HIV/AIDS DATA QUALITY AND

INFORMATION USE TRAINING

PARTICIPANT MANUAL
HIV/AIDS DATA QUALITY
AND INFORMATION USE
TRAINING MANUAL

NOVEMBER 2020

November 2021
PA RT I C I PA N T ’ s M a n ua l 1
© Federal Democratic Republic of Ethiopia, Plan Policy Monitoring and Evaluation Directorate,2021
ICAP at Columbia University in Ethiopia
All rights reserved
First publication, 2021

Development of this publication was supported by ICAP in Ethiopia with a fund from PEPFAR
through CDC-Ethiopia. Its contents are solely the responsibility of the authors and do not necessarily
represent the official views of ICAP or the funder.

2 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
4 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
LIST OF TABLES 4
LIST OF FIGURES 5
ACRONYMS 6
ACKNOWLEDGEMENT 7
MANUAL INTRODUCTION 8
MODULE 8
HIV/AIDS DATA QUALITY 10
SECTION 1: 10
Overview of HIS/M&E of HIV/AIDS program 10
1.1. INTRODUCTION TO M&E OF HIV/AIDS PROGRAM 10
1.2. DATA SOURCES FOR M&E OF HIV/AIDS PROGRAMS 10
1.3. RECORDING AND REPORTING TOOLS FOR HIV/AIDS PROGRAM 12
1.4. HIV/AIDS PROGRAM INDICATORS 13
SECTION 2: 16
Overview of enhanced EMR-ART system 16
2.1. INTRODUCTION TO EMR-ART SYSTEM IN ETHIOPIA 16
2.2. BASIC FEATURES OF THE ENHANCED EMR-ART SYSTEM 16
2.3. BENEFITS OF THE ENHANCED EMR-ART SYSTEM 18
2.4. INTEGRATION OF EMR-ART WITH OTHER HEALTH INFORMATION SYSTEMS 19
SECTION 3: 20
Introduction to Data Quality in HIV/AIDS Programs 20
3.1. DATA QUALITY AND ITS IMPORTANCE 20
3.2. LEADERSHIP IN DATA QUALITY 22
3.3. DETERMINANTS OF DATA QUALITY 22
SECTION 4: 24
Health Data Quality Dimensions 24
4.1. Introduction to Data Quality Dimensions 24
4.2. METRICS OF DATA QUALITY DIMENSIONS 24
SECTION 5: 30
Data Quality Assurance 30
5.1. DATA QUALITY ASSURANCE AND ITS IMPORTANCE 30
5.2. TECHNIQUES OF DATA QUALITY ASSURANCE 30
5.2.1. VISUAL SCANNING (EYE BALLING) 31
5.2.2. LOT QUALITY ASSURANCE SAMPLING (LQAS) 31
5.2.3. DATA QUALITY CHECK BETWEEN ORIGINAL RECORDS AND REGISTERS/ EMR-ART 31
5.2.4. USING EMR-ART TO IMPROVE DATA QUALITY 35
HIV/AIDS INFORMATION USE 37

PA RT I C I PA N T ’ s M a n ua l 5
SECTION 6: 37
Basic concepts of data, data demand and information use 37
6.1. BASIC CONCEPTS OF DATA, DATA DEMAND AND INFORMATION USE 37
6.2. INFORMATION REQUIREMENT AND USE AT DIFFERENT LEVELS 38
SECTION 7: 42
Extraction/ Generation of Patient Level Data from EMR-ART 42
7.1. GENERATING AND EXPORTING PREDEFINED & CUSTOM REPORTS 42
7.2. GENERATING AND EXPORTING OTHER CLINICAL SERVICE REPORTS 45
SECTION 8: 49
Data Analysis, Presentation, and Interpretation 49
8.1. BASIC CONCEPTS OF DATA ANALYSIS 49
8.2. METRICS OF HEALTH AND HEALTH RELATED DATA (SPECIFIC TO HIV/AIDS
PROGRAM) 49
8.3. DATA PRESENTATION/VISUALIZATION 53
8.3.1. OVERVIEW OF DATA PRESENTATION/ VISUALIZATION 53
8.3.2. TECHNIQUES OF DATA VISUALIZATION/PRESENTATION FOR HIV/ADIS PROGRAM 54
8.3.3. E-SYSTEMS FOR DATA VISUALIZATION 59
8.3.4. MINIMUM DISPLAY CHARTS FOR HIV/AIDS PROGRAM UNIT 61
8.4. DATA INTERPRETATION 63
SECTION 9: 65
Data Analysis and visualization using Excel 65
9.1. DATA ANALYSIS USING EXCEL 65
9.2. DATA PRESENTATION & VISUALIZATION USING EXCEL 80
SECTION 10: 94
Information use for Action 94
10.1. STEPS OF INFORMATION USE FOR ACTION 94
10.2. Platforms for information use 98
SECTION 11: 99
Monitoring and evaluation of HIV/AIDS information use 99
REFERENCES 100
LIST OF WORKSHOP PARTICIPANTS 101

LIST OF TABLES
Table 1. Intended users and uses of quality health care data 21
Table 2. Sample data quality assessment result using data quality check 34
Table 3. Health Center /Hospital Data Quality Assessment Action Plan template 34
Table 4. Patient level HIV/AIDS indicators for individual level HIV/AIDS information use 39
Table 5. Aggregate level HIV/AIDS data elements/indicators for HIV/AIDS program data use 40
Table 6. Number of HTC tests performed and positive tests in health center A,
Tir-Sene 2011 EFY 55
Table 7. HIV/ ADIS program Scorecard of zone A in region B, EFY 2011 61
Table 8. Minimum display charts for HIV/AIDS units 61
Table 9. Action Plan Template 99
Table 10. Sample checklist to monitor the presence of information use in HIV/AIDS units 98

6 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
LIST OF FIGURES
Figure 1. Sources of data, classified as population based and facility based sources 11
Figure 2. EMR-ART validation rule (Example for grandfather name) 35
Figure 3. EMR-ART validation rule (example for gender of a patient) 36
Figure 4. EMR-ART data completeness report sample 36
Figure 5. HMIS/DHIS2 report generated from EMR-ART 43
Figure 6. DATIM report generation from EMR-ART 43
Figure 7. Line list report generation and exporting from EMR-ART 44
Figure 8. Cohort Report Generation and Exporting from EMR-ART 44
Figure 9. Custom Report Generation and Exporting from EMR-ART 45
Figure 10. Tracing Line List Report generation and exporting from EMR-ART 46
Figure 11. Generation and Exporting of Scheduled Visit Report from EMR-ART 46
Figure 12: Providers View Line List Report 47
Figure 13. Data Quality Assurance Data Generation and Exporting from EMR-ART 47
Figure 14. Viral Load Eligible Clients Line List Exported from EMR-ART 48
Figure 15. Number of positive HIV tests by region, EFY 2011 56
Figure 16. No. of people tested positive for HIV and no. of PLHIVs newly initiated
on ART, 2011 57
Figure 17. Number of PLHIV newly started on ART by Sex, EFY 2011 57
Figure 18. Number of PLHIV currently on ART in Health Center X from EFY 2000-2011 58
Figure 19. Pie Chart showing nutritional status of PLHIVs in Hospital X, 2011 EFY 59
Figure 20. ART EMR-ART dash board 60
Figure 21 - 22. Steps to rename a sheet name in excel 67
Figure 23 - 24. Steps to change excel data into table 71
Figure 25. Data Filtering in excel 72
Figure 26. Excel functions in excel to calculate measures of central tendency & dispersion 73
Figure 27. Selecting values on a pivot table using excel 75
Figure 28. Analyzing TX_Curr by Regimen and Sex 76
Figure 29. Steps of information use for action 94
Figure 30. Template for prioritization of problems 96
Figure 31. Fishbone Diagram 97

PA RT I C I PA N T ’ s M a n ua l 7
ACRONYMS
AIDS Acquired Immunodeficiency Syndrome
ANC Antenatal Care
ART Antiretroviral Therapy
ARV Antiretroviral
ASM Appointment Spacing
CDC Center for Disease Control
DHIS2 District Health Information System
DHS Demographic and Health Survey
EMR Electronic Medical Record
EPHIA Ethiopia Population Base Impact Assessment
HIT Health Information Technician
HIS Health Information System
HIV Human Immunodeficiency Virus
HMIS Health Management Information System
HTC HIV Testing and Counselling
HTS HIV Testing Service
ICT Index Case Testing
IPD Inpatient Department
LQAS Lot Quality Assurance Sampling
M&E Monitoring and Evaluation
MER Monitoring, Evaluation and Reporting
MOH Ministry Of Health
OPD Out Patient Department
PEP Pre-Exposure Prophylaxis
PEPFAR President’s Emergency Plan For AIDS Relief
PLHIV People Living with HIV
PMT Performance Monitoring Team
PMTCT Prevention of Mother to Child Transmission of HIV
PITC Provider Initiated Testign and Counselling
PrEP Pre-Exposure Prophylaxis
RDQA Routine Data Quality Assessment
STI Sexually Transmitted Infections
TB Tuberculosis
TX_CURR Currently on Treatment
VCT Voluntary Counselling and Testing
VL Viral Load
WoHO Woreda Health Office

8 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
ACKNOWLEDGEMENT
The Ministry of Health would like to gratefully acknowledge the group of experts and organizations, men-
tioned below, who have invested much of their time and energy in the content development, contribution to
the review, refinement and finalization of this training resource package. This manual is developed with expert
inputs from national and regional experts in the area of HIV, Health Information System (HIS) and M&E under
the leadership of Ministry of Health’s Policy, planning, Monitoring and Evaluation Directorate and Disease
Prevention and Control Directorate (HIV/AIDS case team). We thank the ICAP in Ethiopia for the technical
and financial support provided to develop this training manual.

PA RT I C I PA N T ’ s M a n ua l 9
INTRODUCTION
Evidence informed decision making in the health sector is highly dependent on the availability and
use of high quality data. Better clinical and administrative decisions for better health outcomes can
be achieved only if quality data is available and information use at all levels is exercised. HIV preven-
tion, treatment and support programs have been implemented in Ethiopia for the last many years. To
monitor and evaluate the performance of the program and to provide high quality HIV/AIDS service
to clients and patients, data from different sources ranging from individual level to population level is
crucial.
This training manual mainly focuses on HIV program data quality assurance and information use at
health facility level. It focuses on individual and facility level HIV/AIDS data quality and data use for
HIV/AIDS client/patient and program monitoring. It is mainly prepared for health facilities that pro-
vide HIV/AIDS care and treatment service.
It is a participants’ manual and has two main modules:

 Module 1 is about HIV/AIDS data quality: This module focuses on basic concepts of data quality,
data quality dimensions and its metrics. It also includes data quality-assurance methods to be used
at HIV/AIDS units of health facilities.

 Module 2 is about HIV/AIDS information use: It covers topics such as HIV/AIDS data analysis,
data presentation/visualization techniques, using EMR-ART to present HIV data, HIV/AIDS data
interpretation, using MS-excel for data analysis and presentation and use of data for decision-mak-
ing.

10 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
CONTENTS OF THE MANUAL
This training manual covers mainly two areas:
 HIV/AIDS patient level and program level data quality assurance and improvement
 HIV/AIDS information use for evidence informed decision making at health facilities providing
HIV/AIDS treatment service.

LEARNING OBJECTIVES
The overall objective of this training manual is to enhance the capacity of health workers at health cen-
ters and hospitals in assuring the quality of HIV/AIDS data and improving HIV/AIDS data use during
their day-to-day HIV/AIDS service provision.
At the end of the training, participants will be able to:
 Understand the main causes of poor data quality and identify solutions to improve data quality
 Describe dimensions of data quality and measure each dimension
 Identify, describe and use data quality assurance methods for HIV/AIDS data
 Develop and use HIV/AIDS data quality improvement interventions in their facilities
 Define and describe patient and program level HIV/AIDS indicators
 Develop skills of HIV/AIDS data analysis, presentation and interpretation techniques
 Develop basic excel skill for HIV/AIDS data analysis and data presentation
 Use HIV/AIDS data for patient level and program level decision making

TRAINING METHODS
 Interactive lectures using power point presentation
 Group exercises followed by group presentation
 Case studies
 Practical session on EMR-ART and use of MS-Excel
 Exercises and group works
 Plenary discussion
 Case scenario

TARGET GROUPS
Target audience for this training are mainly, but not limited to: -
 ART focal persons
 ART data clerks
 Health Information Technicians
 Health workers working on HIV program
 HIV program managers
 HIV program M&E officers

PA RT I C I PA N T ’ s M a n ua l 11
HIV/AIDS DATA QUALITY
SECTION 1:
Overview of HIS/M&E of HIV/AIDS program
SECTION OBJECTIVES:
At the end of this section, participants will be able to:

 Identify the types of recording and reporting tools for HIV/AIDS program
 Describe data sources of HIV/AIDS program
 Identify and define indicators of HIV/AIDS program

1.1. INTRODUCTION TO M&E OF HIV/AIDS PROGRAM

The global emergency of HIV/AIDS has led to unprecedented attention and commitment from the in-
ternational community to improve access to HIV/AIDS prevention, care and treatment services. The
expansion of HIV/AIDS prevention, care and treatment must be accompanied by effective monitoring
and evaluation (M&E) and operational research to guide implementation and to see that efficiency,
effectiveness, quality of care and acceptability are established and maintained. M&E helps program
and health facility managers assess the effectiveness of interventions and linkages between services
along the cascade of testing, treatment and care for HIV/AIDS and associated conditions. Such infor-
mation is essential to detect and respond to bottlenecks or gaps in program performance and to ade-
quately characterize and respond to patient attrition. Patient monitoring systems are also important
to support people receiving treatment over time and ensure retention in care. As programs mature,
monitoring is also essential for individual- and population-level outcomes, such as toxicity and adverse
events, drug resistance, viral suppression, mortality, survival and incidence, to assess and optimize the
impact of country.

Globally, there is a global momentum towards ambitious targets that measures progress towards HIV
diagnosis, treatment and viral suppression. Ethiopia has adopted the UNAIDS three 95s targets and
working towards it. The three 95 targets of HIV/AIDS are:

 1st 95 target: by 2025, 95% of all people living with HIV will know their HIV status
 2nd 95 target: by 2025, 95% of all people with diagnosed HIV infection will receive sustained
antiretroviral therapy (ART)
 3rd 95 target: By 2025, 95% of all people receiving antiretroviral therapy will have viral suppression

1.2. DATA SOURCES FOR M&E OF HIV/AIDS PROGRAMS

HIV/AIDS data for patient and program monitoring and evaluation can be generated from different
levels in the health system. Data sources can range from individual level up to the population level.

 Individual level data: includes data about individual client’s/ patient’s profile (Socio-demographic
data), health-care needs and service /treatment provided, history of present and previous medical
history, laboratory and diagnostic results. Example: Eg. Type of ARV regimen received by PLHIVs,
viral load test and its result, CD4 count of individual PLHIVs. The source of data for patient level data
includes patient cards, ART intake forms A and B, ART follow up cards, viral load request form, EMR-
ART database etc.

12 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 Health-facility level data: includes aggregated data of HIV/AIDS services, HIV/AIDS related mor-
bidity and mortality data. Example: Number of HTC clients, Number of adults and children with
HIV infection newly started on ART, Number of adults and children currently receiving ART. The
sources of facility level data include: VCT register, ART register, PMTCT register, Viral load logbook,
Partner & Family based index case testing logbook, HIV care and ARV regimen tally sheets, EMR-
ART database etc…
 Population level data: these data are from the population, are essential for public health deci-
sion-making, and generate information not only about those who use the services but also about
those who do not use them. Eg. HIV prevalence/incidence. Population level data sources for HIV/
AIDS include census, ANC sentinel surveillance, Demographic Health Survey (DHS), Ethiopia Pop-
ulation-based HIV Impact Assessment (EPHIA), Civil Registration and Vital Statistics (CRVS) and
other surveys and researches.
 HIV/AIDS Surveillance: HIV/AIDS surveillance is an ongoing, systematic collection, analysis and in-
terpretation of HIV/AIDS data essential to the planning, implementation, and evaluation of HIV/
AIDS program improvement. Example: Number of recent HIV infections identified. Sources for
HIV/AIDS surveillance include HIV case-based surveillance form
Data sources for HIV program can also be categorized into many different other ways. Some of the
common classifications include: Routine and Non-routine sources or Population based and facility/
community based.
 Routine data sources: refers to data that are collected continuously on a routine basis, with pro-
cessing and reporting more often than annually. It includes patient level and aggregated level HIV/
AIDS data. Health Management Information Systems (HMIS) are facility-based and communi-
ty-based subsystems that collect routine information on patients as they use services. Vital reg-
 istration is routine data collection on vital events (births, deaths, and migration data) and occurs
mostly outside the health system.

 Non-routine data sources: refers to data that are collected on a periodic basis. Examples: house-
hold surveys, national survey, researches

Figure 1. Sources of data, classified as population based and facility based sources

PA RT I C I PA N T ’ s M a n ua l 13
1.3. RECORDING AND REPORTING TOOLS FOR HIV/AIDS PROGRAM
Recording and reporting tools are used to systematically collect, monitor and evaluate progress of
individual patient’s treatment outcome as well as the overall performance of a program. In Ethiopia,
the reporting of HIV/AIDS prevention, care and treatment services is integrated into the Health Man-
agement Information System (HMIS) and all client/patient forms and registers are standardized in line
with the HMIS throughout the country. Health facilities are the primary sources of data for HIV/AIDS
program monitoring and evaluation.

The HIV/AIDS program uses standard HMIS tools to record, collect, report and document information
on HIV/AIDS prevention, care and support activities at all levels of the health system. The following
are summary of the standardized HIV/AIDS recording and reporting tools used.

1. Individual client/patient record forms: Individual medical recording forms are those, which are
used to record the medical and clinical information of individual clients and/or patients. At health cen-
ter and hospital levels, each client/patient is expected to have an individual record where all the ser-
vices provided are recorded. Some of the individual record forms that are used for HIV/AIDS clients/
patients include:

 Individual folder: It is a folder that is used as a pouch to contain all the medical records of each client/
patient. It is used to integrate all medical & health service records of an individual patient or client so
that the holistic medical data of individuals can be accessed in one folder, whenever required.

 ART intake form A and B: These are forms that are used to record basic baseline information about a
patient enrolled into ART program. It should be complete by health care workers for all patients that
are enrolled to ART including pregnant and lactating women.

 HIV/AIDS Care/ART Follow-up form: This card is used to record follow up information about a patient
who is receiving HIV/AIDS treatment. It helps to document patient follow up information. Informa-
tion from the follow up card is used to update the ART register, PMTCT register and EMR-ART da-
tabase.

 Referral form: This is a form that is used to refer patients from one facility to another

 Other individual forms: There are other individual record forms such as lab request forms, prescrip-
tion forms, admission discharge forms etc…

2. Registers: Registers are forms that are used to record abstracted summary data elements from
individual client/patient data. It is used to simplify report compilation at facility level. For HIV/AIDS
program, the following registers are used:

 VCT Register: It is used to record and document data of individuals who receive voluntary HIV coun-
selling and testing service. It is used to capture data elements related to HIV testing services, such as
number of people (disaggregated by age, sex and population category) that received counselling and
testing service and number of people tested positive for HIV. Provider initiated HIV testing (PITC)
and its result is integrated with all other HMIS registers (eg. In OPD register, PITC is integrated).

 PMTCT Register: It is used to follow HIV positive pregnant and lactating women and their HIV
exposed infants. The register includes summary information for reporting and calculation of indica-
tors related to PMTCT.

14 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 ART Register: ART register is used to record and document data for PLHIV who are on HIV treat-
ment. It is also used as a patient monitoring tool throughout their ARV therapy. The information re-
quired to complete ART register is obtained from HIV intake and/or follow up form.

 Post Exposure prophylaxis (PEP) register: This register is used to record information about people
who are taking post exposure prophylaxis, either for occupational or non-occupational risks. The
reportable data elements from this register are number of people who received PEP for occupation-
al and non-occupational risks.

3. HIV/AIDS program tally sheets: Tally sheets are forms that are used to simplify counting and ag-
gregating number of people or services provided in a health facility. If a facility is using an electronic
patient recording system such as EMR-ART, tallies are not required since the electronic system can
help us aggregate the number of people or services required for reporting. Some of the tally sheets
used for HIV/AIDS program are:

 VCT tally sheet: This tally is used to simplify reporting of data elements from the VCT register. VCT
data is reported disaggregated by age, sex, population category and test result

 PITC tally sheet: PITC tally sheet is used to document provider initiated counselling and testing and
its result. The tally sheet is completed from different registers that have a PITC component such as
OPD register, IPD, TB, family planning, ANC, delivery registers etc…

 HIV Clinical Care Tally sheet: This tally sheet is used to simplify reporting of ART data elements for
patients who are currently receiving clinical care, disaggregated by age and sex. It includes tallying
the data elements such as number of PLHIV newly started on ART, number of viral load tests and its
result, screening for TB and its result, nutritional screening for PLHIVs and its result etc…

 Currently on ART and regimen tally sheet: This is used to simplify reporting of data elements for
patients who are currently receiving ART by type of regimen, sex and age.

4. HIV/AIDS program reporting form: The HIV/AIDS program monthly reporting format is part of the
routine HMIS and the report will be collected monthly and reported as part of the service delivery
report form. The HIV/AIDS program component of the service delivery reporting format addresses
all HIV/AIDS program indicators related to prevention, care and treatment. Important information/
data/ indicators that are not captured by the monthly routine HIV/AIDS service delivery report but re-
quired for other reporting such as for external donor reporting can be generated from the EMR-ART.

1.4. HIV/AIDS PROGRAM INDICATORS

Indicator is a variable that measures one aspect of a program or project that is directly related to the
program’s objective. Indicators are always directly linked to the objective setting of a program. Indica-
tor measures the value of change of a single aspect of a program or project that is directly related to
the program’s objective with the purpose to indicate change but not explain how or why.

According to the National HMIS indicator guide-2017, standardized indicators to monitor HIV pro-
gram are defined. These indicators are developed to measure the success of comprehensive package
of HIV/AIDS prevention, diagnostic, treatment, care and support services provided for people at risk
of HIV/AIDS infection or living with HIV/AIDS and their families at various levels of the health system.
In addition to the national HMIS indicators, there are additional PEPFAR Monitoring, Evaluation, and
Reporting (MER) indicators that can be used to monitor and evaluate HIV/AIDS response.

PA RT I C I PA N T ’ s M a n ua l 15
Note: These indicators are those that are nationally defined in the 2017 national HMIS indicators. The
list may be updated periodically, as per the national needs and requirements.

National HIV/AIDS/ Indicators as defined in the national HMIS

1. Percentage of people living with HIV/AIDS who know their status
2. Percentage of people living with HIV/AIDS receiving ART
3. Viral load suppression
4. Early viral load suppression rate
5. ART retention rate
6. Number of adults and children with HIV/AIDS infection newly started on ART
7. Proportion of clinically undernourished People Living with HIV/AIDS (PLHIVs) on ART who received
therapeutic or supplementary food
8. Number of persons provided with Post-Exposure prophylaxis
9. Percentage of non-pregnant women living with HIV/AIDS on ART using a modern family planning
method
10. Proportion of Sexually Transmitted Infection (STI) cases tested for HIV/AIDS

TB/HIV Indicators
1. HIV/AIDS screening for TB patients
2. TB Screening for HIV/AIDS positive Clients
3. Latent TB infection treatment for HIV/AIDS positive clients newly enrolled to care
4. Anti-Retroviral Therapy (ART) for HIV/AIDS positive TB patients

PMTCT indicators
1. Percentage of pregnant and lactating women who were tested for HIV/AIDS and know their results
2. Number of HIV/AIDS positive pregnant and lactating women who received ART at ANC, L&D, P&C
for the first time
3. Number of HIV/AIDS positive women who get pregnant while on ART and linked to ANC
4. Percentage of HIV/AIDS infected women on HIV/AIDS care and using modern family planning
method.
5. Percentage of infants born to HIV/AIDS-infected women receiving a virological test for HIV/AIDS
within 12 months of birth
6. Percentage of infants born to HIV/AIDS-infected women who were started on co-trimoxazole pro
phylaxis within 12 months of birth
7. Percentage of infants born to HIV/AIDS-infected women receiving antiretroviral prophylaxis for
prevention of mother-to-child transmission
8. Percentage of HIV/AIDS exposed infants receiving HIV/AIDS (confirmatory test antibody test) by 1
8 months of birth

In addition to the national HMIS indicators, MER indicators are also used to monitor the progress of
HIV program. MER indicators are also requirements for donor external reporting. Some MER indica-
tors are similar to the national HMIS indicators but there are few more indicators in addition to what
is available in the national HMIS indicator reference.

