IEC strategy with
opportunities for intervention principally at the ANC level, while strengthening the
existing points of outlets for ITNs. The original BASICS framework could also be re￾visited to provide guidance in this exercise.
Given that the government fully supports a social marketing strategy, it would be
worthwhile from the onset that this strategy and a community-based distribution strategy
be combined in order to increase access to ITNs. In this respect, Population Services
International (PSI) has had experience with social marketing of bednets in the coastal
17
region and it would be worthwhile to explore links between AMREF and PSI for social
marketing of nets in Western Kenya. This initiative should be facilitated by USAID.
Community-owned resource persons (CORPS) and drug vendors handle drugs such as
SPs. These drugs are gazetted as part II poison. BDMI should seek clarification from the
National Malaria Control Programme and the Pharmaceutical Board to avert the fear by
CORPS of being prosecuted for handling prohibited drugs.
SPs are not always prescribed for pregnant mothers as providers are not sure whether to
charge or not; given that as a policy ANC services are free. It is therefore important for
MOH to ensure that adequate supply of SPs is included in the drug kit for use in
MCH/FP.
Community pharmacy is a cost-effective strategy that fills the gap created by drug
shortage at the health facility level while making drugs more available at the community
level. They are complementary to drug vendors and shopkeepers, ensuring access to
treatment at the community level. This is made possible by the Community Improvement
Fund (CIF). However, the MOH headquarters needs to formalize and regulate CIF within
a legal framework to counter exploitation and abuse. At the moment the MOH at the
district level has no control over the finances and the expenditures of these pharmacies.
Subsequently the team has learnt from DHMT that the Division of health care financing
is in the process of preparing regulatory mechanisms to address the issue but will not
control the funds raised under CIF.
The issue of patient referral requires review at the national level in terms of appropriately
equipping the health facility and especially ensuring that there is a functional ambulance
at the district level. More importantly, community participation should be harnessed to
realize their role in facilitating referral using available local means and resources.
Malaria microscopy is more critical at higher referral levels since most dispensaries do
not have laboratory facilities. But adequate equipment, supervision and quality control as
18
well as refresher training for the laboratory staff should be provided in those facilities
where microscopy is available.
For improved management of IMCI there is need for repeated exposure to a high
concentration of severe cases. This would then facilitate better practice and allow for
frequent assessment and evaluation as required. This may only be possible in large
hospital settings on a rotational basis, six months after training. This arrangement should
be explored for its feasibility.
At the moment only a select group of professional health providers and some DHMT
members have received IMCI training. All Medical Officers and all DHMT members in
Bungoma district should undergo IMCI training to facilitate their role as supervisors and
to ensure appropriate management of children who have been referred from other health
facilities and the community level. IMCI training needs in private NGO facilities should
also be examined as well as pre-service alternatives. Training of CORPs should be
supported by a strategy to ensure that adequate monitoring and supervision are available.
At this point in time basic data will be necessary against which indicators will be tracked
partly to monitor project activities and subsequently measure impact at the end of the
project. At the time of the review, data collection instruments were being piloted in
selected health facilities and at the community level. It will be crucial to liase with the
MOH headquarters at the onset to standardize the instruments for national application.
Again, the team has learnt that this process has been initiated by AMREF.
More importantly, given the fact that there was a prior system in place that has since
collapsed it may be cost-effective to rehabilitate this system through provision of supplies
particularly tally sheets and retraining of data collection staff to re-orient them on the
importance of collecting data. This is crucial for the project in that as it moves into
implementation, indicators for monitoring would have to be tracked on baseline data.
19
The approach of QAP teams was acknowledged to have value as a management concept,
although the discussions around its practical usage remained diffuse. In light of these
findings, the DHMT, QAP and other CAs should re-evaluate the contribution of this
team-based approach towards the achievement of BDMI objectives, and critically assess
the 'value added' of this activity given the remaining time and resource constraints of the
project and modify implementation as needed.
Given the short time remaining in the project, more consideration should be given to
maximising the use of prior research i.e. implementing the recommendations, before
embarking on new ones. Any future research should perhaps focus on assessing the
effectiveness of the interventions currently underway in the true spirit of 'operations
research', although exceptions, such as formative work to explore the problem of
misclassification of severe illness, may be needed.
Dissemination of project activities, including findings from research conducted is yet to
occur. The findings so far have great potential for policy and programmatic change. In
principle the IMCI approach in Bungoma is accepted within the national health policy
and is in line with the national adaptation process. Similarly, linkages to the National
Malaria Control Programme are quite strong. It is expected that the Bungoma experience
would help inform the national agenda for expansion of IMCI and new initiatives such as
RBM. USAID/Kenya could again, effectively broker this. This would be to a broader
audience consisting of researchers in maternal and child health, policy makers, senior
IMCI programme staff, other related health providers especially at district levels and the
donor community working in maternal health and child survival.
20
3. BACKGROUND
Introduction
The gains made in child health previously in Kenya have been lost as reflected in the
latest Kenya Demographic and Health Survey (KDHS) of 1998. The Survey revealed that
the under fives mortality has risen from about 80 deaths per 1,000 live births to currently
112 deaths per 1,000 live births during the1990s. Maternal mortality ratio of 590 per
100,000 live births was also reported. This is in spite of the existence of a health system
that has attempted to address areas of morbidity and mortality in the two decades
preceding 2000. The establishment of the maternal and child health (MCH) programme
was a step toward better health for mothers and their children, but along the way some
obstacles have arisen, influencing negatively the expected outcomes. A key issue is the
resistance to the traditional chloroquine treatment for malaria, a leading cause of
morbidity and mortality.
Malaria remains the leading cause of morbidity and mortality, and therefore a major
public health problem in Africa. Between 300 and 500 million episodes of malaria and
1.5 to 2.5 million deaths occur annually in Africa. At least 14,000 children in Kenya
require admission each year due to complicated malaria. Approximately 26,000 children
die each year from direct consequences of malaria infection. In Bungoma district malaria
prevalence rate across all ages is up to 75%, and accounts for 39% of outpatient
morbidity and 42% of inpatient morbidity. Inpatient mortality of 36% is directly caused
by malaria. Specifically 20% to 25% of childhood deaths are due to malaria. The
resistance to chloroquine has complicated treatment of malaria, which has been easily
available and affordable. Bungoma district is located in the Lake Victoria Basin, which
has a high resistance rate of the malaria parasite to chloroquine to the tune of 85%
(Bungoma District Health Annual Report, 1995).
21
The Kenya Health Policy Framework Paper (1994) proposes strengthening of activities
directed at the reduction of morbidity and mortality due to malaria. In order to achieve
this, one of the priorities of the Ministry of Health is appropriate retraining of health
workers at all levels on current treatment of malaria. In response to this, the Ministry of
Health launched new national guidelines for diagnosis, treatment and prevention of
malaria in 1998. According to the new guidelines sulfadoxine/ sulfalene pyrimethamine
(SP) is the first line drug for treatment of simple malaria. In 1999 SP was gazetted as Part
II Poison drug and is available over the counter while chloroquine was rescheduled to a
prescription drug under Part I Poison drug.
The new guidelines also recommended presumptive treatment of malaria during
pregnancy by administration of a dose of SP at the beginning of the second trimester, and
a second dose at the beginning of the third trimester. The new guidelines are in use in
Bungoma district under the Bungoma District Malaria Initiative (BDMI).
A. The Bungoma District Malaria Initiative (BDMI) 1
The Bungoma District Malaria Initiative (BDMI) is a five-year project (1998-2002)
funded by USAID and the Kenya Government at an estimated cost of US$ 5 million.
BDMI is part of USAID’s regional “The Africa Integrated Malaria Initiative” (AIMI)
which is being implemented in three other African countries: Zambia, Malawi and Benin.
The aim of the initiative is to explore programmatic options for reducing morbidity and
mortality among children under the age of five years and among pregnant women, and to
strengthen local capacity to deliver effective and sustainable integrated malaria control at
the health facility level.
In Kenya, Bungoma District was selected as part of USAID/Kenya's District Focus
component of its Child Survival APHIA Project.2
 The following criteria were used:
 
1
 The project has also been known as the Bungoma District Initiative, or BDI project, but
was renamed the Bungoma District Malaria Initiative, or BDMI, by the DHMT
22
q high malaria burden among vulnerable populations
q variety of service delivery environments as indicated by both urban and rural
service needs
q opportunities to collaborate and build upon prior USAID Integrated case
management (ICM) and FINNIDA investments in infrastructure and capacity
building.
q feasibility of improving management and financial operations as indicated by
adequate potential for cost sharing.
q availability of resources as indicated by adequate ratios of beds and staff per
capita and the mix of facility levels represented.
q adequately diverse service provider pool, indicated by presence of NGOs and
private providers.
q approval by the Ministry of Health
The activities of the BDMI project were begun in 1996, although the project was not
formally launched until March 1998. During this time, the Kenya-Finland Primary Health
Care Programme which had for an extended period supported the rehabilitation of health
sector infrastructure, focusing on upgrading health worker skills as well as strengthening
community-based MCH and water and sanitation services, withdrew its support.
The main project partners are the MOH, the Bungoma District community, and
collaborating agencies (CAs): the Centres for Disease Control and Prevention (CDC), the
Quality Assurance Project (QAP), AMREF, and formerly BASICS (Basic Support for
Institutionalizing Child Survival). The CDC and QAP are USA-based groups providing
technical assistance to BDMI through training and operations research. AMREF, a
 
2
 APHIA Project's District Focus aimed to demonstrate impact and programme synergies
achievable through integrated service delivery in the public and NGO sectors; by
developing and testing innovative and sustainable approaches for maternal and child
health, including management and prevention of malaria-related morbidity and mortality;
and by strengthening managerial and financial operations at district level.
23
Kenya-based NGO, provides both technical inputs as well as project coordination, while
the District Health Management Team (DHMT) is tasked with overall field
implementation as well as technical leadership and management oversight.
B. Project Objectives
The project is expected to contribute to the broader goal of reduced malaria-related
mortality and morbidity in Bungoma District through the achievement of the following
objectives.
· Improved management of fever and anaemia, principally among children under five
years of age by health workers at the health facility.
· Improved capability of mothers and other caretakers to manage fever and anaemia at
household level.
· Improved prevention and management of malaria in pregnancy.
· Increased household use of insecticide-treated materials.
· Effective collection and use of information for planning, monitoring and evaluation.
The activities intended to address these objectives are: -
· Training health workers in correct diagnosis and appropriate treatment of febrile
illness, anaemia and maternal malaria infection.
