e-Health Rwanda Case Study

Document Sample
e-Health Rwanda Case Study Powered By Docstoc
					e-Health Rwanda Case Study
July 1, 2008

Hamish Frasier
Maria A. May
Rohit Wanchoo
Table of Contents
Executive Summary ............................................................................................................ 2
Introduction – Country Overview and Demographics...................................................... 10
Public Health Informatics ................................................................................................. 14
Interoperability.................................................................................................................. 17
Access to Information ....................................................................................................... 20
eHealth Capacity Building................................................................................................ 23
Electronic Health Records ................................................................................................ 26
Mobile eHealth.................................................................................................................. 28
Unlocking the Market for eHealth .................................................................................... 29
National eHealth Policies.................................................................................................. 31
Conclusion ........................................................................................................................ 32

Executive Summary

Health information technology in Rwanda is a quickly growing industry with many
committed stakeholders, including the Government of Rwanda (GoR), several non-
governmental organizations (NGOs), and private sector partners. Particularly in the areas
of electronic health records and national reporting system, Rwanda has been a pioneer in
national initiatives to integrate technology into its expanding health care system. Where
the private market has not emerged, the GoR has provided significant support to help
these fledging industries. As a result, GoR is significantly involved in all major
initiatives and emerging technologies.

There are six significant entities in health information technology in Rwanda to date.
These programs are:
   • OpenMRS – An open-source Medical Records System that tracks patient-level
   • TracPlus and TRACnet – Monthly monitoring of infectious diseases including
       HIV/AIDS, TB, and Malaria
   • CAMERWA – Drug and medical supply management system
   • Telemedicine – Information and communication technology (ICT) used to deliver
       health and healthcare services, information and education to geographically
       separate parties
   • Health Management Information Systems (HMIS) – systems that integrate data
       collection processing, reporting, and use of the information for programmatic
   • E-Learning – use of ICT in instruction of A2-level nurses for promotion to A1

In addition to GoR, Partners in Health and Voxiva, Inc. have played significant roles in
leadership and implementation of HIT in Electronic Health Records and in inventory
management and pharmacy, respectively. The telemedicine, HMIS, and e-learning
programs are in nascent stages, but initial discussions or plans for their implementation
have already begun, often as collaborations between GoR, academic institutions and

These programs will be discussed in additional detail in the following sections:

Introduction. The introduction will give a brief description of the demographics,
politics, economic growth, and current governance of the country. Rwanda’s history is
essential for contextualizing the government’s priorities and attitude, as well as existing
infrastructural strengths and weaknesses that affect ease of implementing HIT.

Medical Overview. This section will provide a general overview of the current medical
state of Rwanda, including human capacity statistics and health-related indicators. This
section will also describe the various Government, Non-Governmental, and Private actors

participating in the health and health-related spaces in Rwanda to illustrate the nature of
relationships and organic collaborations.

Programs. TRACnet, OpenMRS, National Health Information System

Public Health Informatics. This section will describe GoR initiatives to evaluate
population health, monitor health trends, and create a responsive surveillance system. In
particular this section will describe the TRACnet program, and the building of the
National Health Information System.

Electronic Health Records. This section will describe OpenMRS (Open Medical
Records System), which is used by Partners in Health and Columbia University’s
Millennium Village Project (MVP) and has been endorsed by GoR for national rollout.

Mobile e-Health. This section will explore how mobile technology is used to improve
health delivery in Rwanda. Most of the work to date has focused on gathering clinic
level information on infectious diseases through mobile phones using Voxiva’s TRACnet

Interoperability. This section will describe the degree to which the various programs
(e.g. TRACnet, OpenMRS, and National Health Information System) integrate and/or can
communicate with one another. Our initial findings are that the programs have been
developed in isolation from one another, and there is little current interoperability to date.
Some initiatives are now in place to create such links.

Capacity. e-Learning,

Access to Information. This section will describe the ability of patients, providers and
researchers to readily access accurate healthcare information. Patients have almost no
ability to access health information, while providers have varying levels of ability.
Partners-in-Health sites have the ability to access patient level information through their
electronic medical record (EMR) system, OpenMRS. Other clinics have access to the
history of the self-reported aggregated clinic level information as maintained by

e-Health Capacity Building. This section will describe the efforts underway to (a) use
technology to train health providers in standard practice, and (b) to train medical
providers in the various e-Health initiatives. Though to date, few formal training
programs have existed, a program to provide a rigorous 18-month practical program to
nurses will be launched later this year.

Sectoral Responses.

National e-Health Policies. This section will describe the Government’s role in
supporting or creating the initiatives in e-Health to date. It will also suggest policies that
will be required for future success in e-Health. The initiatives that are receiving the most

attention from GoR are the electronic medical records initiative (EMR), the TRACnet
program, and, HMIS, e-Learning. The Government has preliminary plans for work in the
overhaul of the Health Management Information System (HMIS).

Unlocking the Market for e-Health. This section will describe the private sector’s role
in supporting or creating the initiatives in e-Health to date. To date, the largest private
sector partner is Voxiva, through the TRACnet development. Private sector partners,
including MTN, Rwanda-Tel, and alfasoft, are helping to migrate the HMIS from an
Access database to a SQL server.

Conclusion This section will reiterate the current state of health information technology
in Rwanda and evaluate obstacles and remaining challenges. The government’s role as
the main driver in HIT provides an opportunity to streamline efforts, though as yet there
has been little activity in the private sector. Continued economic growth and investment
in health systems may open new doors for the use of technology.


OpenMRS. The most widely used patient-management system in Rwanda, this open-
source medical record system has been endorsed by the government for national rollout.

TRACnet. The government collects monthly data from facilities providing ART to HIV
patients. This section describes the program, its use of mobile technology in collecting
data, and the partnerships involved in developing the system.

Introduction – Country Overview and Demographics


Rwanda is a country of approximately 9.5 million inhabitants and an annual population
growth rate of 2.8%.1 The country is just over 26,000 square kilometers (about the size
of Maryland) and has the highest population density in Africa2. The country is one of the
poorest in the world, with a GDP per capita of approximately $202 ($1,600 adjusted for
purchasing power parity [PPP]), and nominal GDP is $2 billion ($13.7. billion adjusted
for PPP). 3

Rwanda’s poverty is also reflected in its social indicators. The country ranks 161 out of
177 countries on the United Nations Human Development Index4 Sixty percent of the
population lives under $1 per day, sixty-six percent is under the age of twenty years, and
83% is rural dwelling. 5 Most rural Rwandans are smallholder subsistence farmers,
producing bananas, maize, beans, and sweet potatoes as staple crops; droughts are
common and can greatly affect crops, as irrigation systems are limited. There are two
main ethnic groups in Rwanda. Hutus are the largest group and compose 84% of the
population. The Tutsis are the second group and compose 15% of the population. The
remaining Rwandans are Twa.6


Rwanda was colonized by Belgium in 1916 and remained under its rule until 1962. In
the years before Rwanda’s independence , the Party of Hutu Emancipation Movement
came to power and thousands of Tutsis fled to neighboring countries. In 1973, under the
leadership of Maj. Gen. Juvenal Habyarimana, the military took control of the country
and abolished all political activity. In 1990, after several rounds of single party elections,
Tutsi exiles formed the Rwandan Patriotic Front and invaded from Uganda7. While a
ceasefire was negotiated in 1992, the tension of the ethnic divide culminated in the

  EIU Rwanda report, May 2008
  Economist Intelligence Unit (2006) Rwanda: Country Profile 2006
  UN Human Development Report Database accessed on June 13, 2008
  Food and Agriculture Organization (2006) The State of Food Insecurity in the World
2006: Eradicating World Hunger -- Taking Stock Ten Years After the World Food
  CIA. Rwanda. The World Factbook 2008 [Available from:
  United States Department of State (2008), Background Note: Rwanda,
genocide of 1994 after Habyarimana’s plane was shot down. It is estimated that 800,000
Rwandans were killed and millions fled to neighboring countries.8

Current Government

Since the genocide, GoR has made remarkable steps to bring the country together. Under
the leadership of President Paul Kagame, who in 2003 became the first democratically
elected President, Rwanda has become safer, and had more economic development than
any period since the country’s independence in 1962.9 Addressing the harms caused by
the genocide, including the need for reconciliation within society, care for victims and
orphans, rendering justice and promoting equality are also high priorities of the

Governance indicators suggest vast improvement in Rwanda since 1996. According to
the World Bank report, Governance Matters 2007, the most significant improvements
have been made in the categories of Political Stability, Government Effectiveness, Rule
of Law, and Control of Corruption. Moreover, the Government of Rwanda’s
“Effectiveness” is now ranked amongst the highest in sub-Saharan Africa and even above
other large countries including Kenya and Uganda.10 On “Control of Corruption”,
Rwanda ranks behind only Botswana and South Africa.


