Methodology and Assessment Process

The assessment engaged key stakeholders from providers, and users to financiers of health
information. As a stepping stone a steering committee was formed for coordination of the
Tanzania HMN Chapter. Stakeholders were oriented with the HMN framework and assessment
tools before they were engaged in the evaluation process. Steering committee members were
earlier on introduced to the framework and tools by facilitators from HMN Headquarters in
Geneva Orientation on HMN framework and the assessment tools aimed at facilitating further
exchange and feedback on current affairs within HIS. Orientation and adaptation of
the tools was done through workshops (whereby participants were as well trained to orient others
on the framework and tools) and outreach visits to stakeholders who could not participate in the
workshops. The following are the key outcomes from the adaptation process (i) the framework
and assessment tools were accepted as being relevant for the country situation (ii) two main areas
of assessment questions were noted and these are the fact and opinion based questions. The
two categories of assessment questions are important in terms of selecting appropriate
stakeholders participate in different assessment areas and (iii) groups were formed to
participate in specific evaluation areas taking into account nature of the assessment questions

Assessment Findings
Resource
Resources for HIS are generally in place though not adequate (see above). Although there is a
reasonable environment with regard to infrastructure and legislation (covering vital registration,
notifiable diseases and confidentiality), the performance is not adequate in terms of policy and
planning (43%), financing, human resources and institutions to contribute into HIS processes
(49%). Enforcement of legislation is an issue of major concern. On vital statistics for instance
there is inadequate enforcement especially in rural areas due to several reasons including
lack of registration offices and community awareness. Lack of an overall HMIS policy and
HMIS strategic plan is also an important area of limitation. There exist separate strategic plans
for the specific sub systems which lack the required coordinative framework. Regarding human

resource there are limitations in terms of capacity in the core health information sciences,
inadequate skill mix at the central HMIS administrative unit, lack of dedicated HMIS staff at
regional and district levels.

Indicators
Performance on heath indicators was found to be reasonable (74 %) implying capability of the
health sector performances. National minimum core indicators have been identified for national
and regional/district levels covering all categories of health indicators (determinants of health,
health system inputs, outputs, outcomes; health status). A technical committee for Monitoring
and Evaluation has the responsibility for indicator generation. This committee operates under the
Sector Wide Approach structure as such offering good environment for compilation of data from
different sources. The major sources of health indicators include the HMIS; the different surveys
that are coordinated by the NBS which includes the DHS and HIV indicator Survey, National
Sentinels Surveillance sites, HIV and Immunization Surveillances and operational researches.
The assessment has revealed a clear strategy for measuring each of the health- related
Millennium Development Goals (MDG) indicators relevant to the country. In the health sector
there are 33 indicators selected for tracking health sector performance in response to the needs
for the MDG,the National Health Strategy for growth and Poverty Reduction (MKUKUTA)
which are all to be generated on annual bases. On a parallel note the SWAP committee has
selected four key composite indicators for General Budget Support (GBS) which would illustrate
sector progress and these are i) DPT-HpB3 Coverage, ii) Number of people with advanced HIV
infection receiving antiretroviral therapy, iii) Proportion of births attended by a skilled health
workers and iv) TB treatment completion rate. At the regional and district levels there are set of
minimum core indicators used to support Council and Regional Health Management Teams
(CHMT/RHMT) plan and evaluation. The national health indicators are implemented in line with
international indicators e.g. GF- ATM, Reproductive and Child Health indicators.

Data Sources
Census

Population censuses in Tanzania are planned for every ten years. For many reasons, mainly
financial limitations the 10 years cycle have not been strictly adhered to. Since independence
Tanzania has seen four censuses (in 1967, 1978, 1988 and 2002). Overall, census as a source of
health data was found to be adequate (65 percent).However, the assessment has revealed poor
performance (40 %) in terms of contents. Questions relating to recent household deaths
and indirect estimation for child and adult mortality were not adequate. Capacity of the country
to carry out the census was found to be adequate (76 %) and similarly the system performed well
on dissemination of census information (77 %) and use of information products (67 %)

Vital statistics
Vital registration in the country is the responsibility of RITA. Overall, the operation of this
system was found to be adequate (68 %). According to the assessment results, vital statistics
were found to be adequate by capacity and practice (65 %), dissemination (67 %) and highly
adequate for integration and use (83 %), though on contents the assessment has revealed shortfall
(58 %). Among the five sources of data, vital registration was found to be the most adequate.
These results however must be taken with caution because registration of vital events in the
country has been recorded to be very low and almost non-existent in many rural areas(1). Low
birth registration rate have also been documented by UNICEF in 2002 with a marked percent
point difference between rural (3%) and urban areas (22%)(2; 3).

Population-based surveys
Overall performance on population based surveys was adequate (65%). Contents in terms of
nationally-representative survey to measure percentage of the relevant population receiving key
maternal and child health services (family planning, antenatal care, professionally attended
deliveries, immunization) , nationally-representative survey to provide estimates of infant and
under-5 mortality and 1.3 In the pas measurement of prevalence of some priority no
communicable diseases/health problems (e.g. disability, mental illness, hypertension, diabetes,
accidents, violence) and leading risk factors (e.g. smoking, drug use, diet, physical inactivity)
were found to be present but not adequate(54%).Other aspects of capacity and practice,
dissemination and integration and use of data population based surveys.

Health & diseases records

Generally, health and disease records were found to be present but not adequate to meet the
countrys needs (56%). The areas of major weaknesses included lack of mapping of public health
risks and populations at risk, low proportion of investigated outbreaks with laboratory results,
lack of support on quality and continuity of care from individual patient records (patient charts or
patient- retained "health passports") , non use of International Statistical Classification of
Diseases and Related Health Problems (ICD) for reporting hospital discharge diagnoses and lack
of capacity to integrate reporting for disease surveillance and other focused public health
programmes (e.g. maternal care, family planning, growth monitoring). There was a positive note
though on availability and use of case definitions for epidemic and non-epidemic diseases;
capacity to diagnose and record cases of notifiable diseases; capacity to report and transmit
timely and complete data on integrated disease surveillance and response (IDSR) priority
diseases; and (iv) analysis and use of data for outbreak response and planning of public health
interventions. Appropriate case definitions for all 13 diseases (Table 2) under the IDSR strategy
were available. These include epidemic prone diseases, disease targeted for elimination
/eradication and diseases of public health importance. The later group of diseases (malaria,
tuberculosis, diarrhea, pneumonia), are the leading causes of morbidity and mortality in
Tanzania. Similarfindings were reported in 2005 from the baseline Monitoring and Evaluation of
Integrated Disease Surveillance and Response in Tanzania(1;4)
.
Health and disease records are collected in all private and public health facilities (dispensaries,
health centers and hospitals) using tally sheet and registers. IDSR priority diseases are compiled
weekly (epidemic-prone diseases) and monthly (all diseases) and transmitted to the district,
regional and eventually national level. In terms of capacities and practices, districts have some
capacities to diagnose record, analyze and report a limited number of notifiable diseases. Some
regions have capacity to diagnose most bacterial diseases. Capacity to diagnose some viral
infections such as measles and polio is only available at the national level.