· Educating mothers and other caretakers on early symptoms of malaria and
appropriate health seeking behaviour.
· Promoting the distribution of ITMs through micro-enterprise and education of
households on their appropriate use.
· Promotion of the delivery of malaria drugs through antenatal clinics either as routine
prophylaxis or presumptive treatment during pregnancy and the use of ITMs by
pregnant mothers.
24
C. Expected Accomplishments over the Five-year Project Period
People-Level Impact
· Reduction in the numbers of deaths and cases of severe illness attributable to malaria
infection through improved access to and use of service for diagnosis and treatment
of malaria infection.
· Reduction in malaria infection through improved access to and use of insecticide￾treated materials.
Community/household-Level Impact
· Increased capacity at the household level for early recognition of the signs of febrile
illness by mothers and other caretakers, leading to prompt and appropriate health￾seeking behaviour for malaria management.
· Increased recognition of the high risk of maternal infection and demand for
appropriate chemoprophylaxis treatment.
· Increased access to appropriate and affordable antimalarials.
· Increased demand for and access to insecticide-treated materials and insecticides.
Health Facility–Level Impact
· Improved capability of clinicians to diagnose and treat febrile illness, anaemia and
maternal malaria infection, including referral and management of complicated illness,
counselling caretakers, and interpersonal communication.
· More cost effective use of human and material resources through improved planning,
management, training and supervision.
25
National and Regional-Level Impact
· Dissemination of information on how to design and implement facility and
community-based approaches for the delivery of effective and sustainable control of
malaria.
E. BDMI MID-TERM EVALUATION
i. Evaluation Objectives
This evaluation marks the completion of two and a half years of the BDMI project. The
purpose of the evaluation is to review mid-term progress towards the achievement of
project objectives, assess current and potential contribution of the project to child
survival in Kenya, and document challenges and lessons learned with a view to refining
the approach and strengthening future project implementation. (See also attached Scope
of Work - Appendix I)
ii. Methodology
The evaluation team used a variety of approaches to address the terms of reference. At a
two-day workshop, the DHMT and Collaborating Agencies (CAs) presented status
reports on their project activities in relation to the BDMI objectives. The roles and inputs
of these agencies at the different stages of the initiative were discussed, focusing on their
successes and constraints. Due to scheduling difficulties and other reasons, the workshop
was convened very early in the process as the evaluation team had only been in place for
a few days prior to the start of the meetings. As a result, a number of issues identified
during the site visits in the subsequent week could not be adequately vetted with the
implementing partners, although the team attempted to validate findings through
available proxy measures.
26
The evaluation team conducted interviews with community representatives, caretakers,
pregnant women and service providers at health facilities. Issues discussed at community
level included perspectives on malaria and its current management at the health facility,
any changes or improvements they would like to see, as well as their perceived role in the
management of malaria. Key issues related to malaria treatment, its prevention and the
sustainability of project activities were discussed.
In addition to site visits, background documents and annual project reports were also
reviewed, and information was obtained from other key stakeholders. The latter included
interviews with the central MOH, the implementing partners: African Research and
Medical Foundation (AMREF), Centres for Diseases Prevention and Control (CDC), and
the Quality Assurance Project (QAP), USAID/Kenya and USAID/Washington, and
members of the DHMT. The views of technical staff from UNICEF, WHO and DFID
were also sought.
3.0 FINDINGS
3.1 PROJECT MANAGEMENT
A. The Role of AMREF and the DHMT
The project management structure is comprised of a project manager (appointed by
AMREF), a deputy project manager (a member of the DHMT) and a project accountant
(appointed by AMREF). Though the roles and responsibilities of AMREF in partnership
with the DHMT were said to be delineated and understood, the terms of reference of the
DHMT deputy project manager are not formalised contractually.
USAID wished to contract a Kenya-based NGO to manage and coordinate all of USAID's
technical and program support to the initiative. Though initially AMREF signed the
contract to coordinate the BDMI project, AMREF assumed some technical
responsibilities (i.e. implementation of the ITNs component of the project) in order to fill
27
the gaps not taken up by CAs. However, the co-ordination role of the project has had
some shortcoming. The project manager has not received adequate support from the
senior staff of AMREF. In this respect, through the project manager, AMREF has not
sufficiently asserted itself in co-ordinating the activities of the other CAs and advising
USAID accordingly.
The BDMI embraces a local partnership comprising various actors; the DHMT and the
Collaborating Agencies (CAs). The roles and responsibilities are stipulated in the project
document with the DHMT as the implementor and the CAs providing technical support.
There are several modalities of coordination including annual meetings of all the CAs
during which the next set of activities are planned and discussed together as a team.
Initially there appeared to be some confusion in the management and co-ordination of the
various project activities, often leading to ad hoc implementation of activities by the CAs;
who apart from AMREF were not based within the district with less accountability to the
DHMT. The project design allowed for many actors but lacked a strong central
mechanism of control. This administrative arrangement caused planning and
implementation problems for the BDMI project. For example, by design the CAs’
research was expected to inform the selection of intervention strategies. However,
feedback of research findings to the district often took a long time and, as a result, the
project did not move from research to implementation as planned.
The lack of adequate coordination of activities caused undue stress on the DHMT who
had other responsibilities over and above the BDMI. This has since been rectified
following the development of implementation ground rules by the DHMT. Within this
arrangement, the DHMT defines the planning process under which the CAs must operate.
This new arrangement was reportedly working well for all the parties.
The need for a stronger DHMT role in the CAs’ activities was further addressed by
AMREF. A DHMT counterpart was attached to every CA activity and this also ensured
equitable distribution of tasks among DHMT members. The relationship between the
28
various parties had also been strained in other areas. For example, in terms of payment of
allowances the DHMT expected more than the rate as paid by AMREF. This was
discussed and a rate agreeable to all the parties was established.
i. Discussion & Recommendations on the Role of AMREF & DHMT
AMREF as the technical body designated to provide guidance to the project and advice
USAID accordingly has not sufficiently exerted itself. Thus the co-ordination and
advisory roles have not been performed to satisfactorily. Despite some 'hiccups' early on
in the management of the project, the working relationships of the various implementing
partners and accountability to the DHMT were reported to have greatly improved. In
particular, technical leadership and ownership of the project by the DHMT has been re￾enforced through its efforts to develop operational ground rules for the CAs and through
the designation of DHMT counterparts to every CA activity. However, AMREF will need
to ensure that priority areas identified are addressed appropriately and timely feedback to
USAID/Kenya is provided.
That said, USAID should consider formalizing the position of the DHMT deputy
manager in the contract agreement. It was felt that the deputy manager should be given
more authority in decision-making not only to facilitate operations in the absence of the
project manager but also as an element of capacity building. A formal agreement would
give legitimacy to this expressed concern of the District.
B. The role of USAID/Kenya
The role of the USAID/Kenya Mission, as described by the 1st Year Plan, was to provide
direct management and technical assistance, through the AMREF mechanism, in support
of in-service training, behavioural change, improved management and administrative
systems, quality assurance and health financing, operational research, and where
appropriate provide resources for commodity procurements.
29
USAID/Kenya monitors the project principally through three mechanisms namely:
q Quarterly meetings comprising the project manager, members of the DHMT, a
member of the DHMB and the National Malaria Control Programme. At this
meeting the project manager presents a summary of the programme, highlighting
progress and constraints. This affords an opportunity to raise policy issues that the
government representative can respond to or take note of for further action.
q Quarterly programme reports, which are prepared jointly by the project manager
and the DHMT. These usually constitute the objectives and activities for the last
three months. They would also include plans for the next three months. The
quarterly reports are basically narrative rather than analytical in content, making
them less useful for USAID monitoring purposes. The extent to which these
reports were disseminated and read by other internal USAID staff or by the
implementing partners was not reported.
q The USAID programme officer responsible for the project makes quarterly field
trips. Frequent correspondence and communication among the various
implementing partners and with AIMI staff at USAID/Washington as needed
complement these field trips.
 i. Discussion & Recommendations on the Role of USAID/Kenya
USAID/Kenya has provided both managerial and technical support to the formation of
the BDMI project. It was felt, however, that senior health staff at the Mission has not
provided sufficient programmatic direction. Additionally USAID/Kenya has not been
active enough in representing Bungoma accomplishments and issues at the national level.
By way of example, the contribution of relevant operations research findings (discussed
later in this report) to the planning and implementation of child health strategies such as
IMCI, and the policy implications of Bungoma's district-level demonstration activities,
require vetting and dissemination at the national level in order to inform the national
30
planning process. The MOH and some donors are currently taking note of the work being
undertaken in Bungoma and much could be gained, programmatically, by having USAID
broker some of the ongoing discussions around the Bungoma experience.
USAID senior health staff could play a more active role in representing the achievements
of BDMI at the national level by participating in the relevant task forces/technical
committees. This would serve to bridge this pilot USAID-funded project with national
malaria control, IMCI, RBM, and other related strategies currently under development.
Such a link would facilitate leverage of resources from other development partners.
Furthermore, if the current reporting process is not meeting the project monitoring needs
of the Mission, the issues should be raised with AMREF management. This may require
that USAID and implementing partners review the current monitoring and evaluation
strategy. This was not well articulated to the evaluation team and subsequently not
adequately assessed. In particular, the strategy should distinguish between the need for
programmatic or activity-level data that would allow USAID and project management to
make interim adjustments in specific project areas, and the need for more rigorous
measures of district-level performance and impact.
C. Funding Mechanisms
The BDMI is jointly funded by the USAID with in kind contributions from the
Government of Kenya. USAID is providing, on average, funds in the amount of US $ 1
million annually. The funds are channelled through AMREF headquarters (USA); CDC
(Atlanta, USA); and QAP/Bethesda. The Collaborating Agencies (CAs): CDC and QAP
were funded on the basis of agreed upon activities. Local costs, however, were paid by
AMREF.
From March 2000, under the USAID’s bilateral support to Kenya, a direct agreement
between the USAID office in Nairobi and AMREF/Kenya was signed in the amount of
US. 450,000 dollars, excluding the local costs for CAs. The CAs are now expected to
31
incorporate the local cost component in their overall budget. This arrangement is
expected to streamline disbursements and improve the management process of the
project. The affected CAs have indicated their dissatisfaction with this arrangement due
to the possible inconveniences that it could entail. In this regard, the affected CAs have
requested to channel funds for local costs through AMREF under the agreement between
USAID/Washington and AMREF/USA. USAID/Washington has agreed to this proposal
but will implement it only after October 2000.