By 1996, the vast majority of Rwandans who had previously fled the country—Hutus and
Tutsis alike—had resettled in Rwanda. The economy has since recovered rapidly; from
1995 to 2003, Rwanda was sub-Saharan Africa’s second fastest-growing economy. Since
then, growth has been equally startling: the Gross Domestic Product (GDP) per capita
grew 76% from 2002 to 2007. 11

Services accounts for over 53% of GDP. Agriculture accounts for roughly 30%, and
industry/manufacturing provides the balance. The largest exports are coffee, cassiterite,
tin, and coltan, a metallic ore.

Inbound Foreign Direct Investment (FDI) has expanded from USD 3 million in 2003 to
USD 8 million in 2005.12 Official Development Assistance (ODA) has increased in

  Rwanda: how the genocide happened. April 1, 2004. BBC News.
  IMF World Economic Outlook Database, April 2008
   http://info.worldbank.org/governance/wgi2007/mc_chart.asp, accessed on June 12,
   Rwanda EIU Report, March 2008
   OECD, African Economic Outlook 2007, Table 10

dramatically, from USD 321 million to 576 million, since 2003 as the Government has
proven its capacity for governance and efficacy.13

Health and Health Care Overview


The health of Rwanda’s population reflects its high levels of poverty and still developing
health care infrastructure. National statistics look similar to many other countries in the
region. Infant mortality currently is approximately 85 deaths per 1,000 lives births.14
Forty-five percent of children under the age of five meet height-for-age criteria for
chronic malnutrition.15 HIV/AIDS prevalence reached 3% in 2005 and was considered a
generalized epidemic.16 Malaria is the leading cause of morbidity and mortality for both
adults and children in Rwanda.

Health System

As of the end of 2005, Rwanda had about 366 health centers, 33 district hospitals, and 5
referral hospitals.17 Rwanda’s health system includes public, private and traditional
health systems, which are supported by GoR, non-governmental organizations (NGOs)
and civil society.

Mutuelles de santé (mutual health insurance), a community-based health insurance
program, was launched by the GorR in 2006. Subscribers pay approximately USD 2 a
year, with GoR and Global Fund covering the enrollment fee for the poor. Over 83% of
the population has enrolled in the program. Evaluation of the program is preliminary but
data from the past two years suggests that utilization rates of health services increases
with enrollment. 18 Some have commented that the current structure leaves the mutuelle
(community organization) with financial liability and involves burdensome
administrative work. As the Mutuelles approach universal coverage, some issues
diminish as the consequences of adverse selection become less pronounced.

Technical Overview

Government Leadership and Participation
GoR participates in HIT initiatives in diverse ways. The major bodies that lead and
promote HIT are:

   OECD, African Economic Outlook 2007, Table 11
   ORC Macro. Rwanda Demographic and Health Survey 2005. July 2006.
   Rwanda Human Resources Assessment for HIV/AIDS Services Scale-Up (2005).
Please see appendix for a more detailed explanation of each type of facility.
   Community-based Health Insurance in Rwanda: from Case Studies to National Policy.

Ministry of Health. The Ministry of Health is the major Government body responsible
for all health related activities. In particular, the Center for Treatment and Research on
HIV/AIDS, Malaria, Tuberculosis, and other epidemics (TRACPlus) has been a promoter
and user of e-health products, including TRACnet, a system that will be described in
detail in the Public Health Informatics section below.

Ministry of Science, Technology, and Scientific Research. The Ministry of Science,
Technology and Scientific Research sits within the President’s Office and oversees the
Information Communications Technology (ICT) Unit. The ICT Unit has direct influence
over the development of and the expansion of internet connectivity and therefore has the
most direct influence over e-Health initiatives. It oversees the ICT initiatives taking place
in other Ministries and Government related-bodies including the Rwanda Information
Technology Authority, which is described below.19

Rwanda Information Technology Authority (RITA). RITA is housed in the President’s
Office that oversees science, technology and scientific research. Its charge is to design
policies for technology used by GoR and create an IT governance framework and
standards for various ICT strategies. To accomplish these goals, RITA provides ICT
support both in terms of training and consultancy services. It also is the body that is
responsible for standard setting and compatibility in ICT applications throughout the
public sector, and for facilitating the private sector in the development of ICT in Rwanda.

Current State of Connectivity

GoR is known to be one of the governments most committed to ICT development in sub-
Saharan Africa. That having been said, a lot of ICT work remains for e-Health to take off.
Indeed, against all countries in Sub-Saharan Africa, Rwanda had fewer telephone
subscribers, fewer internet users, and fewer personal computers (per 100 people). 20

Rwanda’s connectivity fate is partially explained by the fact that it is not connected to the
international fiber-optic backbone. This implies that Rwanda’s broadband efforts are all
through satellite. Despite this limitation, there has been some initial success in
telemedicine, which will be described below.

  World Bank, Rwanda ICT at a Glance, accessed at
devdata.worldbank.org/ict/rwa_ict.pdf on June 10, 2008

Public Health Informatics


There are two major programs attempting to document and track HIV/AIDS information
in Rwanda. The first, Open Medical Record System (OpenMRS), tracks patient-level
data for clinics.* The second, TRACnet, is a software package used by TracPlus though
which clinics report aggregate data on a monthly or biweekly basis into a repository
managed by the Central Government. Additionally, GoR has plans for an online blood
bank monitoring program.

Overview of Programs:

TRACnet. Please refer to the accompanying profile of TRACnet.

Health MIS

The TRACnet program informs a broader Health Management Information System
(HMIS). Implemented by the MoH in 1997, HMIS has historically been largely paper
based at the health center levels, with a combination of paper and electronic reporting at
the district and central level. The broad goals of the information system are to “integrate
data collection, processing, reporting, and use of the information necessary for improving
health service effectiveness and efficiency through better management at all levels of
health services”21 GoR is currently overhauling the HMIS to make it better-designed to
“assist in the management and planning of health programmes, as opposed to the delivery
of care.”22 A comprehensive HMIS will aggregate patient-level data into clinic and
district level information to enable it to inform drug and medical supplies procurement,
disease surveillance, and programmatic funding.

Current Status of Programs

The Current Management Information System (Système d’Informations Sanitaires, SIS)
is managed on a Microsoft Access database, separate from TRACnet, and has important
limitations. Most significantly, it is not designed to easily pass information/data from
one program area to another or pass it from one system to another. This results in limited
data entries, duplication, loss of critical information, higher costs, and missed
opportunities for timely intervention and prevention.23

  For more information on the OpenMRS section, please refer to the section on Electronic
Health Records.
   WHO, Developing Health Management Information Systems: A Practical Guide for
Developing Countries, p3
   ibid, p3
   United States Agency for International Development, Rwanda HMIS Assessment
Report, p23-26; interview with Dr. Richard Gakuba

Secondly, reports are currently not submitted from the district level to the central MoH. A
USAID Assessment Team in April 2006 found that as of the end of April only 37% of
health centers and 34.3% of hospitals have submitted their SIS Monthly reports for
March 2006.24 As noted above, there is also insufficient capacity at the MoH to
aggregate, analyze, or report national level SIS data. Weekly epidemiological reports also
ceased to exist when the Epidemiological Unit at the MoH closed in January 2006.

A National Health Information System Project is currently in the planning phase, which
will enhance the country’s ability to perform disease surveillance and enhance public
health protection services. The first part of this transition is the migration of data to a
Structured Query Language (SQL)-based relational database server by the end of
summer. Such a migration should make room for a more sophisticated electronic system
to support HMIS efforts.