Health service records

Health services records were judged to be present but not adequate (53 per cent). Despite the fact
that there is a national wide HMIS that is active and running in all public health facilities and
some of the non public facilities the system is facing a lot of challenges. There is lack of trained
cadre of health information specialists at district level, lack of mechanism from district to
national level to verify completeness and consistency of data from facilities, and non use
of findings from surveys, civil registration or DSS to assess the validity of clinic-based data by
managers and analysts at national and regional/district levels. There are an extensive number of
parallel systems known as HMIS complementary data collection tools.These have been created
for special programs like HIV/AIDS, Tuberculosis and Leprosy, Prevention of Mother to Child
Transmission, mainly to track programme specific indicators. Changes in health provision like
Health Insurance Programme, Neonatal care etc has also created demand for new data tools. A
fundamental issue has been lack of coordination of supplementary tools at all levels. As a result
service providers at lower levels are the ones carrying the ultimate burden. On a positive note
though there are mechanisms in place at national and sub national levels for supervision and
feedback on information practices. The data derived from health service records are used to
estimate coverage with key services such as antenatal care, delivery with a skilled attendant and
immunization.

Administrative records
The Health Management Information System entails as well administrative records in the form
of finance management system, Human Resource management system, system to manage drugs,
equipment, buildings, vehicles and other infrastructure. The assessment reported inadequate
performance in this area (53 %). There were shortfalls in terms of contents, capacity and
practices, dissemination as well as integration and use. There are no Global Positioning Satellite
(GPS) coordinates for the majority of facilities which is also paralleled by lack of human
resources and equipment for maintaining and updating the same. For the National Health
Accounts periodicity and timeliness is a serious problem.

Data Management

The system was found to perform adequately on data management (65%). There is a written set
of procedures for data management including data collection, storage, cleaning, quality control,
analysis and presentation for target audiences, and these are implemented throughout the country.
However, there is no integrated data warehouse containing data from all sources (both
population-based and facility-based, including all key health programmes) and this is true at
national and regional/district levels. Unique identifier codes exist for health facilities and
administrative geographical units (e.g. province, district, municipality, etc.) which facilitate
merging of multiple databases from different sources.

Information Products
The assessment revealed satisfactory performance of the system regarding the generation of
products of information in terms of timeliness, periodicity, consistency, representativeness,
disaggregation, and estimation methods (63%). There was adequate performance on these
parameters for health status indicators (69%), mortality (64%), morbidity (74%) and health
systems indicator (62%). However, the performance was inadequate on generation of products
for risk factor indicators (48%)

Health status indicators

Health status indicators in Tanzania are mainly collected through surveys such as the TDHS,
THIS and DSS. The NBS in collaboration with the MOHSW coordinates the collection of DHS
data while MOHSW in collaboration with sentinel districts provide household based data. The
assessed mortality indicators are the under five mortality (all causes), adult mortality (all causes)
and maternal mortality. Others include morbidity indicators like HIV prevalence and
Underweight in children (<59 months or <36 months).The Sentinel Demographic Surveillance
Systems (DSS) on the other hand enables estimation of Disability Adjusted Life Years (DALYs).
Ideally, the most reliable data collection source to provide empirical evidence on mortality
indicatorswould have been the vital registration; however, it has very low coverage in the
country. The findings imply that indicators for health status are adequately produced to provide
evidence base for the health sector needs. The only weakness according to the evaluation is on
periodicity of sources of data as this was found to be inadequate or too spaced.

Health system indicators

Indicators assessed in this category were Outpatient attendance, Measles coverage by 12 months
of age, Deliveries attended by skilled health professionals, tuberculosis treatment success rate
under direct observed treatment strategy (DOTS), proportion of children (<59 months or <36
months) sleeping under insecticide-treated bed nets and general government expenditure on
health (GGHE) per capita (Ministries of Health and Social Welfare, Public Security, Regional
and Local governments, extra budgetary entities), private expenditure on health per capita
(households' out-of-pocket, private health insurance, NGOs, corporations) and density of health
workforce (total and by professional category) by 1,000 population. Health systems indicators
are mainly collected routinely by the MOHSW through the HMIS. These indicators provide info
rmation on performance of the health system including human resources, drug supplies, and
disease burden which includes incidence and case fatality rates. Another set source of health
systems data are the SPA and SAM. In general, all the health systems indicators had adequately
met data quality assessment criteria. An overall, quality of the health system indicators was as
well found to be adequate (62 %).

Risk factor indicators

This area assessed the indicators on smoking prevalence (15 years and older), Condom use and
proportion of households using improved water supply (pipe-borne or borehole or protected
well).The overall performance was poor (48 %).It was noted that the indicators are not very
clear, either rarely done or are done on an ad hoc basis. Most indicators are not timely collected,
and periodicity is also too long. Although information on other risk factors is clear, the risk factor
assessments on smoking do not have clear system or not collected at all. It follows that the HIS
may not have adequate information to provide to the Tanzanian populace regarding risk factors
for health

Dissemination and Use

There is limited evidence on use of information for decision making. Overall performance on
this aspect was 47 %, the level which is generally inadequate. Care-providers other than those at
central level do not use health information for service delivery planning and monitoring.
Graphical presentation in a form of maps, charts and such others are generated but poorly
understood. HIS summary reports especially from the HMIS, the census and surveys are
sufficiently (relatively) analyzed and reports produced. The under five mortality rate,
immunization rate and HIV prevalence was generally known among health focused policy/
decision makers. There is limited systematic use of information on health risk factors to advocate
less-risk behavior in the general public as well as in targeted vulnerable groups. Use of
information for policy and advocacy (53%) or for planning and priority setting (48%) were found
to be in place although inadequate .However there was common use of information for
diagnostic purposes to describe health problems/challenges at Local Government levels (City
municipal, town and district council authorities). There was lack of synchronized use of
information between different planning frameworks. With regard to use of information for
resource allocation it was found that the practice was there but once again not adequate
(48%).Although performance was poor on use of information for resource allocation there
is a need based resource allocation formula for distributing resources from centre to districts.
This approach advocates for geographical equity in allocation of the resources. During the past 5
years there have been some shifts in annual budgets and a general resource allocation, but linking
HIS to these changes is not clear. On another note lack of such a formula within districts and
attachments population where by crude guidelines are used in allocation of resources to various
health facilities poses a challenge. The implication is that equity and increased resources to
disadvantaged groups and communities is not sufficiently addressed. Implementation and action
was found to be an area of critical shortfall with regard to use of information (38 %).