Examples of MER indicator that is not available in the national HMIS indicator list:
 Number of ART patients with no clinical contact (or ARV drug pick-up) for greater than 28 days
since their last expected contact who restarted ARVs within the reporting period
 Indicator Name code: TX_RTT

16 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
  Numerator: Number of ART patients with no clinical contact or ARV pick-up for greater than
28 days since their last expected contact who restarted ARVs within the reporting period
 Denominator: N/A
 Reporting Frequency: Quarterly

 Number of ART patients (who were on ART at the beginning of the quarterly reporting period) and
then had no clinical contact since their last expected contact
 Name: TX_ML (Indicator Code)
 Definition: - Number of ART patients with no clinical contact since their last expected contact
 Numerator: - Number of ART patients with no clinical contact since their last expected
contact
 Denominator: - N/A
 Reporting frequency: Quarterly

EXERCISE:
Konso health center provides HIV prevention, care and treatment service for clients/patients
residing in its catchment area. In Hidar 2012 EFY, the facility has reported the following HIV/
AIDS services:

 800 people were tested for HIV at VCT and 10 of them were tested positive. All the
positives were linked to ART unit and started taking ART
 350 PLHIVs have been receiving ART,
 200 PLHIV have been screened for TB among which 20 were diagnosed to have active
Tuberculosis
 Viral load test was done for 25 PLHIVs on ART, among which 20 had a viral load of less than
1000 copies per ml

Discuss on the following Questions based on the case scenario

 What indicators can be computed from the above case?
 Compute at least three indicators from the above case scario
 What recording tools are used to generate the reports described in the case scenario?
 List the types of registers and tally sheets to be used to generate the reported data elements
above

PA RT I C I PA N T ’ s M a n ua l 17
SECTION 2:
Overview of enhanced EMR-ART system
OBJECTIVE:
At the end of this section, participants will be able to:
 Understand the basics of enhanced EMR-ART system
 Describe the basic features of the Version 5 EMR-ART system
 Understand the benefits of EMR-ART system

In the previous sections, we have learned about HIV/AIDS recording and reporting tools. Development
and use of electronic system for HIV/AIDS program is essential and it helps providers better manage
patients by providing accurate, up-to-date, and complete information about patients at the point of
care. Moreover, it enables quick access to patient records, improves patient and provider interaction
and communication, as well as health care convenience, and reduces costs through decreased paper-
work, improved safety, reduced duplication and other benefits. In Ethiopia, an electronic system called
EMR-ART has been developed and used for HIV/AIDS treatment service. In this section, we will learn
about the basics of the EMR-ART.
2.1. INTRODUCTION TO EMR-ART SYSTEM IN ETHIOPIA
Appropriate HIV/AIDS service record keeping is an essential requirement for improved continuity of
care for HIV/AIDS patient. With significant increase in ART coverage, there has emerged a growing
concern among researchers that many developing countries lack the capacity to support the compli-
cated treatment regimens associated with ART. As a chronic disease without a cure, HIV/AIDS care
necessitates a lifetime of care and treatment, a multidisciplinary approach, and laboratory, pharmacy
and clinical data to monitor patient disease-related processes. Clinicians need to carefully and fre-
quently monitor patient health status and initiate appropriate therapy when needed. Therefore, the
ability of countries to provide and sustain effective long term HIV care with ART requires a patient
monitoring system that integrate care, prevention and treatment, for which development and use of
an electronic system supports such requirements.
EMR-ART system is one of the EMR systems intended to enhance and manage patient level HIV data.
It is a computerized electronic health record system used to record/store, process, retrieve and report
ART patient’s health information. It can provide such record keeping as well as summaries of history
of a patient’s care, allowing health workers to be updated on a patient’s previous medical history and
progress in response to treatment. EMR-ART system has been developed by ICAP at Columbia univer-
sity in collaboration with MOH and CDC-Ethiopia to assist experts who are responsible for providing
care and program monitoring in health facilities where EMR-ART application is implemented. EMR-
ART system uses patient level secondary data encoded by data clerks, that it followed what is said to
be e-Last implementation modality. Since 2018, MOH in collaboration with ICAP at Columbia Univer-
sity have deployed the enhanced EMR-ART system at more than 477 health facilities in 11 regions (as
of September 2020).

2.2. BASIC FEATURES OF THE ENHANCED EMR-ART SYSTEM

Enhanced EMR-ART system has the following features: Dashboard, central patient registry, HTS,
Exposure Prophylaxis, Treatment and Follow-up, Providers view, Pharmacy module, Viral Load Test,
Report and Analytics, Administrative, Data Quality Assurance, Help, logout and quit features. These
features of EMR-ART system are described briefly below. (Please note that this is an overview of the
EMR-ART features. Details of enhanced EMR-ART is covered in a separate EMR-ART manuals)

18 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
1. Dashboard: A data dashboard is an information management tool that visually tracks, analyzes and
displays key performance indicators, metrics and key data points to monitor the health of a service
delivery, department of a specific process. On EMR-ART, the dashboard displays selected HIV/AIDS
indicators.

2. Add Patient: This feature allows adding and registering new patient, entry of ART intake A & B, pos-
itive tracking, case based registration, Retest, Follow-Up, Tracing, Index Case testing, transfer in & out
information are captured. This is the feature where centrally all features are accessed.

3. HIV Testing service (HTS): Data related to all tests performed that identify a new case of HIV are
documented in this feature. The following are the major components of the HTS section: HIV Case
reporting - where a new case of HIV is documented, HIV Positive case tracking, retesting and Index
Case Testing (ICT).

4. Provider view/Module: The Providers View is designed to enable user to automatically check the
completeness of data of clients registered in the system by assigning standard labeling for Address,
Viral Load, TB Preventive Therapy (TPT), Partner & FBICT and ASM status. This Provider View EMR-
ART implementation approach will result accessibility of secondary electronic data of the paper-based
individual records by care providers or data producers in a networked environment. This e-last EMR-
ART implementation approach aimed to improve HIV data quality, use of electronic data for individual
level decision support and to strengthen program monitoring and evaluation.

5. Treatment and follow-up: In this feature, all follow up information of a patient including appoint-
ment spacing is documented. in addition, filling follow up form, tracing and data related to the transfer
out of the client is also documented here.

6. Viral load monitoring: This feature is designed for easy receiving and synchronization of the vi-
ral load test results though inbuilt interoperability layer. Electronic test order and Result request
(ETORRS) for sending and receiving lab request and result. The platform uses a point-to point data
communication with viral load database system by using web APIs. The communication uses EPHI
server as a mediator for sending and receiving electronic test order/result. When viral load result re-
turned from regional labs electronically, the EMR-ART system binds clients’ result to their follow-up
record. In addition, using this version of the software, VL Requisition, VL Register book, VL Eligibility
list, and VL result reports can be generated.

7. Pharmacy Module: The pharmacy module is an automated data entry at pharmacy unit and enable
capturing of ARV and other medications dispensing data, simplified generation of aggregate drug con-
sumption report, and line list of ARV and other inventory management reports for decision-making
purpose.

8. Cervical Cancer Screening and Treatment: The data entry and reporting for cervical Cancer Screen-
ing and Treatment as part of chronic HIV/AIDS care is automated in the latest version of EMR-ART
system.

9. HIV Case Based Reporting: HIV Case based reporting was managed by direct data entry in the
REDCap, a web-based data entry and reporting tool while most of the same data is being entered in
the EMR-ART. This has become an additional burden for data clerks who are responsible for both sys-
tems. This version has automated the new HIV Case Reporting Form (CRF) with complete similarity
to that customized in REDCap. The data clerks need to enter the data only into the EMR-ART and has
functionality to send or upload data to REDCap for reporting.

PA RT I C I PA N T ’ s M a n ua l 19
10. Operation Triple Zero (OTZ): Operation Triple Zero is also new initiatives under piloting in select-
ed facilities with aim of improving access of young and adolescents to the highest possible quality HIV
care in Ethiopia. The data need of this service is fully automated in EMR-ART 5.0 for standardized data
management approaches.

11. Exposure Prophylaxis: One of the cares given at facilities providing ART service includes provision
of ARV for individuals with risks of contracting HIV infection due to some exposure. The proper doc-
umentation of the follow up data of individual receiving this service is essential. This section includes:
Post-Exposure Prophylaxis(PEP) and Pre-Exposure Prophylaxis (PrEP).

12. Report and Analytics: EMR-ART Module enables users to generate monthly service delivery HIV
related reports except reports related to HIV testing and counseling service that performed out of
the ART clinic. HMIS report can be generated and exported so that it can directly be imported to the
DHIS2 software, a platform which is currently used as a national reporting electronic system. Other
reports such as DATIM report, line list, custom report, charts and maps are also included in the report
and analytics module.

13. Administration: Administration module of the system enables the user and system admin to man-
age user accounts and the security of data. In this section: data backup, restore, user management like
account creation and security of the software will be managed.

14. Data Quality Assurance: Despite high automation, much data is and always be typed into elec-
tronic systems by people through various forms and interfaces. Data quality assurance feature is in-
cluded as an additive to support the data clerks manage data quality issues such as duplicate records,
consistency, completeness, restore and delete records.

15. Help: This section of the software contains all the detail procedures or guides that will enable the
users be able to browse through the contents of the application for clarity and advanced use accord-
ingly. This may help users to learn by themselves as required.

16. Logout: Enables the users to logout from a user from the system.
17. Quit: This part provides additional option to close the module directly by clicking from this section.
The other option is the one located at the upper right corner of the page marked by “X” mark.

2.3. BENEFITS OF THE ENHANCED EMR-ART SYSTEM

EMR-ART system creates standardized and comprehensive electronic patient level HIV data manage-
ment tool. Some of the benefits of EMR-ART System include:

 Automate all data entry and reporting standards set by MOH and PEPFAR
 Standardizes patient documentation and reporting systems
 Create conducive environment for the use of high quality HIV data for evidence based decision
making
 Support data clerks and service providers in HIV Data management and use practices
 Continuously update changes related with revisions of the national & PEPFAR indicators
requirements
 It leads to improvement in quality of healthcare, decreased time spent on paper work, increased
patient satisfaction and financial saving
 Improve the confidentiality of a patient’s medical record by enabling password protection, making
confidential patient data accessible only to authorized clinicians.

20 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 Easy receiving and synchronization of the viral load test results though inbuilt interoperability layer,
which may have some impact in reducing unacceptably high turnaround time of the viral load test
result. The feature also enables data clerks to view and manage the viral load requests and results.
 Supports health facility level data quality assurance efforts among data producers and users

2.4. INTEGRATION OF EMR-ART WITH OTHER HEALTH INFORMATION SYSTEMS

The term “interoperability” describes the ability of two or more information systems or components
to exchange information based on standards, and to use the information that is exchanged without
special effort on the part of the user. This means that all users should be able to send, receive, find
and use electronic health information in a manner that is appropriate, secure, timely and reliable to
support the health and wellness of individuals through informed, shared decision-making. It enables
different HIS to work together in and across organizational boundaries to advance the health sta-
tus of individuals and communities and the effective delivery of healthcare to them. An interoperable
health IT ecosystem should support critical public health functions, including real-time case reporting,
disease surveillance and disaster response. In addition, interoperability can support data aggregation
for research, which can lead to improved clinical guidelines and practices. Over time, interoperability
will also need to support the combining of administrative and clinical data to enhance transparency
and enable value-based payment. Interoperability and flexibility are required in a national healthcare
system strategy. Interoperable with other open-source systems such as DHIS2, and flexible for adding
new features.

Currently, the EMR-ART version 5.0 (October,2020 release) has data exchange capability with the al-
ready existing web-based REDCap software. Based on this new feature the EMR-ART can directly
send or upload CBS reports to the web based REDCap software and can download previously sent
reports for editing or other uses, with available internet connection in the facility. Complete communi-
cation enabled with national Web-Based Viral Load & EID dashboard to enable direct sending of Viral
Load requests from facilities to the testing lab and direct receiving of results, with available connec-
tion at facility level.

PA RT I C I PA N T ’ s M a n ua l 21
SECTION 3:
Introduction to Data Quality in HIV/AIDS Programs
SECTION OBJECTIVES
At the end of this Section, participants will be able to:
 Describe the concepts of data quality and its importance
 Understand the role of leadership in data quality management at health facilities
 Describe determinants and possible solutions to data quality

INTRODUCTION
Decisions are part of the day-to-day activities of health workers and managers. Managers at different
levels of the health system make decisions as part of their planning, implementation, monitoring and
evaluation of programs. Similarly, healthcare providers make decisions during diagnosis, treatment,
and follow-up of their clients and patients. Data is one of the most important inputs that all health
workers and managers use in making these decisions.

Availability of high quality HIV/AIDS data is at the heart of a functioning evidence informed decision
making in HIV/AIDS program. Quality data leads to better clinical and health program decisions re-
lated to HIV/AIDS program that results in better health outcomes for patients and health facilities.
Health facilities are expected to maintain the quality of data generated and reported in the facility so
that evidence informed planning, resource allocation and other decisions can properly be done.

In this section, the basic concepts of data quality, data quality problems and determinants of data qual-
ity will be discussed, with a focus on HIV/AIDS program data.

3.1. DATA QUALITY AND ITS IMPORTANCE

What is data quality?

Data quality is often defined as “fitness for use.”

What does it mean when we say “fitness for use”?

 Data are fit for their intended uses in operations, decision-making, and planning.
 Data reflect real value or true performance.
 Data meet reasonable standards when checked against criteria for quality.

In general, terms, quality data represent what was intended or defined by their official
source, are objective, unbiased and comply with known standards.

22 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Importance of Data Quality
Good quality health is dependent on the access to and use of good quality data. Good quality data is
important for both individual clients/patients and organizations.

For patients/Clients
 HIV/AIDS service users are more likely to receive better and safer care if healthcare professionals
have access to accurate and reliable data to support decision making process and for continuing and
future care of patients. Accurate and reliable patient data, such as results of investigations (viral
load, TB screening etc.), information on allergies, past medical history, potential drug interactions,
treatment adherence when readily accessible to the healthcare professionals supports provision of
quality health- care services.

 HIV/AIDS Service users are more likely to receive better care if healthcare performance data is used
regularly to support HIV/AIDS service quality improvement that should reflect actual performance.

For Healthcare Organizations

 Quality data can support healthcare organizations to institute quality improvement initiatives based
on performance measurement for HIV/AIDS program.

 Healthcare organizations can more effectively and efficiently plan and provide for HIV/AIDS service
users’ needs if the data used to support decision-making is of high quality. For example, good quality
viral load result data that highlights low viral load suppression rate will notify the organizations to
review and take actions in improving the quality of HIV service.

Table 1. Intended users and uses of quality health care data

Users of healthcare data Intended/potential use

 Be informed about own health

Patients  For medico legal information
 Make decisions on which health facility to visit

 Investigate and treat syndromes based on local disease patterns

 Consider patient’s past medical histories in current management of patients
Healthcare providers
 Take appropriate performance/ quality improvement actions as individual
and team

 Identify and communicate priority health problems of catchment populations

 For need based resource allocation (human, drug, finance, …)
 Take appropriate performance/quality improvement actions
 Set appropriate targets during planning
 Design effective strategies to improve access and quality of healthcare to
Healthcare managers catchment populations
 Provide relevant and quality support to supervisees
 Request relevant support from supervisors
 Advocate for more investment on effective strategies
 Mobilize communities to address local health problems
 Identify priorities for program evaluations

PA RT I C I PA N T ’ s M a n ua l 23
3.2. LEADERSHIP IN DATA QUALITY
To ensure the quality of data at different levels in the health system, health workers and managers at
all levels of the health system have roles and responsibilities. To ensure better HIV/AIDS quality data
all health workers and managers at each level should convey their role and responsibilities.

What are the roles of data clerks, HITs/M&E officers, Program Managers, Health Care providers
in ensuring HIV/AIDS data quality?

Discuss in groups

Some of the roles and responsibilities of each health cadres working on HIV/AIDS program are as fol-
lows. [Please note that this is not the exhaustive list and consider as an initial list for further discussion]

Data clerks
 Collect and record HIV/AIDS data elements into EMR-ART and registers
 Summarize patient data and check quality of registers and EMR-ART
 Submit and complete summary reports on time
 Routinely analyze and share HIV/AIDS data
 Conduct quality checks/data quality assessment/verification

HITs/HMIS/M&E Officers
 Collect data, review reports collected and submit aggregated reports
 Ensure timeliness and completeness of reporting
 Monitor quality of data captured and reported
 Conduct routine data quality assessments
 Routinely analyze and share data

Health Care Providers and Program Managers

 Follow the proper recording and reporting practice at each service delivery points
 Monitor quality of data captured and reported
 Ensure data quality is a regular agenda at performance monitoring meetings
 Conduct routine supervisory visits to HIV/AIDS units
 Routinely analyze and use data

Note: Maintaining data quality is not the responsibility of only some groups. EVERYONE HAS A ROLE
AND IS RESPONSIBLE!!!

3.3. DETERMINANTS OF DATA QUALITY

Similar to the other health data, HIV/AIDS data quality can be affected by different factors across
system level. The factors can be categorized into three groups, namely; Technical determinants, be-
havioral and organizational determinants, as described below:

Technical determinants
 Lack of guidelines to fill out the data sources and reporting forms
 Non-standardized data collection and reporting forms
 Complex design of data collection and reporting tools
 Lack of electronic system to capture, store, analyze and share HIV/AIDS data

24 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Behavioral determinants
 Personnel not trained in the use of data sources & reporting forms
 Misunderstanding of how to compile data, use tally sheets, and prepare reports
 Mathematical errors occur during data capturing, consolidation from data sources, affecting report
preparation
 Attitude towards the importance of HIV/AIDS data

Organizational determinants
 Lack of a reviewing process, before report submission to the next high level
 Organization reports exaggerated data intending incentivizes for high performance\
 Absence of culture of information use

Possible Solutions to Problems of Data Quality

 Avail standardized and simplified HIV/AIDS guidelines for recording and reporting formats across
the health system
 Standardize and simplify HIV/AIDS recording and reporting formats
 Design and implement HIV/AIDS data collection and reporting tools that are simple and easy to use
 Build capacity of health work force from data generation to information use
 Staffing of health institutions with necessary skilled human power to support the HIS
 Strengthen the Performance Monitoring Team (PMT) at each level of the health system
 Improve awareness among HIV/AIDS care workers and program managers on the intention and
value of rewards and incentives
 Enhance culture of information use at each level of health system
 Integration and institutionalization of health data by voiding using multiple data collection tools

PA RT I C I PA N T ’ s M a n ua l 25
SECTION 4:
Health Data Quality Dimensions
SESSION OBJECTIVES
At the end of this Section, participants will be able to:
 Understand the concept of data quality dimensions
 Describe the different dimensions of data quality
 Explain how data quality dimensions are measured

4.1. Introduction to Data Quality Dimensions

All HIV/AIDS service providers, including data clerks, health professionals, program managers and
health information managers, need to gain a thorough knowledge and understanding of the key com-
ponents of data quality and the requirements so that continuous data improvement can be done. Data
quality attributes (dimensions) may be described in different ways in different references but in most
cases, the following list can summarize the most common data quality dimension.

Major Dimensions of data quality

 Completeness
 Timeliness
 Consistency
 Accuracy and Validity

Other Dimensions
 Legibility
 Accessibility
 Confidentiality
 Precision
 Integrity
 Relevance

Each dimension of data quality and its metrics will be discussed in the next session

4.2. METRICS OF DATA QUALITY DIMENSIONS

4.2.1. Completeness
For a data to be considered complete, all required data should be present and the medical record
should contain all pertinent documents with complete and appropriate documentation.
Completeness examines the extent to which:
 Data reported through the system are available and adequate for the intended purpose
 Data elements in individual records and reports are complete
 All entities that are supposed to report are actually reporting

Completeness should therefore include complete data recording in both data recording tools and in
reporting formats.

26 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 4.2.1.1. Data Completeness on recoding tools (Individual cards/forms, registers)
This refers all the necessary data elements on patient cards/forms and registers should be filled imme-
diately after provision of the service by the care provider.
In HIV program, the following individual level data has to be completed:

 The cover page of integrated individual folder should contain all the necessary identifying data to
uniquely identify an individual patient or client.
 ART intake form A and intake form B has to be completed (all the data elements should be
completed)
 All the data elements in the VCT, PITC, ART and PMTCT registers has to be completed for each
individual patient
Details of checking ART patient card/form completeness based on compulsory data elements will be
discussed in the next section in this manual (Section 5).

4.2.1.2. Data completeness on reporting formats

This refers to the extent to which the data elements in the reporting formats are completely filled for
all the data elements in the reporting form. Health facilities are expected to fill a zero value in the re-
porting form even if the event does not happen in a defined reporting period.

Metrics of data completeness on reporting formats

Completeness of data (%) = # values entered (not missing) in the report

# Total data elements in the report

Completeness of reports (%) = # reports that are complete (all data elements filled out)
# Total reports available or received

4.2.1.3. Report Completeness

This assesses or measures whether all the entities/reporting units, which are supposed to report, ac-
tually do so. This helps to examine the total reports received from all health facilities from the total
reports expected for a given period. Example: If a Woreda Health Office expects 5 monthly service
reports from 5 PHCUs, all the 5 PHCUs should report on a monthly basis to make reporting complete-
ness 100%.

Metric of report completeness

Report completeness (%) = # total reports available or received in a given period

# Total reports expected with a given period

4.2.2. Timeliness
This dimension is about timely availability of data. Information, especially clinical information of pa-
tients, should be documented as soon as the event occurs, treatment is provided or results noted.

PA RT I C I PA N T ’ s M a n ua l 27
Delaying documentation could cause information to be omitted and errors recorded.
Is your information available right when it is needed? That data quality dimension is called timeliness.

Example of timeliness
 A patient’s socio-demographic and other identifying information is recorded at the time of first
attendance and is readily available to identify the patient at any given time
 The patient’s past medical history, a history of the present illness/problem as detailed by the patient,
and results of physical examination, and lab examination reports are recorded at the time of
attendance
 All expected reports are ready within a specified period, having been checked, verified and sent to
the next level with in a due date.

Report Timeliness (%) = # reports submitted or received on time

# total reports available or received

Note: All health facilities and administrative health units should have timeliness and completeness
tracking logbook. If the facilities have electronic version of report tracking mechanism, they should
use that one and keep the print-out as a record.

4.2.3. Accuracy and Validity

Accuracy is the degree of conformity of a measure to a true value. Accurate data are considered cor-
rect: the data measures what they are intended to measure. Accurate data minimize error (e.g., re-
cording or interviewer bias, transcription error, sampling error) to a point of being negligible.

The original data must be accurate in order to be useful. If data are not accurate, then wrong impres-
sions and information are being conveyed to the user. Documentation should reflect the event as it
actually happened. Recording data is subject to human error and steps must be taken to ensure that
errors do not occur or, if they do occur, are picked up immediately.

Examples of accuracy in HIV/AIDS program:

 The patient’s identification details in the patient folder, intake form A and B should correct and
uniquely identify the patient (Example: The name, sex, address of PLHIV who are on ART should
correctly be filled)
 All relevant facts pertaining to the episode of care are accurately recorded (Eg. record the exact ART
regimen provided to the patient)
 All patient/client records (cards, forms) in the integrated individual folder are for the same patient.
 The patient’s address on the record is what the patient says it is.
 Documentation of clinical services in a health facility is of an acceptable predetermined value
 The vital signs are what were originally recorded and are within acceptable value parameters, which
have been predetermined and the entry meets this value
In a computerized system, such as EMR-ART database, the software can be programmed to check
specific fields for validity and alert the user to a potential data collection error. Example: EMR-ART
software has inbuilt validation rules and checks to ensure HIV/AIDS data quality. The details of the
validation rules in EMR-ART will be dealt later in this manual.
Examples of validity in EMR-ART

28 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 A patient must have a unique number and if a duplicate ART number is provided, the system will
send a pop up message
 The date a patient started ART treatment must be the same as or later than the date the patient is
tested positive for HIV

4.2.4. Consistency (Reliability)

Consistency is the level of agreement between source document and reported data. Data should yield
the same results on repeated collection, processing, storing and display of information. In other words,
data should be consistent/reliable on repeated measurement. It is the absence of difference, when
comparing two or more representations of a thing against a definition.
For facility level HIV/AIDS program, internal consistency can be checked by examining:
 Consistency between related indicators: The program indicator compared to other indicators with
which they have a predicable relationship, to determine whether the expected relationship exists
between the two indicators
 Consistency over time: Trends in reporting over time, to identify extreme or implausible values time
to time or year-to-year
 Presence of outliers: This examines if a data value in a series of values is extreme in relation to the
other values in the series
 Consistency of reported data and original records: This examines the consistency of reporting of
selected indicators, by reviewing source documents. It compares consistency between the original
record and the report.
We will learn each of the above four internal consistency checks for HIV/AIDS program data quality
at facility level.

