Using Togaf for Building a National Implementation Strategy for
E-Health Services and Technologies in Burundi
Frank Verbeke
1
, Marc Nyssen
1
, Sandrine Kaze
2
and Etienne Mugisho
3
1
Department of Biostatistics and Medical Informatics, Vrije Universiteit Brussel, Brussels, Belgium
2
Burundi Health Informatics Association, Bujumbura, Burundi
3
Belgian Technical Cooperation, Bujumbura, Burundi
Keywords: e-Health enterprise architecture, TOGAF, Health information systems, Burundi
Abstract: In order to better align existing and future ICT implementations in the health domain with the strategic options
defined by the National Plan for Health Development, the Ministry of Health (MoH) of Burundi initiated in
2014 the development of a national e-health enterprise architecture based on the TOGAF methodology. A
first part of the development cycle consisted of a detailed analysis of regulatory documents and strategic plans
related to the Burundian health system. In a second part, semi-structured interviews were organized with a
representative sample of relevant MoH health structures. The study demonstrated the donor driven unequal
distribution of hardware equipment over health administration components and health facilities. Internet
connectivity remains problematic and few health oriented business applications found their way to the
Burundian health system. Paper based instruments remain predominant in Burundi’s health administration.
The study also identified a series of problems introduced by the uncoordinated development of health ICT in
Burundi such as the lack of standardization, data security risks, varying data quality, inadequate ICT
infrastructures, an unregulated e-health sector and insufficient human capacity. The results confirm the
challenging situation of the Burundian health information system but they also expose a number of bright
spots that provide hope for the future: a political will to reclaim MoH leadership in the health information
management domain, the readiness to develop e-health education and training programs and the opportunity
to capitalize the experiences with DHIS2 deployment, results based financing monitoring and hospital
information management systems implementation.
1 INTRODUCTION
In 2005, the Ministry of Public Health and Fight
against Aids (MoH) of Burundi has developed a
National Health Policy covering the period 2005 to
2015. This policy was later translated by the MoH and
its technical and financial partners into a series of
objectives and results in the National Plan for Health
Development 2011-2015. Amongst the objectives
were the reinforcement of the National Health
Information System and the restoration of the MoH
leadership in the field of health information
management. Therefore, a number of priority actions
have been identified:
 The development of an e-health strategic plan
for strengthening the national health
information system
 The development of an integrated and
competitive health information management
system
 The development of effective tools for planning,
monitoring and evaluation
 Increasing the availability of ICT tools
(hardware, networks and software) at all levels
of the Burundian health system
 The promotion of data driven research activities
in the health sector
98
Verbeke F., Nyssen M., Kaze S. and Mugisho E.
Using Togaf for Building a National Implementation Strategy for E-Health Services and Technologies in Burundi.
DOI: 10.5220/0005890100980103
In Proceedings of the Fourth International Conference on Telecommunications and Remote Sensing (ICTRS 2015), pages 98-103
ISBN: 978-989-758-152-6
Copyright
c
 2015 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

Integrating e-health in the national health policy
yielded from the beginning enthusiasm from the
donor community and in the course of the past
decade, a growing number of ICT tools have found
their way to the Burundian health sector. But most of
these tools have been introduced for supporting
projects lead by NGOs and foreign technical and
financial partners whereas hardware and software
solutions almost systematically served well the donor
objectives, but inter-project coordination and
interfacing remained exceptional. Several successful
e-health tools remained hidden in silo-projects and
only produced a fraction of their potential benefits.
Without corrective action, the Burundian health
sector threatens to evolve towards a cacophony of
divergent non-integrated health informatics
implementations. In order to cope with this threat, the
MoH initiated in 2014, with financial backup of the
Belgian Technical Cooperation, the development of a
national e-health enterprise architecture based on The
Open Group Architecture Framework (TOGAF). In a
first step, an initial analysis of human resources,
business processes, hardware, software,
communication and networking infrastructure related
to health information management, had to be
established. This study describes the objectives,
methods and findings of this analysis.