Planned Interventions

The National Health Information System Project will also include a revamping of
indicators so that nursing staff enters data for only one set of forms, which then
electronically inform the relevant databases. According to the GoR, by the end of 2008
these systems should be coordinated. It intends to create a system that provides the
following capacity:

     Service providers will be able to utilize the same data for the same cases, without
     having to duplicate their collection work. Authorized service providers will be able
     to share clinical information among themselves, about a patient. Public health issues
     and communicable diseases will be quickly identified and managed to mitigate risk
     to the general public. A health surveillance system will provide information on risk
     factors, treatment, health service utilization and outcomes to assist in the
     development and evaluation of policies and programs aimed at the prevention and
     control of infectious and communicable diseases. Aggregated data will be easily
     accessed by the Ministry of Health for reports on statistics and trends to support
     health planning and decision making.25

A simultaneous scale-up in the number of data managers is also planned. Currently, the
head of nursing completes most forms among their other duties. A data manager at each
hospital will not only ease the burden on the nursing staff, but will allow for faster
transitions between systems. Government of Rwanda estimates the costs for this human
resource expansion at USD 100,000 per month. In anticipation of the national roll out,
the International Development Research Centre (IDRC) is funding a formal training
program to be set up by PIH and the GoR for Java programming and medical information
systems with a focus on OpenMRS. The training will create a cadre of Rwandan workers

  ibid, p9
  Government of Rwanda, Ministry of Health, National E-Health Strategic Plan 2007-

who possess high level IT expertise to implement and extend OpenMRS in Rwanda. The
first students are expected to start training in fall of 2008.

Blood bank

The blood bank initiative would serve essentially the same function as TRACnet for the
blood supply. This would allow central agencies to monitor the blood supply by type of
blood and region. At the central level, the project is still in its design phases, but at the
local level, three health centers are linking their blood inventories together. These
activities have no been followed by GoR and results of their efforts could not be located.


There are two major interventions currently underway in Rwanda that pertain to Public
Health Informatics. The first is the use of TRACnet, and the second is the overhaul of
the Health Management Information System. TRACnet currently interoperates with a
drug procurement system (Camerwa), which is in the midst of a significant overhaul, but
not with other MIS-related systems (e.g., medical records, surveillance, and laboratory
tests). Although the drug procurement system provides critical support of the health care
system, the government's limited ability to hold facilities accountable for their
distribution of drugs reduces its effectiveness. Increasing integration of information
systems and strengthening collection of patient-level data would increase the accuracy
and resolution of the database; and could work in tandem with other efforts to reduce
redundancy in forms and costs.



This section will focus on interoperability – the ability of the various Rwandan e-Health
systems to communicate with one another. The major programs in Rwanda that
interoperability applies to are: Open Medical Record System, a patient-level system used
in some clinics; TRACnet, the national aggregate facility-level reporting system for HIV
treatment; Camerwa, a surveillance program; and the Health Management Information
System (HMIS) broadly.. TRACnet interoperates with Camerwa, which is currently
undergoing a significant overhaul, and serves as a proxy for a health surveillance system
However, the current form of the HMIS is not integrated with TRACnet, Camerwa, or
OpenMRS. Another program that needs to interoperate with clinical level systems is the
Access database used in come facilities to support the Mutuelles insurance program.
These programs are described in greater detail below.

Background on the Various Systems:

OpenMRS. Please see accompanying profile.

OpenMRS is currently being extended to interoperate with a variety of district and
national reporting systems. Rwandan programmers trained by PIH have created a module
that allows data from OpenMRS to be automatically submitted to TRACnet. The system
shows the data to be submitted for the monthly report to the user, and then, with the click
of a button, submits the information via the Internet or by mobile phone with guidance
from an automated voice response to TRACnet. In addition, data can now be exported
from OpenMRS in a WHO-supported format called IXF. This format can then used by
district reporting systems and there is agreement to incorporate it in TracNET.

TRACnet. Please see the accompanying profile on TRACnet for more information.

GoR is currently modifying the TRACnet system to accept IXF.
A public-private partnership between Voxiva, PEPFAR, Motorola, MTN, Accenture and
GSM are in the planning process of integrating the existing systems using Tracnet
technology. GOR has approved the project and they have secured a planning grant from
the Gates foundation.

Camerwa. Camerwa is a semi-independent non-profit organization administered by the
GoR which manages the country’s drug inventory. TRACnet aggregates clinic-level
information on the number of HIV and TB patients seen in a given period. Camerwa
uses this information to manage the HIV/AIDS drug and medical supply purchases for
the country. Camerwa is also used to procure and distribute all other essential drugs and
medical consumables.

Health Management Information System (HMIS). Currently in implementation stages,
the broad goals of the HMIS are to “integrate data collection, processing, reporting, and
use of the information necessary for improving health service effectiveness and
efficiency through better management at all levels of health services.26” The
management information system will “assist in the management and planning of health
programmes, as opposed to delivery of care.” A full Management Information System
would aggregate patient-level data into clinic- and district-level information. This would
inform drug and medical supplies procurement, disease surveillance, programmatic
funding, etc.

Rwanda Mutuelles. The Mutuelles program is a community-based health insurance
program coordinated by GoR. It currently utilizes a Microsoft Access database to store
information about patient health utilization for payment purposes in a central repository.
Linking the system with a patient management system, such as OpenMRS, would enable
GoR to evaluate how the insurance scheme improved health and identify remaining
challenges. Additionally, this linkage would also help improve the efficiency of clinics.

Additional improvements are intended to be made in the Mutuelles interface as well.
Currently, clinics complete paper forms that are submitted for data entry at the central
level. While providers and patients have no electronic record of transactions, Mututelles
has created a database for their clients and plans to launch it in the fall of 2008. GoR, as
part of its broader movement towards electronic records, is trying to integrate ICT with
its insurance programs.

Integration and Interoperability

Although there is progress on many e-Health fronts, one of the most important gaps has
been the integration of various systems (OpenMRS, TRACnet, Mutuelles, and HMIS) to
consolidate information. Currently, the various data are entered separately and exist in
virtual silos. Nurses are required to fill out different and mutually exclusive forms,
exacerbating strains on a system facing a scarcity of health professionals.

GoR plans to overhaul the HMIS. The system-level improvement is expected to resolve
many of these issues, but much more work is necessary for successful integration. The
Government has recently endorsed a roll-out of OpenMRS to all hospitals in Rwanda (it
is currently only in use at all PIH and MVP sites). The OpenMRS program would then
feed into the TRACnet system. That would also have the benefit of maintaining and
improving the data in the Camerwa drug procurement program. Drug procurement
appears to be functioning well at a national level, but integration with patient- and clinic-
level information should allow faster access from the drugs to the clinics and more
accountability around their distribution. Moreover, the HMIS system will have the
capacity to gather data regularly for improved surveillance.

  World Health Organisation, Developing Health Management Information Systems: A
Practical Guide for Developing Countries.

GoR has two goals in attempting to overhaul the HMIS system and improve
interoperability. First, it believes that interoperability should create a large reduction in
the paperwork and administrative time spent by nurses. Nurses are the backbone of the
health delivery system, and the current time they spend reporting to multiple sources is
widely considered burdensome. A well functioning system will draw information out of
patient-level records automatically, allowing nurses to spend more time with patients.
Second, an integrated system can provide higher quality data as it is based on collection
at the patient level.

GoR has also implemented performance-based financing at the district level and collects
a set of clinical and managerial performance metrics that influence staff salary levels.
Currently, it uses dedicated software to evaluate performance. Increased collection of
clinical and operational data would allow GoR to identify drivers of performance, refine
its incentive structure, and improve quality within and across sites.

There is little information on migration patterns of patients, but most information seems
to be communicated in English and/or French. Language translation did not surface as a
major issue during interviews.


There is limited interoperability between the various e-Health systems currently used in
Rwanda - OpenMRS, TRACnet, Camerwa, a surveillance program, the Mutuelles
insurance program and the HMIS broadly. TRACnet is integrated with Camerwa, and
serves as a proxy for a health surveillance system.

However, the current form of the HMIS is not integrated with TRACnet, Camerwa, or
OpenMRS. A planned HMIS overhaul is intended to fix many of these issues, but an
absence of funding has curbed efforts in this arena. Ideally, GoR would like the HMIS to
be universally interoperable with current systems and require interoperability with HMIS
as a criterion for future systems and modules.

Access to Information


This section will describe the ability of medical providers, government officials and
researchers to access necessary information in a timely manner. Many of the focus of
Rwandan e-Health initiatives have been on increasing health system quality per se, and
have not yet focused on either provider or patient uptake of these services. Moreover,
limitations to information access are due in part to the lack of Rwanda’s connectivity to
fiber optic. Information access will likely follow general connectivity access.

Access by subpopulation

Medical Providers. Anecdotal evidence suggests that doctors are universally computer
literate but poor electricity supply and slow-at-best Internet connectivity limit the amount
that doctors can rely on ICT for clinical help. At the time of this writing, government
officials estimated that just over half of the district hospitals have Internet access, and all
of the research hospitals have internet access. These access figures may be misleading
however, as computers are likely to be set up in “management” offices and clinicians and
nurses may have limited-at-best access to this equipment.