3.2 Project Implementation
A number of project activities were planned on annual basis although the project
agreement is for a five-year period. For the period under review, reference is made to the
work plans for the years 1998, 1999 and 2000. Due caution has been paid to the work
plan for the year 2000 since the review has occurred about mid-way of that year.
Project activities in the first year (1998) were concentrated on operations research and the
rehabilitation of the District health facilities mainly at the District hospital. In addition, it
was devoted to establishment of the project office through personnel recruitment,
procurement of project equipment and related supplies, and vehicle rehabilitation.
During the first year, 9 operations researches, 3 training programmes and 2 workshops
were planned. These activities were undertaken except for operations research where 7
researches were conducted. Based on the training, health workers began to apply their
skills towards improved management of fever and anaemia at the health facility level.
The second year (1999) was devoted to operations researches, training programmes,
problem-solving teams, IMCI and DHMT supervisory visits. It was also the period when
IEC materials were planned for implementation. In addition procurement of vehicles,
laboratory diagnostic and office equipment were purchased. Most of these activities were
undertaken as planned. However the IEC component was not implemented because of the
expiry of the contract of BASICS with USAID. A few other activities did not fully meet
32
their implementation levels for instance operations research where 5 out of 8 researches
were implemented, training of health workers on the new guidelines on diagnosis and
treatment, prevention and control of malaria and both IMCI and DHMT supervision.
Issues arising from the first two years of the project relate to the design of the project. It
was expected that the first year of the project would be devoted to operations researches
alongside logistical activities. The operations researches were supposed to inform the
interventions to follow in the second year of the project. However this did not happen as
planned as operations researches were carried over to the second year delaying the start￾up of activities.
During the year 2000, planning of specific intervention activities was provided for under
each objective and this marked the beginning of a full-fledged intervention phase of the
project.
The following sections organise the achievements of each activity as well as
recommendations for their improvement around the objective under which they were to
contribute.
B. Objective 1: Improved management of fever and anaemia, principally among
children under five years of age by health workers at the health facility
Eight activities are planned under the case management at facility level objective.
These include training of health workers in case management, adaptation of national
IMCI materials to district level, review of IMCI referral registers and also undertake
DHMT & IMCI supervision. On the whole the implementation of activities under case
management at facility level objective is on schedule. Some of the activities that are
ongoing include DHMT and IMCI supervision, coaching and support of teams. The only
activity that is yet to commence is the development of clinical guidelines for malaria
microscopy.
33
The use of IMCI as the main mechanism for accelerating this BDMI objective is
consistent with global initiatives (such as AIMI and WHO/Rollback Malaria), and with
the MOH's own strategies to reduce malaria-related morbidity/mortality and strengthen
local capacity to deliver effective integrated malaria control. The IMCI activities
undertaken so far within BDMI contribute, in some measure, to all three components of
the WHO implementation strategy:
IMCI's first component primarily involves the improvement of health workers' skills and
practices at the health facility level using a systematic, clinically tested approach to case
management of malaria in children under five years of age and pregnant mothers. The
approach also includes preventive and promotional activities at the facility level
including encouragement of breast-feeding and promotion of Vitamin A supplementation.
In its second component, improving health systems, IMCI requires that SP and pre￾referral drugs like quinine are made available and, through the IMCI adaptation process,
ensures that corresponding changes are made to anti-malarial drug policy. Adequate
monitoring, supervision, and referral systems are also major considerations, as are
mechanisms to improve the organisation of work.
The operational aspects of the third component, improving family and community
practices, are currently being explored for Bungoma District. This component promotes
16 key practices at the family/household level for improved child survival, growth and
development, many of which contribute to malaria control strategies at the community
level. This incidentally is consistent with the recommendation of the APHIA mid-term
evaluation team to adopt an overriding community-based approach in targeted districts.
i. IMCI Training
Training ranks high in the priorities of the Ministry of Health as spelled out in the Health
Sector Policy Framework. The BDMI project integrates training within the DHMT's
planned annual activities.
34
A variety of training activities relating to IMCI have been conducted as follows:
q 125 operational health workers from both the GOK and NGO health facilities;
q Ten teams from ten rural health facilities have undergone a problem￾solving/quality assurance course;
q 813 health workers have been trained on new guidelines on malaria diagnosis,
treatment and prevention;
q 80 public health officers and public health technicians have been trained on the
use of insecticide-treated nets;
q 6 DHMT members have been trained on TOF;
q Two DHMT members have been trained in IMCI in Tanzania.
q An unspecified number of DHMT and facility-level staff have been trained in the
conduct of health facility assessments which utilise IMCI performance indicators
and in relevant community-based surveys
Future training will include some training in the local adaptations for IMCI, refresher
training in the new malaria guidelines as needed, and proposed training for Community
Owned Resource Persons (CORPs) in the simplified malaria guidelines as well as
training of TBAs (see Objective 3).
a. Discussion & Recommendations for IMCI Training
CDC's health facility surveys and post-training IMCI evaluations confirm that training
improves performance, but the data also indicate that significant problems still exist in
the assessment and management of severe illnesses. This may be due in part to the
infrequent number of severe cases seen by IMCI trained health workers at the first level
referral facilities and perhaps could be remedied by exposing these staff to more cases in
a larger hospital setting. Additionally, the problem could be one requiring more focused
monitoring and supervision post-training, or if it is a behavioural issue then some
formative work may be needed to explore the underlying causes.
35
In any case, these are critical issues, which should be investigated through additional
operations research. The relative 'seriousness' of the problem of misclassifying severe
illness by IMCI trained health workers should be examined as this will have
programmatic implications for the expansion of IMCI and the achievement of project
outcomes. At present, the DHMT and implementing partners should re-think the current
strategy being employed to train health workers in IMCI and should explore ways; such
as gaining practical experience with severe cases in a larger hospital setting to strengthen
post-training performance in these areas.
A number of health workers at first level referral sites have been trained in IMCI and the
initiative is ongoing. However some consideration should be given to train medical
officers and other cadres of health staff who have not yet been trained but who will have
responsibility for supervising IMCI trained staff.
On-the-job training for IMCI is being undertaken in some of the government health
facilities but in the NGO sector this is minimal. A CDC report (September 1999)
identified a number of qualities of care issues within non-government facilities. As such,
the DHMT and partners should further explore IMCI performance issues and training
needs in private, NGO facilities.
IMCI training is one of the more costly activities for the MOH. As a cost-effective
alternative, pre-service IMCI training may be an option. The review team learned that
some of the components were already being taught to students at the Medical School,
University of Nairobi, Kenya. There are plans to extend to medical training colleges and
nursing schools. Medical officers posted to the field and who are already acquainted with
IMCI would be informed supervisors, and have the ability to initiate and coordinate
training for their staff at the facility level.
Community Resource Persons (CORPs) completed simplified guidelines in 1998 for use,
but training was held up because the guidelines could not be reproduced and
disseminated. In any case, the training of CORPs is not supported by a strategy to ensure
36
that adequate supervision will be available. Based on past training experience, the idea of
CORPs 'going solo' is reportedly a common and possibly dangerous occurrence. This
suggests that the DHMT should begin to identify possible incentives as well as
supervisory mechanisms to ensure a higher return from this training exercise. Other
training activities and issues are discussed within the context of specific project activities
reported in subsequent sections of this report.
ii. Supervision
Retention and refinement of skills is improved and sustained through frequent and
supportive supervision. Supervision was reported by the supervisors to be inadequate and
health providers confirmed this in the field. There are three types of supervision: The
general DHMT supervision of rural health facilities (both the GOK and NGO), IMCI
supervision and QAP supervision.
a. General DHMT Supervision
Supervision of Governmental and Non-governmental health facilities by the DHMT is
critical. This supervision provides an avenue for the overall improvement of health
services through interpretation and implementation of the Ministry of Health (MOH)
policies at the district level.
The DHMT is provided with a standard checklist for supervision. Initially the DHMT
used to undertake supervisory functions twice every week. However, these have been
reduced presently to once a week. Currently, the BDMI supports supervision with the
provision of transport and allowances.
b. IMCI Supervision
IMCI supervision is geared towards ensuring that IMCI trained health workers maintain
knowledge and skills gained during the training as well as up-dating those skills. It also
37
enlists the support of the Health Centre Development Committees (HCDC) to supplement
drugs and other supplies in the health facilities.
Two IMCI supervisors, one clinical officer and one public health nurse, usually undertake
this supervision. These supervisors use a standard checklist and observe two sick children
being examined by a health worker. The same children are each then re-examined by the
supervisor, who gives feedback on assessment, illness classification, treatment and
counselling to the health worker. The health workers preferred this mode of supervision
as they felt that it was supportive given that feedback was immediate. These supervisors
also check supplies, drugs, dressings, vaccines and equipment using a facility checklist.
The team-based problem solving approach employed by QAP was intended to improve
the performance of health providers in applying the IMCI algorithm .The coach
coordinator reportedly conducts supervision together with DHMT members trained in the
approach.
c. Discussion & Recommendations for Supervision
The constraints experienced in Bungoma district under the BDMI are largely logistical
and relate to resource availability, both financial and human. As observed earlier,
supervision is scheduled to be carried out weekly by the DHMT. Some IMCI trained
providers had been supervised at most once or twice a year since they received the
training. A major obstacle is the lack of transport facilities noting that there is only one
reliable vehicle at the district headquarters that is assigned supervisory activities.
The need for alternatives to the current modes of supervision should be explored given
that transport and allowances are currently provided by BDMI. The expectation is that
the DHMT will somehow absorb these costs once the project ends, but to sustain the
intensity and frequency of current visits will require some considerable planning and
mobilisation of additional resources, including at community level
38
The DHMT is currently considering an integrated approach, which includes IMCI
supervision and QAP coaching which may provide a viable alternative. In addition, CDC
has developed an IMCI register, which could provide a means for monitoring quality of
care between supervisory visits.
iv. Drug Supply
Drugs are often in short supply and patients are expected to buy their drug requirements.
To bridge this gap, the community buys drugs through the Community Improvement
Fund (CIF). The problem is that there is no legal framework to regulate and control the
activities of the CIF. Unscrupulous officials defrauding the kits have often duped
community members and the community has no form of recourse.
a. Discussion & Recommendations for Drug Supply
Community pharmacy is a cost-effective strategy that fills the gap created by drug
shortage at the health facility level while making drugs more available at the community
level. Unlike the traditional BI community pharmacies, these belong to the health facility
and so far they seemed to be operating well as facility income generating activities. They
are supervised and monitored by the DHMT or the in-charges of the relevant health
facility. They complement drug vendors and shopkeepers in the community. This is made
possible by the Community Improvement Fund (CIF). However, the MOH headquarters
needs to formalize and regulate these within a legal framework to counter exploitation
and abuse. At the moment the MOH at the district level has no control over the finances
and the expenditures of these pharmacies. Subsequently the team has learnt that the
DHMT has requested the Division of health care financing to work out regulatory
mechanisms to guide the community on CIF although there still will be no control over
the finances.