2 MATERIALS AND METHODS
The main objective of the study was to provide a
reliable estimation of the existing human and material
resources and issues related to health information
management in Burundi. The research hypothesis was
that an industrial framework like TOGAF could be
used for this, even in the challenging environment of
one of the poorest countries in the world. If
successful, the study results were to become the first
step in a complete e-health enterprise architecture
development cycle according to the TOGAF
methodology, and therefore needed to provide data
for the development of 4 essential sub-architectures:
 Business architecture: what are the MoH
business needs in terms of health information
management?
 Application architecture: which health
information management applications have
already been implemented in the field and to
what extent do they address the business needs?
 Data architecture: what data is needed and
collected today by the MoH and what is the
quality of it?
 Technology architecture: what are the
necessary technologies (software, hardware, and
networking) and which ones are used today in
the health domain in Burundi?
A first part of the study consisted of a detailed
analysis of regulatory documents and strategic plans
related to the Burundian health system.
In a second part, field visits and semi-structured
interviews were organized with a sample of relevant
structures of the MoH. A standardized study-specific
interview guide was developed and systematically
used by the interviewers.
3 RESULTS
3.1 Mission analysis and field visits
The study of regulatory documents and strategic plans
took place in October and November 2014. After that,
a series of field visits and interviews have been
organized with 39 relevant MoH and -related
structures in the Bujumbura province (the permanent
secretary and all MoH directorates, major health
programs, donor agencies, NGOs, public and private
health facilities and educational institutions). In the
period from November to December 2014, the e-
health architecture development team also visited 5
other provinces (Muramvya, Gitega, Ruyigi, Kirundo
and Ngozi), covering 5 provincial health offices, 5
health district administrations and 12 hospitals. In
total, management staff of more than 15% of the MoH
structures have been questioned about the mission,
the mandate and the vision of their organization, their
objectives and the way their work is organized. After
that, a detailed analysis was made of health
information management related human resources,
ICT solutions and non-ICT (paper based) instruments
at their disposal and procedures used for exchanging
health information with other (MoH or non-MoH)
organizations. Finally, an analysis was performed of
health information management problems, expected
benefits and potential threats of health ICT for each
component of the MoH.
3.2 Hardware
The study showed that computer hardware has most
often been supplied to the MoH by donor-driven
intervention programs. There is no organization-wide
management of computer equipment and distribution
of hardware over the different MoH directorates,
provincial- or district administrations and hospitals is
Using Togaf for Building a National Implementation Strategy for E-Health Services and Technologies in Burundi
99

very heterogeneous: some structures which are
supported by several donors are over-equipped,
others remain without any computer hardware at all.
Under impetus of recent national and provincial
policies and international hype, a growing number of
health centers in Burundi started buying computer
hardware with their own funds, unfortunately without
having a clear idea of how to integrate such tools in
their existing activities.
Generally speaking, hardware specifications are
low standard: desktop PCs with Windows XP and
Windows 7 operating systems, of which a large
number have limited functionality due to computer
virus infections (no budget is available for keeping
antivirus software databases up to date and many of
the PCs have no access to internet for performing
these updates anyway). PCs are almost always
accompanied by an uninterruptible power supply
(UPS) but due to the lack of battery maintenance, the
protection offered by these UPSs is minimal.
Many of the executive health staff make use of
laptop computers which in about half of the cases are
their personal privately owned equipment.
Printers are rarely shared in a network and toner
or ink cartridge supply is problematic due to the
unavailability of toner cartridges on the Burundian
market or the lack of budget for operational costs.
Electronic files and documents are commonly
transferred between computers using USB memory
sticks, which constitute an infamous source of virus
infections.
3.3 Networks
Most of the central MoH structures in the Bujumbura
region have access to a wired or Wi-Fi based LAN.
Many times, these networks are only connected to the
internet by grace of donor funding, which is always
limited in time (and sometimes also in data volume).
Few larger structures (central MoH site, reference
hospitals) have been connected to a national optical
fiber network offering reasonable internet
connectivity. However, for most of the small and
medium-sized health facilities, broadband internet
prices remain prohibitive and bandwidth offered by
local ISPs in Bujumbura is poor and unstable
although considerable improvement has been seen in
the past few years.