4.2.4.1. Consistency between related indicators

This examines the consistency of two or more related indicators that have a logic relation in an HIV/
AIDS program report. HIV/AIDS indicators, which have a predictable relationship, are examined to
determine whether the expected relationship exists between those indicators. In other words, this
process examines whether the observed relationship between the indicators, as depicted in the re-
ported data, is as expected or not.

Example: Let’s see a monthly report of the following data elements by a facility

 Number of adult and pediatric ART patients who were tested for viral load in the reporting period
= 400
 Total number of adult and pediatric ART patients with an undetectable viral load (<1000copies/ml)
in the reporting period = 420

From this report, you can see that the total PLHIV on ART that have viral suppression is greater than
total tested for viral load. This is not logically correct and shows that there is a data quality issue. It
needs to be corrected before it is submitted to the next level.

4.2.4.2. Consistency over time

This examines trends in reporting over time, to identify extreme or implausible values from time to
time. The plausibility of reported results for selected HIV/AIDS program indicators is examined in
terms of the history of reporting of the indicators. For national and sub-national (Region and Woreda

PA RT I C I PA N T ’ s M a n ua l 29
levels), there is a calculation to be used for “Consistency over time” [Refer to the national data quality
Manual]. For facility level, we examine the reporting trend and see whether the reported data is plau-
sible or not.
Let us see the “Currently on ART report” of facility X for the last one year and the current (Tir 2012)
report of the facility.

Month No. of PLHIVs currently on ART

Tir 2012 EFY 650
Yekatit 2012 EFY 655
Megabit 2012 EFY 655
Miazia 2012 EFY 660
Ginbot 2012 EFY 658
Sene 2012 EFY 664
Hamle 2012 EFY 668
Nehassie 2012 EFY 670
Meskerem 2013 EFY 670
Tkimt 2013 EFY 675
Hidar 2013 EFY 677
Tahsas 2013 EFY 680
Tir 2013 EFY 900

From the above example, you can see that currently on ART report in Tir 2013 EFY is significantly
higher than the previous month’s report. In the last 12 months, there was a monthly increment by 5
to 10 PLHIV. However, in the month Tir 2013, the number of PLHIV who currently are receiving ART
has jumped to 900, an increase by 220 in one month. This shows that there is a data quality problem
for this data element and the cause need to be investigated and corrected before reporting to the next
level.
You may also compute the mean for the last 12 months and compare with the current month report. In
the above example, the Mean for the last 12 months is 665 (Sum from Tir 2012 to Tahsas 2013 divided
by 12). Divide the current month report (900) by the mean (665) and you will get 1.353. This means
that the current month report is increased by 35.3% compared to the mean of the last 12 months. This
is a high increment in one month for Tx_CURR and the facility should check the current month report
and make corrections accordingly. [NOTE: For Woreda and above levels, such calculations can be used
and the cut-off point is + or – 33%. However, this cut off point is not used to determine whether the
data is consistent or not at facility level.
4.2.4.3. Consistency of reported data and original records
This measures whether there is consistency between the reported data and the original patient re-
cord (patient cards and registers). This requires reviewing source documents by taking randomly se-
lected patient records. Method of consistency check of reported data with original records will be
done by randomly taking patient charts and comparing against registers and reports. The details will
be discussed in the “data quality assurance” section of this manual.

4.2.4.4. Presence of Outliers

This examines whether a data value in a series is extreme in relation to the other values in the series.

30 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Values that looks deviant should be investigated for data quality. [For admin levels, i.e WoHO and
above levels, there is a standard calculation for outlier detection, but for facility level, do it just by
eyeballing]

4.2.5. Legibility
This dimension of data quality is about whether all data is written, transcribed and/or printed are
readable. If abbreviations are used, it should be based on standards.
EXAMPLES OF LEGIBILITY:
 Handwritten demographic data on individual patient folder are clearly written and readable.
 Handwritten notes on individual medical records and registers should be clear, concise, readable
and understandable.
 Handwritten Intake Forms (A and B) and Follow up Form should be clear and easily understandable
to transcribe in to ART Registers and EMR-ART data base

4.2.6. Accessibility
Accessibility – data are available to authorized persons when and where needed; All necessary data
are available when needed for patient care and for all other official purposes. If data is not accessible
when needed, its value is lost even if it is accurately recorded.

EXAMPLES OF ACCESSIBILITY:
 HIV Patients’ medical records are available when and where needed at all times, example when a
patient come for ART follow up
 Abstracted HIV/AIDS data are available for review when and where needed (Registers)
 In an electronic patient record system, clinical information is readily available when needed
 HIV/AIDS reports are accessible when required for Performance monitoring team, planning
meetings and government requirements or for any official need

4.2.7. Precision
This means that the data have sufficient detail.
For example, “Number of PLHIV who are currently on ART” report requires disaggregation by regimen
type, sex and age. An information system lacks precision if it is not designed to record the sex of the
individual who are currently receiving ART.

4.2.8. Confidentiality
Confidentiality means that clients are assured that their data will be maintained according to national
and/or international standards for data. This means that personal data are not disclosed inappropri-
ately, and that data in hard copy and electronic form are treated with appropriate levels of security
(kept in locked cabinets and in password-protected files). This is especially important particularly to
patients and in legal matters.

4.2.9. Integrity
Integrity is the quality of being honest and having strong moral principles or moral uprightness. Data
Integrity can be considered as a polar opposite to data corruption that renders the information as in-
effective in fulfilling desired data requirements.
Data integrity aims to prevent unintentional changes to information. It is not to be confused with data
security, the discipline of protecting data from unauthorized parties. It also aims to prevent uninten-
tional changes to information. Data have integrity when the systems used to generate them are pro-
tected from deliberate bias or manipulation for political or personal reasons.
PA RT I C I PA N T ’ s M a n ua l 31
SECTION 5:
Data Quality Assurance
OBJECTIVES
At the end of this section, participants will be able to:
 Define data quality assurance
 Understand the importance of data quality assurance
 List the different types of data quality assurance techniques
 Understand and apply data quality assurance to check consistency of data in HIV/AIDS program

5.1. DATA QUALITY ASSURANCE AND ITS IMPORTANCE

Data quality assurance is:

 An explicit combination of methods and activities that are carried out for the purpose of reaching
and maintaining high levels of data quality
 A critical factor for generating and sustaining high-quality data
 A system to ensure that data are collected, maintained, monitored, transformed into useful
information, and interpreted in ways that maintain high quality for all users

Data quality assurance is critical for health information systems. It helps to improve data quality by
uncovering hidden problems in data collection, aggregation, and transmission of priority indicator/
data. Knowing about these problems allows health professionals and managers to develop data quali-
ty improvement plan. It also ensures accountability and reporting.

5.2. TECHNIQUES OF DATA QUALITY ASSURANCE

There are different techniques that can be used to perform data quality assurance. These techniques
may be used at facility and administrative health unit levels to determine the level of data quality and
to take corrective measures. The methods used at facility level and administrative level may be differ-
ent. Some of the methods are:

1. Facility level data quality assurance techniques: These are techniques that health facilities conduct
by themselves. It includes:
 Visual scanning/eyeballing
 Data quality check using LQAS (Lot Quality Assurance Sampling) methodology
 Consistency check between medical records, registers/EMR-ART and reports
 Others

2. Administrative level data quality assurance techniques: These techniques are used by administra-
tive levels such as Woreda Health Offices, Zonal Health Departments, Regional Health Bureaus and
Ministry of Health. It includes:
 Data quality Desk review
 Routine Data Quality Assessment (RDQA)
 Data Quality Audit (DQA)
 Performance of Routine Information System Management (PRISM)
 Visual scanning
 Others

32 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
In this training manual, we will discuss on facility level data quality assurance methods, especially fo-
cusing on HIV/AIDS data quality assurance. We will specifically discuss the details of a data quality
check/assessment that aims at assessing quality of HIV/AIDS medical records and its consistency with
ART registers/EMR-ART.

5.2.1. VISUAL SCANNING (EYE BALLING)

This is a simple method used at health facilities to check for consistency of reports before/after con-
ducting data entry. It is about looking across each line of the HIV/AIDS report section to identify miss-
ing data values, unexpected fluctuations beyond maximum/minimum values, inconsistencies between
linked data elements, and for mathematical errors. The data clerk/HMIS officer/M&E expert/health
professional working in the ART unit should visually scan the report before submitting to the HMIS
unit of the health facility.
Example: During visual scanning, the data clerk may find some empty/unfilled data elements, unex-
pectedly high report compared to the previous trend (example: TX_CURR of the facility was 500 in
the previous month and if it is written as 5100 in this month, the data clerk may visually identify the
presence of data entry error). If a third line regimen treatment is not started in the facility and if a third
line regimen is reported, it can easily be identified as a data error by visually scanning the report and
correct it before submission.

5.2.2. LOT QUALITY ASSURANCE SAMPLING (LQAS)

This is a data-quality assurance technique used by health facilities to assess whether the desired level
of data consistency has been achieved by comparing data in relevant summary record forms (i.e., reg-
isters and/or tallies) and the HMIS report. It should be conducted monthly and is based on randomly
selected data elements from a monthly HMIS report.
In this training, LQAS will not be addressed in detail. LQAS will be conducted as part of the whole facil-
ity and HIV/AIDS data elements may be selected randomly during the LQAS process. Specific to HIV/
AIDS care and support related data elements, we will discuss a more detail on data quality assurance
method that includes assuring the quality of individual level medical records as described in the fol-
lowing section.

5.2.3. DATA QUALITY CHECK BETWEEN ORIGINAL RECORDS AND REGISTERS/ EMR-ART
As discussed previously, the major data quality dimensions are completeness, timeliness, consistency,
and validity/accuracy. Determining the level of these dimensions for patients who are receiving ART
is essential. For this purpose, we can use a data quality check on and between original patient records
and registers/EMR-ART. The purpose of this quality check process is to measure the level of com-
pleteness, validity, and consistency of the ART data at health facility level and develop a data quality
improvement action plan based on the findings of the quality check. We named this assurance system
as HIV/AIDS data quality check/assessment and it is developed by ICAP at Columbia University Stra-
tegic Information Project to support ART data quality improvement at health facilities providing ART
service.
Who will conduct the quality check? Facility level ART data quality check/assessment process will be
conducted by a team composed of ART focal person, HITs/M&E officers, and data clerks. It can also be
performed by all relevant stakeholders supporting HIV/AIDS program information system at health
facilities.

PA RT I C I PA N T ’ s M a n ua l 33
Frequency: The frequency of conducting ART data quality check/assessment should be monthly and
should be integrated with the routine performance monitoring of the facility. For very high patient
load facilities (ART patient load of >1000), it has to be conducted at least once per quarter.
Tool to be used for the assessment: To perform data quality check between ART patient cards and reg-
isters/EMR-ART, an excel based data quality check/assessment tool will be used so that assessment
data entry and summarization of the result can be simplified. This tool is primarily designed for use by
all facilities implementing EMR-ART software for ART data management regularly but it can also be
used by facilities that use paper based system only. The excel based tool has four sections:

 Instruction section: This section provides a step-by-step description of the process to be used during
the verification. Operational definition of data elements and data quality dimensions assessed and
measured are also described in this section

 Data quality assessment/Verification Section: This is a section where the assessment data entry and
result is completed.

 Summary sheet/Dashboard: This sheet summarizes the result of the data quality assessment by au
to-filling the result from the “Data Quality assessment/verification” section. This supports the deci-
sion-making process during data quality improvement.

 Action Plan Section: This is a sheet where facilities develop a data quality improvement action plan
based on the findings of the assessment.

Data elements selected for data quality check/verification: The following data elements are selected to
be checked for completeness, validity, and consistency. These data elements are important for ART pa-
tient tracing and follow up. Thirteen data elements are selected for quality check. Even though these
data elements are selected for data quality check of patient cards, it may be changed based on need.

The data elements selected for verification are:

 Age of the patient at entry to ART
 Residence address of the patient: Region, Zone/Subcity, Woreda, Kebele
 ART start date
 Last follow up date
 ART regimen at last visit
 INH start date
 INH completion date
 Viral load result (recent)
 Nutritional assessment result
Assessment procedure
The data quality check/assessment is based on a method of randomly selecting 19 active ART med-
ical records where the completeness and validity of selected data elements on the patient card are
checked. Moreover, the selected data elements will be checked for completeness and validity on regis-
ter/EMR-ART and consistency check between the individual medical records will be done against ART
register and EMR-ART database.
Selection of individual cards for verification: A total of 19 ART patient cards will be selected randomly
for verification monthly. To select the 19 cards, we will use a random selection process from patients
who recently received care and treatment service.

34 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 For hospitals, the sampling frame will be total ART patients that received service a day before the
data quality assessment is being done. If we do verification today, then the 19 patients will be
randomly selected from all patients that have been seen in the previous working day. Total patients
seen in the previous day can be identified by generating a custom report from EMR-ART that can list
all patients that have been seen in the previous day.
 For health centers, the sampling frame will be patients who have received ART service in the last
seven days. Total patients seen in the previous 7 days can be known from EMR-ART custom report.
Steps to conduct data quality check/assessment on ART medical records, ART registers and EMR-ART
database
1. Randomly select 19 sample ART patient cards from the central MRU (Selection will be based on ran-
dom selection process described above). (Note: Please make sure the selected patient cards include
PMTCT and pediatric patients as well).
2. Write the MRN of each card in the first column of the excel based tool
3. For each selected ART patient cards, check the completeness and validity of the data elements de-
scribed above and complete it in the excel data quality check tool.
 For completeness: Mark “completed”, if the data element in the patient card is filled, Mark
“incomplete” if it is not filled and “Not Applicable if the data is not applicable for that specific
data element.
 For validity: Mark “Valid” when the information completed is logical (example: a logical range
of values), appears to be accurate and measures what is intended to measure; mark “invalid”
if the information filled do not have logical relationship and appears inaccurate; mark “Not
Applicable” if it is not applicable. Validity information will be assessed for those variables
that have complete information in the data source (e.g. if the completeness of the “Age at
entry” is ‘Not Completed” then the validity answer is “N/A”)
 Example: If age at entry is recorded as 10, then mark complete and valid. If age at entry is
recorded as 200, mark complete but its validity information will be marked “invalid” since
xage 200 is not logical.
4. Enter data for the ART register and EMR-ART database for completeness and validity for the se-
lected patient cards. In the register and EMR-ART database columns, check the completeness and
validity of the data elements. For the selected patient card, you can find the summary information in
the ART register using the MRN and last visit day, same is true to find the recorded information in the
EMR-ART. Write the result in the excel tool as described in step 3 above.
5. Match all the relevant data elements in the card with the data in the register and EMR-ART and fill
the result in the “consistency” column. Write “Consistent” if there is consistency between what is writ-
ten in the patient card and register or EMR-ART. Otherwise, write “not consistent” if there is no con-
sistency between the two. If it is not applicable for the specific data element, write “Not Applicable”.
6. When the data entry is finished in the “data quality assessment” section of the tool, it will automat-
ically summarize the result for each data element in the “Summary sheet”. Proportion of cards that
have completeness and validity for each data element will appear at the bottom end of the “data quali-
ty assessment” sheet. Moreover, the summary sheet/Dashboard in the assessment tool auto calculate
the values and display the percentage completeness, validity & consistency. It also colors the result as
Green (>95%), Yellow (85-94%), and Red (below 84%).

PA RT I C I PA N T ’ s M a n ua l 35
Note: - It may not be valid to aggregate the finding of multiple facilities to estimate at administrative
level because 19 charts taken randomly from all facilities for may not be adequate or representative
to generalize the finding at higher level.
Please look at separately attached excel verification tool for more information.
Data quality assessment Summary sheet/Dashboard: The data quality check/assessment tool is de-
signed to produce outputs that facilitate analysis and use of the data to understand the current status
of the data quality for selected variables and to develop data quality improvement action plan. When
data entry of the assessment sheet is completed, the tool will auto-calculate the result in the summary
sheet/dashboard as: Green (>95%), Yellow (85-94%), and below 84%.

Patient Chart Registers EMR-ART

S.N Variable
Comp Validity Comp Validity Consistency Comp Validity Consistency
1 Age at Entry 100 100 100 100 100 100 100 100
2 Region 100 76 100 85 100 85 100 85
3 Zone/Sub city 55 90 100 90 100 90 100 90
4 Woreda 100 90 70 90 100 90 100 90
5 Kebele 100 90 100 90 82 90 100 90
6 Phone # 95 90 95 90 95 90 43 90
7 ART Start Date 95 90 95 90 95 90 95 90
8 Last Follow Update 100 100 100 100 100 100 100 100
9 ART regimen isit 43 100 100 100 100 100 100 100
10 INH Start Date 100 100 100 100 100 100 100 100
11 INH Completion Date 100 67 100 67 100 67 100 67
12 VL Load 100 100 100 100 100 100 100 100
13 Nutritional assessment result 95 95 95 95 95 95 95 95

Data Quality Improvement Action Plan

After a data quality assessment is performed, a data quality improvement plan should be developed
based on the result of the assessment. If finding of the data quality assessment on the dashboard show
yellow or red findings in the dashboard/summary sheet, then action plan should be developed with
responsible person and timeliness. The team can use a sample action plan template that is attached
with the DQA tool.

Health facilities should prepare a minute to document the findings of the data quality assessment and
each DQA should be documented in the minute book.

Table 3. Health Center /Hospital Data Quality Assessment Action Plan template

S. N Problem Identified Proposed Solution Responsible Person Timeline Follow Up

1
2
3
3
4
5

36 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
5.2.4. USING EMR-ART TO IMPROVE DATA QUALITY
EMR-ART has several features that can help improve HIV/AIDS data quality. It has validation rules
during data entry to make sure data is captured on the right format and within a reasonable range,
user-defined validation rules based on mathematical relationships between the data being captured
(e.g., subtotals vs totals), as well as reports on data coverage and completeness. More indirectly, sev-
eral of the EMR-ART design principles contribute to improving data quality, such as auto calculating of
nutritional status when height and weight data is entered. In addition, it categorizes viral load results
to “suppressed” and “unsuppressed” based on the value entered.

DATA INPUT VALIDATIONS

The most basic way of data quality assurance system in EMR-ART is to make sure that the data being
captured is on the correct format. EMR-ART will give the users a message that the value entered is
not on the correct format and will not save the value until it has been changed to an accepted value.
E.g. When we register a patient without grandfather name, age or sex, a Pop-Up message appears
and do not allow saving of that specific patient unless the required data elements are entered. EMR-
ART validation rule is based on an expression, which defines a relationship between a number of data
elements. The expression forms a condition, which should assert that certain logical criteria are met.
To stop typing mistakes during data entry, the Smart Care ART Software checks that the value be-
ing entered is within a giver range. For Example, phone Number Contains 10 digits, and the Smart
Care doesn’t save data with a phone number with a digit greater than or less than 10. It also validates
Unique ART Number and don’t save if it is above 12 digits. The validation rules checks are also built
into the data entry process so that when the user completed a form, the rules can be run to check the
data in that form only before closing the form.

Figure 2. EMR-ART validation rule (Example for grandfather name)

PA RT I C I PA N T ’ s M a n ua l 37
Figure 3. EMR-ART validation rule (example for gender of a patient)

PATIENT CHART COMPLETENESS

The system calculates completeness of data elements based on a set of predefined compulsory data
elements recorded on patient’s intake and follow-up forms and will display the completeness status
for each form using color coded naming. Though this labeling is not colored it clearly shows in text la-
beling which is comparable with color coded stickers used in manual data completeness improvement
campaigns.

Figure 4. EMR-ART data completeness report sample

38 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
HIV/AIDS INFORMATION USE
SECTION 6:
Basic concepts of data, data demand and information use
SECTION OBJECTIVES
At the end of this session, participants will able to:
 Understand basic concepts of data, data demand and information use
 Understand use of information at different level
 Identify determinants of data demand and use

6.1. BASIC CONCEPTS OF DATA, DATA DEMAND AND INFORMATION USE

In this section, participants will be able to understand the definitions and basics of data related terms
and concepts

Variable: - The characteristics/attributes of a person, object, or event that can take on different values.
Example: Age and sex of PLHIVs who are currently on ART

Data: It is the raw facts that are collected and form the basis for what we know. It is raw, unorganized
facts that need to be processed.
Example: VCT counselling and testing record of individuals of varying age, sex, population groups and
residence

Information: - The product/result of transforming or processing data to reveal a meaning. It is when

data is processed, organized, structured or presented in a given context so as to make it useful.
Example: The total number of male VCT clients who were tested positive for HIV in 2011 EFY in Ze-
wditu hospital was 100.

Indicators: Indicators are variables that measure one aspect of a health intervention, program, or proj-
ect. They are clues, signs, and markers as to how close we are to our path and how much things are
changing.
Example: Proportion of PLHIV on ART with suppressed viral load
Data demand: The value stakeholders (policy makers, HIV program managers, health workers, part-
ners, etc.) attach to data regardless of the use of data.

Information/Data use: The process through which decisions makers and stakeholders explicitly con-
sider information in one or more steps of the process of policy making, program planning and manage-
ment, or service provision.

In practice, it is difficult to distinguish between data demand and information use and one may choose
to treat them as parts of a single process. However, Demand is a concept distinct from Use.

PA RT I C I PA N T ’ s M a n ua l 39
HIV Data Demand HIV Information Use
The term HIV data demand is related to the value stakehold- The term HIV information use refers to the use
ers attach to data regardless of the use of HIV data. We can of HIV data in the decision-making process. A
say that HIV data demand exists if: decision maker uses HIV information if he/she:

 HIV specific questions are raised, and data are considered  Is aware of the decision to make or question
to answer them and/or to be answered
 HIV data are required when a decision needs to be made  Relevant HIV information is explicitly
 The decision –maker understands and proactively seeks considered in the decision-making process,
HIV information to make evidence-based decision even if the quality of HIV data is sub-optimal.

6.2. INFORMATION REQUIREMENT AND USE AT DIFFERENT LEVELS

HIV/AIDS data is required at individual level, community level, facility level, and population level to
provide individual level care and program level management.

 Patient level: At patient level, different individual level data is required to assess the health care
needs of individuals, provide treatment and counselling services. It is the basis for clinical decision
making for patients/clients during management of HIV care such as ART provision, adherence coun-
seling, management of opportunity infections, TB screening and treatment, IPT provision, nutritional
screening, and management, STI screening, MHI screening, cervical cancer screening, IPV, family plan-
ning counseling and provision, routine viral load monitoring. The table below shows summary of indi-
vidual level data required to be monitored for PLHIV who are on care and support. (See Table 4 below)

 Community level: At community level, health extension workers identify individual patients/clients
and refer to the next level, conduct HIV/AIDS related outreach activities (lost to follow up tracing, part-
ner and family-based index case testing referral), follow up on referral, conduct community awareness
activities on HIV etc.

 Facility level – Health facilities need HIV information on the progress of HIV/AIDS prevention, test-
ing, care and support related activities so that evidence informed planning, resource allocation, and
HIV/AIDS program monitoring and decision making can be done. (See Table 5 below)

 Administrative/ Program levels – Need information on HIV service coverage, incidence and preva-
lence of HIV, resource availability for planning, HIV policy formulation, HIV performance review and
improvement, designing interventions, developing strategies, capacity building, HIV case-based sur-
veillance, research, Innovations and Monitoring and evaluation.