The Bamako Initiative Pharmacies (community-based pharmacies), like those being
implemented in Siaya, utilise trained and certified CHWs to make needed drugs more
39
available to the community. The lessons learned from this initiative should be examined
for their relevance to Bungoma.
Community-owned resource persons (CORPS) and drug vendors handle drugs such as
SPs. These drugs are gazetted as part II poison. BDMI should seek clarification from the
National Malaria Control Programme and the Pharmaceutical Board to avert the fear by
CORPS of being prosecuted for handling prohibited drugs.
v. Patient Referral and Compliance
IMCI is an effective strategy to deal with the major causes of morbidity and mortality
especially in peripheral health facilities that are usually poorly equipped. But IMCI is
designed to operate in the periphery as initial management and not final, which
underscores the importance of having an adequate referral system.
a. Discussion & Recommendations for Referral
Referral remains problematic and it is acknowledged that this is a national problem. The
factors contributing to the observed problem range from structural to institutional, and
include lack of transport, money, poor reception and shortage of essential supplies at the
referral facilities. In the absence of all these essentials, patients opt to remain away from
referral facilities. In many instances patients did not go to the referral facilities but rather
“referred” themselves back to the initial facility whereby providers felt obliged to treat a
condition that was really for higher level management.
An assessment of the behavioural factors, which affect patient compliance with referral,
should be undertaken. The supervision summary provided to the evaluation team by the
DHMT during the 2-day workshop indicates that supervisors will review the referral
registers that record children with severe classifications. A mechanism will presumably
be employed to follow up children who have been referred to identify reasons why they
40
do or do not comply. This will be done in the context of a formal referral study
conducted with assistance from CDC.
In addition, the referral system in Bungoma lacks basic inputs such as ambulances. It was
observed that no government facilities had them and only a few NGO facilities reportedly
have them. The main District referral hospital does not have a functional generator.
These and other shortcomings in the current referral system undermine efforts to improve
the pre-referral practices of IMCI trained health workers and should therefore be given
some consideration in future project planning exercises.
vi. Laboratory
Laboratory diagnosis is crucial in providing a confirmatory support to clinical diagnosis.
This requires that essential equipment be available and functional. At the stakeholders
workshop it was reported that laboratory equipment in the district hospital and the health
centres was obsolete. A presentation at the workshop based on assessment of microscopic
diagnosis of malaria in the health facilities raised issues of concern with regard to the
accuracy of the results noting that the negatives were too high with a predictive value of
93. This raised the issue of quality control, which should be an integral part of laboratory
services. Malaria microscopy is more critical at higher referral levels since most
dispensaries don't have laboratory facilities. But adequate basic laboratory equipment,
supervision and quality control, as well as refresher training for the laboratory staff
should be provided in those facilities where microscopy is available.
B. Objective 2: Improved capability of mothers and other caretakers to manage
fever and anaemia at household level
Four activities were planned under the household management of fever objective for the
year 2000. Under this objective only two activities have been accomplished namely:
41
q Completion of the training materials for drug vendors and their subsequent
training on rational use of drugs; and
q The development of IEC messages for household level, child to parent using
microteaching on fever, rational drug use and on ITNs.
In the current plan period, there will be piloting and adaptation of the simplified malaria
guidelines for CORPS and the development of IMCI guidelines on community practices
(16 household practices/ behaviours). This will be followed by supervisory visits to
trained drug vendors and developing a monitoring system for mystery shoppers of private
drug outlets (see Operations Research section below).
i. Discussion & Recommendations for Improved Household Management
Based on an early baseline study done by BASICS on health seeking behaviour of
caretakers, it was found that over 90% of caretakers treat fever at home ("treatment" as
defined by the caretakers by giving either an antimalarial or antipyretic). The 1996 CDC
household survey found that 47% of children received an antimalarial at home; 90% of
antimalarial treatment at home was started by the second day of fever. Other CDC studies
recommended that home treatment be strengthened through health education and that
home care guidelines be developed for caretakers since home treatment was provided as
often and more promptly than treatment at a facility. Studies such as these should have
triggered the need for a community focus much sooner.
There is good information already collected in Bungoma on behavioural and preventive
aspects of health. The CDC research on malaria in pregnancy also indicates that in terms
of home care and care-seeking practices, caretakers are principle providers of anti￾malarial treatment to children with fever.
These and other studies, again provide a strong evidence base for undertaking
community–based approaches to complement the activities currently being implemented
at the facility level. The approach should build on the existing community structures and
42
activities of CBOs and NGOs working to address the key household and community
behaviours relevant to those circumscribed by the Community IMCI component.
For example, the CARE-Kenya's Community Initiatives for Child Survival in Siaya is
working to reduce mortality and morbidity of children under five and women of
reproductive age and to increase the capacity of community committees, the MOH, and
other local institutions to sustain these reductions. Lessons learned from the community￾based strategies employed here for example; training of CHWs, generation of health data
by the community, use of community based pharmacies and increased IEC should inform
the development of the Community IMCI component for Bungoma District.
The need for multi-sectoral collaborative effort is critical to rationalize use of resources.
IMCI and the current Early Childhood Development (ECD) under the Ministry of
Education are essentially focussing on the same concerns: the child and the caregiver, yet
there is no collaboration between the Ministry of Health and Ministry of Education.
Several ministries could be added to the list, for instance Ministry of Agriculture & Rural
Development and Ministry of Home Affairs, National Heritage & Sports, to adequately
address the issue of resource mobilisation at the household and community levels for
improvement of child and maternal health.
C. Objective 3: Improved prevention and management of malaria in pregnancy
Four activities were planned under the prevention and management of fever in
pregnancy objective. Out of these four planned activities, only the development of IEC
programme on SP, folic acid/iron supplementation for pregnant women had been
undertaken at the time of the review.
From the field visits, the evaluators noted that in NGO facilities, ANC mothers are not
given presumptive treatment with SP. If they are given at all, it is selective for those
mothers who can afford to pay for it. This also applies to folic acid and ferrous sulphate
43
tablets. In all the government health facilities visited, presumptive treatment for pregnant
mothers attending ANC was provided though not always consistently and correctly.
Randomly selected exit interviews with mothers indicated some potentially serious
problems. In one health facility SP was given in the ninth month, yet this is inconsistent
with current guidelines which state it should be given at the beginning of the second and
third trimesters. When SP was given there was no explanation provided to the mother as
to why this was necessary. From the November 1998 CDC study, it was found that health
workers had little understanding of malaria prophylaxis in pregnant women and how
prophylaxis differed from treatment. Given these, the DHMT should review available
data to determine the extent of this problem with a view to exploiting all such
opportunities to improve health worker compliance with the malaria guidelines as well to
impart the relevant messages to ANC mothers regarding malaria prevention and
treatment.
In addition, the evaluation team observed that when the supply of SP is irregular at the
health facility level there is a tendency to use it for general treatment rather than
prophylaxis for ANC mothers. Moreover, in facilities with Facility Improvement Funds
(FIFs) that now charge for services, the provision of SP for ANC mothers is problematic
since the current policy states that maternal and child health services are to be provided
free of charge. The latter indicates that policy level intervention will be needed to ensure
that adequate SP supplies are included in the MCH kit and that ANC mothers get the SP
when needed.
From a baseline study by the CDC on malaria in pregnancy, it was evident that whereas
65% of ANC mothers come for ANC services, they do so late in their pregnancy.
Moreover, a similar percentage delivered at home under Traditional Birth Attendants. To
be able to reach a majority of ANC mothers, interventions beyond the health facility will
be needed. Appropriately, one of the planned activities will be to train CORPS
(Community Owned Resource Persons) especially Traditional Birth Attendants (TBAs)
to effectively support care seeking among ANC mothers.
44
D. Objective 4: Increased household use of insecticide-treated materials
Six activities were planned under the provision of insecticide-treated nets (ITNs)
objective and the implementation of these activities is on course. The activities
implemented so far in the year 2000 are:
q Community mobilisation for ITNs and establishment of Organized Community
Groups (OCG).
q Training of Education Officers/teachers and DHMB on malaria control and ITNs.
q Finalization of training materials and training of PHT/PHO and other health
workers.
q Training of community Initiative sites officials and HCDC officials and
implementation of ITNs programme at health facility.
Based on the year 2000 work plan, 4000 nets and 4000 K-O tabs have been purchased by
16 Community ITNs outlet/ distributors or Organized Community Groups (OCGs).
AMREF and the DHMT have provided training to these groups.
i. Discussion & Recommendations for Increased Household Use of ITNs
ITNs are a cost-effective alternative to other vector control methods, which call for
specialized equipment, technical expertise, a large workforce and widespread application
of chemicals. In areas with seasonal and perennial transmission of Malaria in Africa, the
use of ITNs reduced malaria morbidity in children by between 44% and 63% (Premji et
al 1995, Nevill et al 1996). In areas of seasonal transmission in Africa the use of ITNs
reduces all cause mortality in children by between 17% and 63%(Binka et al 1996, Nevill
et al 1996 ). The use of ITNs has also reduced mosquito population, reduced malnutrition
and led to improvement in child growth. It has further increased communities’
commitment and enthusiasm to adopt the technology in the control of malaria and
willingness to buy nets and participate in net re-impregnation process.
45
AMREF is in-charge of community mobilisation and ensuring that nets reach community
members. In this respect a community-based distribution model seemed the most
appropriate for Bungoma district based on AMREF’s experiences both in Kilifi district
under WHO/KEMRI controlled trials’ study and the Community-based Malaria Control
project in Sagana, Turkana, Taita Taveta and Migori districts. It also echoes aspects of
the Bamako initiative in which bednets were bought and distributed to community groups
at a subsidized rate for sale.
This is not in line with USAID’s view that the way forward is to promote the value of
bednets and thereby create demand for them through social marketing. These bednets
would in turn be supplied in the market by the private sector. The social marketing
strategy is advocated for Bungoma district where various forms of media can be
employed to reach a wider audience.