Installation of internet connections is hardly
coordinated, with some structures sometimes
accumulating several (poorly performing) parallel
connections on the same site: 4 different wired
internet connections have been identified at the site of
the national blood transfusion site, in addition to the
numerous individual 3G-USB modems already
offered by several donor programs. In spite of the
inadequate internet bandwidth, most central level
MoH structures still state that an internet connection
has become indispensable for their daily activities.
Away from the national and provincial capitals,
the situation is worse: wired internet connections are
unavailable and performance of 2G and 3G wireless
data networks is unpredictable. Some donor agencies
(such as EU) have equipped MoH structures with
VSAT connections which provide stable and reliable
bandwidth but come with high operational costs.
Many of these satellite internet connexions remain
unavailable part of the time due to inappropriate use
(downloading movies or audio) consuming all of the
monthly foreseen VSAT credit in only a few days.
3.4 Software
Almost all of the end user computers run Microsoft
Windows operating systems completed with
Microsoft Office applications, with the exception of a
number of desktop and server computers running
Linux Mint or Ubuntu at the directorate of the
national health information system.
Although health specific software
implementations remain rare, a clear tendency
towards web-based business applications is being
noted, often based on Linux/Apache, MySQL
databases and PHP or Java development:
 The MOH started in 2014 pilot implementations
of the DHIS2 data warehouse as a replacement
for the outdated MS Access based GESIS health
data collection solution.
 iHRIS human resource information system
deployment also started end 2014 with the first
implementation pilots scheduled early 2015.
 Hospital information system (HIS)
implementations remain rare (less than 10% of
the hospitals), with all of the health facilities in
our study sample running OpenClinic GA. The
majority of the HIS solutions are concentrated in
third level reference health facilities.
 OpenRBF has been implemented for monitoring
of results based financing (RBF) programs at the
central and provincial levels.
 Joomla and Drupal are the most popular
solutions for website content development.
Some successful m-Health applications (the
RapidSMS based KIRA Mama project and SIDA-
info) provide promising results today.
Fourth International Conference on Telecommunications and Remote Sensing
100

Epi-Info and SPSS are the leading statistics
software solutions. General and analytical accounting
systems are used by several health sector structures of
Burundi: Asyst and QuickSoft (local development),
SAGE Saari, Popsy, and Banana were found in about
half of the interviewed health facilities while Tompro
was recently introduced for project-oriented
accounting at the central MoH level.
3.5 Paper based instruments
The vast majority of the provincial and health district
administrations are using ICT-tools for reporting
health data to the central level (GESIS), but a number
of hospitals and almost all health centers still rely on
paper based instruments for routine data collection.
Information is written down in registers by
administrative clerks and clinical staff and sent on a
monthly basis to the health district administration
(emergency surveillance information is sometimes
reported more quickly using SMS). Health districts
then forward compiled health facility data to the
provincial level, where eventually provincial reports
are sent to the central level in Bujumbura.
A minimum of 25 paper registers must be kept by
all health centers and around 75 registers are in use in
an average district hospital. Additionally, donors and
health intervention programs claim parallel and
redundant reporting from the health facilities and
district administrations they support, all of which
represents an impressive administrative overhead.
Paper based instruments are also predominant for
health record keeping in most (90%) of the hospitals.
All of them are facing health information quality
management issues.
3.6 Health information management
problems detected
Over the past 10 years, the existing health sector ICT
landscape of Burundi grew organically, with most of
the project-oriented solutions being provided by
donors and health programs. This happened in an
uncoordinated way, leading to:
 Lack of standardization: health information
representation is hardly standardized and few
international classifications or coding systems
are in use (with the exception of some of the
DHIS2 and HIS modules using ICD-10).
 Data availability risks: many databases are
hosted in donor countries outside Burundi, with
true data accessibility risks for the MoH. Also,
many MoH agents use personal computer
equipment without appropriate backup
procedures or anti-virus protection.
 Data protection risks: data access rights are not
being organized in layers according to the role
people fulfil in the health administration; most
often one has full access to all of the information
or no access at all.
 Varying data quality: multiple issues explain the
poor quality of data collected in the field. There
is (1) the lack of intrinsic motivation of MoH
staff that don’t produce data for their own
purpose; (2) the important administrative burden
caused by redundant health data collection
processes; (3) many MoH agents don’t have the
necessary qualifications for producing reliable
data; (4) the absence of personal consequences
linked to the production of erroneous
information; (5) donor funding for the collection
of project specific health data at the same time
compromising the global and systemic
collection of routine data for which no financial
incentives exist (RBF).
 Varying data promptness: the lack of reliable
(electronic) communication instruments delays
the transmission of health information between
different levels of the health system.
 Lack of data completeness: data is sometimes
considered a factor of power and the lack of
perceived personal interest in information
sharing may hinder the effective, complete and
systematic exchange of health sector data.
 Defective and insufficient computer equipment:
a number of MoH structures have no access to
appropriate ICT hardware and due to the lack of
maintenance procedures, many of the existing
equipment has become defective. Computer
virus infections also constitute a major problem
for the MoH administration.
 Inadequate ICT infrastructure: today, access to
stable electric power is out of reach for many of
the MoH structures, even in the larger cities.
UPSs have been provided with most of the
computers, but their defective batteries often
don’t provide any protection against power
failures (sometimes power failures can last for
several days, which heavily compromises the
reliability of electronics in every day’s work).
Affordable broadband internet is unavailable for
most of the MoH components. Donor project-
funded internet connectivity is always limited in
time and does rarely bring a sustainable solution.
 Unregulated e-health market: although e-health
solutions are being considered “medical
devices” by WHO, no standards or regulations
Using Togaf for Building a National Implementation Strategy for E-Health Services and Technologies in Burundi
101

have been put in place for introducing ICT-tools
in Burundi’s health system. E-Health solutions
deployment therefore escapes today from any
health authority control.
 Lack of health applications: most of the software
solutions deployed in the health sector are
generic office applications, statistical analysis
applications or aggregate data reporting
instruments. Few health application
implementations such as hospital-, laboratory-
or pharmacy information systems have found
their way to Burundi’s health system.
 Insufficient human capacity: on the one hand,
qualified staff who are capable of effectively
using ICT-tools in their work environment are
missing in most of the MoH structures. On the
other hand, there is a plethora of unmotivated
and underqualified staff occupying positions in
the MoH administration preventing young and
better qualified workers from being recruited.
Additionally, health-ICT related training and
education opportunities are not aligned to the
needs expressed by the different directorates and
health facilities.
 Organizational problems: the organizational
structure of the MoH reflects in no way the
important transversal role of ICT in today’s
healthcare. The statute of ICT professionals at
the MoH is far from attractive; they are
considered an administrative burden rather than
a valuable asset of the organization.
 Ineffective dissemination of information: the
absence of a reliable communication network
limits the dissemination of regulations, practice
guidelines and policies from the central MoH
level to the peripheral structures.
4 CONCLUSIONS
The TOGAF methodology, after applying some
simplifications, offered the appropriate instruments to
quantitatively and qualitatively describe the status of
health ICT tools deployment in the health sector of a
low-resource country like Burundi. The output of the
study was later used as a starting point for the further
development of an e-Health Enterprise Architecture
for Burundi’s MoH, which has been officially
validated on July 29
th
, 2015.
The study results more or less confirmed the
challenging situation of the Burundian health
information system, but they also exposed a number
of bright spots that provide hope for the future:
 There is a political will to reclaim MoH
leadership in the health information
management domain by enforcing compliance
with international consensus and standards for
future e-health initiatives, putting the MoH in a
regulator/gatekeeper position.
 The human resource deficit in health
informatics is huge and many of the country’s
education institutions will have to collaborate
on national and international levels to provide
necessary ICT training programs. Burundian
academic institutions and the donor community
seem to be willing to invest in this.
 DHIS2 implementation got substantial support
from the government and donor agencies.
Extensive training programs have started in
December 2014 and a lot of enthusiasm exists
to make the implementation of a flexible
national health data warehouse a reality.
 Hospital information management systems
implementation has been convincingly
successful in several hospitals and provides
clear evidence for the feasibility of HIS
implementation in Burundi.
An important challenge remains to capitalize the
experiences from the few success stories and to
integrate these in a new coordinated, well adapted and
appropriately funded e-health strategy for the country
in the next 5 to 10 years. According to the architecture
vision developed in this study, such a strategy should
account for:
 The creation of a national MoH datacenter in
Bujumbura that centralizes shared databases and
applications and provides a professional
infrastructure with stable electricity, access
control, data backup and redundancy.
 The development of a multi-technology (optical
fiber, 3G and VSAT) VPN-based health care
intranet connecting central, provincial and
district level structures.
 The implementation of shared generic
applications for the public health sector:
accounting software, workflow management, a
unique central website, a virtual library, a
geographic information system and an MoH
owned mail server (preventing loss of valuable
information when staff using gmail.com of
yahoo.fr accounts leave the organization)
 The implementation/strengthening of a series of
health specific business applications such as
DHIS2, iHRIS, OpenRBF, OpenClinic GA HIS,
Fourth International Conference on Telecommunications and Remote Sensing
102

LMIS and a series of health resource registries
(a facility registry being one of them)
 The implementation of tablet and smartphone
based patient oriented health data collection
tools in health centers and at the community
level (KIRA Mama and SIDA-Info)
 The implementation of an SMS-to-IP gateway
enabling health facilities that have only access to
plain GSM and SMS connectivity to participate
in the country’s electronic data collection
mechanisms.
 The development of 3 health informatics
teaching programs to cope with the important
human capacity building needs: (1) a Master in
Health Informatics program in collaboration
with universities from neighbouring countries,
(2) a specialization program in applied health
informatics for health professionals and (3) the
creation of a biomedical technician bachelor
program.
 The creation of an autonomous health
informatics directorate at the MoH with
departments in charge of (1) standardization and
regulation, (2) health informatics infrastructure
management (datacenter and intranet), (3) health
informatics education and promotion and (4)
helpdesk and support functions.
REFERENCES
Plan National de Développement Sanitaire II 2011-2015,
Ministry of Public Health and Fight against Aids,
Bujumbura, Burundi
The Open Group Architecture Framework (TOGAF),
http://www.opengroup.org/togaf/
Politique Nationale de la Santé 2005-2015, September
2004, Ministry of Public Health and Fight against Aids,
Bujumbura, Burundi
Politique Nationale de Développement des Technologies de
l’Information et de la Communication du Burundi
2010-2025, November 2010, Council of Ministers,
Bujumbura, Burundi
Plan Stratégique du Système National d’Information
Sanitaire 2011-2015, Ministry of Public Health and
Fight against Aids, Bujumbura, Burundi
Plan Stratégique National de Développement des
Ressources Humaines pour la Santé 2010-2014,
Ministry of Public Health and Fight against Aids,
Bujumbura, Burundi
Human resources for health: overcoming the crisis. WHO;
2004. Available from: http://www.who.int/hrh
documents/JLi_hrh_report.pdf
Kruse S-E. Health Information Systems in Africa:
Resources, Indicators, Data Management,
Dissemination and Use. Technical papers. Algiers:
Centre for Health and Social Development (HeSo);
2008. Available from: http://www.tropika.net/
specials/algiers2008/technical-reviews/paper-8-en.pdf
Kamadjeu RM, Tapang EM, Moluh RN. Designing and
implementing an electronic health record system in
primary care practice in sub-Saharan Africa: a case
study from Cameroon. Informatics in primary care.
2005;2005(13):179–86.
Verbeke F, Karara G, Nyssen M, Evaluating the Impact of
ICT-tools on Health Care Delivery in Sub-Saharan
Hospitals, Stud Health Technol Inform, August 22,
2013
KIRA Mama project, Unicef, 2014, https://
unicefstories.files.wordpress.com/2014/05/projet-kira-
mama-pdf-final-1.pdf
SIDA Info, SWAA Burundi, http://www.sida-info-
service.org
OpenClinic GA, http://sourceforge.net/projects/open-clinic
Using Togaf for Building a National Implementation Strategy for E-Health Services and Technologies in Burundi
103