40 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Table 4. Patient level HIV/AIDS indicators for individual level HIV/AIDS information use

S. No Indicator name Description Data source Review period

Time gap between HIV positivity Number of Days between detection and ART EMR-ART, Patient Monthly
1
and ART start date start date card
Blood sample collection and referral to test- EMR-ART, Patient Monthly
Viral load (VL) testing for eligible
2 ing site at 3, 6, 12 months and every year af- card and ART register
patients
ter initiation
Time period between blood sample collec- EMR-ART, Patient Monthly
3 VL test result reception and TAT
tion and viral load test result card

VL suppression status of individual VL result >= 1000/ml or < 1000/ml EMR-ART, Patient Monthly
4 card and ART register
patients

6 months TB Prophylaxis completion (start- EMR-ART, Patient Monthly

IPT/TPT completion status of indi-
5 ed, completed, stopped, never taken) or his- card and ART register
vidual patient
tory of previous TB treatment
A patient who missed follow up for 31-60 EMR-ART, Patient Monthly
6 Patient first lost to follow up
days to pick ARV card and ART register
A patient who missed follow up for 61-90 EMR-ART, Patient Monthly
7 Patient Second lost to follow up
days to pick ARV card and ART register
A patient who missed follow up for more EMR-ART, Patient Monthly
8 Drop from ART treatment
than 90 days to pick ARV card and ART register
Level of adherence to ARV treatment as: EMR-ART, Patient Monthly
9 Adherence to ART treatment
good, fair or poor adherence to ARV card
Status of index case contact disclosure and EMR-ART, patient Monthly
Index case contact disclosure and
10 testing status card Index case reg-
testing status
ister
Regimen switching to another regimen line EMR-ART, Patient Monthly
ARV regimen switching and substi- or substitution to another regimen with- card and ART register
11
tution status in the same regimen line and reasons for
switching and substitution
Assessment of patients for ASM and coun- EMR-ART, patient Monthly
Status of patient assessment, coun-
selling for ASM card and ASM register
12 selling, enrolment and termination
Enrolment to ASM
from ASM
Termination from ASM
Individual schedule visits for rou- Schedule visit date EMR-ART, patient Monthly
13
tine follow up, VL testing, EAC etc card
Monthly TB screening for individual TB Screening status during a monthly follow EMR-ART, Patient Monthly
14
PLHIVs up card and ART register
Individual nutritional status of Nutritional screening and its result as Nor- EMR-ART, Patient Monthly
15
PLHIVs mal, mild, MAM, or SAM card and ART register
Provision of therapeutic or supplementary EMR-ART, Patient Monthly
Nutritional support to undernour-
16 food for PLHIVs that are found to be clinical- card and ART register
ished PLHIVs
ly undernourished
Started, completed EMR-ART, Patient Monthly
17 Co-trimoxazole completion status
card and ART register
Family Planning for non-pregnant Monthly use of contraceptives for non-preg- EMR-ART, Patient Monthly
18
PLHIV women nant women receiving ART (15-49) card and ART register
Completeness of demographic in- Full demographic information of patient data Individual folder Monthly
19
formation in patient chart

PA RT I C I PA N T ’ s M a n ua l 41
Table 5. Aggregate level HIV/AIDS data elements/indicators for HIV/AIDS program data use

S.no Indicator name Description Data source Review period

Number of PLHIV who are currently taking EMR-ART, ART and Monthly
1 Number of PLHIV currently on ART ART PMTCT Register,
DHIS2
Percentage of patients on ART with a sup- EMR-ART, ART and Monthly
2 Viral load suppression rate pressed viral load (<1000 copies/ml) in the PMTCT Register,
past 12 months DHIS2
Percentage of ART patients with an unde- EMR-ART, ART and Monthly
3 Early viral load suppression rate tectable viral load at 6 month after initiation PMTCT Register,
of ART DHIS2
Number of adults and children newly started EMR-ART, ART and Monthly
Number of adults and children with
4 on antiretroviral therapy (ART) PMTCT Register,
HIV infection newly started on ART
DHIS2
The proportion of individuals receiving ther- EMR-ART, ART and Monthly
Proportion of clinically undernour-
apeutic or supplementary food among those PMTCT Register, Clin-
5 ished PLHIV on ART who received
whose nutritional status was assessed and ical care tally sheet,
therapeutic or supplementary food
found to be undernourished DHIS2
Percentage of adults and children known to DHIS2, EMR-ART, Monthly
6 ART retention rate be on treatment 12 months after initiation ART Register, PMTCT
of AR register
Percentage of non-pregnant women living DHIS2, EMR-ART, Monthly
Percentage of non-pregnant wom-
with HIV on ART using a modern family plan- ART register, PMTCT
7 en living with HIV on ART using a
ning method register, Clinical care
modern family planning method
tally sheet
Proportion of STI cases tested for HIV in the DHIS2, EMR-ART, Monthly
Proportion of Sexually Transmitted
8 reporting period PICT Tally, OPD and
Infection (STI) cases tested for HIV
IPD registers
The proportion of patients on ART who DHIS2, EMR-ART, Monthly
TB Screening for HIV positive Cli-
9 were screened for TB during the reporting ART register, PMTCT
ents
period register
Latent TB infection (LTBI) treat- Proportion of newly enrolled HIV-positive DHIS2, EMR-ART, Monthly
10 ment for HIV positive clients newly people started on LTBI treatment during the ART register, PMTCT
enrolled to care reporting period register
Number of HIV-positive TB patients who are DHIS2, EMR-ART, Monthly
Anti-Retroviral Therapy (ART) for started on or continue previously initiated Unit TB register
11
HIV positive TB patients ART during their TB treatment, expressed as
a proportion of all HIV-positive TB patients
Percentage of HIV-positive pregnant wom- DHIS2, EMR-ART, Monthly
en who received ART to reduce the risk of PMTCT Register
Percentage of HIV-positive preg- motherto child-transmission (MTCT) during
nant women who received ART pregnancy, L&D and postnatal. It includes
to reduce the risk of mother-to number of HIV positive pregnant, laboring
12
child-transmission during pregnan- and lactating women who received ART at
cy, labor & delivery (L&D) and post- ANC+L&D+PNC for the first time and HIV
partum positive pregnant, laboring and lactating
women who get pregnant while on ART and
linked to ANC
Percentage of infants born to HIV-positive DHIS2, EMR-ART, Monthly
Proportion of HIV exposed infants
13 women who received a virological (DNA/ PMTCT Register
with virological test
PCR) HIV test within 12 months of birth

42 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 Percentage of infants born to Percentage of infants born to HIV-positive DHIS2, EMR-ART, Monthly
HIV-infected women who were women who started on co-trimoxazole pro- PMTCT Register
14
started on co-trimoxazole prophy- phylaxis within two months of birth
laxis within two months of birth
Percentage of infants born to Percentage of infants born to HIV positive DHIS2, EMR-ART, Monthly
HIV-infected women receiving women who received ARV prophylaxis to PMTCT Register
15 antiretroviral (ARV) prophylaxis reduce risk of mother-to-child transmission.
for prevention of mother-to-child
transmission (PMTCT)
Percentage of partners of preg- Percentage of partners of pregnant, laboring DHIS2, EMR-ART, Monthly
nant, laboring and lactating women and lactating women tested for HIV during ANC, Delivery and
16
tested for HIV during the reporting the reporting month PNC registers
month
Percentage of HIV exposed infants Percentage of HIV exposed infants tested DHIS2, EMR-ART, Monthly
17 receiving HIV confirmatory (anti- and confirmed HIV status at 18 months by PMTCT Register
body test) test by 18 months rapid antibody test

PA RT I C I PA N T ’ s M a n ua l 43
SECTION 7:
Extraction/ Generation of Patient Level Data from EMR-ART
OBJECTIVE
At the end of this session, participants will be able to:
 Describe the steps to be followed by users to generate and export all kinds of reports from
EMR-ART for reporting, documentation, and analysis purposes.
 Generate other clinical service reports for clinical service improvement at facility level Enhanced
EMR-ART as a comprehensive data management tool that captures patient level data for patients
who are receiving ART. It has features that helps to improve data use for patient and program level
HIV/AIDS decision making. Two types of reports can be generated from this system: Predefined
Reports and Custom Reports.

Before the details of data analysis, presentation and use for decision making, we will discuss briefly
on how we can generate data from the enhanced EMR-ART and make data ready for further analysis.

7.1. GENERATING AND EXPORTING PREDEFINED & CUSTOM REPORTS

GENERATING PREDEFINED REPORTS

Four types of predefined reports are included in EMR-ART, namely

A. HMIS/DHIS2.3 Report
B. DATIM Report
C. Line List Reports
D. Cohort Report

Steps to Generate and Export HMIS/DHIS 2.3 Report

1. From the home page of EMR-ART click on “Report and Analytics”

2. Click on the (+) (Plus Sign) on Aggregated Reports
3. Then from drop down click on “HMIS/DHIS2”
4. Enter the appropriate period using calendar or use the month and year icons above
5. Wait till the HMIS/DHIS2 report displays
6. To print the report, click the “Print” icon at the left upper corner of the page and keep the copy of the
report in the facility for documentation
7. To Export to excel click on “Export to Excel” icon & rename save in folder used for data analysis

44 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Figure 5. HMIS/DHIS2 report generated from EMR-ART

Steps to Generate and Export DATIM 2.4 Report

1. From the home page of EMR-ART click on “Report and Analytics”

2. Click on the (+) (Plus Sign) on Aggregated Reports
3. Then from drop down click on “DATIM”
4. Then from drop down click on the indicator category intended and the specific indicator as
numerator or denominator
5. Enter the appropriate period using calendar by defining the Indicator’s reporting period. Use
calendar to enter reporting start date and end date & click show report in the left upper corner of
the page
6. To print the report, click the “Print” icon at the left upper corner of the page and keep the copy of the
report in the facility for documentation
7. To Export to multiple file formats like (Excel, Word, PDF…) click on “Inverted Arrow” at the left
upper corner of the page & rename save in the folder used for data analysis

Figure 6. DATIM report generation from EMR-ART

Note: if the generate DATIM report content in more than one page, please don’t forget to click on
“Go to next page” icon in the upper part of the page

PA RT I C I PA N T ’ s M a n ua l 45
Steps to Generate and Export Line List Reports
1. From the home page click the “Report and Analytics”
2. Click on the report sub-icon
3. Then from the opening page select the type of line list report on the left side
a. Click on the (+) (Plus Sign) on Line List Reports
b. Then click on the required line list report from the drop-down
4. Enter appropriate time period using calendar and enter filtering criteria if applicable and necessary
5. Click shows reports and wait till the report is loaded
6. Click on “Print/Export” icon at the left upper corner of the page to print the report
7. In the displayed dialogue box select the file type (Word, PDF, Excel...) and click Save

Figure 7. Line list report generation and exporting from EMR-ART

Steps to Generate and Export Cohort Report

1. From the home page of EMR-ART click on “Report and Analytics”
2. Click on the (+) (Plus Sign) on “Cohort Report” & from drop down click on “Cohort Report”
3. In the Cohort selector dialogue box, select “Cohort Year” and click show chart
4. To export all the clients line list report from the cohort aggregate report, select the “Export Data”
in the box
5. Finally, the line list will be exported in excel format and aggregate report will be displayed in the
dialogue box for selection of file types
6. In the dialogue box save as type part, select the file types intended (excel, Pdf…) save

Figure 8. Cohort Report Generation and Exporting from EMR-ART

46 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Generating and Exporting Custom Reports
Use the following steps to generate and export custom report from Smart Care
1. From the home page of EMR-ART click on “Report and Analytics”
2. In the upper section select report type as “Last Follow-Up, Follow at enrollment, Follow-Up on ART
start or patient Info”
3. Click on filtering criteria in the middle of the page
4. On the right upper corner select all the variables intended to be included in the report
5. To Export to Excel click on “Export” at the upper part of the generated grid table

Figure 9. Custom Report Generation and Exporting from EMR-ART

7.2. GENERATING AND EXPORTING OTHER CLINICAL SERVICE REPORTS

TRACING/MISSED APPOINTMENT REPORT
Tracing Report in the EMR-ART has a mission to show all clients with possible appointment but miss-
ing appointment for more than or equal to one day. So, these clients are classified by the software as
missed, 1st Lost, 2nd Lost and Dropped based on the number of days missed from last appointment
date.

Definition of terms
• Missed: - Clients missing 1-29 days
• 1st Lost – clients missed appointment 30-59 days
• 2nd Lost – clients missed appointment 60-89 days
• Dropped – clients missed above or equal to 90 days

Steps to generate and export tracing data

1. Click on the “Treatment and Follow Up” from the home page
2. Click Tracing and click on show/Export report option
3. Rename in the Dialogue box and save the report in excel document

PA RT I C I PA N T ’ s M a n ua l 47
Figure 10. Tracing Line List Report generation and exporting from EMR-ART

SCHEDULED VISIT REPORT

Viewing and managing the appointment date of the client is one of the essential activities needed to
be conducted for proper tracing & making necessary preparations ahead of time before the clients’
appointment date is due. To generate this report, use the following steps,

1. Click on the “Treatment and Follow Up” from the home page
2. Click Scheduled Visit and click on “Show Appointment Report” option
3. In the dialogue box, select the time range to generate the appointment report (Today, Tomorrow,
This Week, This Month and Range)
4. Click on the Inverted Arrow in the left upper corner of the page & select the file type
5. Rename in the Dialogue box and save the report in excel document

Figure 11. Generation and Exporting of Scheduled Visit Report from EMR-ART

PROVIDERS VIEW LINE LIST REPORT (COMPLETENESS CHECK)

Providers View Line List report helps to program monitoring and completeness check for specific vari-
ables in areas of Client Address, TPT, Viral Load, ASM and P&FBICT services. To generate this report,
use the following steps,

1. Click on the “Providers View” from the home page

2. Click Completeness Check and Clinician View List page will appear
3. Click ‘Show Report’ to View all clients status
4. Click on Print/Export in the left upper corner of the page & select the file type
5. Rename in the Dialogue box and save the report in excel document

48 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Figure 12: Providers View Line List Report

DATA QUALITY ASSURANCE REPORT GENERATION

Data Quality assurance feature in EMR-ART is planned to check the label of completeness, Duplica-
tion and consistency of data entered data in the system.
Steps to generate clients list with data quality issues
1. From the home page click “Data Quality Assurance”
2. Select the data quality type and completeness check criteria from the upper section of the page
3. Click search
4. If you want to delete selected client, select, and click “DELETE”
5. To restore already deleted client click on “Restore Deleted Client”
6. To Export already generated list of clients with records of data quality issue,
a. click on the cell at the left upper corner of the listed clients’ table
b. Hold keys “CTRL” and “C” & open new excel file
c. Select a cell in the opened new excel document
d. Finally label and save list of clients and use for intended analysis
Deduplication and solving completeness issue of charts in the facility requires the involvement of both
data clerks and Medical Record Unit clerks with appropriate documentation of changes done.

Figure 13. Data Quality Assurance Data Generation and Exporting from EMR-ART

PA RT I C I PA N T ’ s M a n ua l 49
VIRAL LOAD ELIGIBLE CLIENTS REPORT GENERATION
All eligible clients documented in the EMR-ART can be generated as eligible or not based on previously
documented viral load results in the system and the current date of the computer used for EMR-ART.
Use the following steps to generate line list of these clients:
1. Click on Line List then “Viral Load (+)” sign to expand the Lists
2. In the displayed list “Eligible Clients for VL”
3. Indicate the Eligibility Data from _________ to _________________
4. Select the Status of the client from Drop down as Alive, Restart, Lost or Dropped
5. Click show report and we will get line list of clients
6. Use the same steps mentioned above to export & print list into different file formats
Figure 14. Viral Load Eligible Clients Line List Exported from EMR-ART

Note: - The viral load testing eligible client’s list validity will be determined by the rate at which viral
load results are updated in the EMR-ART from the results received in the paper format. If results are
not being updated regularly wrong number of clients may be generated as being eligible and this issue
can be taken as input in improving the data updating quality aspect, prevailing in many facilities.
In the Line list report in EMR-ART version 5 contains more than 24 types of lists generated for facility
level data management and program monitoring. This includes

• Lost to Follow Up • Index Case

• Missed Appointment • Family planning
• Dead and Stop • Treatment Current Report Analysis
• Viral Load (3) • Retest Report
• ART Patient and others • Positive Tracking
• ART Initiation • Cervical Cancer (2)
• Transfer IN/Out • Pre-Exposure Prophylaxis
• Enrolled to chronic Care/ART • OTZ
• On TPT/CPT/FPT • TX_ML
• TB Treatment • TX_RTT
• ASM

50 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
SECTION 8:
Data Analysis, Presentation, and Interpretation
OBJECTIVES:
At the end of this section, participants will be able to:
 Define data analysis, visualization, and interpretation
 Describe the purposes of data analysis, visualization, and interpretation
 Calculate HIV/AIDS programmatic performance measures
 Describe metrics of HIV/AIDS data
 Describe data presentation techniques

CONTENTS:
 Basic concepts of data analysis
 Metrics of health and health related data
 Data presentation/visualization
 Data interpretation

8.1. BASIC CONCEPTS OF DATA ANALYSIS

WHAT IS DATA ANALYSIS?
Data analysis is the process of systematically applying different techniques to describe, summarize
and compare data. It is the iterative process of examining data for patterns, trends, and comparisons.
It is turning raw data into useful information and taking the data that you collected and looking at it
in comparison to the questions that you need to answer about clients/patients, programmatic and re-
search.

PURPOSE OF DATA ANALYSIS

The purpose of analyzing data is to obtain usable and useful information. The analysis, irrespective of
whether the data is qualitative or quantitative, may:
 Describe and summarize the data
 Identify relationships between variables
 Compare variables
 Identify the difference between variables
 Forecast outcomes

 Data analysis is turning raw data into useful information. Its purpose is to provide answers to
questions being asked by a health program. Even the greatest amount and best quality of data
mean nothing if data are not properly analyzed.

8.2. METRICS OF HEALTH AND HEALTH RELATED DATA (SPECIFIC TO HIV/AIDS PROGRAM)
The most important part of what comprises an indicator is the metric. A metric is the precise expla-
nation of the data and the calculation that will give the measurement or value of the indicator. A good
metric clarifies the single dimension of the result that is being measured by the indicator. A good met-
ric does this in such a way that each value measured for the indicator is exactly comparable to values
measured at another time.
We use various tools to measure the frequency of the occurrence of disease, death and other health
related conditions and health services in a population. Some of the measures include: absolute num-
bers (counts), ratios, proportions and rates. We will discuss about each metric as follows.

PA RT I C I PA N T ’ s M a n ua l 51
1. ABSOLUTE NUMBERS (COUNTS)
This is the simplest and most frequently performed quantitative measurement. It describes the num-
ber of persons who received a particular service or who have a particular disease, event or character-
istic. It is the number of entities, events, or some other countable phenomenon, for which the question
“how many” is relevant.
Uses of absolute numbers (counts):
 Helpful to understand the total number of people who receive a certain service or who have a
certain type of disease or event
 Mainly useful to monitor the occurrence of important infectious diseases, especially outbreaks
 Used to draw epidemic curve of new cases of a disease over time.

In HIV/AIDS program, some of the indicators that can be more relevantly described with absolute
numbers (counts) are:
 Number of individuals who have been tested for HIV and who received their results
 Number of HIV positive clients identified
 Number of PLHIV who newly started ART
 Number of PLHIV who are currently receiving ART, disaggregated by age, sex and regimen
 Number of PLHIV on ART who are lost to follow up

Counts alone, however, are insufficient for describing the characteristics of a population and for de-
termining risk. The key is to relate the frequency of an event to an appropriate population. We cannot
use counts to compare services or events with other population groups. For this purpose, we mostly
use other metrics such as ratios, proportions, and rates.

2. RATIO
A ratio expresses a relationship between two items in the form of X: Y. Ratios are universal in epidemi-
ology, since they enable the number of cases to be expressed relative to against target population. It
compares the relative frequency of the occurrence of some event to the other event. It is any fraction
obtained by dividing one by another, where the numerator and denominator are not related. Example:
you can say ratio of male to female population in woreda X is 1:1.
Example 1: If you want to compare the ratio of HIV positivity in males versus females, you may say that
male to female ratio of HIV positivity is X: Y. If Male to female ratio, for example, is 1:2, this means that
females are two times positive than males.
Example 2: The ratio of physician to population ratio in Woreda X is 1:12,000 means that for every
12,000 populations, there is one physician in that specific Woreda.

3. PROPORTION
Proportion is one of the basic ways to describe a group. A proportion is a type of ratio in which the
numerator is included in the denominator and the result is expressed as percentages, per 1000, per
100,000 etc. In order for a count to be descriptive of a group, it must be seen in proportion to it, i.e. it
must be divided by the total number in the group.
Ratios, proportions, and rates all include both a numerator and a denominator.
During calculation of indicators, we usually set up fractions. Numerator is the top number and denom-
inator is the bottom number in a fraction. In public health, numerator is often the number of health
events and/or services that we provided.

52 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Example: - In the calculation of the indicator “Proportion of Sexually Transmitted Infection (STI) cases
tested for HIV”, the numerator is “Number of STI cases tested for HIV in the reporting period” and the
denominator is “Total number of STI cases in the reporting period”.
In order to clearly understand the above indicator, we need to be clear about both the numerator and
denominator. For HIV/AIDS program analysis, numerators are usually taken from EMR-ART or from
the routine DHIS2/MER reports (weekly, monthly, quarterly and/or annual reports). Denominators
can be taken from the population data in most cases and may be from DHIS2/MER reports.

EXAMPLE 1:
Indicator: Percentage of adults and children known to be on treatment 12 months after initiation of
ART
 Numerator: Number of adults and children who are still on treatment at 12 months after
initiating ART
 Denominator: Total number of adults and children who initiated ART in the 12 months prior
to the beginning of the reporting period (net current cohort)
In this example: the denominator is also taken from monthly report
EXAMPLE 2:
Indicator: Percentage of people living with HIV receiving ART
 Numerator: Number of adults & children receiving ART at the end of the reporting period
 Denominator: Estimated number of people living with HIV
In this example, the denominator is population-based estimate

EXERCISE:
For the following indicators, what are the numerators and the denominators? Discuss in groups and
present to the participants
 Viral load suppression rate
 Early viral load suppression rate
 ART retention rate
 Number of adults and children with HIV infection newly started on ART
 Proportion of clinically undernourished People Living with HIV (PLHIV) on ART who received
therapeutic or supplementary food
 Number of persons provided with Post-Exposure prophylaxis

4. PERCENTAGE (NUMBER PER 100)

Percentage (number per 100), is one of the most common ways of expressing proportions. Number
per 1000 or per 100,000 and per 1 million or any other convenient base may also be used.

A percentage is just a proportion multiplied by 100 and they are often used instead of proportions
because most people are more familiar with percentages. Several of the indicators in the HMIS/MER
are called “proportions” but the formulas require that you multiply the proportion by 100, which mean
we are really presenting percentages.

Please refer to your HMIS/MER Indicator Reference Guide to calculate some percentages. Some indi-
cators may be expressed as number per 1,000 or per 100,000 because the events are rare to express
them as number per 100.

PA RT I C I PA N T ’ s M a n ua l 53
Example: - Infant mortality rate: This indicator has a base of 1,000 and will be expressed as number of
infant deaths per 1000 livebirths. Maternal mortality ratio is another indicator, which is expressed as
per 100,000.

5. RATE
Rate measures the relative frequency of cases per unit of population per unit of time. It can be seen
as a Proportion with a time dimension. It measures the occurrence of deaths (mortality), births and
disease (morbidity). Due to the variation in population of different woredas and zones, it can be more
helpful to calculate rates per population.

Example: - Let us see the following example for two facilities namely, Health Facility X and Y. Facility
provides service to a total of 100,000 catchment population and Facility B provide service to 25,000
catchment population. In 2011 EFY, both facility X and facility Y have provided HTC service to 10,000
people. Using an “HTC testing rate” calculation, HTC testing rate for Facility X is 10% (10,000/100000)
and for Facility Y is 20% (5,000/25,000). By just observing the total number of tests done by Facility X
and Facility Y, we may wrongly conclude that both Facilities have performed equal. But due to differ-
ence in the catchment population they serve, Facility Y have a better testing rate (20%) than Facility
X (10%).

CASE SCENARIO - EXERCISE ON DATA ANALYSIS

CASE 1: Dejen Health Center provide service to a catchment population of 40,000. In the
first three months of 2012 EFY (Hamle 2011, Nehassie 2011 and Meskerem 2012), the ART
clinic has reported the following HIV Care and support related activities.
 A total of 810 adult PLHIV and 30 children <15 PLHIVs were receiving ART at the end of
Meskerem 2012.
 100 PLHIVs on ART were tested for viral load and 80 of them had a viral load less than
1,000 copies per ml
 In the three months’ period, there were 20 PLHIVs who were lost from the ART follow up
 In the health center, 25 PLHIVs started ART 12 months ago, among which 5 were
transferred out to Debre Markos hospital, 10 PLHIVs of the same cohort were transferred
in from Bichena health center. After 12 months, the data clerk in the ART clinic found that
25 adults and children were still on treatment at 12 months after initiating ART.
 ART in the original cohort including those transferred in, minus those transferred out (net
current cohort) were
 Regarding screening for Tuberculosis, 500 PLHIVs were screened for tuberculosis in
Hamle 2011 EFY, 100 screened in Nehassie month and 600 PLHIVs screened in
September 2012 EFY
 200 PLHIVs were nutritionally assessed and 50 of them were found to be clinically
undernourished. Supplementary food support was provided to 35 PLHIVs that were
clinically undernourished

What indicators can be computed from the above case scenario? List all possible HIV/AIDS
program indicators that can be computed from it.