One also wants to make sure that the products for social marketing are available in the
community. For this reason AMREF community based approach is an initial way of
addressing availability; elsewhere referred to as “pump-priming”, before social
marketing. To increase the coverage it may therefore be worthwhile to employ the
strategy of social marketing at this early stage as a complement to community-based
distribution. Given that the government fully supports a social marketing strategy, it
would be worthwhile from the onset that this strategy and a community-based
distribution strategy be combined in order to increase access. In this respect, Population
Services International (PSI) has had experience with social marketing of bednets in the
coastal region and it would be worthwhile to explore links between AMREF and PSI for
social marketing of nets in Western Kenya.
The creation of demand for bednets through identified community outlets and willingness
to pay for them as a means of protection against malaria will be a major achievement
towards the project goals. Community members recognise the value of nets as prevention
against malaria. However the issue of availability and particularly affordability may
prohibit the actual use of ITNs.
46
The issue of poverty resonated in all the discussion groups. A visit to the health facilities
and a discussion with the ITN providers revealed that there is willingness among
community members to use bednets as protective mechanism against malaria. However,
the issue of cost is real and it is not quite clear how behaviour change can be achieved at
community level in the absence of uplifted standard of living among the population to
acceptable levels to enable them participate fully in the development processes.
E. Objective 5: Effective collection and use of information for planning, monitoring
and evaluation
Five activities were planned around the effective collection and use of information for
planning, monitoring and evaluation objective.
The effective collection and use of information is poor. Generally government health
facility reporting rates on vital statistics are low (43% in 1999) while no information is
received from the NGOs. Of the planned activities, two are ongoing: Analysis and
distribution of DHMT/IMCI supervisory visit reports and End of Training (EOT)
assessments. These are intended for monitoring the project activities.
Operations research is undertaken on the premise that although the issues are known the
studies provide evidence base for interventions. However without baseline data it will not
be possible to measure impact. This is compounded by the fact that record keeping at the
moment is very poor. At the workshop it was stated that reporting from the various
facilities had dropped from 90% in 1995 to 43% in 1999. Furthermore the available
facility records were only from five facilities and this was sporadic such that they could
not be used for comparative purposes. Yet for the project, effectiveness of monitoring
and evaluation tools would depend on available and reliable data, which are not being
collected at the moment.
In November 1999 an attempt to collect morbidity and mortality data on Malaria in 9
health facilities in Bungoma district revealed gaps in data. The data that was availed to
47
the team shows that there is no systematic form of data collection and for this reason the
data sets from the different facilities were not amenable to comparison. A cursory look at
records by the review team in some of the health facilities revealed that even where there
were tally sheets these were not routinely used and in others the tallying was inconsistent.
In cases where data were collected and forwarded to the district headquarters, there was
no evidence of analysis and feedback to the collecting health facilities.
With this realisation the DHMT has developed data collection forms which are being pre￾tested at selected health facility and community levels. It will be crucial to harmonize
these forms with the HMIS at the headquarters to avoid parallel systems of data
collection.
More importantly given the fact that there was a prior system in place that has since
collapsed it may be cost-effective to rehabilitate this system through provision of supplies
for instance; tally sheets and retraining of data collection staff to re-orient them on the
importance of collecting data. This is crucial for the project in that as it moves into
implementation, indicators for monitoring would have to be tracked on baseline data.
F. Operations Research: Appropriateness and Timeliness
As stated earlier, operations research was intended to provide the evidence base for the
selection of project interventions. To this end, various collaborating agencies undertook a
number of operations research studies within the project area. However, the sense was
that the project should have moved from research to implementation much sooner. The
delay may have been due in large part to the length of time it took for the CAs to start up
activities as well as to process and feed back their results. With so little time left for the
project and given that BDMI activities are expected to contribute to the goal of reduced
malaria related morbidity and mortality in Bungoma, it is uncertain how the current and
proposed activities will be able to operate at such a scale as to achieve impact.
48
In addition, dissemination of project activities, including findings from research
conducted has been inadequate and slow. Due attention should have been paid to the
dissemination of this body of work to a wider audience given their potential contribution
to national-level program planning, as is the case for IMCI and Roll Back Malaria. As
such, the current study activities and findings should be made more accessible to relevant
units within the MOH, research institutions, and other local partners who could benefit
from the technical exchange of information around malaria control and the mechanisms
such as IMCI and ITNs being used to achieve impact. To give the results a high profile
they should be disseminated under the auspices of the MOH with the support of USAID
either through organised workshops or special meetings around programmatic themes.
The participation and ownership of these studies was reportedly also an issue with the
DHMT, particularly with regard to the CDC research. The DHMT were involved in the
research process principally at the level of data collection. They expressed dissatisfaction
that they were not, however, adequately involved in data analysis, an area they felt would
contribute to capacity building in terms of data management and report writing. CDC
acknowledged this fact but felt that their schedule did not always allow for in-country
data analysis, at times the DHMT staff was busy with other responsibilities at the time.
If additional research is needed, BDMI and its implementing partners should consider
undertaking more operational, more timely, and perhaps less rigorous studies around the
current program interventions in order to better gauge their 'effectiveness'. In the case of
IMCI, the use of the approach as an investment strategy for the health sector as well as a
viable means for accelerating BDMI objectives needs to be further demonstrated. To that
end, studies such as those to address the cost-effectiveness of alternatives to undertaking
IMCI supervision and training, cost-savings resulting from its application, and effects on
client utilisation and health worker performance at facility-level. Similarly, use of a
simulated client approach (not unlike the current QAP/DHMT activity) or other tools to
monitor IMCI compliance, among others, could further inform the national process which
is now beginning to roll out and serve to build advocacy within USAID as well as among
its collaborators.
49
The following sections describe the main Operations Research (OR) activities
implemented by each CA and the corresponding findings of the evaluation team.
i. The Centres for Disease Control and Prevention (CDC)
To date, CDC has conducted studies in the following areas:
q Pre-BDMI and BDMI quality of care studies at health facilities (including
baseline and follow-up GOK health facility surveys, a referral study, and a
baseline of Health Worker performance in non-GOK facilities)
q Malaria in Pregnancy
q Assessment of Microscopic Diagnosis of Malaria
CDC on balance has produced studies that are scientifically valid and of programmatic
value. Some of the research has provided critical input to the development of IEC and
social-marketing strategies. Other studies have leant further support to previously
identified areas for intervention e.g., the use of Traditional Birth Attendants (TBAs) to
support care seeking among antenatal mothers.
However, it would seem that much of the research data generated has not been utilized,
disseminated, or the resulting issues vetted in such a way as to adequately inform the
BDMI planning process or facilitate implementation on the ground. For example, data on
the classification of severe illness by trained IMCI health workers gathered during the
1994 baseline health facility survey, immediately after IMCI training in 1996, during
supervisory visits conducted 1-3 months post-training, during the 1997 follow-up facility
survey, and during supervisory visits conducted between July 1998 and October 1999,
indicate that the failure of health workers to correctly classify and treat severe illness is a
consistent deficiency. Such findings should have lead to more timely investigation or
intervention. Only recently has discussion of the issue been revived, and this is largely in
response to USAID/Washington concerns that no remedial actions have yet been
undertaken.
50
As described earlier under Objective 2, CDC studies had indicated that home treatment
was provided as often and more promptly than treatment at a facility and that caretakers
were principle providers of antimalarial treatment. Yet, these had not lead to the timely
development of a strategy focused at the household and community level to improve the
capability of mothers and other caretakers to manage fever and anaemia (BDMI's
Objective 2).
The November 1998 Malaria in Pregnancy study underscored some important issues:
deficiencies in health worker knowledge of malaria treatment and prevention, selective
provision of ANC services, and mothers' misperceptions about malaria, among others.
Yet although an objective was to assess the extent and impact of malaria infection during
pregnancy, the study was conducted in the low malaria transmission season thus
presenting lower than anticipated prevalence of parasitemia. A repeat survey had to be
scheduled for the high malaria transmission season.
As discussed earlier, CDC and other implementing partners should work to more
effectively and expeditiously disseminate study findings to the multiple 'end-users' i.e. the
DHMT, USAID, CAs, MOH, and others in the research community such as KEMRI.
Given the short time remaining in the project, more consideration should be given to
maximising the use of prior research by implementing the recommendations, before
embarking on new ones. Any future research should perhaps focus more on assessing the
effectiveness of the interventions currently underway in the true spirit of 'operations
research', although exceptions, such as formative work to explore the problem of
misclassification of severe illness, may be needed.
ii. African Medical and Research Foundation (AMREF)
AMREF has a double role of manager/coordinator and implementor, with 75% of the
time devoted to management and coordination, and 25% on implementation. So far
51
AMREF has conducted three studies and some of their recommendations have been
utilised in defining interventions. These researches include:
q Assessment of Non-technical Barriers to Supervision that led to establishing a
strategy of supportive supervision based on work plans with set targets,
supervisory guidelines and checklists.
q Community KAP on Insecticide-treated Materials (ITMs) & Potential Delivery
Systems, and
q Assessment of Community Financing Systems for Mosquito Nets & Insecticides
by AMREF and the DHMT.
iii. Basic Support for Institutionalizing Child Survival (BASICS)
The operations research carried out by BASICS prior to the expiry of its contract
included formative research on Health Seeking Behaviour of Caretakers and Assessment
of IEC Materials. These studies formed the basis of the IEC Strategy developed by
BASICS in collaboration with the DHMT and community representatives. Some of the
activities that resulted from this research included:
q IEC messages’ development workshop.
q IEC materials development workshop and subsequent pre-testing of both
messages and materials for ANC mothers emphasizing presumptive treatment
using SP; for caretakers emphasizing early recognition of symptoms of Malaria
and early health seeking.
q A poster emphasizing fathers’ role in facilitating early health seeking.
q A poster for ITNs’ promotion.
q Training and support of Shopkeepers as drug outlets.
However, given the significance of the findings contained in the research on health
seeking behaviour one would have expected more community-focused interventions.
52
iv. The Quality Assurance Project (QAP)
The Quality Assurance Project currently implements two research activities under BDMI:
q assessing the cost-effectiveness of using facility-based problem-solving teams to
improve health worker compliance with IMCI standards
q improving private drug outlets' compliance with malaria treatment guidelines.
Based on the July 1997 CDC findings that showed declines in health worker performance
levels with regard to IMCI compliance, CDC requested the assistance of QAP to address
this issue. QAP has been working in a number of other contexts to apply the techniques
and tools of quality management, including the use of work teams, as a means for
identifying and resolving quality of care problems. The working assumption of the team
approach is that, with proper training and 'coaching', teams comprised of health workers
will be best able to understand the causes of problems in their facilities and propose
suitable solutions to address them. Thus for BDMI, the expectation was that healthcare
providers trained in IMCI would apply this approach to identify specific problems in the
application of the IMCI standards and subsequently improve performance in those areas.