Currently, even simple “Google” searches are not regularly used because of slow
connections and/or lack of regular access. Some providers use professional list-serves to
communicate with each other, but again these appear to be of limited value in time-
sensitive matters. GoR plans to have established LAN connectivity in all district-level
facilities by 2010.

Clinics managed by Partners in Health (PIH) or Columbia University’s “Millennium
Villages” Project (MVP) can individually access patient-level records using OpenMRS
software27. At present, health facilities and hospitals managed by other groups are not
using the OpenMRS software.* For a more detailed discussion on OpenMRS, please see
the Section titled, “Electronic Medical Records”. The King Faisal hospital is also
implementing health information systems.

PIH recently developed modules to be used with OpenMRS to assist the providers. One
is the “patient dashboard,” which provides key information on a patient to the provider in
an easy-to-understand format. Particularly for patients that have coming to the facility
for some time, navigating through paper charts can be challenging.

Health facilities. Health center information (at the clinic level) is available to respective
district hospitals through the TRACnet program, which is the software used by the

  Currently available at six out of the seven sites managed by PIH
 For more information on OpenMRS, please refer to the section titled “Electronic
Medical Records.”

Government of Rwanda to collect and store clinic-level HIV/AIDS information at the
central level. These hospitals receive a copy of their own clinic’s statistics and data for
health centers in their jurisdiction. The aggregated nature of the data also makes it
impossible for physicians to discern the quality of their performance relative to others’.
Though reporting rates for TRACnet are consistently high and data is collected at
frequent intervals, little research has been conducted verifying the accuracy of the
information collected.

PIH has begun to collect clinical indicators across sites to compare outcomes and
distributes them among the sites. It also reviews longitudinal data aggregated from all
the sites to evaluate the programmatic performance over time. Their installation of
OpenMRS also allows them to create alert lists; that is, lists of patients that have low
CD4 counts or other important markers and are not on ARV treatment. Such patients can
then be sought out and treated.

In addition, OpenMRS has enabled PIH to better understand its quality of care not only
clinically, but also programmatically. Recently they have begun to explore different
measures of quality, such as how long after a patient gets diagnosed with HIV they their
first appointment, and the percentage of patients missing follow-up appointments.

Government and researchers have access to information from the TRACnet application.
Additionally both groups have information provided by the Health Management
Information System (HMIS), but this data is very limited and can be out of date. The
planned HMIS overhaul may reduce duplicative data entry, and improve the quality and
frequency of the reported data.

Patients. Adult literacy rates are approximately at 65%, presenting an upper bound for
the medical literacy in the country and suggesting a need to communicate through non-
written media. 28 According to first-hand accounts, radio has been used for this purpose,
but GoR has largely centered its efforts in other areas.

Planned Programs

Government of Rwanda has other initiatives in the planning stages as well, including
setting up and an electronic laboratory that will allow clinicians to access a patient’s lab
test history electronically and to determine whether a patient’s lab test have already been
ordered by another physician. No date has been set for when this system will be piloted.

The GoR sanctioned plan to roll out the OpenMRS EMR system will provide better
access to patient data in clinics. This should help to support monitoring of patient care
and outcomes as well as better reporting.


     World Bank EdStats April 2008

Access to information at the clinic and national level exists though with little granularity.
In some clinics, the implementation of electronic medical record systems enables
clinicians to access comprehensive patient data efficiently. The use of the EMR by
physicians is unclear. Little effort has been made to distribute information on HIV to the
general population, and the literacy rate indicates that in order to penetrate effectively,
visual or auditory media (including TV and radio spots) will be necessary. While
researchers are given access to high-level data, the lack of intra-clinical data collection
limits its research potential, particularly around quality of care.

E-Health Capacity Building

This section will describe the telemedicine and e-Learning efforts in Rwanda. This report
will define telemedicine as “the use of communications and information technology to
deliver health and healthcare services, information and education, where the participants
are geographically separated."29 In resource-poor environments, telemedicine can help
mitigate the absence of specialist doctors by connecting other health providers with
specialists located elsewhere, or be used to train future doctors in rural areas. In general,
telemedicine “facilitates clinical consultation, continuing professional education, health
promotion, and healthcare management.”

Current State of Telemedicine

There are two major telemedicine efforts already under way. The first of these efforts is
to connect district hospitals to referral hospitals to store and forward asynchronous
telemedicine; most immediately, teleradiology. X-rays taken in one facility, for example,
the referral hospital, can be sent to anotherr facility, such as a district hospital, to be read
by a radiologist. Importantly, computed radiography machines for this effort are already
installed in two clinics and the procurement process for an additional two machines has
already begun, with funding from the World Bank.

The second effort is to create a universal platform for biomedical imaging using the
Digital Imaging and Communications in Medicine (DiCom) platform in Rwanda. These
efforts are being facilitated through the World Bank. As of the end of May, the tender
phase of the formal Request for Proposal had been completed.

Current State of e-Learning

Although a formal e-Learning program has yet to begin in Rwanda, there have been
isolated instances of e-Learning being used in limited settings. In the university setting,
many medically relevant lectures have been broadcast over a local area network that
connects all three referral hospitals. Moreover, there have been telecasts of open-heart
surgeries and various meetings. These efforts and others along the same vein are
instrumental in increasing learning in health practices.

GoR is also attempting to start a formal program in e-Learning for nurses. Rwanda,
along with many other countries, suffers from a shortage of A1-nurses, or nurses who
have completed secondary school and have two additional years of nurse training. A2
nurses, who make up the bulk of the health work force, have two years of secondary
education and two years of nursing education. The e-Learning program is currently
focused on additional training for A2 nurses to graduate to A1 level. The training

  Government of Rwanda, Ministry of Health, National E-Health Strategic Plan 2007-

program will be funded by the International Development Research Centre (IDRC) and
will use web-based training and exams. In its first stage, the program hopes to take 10
nurses through a one-year, practical curriculum. The program is slated to start in late

Telemedicine in Rwanda will be primarily provided through the use of communications
technologies alongside dedicated telemedicine software and diagnostic medical
equipments. The GoR hopes to connect the telemedicine efforts and the Electronic
Health Records (EHR). For example, in the case of diagnostic imaging, GoR envisions
the transmission of live echocardiogram and ultrasound images for interpretation, along
with store-and-forward transfer of digital images for review and assessment.

Although there are health programs that use ICT, there are no formal training programs in
e-Health technology per se. There are informal training programs that take place in
specific systems (e.g., TRACnet training), but there are few programs that focus
exclusively on capacity building. The International Development Research Centre
(IDRC) have now funded a formal training program to be set up by PIH and the GoR for
Java programming and medical information systems with a focus on OpenMRS. This will
create the high level IT capacity to implement and extend OpenMRS in Rwanda. The
first students are expected to start training in fall of 2008.

Kigali Health Institute (KHI) – the major teaching hospital in Rwanda – has been
teaching basic computing skills as part of its nurse training. These training programs are
fairly new, so only new graduates are likely to have acquired these skills.

Planned Activities

Government of Rwanda Goal for e-Learning

By 2009, GoR plans to for all 30 district hospitals will have Internet access, to improve
capacity for Telemedicine.30 As part of infrastructure for e-health, VPNs will be set up in
order to allow high bandwidth interconnectivity. Its plan for rolling out this activity also
involves procuring new equipment and coordinating the implementation efforts. King
Faisal Hospital will be the Telemedicine Hub, and all hospitals will have a functional
telemedicine platform installed by 2011.


There are two initiatives to raise the capacity of health workers in Rwanda. The first is
telemedicine, and the second is e-Learning. Telemedicine efforts have been growing at
the ground level, and the central government would like to build upon these efforts. GoR
is also attempting to develop an e-Learning program that would facilitate nurse training
and hopes to leverage its e-Health initiatives to build capacity broadly within its medical
professionals. Currently, capacity building in e-Health initiatives occurs primarily at an


application-specific level; that is to say that users are trained in TRACnet, OpenMRS,
HMIS separately.

The IDRC have now funded a formal training program to be set up by PIH and the GoR
for Java programming and medical information systems with a focus on OpenMRS. This
will create the high level IT capacity to implement and extend OpenMRS in Rwanda. The
first students are expected to start training in fall of 2008.

Electronic Medical Records

Electronic Medical Records (EMR) in Rwanda – Overview

Electronic Medical Records allow clinics to collect, track, and analyze patient level data
over time and potentially across health facilities. There are currently two primary
systems in use - Open Medical Record System (OpenMRS), and Fuchia – which are
detailed below.

OpenMRS. Please see accompanying profile.