Policy Implication of The Findings

Supportive efforts made previously (since 1990) within the MoHSW and a number of external
pushes from other ministries and development partners had played a major role in attaining the
level of HIS development recorded in this assessment. Existence of health sector management
instruments such as the SWAP Committee and Basket Financing Committee has lead to creation
of mechanisms with direct push on HIS including the on-going Health Sector Support

Programme (HSSP), the Health Sector Review (HSR) process, establishment of M & E technical
committee and several others. The arrangements have been complimentary to government efforts
in the implementation of HIS in terms of setting performance indicators and strategies. Such
management instruments at policy level are likely to yield additional benefits if they are aligned
to mechanisms to stimulate effective use of information particularly at sub national level. It is of
paramount importance that such policy level mechanisms have clear links to HIS in a way that
puts to task the sub national levels with regard to use of information for implementation and
action. Already existing arrangements such as the resource allocation formula that aim to direct
resources to the main priorities in the health sector, existence of Comprehensive Council Health
Plans and the growing necessity for proof of performance at all levels provide good environment
for stimulating effective use of information. Moreover in order to enhance equity and increased
attention to disadvantaged groups it is vital that catchments populations within districts (in this
case health facilities) are also brought into the playing arena with regard to employing
mechanisms to ensure effective use of information. Such an emphasis on use of information that
dissects the districts is likely to bring about meaningful changes with ultimate feed into policy
review processes. The general inadequacy noted in HIS sub-systems should challenge HIS
stakeholders to a deeper level of analysis of these results and devise effective improvement
strategies. Better coordination of the stakeholders shall be a pre-requisite to coming up with an
HIS policy and strategy for the medium to long term.

Opportunity For Donor Coordination

Information on performance of the health sector is a key input to all stakeholders on board.
Performance monitoring is becoming critical to donors who have opted to coordinate, harmonize
and align their support to the sector through the general budget support and Health Basket
Funding. In this regard the agenda on strengthening HIS is becoming more and more a priority to
all the players. The already existing M&E technical committee for HIS and the Health Sector
Review (HSR) process within the MOH&SW along with other similar inter ministerial structures
such as the analysis working group for the National Strategy for Growth and Poverty Reduction
(MKUKUTA) and the SWAP committee provide good environment for coordination of technical
and financial efforts.

Analysis of Strengths, Weakness, Opportunities and Threats

HIS Strengths
Has good organization framework that is implemented countrywide and at all levels
Collects comprehensive data to suit most of the needs by stakeholders
It is supported by government and a wide group of stakeholders and in particular the HSSP
HIS Weakness
Has allowed mushrooming of parallel sub systems that continue to frustrate the mainstream
Has failed to accommodate current changes in the health sector a reasonable pace as such
helping to emerging of parallel systems HIS Opportunity
Development partners window to support harmonization of the system to compliment
Decentralization by Devolution (D by D) policy
Harmonization of HIS with the Poverty Monitoring System (PMS) provides a niche for utilizing
government resources in refining the HIS.
Threats to future development
Inadequate financial resources to support integration
If parallel funding by partners favors some subsystems in uncoordinated fashion.

Project Implementation
The project objective created to improve the management of health facilities and to that end
develop a user friendly HMIS to collect reliable information on all activities, and to produce the
reports needed by the hospital administration and the Ministry of Health. The HMIS aims to
provide
timely and accurate information leading to better healthcare planning, improved diagnosis and
increased patient access to health services. Direct users are the health workers, who are expected
to:
get better access to patient information
be able to spend more time on patients

 be able to spend less time on administration

It is expected that patients will also benefit, as they will:
spend less time in the hospital
get possibilities to do appointment scheduling
get more insight in bills
g et better access to medicines due to better inventory keeping
get better quality control
The specific objectives of the HIS project were as follows:
To digitally record all standard transactions in the hospital by means of customised software
To digitally produce all reports according to the standard specifications, including the
government health registration system Mfumo wa Taarifa za Uendeshaji wa Huduma za
Afya (or MTUHA, the Swahili name for HMIS), by means of the same software
To have satisfactory operation of the HMIS within the health facilities by ensuring the required
hardware is in place
To have trained and motivated staff who can use the HMIS system, maintain it and generate
reliable outputs
To have support services in place that can troubleshoot at distance within half a day
To assure user friendliness and improved work processes bycontinuously generating end user
and management feedback and integrating these results in the software and ways of operating
To roll-out the HMIS to 26 health facilities, supported by an introduction package including a
guideline manual, assistance in organisational change processes, user training and ICT support
and back up services

Software package and modules The project

team could build on earlier experiences, as sELCT had pioneered in Tanzania with the
development of a
DOS based HMIS system in 1997. It was used in the
beginning by six hospitals, but only one continued to use it
after a few years. This experience showed that there is
interest, because people indicated it helped them in their
work and they have since asked for a better version of the
system. But it also clearly showed the problems that inhibit
the effective use of such a system. Most problems related to:
the integration of ICT in the workflow, processes and
responsibilities
the robustness, effectiveness and user friendliness of
the system
t he computer skills of the hospital staff
t he technical support structure
The project team compared many alternatives which were
either too expensive, of poor quality or without affordable
long term support. After a careful review of available
solutions with a broad collection of stakeholders from
amongst others the Ministry of Health and the University of
Dar es Salaam Computer Centre (UCC), ELCT decided to
develop a generic Tanzanian version of Care2X, an open
source HMIS that is used in several countries over the
world. ELCT programmers cooperated with UCC, COSTECH
and a German software company to adapt the software to
the government health registration system MTUHA, and to
specific hospital requirements. Drafting of requirements
was done with hospital managers. Experiences and lessons
learned from the hospitals were continuously translated
into requirements for further development of the Care2X
software. The process of developing a Tanzanian generic
version of Care2X took about two years.
The choice for Care2X was based on the following criteria:
The software is free
Local support is available
Care2X has a proven track record

 It can be adapted to contribute to government (MTUHA)

reporting
Care2X is client/server software, which is scalable from a
single computer to a big hospital network. It is modular and
health facilities can choose which modules they need or are
able to use. Modules used in this project are patient
registration, billing, laboratory, pharmacy, radiology,
diagnosis and treatment, appointments, in-patient, outpatient,
eye-clinic and nursing

Software and system integration. Strengthening the MOHSW

HMIS unit and building the integrated district and national
data warehouses
General training of MOHSW staff
Develop Software team A dedicated software team is being built consisting of 4 MOHSW staff,
4 UDSM /HISP staff
This team is partly overlapping with the technical rollout team described above The software
teams main tasks are to develop the national and district data warehouses and to customize DHIS
and additional software tools as according to the needs of the MOHSW, Furthermore, to integrate
the data warehouse with other data sources by extracting the needed data. GIS functionality will
be developed as part of the collaboration with WHO on the OpenHealth project. The software
team will work with, and be part of, the global network of DHIS developers and work closely
with the WHO and HMN on the developing of appropriate Open Source tools. This work has
already started as three members of the software team (one of them form the MOHSW) took part
in a two weeks DHIS workshop in New Delhi, India, in March.
Regional and international experts will assist the MOHSW in the development of the data
warehouse, at national, regional and district levels, as well as in the rollout of the system.
Capacity building MOHSW; data management, DHIS data warehouse and open source
technologies

Capacity will be developed in the MOHSW through the activities of the software team
described above but also through the building of a solid base for managing the overall
National HMIS. All staff in the HMIS and ICT units will be part of this effort. As the building of
the National data warehouse and the National HMIS will take place in the HMIS unit, capacity
building will be part of every step of the process. In order to achieve capacity development in the
MOHSW, the project needs to recruit additional regular MOHSW staff, train existing staf
f, allocate skilled project staff to the HIR section and engage them fully in the development of
the system and its rollout. Based at the UDSM and the NOMA Masters Programme, which is
running courses on these issues, the project will engage in a wider national capacity development
in this area, as well as targeting individual HMIS and ICT staff for trainin
g (several are already included). Training of regional and zonal HMIS staff will also
be part of this effort.
Software development and systems integration
The software development and integration work package will be a continuous activity and will
be handled by the UDSM-MOHSW software team.
The team will be responsible to:

Design database and software for rapid rollout based on existing HMIS.
Update software and data warehouse design based on revised tools.
Integrate DHIS with other related applications (e.g. LGMD)
Revise database & software once new registers & tools finalised

The customization and further development of the DHIS software and its integration with the
WHO OpenHealth application for web based Geographical Information System (GIS), is
specified in a separate document. The DHIS-OpenHealth development is in cooperation with
WHO and Health Metrics Network. The WHO-OpenHealth is the new web-based application
developed by WHO to replace the HealthMapper.
The DHIS-OpenHealth is a scalable district and webbased data warehouse, which may work
as a stand alone application without internet connection in a district, or as a web based

national repository at MOHSW. Integration with other computer based application such as
Electronic Patient Record systems in hospitals and other facilities is handled through a
standardised data interchange platform.This, of course, is depending on the use of open standards
also by other computer applications, which will be ensured through the strategic planning
process. and other computerised data sources; establishing web-based data warehouse, electronic
reporting etc. The basic principle underlying the various software applications involved in this
work package is that first a first customised, stable and useful application is implemented, and
thereafter it is continuously further developed andintegrated with the patient record systems
such as the Care2X, and other systems in place.

Approaches in Software design:

The top-down approach

Rapid implementation in all districts over 3 years including 6 months initial phase
Overall objectives:

Implement and strengthen the integrated HMIS and DHIS in all districts and regions and
establish effective data flows to the MOHSW. The aim and indicator of success is to

reach level 1 of information usage according to the TALI tool (see annex),
Level 1; establish the basic system with quality data at all levels (including completeness)

and with the effective transmission of data between the levels

Develop a national data warehouse integrating all routine reporting and other data sources

and build capacity at the MOHSW.

Establish the Coast region as a test region for the development and testing of tools later to
be rolled out to all districts (an additional test region may be added later)

The work is organized in two components called work packages:

Work packages (WP):

WP1 National rollout of strengthened HMIS to all districts and regions. The rollout is labelled
top-down because an initial package of the revised HMIS is, first, tested in the Coast region
(WP2), second, rapidly deployed to all districts and regions, and third, consolidated over the
following two years, all in all a three years intervention. The initial revised HMIS consists of
HMIS and routine data sets and data reporting forms that are currently in use or recently revised,
and the DHIS software which is used for managing, analysing, presenting and transferring the
data. The HMIS data sets and reporting form will be further revised during the process. These
revisions will be accommodated when ready. Rollout approach: Each district region will have
two sessions of formal training (initial training, and repeat training about 8-12 months later) and
three sessions of on-the-job training and support to be carried out by a team of facilitators. While
the initial rollout and training will be rapid, the following technical support and training covered
by this WP will be carried out over about 2.5 years; July, 2009 - December, 2011.
WP2 Coast Region initial test region. Implement DHIS using the existing and revised
HMIS reporting forms in all districts and regional administration in the Coast region.
Establish a first version of the data reporting forms and data sets to be included in the national
roll-out. Including the revised reproductive health and human resources data sets. Later revisions
of data reporting and data sets will be included when they are ready for implementation. Changes

in data sets and data reporting forms are easily accommodated by the software (which is soft),
but more problematic in terms of paper forms (which are hard copied).
The first phase of this test region is planned for 6 months, until the contracts are formalised.
Thereafter the test region will be used to develop and test all intervention packages of the overall
project (another test region may be added).The Test region will be used to test all new tools,
software customisation, and understand user needs at facility, district, hospital & regional le
vels. Adapt support, training & tools accordingly, and then roll-out to other regions.
WP3 Software development and Systems Integration. Developing District based National data
warehouse and building capacity at MOHSW. The first version of the national data warehouse
will be similar to the DHIS used in the Coast region and will be used to receive data from all
over the country as the rollout advances. Gradually additional data sources will be included and
improved functionality such as Geographical Information Systems and web reporting will be
developed. A key issue is to integrate and include the routine data being collected at facility level
and to remove double reporting and redundant data reporting. The aim is to achieve integration
and unification of all data reporting. This work on data and indicator sets is contributing to, and
is integrated with, the revisions of data sets in WP4.A software team including the MOHSW will
be built and become in charge of the further development and improving of the DHIS software
and related technical aspects at district and national levels. A program for training and capacity
building for MOHSW staff will be established.
Total budget for WP 1-3 and top-down approach: about 2.5 mill USD Was funded by the
Royal Norwegian Embassy.
Attached minimum budget 30 months test region: 310,000 USD
National rollout: Institutional base and coordination
The three WPs in the National Rollout are naturally based at the HMIS unit, MOHSW, and
will be located in an office here and managed and coordinated from the HMIS unit. However,
University of Dar es Salaam (UDSM )and the University of Oslo were responsible for the
software component (WP3) and the capacity building at the MOHSW in this regard. The funds
are to be managed in relation to the HMIS unit, MOHSW, with one part of the budget allocated
to UDSM and the University of Oslo for the software component.

The Bottom-up approach;

System strengthening, information usage and capacity building;

Objectives:

Develop capacity and ensure that information is analysed and used for informed decision

making and action in all districts and regions.

The aim and indicator of achievements is to achieve Level 2 and 3 of information

usage according to the TALI tool (see annex) for 80% of the districts;
Level 2; Information is analysed, disseminated and used (graphs on the walls, routine

district quarterly reports, etc.)

Level 3; Information is (documented) used for (district) planning and the evaluation of

the implementation of these plans

Institutionalise quarterly / regular data-use workshops in all districts and regions;
information from the HMIS is presented, analysed, discussed in the workshop and
concrete action plans are made for the improvement of the HMIS and data quality as
well as for the improvements of the services as according to the data.

Work packages
WP4 Revisions of data and indicator sets and HMIS procedures - address and include
new data needs as they emerge during the process. Revision of data and indicator sets is an
ongoing process and includes interaction with the health programs. Initially the rollout will be
based on the existing HMIS forms and already revised data sets. Revisions will be included as
they are ready for implementation. Aim is to arrive at a situation where a national co
mmittee is responsible for the harmonisation of the total array of reporting requirements to the
health facilities. Programs and agencies wanting to introduce new data reporting forms will
then have to approach this committee in order to get acceptance for additional requirements.
Requirements when introducing new data reporting forms will include that they are harmonised
within the overall HMIS framework and that no double reporting is accepted.
WP5 Building district and regional capacity. Establish a HMIS training scheme on analysis
and use of information for management and decision making for program and district
managers at district and regional levels in the entire country. The training will consist of repeated
formal training sessions with individual and group assignments and support in between.

WP6 System strengthening; Improving data quality, information usage and health
management.
This WP consists of three major set of activities;
1: Institutionalise quarterly /regular data-use workshops in all districts and regions, where data
is analysed and discussed, problems identified (both regarding HMIS and health services),
and action decided upon.
2: Institutionalise the production and dissemination of quality district and regional quarterly
and annual reports based on the HMIS data
3: Establish and strengthen the district and regional information office and DHIS data warehouse
as an active and competent information resource centre that is effectively responding to the needs
of the district /region: Create District Information Towers Approaches to enable these activities
in districts and regions are to establish a team of HMIS facilitators and a support structure
covering the country. WP 5 trainers and assignments will make up an important part of the
support structure to WP5.
WP7 Establish and use the Test region to develop and test tools and methodologies including
revised datasets. (This WP will be combined with WP1, which has a preliminary status until the
project is formally established). The test region will be used to develop, initially implement and
test all the interventions described in the above WPs; data sets, software, training programs,
guidelines for data workshops, quarterly and annual district reports, strengthening the district
information office and data warehouse, etc. The test region will be run by hiring additional staff
based at the region, conduct particular investigations and have increased support as compar
ed with the other regions.
Total budget for WP 4-7 and bottom-up approach: about 5 mill USD was be funded by
the Dutch Embassy. Some basic components of the test region is included in the roll-out budget
to be covered by the Norwegian Embassy; 2 staff, vehicle, support, training of district and
program managers. This was done because of uncertainty as to when funding and contracts
would be available.
System strengthening bottom-up: Institutional base and coordination
WP 4-7 consist of distinct tasks, such as running and coordinating
Revisions of data and indicator sets and proceduresfor their collection, management

and use (WP4); a committee of programs and stakeholders will be established (based
on the M&E committee), responsibility, coordination and management, clearly located
to the MOHSW.
Training scheme (WP5); while being coordinated through the MOHSW, clear responsibilities
for the development of programs and the running of the courses are given Ifakara and UDSM.
Test region (WP7); need to be closely linked to the other WPs as a test-bed, but the
responsibility to run it may be delegated Ifakara
Continuous system strengthening (WP6), which is a further strengthening and after
the first three years the continuation of the National rollout process, should be
managed and coordinated by the HMIS unit, MOHSW.
The funds are to be managed in relation to the HMIS unit, MOHSW, with items of the budget
allocated to Ifakara in relation to the training scheme and the Test region..

Management and coordination

Specification of the Bottom-up approach - System

strengthening, information usage and capacity building

The bottom-up approach consists of 3 interlinked work packages:

WP4 - Revisions of data and indicator sets and HMIS procedures (continuous, but with initial
baseline revision)
WP5 - Building district and regional capacity training scheme
WP6 - System strengthening; Improving data quality, information usage and health management
WP7 - Establish and use the Test region to develop and test tools, methodologies and revisions
Work Package 4 Revisions of data and indicator sets and HMIS procedures (continuous, but
with initial baseline revision)
An indicator set has existed in Tanzania since the 1990s, but this is now seen as being out of
date, fragmented and inadequate and needs to be reviewed in conjunction with all relevant
stakeholders. Numerous additional program specific data collection tools and routines are in
use. There is a need for international best practice standards to be applied to indicator
selection as well as to the harmonization of the routine data collection system. There is a
perceived need for harmonized set of indicators and corresponding data sets and
data collection tools and routines, clearly acknowledged by MoH&SW top management,
M&E chapter of HSSP III, donors and programs.
Objectives
First phase (first 2 months)
Review existing data collection tools and indicators being used by MOHSW and the
relevant programs
Identify key performance indicators for the health sector in Tanzania
Identify data sets and data collection tools and HMIS procedures to be part of the
initial phase. Make sure indicator numerator and denominator, with clear definitions of
all components, are covered (when possible)
Identify sources and frequency of all data and indicators
Hospitals; same procedure for hospitals. The first phase approach is to assess current
forms and effectively harmonize and use what is there.
Define and establish the data sets that are going to be part of the first rollout:

Vertical programs: HIV, and subsequently EPI, Leprosy/TB, Malaria.

Handle the overlaps: same data collected by several programs and data collection tools:
These data will as a start be harmonized within the DHIS database (while one single data
item might be collected in many forms, such as e.g. First ANC visit, it will only be
captured and registered once in the database /DHIS). During phase 2, they will also be
harmonized in the data collection tools.

Second phase (first 18 months)

Harmonize all relevant data collection tools and routines to be used at facility and district levels,
and design produce and distribute new forms that will be the primary monthly manual return
from facility to district. Given the slower process to print and replace existing paper based data
collection tools, this process will go on beyond the initial phase.
Establish a permanent national committee to oversee ongoing revisions of data and indicator
sets. The aim is to arrive at a situation where programs needing more data and reports will
have to present their new data demands for this national committee and arrive at an
agreement on whether the new demands are justified, whether the needed data are already
collected, how, eventually, the new data needs may be incorporated in the existing routine data
collection system, or whether new data collection tools, provided they are not overlapping any
existing tool, are needed. This measure is to stop the process of fragmentation.
Special focus on hospitals. Harmonize all relevant data collection tools, data sets and
indicator sets.
Scope of Work
Review existing data collection tools, indicator set and data definitions
Conduct consultation meeting with all key HIS stakeholders and their implementation
teams
oMOHSW Policy makers, cooperating partners
oRelevant vertical program managers,
oHMIS managers, District, provincial managers
Develop draft national indicator set with defined numerator and denominator and source, in
accordance with international norms and HMN framework
Circulate indicators with numerator, denominator, data source, rationale, use, related

indicators
Organize and facilitate consensus meeting on indicators with each program
Design, print and distribute new paper forms which are shared between MOHSW and
all relevant health programs. There will be no duplication and overlapping data
collection (paper) forms. HMIS, RCH, EPI, etc, willall be based on a shared set of forms
Continuously incorporate revised forms and definitions of data, data forms and indicators into
DHIS database /data warehouse
Revise the entire package of paper based data registration tools used at the facility
level; design, print, distribute the new tools. IMPORTANT; design a strategy for their
continuous update and revision over time, i.e. design for addendums according to the
future updates
Devise a mechanism to revise data and indicator sets annually (i.e. phase 2)
Hospitals
Similar scope of work, but in a different context. Many hospitals are introducing
computer based systems. The program needs to take this into account and plan for a
gradual integration of these systems based on electronic patient records.
The scope of work includes
O The aggregated data flows within the hospitals (wards), to be managed within
DHIS at each hospital (gradually)
O Reporting from all hospitals to the HMIS/DHIS
O Integration of the electronic patient based systems being implemented in the
hospitals with the system for aggregated statistica
l data reporting; HMIS and
DHIS
Establish a sub-group within the program for the revision of the hospital HMIS
Develop hospital indicators and data sets
Develop revised reporting forms for hospitals
Incorporate revisions in the DHIS
Implement DHIS in selected hospitals