Around the first of the two QAP activities above, the evaluation team identified a number
of critical shortcomings based on the presentation of preliminary study findings, review
of available documents, interviews with the relevant DHMT and health facility staff, and
subsequent discussions with QAP headquarters staff. These issues, which are related to
the design, implementation, and outcome of the study intervention, included the
following:
One of the critical concerns with regard to IMCI implementation was/is the non-use of
the approach in the absence of a study observer or IMCI supervisor. The evaluation team
gathered similar anecdotal evidence from ad hoc interviews with IMCI-trained health
workers who cited time and other constraints as factors limiting the effective application
of IMCI. Recognising the difficulties faced in collecting such performance data, QAP
53
nonetheless tried to engage the teams to address this issue through a prescribed problem￾solving approach. Despite a sincere and concerted effort on the part of QAP, AMREF,
and the DHMT, to address this particular concern as well as other possible underlying
problems, the preliminary results were not encouraging.
The team-based activity appeared to be heavily resource and time dependent with no
demonstrated effect on health worker compliance (the main dependent variable). Even
though the health staff often mentioned other benefits such as teamwork, better
organisation of work, and the crosscutting nature of the approach as a management tool
interviewed.
.
The study intervention i.e. problem-solving teams may not have been adequately
understood or defined prior to the start of the study as evidenced by the difficult start-up
as well as the use of narrowly circumscribed performance indicators. An inexperienced
trainer was used to introduce the approach and the subsequent need was to re-train the
teams as well as add on substantial 'coaching' or external support to apply the problem￾solving steps.
Regarding performance measures, the QAP pre- and post- assessments primarily
evaluated how well the health workers were following the specifications of the IMCI
algorithm. Although QAP was presumably working to address the issue of non-use of
IMCI as described above, and stated that it was still too early in the process to be seeing
significant changes in performance, the researchers did not change the performance
measures or otherwise define the parameters of improvement that could be attributed to
the work of the teams.
The introduction of the team approach requires that team members gain new
competencies not only in teamwork but in the use of problem solving tools. The desired
performance outcome would thus depend in part on the coach's and team's skill in
conducting problem analysis. These skills may not have been sufficiently measured or
assessed as such. From the conclusions presented at the workshop, "teams in general have
54
not adequately mastered problem-solving tools"… "..some coaches have not adequately
mastered problem-solving tools", "the tools developed for team monitoring were not
analyzed and used effectively by coaches to improve team performance, in part because
they were not fully mastered," and "few coaches demonstrated a high level of coach
competence," among others.
In addition, facility-level issues such as organisation of work, time management, staffing
patterns, system improvements -- all of which may have particular relevance to the
creation of an enabling environment for IMCI and, moreover, are often 'default' problem
areas that QAP teams elsewhere are commonly known to address, were not
conceptualised as part of the original study design.
The teams were not coached to use the baseline data, particularly that regarding clinical
service quality, in their problem-solving activities. The resulting solutions for example
procurement of drugs, time allocation and scheduling of staff, on-the-job training in
IMCI, punctuality and other mundane problems such as getting water to the facility, were
not seen by the evaluation team to contribute substantively to the expected IMCI
performance outcomes.
The review team felt that supportive supervision, whether internally done by an in-charge
or externally through designated IMCI supervisors, could better address many individual
performance problems identified in the baseline assessment rather than a team-based
approach. The rigorous stepwise process prescribed by the QAP trainers, along with the
implied need for consensus among team members, should not be utilised for every
problem, particularly when common sense and experience should prevail. Yet the few
team members interviewed by the evaluators were insistent that the complete process as
opposed to a more selective approach to using various tools had to be applied every time.
Other factors may have contributed to the failure of this activity to achieve results. For
example, although the DHMT was said to have nominated the participants for training,
the activity appeared to lack adequate and informed ownership at the District level. Some
55
BDMI staff viewed the QAP work as an adjunct activity, not clearly contributing to the
broader aims of the project.
In summary, the desired improvements in IMCI performance as a result of the QAP
teams were not realised. The approach was acknowledged to have value as a management
concept, although the discussions around its practical usage remained diffuse and
sufficient evidence of its achievements with regard to facility-level improvements was
not provided. In light of these findings, the DHMT and implementing partners should re￾examine the contribution of this team-based approach towards the achievement of BDMI
objectives, critically assess the 'value added' of this activity given the remaining time and
resource constraints of the project, and modify implementation as needed.
The second QAP activity is testing a social marketing strategy to increase the likelihood
that private sector vendors will adhere to malaria treatment guidelines. The innovative
approach has been well received by the DHMT and ownership of this activity has been
largely taken up by DHMT staff as described below.
v. District Health Management Team (DHMT)
In 1998 the DHMT conducted a survey to identify groups and individuals involved in the
actual distribution of anti-malarials and anti-pyretics in the district. Shops or kiosks were
found to be the main source of drugs for people in rural areas, and while most
shopkeepers had never received training on drug use, 87% were giving some instruction
on dosages to their customers and 27% couldn't differentiate between anti-malarials and
anti-pyretics. The joint QAP/DHMT study currently being undertaken is testing an
intervention to use drug wholesalers (wholesalers, mobile venders and large retail
pharmacies) to communicate the malaria treatment guidelines to drug outlets
(pharmacists, pharmacy attendants and shopkeepers). The intervention has two main
components:
q the "social marketing" of job aids using a vendor-to-vendor approach and,
56
q the use of mystery shoppers and supervisors to monitor the implementation of
standards
As was presented by the study team, this ongoing activity shows promise as an
alternative to traditional and more costly approaches to training drug retailers. By taking
advantage of the existing interactions between wholesalers and drug outlets, and creating
mechanisms to ensure that pharmacists and shopkeepers give SP correctly and safely, the
intervention could help minimise the adverse outcomes that may be associated with the
dispensing of inefficacious or expired drugs and the receipt of incorrect information by
the community.
G. Information, Education and Communication (IEC) Strategy
IEC is crosscutting and therefore is expected to be addressed in all the objectives. The
move towards early recognition of malaria, prompt health care seeking by the caretakers,
presumptive (therapy) treatment with SP for pregnant mothers, use of ITNs and
subsequent reduction in morbidity and mortality was supposed to be propelled by
strategically translating the developed IEC strategy into action. To facilitate this,
BASICS and the DHMT identified the key activities for implementation. Unfortunately,
BASICS contract expired and apparently, it was not clear who would manage or fund
some of the activities. AMREF took over this component but in practice other CAs shares
it. For instance, Quality Assurance Project (QAP) took over the vendor to vendor IEC
component to improve the community drug outlets. Job Aids were developed for the
vendors and posters for caretakers. Additionally vendors were trained on how to use the
Job Aids as trainers of caretakers. QAP plans to train 'mystery shoppers' as well.
The John Hopkins University (JHU) undertook a related IEC activity focusing on Acute
Respiratory Infections (ARI) based on the DHMT recommendations that pneumonia is
one of the complications of malaria in the district. Although it was not a BDMI activity, it
was complimentary to the project. Calendars and flipcharts on ARI were produced and
used within the context of IMCI for the whole district.
57
The co-ordination of the various CAs’ activities particularly the IEC component was
problematic, resulting in overload and overlap of information imparted to the community
whether through mass media, counselling at the health facility, or any other form of
outlets. Harmonization of messages is therefore a key challenge for the project. At the
moment AMREF is alone in the IEC area and will need to harmonize the messages to
reduce the information overload to the community.
4. KEY ISSUES
A. Capacity Building & Sustainability
One of the key areas of concern is the capacity building of the DHMT members to be
able to carry out activities once BDMI has concluded. This will depend on the extent to
which they have been prepared for the task. The DHMT is highly motivated and
identifies positively with the project. Training has equipped the DHMT members with
skills and they are able to impart the same to others through on-the-job training thus
having a multiplier effect at the local level. This will become increasingly so nationally
as the IMCI adaptation process takes root.
However, the need for management skills for the DHMT and other health providers
manning health facilities resonated in the discussions. This was based on the observation
that with decentralisation the management of resources, particularly finance is pausing a
challenge to health providers who have no experience or training in management and yet
are expected to undertake this role within their facilities and those below them. Thus
management training was identified, as a priority need.
The DHMT members have been involved in research alongside the CAs and were proud
to have undertaken one research on their own and translated it into an intervention.
However, the DHMT members decried their not being involved in the data analysis,
report writing and dissemination activities by the CDC group. The involvement of the
58
DHMT would facilitate a sense of ownership as well as demonstrate partnership and
capacity building as stipulated in the project goal and objectives.
During the evaluation interview, the CDC group conceded to this concern but felt
handicapped in that their work schedule in-country does not allow for undertaking data
analysis locally, a point that needs further consideration to ensure professional
involvement of local project personnel in research activities at all levels. For
empowerment and project sustainability it is imperative that the relevant DHMT
members be involved in all stages of the research and data management processes.
B. Perspectives on IMCI at the Health Facilities
Some additional qualitative data was obtained through discussions and exit interviews
held at the health facilities. Discussions with 9 health providers at sampled health
facilities indicated that the IMCI approach was useful and effective in diagnosis and
treatment. They indicated that they were able to systematically examine a child and arrive
at a diagnosis with ease, give the correct treatment and use drugs rationally. However,
the actual use of IMCI by these staff, in the absence of any observer or other such
prompting, could not be ascertained.
The health providers felt that correct treatment in the long run would reduce the number
of cases unnecessarily being referred to the next level of care. This was more so the case
in the Government facilities. In one NGO facility it was noted that the referral cases
coming from IMCI trained providers suggested that the initial management from those
facilities was appropriate and had improved greatly. However in this same NGO facility
the practice of IMCI was perceived as “first aid,” too time consuming, and would lead to
wastage of drugs. For other facilities, IMCI was reportedly not easy to apply unless the
patients were few.
Caretakers interviewed were impressed with IMCI and appreciate the comprehensive
manner in which a child was handled. The stripping of a child was an indication of the
59
health worker’s concern for the child and desire to give the best. Whereas those in the
queue grumbled that a health provider took too long with patients, once in the
consultation room, the caretaker began to appreciate the rationale for the extra time spent.
Caretakers were happy with the fact that they were told the diagnosis and that drugs were
administered instantly.