Fuchia is a partially free software tool, originally developed by Medicines Sans
Frontieres (MSF) and Epicentre, based on Delphi and Microsoft Access. The product is
partially free because the software to run it is free and distributed by MSF, but in order to
use it, you must have Microsoft Access, which in turn requires Microsoft Windows –
both of which incur cost. The installation of Fuchia at the TRAC clinic allows collection
of intake and follow-up data for HIV patients, but all changes to the code have to be
made by the original developers so the forms are fixed in the system and cannot be edited

Currently, the TRAC Clinic in Kigali has an installation of Fuchia with approximately
6,000 patients registered. However, TRAC slowly suspended usage of the system in
2007, due to lack of data accuracy and insufficient reporting and clinical tools. It was
discovered that the paper registers contained more accurate and up-to-date information on
the same set of patients.

Fuchia is also used by LUX Development at two locations – Rwamagana Hospital and
Kimironko Health Center. Currently, it is estimated that the Fuchia database used by
LUX contains data for 3,000 patients.

The installation of Fuchia at the TRAC clinic allows collection of intake and follow-up
data for HIV patients, but the forms are fixed in the system and cannot be edited. The
system also can produce printed patient summaries for clinicians and has a few built-in
reporting forms for monitoring enrollment over time and some other high-level
indicators. The system is not web-based nor is it built enterprise-style, so each computer
utilized for data collection requires its own program installation and generates a separate

Other EMR systems

Some sites are using homemade tools to generate a database of patient-level data, often
built in Microsoft Excel or Microsoft Access. It is likely that Excel and Access are used
often because they are widely available, moderately priced and well known. While these

forms provide clinics with basic information, these programs systems are hard to use for
the complex, longitudinal, and multi-domain information that clinics hope to include, and
often lack the security features necessary to protect patient confidentiality. They also do
not readily facilitate data analysis from a quality or programmatic perspective and rarely
streamline reporting except for relatively simple data. Access has however been used in
many countries very successfully for smaller projects. The number of such systems and
their capabilities is difficult to estimate and no formal research has been conducted by

Plans for EMR scale up
In 2006, a letter from Dr. Innocent Nyaruhirira, the Minister of HIV/AIDS at the time,
indicated that the Rwandan government had selected OpenMRS as the system that would
be used universally across Rwanda for patient-level collection and monitoring. Since
then, the government has been working to secure funds and expertise for such a venture,
and is beginning deployment at the TRAC Clinic–the main research clinic in the country–
in central Kigali.

In anticipation of the national roll out, the International Development Research Centre
(IDRC) is funding a formal training program to be set up by PIH and the GoR for Java
programming and medical information systems with a focus on OpenMRS. This will
create the high level IT capacity to implement and extend OpenMRS in Rwanda. An
initial round of students trained in the spring of 2008, and the first round of funded
students are expected to start training in fall of 2008.


TRAC has implemented a fairly robust aggregate-level data collection infrastructure for
HIV services in Rwanda. Though TRACnet’s ability to allow its users access high
quality patient-level data are limited, TRAC now can watch HIV trends over time.
Patient-level data gives clinics and the government a much more granular view of trends
and activity, and the government’s recent endorsement of OpenMRS demonstrates their
desire to increase the quality of data collected in clinics nationwide. While OpenMRS
provides many useful features and flexibility for clinics to customize the software to their
needs, in its current state it requires a considerable level of technical expertise to be
installed and maintained so that it can be utilized in multiple sites with many users. Basic
implementations of OpenMRS on one Windows server for specific tasks requires a lot
less support but requires all interactions with clinic staff to mediated by paper forms and
reports. Creating efficient communication between TRACnet and OpenMRS will be
essential for maximizing the value of both systems. In addition, creative thought about
how to ensure quality of data collection and efficiency in entry will be important
components of the rollout’s success.

Mobile eHealth

While mobile technology holds great potential for health care in Rwanda, at present,
there is still little use of phones in Rwanda. Only 3% of the population had a mobile
phone in 2006, though over 75% of the population lives in an area with cell phone
coverage.31 It is worth noting that in 2000, only 0.5% of the population had a mobile
phone, so the per capita phone rate grew seven-fold over, which may indicate that mobile
phones will soon become more commonplace in Rwanda.


TRACnet has been able to utilize mobile phones as a tool for data submission quite
successfully; virtually all facilities providing ART to HIV patients submit monthly
reports using mobile phones. This process is outlined in detail in the accompanying
profile on TRACnet.


The Millennium Villages Project (MVP) has creating a program that allows community
health works to enter the vital statistics of patients into the medical record system
(OpenMRS) using a mobile phone. MVP is currently piloting the program in its site in
Uganda and plans to roll it out in Rwanda in late 2008.

Mobile phone usage in health thus far has been limited thus far but may be an untapped
resource. In Rwanda, where internet connectivity is extremely limited, mobile phones
have served as a tool in aggregating facility-level data, in the case of TRACnet, and will
soon be used to enter patient-level data in the MVP sites.

     World Bank, Rwanda ICT at a Glance. devdata.worldbank.org/ict/rwa_ict.pdf

Unlocking the Market for eHealth
The private sector has played a partnering role in Rwanda’s eHealth initiatives. The main
coordinators of efforts appear to be the Ministry of Heath and the Rwanda Information
Technology Authority, which then support ongoing private sector efforts. A foreign
private player, Voxiva, Inc. has been awarded USAID funding contracts for the
development of TRACnet, a system to track aggregate (clinic-level) information for
monitoring purposes.32 MTN Rwanda – the largest local cellular provider – has also
donated airtime for this monitoring effort.


Voxiva is the largest player in the e-Health space to date. GoR offers a tax holiday for
private companies investing in e-Health and it is remains one of the best countries in Sub-
Saharan Africa for setting up a business and enforcing contracts. However, Rwanda is
average for the region in terms of dealing with licenses and registering property.
According to the World Bank, the country requires substantial work in access to credit
markets, protecting investors, trading across borders, and closing a business.

Related to these indicators is the small size of Rwanda. As a country of approximately
nine million people, any country-specific “market opportunity” would be small relative to
other countries.33

Telecommunications has had mixed results in Rwanda. In late 2007, the GoR privatized
its national telephone company and sold its 80% stake in Rwanda-Tel to Lap Green
Networks, a Libyan Company.34 MTN Rwanda – the local branch of cellular provider
MTN Group (a multi-national company), has had large success however.

In the e-Health sector particularly, the anecdotal evidence is mixed as well. The private
sector is involved in installing the information system, and a private local company is
designing the new HMIS database.35 However, the opportunities that would have been
the most promising for the private sector are financed through public monies. Voxiva’s
TRACnet was financed through USAID/PEPFAR and better fits the description of a
public-private partnership than a pure private sector opportunity.

At the time of this writing, the Rwanda Information Technology Authority (RITA) is
unaware of other efforts to include the private sector per se.36

   TRACnet is referenced in detail in the accompanying profile.
   See appendixes for World Bank Doing Business ranking and summary of Doing
Business Profile for Rwanda.
   The HMIS database is described in detail in the Public Health Informatics section.
   Interview with Mr. Patrick Nyirishema

Future possibilities

GoR has committed to a national rollout of OpenMRS. The implementation and
management of the system will require significant training and sustained investment.
In anticipation of the national roll out, the IDRC are funding a formal training program to
be set up by Partners in Health and the GoR for Java programming and medical
information systems with a focus on OpenMRS. The training will create a cadre of
Rwandan workers who possess high level IT expertise to implement and extend
OpenMRS in Rwanda. Once trained, these individuals will possess valuable
programming skills that could enable them to create private companies specializing in
OpenMRS management, or apply their skills to other pieces of health information


The private sector has participated in some of the largest e-Health efforts in Rwanda
(TRACnet, HMIS overhaul). These efforts have largely been at the request of the main
player in the domain – GoR. Private innovation in this space has been somewhat limited
due to small potential returns. The country’s small size, and large amounts of poverty
create a difficult environment to create profits. Private efforts in the space are likely to
be dominated by private-public partnerships, or multilateral/bilateral government
financing (e.g., PEPFAR, USAID).