Plan for implementation of computerized HMIS and DHIS at all 200 hospitals. Make
this part of the strategy to integrate electronic patient records with the DHIS and
HMIS
Develop a strategic plan for the Hospital Information Systems in Tanzania, where the
process of introducing electronic patient based record systems in hospitals are
coordinated and integrated within the wider M&E framework as according to the HMN
technical framework
Work
To establish a national team responsible for co-ordination, consultation, and finalizing new
registers, tools, indicators. Each member of the consortium (MoHSW, Ifakara, UDSM, UiO)
will assign a staff member full time for this national team.
Phase 1 the first 2 months; establishing the initial data and indicator sets This task, in Dar
es Salaam, will start togethers with the initial phase in the test region and go on for about 2
months.
During the two months, one staff (equivalent) from MOHSW, Ifakara, UDSM and UiO will each
work 6 weeks; 3X30 days= 90 days @ 100.000 = 9.000.000 Tshs
Phase 2 revision and replacements and other needed tools (Month 7 12)
The same/equivalent task force will continue and work on the more radical revision of the
paper based tools used in all facilities. This work will start 4 months after the initial data sets
are implemented (in order to gain some experience),go on for 6 months, with slightly less
intensively; 250 days = 25.000.000.
Important will be to be realistic in designing the additional tools and books supporting book
2; paper based books/registers/forms are difficult to replace once printed. They need to
adhere to certain flexibility, i.e. include blank columns.
Hospitals Phase 1 and 2

A task force from the same partners and of the same size and intensity as for Phase 2 (250
days, 25.0000.000) will carry out this work over the first year of the project. While the first
phase, reporting from district hospitals will be part of Phase 1, referral hospitals need
additional attention. Plans for further computerization and the electronic integration of
patient record systems such as Care2X, and the collaboration with projects of this type, will
make up an important part of this work. The task force will also develop a strategic plan a
nd framework for the development of the overall Hospital information system in Tanzania, an
d its integration within the M&E framework, beyond the initial 18 months. Printing and
distribution need to be budgeted separately.
Budget: 250 days, 25.0000.000 Tsh
For all phases above: UiO is covered through the international consultancy budget.
WP5 Building district and regional capacity training scheme
Objectives: Run courses and develop capacity in HMIS and data use at
Regional and national levels (training of trainers of district staff)
District level staff (training of trainers of facility level staff)
Facility level staff
Develop training materials and guidelines for the above training scheme
This capacity development scheme will have to be carried out in stages starting with the
regional and national levels, and about 3 months later, also with the district level staff in the
test region(s) The first objective and stage for this work package is to establish a training
program for the training of trainers; regional and national staff.
Three staff from each region, including the HMIS focal person, totalling 63, and about 12
from national and zonal level, totalling about 75. Four training sessions of 2 weeks will be
conducted over 2 years. The topics of the training will address HMIS and management,
analysis and use of information for health management and health services delivery.
Assignments to be completed between the training sessions will include; use HMIS and other
information for situation analysis, planning and target setting, as well as the organisation of

data use workshops at district and regional (i.e. for all districts in the region) levels. The training
will be conducted using the zonal training centres where appropriate. The second objective is to
devolve an adapted part of this training program to the district level, starting in the test region(s).
Three times one week training over about 1.5-2 years will be carried out at the regional level
for 3-4 staff from each district. With about 25 persons per training session, some regions may
be combined. The training will include the same issues as for the regional staff, but with an
additional emphasise on facility supervision and the training of facility level staff. Assignments
to be completed between training sessions will include, as for the regional level staff; use HMIS
and other information for situation analysis, planning and target setting, as well as the
organisation of data use workshops at district level for district and facility staff. Regional level
staff will be responsible for conducting the training, but with support from national level, in
particular during the first session. The third objective is to devolve the training scheme to the
facility level. The test region(s) will be used to develop cost effective methods to train facility
level staff (data use workshops may be the primary methodology, linked to supervision an
additional one). Details will be developed later. For long term HMIS sustainability, an HMIS
module will be designed and integrated into pre-service training. In addition, a diploma in Health
Informatics programme will be established for HMIS cadre.
Development of training materials
Regional Training Program - one staff equivalent from each of the consortium members;
MOHSW, Ifakara, UDSM and University of Oslo, will each work 4 weeks up to and including
thefirst 2 weeks training session for
regional and national staff.
District Training Program - the team of about 4 persons develops a down-scaled version of the
regional training program for the district level. This work will be conductedin the test region(s)
and will include the practical development of a methodology and guidelines for data use
workshops both at the regional level (for the districts) and at the district level (for district
managers and facility staff). The practical development of data workshop methodology and
guidelines will also feed into the regional training program. Four staff equivalents will work 8
weeks on this task.

The implementation of the HMIS system is ongoing. At themoment of writing it is operational in

six hospitals and fivehealth centres: St. Elisabeth Hospital, Arusha LutheranMedical Centre
(ALMC), Haydom Lutheran Hospital, Wasso Designated District Hospital, Kilimanjaro Christian
MedicalCentre (KCMC), Marangu Hospital, Tumaini Health Centre,Kijenge Health Centre, USA
Health Centre, Njiro HealthCentre, Mtoni Health Centre.
Change management
The project hired a change manager responsible for the organisational change management
process. Workshops were organised to prepare staff for the changes ahead and to determine their
attitudes towards the changes, at three operating levels:
ELCT bishops
Management of the hospitals
Health staff of the hospitals
Since independence, healthcare in Tanzania has moved fromcharity, mostly provided by religious
entities, to a publicservice. Government as well as donor organisations stimulate the
professionalization of health service provision to ensure this public service can be accessed by
all. Hence faith based hospitals increasingly rely on funding from performance based financing
schemes and service level agreements with government, which is based on hospital output data.
IICD facilitated change management sessions in which the bishops and the project team mapped
the changing contextin which hospitals operate, to come to a better undersanding of how HMIS
can assist hospitals in dealing with these changes. The IT Unit facilitated all the change
management workshops in the hospitals. In the course of the project, IICD and the ELCT IT Unit
developed a guide for change management. This guide is based on a proven and successful
approach developed by IICD in the setting of district governance. It uses a 14-step approach,
starting with the germination of the original idea to set up an HMIS right through to the
establishment of a fully operational system. Steps 1 to 5 are about preparation and training:
laying a strong foundation is the first requirement for constructing a sturdy building. Step 6, 7
and 8 are the core of the approach. Together, management and staff decide on how
to implement the HMIS. This includes identifying the specific problems of the health facility and
then developing a joint vision on where to go. Steps 9 to 14 are all about troubleshooting,
implementing the HMIS, and finally handing over the full-fledged system to a capable workforce

that is able to work with it and maintain it. Training of IT Unit, local software developers and
health staff At the start of the project, a German software companyassisted in the programming
and testing. The IICD Capacity Development programme supported the IT Unit in training
local developers. The University Computer Centre in Dar es Salaam was also working on the
generic version of Care2X and IICD facilitated face to face exchanges as well as online sharing
of code.
IICD also supported the IT Unit in providing highquality training programmes for health staff.
Members ofthe IT Unit followed a Training-of-Trainers programme, to sharpen training needs
assessment, planning and implementation skills.The IT Unit organised ICT training at hospital
level (on the job training). Most hospital staff members did not have any computer skills and had
to start from scratch. In total,about 130 staff members were trained in basic ICT skills and the
use of Care2X, usually a mix of nurses, doctors and medical recorders.From each hospital, one
staff member - usually a health worker with special interest in ICT - was selected and trained
in basic computer and network maintenance and troubleshooting. Five hospitals recruited their
own ICT technician. Support services Software problems in the hospitals are tested and fixed
remotely from the IT Units office in Arusha. Also software updates are installed remotely. At
hospital level, technical people are trained to repair server or internet connection breakdowns. In
this way the project team has been able to work with the eleven health facilities where Care2X is
implemented. On average, the team visits each site once every two months.
Integration with other software Two hospitals integrated an open source accounting package
(WebERP) in the HMIS, allowing the export of aggregated billing information from Care2X to
WebERP. As explained above, the HMIS generates data that improve the management at the
level of individual health facilities. An interface has been built between Care2X and the District
Health Information System DHIS, which enables aggregation of health facility data at district
level and official reporting to the national health authorities. This way aggregated health
facility data from a certain district or region can be used to make informed health policy
decisions.
Impact

The number of questionnaires collected over the years is too low to measure impact as perceived
by users. The openanswers in the 29 questionnaires do give an impression of how health staff is
using the HMIS. When asked what actions they undertook as a result of this project in order
to improve their work, users provided answers like: I managed to register all patients diagnoses
and prescriptions electronically and to produce all the reports electronically. Another user
replied: Keeping records ofpatients on the computer. Get reports of the hospitaleasily. When
asked why they have achieved their goals, a user answered: I can register and enrol a patient
without delay. Another user replied that the patient flow can easily be monitored.
When asked why they have not achieved their goals, users provided answers like: Lack of
computers, Lack of full management support to the project, and Need for more training.
Economic impact
In an economic impact study carried out by an independent consultant at the end of 2009, it was
found that health facilities using HMIS had recorded an increase of income collected from
different collecting sites in the hospitals. This was evidenced through interviews and focus group
discussions with management and staff members from billing stations, pharmacies and
laboratories in six health facilities. The responses also indicate an improvement in
logistics and drugs management. The respondents said this is seen as the result of a good and
effective HMIS. Unfortunately, only one health facility could provide data on revenue collection
before (2006) and after HMIS implementation (2007 and 2008). Revenues increased 33%
between 2006 and 2007 and 34% between 2007 and 2008. These are indicators that the HMIS
contributed to an increase in revenues, although there are no data on patient numbers and other
causes cannot be excluded. The respondents noted that the increase in income collected from
health facilities could be related to:
C ontrol of fraud points: all patients attended are recorded in the system. Hence money
collected is always equal to the number of patients attended at the health facility.
R eduction of so-called corridor clinics, which implies the treatment of unrecorded patients by
health providers without recording the patients and their payments.
B etter logistics and drugs management, through use of an effective HMIS to minimise
wastage.

Impact on health care

Time saving In the impact study, health workers pointed out that they were able to retrieve
patient files in the HMIS much faster than in the manual system, increasing efficiency by saving
both patient and staff time, so patients can be attended faster. Anecdotal evidence from
interviews with 30 patients states that the HMIS urged doctors to do ward rounds daily
to update their HMIS records, hence providing health care at the right time and reducing bed
occupancy time. Also, estimates of reduced patient waiting time from 24 hours to 30 minutes are
mentioned, as well as reduced loss of patient files from 30% to 1%. On the other hand, health
workers noted that in case of power cuts or system failure, they cannot access patients
records, causing delays and inconveniences. They areforced to go back to the manual system.
After the power comes back, manually collected data are not always etered into the system.
Improved data security
The IT Unit ensures that all the records stored by the system are encrypted and saved with one
local backup and a remote backup for each hospital. In the impact study, respondents from
administration and management observed that data are nowadays safe from damage, corruption,
or modifications by unauthorised individuals. Improved data safety was for all respondents a
major benefit from the project. Side effects The project has been found to be beneficial to the
users and health providers who are now computer literates, increasing their professional status
and career opportunities. In the M&E questionnaire a user comments: I use the computer with
more confidence, I work more with modern technology, and I can explore other technology.
Lessons learned
In theory, all aspects have been thought of; the change management guide with the 14 step
approach that was developed during the project roll out prescribes a thorough preparation in
terms of awareness raising and training among management and staff to enable them to take
ownership and decide on the HMIS implementation. In practice, it still turned out to be
extremely complex for hospitals to take full ownership and implement an HMIS.What can be
learned from the differences between expectations and results so far? Manage expectations

When looking back, expectations set at the start of the project were too high. Project developers
expected that the software could be developed and customized in one year; it took two years. It
was expected that every two months a hospital would start implementing the HMIS. Within four
years, six hospitals and five smaller health units have started to use the HMIS, with varying
intensity. The most successful hospital is implementing ten modules of the HMIS. The least
successful hospitals are only (partly) using the patient registration module, which means there is
still a risk of turning back to manual. Organisational change needs a lot of time and patience.
This is also reflected in the M&E questionnaire, where a user comments on the HMIS: People
should get more time to understand its importance.Identify key agents of change at several
levelsChange agents need to be identified at the levels of government, hospital management,
hospital departments, and health workers. Since the dioceses of the Lutheran Church are the
owners of the hospitals, also church leaders need to be involved. From the most advanced
hospitals the IT Unit learned that the attitude of the hospital management towards the
HMIS is essential. If management prioritises and pushes for HMIS, everything is possible: staff
resistance can be overcome, electricity problems are dealt with. It just takes time.The IT Unit
learned that some hospitals are more motivated to use HMIS than others. Therefore, priority in
rolling out the project is on those hospitals that are most motivated and ready for HMIS
implementation. Creating champions in this way will inspire the more reluctant hospitals.Apart
from hospital management, also church leaders like bishops and administrators need to be
actively involved and understand the importance of the HMIS, since they influence decisions on
how the hospitals are run. What is said or recommended by church leaders often has implications
for service delivery. Recently the IT Unit also started to involve government officials to enhance
public private partnerships and get government support for the system.
Similarly, departments as well as individuals can be agents of change. In general, junior staff
members take up ICT (Information Communication Technology) more easily than older staff
and can thus set the example. Some departments pick up the HMIS faster than others. A certain
department was very reluctant in participating. When its staff members saw the others going
ahead, they felt ashamed, joined and made up quickly for the delays.
Make benefits to management explicit from the start

Benefits like reduced time spent in hospital, reduced loss of medicine and increased financial
revenues are or should be - strong incentives for management to incorporate the HMIS in the
work flow. The project team noted that some hospital managers are more receptive to these
benefits than others. Whether it is used as a carrot or a stick, the
project team needs be able to provid e hard proof of these benefits by monitoring indicators
from the start. In thecurrent project, no baseline measurement was done and noindicators for
management benefits were set. This lesson was picked up in a similar, more recent project
supported by IICD in Malawi, where indicators like financial revenues and patients length of
stay have been included in the project design from the start.
Provide coaching on information-based management
In the project design it was assumed that increased access to health data would automatically
result in better managerial planning and decision making; no guidance for management was
offered. During the roll out of the project it turned out that support in the use of (HMIS) data for
management and planning was needed. This was translated in additional training in financial
management and special staff was recruited for that purpose. IICD applied this lesson in a related
HMIS project in Mwanza region, where the project extension now includes peer learning
sessions for hospital managers, to exchange on the use of data to improve planning and decision
making. At facility level, there is a need for more guidance on data entrance and for feedback on
how the data are used. The IT Unit noted that data entered in the national MTUHA system are
often estimates or cooked data. If health workers get feedback on the collected data, especially
their own outputs, they will be aware of the importance of data and have an interest to collect
them.
Involve everyone in the change process
When implementing an HMIS, more effort is needed in change management and working on
peoples attitude than in the technical part. Management and healthcare staff have to first get
familiar with ICT. It is very important that everyone involved in the health facility can speak out
about the problems they encounter, as well as suggest possible solutions, and that they can
contribute to how to implement an HMIS. By involving everyone in the design and
implementation of the HMIS, all can contribute and each individual can adapt to the changes this

inevitably brings. Change management workshops were organized separately for bishops,
hospital managers and health workers, as the project team noticed that health staff was
reluctant to speak out in presence of senior management. Typical concerns expressed by health
personnel were their fear to loose their jobs if computer use would result in staff cuts, and the
fear to get more work due to both manual and digital recording. As the IT Unit became aware of
these fears, they were able to address them and hence reduce the resistance to the system.
However, not everyone participated in the workshops and resistance could not be taken away
entirely. Also it was noted that in some units only the people directly working with the HMIS
were aware of the benefits of the system. This limited the support of the project by the
entire staff. Involving the entire staff would also create more ownership and broaden the interest
of all staff to be computer literate.
Strengthen IT support at hospital level
At the start of the project, the IT Unit reckoned that the ICT support at hospital level could be
created by training a health worker from the hospital with special interest in ICT. Yet the IT Unit
had to visit the hospitals constantly for troubleshooting and hardware problems. Hence it decided
to review its strategy, by advising hospitals to employ an IT technician before Care2x is
implemented, to have ownership from the start. The IT Unit now requires this from hospitals. As
it is difficult to find ICT technicians in remote areas, the IT Unit trained two interns to become
ICT technicians and transferred them to different hospitals. The successful implementation of
Care2X in Arusha Lutheran Medical Centre, one of the first hospitals to be introduced to the use
of HMIS, can be attributed to the full time young ICT university graduate that the hospital
employed and who took responsibility for all ICT matters. In total, five hospitals recruited their
own ICT technician. Provide tailor made ICT training for hospital staff It was difficult to
determine what motivates health workers to learn to use a computer. Many of them are
overworked and underpaid, which makes them reluctant to participate unless they have a direct
interest. Some hospitals started to include basic computer knowledge as a requirement when
recruiting new medical staff. Several lessons were learned about ICT training for hospital staff:
Management support is key and needs to be developed: management has to allow trainings to
take place duringworking hours. As soon as these kinds of events are organised outside official
working hours, people will not attend unless financially compensated.

 Atraining room on site increases the ability and willingness of staff to attend training and is
cheaper than sending staff to a training centre off site. However, it is more difficult to
control attendances.
Due to the general high staff turnover in hospitals, ICT training needs to be provided
continuously.
An onsite training room enables staff to continuously and independently upgrade their skills by
using the training room in their free time whenever it is convenient for them.
Independent and peer to peer training is to be preferred over classical training conducted by a
paid trainer. While there can be reluctance from experienced staff, the young newly recruited
staffs swallow it easily and can teach others.
Preferably trainings are conducted in shifts according to the different working schedules, type
of health worker and skills as determined through a skills and needs assessment
Challenges
Tacit resistance to change

Although the IT Unit tried to guide the hospitals through all the steps of the change management
guide, hospitals did not always follow them. The project team stumbled upon tacit resistance: in
general, hospital managers express their commitment to the process, in practice there is a lack of
collaboration. Pauline Kimollo, project manager, explains how such resistance is manifest
expressed: The facility is giving a lot of excuses when you plan a certain activity. Or
when you visit the facility for an agreed activity, no one is bothering your presence and nothing
is organised for the activity. In several cases, the IT Unit decided to stop working with unwilling
hospitals and continued in facilities showing genuine commitment. The project syndrome
Among the project developers there was a general belief that hospitals were ready to accept the
HMIS and would make a deliberate move to own and support the system. However, HMIS
uptake, especially the financial integration of the system, took longer than expected and in most
hospitals it is still not fully integrated. When health facilities started to use the HMIS, the IT Unit
managed to make agreements with them to share the accommodation and transport costs for
technical support visits. Health facilities partially pay the costs of ICT equipment and sometimes
the salary of an IT technician. Yet the majority is not ready or able to incur all the costs of an
HMIS; the switch from external project financing to internal budgeting seems to be difficult. The
hospitals still depend heavily on external funding. In the impact study it was stated that most

health facilities consider the HMIS as an external project of IICD and the IT Unit. This makes
them passive in investing in the HMIS project.
Partial use of the HMIS
The least successful hospitals are only (partly) using the patient registration module. In the M&E
questionnaire, a user states: we are finding some of the data are not filled so we are ending up
with missing information. As long as the HMIS is not sufficiently integrated in the workflow,
there is still a risk of turning back to manual.
Power and connectivity cuts
The system is mimicking the manual system so that when power goes off or the server is down,
the old manual system can be used. However, power cuts are so frequent that they endanger the
proper use of the system. Power back up systems like standby generators are expensive to run.
Some hospitals are not even connected to the national grid and have to get electricity through
expensive alternative power solutions. Hospitals are increasingly using low power solutions like
Inveneo desktops.
Lack of software developers
The software developers that were trained with support of IICD all went for other opportunities
and are not working for the project anymore. The IT Unit is still looking for a way to attract a
pool of strong developers. Israel Pascal, system administrator: We consider selecting students
from colleges in Arusha by giving them a specific task; the best performers will be offered a
job.
Staff turnover
There is a tremendous movement of health staff from one facility to another which affects the
performance and progress of the HMIS. It is difficult to maintain the system when there is a
serious staff turnover. For example, at one facility implementing the HMIS, five nurses, one
medical recorder and two doctors left their job and moved to a government health facility. The
project team needs to have a sustainable training strategy to deal with this kind of challenge. It
plans to train three to five staff members in each facility to be ICT trainers, who can schedule

training whenever needed. On the longer term, basic computer skills and e-health applications
need to be integrated in the curricula of the national health education system.
Next steps and future plans
At the moment of writing, the Ministry of Health and Social Welfare of Tanzania is developing a
national eHealth strategy. It is a multi-stakeholder process andsupported by IICD. The eHealth
strategy defines how ICT can reinforce the implementation of the Health Sector Strategic Plan
2009 2015, and aims to leverage existing initiatives, among which the HMIS project of ELCT.
The endorsement of the eHealth strategy will therefore facilitate the roll-out of the HMIS project.
In the extension phase of the project, a financial plan for the HMIS will be made with each
hospital before implementation, to ensure ownership from the start. IICD, Cordaid and FELM
intend to support the expansion of the project.
Bibliography
http://rchiips.org/NFHS/index.shtml