Calculate the indicators and discuss in groups

54 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
8.3. DATA PRESENTATION/VISUALIZATION

SESSION OBJECTIVES:
At the end of this session, participants will be able to:
 Describe the purposes visualization/presentation
 Describe different ways to best summarize and present HIV/AIDS data
 Determine and choose the appropriate format for data presentation/visualization
 Understand Smart-Care ART dashboard
 Develop graphs that display performance measures
In this section, we will deal about the different data presentation/visualization methods followed by
examples for each data presentation method.
“A picture is worth a thousand words”

8.3.1. OVERVIEW OF DATA PRESENTATION/ VISUALIZATION

The manner in which data is translated into visual images for improved interpretation by the reader
is called data visualization. Information needs to be displayed/ presented in ways that can be easily
seen, understood and discussed by managers, health workers and other people. Using charts, graphs
and pictures makes ideas and trends to be easily communicated to people. The human brain can more
accurately interpret an image or picture rather than rows and columns of numbers, data or words.
Clearly communicating ideas and trends is necessary to guide initiatives toward achieving objectives.
The use of data visualization tools is particularly important for health care executives, who often need
to convey complex concepts across a wide range of stakeholders that may include representatives
from finance, human resources, clinical staff, and more. The uses of visualization tools help simplify the
interpretation of the data by the audience or reader.
Data presentation is the systematic process of making information available and accessible to poten-
tial users, stakeholders and/or beneficiaries.
Some information is effectively displayed as a table, while other information is more easily understood
when presented in a graph or a map. It is important to select the best type of display format that best
present the information. Irrespective of the method used for displaying data/information, the follow-
ing basic principles should be adhered to:
Data Visualization Consideration
Before selecting the right type of data visualization, begin with the right questions:
 Why are you even doing this?
 What story to tell?
 What data to use?
 What are your available resources?
 Who will be involved?
 Who is your audience?

Data presentation: Is the systematic process of making information available and accessible to poten-
tial users, stakeholders and/or beneficiaries.
Some information is effectively displayed as a table, while other information is more easily understood
when presented in a graph or a map. It is important to select the best type of display format that best
present the information. Irrespective of the method used for displaying data/information, the follow-
ing basic principles should be adhered to:

PA RT I C I PA N T ’ s M a n ua l 55
  Titles / labels should clearly indicate the contents/data displayed in terms of person, place,
and time (what, who, where, when). The axes of graphs should be well labeled. Example (title):
Number of PLHIV currently on ART in Hawassa Health center from Hamle 2008 to Tikimt
2012.
 Indicate the source of the data: When data from EMR-ART/DHIS2/DATIM/ report or other
source is used, indicate the date when the data was extracted.
 Amount of information: Do not put too much information in one table, graph or map. Keep
them simple to convey clear messages that enable users to draw the necessary conclusions
from what is presented?

8.3.2. TECHNIQUES OF DATA VISUALIZATION/PRESENTATION FOR HIV/ADIS PROGRAM

There are a number of visual data summarization and presentation formats such as tables, graphs,
maps and diagrams. In this section, we will learn about the details of each presentation format with
examples from EMR-ART/DHIS2/DATIM reports.

1. TABLES
A table is the simplest means of summarizing a set of observations and can be used for all types of
numerical data. Tables are often used in reports. If tables are used properly, messages can more ef-
fectively be conveyed than the written text and enables the presenter to explain the data. Tables are
easy to make but may be difficult to use, especially if they are big. It is critical that rows and columns be
clearly labeled and, where appropriate, all the categories should be clearly shown.
A table is a method of numerical data organization/presentation in rows and columns. Rows are hori-
zontal and columns are vertical arrangements. The use of tables for organizing data involves grouping
the data into mutually exclusive categories of the variables and counting the number of observations
(frequency) to each category.
A table is the simplest means of summarizing a set of observations and can be used for all types of
numerical data. Tables are often used in reports. If tables are used properly, messages can more ef-
fectively be conveyed than the written text and enables the presenter to explain the data. Tables are
easy to make but may be difficult to use, especially if they are big. It is critical that rows and columns be
clearly labeled and, where appropriate, all the categories should be clearly shown.
The following general principles should be addressed in constructing tables.
 Tables should be as simple as possible.
 If data are not original, their source should be given in a footnote.
 Tables should be self-explanatory
 Title should be clear and to the point (a good title answers: what? when? where?) and it
should be placed above the table.
 Each row and column should be labeled.
 Numerical entities of zero should be explicitly written rather than indicated by a dash.
 Totals should be shown both in row and column wise

56 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Example:
Table 6. Number of HTC tests performed and positive tests in health center A, Tir-Sene 2011 EFY

No. of people tested at VCT and PITC Number of positive tests at VCT and PITC
Tir 2011 400 5
Yekatit 2011 300 6
Megabit 2011 500 8
Miazia 2011 600 10
Ginbot 2011 200 3
Sene 2011 700 5
Total 2700 37
2. GRAPHS
Graphs are very important for making sure that information is fully understood. Graphs should be
designed so that they convey the general patterns in a set of observations at a single glance. Although
they are easier to read than tables, graphs often supply a lesser degree of detail. However, the loss of
detail may be accompanied by a gain in understanding of the data. The most informative graphs are
relatively simple and self-explanatory. Like tables, they should be clearly labeled, and units of mea-
surement should be indicated.
Graphs should tell a ‘story’ by themselves and are best used to detect trends over time, search for pat-
terns among large amounts of data and display the relationships between variables.

When using graphs to present data/information:

 Every graph should be self-explanatory and as simple as possible.
 Title should be written clearly, and usually placed below the graph
 Label axes should clearly be stated
 Provide a legend which explains each of the lines or bars
 Select scales that fill the entire graph on both axes. Use scales that best illustrate what is being
shown, e.g., percentages may work better than raw numbers.
 Use same scale consistently in graphs
 Where possible, show a target line or reference point
 The numerical scale representing frequency must start at zero or a break in the line should be
shown.

TYPES OF GRAPHS
Different types of graphs are used for different purposes. It is important to think which kind of graph
will work best to show the information. Example: If you want to see trend over time, a line graph is the
best method to visualize your data.

BAR GRAPHS: Bar charts are a popular type of graph used to display a frequency distribution for cate-
gorical and discrete nominal variables. Bar charts are used to plot individual data values next to each
other, for example to compare different facilities, activities or indicator values. In a bar chart, the var-
ious categories into which the observations fall are presented along a horizontal axis. A vertical bar
is drawn above each category such that the height of the bar represents either the frequency or the
relative frequency of observations within that class. The bars should be of equal width and separated
from one another so as not to imply continuity.

PA RT I C I PA N T ’ s M a n ua l 57
 Bars should be considerably wider than the space between them; a principle of allowing the data to
dominate the chart.
 Grid lines help to make comparisons
 Avoid choosing very close grid lines
 A well-chosen order for the variable values gives a better chart
 Horizontal bar graphs are used when the variable values have long names or when three are too
many variables
There are different types of bar graphs; the most important ones are simple bar graph, multiple/
grouped bar graphs, and stacked bar graph.

SIMPLE BAR GRAPH: It is a one-dimensional diagram in which the height or length of each bar indi-
cates the size (frequency) of the figure represented.
In the above table which displays the HIV tested positive clients of Ethiopia by Region EFY 2011, it can
be displayed in a bar graph form as follows. The data can be easily understandable when displayed in
graph than in table form.

Figure 15. Number of positive HIV tests by region, EFY 2011

Using the above bar graph, one can allow us to easily compare each region than when it is displayed in
table or word.
MULTIPLE/GROUPED BAR GRAPH: In this type of graph, the component figures are shown as separate
bars adjoining each other. It is used when two or more categories of qualitative data put alongside
each other.
The height of each bar represents the actual value of the component figure. It depicts distribution pat-
tern of more than one variable. The limitation is that it doesn’t show the total within a given category.
If we also try to cram in too many categories, the chart becomes difficult to understand; hence better
to draw several ordinary/normal charts instead.
Example: The national HIV/AIDS program manager wants to see the regional performance of the num-
ber of HIV positives identified and number of PLHIV newly initiated on ART in 2011 EFY. In this case,
the total HIV positives identified and newly initiated on ART can be put side by side so that the differ-
ence can easily been seen.

58 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Figure 16. No. of people tested positive for HIV and no. of PLHIVs newly initiated on ART, 2011

STACKED BAR CHART: In this type of graph, bars that represent the frequency of observations in two
or more different subgroups are placed on top of one another. Example: Male to female proportion of
ART initiation by region for the EFY 2011.
Bars are sub-divided into component parts of the bar. This type of graph is constructed when each
total is built up from two or more components. It gives a picture of how a total is broken into its parts.
Total is clearly visible, but the size of each categories of the variable takes second place. Only the size
of the bottom category is easy to read precisely, and it might be difficult to underhand if lots of catego-
ries present and stacked bar graphs can be constructed using actual figures or percentages.

Figure 17. Number of PLHIV newly started on ART by Sex, EFY 2011

From the above stacked graph, you can easily convene different information in one graph but in a sim-
ple manner. You can see the overall Male to Female proportion ART initiated by region.
Histogram: A histogram is a graph of the frequency distribution of continuous measurement variables.
It is constructed on the basis of the following principles:
 A histogram is constructed by choosing a set of non-overlapping intervals (class intervals) and
counting the number of observations that fall in each class. The number of observations in each class
is called the frequency. Hence histograms are also called frequency distributions

PA RT I C I PA N T ’ s M a n ua l 59
 The horizontal axis is a continuous scale running from one extreme end of the distribution to the
other. It should be labeled with the name of the variable and the units of measurement.
 For each class in the distribution a vertical rectangle is drawn with
- Its base on the horizontal axis extending from one class boundary of the class to the other
class boundary, there will never be any gap between the histogram rectangles.
- The bases of all rectangles will be determined by the width of the class intervals
 Class intervals are usually chosen to be of equal width. If this is not the case, the histogram could give
a misleading impression of the shape of the data
 Classes of equal size
 The base of each rectangle is the same
 Height of the class is equal/proportional/corresponds to its frequency
 Classes of different size
 Relatively complicated to draw and read
 Height of the class is not equal/proportional/to its frequency
 Only area is proportional to frequency
 We therefore write nothing on the y-axis, except perhaps for ticks

LINE GRAPHS: A line graph is used to illustrate the relationship between two different continuous
measurements. The line graph is especially useful for the study of some variables according to the pas-
sage of time. Each point on the graph represents a pair of values: the scale for one quantity is marked
on the horizontal axis, or x-axis, and the scale for the other on the vertical axis, or y-axis. Each point
on the graph represents a pair of values. Each value on the x-axis has a single corresponding measure-
ment on the y-axis.
Adjacent points are connected by straight lines. The time, in weeks, months or years is marked along
the horizontal axis and the value of the quantity that is being presented is marked on the vertical axis.
The distance of each plotted point above the baseline indicates its numerical value. The line graph is
suitable for depicting a consecutive trend of a series over a long period.
Example X: Let us say a Health facility head wants to see the trend of TX_CURR in the facilities from
2005 EFY to 2011 EFY.
To easily understand whether TX_CURR is increasing or decreasing in the health facility. A line graph
is an appropriate method of data presentation. See the graph below

Figure 18. Number of PLHIV currently on ART in Health Center X from EFY 2000-2011

60 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
PIE CHARTS: These types of charts are used to show the proportion of an activity as part of the whole
(like the slice of a pie) as a ‘slice’ in a circle. It is a circle divided into sectors so that the areas of the sec-
tors are proportional to the frequencies. Used to represent and compare the frequency distribution
of categorical variables. It is an alternative to bar charts. We should not have too many sectors in a pie
chart; five or six is a reasonable upper limit for a lucid chart.
Example: Let us make a chart for the indicator TPT “Z health facility Proportion of Clients on ART
whose charts labeled with Gold, Silver and bronze on August 2019.
Figure 19. Pie Chart showing nutritional status of PLHIVs in Hospital X, 2011 EFY

MAPS: Maps can be used to display several other types of health information. It can provide a global,
national, zonal or woreda view of the prevalence of a disease. This can give a clear picture of best
performing areas and where most support is needed. Whilst similar data can be illustrated in a graph
or a map, a map provides a geographical ‘picture’ at one glance whilst a graph can be used to compare
indicator values for different years in each province and differences between provinces over time in
an understandable way.
Maps are the most useful visualization methods. There are various map types you can use, such as
bubble maps, spot maps to indicate different geographies or administrations.
A catchment area map is an effective tool to enable facility staff to understand the area and the pop-
ulation they are serving. These maps can be used to depict problems in terms of availability and acces-
sibility to health care and distribution of population and facilities.
Spot maps show where facilities are situated and what effect location may have on health indicators
and health care results.

8.3.3. E-SYSTEMS FOR DATA VISUALIZATION

Multiple software platforms, either open-source or proprietary, are available to facilitate data visual-
ization. These software applications, many of which are interactive, provide tools to develop charts,
maps, infographics, timelines, and other visual tools.

PA RT I C I PA N T ’ s M a n ua l 61
The common tools used for data visualization using simple electronic tools are Scorecard and Dash-
board. Scorecards tell how the health systems is doing overall while dashboards tell what’s happening
now using interactive metrics with drill-down capabilities. Dashboard visualizations are more in-depth

8.3.3.1. EMR-ART Dashboard

EMR-ART_ART software has a dashboard, which is displayed as a homepage of the software. The
dashboard displays some HIV/AIDS calculated indicators such as “Trend of ART initiation and follow
up status”, “Viral load result”, Number of PLHIV that missed AR, by number of missing days” etc.…

Figure 20. ART EMR-ART dashboard

62 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 8.3.3.2. Scorecard
Scorecard is a powerful data analysis and visualization tool that provides leaders at all levels with easy
performance tracking and comparison at subsequent levels of the health system for key indicators,
thus enhancing transparency and encouraging action. The scorecard colors; red, yellow and green to
enable managers to easily visualize and identify areas in which progress has been good or is sub-opti-
mal.
Table 7. HIV/ ADIS program Scorecard of zone A in region B, EFY 2011

% of STI cases Retention rate at % VL suppression % VL suppression

Name of Woreda
tested for HIV 12 months at 6 month at 12 month
Ankasha Guagusa 84% ND ND 87%
Awi ZHD 85% 91% 91 88%
Ayehu Guagusa 51% 100% ND 88%
Banja 97% ND ND 92%
Chagni Town 90% 100% 100 87%
Dangla Town 90% 91% 100 96%
Dangla Zuria 100% 100% ND 86%
Fagitalekoma 100% 86% 100 94%
Guagusa Shikuda 90% 71% 83 91%
Guangua 96% 10% ND 84%
Injibara Town 37% 85% 82 93%
Jawi 73% 80% 100 77%
Zigem 100% ND ND 71%

From this scorecard we can easily understand the performance of woredas using colors: Green; meets
the expected, yellow; near to meet the expected and red need attention to meet the expected perfor-
mance.
8.3.4. MINIMUM DISPLAY CHARTS FOR HIV/AIDS PROGRAM UNIT
HIV/AIDS units of each health facility should maintain a minimum set of standard charts for selected
HIV/AIDS indicators. The purpose of minimum standard charts is to ensure that:
 Basic HIV/AIDS program information is regularly updated and monitored
 Basic HIV/AIDS program information is displayed where it will have visibility to health workers,
supervisors, and visitors
The following display charts are charts that are just MINIMUM. It is highly recommended that each
health facility can have additional HIV/AIDS charts as appropriate.

Table 8. Minimum display charts for HIV/AIDS units

Name of Chart Visualization Format Frequency of update Place
PLHIV currently on ART Trend Monthly ART Unit, HMIS unit
Viral Suppression Rate Bar graph Monthly ART unit
ART Cohort monitoring chart Table Monthly ART Unit
HIV Testing and Result Graph Monthly VCT Unit

PA RT I C I PA N T ’ s M a n ua l 63
CASE SCENARIO - EXERCISE ON DATA PRESENTATION/VISUALIZATION
Case: Refer the cases scenario (Dejen Health Center) that you have worked in the data analysis
section and use the following additional information for this exercise.
Dejen Health Center provide service to a catchment population of 40,000. In the first three
months of 2012 EFY (Hamle 2011, Nehassie 2011 and Meskerem 2012), the ART clinic has re-
ported the following HIV Care and support related activities.
 The number of Adult and child PLHIVs in the health center, by month was as follows:
 Yekatit 2011: Adult on ART = 750; children on ART= 20
 Megabit 2011: Adult on ART = 760; children on ART= 22
 Miazia 2011: Adult on ART = 770; children on ART= 23
 Ginbot 2011: Adult on ART = 775; children on ART= 26
 Miazia 2011: Adult on ART = 790; children on ART= 26
 Sene 2011: Adult on ART = 790; children on ART= 28
 Hamle 2011: Adult on ART = 799; children on ART= 29
 Nehassie 2011: Adult on ART = 800; children on ART= 29
 Meskerem 2012: Adult on ART = 810; children on ART= 30
 100 PLHIVs on ART were tested for viral load and 80 of them had a viral load less than 1,000
copies per ml
 In the three months’ period, there were 20 PLHIVs who were lost from the ART follow up
 In the health center, 25 PLHIVs started ART 12 months ago, among which 5 were transferred
out to Debre Markos hospital, 10 PLHIVs of the same cohort were transferred in from Bichena
health center. After 12 months, the data clerk in the ART clinic found that 25 adults and
children were still on treatment at 12 months after initiating ART.
 ART in the original cohort including those transferred in, minus those transferred out (net
current cohort) were
 Regarding screening for Tuberculosis, 500 PLHIVs were screened for tuberculosis in Hamle
2011 EFY, 100 screened in Nehassie month and 600 PLHIVs screened in September 2012 EFY.
The screening result shows that a total of 10 patients in Hamle, 2 patients in Nehassie and 15
patients in Meskerem were diagnosed to have Tuberculosis.
 200 PLHIVs were nutritionally assessed and 50 of them were found to be clinically undernour-
ished. Supplementary food support was provided to 35 PLHIVs that were clinically undernour-
ished

What visualization method will you use for each type of indicator from the above case scenario?
Develop the following graphs:
 Line graph for TX-CURR trend in Dejen health center
 Bar graph for an indicator that you select from the case scenario
 Pie chart

64 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
8.4. DATA INTERPRETATION
SESSION OBJECTIVES
By the end of this session, participants will be able to:
 Understand the difference between data analysis and interpretation
 Define data interpretation
 Understand how data interpretation can be done for HIV/AIDS program
In the previous sections, you have learnt about how data can be transformed into information by an-
alyzing it and in the next sessions we will cover how data can be summarized using tables, charts or
through narration. Analyzing data helps to make the data to be meaningful. Once the data is changed
to meaningful information, we need to interpret it. We need to consider the relevance of the findings
to our program, identify the relevance of the findings to our programs and set the possible next steps.
Data interpretation is the process of making sense of the information. It is attaching meaning to infor-
mation by making connections and comparisons and exploring causes and consequences. The purpose
of data interpretation is to get answers to program or study questions. It answers the ‘why’ of happen-
ings in our program.

Data interpretation involves examining the following:

 How are our services performing in terms of our goals, objectives, and targets? Does the indicator
meet the target? How far is it from the target?
 How are our services performing compared with benchmarks?
 What is happening in our services over time? (trends)
 How is our facility performing compared to others? How are our services performing compared to
services in other facilities?
 Why are we doing well (or badly)?
 How can we do better?
 Can we improve quality of care with existing resources?
 How can we be more effective or efficient?”
Data analysis and interpretation are related and sometimes difficult to separate. Analysis answers the
“what” of a program and interpretation answers the “why” of a program. Let us see the concepts of
analysis and interpretation with an example. Suppose the HIV/AIDS unit of health center X wants to
know if the health center’s ART program is performing well or not. The program manager will perform
the following:
1. The manager should first define the actual ART performance by calculating ART program indica-
tors. He/she will also look at the ART targets set at the beginning of the fiscal year. Then the program
manager will compare the ART program indicator target with the actual performance for the specific
time. Performing all this is called “data analysis”. [Note that in the previous sections, you have learned
about HIV/AIDS program indicators and how it can be computed].
2. Using the above analysis, further exploration of the findings should be done to better understand
the relevance of the findings to the program and how the findings are going compared to the plan. The
potential reasons for the findings should be explored. At this step, you have moved from the ‘what’ is
happening in the ART programs to the ‘why’ it is happening. This second step is called “interpretation”.
Example 1: Suppose a clinician working in the ART clinic wants to know the nutritional status of PLHIV
who are on ART and nutritional supplementation of PLHIVs who are found to be clinically malnour-
ished. He/she performs the following:

PA RT I C I PA N T ’ s M a n ua l 65
 Determine the nutritional status of PLHIV on ART by calculating Body-Mass-Index (BMI). He/she
will compare it with the normal/acceptable range values. From the report, he/she should determine
the proportion of clinically malnourished PLHIVs for whom nutritional supplementation is provided.
Performing all this is called “data analysis”. [Note: EMR-ART supports clinicians and data managers
to compute BMI by auto-calculating it when weight and height are entered into the software].
 Using the above analysis, further exploration of the findings should be done to better understand
the relevance of the findings. The potential reasons for the findings should be explored. This step is
“interpretation”.
Example 2: Imagine that you are the head of the ART clinic in health center X and you want to deter-
mine the performance of HIV/AIDS programs. In this case, you have to look at the HIV/AIDS data of the
health center for the period that you want to determine, summarize the data and interpret it. During
interpretation, you may be interested to know the following:
 Did the health center achieve the HIV/AIDS program targets? Which program area has achieved the
target, and which is below the target?
 How do the HIV/AIDS program indicators compare with national & regional benchmarks?
 How is trend of HIV indicators over time?
 What are the reasons for our best or bad performances?
 How can we improve HIV/AIDS indictors?
Example 3: Answering HIV/AIDS program questions through interpretation
Question: Is viral suppression among PLHIV receiving ART meeting its target in Health Center X?
Analysis: Compare the target for viral suppression for the health center (E.g., 90%) with the actual
viral suppression (e.g., 70%) so that you can see the gap between your target with the performance
Interpretation: What does 70% viral suppression mean to the HIV/AIDS ART program in the facility?
It means that suppression is far below the target set for the facility and many PLHIV on ART does not
have viral suppression as planned. The reason why performance is low may require further investiga-
tion and additional data.

66 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
SECTION 9:
Data Analysis and visualization using Excel
OBJECTIVE
At the end of this section, participants will be able to:
 Understand & use basic Excel skills to analyze data
 Use Microsoft excel to develop different charts
 Build Dashboard for ART Programs and Use Updating options when new data is entered
Contents
1. Data Analysis using excel
 Getting data in excel from EMR-ART system
 Basic Microsoft Excel
 Data Preparation for Analysis
 Data filtering & Sorting Data in Excel
 Using Pivot Table in Excel
2. Data Presentation & Visualization using Excel (Charts and Dashboard)
a. Excel Charts
b. Excel Dashboard

9.1. DATA ANALYSIS USING EXCEL

Getting data in excel from EMR-ART System
In most electronic data management tools or software, we can get individual level or aggregate client
data by using different options embedded in the systems. In the EMR-ART patient level can be fetched
by using the Print/Export option at the right upper corner of the Line List or Aggregate reports gen-
erated for use in each reporting section in the system. The following are some of the aggregate and
patient Level and Individual level reports.

What is the difference between Aggregate Data and Individual/Patient level

from the EMR-ART.? What are the common types of Patient level and Aggregate
from EMR-ART?

Individual/Client Leve Data

Reports can be prepared containing detail client personal identifier and other detail information as re-
quired; these reports are provided for facility level use only for improvement of clinical service quality
improvement by identifying gaps in provision of service by identifying each client uniquely. Some of
these client level reports are the following.
1. Treatment Current, New and Ever Enrolled Clients List
2. Clients Receiving Modern Family Planning Method
3. Clients Received Cervical Can Screening and Eligible for cervical Cancer
4. Viral Load Sent and Received, Eligible Clients for VL and High Viral Load Line List
5. List of clients received TPT, CPT of Fluconazole Preventive Therapy
6. List of Clients Based on Data Completeness Status and Eligibility for TPT, Viral Load, Appointment
Spacing Model and Partner and Family Based Index Case Testing and others

PA RT I C I PA N T ’ s M a n ua l 67
Exercise: - Generate Line List report of the above line list reports from the EMR-ART and export in the
excel, Word and PDF format. For detail steps please look at the mentioned in section 7 of this document.

Note: In Exporting individual level patient in excel for analysis make sure that type of
excel format should only be with extension Excel (97-2003) Data-Only (*.xls) or Excel
Workbook Data-Only (*.xls)

Aggregate Reports from EMR-ART

In this section, we will discuss about the basics of using excel for data analysis.

Data Preparation for Analysis

Data can be analyzed by using various approaches and tools; some of tools used for data analysis and
visualization includes
1. Microsoft Excel
2. Microsoft Power BI
3. Tableau
4. Others
Aggregate Repots
Aggregate reports are those reports summarized by some predefined criteria set by FMOH, PEPFAR,
other partners and evolving data demands at all levels. These reports include HMIS/DHIS, DATIM and
other Aggregate reports generated by using data extraction template. See section seven for the detail
of the aggregate reports from EMR-ART.

What is the EMR-ART Excel Data Extraction Tool and explain the steps to extract
data from the EMR-ART to improve the program monitoring and continuous data
completeness improvement?

68 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
EMR-ART Data Extraction Tool or Excel
EMR-ART Data Extraction Tool is prepared for the purpose of extracting pertinent HIV/AIDS program
priority data to be used at facility level and sharing it along the hierarchy of all administrative levels
and other partners for gauging the program performance and for timely action of the gaps identified
in the context of high priority areas.

Steps to Use the EMR-ART Data Extraction Tools and Sharing for Use
1. Make sure the EMR-ART system is functional and all necessary data from the paper charts is fully
updated in the system before using the extraction tool
2. Make sure Data/Internet connectivity is available in the EMR-ART computer
3. Get the correct version of the Data Extraction Tool and open the template on the computer the
EMR-ART system already running. Make sure you run the tool on the server computer If the client
computers are connected using the Local Area Network.
4. Follow the next steps to extract EMR-ART data using the extraction excel tool
a. Click Clear Value
b. Select the month and year of report
c. Click Generate Report
d. Click Send Email, with this step the deidentified data to relevant data users and stakeholders

Note: EMR-ART data extraction tool will generate and export patient level data
excluding the Client Identifiers like Client Name, Unique ART Number and Medical
Registration Number. The following reports could be analyzed from the extracted
data using this tool
1. Treatment Current Report
2. Viral Load Data including HVL and wit EAC Cascade
3. Cervical Cancer Screening and Treatment
4. TB Preventive Therapy

PARTICIPANTS MANUAL
Figure 21. EMR-ART Data Extraction Tool, Data Generation Steps
PARTICIPANTS MANUAL

A
B C

Figure 21. EMR-ART Data Extraction Tool, Data Generation Steps

Check the version of

Data extraction tool
PA RT I C I PA N T ’ s M a n ua l 69
 D

Figure 21. EMR-ART Data Extraction Tool, Data Generation Steps

Check the version of

Data extraction tool

The Five Line List Report of Patient Level reports

generated by the EMR-ART data extraction tool

Data Extraction
page

In this document, basic data analysis and visualizations are described on the basis of Microsoft Excel
In this document,
that is widely and freelybasic data analysis
available in mostand visualizations
facilities are described
implementing on the basis
EMR-ART of Microsoft
for HIV data management.
Excel that is widely and freely available in most facilities implementing
Basic excel skills may be required to understand and use the contents included in this EMR-ART for HIV data
document but
management.
naïve users can copeBasic
withexcel skills may be
the contents withrequired
little to understand
efforts throughandtheuse ordinary
the contents
useincluded
of excelinfor routine
thisOther
activities. document but naïve
business users can cope
intelligence with the
software contents with
mentioned abovelittle
mayefforts
havethrough the features but not
advanced
ordinary use
freely available andofneed
excel intensive
for routine training,
activities. Other
whichbusiness intelligence
may render software
limitation onmentioned
the data analysis use at
above may
lower (facility) have advanced features but not freely available and need intensive training, which
level.
may render limitation on the data analysis use at lower (facility) level.
Prerequisites before using the EMR-ART Data for Further Analysis
Preparation of data for analysis is one of the mandatory activities that should be performed before
any data is used for further analysis and decision making. Data quality assurance activities94mentioned
in the above section of this document should be regularly completed and documented based on the
standards; These quality assurance activities include
1. Make sure that all the data are properly updated in the EMR-ART timely (Perform the Treatment
Current Analysis on weekly and Monthly basis)
2. Make sure that all the completeness, validity, and consistency checks to be done in the facility and
proper action points are identified and documented on weekly/monthly
3. Comparison should be between the reported values from EMR-ART generated and sent reports to
the higher level are the same based on selected data elements
4. Make sure that the LQAS score of the overall report of the facility is acceptable
5. Conduct a monthly data quality check/assessment on randomly selected patient cards and verify its
consistency with registers and EMR-ART-ART as described in the data quality section,
The following contents are included to help participants understand about basic excel skills needed
for analysis of data that is extracted from EMR-ART.
1. Basic functions/parts of excel
2. Calculate measures of central tendency and measures of dispersion using excel
3. Prepare tables, graphs, and charts in excel
4. Create interactive Dashboard in Excel using data from EMR-ART

70 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
BASIC OF MICROSOFT EXCEL
What is excel, columns and rows PARTICIPANTS MANUAL
Excel is one of the spreadsheets programs used to record and analyze numerical and other forms of
PARTICIPANTS MANUAL PARTICIPANTS MANUAL
data. Excel interaction between columns (represented by letter) and rows (represented by number)
that forms a table. It arranges data in rows and columns. One row represents one individual (entity)
PARTICIPANTS MANUAL
data whereas columns describe attribute or variable for that individual or entity.

ColumnColumn
Row

Row

Example: In the above excel table exported from EMR-ART treatment current line list, a client
currently on ART is represented by a row containing only a single individual’s ARV treatment data.
Example:Example:
In the
The columns
In theexcel
above above excel
containtable
table exported
exported
description of from
from EMR-ART
EMR-ART
that individual
treatment
bytreatment
current
current
various variables lineline
like list,list,
Medical
a client
a client
Recordcurrent-
currently on ART is represented by a row containing only a single individual’s ARV
ly on ART is represented by a row containing only a single individual’s ARV treatment data. The col- treatment data.
Number (MRN), Unique ART Number, Patient Name, Age, Sex, Type of Regimen, and other
The columns
umns contain contain
description description
of that of that
individual individual
by various by various
variables likevariables
Medicallike Medical
Record Record
Number (MRN),
variables.
Unique ART Number,
Number (MRN),Patient
UniqueName, Age, Sex,Patient
ART Number, Type ofName,
Regimen,
Age, and
Sex, other variables.
Type of Regimen, and other
UPPER AND LOWER PARTS OF AN EXCEL
variables.
UPPER AND LOWER PARTS OF AN EXCEL
UPPER
In the
In the upper ANDsection,
upper
section, LOWER
knownknown PARTS
as asOF
“Ribbon AN
“RibbonEXCEL
Tables”,
Tables”, it contains
it contains all basic
all basic functions
functions in in excelused
excel usedtotoperform
functionsperform
in excelfunctions
sheet. in excel sheet.
In the upper section, known as “Ribbon Tables”, it contains all basic functions in excel used to
perform functions in excel sheet.

In the lower
In thepart of part
lower excel,
of sheets of excel
excel, sheets document
of excel opened
document openedareare
displayed.
displayed.The
Thenumber of sheets will
number of
be determined based
sheets will on the need
be determined of the
based user
on the andofcan
need the be renamed
user byrenamed
and can be double clicking onclicking
by double the default
In the lower part of excel,
names written as sheet1, Sheet2 etc. sheets of excel document opened are displayed. The number of
on the default names written as sheet1, Sheet2 etc.
sheets will be determined based on the need of the user and can be renamed by double clicking
on the default names written as sheet1, Sheet2 etc.

96
PA RT I C I PA N T ’ s M a n ua l 71
96
 PARTICIPANTS MANUAL

Figure 22. Steps to rename a sheet name in excel

PARTICIPANTS MANUAL

Figure 22. Steps to rename a sheet name in excel

DATA ENTRY IN EXCEL
DATA ENTRY IN EXCEL
Data in excel can be entered directly by typing into each cells or data can be directly generated from
EMR-ARTData andinexported
excel canas beexcel
entered document.
directly by Totyping
view details
into each ofcells
getting readymade
or data data in
can be directly excel from
generated
EMR-ART, please refer to the previous section of this document. It is described in Section Seven above
from EMR-ART and exported as excel document. To view details of getting readymade data in
(Generation/Extraction of Data from EMR-ART).
excel from EMR-ART, please refer to the previous section of this document. It is described in
CREATING Section Seven FOR
DATASET aboveANALYSIS
(Generation/Extraction of Data from EMR-ART).
Steps to get data ready for analysis using excel
CREATING
• Use EMR-ART line DATASET
list report to FORgetANALYSIS
data for analysis as mentioned in the above section; we can also
get lineSteps
list report
to get of TX_Curr,
data ready for Viral Load, using
analysis TPT andexcelother client level data from the EMR-ART
extraction tool as mentioned in the above section.
Basic Steps in• DataUseAnalysis
EMR-ARTUsing line list report to get data for analysis as mentioned in the above section;
Excel
1. Get the wedata
can in
also get format,
table line list report of TX_Curr,
i.e., data arrangedViral Load,(horizontal)
in rows TPT and other
andclient level(vertical)
columns data
2. Make surefrom that there areextraction
the EMR-ART no blank columns and rowsininthe
tool as mentioned theabove
data presented
section.
3. Make sure each cell this labelled with unique column labeling or header
4. Basic
AfterSteps in Data Analysis
that tabularize Usingby
your data Excel
using the following steps
• Select a cell containing data
1. Get the data in table format, i.e., data arranged in rows (horizontal) and columns
• Click on insert on the excel menu tab and select table
(vertical)
• In the opened dialogue box, click ok after making sure that the “My Table has Headers”
2. check
Make sure
box is that there are no blank columns and rows in the data presented
ticked
5. Go3.to the
Make sure menu
Insert each cell thisand
again labelled
click with unique
on insert column
then click labeling or header
on “PivotTable”
• 4. After
In thethat tabularize
opened yourBox
dialogue dataselect
by using
thethe
rangefollowing
of datasteps
to be analyzed
• Then Selectwhere
• select a cell containing
you want to data
put the PivotTable analysis either the “The New
Worksheet”
• Click onorinsert“Existing
on the Worksheet”;
excel menu tab Forand
convenience select the New Worksheet
select table
and•it will
In the opened dialogue box, click ok after making sure data
display analysis option at the left of our table of we “My
that the are analyzing.
Table has
• Get excel document containing
Headers” check box is ticked list of clients who are currently on ART in your facility
• In the opened data set, rename the sheet by double clicking on the sheet name
5. Go to the Insert menu again and click on insert then click on “PivotTable”
• Select all data in the sheet by select all the sheet (You can use Ctrl + A as a shortcut)
• To Insert • In thetableopened
>> Godialogue
to Ribbon BoxTab>>
selectInsert>>Table
the range of data (Orto beaanalyzed
use shortcut as “Ctr + T”>>
click Ok.

97

72 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 PARTICIPANTS MANUAL

Then select where you want to put the PivotTable analysis either the “The New
•
Worksheet” or “Existing Worksheet”; For convenience select the New
Worksheet and it will display analysis option at the left of our table of data we
are analyzing.
Figure 23. Steps to change the unstructured Excel data into structured tabularized form

4 2

Tabularized data f oTX_Curr

data ready for Analysis 3

Click a cell and

Press ctrl+A

5
Select where to put the
analysis PivotTable

Figure 23. Steps to change the unstructured Excel data into structured tabularized form
Advantages of data tabularization
1. Every record is housed on one row and automatically named columns 98
2. Each Column contains a type of data e.g., MRN, UAN, Sex, Age etc.
3. There are no blank rows or columns. Note: blank rows will create difficulties during data analysis
4. All inbuilt excel functions can be easily used in tabularized format
5. When new individual data is inserted it will automatically incorporate it in the table without need
to adjust the already built dashboards or calculation.

PA RT I C I PA N T ’ s M a n ua l 73
 Exercise: Generate line list of clients currently on ART and Tabularize the data it
using above steps

FILTERING DATA IN EXCEL

Data filtering is used to filter the content of a column based on the intended variable or attribute;
Steps to filter content on excel
1. Click the column with the variable for filtering
2. Click “Data” tab on the Ribbon
3. Click on the “down Arrow” displayed to filter
4. Check or uncheck the intended variable in the displayed box below

Figure 25. Data Filtering in excel

Exercise: In the above tabularized table filter list of currently on ART clients to view
only Male clients taking Regimen 1J (TDF + 3TC + DTG).

SORTING DATA IN EXCEL

Steps to sort data in excel
1. Click the variable intended for sorting from any column
2. Click Data tab on the Ribbon, then click the Sort command.
3. The Sort dialog box will appear
4. Decide the sorting order (either ascending or descending) ....
5. Once you’re satisfied with your selection, click OK; the cell range will be sorted by the selected
variable and sorting criteria

Exercise: In the above tabularized table sort list of currently on ART clients by increasing
age order (from Smallest to Largest). What is the minimum age of client on ART from this
example? What is the maximum age?

74 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Calculation of Measures of central tendency and dispersion using excel
Objectives:
• Calculate the measures of central tendency (Mean, Median and Mode) from individual level data
generated as report from EMR-ART using excel for analysis
• Calculate the measures dispersion from individual level data generated as report from EMR-ART
using excel for analysis
• Check outliers, incomplete and inconsistent data generated as report from EMR-ART using excel for
analysis

ART focal person in the facility wants to understand the Average, Minimum or Maximum age of
clients currently on ART in the facility from the current month of EMR-ART report. He/she also
wants to know the types of regimen being used by clients on Care for estimation of ARV drug of
the facility. How can the focal person know such data?
To know such measures, we need to calculate measures of central tendency (mean, median, mode)
and this can be done by excel. We will discuss this below.

Calculation of central tendency from tabularized excel data

The following are the steps used to calculate measures of central tendency (Mean, Median and Mode)
and measures of dispersion (Variance and standard deviation) using excel
1. Get the line list TX_Curr report from EMR-ART database
2. Change the data into table form (tabularize it) as described above
3. Click on one cell with data in the table & then click on “Design” tab in the ribbon
4. Check the “Row Total” in the design ribbon. At the bottom of the table (last row of the table), it will
auto calculate, and row total will be displayed. The row total will be added with the following inbuilt
functions
a. Average
b. Count
c. Count Numbers
d. Maximum
e. Minimum
f. Sum
g. Standard Variation
h. Variance
i. More Functions (can proceed to further analysis options accordingly)
Figure 6. Excel functions in excel to calculate measures of central tendency & dispersion

PA RT I C I PA N T ’ s M a n ua l 75
5. Select the column or variable intended for calculation of central tendency, variance, or simple count,
such as age of the client. When the measure is selected for the column automatically the value will
be displayed in the Row total in the bottom.
Note: The functions to be used for analysis should be based on the type of data for analysis. Mean,
Median, Mode and variance are done for numerical data like age and viral load. Categorical data like
Sex, Regimen type and Follow Up status will be summarized by count (frequency) included in the same
drop-down content.
For more functions like mode, median or other calculations, if needed to be calculated from the row
total use the following steps,
1. Click on the More Functions at the bottom the drop down
2. In the insert function dialogue box write the function intended as “Median” or “Mode”. To locate
the function intended please make sure the spelling of the function is correct
3. Then select the name & category of the function, click Go then click ok

Exercise
Generate last month’s Treatment Current (TX_Curr) Report of Yekatit 2012 of your facility from the
EMR-ART and answer the following questions. Use steps in the above section of this document
1. From the list of clients currently on ART, count the total number of clients using unique ART
number, ARV Regimen or client Name from Total Row.
2. From the same list, calculate the Mean, Median, Mode, Minimum, Maximum and range of age of
clients currently on ART
3. Explain the normality of the age of the client based on the age of the client by using the values for
the mean, median, mode and range of clients’ age. Do you observe any outliers?
4. Calculate the measures of dispersion. i.e., variance and standard deviation for age of clients
currently on ART
5. Based on your finding, explain how to solve data quality issues identified on this analysis using excel,
Explain the reason for abnormal age like maximum age 120.

“PIVOT TABLE” IN MICROSOFT EXCEL

Pivot table is one of the powerful functions in excel that can effectively be used to summarize and
make sense or drive meaning from a big unstructured data. Pivot table can be used to perform the
following basic but rather difficult analysis to perform using manual methods.
1. Compare clients by treatment outcomes and by regimen category
2. Calculate percentage of clients by viral suppression & other statuses
3. Add default values or create calculated columns & other calculations
4. Build refreshable analysis and update automatically using inbuilt functionalities
5. Others more….
STEPS TO CREATE A PIVOT TABLE
1. Enter your data into a range of rows and columns or get readymade exported data in excel from
EMR-ART. E.g., Get Treatment Current line list of clients with TPT, Viral Load, Cervical ca Screening
and Treatment data from the Excel extractor tool.
2. Sort your data and arrange data in rows and columns (tabularize it). See the above section on how
to tabularize unstructured data.
3. Select/Highlight, select a table or range of data to create pivot table. If all data in a sheet is to be
analyzed, click select all or use (Ctrl+A) after selecting one cell from a sheet

76 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 PARTICIPANTS MANUAL
4. Click on “insert tab” and click on “PIVOT” on the left upper corner in the excel ribbon
PARTICIPANTS MANUAL
5. In the opened dialogue box select if the PIVOT to be included in new sheet or existing (for more look
at figure three in the above section)
5. In the opened dialogue box select if the PIVOT to be included in new sheet or existing
Decide how to display
(for a data
more look in thethree
at figure pivotinanalysis
the aboveby selecting variables to put in the rows and columns.
section)
Column will contain variables for disaggregation and rows will represent category of data intended
Decide how to display a data in the pivot analysis by selecting variables to put in the rows and
for display or analysis.
columns. Column will contain variables for disaggregation and rows will represent category of
6. Drag andintended
data drop a for
field into or
display “ROWS”, “COLUMNS” and “VALUES” in PivotTable Fields.
theanalysis.
7. Drag and drop a field into the “VALUES” area.
8. Remove6. fields
Drag no
andlonger
drop a field
wantedinto by
thedrag
"ROWS",
and “COLUMNS” and “VALUES”
drop them out in PivotTable
of the PivotTable Fields.
Fields box
7. Drag and drop a field into the "VALUES" area.
8. Remove
Figure 7. Selecting fields
values onno longertable
a pivot wanted by drag and drop them out of the PivotTable Fields box

PIVOT
Table Area PIVOT
Table Fields

Select Field to be
analyzed from here

Drag & Drop Field to be

analyzed here (In Filters,
Columns, Rows & Values

Note: -Figure
to effectively interpreted
26. Selecting values data analyzed in PIVOT, consider the following points
on a pivot table

Note: to
• Variables - tobeeffectively
droppedinterpreted
in COLUMNS, data analyzed
ROWSin PIVOT,
and consider
FILTERS be categorical
the following
should points variables (Like sex,
regimen• typeVariables
or agetocategory
be dropped butin noy
COLUMNS, ROWS and FILTERS
age (continuous should be categorical
variable)
• Put (drag &variables (Like sex, regimen
drop) variables intended type
fororcounting
age category but noy age
categorical (continuous
variables variable) and ROWS
in COLUMNS
• to
• Variables Putbe(drag & drop)
dropped in variables intended
values should notfor counting
have categorical
null value. Prefer Name
E.g.,variables of client
in COLUMNS andor Sex which
ROWS
is required field than VL Result which may not be available for some clients if test is not done in that
Variables
specific•visit. to be dropped
Sex cannot in values
be (required) shouldin
missed not data. E.g., Prefer Name of client
have null value.
EMR-ART
• After data or Sex which using
is analyzed is required field than
PivotTable VLoriginal
the Result which
datamay
willnot
be be available
kept for some
unaffected clients
in the original table,
if test is not done in that specific visit.
even if we perform any manipulations in the PivotTable. Sex cannot be (required) missed in EMR-ART data.
• After data is analyzed using PivotTable the original data will be kept unaffected in the
Example: Performance monitoring
original table, even if we team ofany
perform Adama Hospitalin
manipulations wants to look at the report of 7,567 clients
the PivotTable.
who are currently on ART on Tir 2012 EFY by regimen type, sex, and age categories. The data clerk,
Mr. Tolera used PIVOT table to summarize the report. Before performing the analysis, the data clerk
performed the following activities.
104
• Used EMR-ART to generate and export list of reported clients from line list of “Treatment Current
Report Analysis” after full updating of clients on ART.
• Reviewed the data and converted it into table using (Ctrl+T) table

PA RT I C I PA N T ’ s M a n ua l 77
 clients who are currently on ART on Tir 2012 EFY by regimen type, sex, and age categories. The
data clerk, Mr. Tolera used PIVOT table to summarize the report. Before performing the
analysis, the data clerk performed the following activities.
• Used EMR-ART to generate and export list of reported clients from line list of
“Treatment_Current_Report Analysis” after full updating of clients on ART.
• Reviewed the data and converted it into table using (Ctrl+T) table
• Inserted PIVOT PIVOT
• Inserted table table
to summarize
to summarizethethedata
dataafter making
after making sure
sure notnot blank
blank row row
• Dragged• Dragged and dropped
and dropped Regimen
Regimen ROWS,
intointo ROWS,Sex
Sex in COLUMNS
in COLUMNS & Count of Patient
& Count Name Name into
of Patient
VALUES.into TheVALUES.
reportThehasreport
beenhas been summarized
summarized in table,
in table, Tolera
Tolera wantsto
wants tochange
change the
the counted numbers
counted numbers into proportion from the total
into proportion from the total clients in each sex category clients in each sex category

Figure 8. Analyzing TX_Curr by Regimen and Sex

See the
result here
Drag &
Drop

Figure 27. Analyzing TX_Curr by Regimen and Sex

From the data summarized in PIVOT we can simply tell total number of clients on ART in each sex and
regimen category. In addition to this the proportion of the clients from each regimen and sex category
From
can be the data
simply summarized
generated in PIVOT
by the we can steps.
following simply tell total number of clients on ART in each
sex and regimen category. In addition to this the proportion of the clients from each regimen
1. Click on one of the values generated as Grand Total or Male/Female total
and sex category can be simply generated by the following steps.
2. Right click and select “shows value as” and select “% of Column Total” or “% of Row Total” or any
105
other intended calculations. The calculation will change automatically
3. If you want to save the previous and current calculation separately,
a. Before performing this calculation Select Previous Table generated
b. Right click copy or “Ctrl+C” and paste in any other cell
4. If another PIVOT table needs to be added in the same page, simply click new cell on the same sheet
and repeat steps to create PIVOT table as mentioned above

78 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 a. Before performing this calculation Select Previous Table generated
b. Right click copy or “Ctrl+C” and paste in any other cell
4. If another PIVOT table needs to be added in the same page, simply click new cell on the
same sheet and repeat steps to create PIVOT table as mentioned above

Double click & rename column label

See the Result Here

GROUPING CONTINUOUS VARIABLES INTO CATEGORICAL DATA

GROUPING CONTINUOUS VARIABLES INTO CATEGORICAL DATA
Some continuous
Some continuousvariables
variableslike
likeage,
age,viral
viralload
load or
or others maybe
others may beneeded
neededtotochange
changetoto categorical data
categorical
for calculation, summarization or for better understanding of data for decision making.
data for calculation, summarization or for better understanding of data for decision making.
UseUse
thethe
following tips
following toto
tips group
groupororcategorize
categorizecontinuous data into
continuous data intocategorical:
categorical:
1. Insert PIVOT in new blank cell of the same sheet
2. From tabularized excel table select one column with continuous data like Age
3. Drag and drop it on ROWS 106
4. In VALUES drag and drop Client Name
5. In the displayed result table select the range of ages need to be grouped & group name will display
above, right click and select group, rename by double clicking>> continue for other age ranges to be
grouped
Example: - After calculation of clients currently on ART by regimen and sex category; the data clerk
wants to explore the number of clients by age categories like <1, 1-4, 5–9, 10–14, 15-19, 20–24, 25–
49, 50+ years and explore more differences among various age categories, regimen & sex
1. Create new PIVOT in new cell
2. Dragged and dropped age of the client in the COLUMNS
3. Drag and drop Client Name in the ROWS

PA RT I C I PA N T ’ s M a n ua l 79
4. Select range of values to be grouped>>0-4 and right click>>Group>>Rename Group 1 Created as
1-4>>do the same renaming for all other age ranges after selection. Finally, we can count all clients
by age category.
Example: - from the exported line list report of the viral load line of the EMR-ART make the following
analysis using PIVOT table
1. How many clients in the line list are having suppressed viral load and how many of them are having
unsuppressed results?
Solution:
• Get Line list of Viral Load clients by Going to the EMR-ART Line List.
• Change to the line list report into table and insert PivotTable
• Select the Viral Load status on the row and Sex in the values (Count of sex)
• The viral load status will be presented in four different Categories as Detectable, Suppressed,
Undetectable and Unsuppressed. Group the detectable and Unsuppressed as Unsuppressed
and Suppressed and Undetectable as Suppressed.
2. How many and what proportion of the viral loads results are with indication first viral load with
indication 6 months’ viral load?
Solution:
• In the PivotTable drag and drop routine_viral_load and target separately
• Drop count of MRN in values section the blanks will shows undocumented indication which
should be considered for data completeness issues
3. How many of the clients in this list are aged under 15 and above?
Solution:
a. Select age of the client and drop in the rows section
b. Count of Sex in the values section
c. Select one of the age and right click, group as under 15 and 15 and above

Exercise on PIVOT Table

1. Extract Treatment Current Data of your facility for Yekatit 2013 with TPT data and perform the
following analysis using PivotTable
a. What Many Clients currently on ART are Males and Females?
b. Divide clients in the 5 years age group starting from zero
 How clients are in the age group less than 15 and above 15 years age category?
 How Many of the clients in the Age group >= 15 are Females?
 In which age category are most of the clients on ART in your facility?
 Are the any clients in age category of 80 and above? What should be done to valid the
age of the clients who can be considered as outliers?
 How could we identify the list of clients as outliers based on age distribution?
c. Show the distribution of the clients on ART by their ART start date and Sex.
 In which age category are most clients currently on ART in your facility?
 How many of clients under 15 years age are females?
d. From the Treatment Current Line List and TPT data answer the following question related to
TB Preventive Therapy
 Out of clients currently on ART, how many of them started and how many of them
have completed TPT?
 Show the distribution of TPT start date and completed date.

80 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
  Are there any clients with TPT start date documented without TPT start date? What
could be the possible reason for this and how we should approach such kinds of data
quality issues in your facility?
 How many clients on ART have incomplete TPT Start date? Is there any difference In
the TPT completed date recording for clients Transfer-In and other clients started
ART in the facility?
 How many clients on ART have incomplete TPT type (use column TPT_Typechar for
analysis)? How many clients have incomplete TPT type and discuss how this data
incompleteness could be improved
 Are there any clients with invalid TPT start date and completed date? What should be
done to correct such kinds of data quality issues?
e. From the treatment current line list extracted report, answer the following questions
related to the nutritional screening result of the clients
 How many of the clients have undocumented nutritional screening result? What
should be done to correct this for consecutive months?
 Explain the relationship of proportion of clients Overweight, Undernourished and
Normal?
 Out of clients documented as Undernourished in the facility how many of them
therapeutic food provided? Do you think there is a reason to suspect of presence of
data quality issue? If yes, what could be done to correct that?
f. Analyze clients currently on ART by their follow-up date and explain the frequency of clients
in relation to the follow-up date they appear for care in the facility. How could we plan clinical
schedule by using distribution of clients next visit date?
2. Extract Treatment Current report for Yekatit 2013 of your facility with viral load data and answer
the following questions by analyzing the data using PivotTable.
a. Analyze number of clients eligible for viral by Next visit date. In which month, more clients
eligible for viral load? Explain how this data could be used for patient level clinical decision
making?
b. How could the change of this data through time be used for continuous decision making
for viral load service quality improvement by comparing clients eligible at baseline and
consecutive months on care on ART?
c. How many clients in your facility has documented pregnancy status from the Treatment
Current Line List with Viral Load data? (Assume all female clients aged 15 years and above
should have documented Pregnancy status)
 Among clients having appointment in February 2021 how many of them are pregnant
and how many of them have pregnancy status is unknown/undocumented?
 What do you think the reason for high number of female clients’ pregnancy status is
not known or documented? What should be done to correct this?
d. What proportion of viral loads received in 2020 have documented suppressed viral load
result?
 What proportion of females of this category has documented pregnancy status? If not,
what should be the measure to correct this situation?

PA RT I C I PA N T ’ s M a n ua l 81
9.2. DATA PRESENTATION & VISUALIZATION USING EXCEL
CHARTS OR GRAPHS

ART focal person in the facility wants to view multiple data in the exported re-
port of the EMR-ART to be displayed in various displays like Pie, Bar Graphs, Line,
Combo box and other charts. This has been repeatedly requested by the facility
head for proper display & monitoring of the performance of the facility for prop-
er decision making & progress follow. These displays are set as standard by the
facility for proper Art care program monitoring monthly by the Data Quality Im-
provement Team of the department.
In the above section we have seen multiple data summarization and graphical presentation of data
using PivotTable. This section will elaborate more on steps to follow in processing of individual level
data exported from EMR-ART to be displayed in attractive display charts or graphical presentation
that could properly show the reality of performance in these facility in understandable and simple
representations.
Examples for pie-chart: Use the treatment current analysis from EMR-ART to display clients by sex in
simple pie chart and label the chart properly according to standard.
1. Export Treatment current report from the EMR-ART and use steps in the using of PivotTable and
summarize clients by sex in table. Put Sex in the ROWS and Patient Name in the columns field.
2. Select content of this table>>Insert>>PIVOT Charts>> Select Pie and specify the type that
interests you from the options displayed
3. Click on the displayed chart & from the (+) sign in the right upper corner of the page select data
labels and percentage of each proportion will be included in the chart
4. Double click the Title and rename it as you like but indicating “What, Where, When”. Finally copy
the chart by right clicking on the chart and COPY or (Ctrl+C). Paste in any document like
power-point, word, excel or word as you like &/or print
Note: - when labeling a chart title make sure that following requirements are mentioned in the content
• What: - content of the chart
PARTICIPANTS MANUAL
• Where: - it is happening
• When: - the time of anMANUAL
PARTICIPANTS event
• Context of the data: - from which clients is the variable has been calculated
• Data Source: is it from EMR-ART, DHIS 2.0 or Other Sources
2

Sex Number of Clients

4. Rename the chart titles properly
F 449
M 227
Grand Total 676

Start by Selecting data 1

from PIVOT table

82 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
Examples Column or Bar Graph: - From the same treatment Currently on ART line list from
EMR-ART Tolera wants to compare clients by age group starting from 0 by 5 years age interval
and display it in column or Bar chart. He followed the following steps to represent these clients
Examples Column or Bar Graph: - From the same treatment Currently on ART line list from EMR-ART
Tolera wants to compare clients by age group starting from 0 by 5 years age interval and display it in
column or Bar chart. He followed the following steps to represent these clients in the display
1. Export Treatment current report from the EMR-ART and use PIVOT tables and summarize clients
by age. From PIVOT table fields put Age in the ROWS and Patient Name in the columns field.
2. But to group the age right click>>select group and select starting point 0 and end the default higher
age and interval as 5
3. Select content of this PivotTable>>Insert>> PivotChart>> column and click ok after selecting type
of chart that interests you
4. Click on the displayed chart & from the (+) sign in the right upper corner of the page select data
labels and chart tile
5. Double click the Title and rename it as you like. Finally copy the chart by right clicking
PARTICIPANTS on the chart
MANUAL
and COPY or (Ctrl+C). Paste in any document like ppt as you like &/or print
6. Rename one sheetMANUAL
PARTICIPANTS of the excel as “Dashboard” and paste this chart for later building of interactive
dashboard

2. Click on Insert
3. Select PivotChart

7. Click here to
insert slicer
variable

6. Click Here to Rename charts

according to standard

1. Select PivotTable

5. Tick Here to enable

chart Titles & Data Labels

4. Chart Appears Here

In the above chart let us assume we wanted to view trend of age change for male and female
In the above chart
clients letsame
in the us assume
displaywe wanted
chart, followto view
next trend of age change for male and female clients in
steps
the same display chart, follow next steps
 Click Slicer in the right section>select sex as Criteria and click ok>when each sex is
 Click Slicer in the right section>select sex as Criteria and click ok>when each sex is selected the
selected
display will change the display will
automatically change automatically
changing changing content for viewers.
content for viewers.
Example Line List Report: Tolerate wants to compare number of clients visiting the facility by visit or
follow-up date and display the trends in line chart to monitor the distribution of clients currently on
Example Line List Report: Tolerate wants to compare number of clients visiting the facility by
ART by their visit date.
visit or follow-up date and display the trends in line chart to monitor the distribution of clients
currently on ART by their visit date.

• Insert PivotTables from the above Treatment Current Line List with Column Name
PA RT I C I PA N T ’ s M a n ua l 83
FollowUpDate_GC and count of MRN in the values section
• Select the PivotTable and Insert Pivot Charts>>Select Line chart and label the chart and
• Insert PivotTables from the above Treatment Current Line List with Column Name FollowUpDate_
GC and count of MRN in the values section
• Select the PivotTable and Insert Pivot Charts>>Select Line chart and label the chart and titles as
mentioned in the above section of the chart
• Resize and correct the color composition-based ion your preference and make the necessary
adjustments for exporting for printing of the charts

Dashboard

ART focal person in the facility wants to build interactive dashboard by using ex-
cel exported data from the EMR-ART for clients Currently on Treatment by their
Sex, Age Category, Regimen type and ARV Doses. He only wants to change list of
clients for future months and use this dashboard to shows changes over time by
only refreshing the dashboard prepared already.

Dashboard can effectively be used to graphically represent detail data for a glance views of key per-
formance indicators (KPIs) relevant to a particular objective or intended performance standards.
Dashboard can be designed to be interactive based on the data source to be changed based on need.
The following are some of the basic uses of dashboard in a facility
 Show the changing performance of a facility based on some basic Key Performance Indicators or
Metrics like Treatment Current, Viral Load, Age change in clients etc.
 Show performance gaps based on real time data to correctly
 Displays data in easy-to-read fashion
 Help decision makers identify gaps without great efforts & help them make efficient decision
 Make the reporting easier by avoiding disconnected different parts
Before trying to build dashboard using the EMR-ART data and available line list or extracted data, the
following main points should be addressed.
1. What kind of performance questions will my dashboard answers/addresses?
a. What is the Treatment Current of my facility by age category and sex?
b. What factors affected the TX_Curr when compared to previous months?
c. What do the TX_Curr clients distribution look like when seen by Follow-Up Date?
d. How many clients are having appointment by specific date and month?
e. What the type of ARV regimens and dose are clients visiting the facility in the next two
months taking?
2. What kind of data could be represented in the dashboard?
a. The dashboard question that could be answered by dashboard could vary based on the data
available, but it can enable viewers drill though, slice and dice data using various techniques.
Two to three KPIs could be sliced using slicers interconnected to each table and be able to
produce the necessary data view effects accordingly
3. Dashboard produces interactive data effects that can help user/decision makers use their imagina-
tion powers for prediction of future trends & adjust necessary factors
Example Dashboard: -
Use the above treatment current line list from EMR-ART, of all regions and explain the followings a
single dashboard that can support appropriate data updating features.
1. Show the report completeness for each facility included under each region

84 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
2. the trend of the treatment Current report of the five regions Addis Ababa, Amhara, Oromia, Dire
Dawa Tigray and Gambela.
3. Show the Treatment Current Trend by ARV Regimen Type and Dose,
 Try to explain the ASM enrollment status based on ARV dose days starting from the
appearance of Covid-19 pandemic in Ethiopia
 Is there any difference between the ARV regimen type being used when compared starting
from Tikimt 2012?
4. Try to show the age change of the treatment current population by making comparison between
Tikimt 2012 and Tikimt 2013 for under 15 vs 15 & Above yrs. Clients. What proportion of the
population has grown and moved to 15 & above yrs. of age?
Insert appropriate display charts for each category of data mentioned above and prepare interactive
dashboard that can be used to monitor the Treatment current related changes by using a treatment
current Dashboard for your facility.
Steps to Follow
1. Get the data source like the EMR-ART extracted data
2. Prepare a standardized excel sheet labeled properly with standardized rows and columns
3. Change this content into table (Ctrl+T) and insert PivotTable as shown above
4. Insert Display Tools for each category of data based on the nature of data. E.g. Insert Line chart to
monitor the trend of the TX_Curr of the facility from month to month.
5. After preparation of as much relevant displays in this pivotable showing Key Performance
Indicators, create a sheet Name TX_Curr_DashBoard.
6. Copy each relevant chart into TX_Curr_DashBoard sheet and use appropriate slicers and connect
to the necessary analysis tables accordingly.
The following example demonstrates how to prepare Treatment Current Dashboard to easily view
the trend in Treatment Current Report, Clients TPT Completion Status, Nutritional Screening Result
and Multi-Month Dispensing Trend. This data is found from the EMR-ART extractor tool version 2.0
and the following procedure demonstrates each step required to build an interactive dashboard which
could be named as TX_Curr_DashBaord
Preliminary Stage to Build TX_Curr_DashBaord
Get the data from any Facility EMR-ART by using the Excel extraction tool as demonstrated in the
above sections and build the Treatment Current trend of your facility by using simple line chart from
pivot charts for the number of months you could extract from the system.
• Generate a month TX_Curr data (Tikimt 2012) and save in excel sheet separately and tabularize
using the steps specified in the above section
• Generate next month TX_Curr data (Hidar 20112) and copy this month TX_Curr except the header
and Paste (append) at the end of the previous month TX_Curr tabularized data. For each month fol
low the same steps and save this data as a database.
o Just add consecutive month data to build a database by adding new month TX_Curr at the
end of this original file, which helps us to make comparison for each month
o No need to build new visualizations, only perform data refresh to update charts from the
added new month treatment current report of the facility
This example demonstrates a sample data taken from the EMR-ART extracted data of Zewditu Memo-
rial Hospital from Tikimt-Nehase 2012 in Ethiopian Calendar.

PA RT I C I PA N T ’ s M a n ua l 85
STAGE ONE:
• Build a Treatment Current PivotTable as displayed below by putting the right variables in the rows
and columns. On the filters put Report Year to select only 2012 (if more the table)
• Insert Pivot charts and select Line Chart with proper labeling and Title naming
• Make appropriate axis titles, Chart Titles and Data labels from the plusPARTICIPANTS
Sign atMANUAL
the right top of the
chart inserted.
PARTICIPANTS
• Insert & SLICER MANUAL
from the ribbon above to slice along the data layers (see below)
This chart This
demonstrates a clear
chart demonstrates visualization
a clear for
visualization for thethe trend
trend of theof the TX_Curr
TX_Curr ofwhich
of the facility the facility
is which is ap-
propriate for properfor
appropriate comparison for any
proper comparison increase
for any increaseand decrease
and decrease in value.
in value.

TX_Curr Trend of Zewditu Hospital, Tikimt-Nahase 2012, from EMR-ART

Extracted Data
7750

7721
7700 7704
7689 7691

7664 7659
7650
# Cients

7631
7625
7613
7600

7584 7580
Insert SLICER as above
7550 to slice the data
presented for proper
variables.
7500
Tikimit Hidar Tahisas Tir Yekatit Megabit Miazia Ginbot Sene Hamle Nehassie

Reporting Month

SLICERs are visual filters or drilling option in the data visualization for simple examination for each contents by
excluding unnecessary references. To insert slicer, click the chart, go to insert, select slicer, and select the variable to be used
for slicing the visualization.

After this step

After this step
1. Select in the
1. same
Selectexcel doc,excel
in the same create doc,acreate
newasheet andand
new sheet rename
renameit it as TX_Curr_DashBaord
as TX_Curr_DashBaord
2. Copy the above
2. Copy chart and related
the above chart and slicers and paste
related slicers in inthe
and paste theTX_Curr_DashBaord
TX_Curr_DashBaord
3. All the next3.displays to be
All the next prepared
displays will bewill
to be prepared copied with
be copied withrelated slicers
related slicers and
and stored
stored in thein the appropriate
appropriate location for the dashboard and will be connected
location for the dashboard and will be connected to each other to build interactive to each other to build interactive
contents
contents appropriate to
appropriate to view and for data monitoringview and for data monitoring

STAGE
STAGE TWO: - TWO: -
• Build theCurrent
• Build the Treatment TreatmentDataCurrent Data visualization
visualization usingusinga acolumn
column chart
chart by by
ARVARV
Dose Days
Dose asDays
shown as shown
in the next section.
in the next section.
o Take the ARVDoseDays in the Rows, count of ID in the values and Report Month in the
o Take the ARVDoseDays in the Rows, count of ID in the values and Report Month in the
columns of the PivotTable Fields
columns ofothe PivotTable
Insert Bar Graph Fields
for the PivotTable and make the proper labeling of chart title, axis title
o Insert Bar Graph and data the
for label PivotTable and make the proper labeling of chart title, axis title and
data label o Insert the necessary slicer if needed like reporting month to simplify the view of the
o Insert the necessary slicernumber
data in limited if needed like reporting
of columns. E.g., Compare month
the firstto simplify
three monthsthe view of the data in
of 2012
• This display shows the same treatment current clients
limited number of columns. E.g., Compare the first three months of 2012by the number of ARV Dose Days
dispensed
• This display shows the for
samethe client at the last
treatment visit the client
current clients is counted
by thefor treatment
number ofcurrent of the month.
ARV Dose Days dispensed
See below the display for this chart
for the client at the last visit the client is counted for treatment current of the month. See below the
display for this chart
116

86 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 PARTICIPANTS MANUAL

PARTICIPANTS MANUAL

TX_Curr By ARV Dose Days Zewditu Hospital, Tikimt-Nahase 2012, from EMR-ART
5000

4659
4328
4297
4500

4062
3986

3957
3718
4000

3201
3500

2850
2828
Yekatit
3000
Megabit
# Clients

2224
2500 Miazia

1933
Ginbot
2000
Sene

1294
Hamle
1500
1107
1062

Nehassie
980

808
1000
738
723

574
407
356

352
324
309
272
267

265

500
219
198
192
161

153
150
102

97
88
80
75

70
56
42

0
30 60 90 120 150 180
ARV-Dose-Days

The above comparison of the same clients by the ARV Dose Days taken in the respective month for
The above comparison of the same clients by the ARV Dose Days taken in the respective month for
proper comparison of the of
proper comparison dose changes
the dose changesbased onstandards
based on standards of comparison
of comparison between thebetween the months. The
months. The
above data above
clearly
datashows the ARV
clearly shows the ARVdose
dose provision trend
provision trend changeschanges related
related with with the
the occurrence occurrence of the
of the
COVID 19 pandemic in Ethiopia
COVID 19 pandemic starting
in Ethiopia from
starting from March
March 2020. 2020. We can
We can clearly clearly
see from seethat
the chart from the chart that
continuous continuous
decline indecline
number of clients
in number taking
of clients oneand
taking one andtwotwo months
months of ARVof ARV
dose and dose
sloppy and
rise insloppy
the rise in the
number of clients put on 6 months of ARV dose from Yekatit
number of clients put on 6 months of ARV dose from Yekatit to Nehase, 2012. to Nehase, 2012.

As mentioned
As mentioned above copyabovethe
copychart
the chart prepared and
prepared andslicer andand
slicer pastepaste
it in theitTX_Curr_DashBaord sheet
in the TX_Curr_DashBaord sheet
and continue to the next step.
and continue to the next step.
SLICER inserted for one type of chart should not be repeated for other kind and if the same slicer needs to be
SLICER inserted foruse
used, we will one type
another of chart
technique should
to connect notby be
the charts usingrepeated for other
the same SLICER kind and if the same slicer
as required.
needs to beInused, we will
the Dashboard use
sheet another
make technique
the necessary to connect
space adjustments and arrangethe charts
the charts by using
in space the same
efficient manner to SLICER as
required. make all charts visible in a single view without any scrolling by decision makers.
In the Dashboard sheet make the necessary space adjustments and arrange the charts in space
Try to make all visuals as simple as possible and labeled clearly as of standards
efficient manner to make all charts visible in a single view without any scrolling by decision makers.
Try to make all visuals as simple as possible and labeled clearly as of standards

STAGE THREE: -
• Build the Treatment Current Data visualization using a bar chart by for clients started and
completed TB Preventive Therapy (TPT) as shown in the next section 117

 Take the INH Start Date GC and INH Completed Date GC in the Rows, count of INH Start
Date GC and INH Completed Date GC in the values and Report Month in the columns of the
PivotTable Fields. From the and in the rows only include the Year for analysis (exclude Quar-
ter and Date from analysis). By doing data will be counted by the TPT start and completed
year.
 Insert Bar chart from the PivotTable and make the proper labeling of chart title, axis title and
data label.

PA RT I C I PA N T ’ s M a n ua l 87
 Date GC and INH Completed Date GC in the values and Report Month in the columns of
the PivotTable Fields. From the and in the rows only include the Year for analysis
(exclude Quarter and Date from analysis). By doing data will be counted by the TPT start
and completed year.
 Insert Bar chart from the PivotTable and make the proper labeling of chart title, axis title
and data label.
 Insert the necessary
 Insert slicer
the necessary if needed
slicer if neededlike
likereporting month
reporting month toto simplify
simplify the the
viewview
of theof the data in
limiteddata
number of columns.
in limited number ofE.g., Compare
columns. the firstthe
E.g., Compare three
firstmonths of 2012
three months of 2012

TX_Curr clients stared & Completed TPT, Zewditu Hospital, Tikimt-Nahase

2012, from EMR-ART
1000
871
900 826 816
773
800
700
Number of Clients

600 539
493
490
500
400 Started INH
300
INH started & Completed
171 170 187 Completed INH from TX_Curr sliced by
200 141
139 130
128 152 148
106
106 9090 99
3837 6262
Reporting Month
100
2 2 2121 1515 4
0

TPT Start or Completion Year

In the above display clients are counted by their TPT start and completion year and will help to
In the above display clients are counted by their TPT start and completion year and will help to
 Clearly visualize distribution of clients started and completed by each respective year for sliced
 Clearly
month visualizeSo,
of reporting. distribution
by only of clients at
looking started and completed
and comparing by each
clients respective
started and year for sliced TPT in
not completed
month of reporting. So, by only looking at and comparing clients started and not
a specific year, we can take corrective actions for follow-up of clients started but no completion completed TPT date
in a specific
not recorded, year, wedid
or clients cannot
takecomplete
correctivetheactions for follow-up
therapy. of clients
For example, in started
2019 outbut of
no539 clients started
TPT incompletion
2019 onlydate notthem
99 of recorded,
haveorcompleted
clients did not complete
TPT, the therapy.
the remaining For example,
significant numberin 2019 out clients
of 440
of 539 clients started TPT in 2019 only 99 of them have completed TPT, the remaining
can be identified from the line list and proper data or treatment completion measures could be per significant
number of 440 clients can be identified from the line list and proper data or treatment
formed.
completion measures could be performed.
 Wecan Wealso
canlook at data
also look quality
at data issues
quality issueslike
likeclients withTPT
clients with TPT completion
completion yearyear documented
documented but but TPT
start date not documented
TPT start in the database
date not documented and make
in the database necessary
and make corrections
necessary byidentifying
corrections by identifying specific
clientsspecific
by name and solve the data documentation problem.
clients by name and solve the data documentation problem.
As mentioned above copy the chart and slicer and paste it in the TX_Curr_DashBaord sheet prepared
As mentioned above copy the chart and slicer and paste it in the TX_Curr_DashBaord sheet prepared
and continue to the next step.
and continue to the next step.
STAGE FOUR:
118 Result
 Build the Treatment Current Data visualization using a Pie Chart for Nutritional Screening
included in the line list report
 Take the Nutritional Screening Status in the ROWS and count of Nutritional Screening Result in the
columns.
 Insert Pie chart from the PivotTable and make the proper labeling of chart title, axis title and data
label.

88 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
  Build the Treatment Current Data visualization using a Pie Chart for Nutritional Screening Result
included in the line list report
 Take the Nutritional Screening Status in the ROWS and count of Nutritional Screening Result in
the columns.
 Insert Pie chart from the PivotTable and make the proper labeling of chart title, axis title and
data label.

TX_Curr clients by Nytritional Screening Result,

Zewditu Hospital, Tikimt-Nahase 2012, from EMR-
ART

1318, 17%
Normal
584, 8% 3214, 42%
OverWeight
UnderNourished
2497, 33% Incomplete

In the
In theabove
abovechart
chartclients
clientsareare
counted basedbased
counted On theOnnutritonal screening
the nutritonal result andresult
screening follow-up
andoffollow-up
clients of clients
based on
based ontheir screening
their screening result will help
result willfor thefor
help better
thetreatment outcome ofoutcome
better treatment clients onofchronic care.
clients on chronic care.
The pie
The piechart
chartcan bebe
can sliced by the
sliced by Reporting monthmonth
the Reporting and each month
and eachperformance can be compared
month performance can befor compared for
improvement and data quality issues. In the above display one third of the clients currently
improvement and data quality issues. In the above display one third of the clients currently on ART on ART are
overweight
are which which
overweight may bemay due to
besome
due tokind of data
some quality
kind issues
of data and which
quality issuesmay bewhich
and investigated
may be at investigated at
individual level from line list to ensure implementation of necessary corrective
individual level from line list to ensure implementation of necessary corrective measures.measures.