National e-Health Policies

Government of Rwanda (GoR) has made explicit, proactive effort in promoting both e-
Health and Information and Communication Technology (ICT). The e-Health initiatives
in large measure, began in 2007, and have multiple parts. In particular, GoR has
organized around the following efforts:

      •   OpenMRS – An open-source medical record system that tracks patient-level data
      •   TracPlus and TRACnet – Monthly monitoring of infectious diseases including
          HIV/AIDS, TB, and Malaria
      •   CAMERWA – A drug and medical supply management system
      •   Telemedicine – ICT used to deliver health and healthcare services, information
          and education to geographically separate parties
      •   Health Management Information Systems – systems that integrate data collection
          processing, reporting, and use of the information for programmatic decision-
      •   E-Learning – use of ICT in instruction of A2-level nurses for promotion to A1

What is most striking about the evolution of the country’s e-Health program is the type of
role that GoR plays. Rather than the simply regulating other participant’s efforts, GoR
acts as an active facilitator – it fundraises for new initiatives, it is active in program
design, and it assists with implementation. This focus on active participation has perhaps
allowed GoR to focus on programmatic policy (i.e., policies on which programs to
implement), rather than regulatory or standard setting policies. Indeed, even the national
e-Health documents do not bear the “policy” name, but rather are described as “strategic

The success of many of the e-Health initiatives will be on the back of increased Internet
connectivity. Although the Government has promoted ICT, in an absolute sense a lot
more work remains to be done. In particular, ICT indicators are still quite low for Sub-
Sahara African countries (as of 2006). Only 3% of the population has a mobile phone,
compared to 13% for the region.37

Healthcare providers suggest that the two largest programs are TRACnet, a national data
aggregation system focused on ARV therapy enrollment and outcomes, and OpenMRS,
an open source medical record system implemented in several sites that the government
is planning to roll out nationally.38 The decision facing GoR at the moment is how best
to use OpenMRS as the primary data collection mechanism for TRACnet. OpenMRS can
provide richer data for TRAC to analyze and to use to inform programmatic decisions,

     World Bank, ICT at a Glance.
     See accompanying profiles on OpenMRS and TRACnet for more information.

but is logistically and technically much more demanding. GoR has in principle suggested
that OpenMRS should inform the TRACnet program which can then be used for drug and
supply management (through Camerwa).

GoR has created an internal publication outlining its goals and activities through 2011.
Additionally, the GoR will need to work to develop appropriate patient confidentiality
policies, as there appear to be none in place. Other government initiatives, such as
creating universal standards for patient forms, have great potential to streamline care, in
this case by reducing the number of forms that nurses are required to fill out on a regular
basis allowing them to focus on their clinical duties. These measures and others will go a
long way in creating a more efficient e-Health system. The Government is clearly aware
of these needs, and is making inroads into creating a coordinated set of systems.


GoR’s efforts in e-Health have been mostly strategic and programmatic rather than
regulatory. That is to say that the GoR has a vision of which segments ought to be built
up in the industry and partners with organizations that can effectively create the vision.
Because the GoR is so active in the promotion of ICT and e-Health, its support acts as de
facto policy making. That having been said, some initiatives, such as medical records
systems, are becoming increasingly complicated and require both support and regulation.
To manage the growing systems, GoR will need to develop appropriate patient
confidentiality policies and universal standards for user forms.

The implementation of health information technology faces great challenges in Rwanda.
In the wake of a societal tragedy and in the midst of the realities of poverty and disease,
GoR has many competing priorities and limited resources. Its embrace of information
technology, particularly in the health sector and HIV/AIDS treatment, and willingness to
partner with organizations like Partners in Health, the Clinton Foundation and the
Millenium Villages Project has been pivotal to creating the current landscape. While
many systems, such as TRACnet and OpenMRS, exhibit great shortcomings, they also
demonstrate commitment to an HIT infrastructure and a chance to evaluate systems en
vivo to integrate into future systems and improvements in current ones. Rwanda’s HIT
can simultaneously help it leapfrog into a relatively advanced health care system, but will
also be limited by the same factors that limit the health care system: lack of electricity,
lack of good roads, and limited trained health professionals.

At this time, the government has created a virtual monopsony for health information
technology. Because a great deal of international aid is channeled through the
government and its partners, they control the incentives that exist for innovation and
implementation within Rwanda. Whether in the long run, this strategy is sustainable or
optimal is debatable, but at the moment, the size of the domestic market and the poverty
of the population may deter private investors. For now, the power they weld gives them
the ability to create a unified and interoperable system that might not arise in free
market. OpenMRS, TRACnet, Camerwa, and HMIS all represent necessary parts of
managing a complex health system, but will require significant investment and
restructuring to be truly optimal. The value that they can add to the system as a sum, in
comparison to discrete systems that don't communicate, is profound.

Currently, Rwanda lacks a surveillance system that can identify disease outbreaks and
monitor health trends at a population level. TRACnet, because it receives information on
virtually all ART patients, does provide a rough measure of HIV treatment and patterns
in Rwanda. Unfortunately, it is difficult to validate the quality of the data collected from
the facilities where there is no patient-level data maintained. OpenMRS, a patient
management system currently used mainly by Partners in Health and the Millennium
Village Project, may give TRACnet legs once it is rolled out nationally if it improves the
quality of data collection and maintenance. Furthermore, to maximize the benefit of
these systems, it will be necessary to expand their use beyond HIV/AIDS. However,
designing systems like TracNET in an open fashion facilitates ease of modifications,
allowing the system to be customized for Rwanda’s needs and allow interoperability.

While work remains to be done, the Government has set an ambitious agenda. E-training
programs for nurses and technical training for the OpenMRS rollout will begin later this
year. They are designing a new Health Management Information System that will link
the variety of other systems and have improved capacity. Mutuelles is exploring ways to
create a database with more capacities that could potentially link with patient
management information. With the economic growth trends in Rwanda, the health
infrastructure should continue to strengthen, with new opportunities for mobile and

Internet-based health services increasing as these technologies spread through the


BBC News. Rwanda: how the genocide happened. April 1, 2004.

BBC News. Rwanda genocide “failure” berated. April 5, 2004.

Boned-Ombuena (2007), Development Assistance to Health Information Systems
Strengthening, World Bank, Health, Nutrition and Population Unit.

CIA. Rwanda. The World Factbook 2008 [Available from:

Economist Intelligence Unit (2006), Rwanda Country Report

Economist Intelligence Unit (2008), Rwanda Country Report, May 2008

Food and Agriculture Organization (2006) The State of Food Insecurity in the World
2006: Eradicating World Hunger -- Taking Stock Ten Years After the World Food

Gahigana, Innocent, “Rwanda: Rwandatel Deal Sealed,” The New Times, October 26,

Gahigana, Innocent, “Rwanda: All District Hospitals to Be Connected in 2008,” The New
Times, December 31 2007

Government of Rwanda, Ministry of Infrastructure (2008),
=1129472269&ucat=2&lang=en \

Government of Rwanda, Ministry of Infrastructure (2008), Network Organisational Chart
of Ministry of Infrastructure,

Organisation for Economic Cooperation and Development (2007), African Economic
Outlook 2007

Shepard, Donald; Rwiyereka, Angelique K; Beaston-Blaakman Aaron, Community-based
Health Insurance in Rwanda: from Case Studies to National Policy. Rwanda Medical
Journal. In press.

United States Agency for International Development (2005), Rwanda Human Resources
Assessment for HIV/AIDS Services Scale-Up

United Nations (2008), United Nations Human Development Report Database,

United Nations Department of Education and Social Affairs, Division for Sustainable
Development (2008), TRACnet, Rwanda: Fighting Pandemics through Information
Technology, www.un.org/esa/sustdev/publications/africa_casestudies/TRACnet.pdf

United States Agency for International Development (2006), Rwanda HMIS Assessment
Report, May 9 2006.

Voxiva, Inc. website, http://www.voxiva.com/rwanda.asp, accessed June 12, 2008

World Bank (2008). Education Statistics, http://www.worldbank.org/education/edstats

World Bank (2008), Worldwide Governance Indicators (WGI) project,
http://info.worldbank.org/governance/wgi/index.asp; accessed on June 12, 2008

World Bank (2008), Rwanda: ICT at a Glance,

World Health Organisation, Developing Health Management Information Systems: A
Practical Guide for Developing Countries

   I.      Interview List

Dr. Michael Kremer, Director General, TRACPlus- CIDC

Dr. Anita Asiiwame, Director of HIV/AIDS for the STI units, TRACPlus-CIDC

Dr. Richard Gakuba, e-Health Coordinator for the Ministry of Health

Shabani Cishahayo, Interim Head of Surveillance of the Bioinformatics and IT Division,

Jean Baptiste Koama, Voxiva

Patrick Nyirishema, http://inside.pih.org/node/754Deputy Executive Director of the
Rwanda Information Technology Authority (RITA)

Christian Allen, Head of Technology for Partners in Health, Rwanda

Neal Lesh, Chief Technology Officer of D-tree International

Dr. Jean Kagubare, Senior Program Associate, Center for Health Outcomes, Management
Sciences for Health

Jonathan Jackson, Co-Founder and Chief Executive Officer of Dimagi

Linea Rowe, Director of Product Management of the Global Health Delivery Program,
Harvard University

Dr. Andrew Kanter, Director of Health Information Systems/Medical Informatics of the
Millennium Villages Project, Earth Institute at Columbia University

Darius Jazayeri, Lead Programmer for Partners in Health

Dr. Lisa Hirschhorn, Assistant Clinical Professor in the Department of Global Health and
Social Medicine, Harvard Medical School and Senior Clinical Advisor on HIV/AIDS at
John Snow, Inc. Research and Training.