C. Issues for Expansion of the Bungoma Experience
Bungoma district has just embarked on a range of activities that simultaneously touch on
all five project objectives. Consolidation of the IMCI, activities within Bungoma rather
than expansion of these approaches to other Districts in Kenya should be a priority for the
project. A community-based component, which is already evolving, and which can draw
from the experiences of other donors and collaborating agencies, would contribute to
achieving the project outputs and goals. However, the issues around the facility
implementation of IMCI, including health worker compliance, training and post-training
interventions, reorganisation of work and other ‘enabling factors’ within the facility,
should remain a critical focus of the project.
Bungoma could contribute to expansion efforts by demonstrating a well-documented
operational BDMI model, one that could be replicated nationally with appropriate
modifications through the national implementation strategies for IMCI, Roll Back
Malaria, or other related initiatives. BDMI presents an opportunity to identify the critical
elements required for implementing IMCI at such scale as to achieve impact. Some
retrospective data analysis could identify the start-up and recurrent costs involved in
implementing the approach in Bungoma, and operational studies around its
‘effectiveness’ will inform the current IMCI planning process. USAID can contribute to
the policy dialogue around these and other relevant issues by participating more actively
in the review of the adaptation process and facilitating the dissemination of project
findings.
60
D. Linkages to National Health Policy and Programmes
The inception of the BDMI project was based on an agreement between the Kenya
Government and USAID, thus paving the way for acceptability even though IMCI had
not been fully incorporated in the MOH policy. In principle the IMCI approach is
accepted within the national health policy. This is based on the recognition that malaria
is a major killer among the vulnerable groups especially children and women. In respect
of the latter, the use of SP for treatment of malaria and as chemoprophylaxis is based on
the technical recommendations of the National Malaria Control Programme.
There are various instances, which demonstrate that the BDMI project is linked to
national health programmes. At the district level the DHMT acknowledged that the
project’s strategic approach had facilitated the integration of what are often parallel
programmes at this level. It was observed that health providers examined a child in a
holistic manner rather than the traditional focus on a narrow area of expertise. They gave
examples of the outreach programmes that go beyond nutrition and immunisation status
to incorporate the current health status of the child including identification of anaemia.
The national adaptation process of the IMCI has been finalized. This finalization has thus
legitimized the position of the BDMI in many respects. The discussion with WHO
indicated that the IMCI in Bungoma district is in line with the national adaptation
process. It was also acknowledged that although Bungoma district project used generic
materials for training, these have since been revised based on the national version of
training modules. It is expected that the experiences arising from Bungoma district would
be valuable to the national agenda.
All along a member of the Bungoma DHMT was seconded to the IMCI National
Adaptation Task Force. With the completion of the adaptation process he has been
involved in incorporating the necessary changes to the generic materials and district
specific adaptation, particularly local terminology. Bungoma DHMT members who have
been trained as facilitators will be involved in IMCI training activities for other Districts.
61
Although drug vendors have been trained and are motivated to participate in the
prevention and control of malaria their legitimacy in this respect is a source of concern to
health providers who feel responsible for them. The gazettement of SP has rescheduled it
from Part I poison to Part II poison but has not been clear on whether community drug
outlets can stock it or not. Therefore the gazettement is contentious, as it does not allow
extensive availability of SP in community drug outlets such as kiosks, shops and
community pharmacies. The existing policy requires that for one to stock SP, one has to
obtain a part II licence worth Kshs. 1000/= which is prohibitive for small entrepreneurs.
National Malaria Control Programme plans to readdress the issue so that SP can be re￾gazetted to allow for availability of the drugs in the community outlets with ease.
The experiences in Bungoma district so far are already feeding back into policy
discussions at the national level. For example health providers have already noted an
anomaly in the management of children. There is a problem with harmonization of
dosages of SP at this level and this has been brought to the attention of the MOH. The
stakeholders comprising MOH and representatives of pharmaceutical companies are
planning a meeting to address this issue.
A representative of the National Malaria Control Programme attends the BDMI Quarterly
meetings that serve as a venue for continuous BDMI project review and monitoring.
Besides members of the Malaria Control Programme participate in the training of
Bungoma health workers on the new malaria guidelines. Furthermore, the National
Malaria Control Programme has provided guidelines for training of Community Owned
Resource Persons (CORPs) and Bungoma will be the first District to use these guidelines.
With regard to ITNs the MOH is working on an ITNs’ Strategy document that will
address the issue of accessibility and availability in the context of taxation exemptions on
nets and netting materials. A stakeholder meeting for the development of an ITNs’
Strategy Development held in Nairobi on the 14th of June 2000 formed an ITNs’ task
force that would coordinate the activities of research partners and source for the support
of donor partners towards demand creation for bednets.
62
The promotion of bed-nets as a malaria preventive measure is consistent with the
National Plan of Action for Malaria Control in Kenya. This is also reinforced by the
participation of His Excellency the President of the Republic of Kenya in the Africa
Summit on Roll Back malaria in Abuja, Nigeria in April 2000 and expressed
unequivocally his support for the declaration that pledged to
“… reduce or waive taxes and tariffs for mosquito nets and materials,
insecticides, antimalarial and drugs and other recommended goods and services
that are needed for malaria control strategies …”
To this end, it was resolved to
“… initiate sustainable action to strengthen health systems to ensure that by the
year 2005, at least 60% of those at risk of malaria particularly pregnant women
and children under five years of age, benefit from the most suitable combination
of personal and community protective measures such as insecticide treated
mosquito nets…”
On the issue of the large stocks of chloroquine in health facilities, it was clarified that
during this transition period health workers would use and clear the stocks, and that the
MOH is not procuring any more chloroquine (CQ). CQ has not been banned, but
rescheduled and is still useful in some cases. In the absence of SP, health workers can fall
back on CQ.
DFID has supplied 19 million tablets of SP to the MOH and these were expected to be in
the provinces and the districts by the time of the review. Some of these were to be used
for presumptive treatment at ANC. The scaling up of SP implementation is in place.
JIPIEGO under DFID is working in both Busia and Kilifi districts, while the Population
Council with the support of DFID will incorporate the same in a safe motherhood
initiative in both Kakamega and Vihiga districts.
63
 It is evident that a variety of training programmes has been implemented under the
BDMI. This will go along way in upgrading the knowledge and skills of the personnel
involved thus enhancing the capacity-building component of the Bungoma District
Health Team who will then be utilised by others for training and implementation in other
Districts.
Conclusion
BDMI as an integrated child survival project is doing more than IMCI. Although the
programmatic focus of BDMI has been on malaria, the introduction of IMCI as a
mechanism for accelerating its objectives reintroduces a broader child health perspective
to the project. It provides a logical platform for cross- sectional collaboration and
integration of child health strategies in immunisation, nutrition, HIV/AIDS, among others
to occur. Given the point that it includes a substantial component of maternal health, it is
simultaneously addressing the issue of safe motherhood and child survival, which is a key
component in the National Reproductive Health strategy.
Project activities are on course and research findings are available even though these
have not been systematically disseminated. This arises partly from the fact that AMREF
as the co-ordinator of the project has not asserted itself enough to provide technical
leadership and a framework within which the CAs were to operate.
This is exacerbated by a limited oversight role of USAID/ Nairobi. The Mission through
its senior staff has not been proactive in providing programmatic support to the project.
Furthermore communication and co-ordination in general were said to be problematic
This is particularly in reference to internal sharing of reports among the CAs. Whereas it
had been agreed that quarterly reports would be circulated to all CAs to date this has not
happened. Similarly at the November 1999 planning meeting it was agreed that an
electronic database of BDMI documents would be created to which all CAs would
contribute. However only CDC is reported to have undertaken this.
64
 The DHMT raised their lack of management skills. Equipping the DHMT with these
skills and their participation would allow for a sense of ownership as well as
demonstrated partnership and capacity building as stipulated in the project goal and
objectives. In addition it was felt that the roles and responsibilities of the project deputy
manager were not clear and formalised. Both concerns are critical for capacity building.
Bungoma district has just embarked on a range of activities that simultaneously touch on
all the five project objectives. A community-based component is already evolving, and
drawing from experiences of other donors and collaborating agencies, it is likely to
contribute to achieving the project outputs and consequently its goals. This is likely to
result in a BDMI working model that could be replicated nationally with appropriate
modifications through the planned national adaptation of IMCI.
There is potential for community participation particularly in drug supply although the
issue of gazettement will need to be clarified by the MOH.
Patient referral remains problematic and is a major challenge to practice and success of
IMCI. To address the problems presented by patient referral will require harnessing both
facility-based and community-based resources.
The study intervention namely, problem-solving teams may not have been adequately
understood or defined prior to the start of the study as evidenced by the difficult start-up.
An inexperienced trainer was used to introduce the approach and the subsequent need
was to re-train the teams as well as provide substantial 'coaching' or external support to
apply the problem-solving steps. It is acknowledged that 10 senior health workers, 5 of
whom are DHMT members were trained in Quality Assurance concepts and
methodology. Yet from the discussions with the DHMT, the activity seemed to lack
adequate ownership at this level and the expected improvements in IMCI as a result of
this approach were not realised.
65
To monitor trends in malaria morbidity and service utilisation rates, a basic tool is record
keeping at the health facilities. From the workshop presentation on Health Information
System (HIS), follow-up discussions with relevant officers, health providers and
observations at health facilities, it is clear that this component of BDMI is the weakest.
At the time of the review there was an exercise to pilot data collection forms in selected
health facilities and at the community level. However, at the moment these are project
specific and there is a danger that they may not be applicable nationally given that the
MOH headquarters has not been involved in their development for standardisation.
On evaluation and observation, health providers still have difficulty in recognising severe
illness among children. In one facility a pregnant woman was given SPs in the ninth
month. This is a pointer to the need to revisit the training strategy and supervisory
mechanisms that would facilitate retention of skills through practice and assessments.
The results from the study on assessment of microscopic diagnosis of malaria raised the
issue of quality control. This calls for training of laboratory personnel given the
importance of malaria microscopy at the higher referral levels.
Training and supervision are the most expensive aspects of the IMCI. For cost￾effectiveness of the IMCI strategy pre-service training is an option. This is already being
implemented at the Universities of Addis Ababa, Ethiopia and Dares-Salaam, Tanzania.
The review team learned that some of the components are already being taught to
students at the Medical School, University of Nairobi, Kenya. There are plans to extend
to medical training colleges and nursing schools. Medical officers posted to the field and
who are already acquainted with IMCI would be informed supervisors who would steer
child survival activities within the district.
At the beginning of the initiative, there may have been some misgivings at the national
level, principally because the project approaches and activities may not have been very
clear to all parties, but with time this has changed. Recently different MOH personnel
have participated at various levels as part of the project implementation including
66
training, provision of malaria guidelines for training and participation in the project
quarterly meetings. From the discussions with MOH senior staff, it was evident that
BDMI is expected to provide insights into the implementation of IMCI nationally.
Recommendations
Consolidation of the IMCI activities in Bungoma rather than expansion of the same to
other parts of Kenya should be a priority for the project. Expansion could still be
achieved by providing a working BDMI model, which could be replicated nationally,
with appropriate modifications, through the planned national adaptation of IMCI. To this
end, USAID can contribute at the policy level by participating more actively in the
review of the adaptation process and facilitating the dissemination of findings of the
project so far.
USAID senior health staff could play a more active role in representing the achievements
of BDMI at the national level by participating in relevant task forces/technical
committees. This would serve to bridge this pilot USAID-funded activity with national
malaria control, IMCI, RBM and other related strategies currently under development as
well as help leverage the resources of other development partners.
In project management USAID/Kenya will need to strengthen its oversight role such that
if the implementation and monitoring needs of the Mission are not being met by the
current reporting process the issues should be raised promptly with AMREF
management. In the same vein, there is need for greater AMREF/Nairobi senior staff
involvement in the project to provide the needed technical backup to the project manager
in Bungoma. This would ensure that priority areas agreed upon at the planning meetings
are addressed in a timely manner and that the CAs adhere to these requirements.
Additionally, AMREF as the co-ordinator should be more proactive in facilitating
communication between the CAs. This is especially so with regard to distribution of
reports among the CAs to promote awareness of each other's activities and how they fit
together in the overall objectives and goal of the project.
67
USAID and implementing partners should review the current monitoring and evaluation
strategy. This should distinguish between the need for programmatic (activity-level) data
that would allow USAID and project managers to make interim adjustments/corrections
in specific project areas, and the need for more rigorous measures of district-level
performance and impact.
The working relationships of the CAs and accountability to the DHMT were reported to
have greatly improved. However, USAID should consider formalizing the position of the
DHMT deputy manager in the contract agreement in order to facilitate the administration
of the project, particularly in the absence of the AMREF coordinator.
To facilitate acquisition of management skills for DHMT members and health facility in￾charges, the project will need to identify and organise for short-term management courses
that are available locally.
In the second half of the project it will be timely to forge a true linkage between
organisations that work in child survival activities, the Division of Primary Health Care
(DPHC) as well as the BDMI. Presently Siaya district under Care-Kenya offers a
Community-based model for IMCI and may provide insights for replication in Bungoma
district.
There is good information already collected on behavioral and preventive aspects of
health which provides a strong evidence base for undertaking community–based
approaches. Should such an approach be undertaken, they should build on the existing
community structures and activities of CBOs and NGOs working to address the key
68
IMCI household and community behaviors, and forge multi-sectoral linkages across
Ministry departments and among organizations working in child survival.
The need for alternatives to the current modes of supervision should be explored given
that transport and allowances are currently provided by BDMI. The expectation is that
the DHMT will somehow absorb these costs once the project ends, but to sustain the
intensity and frequency of current visits will require some considerable planning and
mobilisation of additional resources, including at community level.
Based on the premise that routine points of intervention work better; building on them
rather than creating new structures is a cost-effective approach to improving maternal and
child health. In this respect, AMREF should harmonise the IEC strategy with
opportunities for intervention principally at the ANC level, while strengthening the
existing points of outlets for ITNs. The original BASICS framework could also be re￾visited to provide guidance in this exercise.
Given that the government fully supports a social marketing strategy, it would be
worthwhile from the onset that this strategy and a community-based distribution strategy
be combined in order to increase access to ITNs. In this respect, Population Services
International (PSI) has had experience with social marketing of bednets in the coastal
region and it would be worthwhile to explore links between AMREF and PSI for social
marketing of nets in Western Kenya. This initiative should be facilitated by USAID.
Community-owned resource persons (CORPS) and drug vendors handle drugs such as
SPs. These drugs are gazetted as part II poison. BDMI should seek clarification from the
National Malaria Control Programme and the Pharmaceutical Board to avert the fear by
CORPS of being prosecuted for handling prohibited drugs.
SPs are not always prescribed for pregnant mothers as providers are not sure whether to
charge or not; given that as a policy ANC services are free. It is therefore important for
69
MOH to ensure that adequate supply of SPs is included in the drug kit for use in
MCH/FP.
Community pharmacy is a cost-effective strategy that fills the gap created by drug
shortage at the health facility level while making drugs more available at the community
level. They are complementary to drug vendors and shopkeepers, ensuring access to
treatment at the community level. This is made possible by the Community Improvement
Fund (CIF). However, the MOH headquarters needs to formalise and regulate CIF within
a legal framework to counter exploitation and abuse. At the moment the MOH at the
district level has no control over the finances and the expenditures of these pharmacies.
Subsequently the team has learnt from DHMT that the Division of health care financing
is in the process of preparing regulatory mechanisms to address the issue but will not
control the funds raised under CIF.
The issue of patient referral requires review at the national level in terms of appropriately
equipping the health facility and especially ensuring that there is a functional ambulance
at the district level. More importantly, community participation should be harnessed to
realise their role in facilitating referral using available local means and resources.
Malaria microscopy is more critical at higher referral levels since most dispensaries do
not have laboratory facilities. But adequate equipment, supervision and quality control as
well as refresher training for the laboratory staff should be provided in those facilities
where microscopy is available.
For improved management of IMCI there is need for repeated exposure to a high
concentration of severe cases. This would then facilitate better practice and allow for
frequent assessment and evaluation as required. This may only be possible in large
hospital settings on a rotational basis, six months after training. This arrangement should
be explored for its feasibility.
70
At the moment only a select group of professional health providers and some DHMT
members have received IMCI training. All Medical Officers and all DHMT members in
Bungoma district should undergo IMCI training to facilitate their role as supervisors and
to ensure appropriate management of children who have been referred from other health
facilities and the community level. IMCI training needs in private NGO facilities should
also be examined as well as pre-service alternatives. Training of CORPs should be
supported by a strategy to ensure that adequate monitoring and supervision are available.
At this point in time basic data will be necessary against which indicators will be tracked
partly to monitor project activities and subsequently measure impact at the end of the
project. At the time of the review, data collection instruments were being piloted in
selected health facilities and at the community level. It will be crucial to liase with the
MOH headquarters at the onset to standardise the instruments for national application.
Again, the team has learnt that this process has been initiated by AMREF.
More importantly, given the fact that there was a prior system in place that has since
collapsed it may be cost-effective to rehabilitate this system through provision of supplies
particularly tally sheets and retraining of data collection staff to re-orient them on the
importance of collecting data. This is crucial for the project in that as it moves into
implementation, indicators for monitoring would have to be tracked on baseline data.
The approach of QAP teams was acknowledged to have value as a management concept,
although the discussions around its practical usage remained diffuse. In light of these
findings, the DHMT, QAP and other CAs should re-evaluate the contribution of this
team-based approach towards the achievement of BDMI objectives, and critically assess
the 'value added' of this activity given the remaining time and resource constraints of the
project and modify implementation as needed.
Given the short time remaining in the project, more consideration should be given to
maximising the use of prior research i.e. implementing the recommendations, before
embarking on new ones. Any future research should perhaps focus on assessing the
71
effectiveness of the interventions currently underway in the true spirit of 'operations
research', although exceptions, such as formative work to explore the problem of
misclassification of severe illness, may be needed.
Dissemination of project activities, including findings from research conducted is yet to
occur. The findings so far have great potential for policy and programmatic change. In
principle the IMCI approach in Bungoma is accepted within the national health policy
and is in line with the national adaptation process. Similarly, linkages to the National
Malaria Control Programme are quite strong. It is expected that the Bungoma experience
would help inform the national agenda for expansion of IMCI and new initiatives such as
RBM. USAID/Kenya could again, effectively broker this. This would be to a broader
audience consisting of researchers in maternal and child health, policy makers, senior
IMCI programme staff, other related health providers especially at district levels and the
donor community working in maternal health and child survival.
72
REFERENCES
1. Bungoma District Initiative (BDI) Five year strategy and first year plan 1997-1998
2. APHIA Mid-term Review "Issue Papers"
3. APHIA Mid-term Review "Annexes"
4. AIDS Population and Health Integrated Assistance Project
USAID/KENYA, Report of the MID-TERM REVIEW
5. National Guideline for Diagnosis Treatment and Prevention of Malaria for Health
Workers,MOH – 1998
6. Bungoma District Malaria Initiative: Promoting the Use of Pesticide Treated Nets
Hezron Ngugi, Project Manager - May 2000
7. Annual Report Bungoma District 1999
8. Plan of Action on the Promotion of Large Scale Use of Insecticide - Treated
materials, Malaria Control Programme - Kenya 1998
9. Bungoma District Malaria Initiative: Summary Report on Operational Research
Findings and Recommendation 1998-1999
10. IMCI Orientation Meeting at KCCTI Mbagathi, 23-24 November 1998. Report of the
Proceedings. Division of Primary Health Care
11. Bungoma District Malaria Initiative: Annual Technical Project Progress Report
January to December 1998
73
 12. Report on Research on Community Distribution outlets for Drugs supply (current
and potential) Conducted by DHMT and supported by - BDMI Project. Bungoma District
August 1998
13. "Rapid Assessment for District Based Malaria Prevention During pregnancy:
Evaluating the Problem and the Opportunities for Intervention - Bungoma District,
Kenya"
14. Therapeutic Efficacy of Anti-malarial Drugs for Uncomplicated Falciparum Malaria
in Areas with Intense Transmission. World Health Organization,
Division of Control of Tropical Diseases
15. Quality Assurance Project: TRIP REPORT: Kenya October 29 - November 7, 1998
by Paula Tavrow PhD
16. HMIS Designers Workshop for Bungoma District
Held at Parkvilla Hotel, Webuye from 14th to 18th Dec 1999 and Debriefing from 27th
to 28th January 2000
17. Bungoma Malaria Initiative Programme: Assessment of Non-Technical Barriers to
Supervision at District, Facility and Community Levels
18. Knowledge, Attitude and Practice on Malaria with special Reference to Insecticide -
Treated mosquito Nets in Bungoma District, KENYA. A study report prepared for the
Development of Implementation Plan for Bungoma BDMI
19. Assessment of Community Distribution and Financing Systems for Insecticide
Treated Mosquito Nets in Bungoma District: A study report prepared for the
Development of Implementation Plan for Bungoma BDMI. AMREF 1999
74
20. Potential Insecticide - Treatment Mosquito Net Delivery System in Bungoma District,
KENYA. AMREF/DHMT. 25th November 1999