As mentioned
As mentioned above
abovecopy the the
copy chartchart
and slicer
and and paste
slicer andit paste
in the TX_Curr_DashBaord sheet prepared
it in the TX_Curr_DashBaord sheet prepared
and continue
and continuetoto
thethe
next step.
next step.
STAGE FIVE:
STAGE FIVE: - Dashboard
- Dashboard Activation
Activation and
and Data Data Refresh
Refresh
After
After allallthe
the necessary
necessary charts
charts and slicers
and slicers are together
are brought broughtinto
together intothe
one sheet, one sheet,adjustments
following the following adjust-
ments
should be should
done be
for done forand
effective effective and
attractive attractive
dashboard dashboard content
content
 Arrange the chart or visuals into logically coherent order
 Arrange the chart or visuals into logically coherent order
 Use Usespace
 the the space available
available efficiently and
efficiently and make
makevisuals
visualsonon
thethe
right and and
right slicers on theon
slicers left
theor left
upperor upper
section of the
section of dashboard
the dashboard
 Connect
 Connect eacheach slicers
slicers andandtables
tablesto
to create
create interactive
interactiveor or
dynamic display
dynamic display
 Do not create too may displays, only focus on pertinent data from the raw line list data
 Do not create too may displays, only focus on pertinent data from the raw line list data

TX_Curr_DashBoard 119

After merging all the previously prepared charts and relevant slicers the following type of dashboard
will finally be produced. Click contents included in the slicer to slice and drill down to each detail of the
data required. This Dashboard could be used as input for the Data Quality Improvement /performance
monitoring team of HIV program on regular performance and data quality reviews.

PA RT I C I PA N T ’ s M a n ua l 89
 PARTICIPANTS MANUAL

TX_Curr_DashBoard

After merging all the previously prepared charts and relevant slicers the following type of dashboard will finally be produced. Click contents
included in the slicer to slice and drill down to each detail of the data required. This Dashboard could be used as input for the Data Quality
Improvement /performance monitoring team of HIV program on regular performance and data quality reviews.
Figure 29. View of Sample EMR-ART TX_Curr Dashboard

Figure 28. View of Sample EMR-ART TX_Curr Dashboard

Exercise on Pivot Table and Excel Dashboard
Prepare ART Data Completeness Dashboard by using the EMR-ART Providers View Line List and
120to
answer the following questions.
Go to provider’s view on the page of the EMR-ART software and generate the Treatment Current cli-
ents line list for Yekatit 2013 and build interactive dashboard that can answer the following questions.
 What proportion of the Treatment Current clients in your facility have complete documentation of
the Address data?
 What proportion of the Treatment Current clients in your facility have completed TPT and how
many are currently taking TPT?
o Prepare the list of clients not completed TPT for printing to be provided to the care providers
to be used as evidence for daily clinical decision
o List out clients currently taking TPT and list out the expected date of TPT completion
 What proportion of the Treatment Current clients in your facility are eligible for viral load and what
percent of them are viral load done in appropriate time.
o Prepare list of clients eligible for viral load to be provided to the clinicians for decision
support in provision of day-to-day chronic care
 What are proportion of clients currently on ART are enrolled in Appointment Spacing model and
how many of them are not eligible for ASM but not enrolled in care
o Prepare list of clients eligible for ASM and provide the printout to care providers to be used
as a reference for provision of the care in facility visit of the client.
o How many of the clients eligible for ASM have one years or less stay on ART Care
 How many of the client’s ICT screening data documented in your facility at least one time and list out
clients never been screened for ICT before?
 Build a Dashboard for ART data quality in separate sheet including the following charts and slicers
for proper view of the quality of data and evidence-based decision making by the performance
monitoring team or decision makers

90 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 o Pie Chart for clients labeled as Green and Yellow for address data completeness, for TPT,
Viral Load, Appointment Spacing and Partner and Family Based ICT.
o Insert Slicers using the following variables, Sex, Age Category (less than 15 & 15 & above), by
ART dose code
o Is there any difference between the completeness of Address, TPT, and Viral Load data for
clients started ART in the past 12 months and earlier? To answer this question, insert new
column as Months on ART less than 12 and 12 & and above months and use it as slicer.
 Demonstrate how we could insert each month new data in the already built ART Data Quality Dash-
board for Analysis and Comparison. Think of how to keep copy of the previous month data quality
status before replacing it all with current month data for facilitation of the comparison and docu-
mentation of the progress towards better quality data.

9.3. DATA VISUALIZATION USING POWER BI

9.3.1 Overview
Power BI is a data visualization tool that converts data from different data sources to interactive dash-
boards and reports.
It allows its users to create visualizations, reports, and dashboards. A Power BI report is one or more
pages of visualizations, graphics, and text, whereas A dashboard in Power BI is an overview of key
indicators of interest. A power BI is a data analytics and visualization software. It helps users analyze
data from various sources and build reports & dashboards. It can be connected to any dataset format
(Excel files, SQL databases, BI warehouses, Cloud data, web pages, and more), link one table with oth-
ers, creates clickable visualizations, and then share them with your audience.
Power BI is available for use in two ways: Desktop and Service. Desktop is the version that is accessi-
ble through ones’ local computer. It is free to use and includes many features such as the Power Query
Editor. Service is the cloud-based version of Power BI.

9.3.2. Steps in Power BI using

Step 1. Connecting or Import data to Power BI
To do so in Power BI Desktop Dashboard press on Home > Get Data > Choose a data source > Connect.
Step 2: Format data in Power BI Dashboard
The next step is to import data from any data source and visualize it in Power BI. To do so, press on
Home > Get Data > Choose a data source > Connect as shown below.

PA RT I C I PA N T ’ s M a n ua l 91
Step 3: Formatting the Data in Power BI
Before loading the data, you have an option to format it. To do this, select the “Transform Data” avail-
able at the bottom of the page. In case you want to load the data without any formatting, click on
“Load”.

When you select Transform data, a summarized form of the table is shown. This is shown below.

Step 4: Creating Visualizations for your Power BI Dashboards

After transforming raw data into a workable format, the next step is disseminating that refined data.
In Power BI, we can create a dashboard by pinning visualizations from BI reports that are published
using the Power BI desktop.

9.3.3. Power BI features

Upon opening Power BI have three views on the interface for datasets: Report, Data, and Model. On
the Report view, you can create visualizations and reports. On the Data view, you see the dataset used
for your reports. On the Model view, you can see the relationships between different tables for the
data model.

92 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
In the Power BI Desktop Report view, you can build visualizations and reports. The Report view has
six main areas:

1. The ribbon at the top, which displays common tasks associated with reports and visualizations.
2. The canvas area is in the middle, where visualizations are created and arranged.
3. The page tab area at the bottom, which lets you select or add report pages.
4. The Filters pane, where you can filter data visualizations.
5. The Visualizations pane, where you can add, change, or customize visualizations, and apply
drill-through.
6. The Fields pane, which shows the available fields in your queries. You can drag these fields onto the
canvas, the Filters pane, or the Visualizations pane to create or modify visualizations.

PA RT I C I PA N T ’ s M a n ua l 93
Power BI Filters
Filters in Power BI sort data and information based on some selected criteria. That is, you can select
particular fields or values within fields and view only the information related to that.
There are four types of filters in Power BI.
1. Visual-level Filters
The filters applied to the visual-level are applied directly on individual visualizations. Such filters are
applied both on data and calculation conditions used within a visualization.
2. Page-level Filters
Page-level filters are for a particular page within a report whereas reports are usually of multiple pag-
es. You can apply certain filter conditions on a selected page within a report. Each page in a report can
have a different set of filter conditions applied to it.
3. Report-level Filters
The report-level filters are the filters that you use to apply a filter condition on the entire report. The
report-level filter will get applied to every visualization and page of a report. Thus, unlike visual-level
and page-level filters, report-level filters are generalized filters.
4. Drill-through Filters
Using drill-through filters, you can create a page that mainly focuses on the specific entities.

9.3.4. Power Query

Power Query is a data transformation and data preparation engine. We can use the Power Query Ed-
itor to connect to one or many data sources, shape and transform the data, and then load that model
into Power BI Desktop.
To get to the Query Editor, select Edit Queries (Transformation data) from the Home tab of Power BI
Desktop. Once a data connection is established, the Query Editor appears to be something like the
following:
1. The ribbon has many buttons, which are now active to interact with the data in the query.
2. In the left pane, queries are listed and available for selection, viewing, and shaping.
3. In the Centre pane, data from the selected query is displayed and available for shaping.
4. The Query Settings window appears, listing the query’s properties and applied steps.
The following illustration shows a few of the transformations available in Power Query Editor.
Working with Power BI.

94 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 PARTICIPANTS MANUAL

PARTICIPANTS MANUAL

When your query is where you want it, you can have the Editor apply the changes to the
data model.
When your query is where you want it, you can have the Editor apply the changes to the data model.
To do so, select Close & Apply from Power Query Editor’s File menu.
To do so, select Close & Apply from Power Query Editor’s File menu.

128
PA RT I C I PA N T ’ s M a n ua l 95
SECTION 10:
Information use for Action
Section Objectives
By the end of this section, participants will be able to:
- Understand steps of information use for patient level and program level decision making
- Identify and prioritize problems and identify solutions
- Understand root cause analysis techniques
- Define platforms for information use and dissemination

10.1. STEPS OF INFORMATION USE FOR ACTION

Using information for evidence informed decision making requires systematic analysis and interpreta-
tion of both patient level and aggregate data. To facilitate information use for action, we use step wise
approach that can help health care providers and managers identify problems, prioritize problems,
analyze, and identify root causes of problems, develop interventions and follow the implementation
of interventions. We use a five steps approach to information use, that is customized from the national
health information use manual.

Figure 27. Steps of information use for action

STEP 1: IDENTIFY PROBLEMS

In the data analysis section, you have been able to determine whether patient level and program level
HIV/AIDS indicators are progressing well or not. Those indicators that have low performance level
shows that there is a problem associated with it. Identification of problems in providing care to pa-
tients or identification of gaps in program implementation is the first step in information use. It is only
when you can identify problems that you can improve HIV/AIDS programs or clinical outcome of pa-
tients.

96 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
 How do we identify that there is a performance gap?
We identify performance related problems when there is:
 Performance gap compared with the plan (target)
 Individual level clinical service provision is not as per the standard
 When the existing phenomena differ from our clients’ expectations
 Concerns from individuals or from the community

Example
Health Center X provides service to a catchment population of 25,000. The ART clinic in the health
center provides clinical care and support service to PLHIV. The health center has planned and achieved
HIV/AIDS related activities in 2012 EFY as described in the table below.
Table X: HIV/AIDS Performance in Health Center X

Plan for 2012 Achievement (First Last Year’s

Indicator
EFY 6 months) performance
Number of PLHIV currently on ART 850 860 810
Early viral load suppression rate 90% 75% 85%
ART Retention Rate (at 12 months) 85% 90% 70%
Percentage of non-pregnant women
living with HIV on ART using a modern 90% 95% 85%
family planning method
Proportion of Sexually Transmitted
100% 80% 65%
Infection (STI) cases tested for HIV
TB Screening for HIV positive Clients 100% 70% 90%

The data clerk in the clinic also identified that 10 patients are lost to follow up at the end of the second
quarter, and viral load test is not done for 20 PLHIV who are on the 6th month after ART is initiated.
Based on the case scenario given above, discuss the questions that follows.
 How do you identify performance problems in HC X?
 Which indicator has performance gap? Which one is on the right track?

How can a clinician identify problems during HIV/AIDS patient care? What would
be the role of the providers view feature of the EMR-ART in your facility for
patient level and program level decision making in your facility.
Discuss in your group, with documented examples and present to the large
audience.

During clinical care of individual patients, clinicians can identify the presence of problems by taking
patient history, Physical and lab examination. Moreover, the EMR-ART will support clinicians and data
clerks to identify patients whether specific HIV/AIDS services are provided or not. Example: EMR-
ART will show clinicians/data clerks whether a viral load test is done or not; shows number of patients
lost or dropped from ART. Based on this, the clinician can take appropriate actions.
Example:

PA RT I C I PA N T ’ s M a n ua l 97
 A physician working in the ART clinic of health Center X identified that 4 of his patients that attend-
ed ART clinic did not have viral load test in the last three years, but they have been taking ARVs con-
tinuously. Checking the status of viral load test for individual patients help identify individual patient
problems and appropriate action can be taken
 A physician working in the ART clinic has identified that the nutritional status of one of his patients,
by assessing his BMI, shows severe malnutrition (SAM). Identification of the nutritional status of the
patient can help manage the patient accordingly.
Key Points
Problem identification can be done both at patient level and program level
 At patient level: problem identification for individual patients can be done during clinical care inter-
action and/or by checking the individual record of patients on the EMR-ART application. For example:
You can identify patients that are lost, patients that are eligible for viral load but not tested, INH/TPT
status of patients etc.…
 At program level: HIV/AIDS problem identification can be done by analyzing the program indicators
using the analysis technics described previously and comparing performance with plan, with bench-
marks, previous time trends etc.…
STEP 2: Prioritize PROBLEMS
Is it possible to tackle all identified problems at once? If not possible Why?
What can we do if we have identified many problems at once?

After problems are identified, it is good if all the problems are solved as soon as possible. However, due
to resource constraints, all problems might not be able to be solved at once. As a result, prioritization
of problems using different prioritization criteria can be used to identify priority problems that need
immediate action. Prioritization is most useful for program level problems. Patient level problems
should be addressed immediately. Example: Patients that are lost from treatment should be searched
and get back to treatment as soon as possible. Patients with severe drug side effects should be man-
aged immediately.
Prioritization is making decisions on how limited resources could be best allocated to priority health
problems or needs. For HIV/AIDS program level performance, the following prioritization criteria can
be used to prioritize problems.
- Magnitude of the problem: How big is the problem?
- Severity of the problem: how serious is the problem to individuals or to the community?
- Feasibility: How easy is it to reverse the problem?
- Community Concern: Is it a pressing issue of the community?
Using the criteria, we rank problems and select the top priority problems and solve them first
Figure 30. Template for prioritization of problems
Problem Magnitude Severity Feasibility Community Total Rank
Concern

Point System: 3 High, 2 Medium, 1 Low

98 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E
STEP 3: Investigate the Cause
In the above two steps, you have learnt how you can identify and prioritize problems of HIV/AIDS pro-
gram performance. In this third step, you will learn how to investigate the underlying and root causes
of problems.

Root cause analysis (RCA) is a systematic process for identifying “root causes” of problems or events
and an approach for responding to them. There are several methods to analyze and investigate root
causes of problems. Some of these techniques and tools that can be used by health care providers
and managers include: Fishbone Diagram, Pareto diagram, bar charts, flow charts, the logic tree etc.…
These formal tools can be used to group causes into categories and can help to dissect an occurrence
into its contributing factors and component parts and pinpoint where an error or failure occurred.

Fishbone Diagram

This is one of the root cause analysis tools that can help identify the root causes of a given problem.
This type of analysis enables you to discover the root cause of a problem. This tool is also called a
cause-and-effect diagram or an Ishikawa diagram. Major causes can be categorized in different forms.
It can be categorized based on Supply-demand categories of the problem or can be categorized based
on the six building blocks of a health system: leadership/governance, service delivery related, human
resource, supplies/drug related, health information, and finance or any other category. Selection of
the categories should be done in a participatory manner through consensus usingMANUAL
PARTICIPANTS prior knowledge
and understanding of the program. The type of categories chosen depends on the context.
PARTICIPANTS MANUAL
Figure 29. Fishbone Diagram

Figure 31. Fishbone Diagram

The Tree Diagram– 5 WHYs
The tree
The diagram analysis
Tree Diagram– is done by constructing a problem tree, that includes a set of assumptions
5 WHYs
associated with the problem and its consequences. Each major cause is then further explored by ask-
The tree diagram analysis is done by constructing a problem tree, that includes a set of
ing “Why did this happen?” The second layer of causes is further explained by repeating the question
“Whyassumptions associated
did this happen?” Suchwith the problem
questioning and its consequences.
continues Each major
for looking in-depth cause
to find theisroot
thencause.
further
explored by asking “Why did this happen?” The second layer of causes is further explained by
repeating the question “Why did this happen?” Such questioning continues for looking in-depth
to find the root cause.

Viral
Suppression PA RT I C I PA N T ’ s M a n ua l 99
rate is low
 The tree diagram analysis is done by constructing a problem tree, that includes a set of
assumptions associated with the problem and its consequences. Each major cause is then further
explored by asking “Why did this happen?” The second layer of causes is further explained by
repeating the question “Why did this happen?” Such questioning continues for looking in-depth
to find the root cause.

Viral
Suppression
rate is low

Poor
Repeated OI
adherence

Why? Why? Why? Why?

Exercise on root cause analysis

Exercise
Suppose that a health on root
center causeofanalysis
has 10% its patients on ART lost to follow up. Discuss on the problem
and do a root causeSuppose that
analysis a health
using centerdiagram
fish bone has 10%orofthe
its patients on ART lost to follow up. Discuss on the
tree diagram
problem and do a root cause analysis using fish bone diagram or the tree diagram
STEP 4: Develop Intervention
Once problems are identified and the causes are identified, we need to develop an intervention that
can improve the performance of the program. A detail action plan has to be developed for each identi-
fied problem based on the root cause analysis. The action plan should include details of the identified
solutions, the responsible person for each action point, the detail timeline and cost of each interven-
tion. The following template can be used as a guide for it. 134

Table 9. Action Plan Template

Identified Problem Possible Causes Solutions /Action Responsible person/ time- Cost
(Performance Gap) Points Unit line

STEP 5: Implement the action Plan and follow up

Once an action plan is developed in step 4 above, it has to be implemented and follow up is important.
The team leader or the person responsible for the activity has to follow the implementation of the ac-
tivities to ensure that the various tasks are accomplished within the set deadlines.
10.2. Platforms for information use
For HIV/AIDS program monitoring and evaluation, the major platforms for information use are:
- Annual planning sessions
- performance data should be used
- HIV/AIDS case team meetings
- Performance Monitoring Meetings
- Facility Quality Improvement Team Meetings
- Review Meetings
- Catchment area meetings
During the above meetings, HIV/AIDS data should be analyzed, interpreted, problems identified, and
appropriate actions should be taken to improve program performance.

100 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E

SECTION 11:
Monitoring and evaluation of HIV/AIDS information use
Measuring the level of HIV/AIDS information use at health facilities is essential. To monitor whether
HIV/AIDS unit of a health facility has been using its data for evidence informed decision making, we
need to have indicators and checklist that we should use to monitor it periodically. The indicators and
checklist help the facility and external supervisors to assess the level of HIV/AIDS data use in the fa-
cility.
Indicators to monitor HIV/AIDS information use
- Availability of data use platforms in the facility: PMT, QIT
- Number of data use meetings held
- Regular comparison of HIV/AIDS program performance versus target
- Performed root cause analysis for HIV/AIDS indicators
- Developed action plan for HIV/AIDS program gaps
- Availability of minimum display charts in the HIV/AIDS unit
- Number of review meetings held for HIV/AIDS program
- Number of supportive supervisions provided to the next lower level
Checklist (Sample):
Table 10. Sample checklist to monitor the presence of information use in HIV/AIDS units
S. No Question Answer Remark
Does the facility perform data completeness desk review at Yes
HIV/AIDS unit in the past one week with documented minute? No
Did the health facility perform data quality assurance for HIV/
Yes
1 AIDS program?
No
If yes, what
Does the facility have a performance Monitoring team on HIV/ Yes
2
AIDS department in the past one month? No
How many performance monitoring meetings were held
3
during the last 3 months?
4 Information use minute book available?
Comparison of Plan versus Achievement done based on key Yes
5
HIV/AIDS indicators? No
HIV/AIDS program performance gaps are identified by com- Yes
6
paring achievement against target in the past one month No
Is Root cause analysis is done for low performing key indica- Yes
7
tors, at least once in the review period No
Minimum Display charts are available in the HIV/AIDS unit
(ART unit)
Cohort Monitoring Chart
8
TX_CURR Trend chart
Viral Load Tests and Suppression
Others, specify………
Yes
9 Review meeting on HIV/AIDS program performance?
No

PA RT I C I PA N T ’ s M a n ua l 101
REFERENCES
1. Federal Ministry of Health (2018). Health Data Quality Training Manual. Addis Ababa, Ethiopia
2. Federal Ministry of Health (2013). HMIS Information Use Guide. Addis Ababa, Ethiopia.
3. Federal Ministry of Health (2018). Information use Training Manual. Addis Ababa, Ethiopia
4. Federal Ministry of Health (2017). HMIS Indicator Reference Guide. Addis Ababa, Ethiopia
5. ICAP at Columbia University (2019). EMR-ART module training manual. Addis Ababa, Ethiopia
6. Measure evaluation (2017). Data Quality for Monitoring and Evaluation Systems.
https://www.measureevaluation.org/resources/publications/fs-16-170-en
7. Measure Evaluation, USAID and PEPFAR (2012). Data Demand and Use: An Introduction to
Concepts and Tools.
8. Mikael Gebre-Mariam et al. An Electronic Medical Record (EMR) Implementation Framework for
HIV Care and Treatment Facilities in Ethiopia
9. PEPFAR (2019). Monitoring, Evaluation, and Reporting Indicator Reference Guide – MER 2.4.
10. UNAIDS (2014) 90-90-90: An ambitious treatment target to help end the AIDS epidemic
11. World Health Organization (2004). Improving Data Quality: A guide to developing countries.
Geneva, Switzerland
12. World Health Organization (2008). Health Metrics Network: Framework and standards for
country information systems. Geneva, Switzerland
13. World Health Organization (2017). Data quality review: a toolkit for facility data quality
assessment, Module 1 Framework and Metrics. Geneva, Switzerland
14. World Health Organization (2017). Data quality review: a toolkit for facility data quality
assessment, Module 2 Desk Review of Data Quality. Geneva, Switzerland
15. World Health Organization (2017). Data quality review: a toolkit for facility data quality
assessment, Module 3 Data Verification and System Assessment. Geneva, Switzerland
16. World Health Organization. 2008. Antiretroviral Therapy (ART). Geneva, Switzerland
http://www.who.int/hiv/topics/arv/en/

102 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E

List of Workshop Participants
S.N Full Name Organization
1 Biruk Abate MOH- PPMED
2 Mirtie Getachew MOH-DPCD
3 Teklu Lemessa MOH
4 Alemtshay Abebe MOH
5 Yakob Wondrad MOH
6 Seidu Fentie RHB - Amhara
7 Bogalech Yadess RHB-Benishangul G
8 Masresha Molla RHB - Amhara
9 Dinkayehu Tadele RHB- Oromia
10 Rania Mohammed RHB - Harari
11 Abdusemed Ali RHB - Harari
12 Yonas Mekonnen RHB - Oromia
13 Seyfedin Yisehma RHB-Benishangul G
14 Abdi Shlbeshir RHB - Somali
15 Mesud Mohammed RHB-Tigray
16 Daniel Aseffa RHB - Tigray
17 Fasil H/Meskel RHB - Addis Ababa
18 Nebsu Asamerew RHB - Addis Ababa
19 Berhanu Zewdi RHB - Addis Ababa
20 Abreham Wondimu RHB-SNNP
21 Bashir Ahmed RHB - Somali
22 Assiya Jaylan RHB- Oromia
23 Dhaba Ejara RHB - Oromia
24 Dejene Tolla RHB - Addis Ababa
25 Demitu Crensch RHB - Oromia
26 Deme Bedada RHB - Addis Ababa
27 Dr. Jemal Aliy ICAP
28 Bedri Ahmed ICAP
29 Desalegn Lulu ICAP
30 Fesseha Taddese ICAP
31 Getaneh Derseh ICAP
32 Tenaye Abate ICAP
33 Chaltu Yasin ICAP
34 Jemal Tadesse ICAP
35 Refissa Bekele ICAP
36 Shegaw Mulu ICAP

PA RT I C I PA N T ’ s M a n ua l 103
37 Oumer Mohammed ICAP
38 Zaid Taddese ICAP
39 Mahlet Kassahun ICAP
40 Reay Zewdi ICAP
41 Abraraw Gebre ICAP
42 Ayalew Jembere ICAP
43 Tadesse Alamir ICAP
44 Tarekegn Walacho ICAP
45 Takele Halfom ICAP
46 Yosef Zeru ICAP

104 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E

PA RT I C I PA N T ’ s M a n ua l 105
 Printed by: maxgreen 0920 20 82 85

106 H I V/A I D S DATA Q UA L I T Y A N D I N F O R M AT I O N U S E