     II.       Appendixes

     Appendix 1: Map of Rwanda and ART health facilities

Graphic from Shabani Cishahayo’s presentation at Rwanda Health Education and
Information Network (RHEIN) Workshop. Kigali, May 29-30, 2008.

Appendix 2: Health Indicators for Rwanda and sub-Saharan Africa

Indicator                                                  Rwanda    Sub Saharan Africa,
                                                           Rural     all (2005) +++
A. Socio-Economic Characteristics
% poor population                                          45%       --
% of households with access to clean water                 43%       56%
Primary school attainment                                  54%       61%
B. Mortality Rates
Under-5 mortality rate 2000-2005 (per 1000)*               116       163

Infant mortality rate 2000-2005 (per 1000)*                                               84            93
C. Reproductive Health
Total fertility rate for 3 years prior to survey (all women 15-49)                        6.1           5.2
Proportion of pregnant women who received any antenatal care at last pregnancy            93%           70%
Proportion of assisted deliveries by trained personnel                                    35%           44%
Proportion of deliveries taken place in health facility                                   24%           --
D. Child Health
Proportion of children WITHOUT full basic vaccine coverage                                26%           --
Proportion of children under 5 sleeping under a treated bednet                            20%           --
Prevalence of moderate and severely underweight children                                  24%           25%
Prevalence of fever**                                                                     26%           --
Prevalence of diarrhea**                                                                  15%           --
E. Utilization of Health Services
% of women who reported lack of money for treatment                                       74%           --
% of women who reported distance to health service as a large barrier to accessing care   43%           --
% of utilization of modern health services for diarrhea                                   14%           --
% of utilization of modern health services for fever                                      23%           --
F: Population Health                                                                      Rwanda, All   Sub-Saharan Africa,
                                                                                                        all (2005)
HIV prevalence (per 100,000)                                                              3133          --
TB prevalence 2006 (per 100,000) +                                                        562           348 (2005)
Malaria cases 2000 (per 1000 population) ++                                               120           --

*direct calculation used: (#died/#born alive 2000-2005)*1000.
** of children who had had diarrhea or fever reported in the weeks prior to interview
+TB prevalence taken from WHO Report 2008, Global Tuberculosis Control
++ Malaria taken from most recent WHO Malaria Country Profile, Rwanda
+++Sub-Saharan Africa data from Health Systems 2020 Country Brief and World Bank
Development Indicators 2

Appendix 3: Ease of Doing Business
             Ease of
              Doing Starting Dealing Employi                                       Trading
             Business  a      with     ng    Registering Getting Protecting Paying Across Enforcing Closing a
 Economy      Rank Business Licenses Workers Property Credit Investors Taxes Borders Contracts Business
Mauritius           1      1       4       7          33       10         2      1        1      12         6
South Africa        2      3       6      14           8        2         1     13       20      13         7
Namibia             3     13       3       4          21        4         9      8       26        1        2
Botswana            4     12      21      10           3        2        16      2       27      17         1
Kenya               5     17       1       8          15        1        12     38       30      19        11
Ghana               6     24      29      27           1       15         3     15        3        4       17
Seychelles          7      2       7      16           4       39         5      5        6        8       34
Swaziland           8     27       2       5          27        4        46      6       28      25         4
Ethiopia            9     15       9      12          29       10        16      3       31       11        8
Nigeria            10      6      40       3          46        7         5     21       22      16        13
Zambia             11      7      34      24          18       10         9      4       36      14        12
Uganda             12     18      15       1          38       39        20     11       24      22         3
Lesotho            13     21      32       9          22       15        30      9       16      18         5
Malawi             14     16      20      13          11        7         9     16       37      27        27
Tanzania           15     10      42      33          37       15        12     20        9        2       20
Gambia             16      9      13       2          23       27        43     44        4        7       23
Cape Verde         17     36      14      29          20        6        20     25        2        6       34

Mozambique       18       20      33      38        19       10         3     14       23        29       26
Sudan            19       10      24      28         2       27        30     12       25        31       34
Gabon            20       31       5      39        30       15        34     18       11        32       24
Comoros          21       29       8      35        10       39        20      7       13        36       34
Madagascar       22        4      28      32        40       46         5     17       15        35       34
Rwanda           23        5      23      15        25       39        43     10       41         3       34
Benin            24       23      22      21        14       15        34     39       14        41       18
Zimbabwe         25       28      44      25         9       10        16     33       43        10       33
Cameroon         26       38      36      23        24       15        16     42       18        44       16

C™te d'Ivoire    27       35      38      20        33       27        34     32       29        23        9
Togo             28       44      31      30        31       27        27     31        5        37       14
Mauritania       29       39      30      22         5       15        30     43       32        15       30
Mali             30       32      17      11        12       27        34     36       38        39       19
Sierra Leone     31        8      35      41        45       15        14     34       17        30       29

Burkina Faso     32       14      41      34        44       15        27     29       44        20       15
Senegal          33       37      10      36        35       27        42     41       21        33       10
S‹o Tomˇ
and Principe     34       21      19      46        28       15        20     37        7        21       34
Guinea           35       41      16      45         6       27        30     30       19         9       34
Guinea           36       40      39      17        32       27        43     40       10        24       22
Angola           37       42      26      43        41        7         5     27       40        46       28
Niger            38       33      37      37         7       27        34     24       39        26       25
Liberia          39       26      45      19        42       27        27     26        8        40       31
Eritrea          40       43      46       6        36       39        14     19       35         5       34
Chad             41       45      12      26        17       27        20     28       34        42       34
Burundi          42       19      43      17        16       45        34     22       42        33       34

Congo, Rep.      43       34      11      40        43       15        34     46       45        38       21
Bissau           44       46      17      44        39       27        20     23       12        28       34
Republic         45       25      25      31        13       15        20     45       46        43       34
Congo, Dem.
Rep.             46       30      27      42        26       39        34     35       33        45       32

Appendix 4: Ease of Doing Business

Starting a Business            Business Procedures (number)      9
                               Duration (days)                   16
                               Cost (% GNI per capita)           171.5
                               Paid in Min. Capital (% of GNI    0.0
                               per capita)
Dealing with Licenses          Procedures (number)               16
                               Duration (days)                   227
                                Cost (% of income per capita)    822.1
Employing Workers              Difficulty of Hiring Index        56
                               Rigidity of Hours Index           40
                               Difficulty of Firing Index        30
                               Rigidity of Employment Index      42
                               Nonwage labor cost (% of          5
                               Firing costs (weeks of wages)     26
Registering Property           Procedures (number)               5
                               Duration (days)                   371
                               Cost (% of property value)        9.4
Getting Credit                 Legal Rights Index                1
                               Credit Information Index          2
                                 Public registry coverage (%     .2
                               Private bureau coverage (%        0.0
Protecting Investors           Disclosure Index                  2
                               Director Liability Index          5
                               Shareholder Suits Index           1
                               Investor Protection Index         2.7
Paying Taxes                   Payments (number)                 34
                               Time (hours)                      168
                               Profit tax (%)                    20.2
                               Labor tax and contributions (%)   5.7
                               Other taxes (%)                   7.9
                               Total tax rate (% profit)         33.8
Trading Across Borders         Documents for export (number)     9
                               Time for export (days)            47
                               Cost to export (US$ per           2975
                               Documents for import (number)     9
                               Time for import (days)            69
                               Cost to import (US$ per           4970
Enforcing Contracts            Procedures (number)               24
                               Duration (days)                   30
                               Cost (% of claim)                 78.7
Closing a Business             Time (years)                      No practice
                               Cost (% of estate)                No practice
                               Recovery rate (cents on the       0.0

World Bank (2008), Doing Business Rwanda Case

Profile: TRACnet
TRACnet is a software used by the GoR to collect and store clinic-level health
information at the central level since 2005.1 The software was developed by Voxiva inc
(a Washington D.C. based software and telecommunications company), but is managed
and used by the Treatment and Research AIDS Center (TRACplus) division of the
Ministry of Health.


TRACplus is the new broader incarnation of the Treatment and Research AIDS Center
(TRAC) division of the Ministry of Health. Originally, the group focused entirely on
AIDS, but has expanded to encapsulate malaria and tuberculosis as part of its broader
mandate, and plans to later include other conditions. It has changed its title to TRACplus
to reflect these changes. Additionally TracPlus is the primary national agency
responsible for Preventing Mother-to-Child Transmission (PMTCT) and Voluntary
Counseling and HIV testing (VCT), Epidemiology Surveillance, and Health ICT/
Information Management. As part of this mandate, TracPlus uses the TracNet software
developed by Voxiva to store facility-level data.

TracNet is a collaboration between both the private and public sectors. MTN and
Rwanda-Tel, the local cell phone carriers, donated network time for facilities to use when
reporting their data. Voxiva Inc., provided ICT support to the project. The United States
Centers for Disease Control and Prevention (CDC) provided the financial and
administrative support through the President's Emergency Plan for AIDS Relief

The breadth of implementation to date is impressive. TRACnet has been deployed in all
94 health facilities offering ART in Rwanda, thus capturing virtually all ART treatment
nationwide.2 In addition, approximately 6,000 individual case records are monitored
using the system.3

According to the UN:

        “TracNet is a dynamic information technology system designed to collect, store,
       retrieve, display and disseminate critical program information, as well as to
       manage drug distribution and patient information related to the care and treatment
       of HIV/AIDS. This system enables practitioners involved in anti-retroviral (ARV)
       treatment programs to submit reports electronically and have timely access to
       vital information. By dialing 3456, a Rwanda toll free number, or logging onto a

  From Voxiva website, with updated information from report, “Development Assistance
to Health Information Systems Strengthening,” World Bank September 2007.
       bilingual website (English and French), health centre staffers can submit or
       receive program results on HIV/AIDS patients as soon as they are processed.
       TRACnet uses solar energy chargeable mobile phones, which can be used in the
       most remote parts of the country.”4

In essence, TracNet provides a direct electronic means of transmitting consolidated data
for programmatic decision making, including national level drug procurement. A process
that once only provided one-way information and took months, has been reduced to
minutes, and can now provide two-way information.

The aggregated data guides healthcare delivery activities at the national level. For
example, TracNet data informs drug procurement at a national level. The TracNet data is
transferred to Camerwa, a Rwanda-based pharmaceutical company, that then keeps stock
of the availability of ARV drugs. TRAC monitors and supervises health facilities that
provide ARV treatment in the country. TRAC also has a team of IT personnel, who have
trained over 200 health care providers in health facilities on how to submit data to
TRACnet, and who also monitor reporting into TRACnet and publish monthly reports.
Data is entered either on a biweekly or monthly basis (depending on the statistic) and
then collected in a national repository.

Data Quality

Some challenges remain in the implementation of TRACnet. Although TracNet’s
coverage is wide, the depth and accuracy of data remain unclear. One evaluation
suggested that there is wide variance in the completeness of data depending on the size of
the facility. All ART sites had data for over 90% of patients at the time of abstraction.
However, 6 month follow-up rates for patients who were alive on ART were 56%, 60%
and 14% for small-, medium, and large-sized ART clinics. Moreover, 12-month CD4
counts were only recorded for 35%, 30% and 25% of small-, medium-, and large- ART
cites respectively. 5

Additionally, because patient-level data is not maintained electronically at most facilities,
cross-checking clinic-level data is difficult and does not appear to occur under normal
circumstances. There has been limited evaluation of data completeness to date but no
comparison of individual records at facilities with data in TracNET. The difficulty of
verification may provide perverse incentives to clinics; since drug supply is based on the
number of patients they report, over-reporting might be rewarded with excess
medications and supplies. Further evaluations of the system are planned.

While this type of surveillance is valuable for capturing a broad understanding of groups
of patients on ART, it does not identify the segment of the population whose HIV status

 TracNet, Rwanda: Fighting Pandemics through Information Technology
 Government of Rwanda (2008) TRAC Report on the Evaluation of Clinical and
Immunologic Outcomes from the National Antiretroviral Treatment Program in Rwanda,
2004 – 2005, p47
is unknown. These data collection efforts offer only a partial picture of the epidemic
unless coupled with population-based survey efforts.


Screen Shot of TRACnet
Architecture of the integration of information systems at TRACplus

Graphic from Shabani Cishahayo’s presentation at the Rwanda Health Education and
Information Network (RHEIN) Workshop. Kigali, May 29-30, 2008.
Profile: OpenMRS
OpenMRS, developed by the OpenMRS collaborative, is a free and open source medical
record system that has been in development since 2004. In Rwanda, it is used by Partners
In Health/Clinton Foundation (PIH) and the United Nations Millennium Village Project
(MVP). PIH currently operates in seven sites (as of May 2008) and their OpenMRS
installation contains data for nearly 7,000 HIV patients, over 4,000 on HIV or TB

Partners in Health

The OpenMRS installation that PIH-Rwanda uses includes many country-specific
features. Specifically, it can generate reports that meet Rwanda reporting requirements,
including submissions to TRACnet, the national aggregation system.* It also contains
templates for quality monitoring and administrative overview reports. Patient
information is available across sites, and the data synchronizes automatically when a
computer goes online (though information is available offline as well). OpenMRS uses
Secure Socket Layer Protocol and role based authentication to ensure confidentiality of
medical data – the combination represents the same industry-grade security used by
banks and other highly-secure institutions around the world. There are data export, report
building, and patient form building tools in the software package as well. It includes
patient information lookup tools, and a variety of visualizations of patient data to help
clinicians quickly assess a patient’s progress over time.

As a result of the flexible nature of OpenMRS and the number of different functions
implemented at PIH sites, installation and maintenance of the system requires substantial
technical expertise. For example, PIH employs two full-time IT technicians to maintain
all harware system, and these technicians also have the part-time responsibility of
troubleshooting EMR hardware issues across the seven sites. Particularly challenging
issues have been stable power (now mainly addressed with solar systems), and lightening.

Use of electronic medical record (EMR) systems like OpenMRS can be very labor
intensive. It allows clinics to collect a great deal of data about their patients, but the data
entry and quality control requires human attention. PIH collects five pages of
information about HIV patients on an intake form and two pages in follow-up visits. To
cope with this inflow of data, PIH employs two data managers, four lab data officers, and
eight patient data officers, for a total staff of 14 people responsible for the collection,
cleaning, reporting, and analysis of data. If the database continues to grow, more staff
will be required to manage the data effectively. Many projects providing HIV treatment
and funded by the PEPFAR program or the Global Fund receive direct financial support
for reporting. These funds usually support the salaries of data entry staff, data managers
and also IT costs. Such funding will be important in speeding the rollout of the EMR in

    For more information on TRACnet, please refer to TRACnet profile.
The implementation of the EMR at PIH sites is one in which clinicians/providers consult
with patients and write down information on paper. This paper-based information is then
given to the data team for entry. Later, the data team produces lists of patients, patient
summaries, reports, and alerts when appropriate. This information is usually printed on
paper and delivered back to a member of the clinical team. PIH has implemented patient
summaries and flow sheets within the software, and is now working to make them
directly accessible during clinical visits by clinicians. They are also working to capture
data on patient follow-up and medication collection directly from clinic staff into the

Millennium Villages

The implementation of OpenMRS at MVP sites is more recent and contains a subset of
functionality to the PIH installation. Neither organization is implementing real-time live
entry during clinical visit for the time being.

MVP uses OpenMRS as part of its health information system in the Bugesera District. It
runs on an Atheon server linking 6 Inveneo ION workstations within the clinic. The
OpenMRS system is being used to capture primary care data (e.g. adult and pediatric visit
forms and pharmacy data) as well as vital statistics (new birth registrations). Unlike the
PIH installation, the system uses a centralized data dictionary that is shared by other
Millennium Village clinics in multiple languages to ensure that the data will be
interoperable between the MV sites and will allow aggregation across linguistic and
geographic boundaries. In Rwanda, French is the most commonly used language, though
MVP plans to translate the antenatal and community health workers forms into

MVP hopes to expand its installation of OpenMRS to include other primary care clinics
in Rwanda through the Access Health Project in Kigali. In addition, MVP is working
with other partners, such as ICAP at Columbia University and Loma Linda University to
include geographical information systems into its data analysis.
Exhibit 1: PIH Patient Dashboard
Exhibit 2: Screenshot of patient information on PIH installation of OpenMRS
Exhibit 3: List of patients and clinical summaries and alerts with scheduled appointments

Shared By: