Download General Findings from the four situational - MeSH

HIV CASE BASED SURVEILLANCE AND PATIENT TRACKING SYSTEMS
RESULTS FROM FOUR SITUATIONAL ASSESSMENTS IN
HIGH BURDEN, LOW-RESOURCED COUNTRIES
Conducted by the Measurement and Surveillance of HIV Epidemics (MeSH) Consortium
Members of the MeSH Routine Case-based Surveillance Working Group
Andrew Boullé, University of Cape Town
Sandy Schwarcz, University of California San Francisco
Jim Todd, London School of Hygiene and Tropical Medicine
Brian Rice, London School of Hygiene and Tropical Medicine
Richelle Harklerode, University of California San Francisco
Serge Xueref, World Health Organization
Mary Ann Davies, University of Cape Town
Mark Griswold, National Alliance of State AIDS Directors, associate member organization
Lucy Slater, National Alliance of State AIDS Directors, associate member organization
Released June 20, 2016
Page 1
Contents
ACKNOWLEDGMENTS ................................................................................................................................................ 3
EXECUTIVE SUMMARY ............................................................................................................................................... 4
INTRODUCTION .......................................................................................................................................................... 5
METHODS ................................................................................................................................................................... 6
CASE STUDIES ............................................................................................................................................................. 7
UNITED REPUBLIC OF TANZANIA ................................................................................................................ 7
Background .....................................................................................................................................7
Scope of the assessment ................................................................................................................7
Findings ...........................................................................................................................................9
Interpretation of findings .............................................................................................................13
Recommendations........................................................................................................................13
REPUBLIC OF SOUTH AFRICA .................................................................................................................... 14
Background ...................................................................................................................................14
Scope of the assessment ..............................................................................................................14
Findings .........................................................................................................................................14
Interpretation of findings .............................................................................................................19
Recommendations........................................................................................................................19
REPUBLIC OF KENYA ................................................................................................................................. 20
Background ...................................................................................................................................20
Scope of the assessment ..............................................................................................................20
Findings .........................................................................................................................................20
Interpretation of findings .............................................................................................................26
Recommendations........................................................................................................................26
REPUBLIC OF HAITI ................................................................................................................................... 28
Background ...................................................................................................................................28
Scope of the assessment ..............................................................................................................28
Findings .........................................................................................................................................29
Interpretation of findings .............................................................................................................32
Recommendations........................................................................................................................32
GENERAL FINDINGS .................................................................................................................................................. 32
RECOMMENDATIONS ............................................................................................................................................... 36
APPENDIX ................................................................................................................................................................. 39
Protocol for the situational assessment of HIV case reporting................................................................ 41
Page 2
ACKNOWLEDGMENTS
Funding for this work was provided by the Bill and Melinda Gates Foundation, the Global Fund to Fight
AIDS, Tuberculosis, and Malaria, and the World Health Organization (WHO).
We wish to thank all of the individuals from the international, national, regional, district and facility
programs who graciously participated in these situational analyses. Several individuals contributed
substantial effort to the preparation and/or conduct of these situational analyses and to them we offer
our special appreciation.
United Republic of Tanzania
Dr, Geoffrey Somi, National AIDS Control Programme, Ministry of Health and Social Welfare, Tanzania
University of California-Tanzania staff (Joel Ndayongeje, Christen Said, Susie Welty)
Republic of South Africa
Nicki Tiffin, University of Cape Town
Republic of Kenya
Emily Zielinski-Gutierrez, Centers for Disease Control and Prevention -Kenya
Wanjiru Waruiru, University of California San Francisco-Kenya
Republic of Haiti
République d’Haïti Ministère de la Sante Publique et de la Population (MSPP)/Programme National de
Lutte contre le Sida (PNLS) staff (Ermane Garcia Robin, Guethina Galbaud, Jethro Guerrier, Jean-Daniel
Hyacinthe, Rachelle Cheristin, Charles Emmanuel)
Valerie Pelletier, Centers for Disease Control and Prevention -Haïti
National Alliance of State and Territorial AIDS Directors staff (Mark Griswold, Nadjy Joseph, Moise
Weedner, Jean Rene Dorfils)
Anthony Monfiston, Organisation Panaméricaine de la Sante/Organisation Mondial de la Sante
(OPS/OMS)
Miriam Sabin, Fonds Mondial
Page 3
EXECUTIVE SUMMARY
HIV case-based surveillance (CBS) refers to an approach to surveillance that involves the reporting of
individual-level information from each person diagnosed with HIV to the public health agency responsible
for monitoring and controlling the HIV epidemic. Information along the course of disease from diagnosis
to entry into care, initiation of antiretroviral therapy (ART), viral suppression, and death are necessary to
measure progress towards epidemic control, these events are collected and maintained longitudinally.
To facilitate the development of CBS in resource-limited countries the Measurement and Surveillance of
HIV Epidemics (MeSH) Consortium conducted situational assessments in Tanzania, South Africa, Kenya
and Haiti from August 2015 through April 2016. A standardized protocol and data collection tool were
utilized to ensure a uniform approach to the situational assessments that took into consideration the
context for each country in relation to CBS. The assessment included: 1) a desk review of relevant
materials on HIV surveillance, 2) interviews with stakeholders, and 3) site visits. Prior to each assessment,
the MeSH assessment team met with government and other stakeholders from the countries to confirm
the specific focus of each assessment, adapt the protocol and tool, develop a list of documents to review,
and identify stakeholders for interviews and sites for visits. Findings from the assessments were
categorized as strengths, weaknesses, opportunities, and threats (SWOT). These findings were
summarized and presented at a debriefing session with stakeholders followed by discussion of
opportunities and recommendations.
Countries were selected to include variation on the stage of readiness for CBS. Tanzania had expressed a
strong interest in development of CBS. South Africa has a multi-facility-based integrated patient
monitoring system and a national electronic repository of all CD4 and viral load tests that may serve as a
model for capturing data from electronic systems for use in CBS. Kenya is in the early stages of CBS
implementation and Haiti has a moderately well-established CBS system that was built off of existing
electronic patient monitoring systems.
Detailed summaries were developed as a case study for each country and findings were then generalized.
The assessments revealed substantial interest in, and support for CBS as a way to more effectively
monitor the epidemic with the recognition that CBS requires additional resource. All of the countries
assessed conduct routine HIV programme monitoring that collects information on the number of persons
tested for HIV, number of new HIV diagnoses and number of persons accessing HIV care and treatment
services. At the facilities, individual level data are collected and entered into registers and individual
patient records. Current reporting of these data to the Ministry of Health is done in aggregate. Review of
systems and discussions with stakeholders revealed significant problems with data quality. Unstable
power sources and internet access was a problem in many regions which adversely impacts all electronic
and networked systems.
The findings from the situational analyses support the development of CBS systems to systematically
capture routinely collected health data to describe and monitor the epidemic for program planning and
evaluation and ultimately disease control. Although it may be possible to adapt some of the current
program and patient monitoring systems to function effectively for CBS, there was wide variation in the
systems examined and few that were open-source. The recommendations provide guidance to countries
on the development of CBS. The approach to develop CBS should match the capacity and resources of
the country and leverage existing systems. Often policies and procedures need to be updated for
implementation of CBS to ensure standard reporting and data security and confidentiality. Data
collection and use should be specified to evaluate and enhance quality of data. The MeSH protocol and
tool is provided as a resource to be utilized by countries to assess the feasibility of implementing CBS or
to evaluate and enhance an existing CBS system.
Page 4
INTRODUCTION
Universal health coverage is one of the key health targets of the Sustainable Development Goals (1). The
draft WHO Global Health Sector Strategy on HIV 2016–2021 maps the way forward towards universal
health coverage for HIV (2). The draft strategy makes clear that to achieve universal coverage for HIV it is
essential that HIV services are accurately targeted and effective. To measure the precision and impact of
HIV services, WHO released consolidated guidelines in 2015 recommending the use of ten global
indicators to collect information along the cascade of HIV care and treatment (3). Six of these indicators
use data that originate from patient diagnosis, testing and medical records.
The Sustainable Development Goals (SDG 17.18), the draft Global Health Sector Strategy, and the draft
WHO Global Health Sector Strategy recommend developing a comprehensive strategic information
system to provide high-quality, timely and reliable data disaggregated by population characteristics
across the different levels of a health care system (1-3). HIV case-based surveillance (CBS) is such a
system and refers to an approach to surveillance that involves the reporting of individual-level
information from each person diagnosed with HIV to the public health agency responsible for monitoring
and controlling the HIV epidemic. CBS is also called case notification and is conducted for many infectious
diseases in order to control and prevent disease.
In CBS, information from individual cases (i.e. persons) is maintained as disaggregate data in a database
at the national and often also sub-national level. As information along the course of disease from
diagnosis to entry into care, initiation of ART, viral suppression, and death are needed to measure
progress towards epidemic control, data on each of these events is collected and maintained
longitudinally. This is a distinctive characteristic of CBS systems and distinguishes it from aggregate
reporting. In order to accurately add new information pertaining to an existing case record, and to also
distinguish duplicate case reports (i.e. two or more reports for the same individual), a CBS system must
be designed to permit matching to identify information relating to the same individual from information
concerning a new, previously unreported individual.
A well-functioning HIV CBS system can provide timely longitudinal population-based data collected from
medical records to the local, regional, and national levels. These data can be used to determine and
describe the geographic, demographic and risk factor distributions of HIV. This in turn will improve the
quality of information provided to national AIDS program coordinators and other relevant partners for
program planning and evaluation. Interest for developing and implementing CBS systems that takes into
consideration current routine patient information systems and emerging technologies is growing among
sub-Saharan African countries and other resource limited countries.
To facilitate the development of CBS there is a need to evaluate existing data collection systems to
identify what works and understand how to leverage these for CBS. It has been suggested that once the
limitations of current systems are identified it becomes easier to elaborate a national plan for
surveillance, and that one of the best approaches to reaching a consensus for a sound surveillance plan is
to discuss the country’s needs with the main partners (4).
To identify systems that are context appropriate, feasible, scalable, and sustainable to implement CBS
the Measurement and Surveillance of HIV Epidemics (MeSH) consortium conducted situational
assessments in Tanzania, South Africa, Kenya and Haiti from August 2015 through April 2016. With
funding from the Bill & Melinda Gates Foundation, the assessments were conducted in collaboration with
WHO, the Centers for Disease Control and Prevention (CDC), Joint United Nations Programme on
HIV/AIDS (UNAIDS) , the Global Fund to Fight AIDS, Tuberculosis, and Malaria, ministries of health (MOH)
and other within country stakeholders. The main objective of the assessments was to identify the
strengths, weaknesses, opportunities and threats to CBS.
Page 5
METHODS
A protocol and data collection tool (appendix) were developed to ensure a standardized approach to
situational assessments that took into consideration the context and focus for each country in relation to
CBS. The assessment included three parts: 1) a desk review of relevant materials on HIV surveillance for
the county, 2) interviews with stakeholders, and 3) site visits.
The desk review examined standard registers and medical records that capture individual level data,
surveillance and program monitoring reporting requirements and regulations, death certificates, strategic
plans, reports and system evaluations. Interviews were conducted with persons knowledgeable about
HIV strategic information, involved with other infectious disease surveillance activities, and working in
HIV testing and care and treatment programs, vital registration, and laboratories. Site visits were
conducted to understand human resource capacity, availability of data relevant to CBS, electronic
systems, data flow, storage, transfer, quality, standards, and laboratory systems.
The protocol was intentionally designed to be a comprehensive tool that could be adapted for the
particular focus of the assessment within each country. For example, in countries with established CBS
systems the tool was used to evaluate the system whereas, in countries planning for CBS the tool was
used to conduct a situational analysis to inform the development and implementation of CBS. The
following standardized tools were developed for the assessment: Document Review Checklist, Document
Request Checklist, Interview Guide, Site Visit Checklist, and Debrief Form.
Countries were selected to include variation based on their stage of readiness for CBS. Tanzania had
expressed a strong interest in developing CBS. South Africa was included as it has a multi-facility-based
integrated electronic patient monitoring system with interoperability and a national repository of all CD4
and viral load tests that may serve as a model for capturing data from electronic systems for use in CBS.
Kenya is in the early stages of CBS implementation and Haiti has a moderately well-established CBS
system that was built off of existing electronic patient monitoring systems.
Prior to each assessment, the MeSH assessment teams (MeSH members and in-country participants) met
with government and other stakeholders from the countries to confirm the specific focus of each
assessment, adapt the protocol and tool, develop a list of documents to review, and identify stakeholders
for interviews and sites for visit. Findings from the assessments were categorized as strengths,
weaknesses, opportunities, and threats (SWOT). These findings were summarized and presented at a
debriefing session to stakeholders followed by discussion of opportunities and recommendations. More
detailed summaries were developed as case studies.
.
Page 6
CASE STUDIES
UNITED REPUBLIC OF TANZANIA
Background
In Tanzania data are collected for routine programme monitoring activities and are reported in aggregate
to a district focal person who is responsible for entering the data into the Health Management
Information System (HMIS) through the District Health Information System (DHIS-2). At present, DHIS-2
contains aggregate data from every health facility offering HIV services which is available to district,
regional and national level planners.
Various programmes and projects within the Ministry of Health and Social Welfare (MoHSW) also have
their own reporting mechanisms, including the prevention of mother-to-child transmission (PMTCT) and
the care and treatment clinic (CTC) programmes within the HIV sector. Many facilities report individual
level data to the national CTC system through the CTC2 database. The CTC2 database collects the facilityspecific CTC2 patient number but not names or a national identity number. Because they are line-listed
there is substantially greater flexibility in data analysis than in the aggregate data systems and as such,
the CTC2 data are analysed and used by the National AIDS Control Programme, (NACP) and stakeholders
for programme planning.
Both the DHIS-2 and CTC data have limitations. For example, there is a high probability that duplicate
records are present in the CTC2 and DHIS-2 databases, thereby preventing an accurate assessment of the
number of diagnosed persons or of individual care pathways.
Scope of the assessment
The goal of this assessment was to evaluate the feasibility and acceptability of and recommendations for
the next steps towards developing a secure, high-quality and timely HIV CBS system. Topics assessed
included degree of interest in CBS from stakeholders, the need for changes in policies and additional
resources (financial and human), data sources and quality of data, and the presence of unique identifiers
that could be used to link, and where relevant de-duplicate, records pertaining to the same individual.
The assessment began with a meeting of stakeholders to review and adapt the SWOT data collection tool
to meet the specific country needs, and to provide guidance on the number of facilities for site visits
based on the systems currently in use. To assess needs in both urban and rural areas with substantial
disease burden, the SWOT exercise was focused in two distinct regions in Tanzania, Dar-es-Salaam and
Mwanza.
Interviews were conducted at several levels. Nationally, interviews were conducted with individuals
delivering health services (primarily HIV), monitoring infectious diseases (including tuberculosis, sexually
transmitted infections and other diseases), and supporting interventions and services. In addition,
Tanzania-based personnel from international organisations (WHO, UNAIDS,CDC), and local implementing
partners working with the MoHSW, were interviewed. At the sub-national level, regional and district
health officers were interviewed, including the Regional Medical Officer, District Medical Officer,
Regional AIDS Control Coordinator, District AIDS Control Coordinator, Regional Tuberculosis and Leprosy
Control officer, and District Infectious Disease Surveillance personnel. Site visits were made to select
health facilities and laboratories providing HIV services in the Dar-es-Salaam and Mwanza regions. Sites
were selected based on the facility ownership (i.e. private and government) and size (i.e. regional
hospital, health centre, and dispensary). Twenty-three interviews were conducted with stakeholders at
the national, regional, district, and facility levels and nine facilities (tertiary, secondary and primary sites)
were visited.
Page 7
Page 8
Findings
The main findings from the interviews and site visits are summarized in Table 1.
Few of those interviewed were familiar with CBS. However, most were aware of other HIV reporting
systems, including CTC2. Once the limitations of aggregate data and the benefits of CBS were explained,
there was consistent support for CBS. Persons interviewed reported that selected infectious diseases are
reported on a case-basis. Confidentiality and security of monitoring and evaluation data were addressed
by excluding personally identifying variables beyond the facilities.
Review of registers and medical records indicated that the variables required for CBS to produce key
indicators along the care cascade are available. Patient-level information including names and dates of
birth are available in patient medical records. CTC and the Prevention of Mother-to-Child Transmission
(PMTCT) data collection tools are standardized but referral and transfer forms are not. The CTC2
database includes data from selected care and treatment facilities but not from HIV testing sites. It is
worth noting that most service-level systems are paper-based.
There is interest in adapting current data systems for use in CBS but these systems have limitations. The
CTC2 system is line-listed and contains all of the variables needed for CBS but only collects data once a
patient has entered care. Data from the CTC2 system are sent to the national level but without the
patient name or other personal identifiers. The CTC2 system starts at entry into care, therefore losing
anyone with an HIV diagnosis who does not obtain care. In reporting facilities the CTC2 paper forms are
completed, but not all facilities report to the CTC2 database (at the time of the assessment
approximately 50% of sites providing HIV care and treatment reported to the CTC2 database). The DHIS2
database includes HIV diagnosis as well as PMTCT data but only collects aggregate data and does not
include all of the data required for CBS. Although it appears that there is adequate staffing to meet
current reporting requirements (e.g. into DHIS2 and CTC2), CBS will likely need additional human
resources as currently data quality assurance is not being enforced (including reviews of completeness
and accuracy of data), and there is limited use of data beyond mandated program reporting.
Various information technology issues were observed: there are limitations in infrastructure,
connectivity, and interoperability (for example CTC2 and DHIS2 are not interoperable, although plans to
extract data from CTC2 into the DHIS2 reports are well advanced), and vital statistics data cannot be
linked with current HIV systems. There is also a shortage in the human resource capacity for information
technology.
Unique identifiers
Unique identifiers that would work for CBS do not currently exist in HIV testing and counselling (HTC) and
care programs. The CTC2 number appears to be unique within each facility but not across facilities. As
such, record matching and de-duplication beyond the facility is not possible. Names and dates are
collected, although not comprehensively. Currently there is no national identify number. However, there
have been discussions focused on exploring the possibility of enhancing the CTC system to maintain
identifiers at the national level.
Policy
National policies that are applicable across all programs regarding case-based disease reporting, security,
and confidentiality of paper and electronic records do not exist. There are no data release policies.
Page 9
Table 1. Summary findings from SWOT analysis for key areas of interest in Tanzania.
Overall findings
Political and
organisational
support,
governance, and
human
resources
HIV care and
treatment
cascade; data
collection and
reporting
STRENGTHS
The current system is a strong
foundation for CBS.
Systems for collecting data from all
points of the cascade exist.
Procedure for reporting aggregate
data from clinics to sub-national /
national level can be built on.
Understanding of the need for a
unique identifier is present
throughout HIV program monitoring
system.
Strong support from PEPFAR partners
for clinics, regions and MoHSW.
Committed, hard-working health
workers in many clinics.
Data collection in CTC and PMTCT is
comprehensive and standardised.
Referral and transfer process
between clinics exists (not operated
universally and documentation is
weak).
WEAKNESSES
Over reliance on paper-based registers
and aggregate reporting.
The unique identifier for patients (facility
CTC number) is inadequate for CBS.
Full potential of data are rarely utilized at
all levels.
The interconnectivity between the patient
records and DHIS reporting system
needs to be developed.
Data quality needs to be addressed.
OPPORTUNITIES
Define a unique identifier which
would link HIV services and
facilitate de-duplication.
Enhance data quality systems
and supervision, including the
use of the data.
Build a career structure for
those working with the data to
improve understanding of the
data at the various levels.
THREATS
If CBS requires additional data
collection, it will put pressure
on the resources and staff
time, which may undermine
clinical care and/or data
quality.
Data quality and utilization
needs to be addressed.
Separate policies within different
departments and programmes (e.g. key
populations).
Lack of training and resources.
Reporting / data staff does not have
sufficient time available.
Partner activities not harmonised.
Financial sustainability of comprehensive,
high-quality, timely reporting system.
Integrate reporting within
MoHSW, including
tuberculosis and infectious
disease surveillance as part of
an improved reporting
system.
Commitment from partners
and donors to support CBS.
Build a career structure for
those working with data to
improve understanding at the
all levels.
Increase links between clinics
and programmes and
strengthen links between
HTC, PMTCT and CTC bringing
in communities / key
populations.
Needs of donors for particular
reports / data may
undermine single reporting
mechanism.
Over reliance on partners may
limit sustainability.
New technology often seen as
a threat to existing jobs.
Subsequent testing among persons
diagnosed with HIV.
High attrition from HTC to CTC.
CTC electronic data is not a surveillance
tool, more a basic electronic medical
record.
Multiple testing points of
entry. Need to ensure
comprehensive reporting
including community and key
populations.
CTC does not focus on
Page 10
Data are collected in health facilities
(paper record) and sometimes used
for clinical management of patients.
Data are reported to district and
national level (indicators).
Tracing and tracking missed
appointments exists (not operated
universally).
Unique identifier for CTC (useful
within but not necessarily between
clinics).
Feasible to add name to HIV testing
and counselling (HTC) register.
Reporting
systems for HIV
and other
notifiable
diseases
DHIS-2 is universal reporting
mechanism in Tanzania; monthly
reports of aggregate data.
Quarterly report of aggregate CTC
data.
District reporting meetings provide
means for identifying and sharing
best practice and for networking.
Periodic analysis of CTC data by the
National AIDS Control Program.
Programming at national level exists
and is trusted by MoHSW.
Linking clinical
data with
laboratory,
pharmacy and
Lab systems exist with CD4 in many
clinics and viral load in regional labs.
Most labs have electronic systems,
(not linked to other systems).
Electronic data seen as an “extra” task not much utilisation within clinics.
UIs not based on master facility list, and
names not consistently available for deduplication.
Guidelines and standard operating
procedures are not widely understood /
used.
Data quality assessment infrequent; most
do not check quality regularly.
Over estimation of lost to follow-up due to
lack of data linkage between care
facilities.
Variable support from partners and not
available to all clinics.
DHIS-2, CTC and the Presidents Emergency
Plan for AIDS Relief (PEPFAR) all have
separate reporting requirements and all
relate to aggregate data.
Potential duplicate counting of HIVdiagnosed people due to repeat testing
and lack of de-duplication.
Potential underestimation of HIV
diagnoses due to incomplete reporting.
Data printed out at clinic and re-entered
at districts into DHIS-2.
Private clinics do not report fully.
Little use of available data.
No links between clinic and lab data.
No links between clinic and pharmacy
data.
Rapid tests used for HTC are not reported
Enhance referral systems.
Update CTC forms to include
CD4 counts and viral load
results.
Develop and universally
introduce UI at time of HIV
diagnosis.
Tanzanian national identifier
and the master facility list
being developed and would
enable better de-duplication
if collected as part of patient
care at all health facilities.
monitoring patient care or
facilitating linkage between
sites.
Data quality assessment
systems are not in place at all
levels.
Universally applied quality
management procedures,
including quality control in
laboratories, not in place.
Lost referral and transfer
forms inhibit patient access
to services.
Tanzania government create a
single reporting system.
DHIS-2 Tracker (a feature
within DHIS2) can
accommodate patient level
reporting.
Since much data exists in CTC2,
develop a system to
automatically extract and
report on clinic-level data.
De-duplication procedures
needed.
Currently unable to extract
variables from clinic and
laboratory databases.
Lack of resources and staff at
paper-based clinics and
private health facilities for
reporting.
Link lab data and pharmacy
data from clinics through a
unique identifier.
Enhance vital registration to
Staff shortages and lack of
training will compromise
completeness and quality of
data.
Page 11
other data
Many pharmacies have electronic
system for drug ordering (not linked
to CTC).
to lab.
Lack of resources to identify deaths that
occur in the community.
link deaths to clinical records.
Page 12
Interpretation of findings
Based on the assessment, it was determined that HIV case-based surveillance is possible for Tanzania and
there is a high degree of acceptability from stakeholders and implementing partners. Basic systems are in
place to build a strong CBS system in the country. Although not contained in a single national database
currently, the data regarding diagnosis, entry into care, CD4 results, and ART initiation are available.
Tanzania has not yet adopted routine viral load monitoring for patients on ART. Gaps in data for CBS are
mainly in obtaining unique identifiers; although data quality was also found to be a significant issue.
There is staffing at the national, sub-national and facility level specific to HIV monitoring and evaluation
and surveillance. In implementing CBS, it needs to be determined what role these existing staff would
have and/or what new cadre of personnel would be required. It will be essential that staff with some
capacity to manage and report out data be included within the organizational structure of facilities. There
is a need for additional trained staff that can effectively manage and analyse surveillance data.
It was determined that there are gaps in policies/laws regarding HIV reporting and data confidentiality.
These gaps will need to be addressed to facilitate successful implementation of CBS.
Recommendations








For successful CBS implementation, policies in the following areas are required: HIV reporting,
patient confidentiality, data security, data release, and vital registration (reporting of deaths).
Specific areas for CBS must be addressed:
o Explore options for unique identifiers that can effectively de-duplicate records; this could
include enhancing the CTC system to maintain identifiers at the national level; a pilot is
recommended to test various options for unique identifiers
o Determine sentinel events to include in CBS and identify methods to obtain these
o Develop data quality assurance including internal checks
o Determine role of facilities, district and regional level staff in CBS
o Identify technology systems to develop or enhance for CBS
o Determine frequency of reporting and reportable variables
o Develop approach for implementation (e.g. staggered or national, with or without pilot
testing)
Relevant Technical Working Group to coordinate information technology activities so systems are
not being created in parallel and to ensure that these systems facilitate CBS.
Implement routine systems for systematic data validation and quality assurance that include
internal checks.
Streamline paper-based data collection tools and reporting processes.
Develop methods to ensure completeness of death records.
Enhance efforts to improve and maintain quality of laboratory techniques including improved
and utilized standard operating procedures.
Encourage a culture of data use through training ministry staff in data analysis, interpretation,
and exploration of outcomes and endpoints.
Page 13
REPUBLIC OF SOUTH AFRICA
Background
To create a comprehensive national health information system, the South African National Department
of Health (NDOH) is required, under the National Health Act, to facilitate and co-ordinate the
establishment, implementation and maintenance of health information systems at the national,
provincial and local levels. HIV care and treatment is included in the national system whereas HIV
surveillance is part of the HIV monitoring and evaluation system.
South Africa has implemented a national, longitudinal patient monitoring system for persons on ART Three Interlinked Electronic Registers (TIER.Net). Prior to national implementation of TIER.Net there were
over 40 HIV program based data systems in South Africa, each of which collects different variables.
TIER.Net includes three options for patient management systems; a paper-based system (Tier 1), an
electronic version of the paper-based system (Tier 2), and a fully networked electronic medical record
(EMR) system (Tier 3). This approach allows facilities to choose the tier that meets their resources, with
outputs from all three tiers being aligned so that they can be consolidated into a single reporting system.
To date, the middle tier has been widely deployed, with a limited number of sites using the other tiers.
TIER.Net uses standard clinical records, paper registers, software, and data exchange standards for
moving data between systems. Individual-level data are collected and reported nationally including
patient names and other identifiers. Although TIER.Net has modules for HIV testing and pre-ART care,
neither of these is currently used.
South Africa also has a national laboratory data warehouse that collects all CD4 cell and viral load tests
from person receiving HIV care. The National Health Laboratory System (NHLS) provides laboratory and
related public health services to over 80% of the population, and nearly all public sector services, through
a national network of laboratories. CD4 and viral load testing is standard for ART patients in South Africa
and is performed at centralized laboratories all using the same national laboratory information system
(LIS), which facilitates electronic reporting. The LIS captures individual level data, including identifiers
provided by the facilities ordering the tests. A scannable barcode is provided for each unique patient.
There are a number of distinct personal identifiers in use in South Africa. Some of these are HIV-specific,
while others are for general use.
In summary, in South Africa patient-level, longitudinal data are collected in a single patient monitoring
system, regional laboratories conduct CD4 and viral load test results, and systems include personally
identifying data. These may serve as a model for a national case surveillance system.
Scope of the assessment
The goal of this assessment was to assess how the TIER.Net system, the NHLS, and patient identifiers
could be used for CBS. The Western Cape Province was selected as a region where the TIER.Net system
and identifiers could be explored. The NHLS is based in Johannesburg.
Seven interviews were conducted with stakeholders at the national and regional level, and the NHLS data
warehouse. Projects were examined in order to understand the different methods used to uniquely
identify patients (Jembi, Health Systems, Provincial data centre, HE2RO), and two facilities and one
regional laboratory were visited.
Findings
The main findings from the interviews and site visits are summarized in Table 2.
TIER.Net
Page 14
The TIER.Net system has been successfully implemented. However, most facilities lack reliable internet
and infrastructure required for the Tier 3 system. Site visits are performed as part of monitoring and
evaluation standard operating procedures to ensure data quality. Available TIER.Net modules for HIV
diagnosis and pre-ART are currently not used, although it seems feasible to implement these given
adequate resources. TIER.Net can be used to estimate the number of patients on ART and, with the
addition of the HTC and pre-ART modules, could be used for CBS.
NHLS
The NHLS includes scannable barcodes for each test that uniquely identifies the patients and assists with
data quality. The NHLS data warehouse has developed an algorithm utilizing the various identifiers to
match and de-duplicate tests. The NHLS data warehouse currently uses the data to estimate care cascade
measures by taking the number of people with a CD4 test as a proxy for the number entering into care,
the number of people with a viral load test as a proxy for the number on ART, and the result of the viral
load test to indicate the number virally suppressed.
Unique identifiers
A number of systems produce personal identity numbers that may be useful in de-duplicating patient
records (within all levels of TIER.Net and at the NHLS) if collection of these identifiers is routinely
conducted in care and laboratory settings.







National civil identifier - the national civil identifier is available to all citizens and is currently
utilized as an identifier in health systems. Approximately 10% of population are non-citizens and
do not have this identifier, therefore health system rely on name and date of birth to unique
identify individuals.
NDOH health patient registration system (HPRS) –health care is available to everyone; therefore a
health patient registration system is being developed to provide everyone with a health identity
number in support of the eHealth Normative Standards Framework (eHNSF) which is promoting
system interoperability. The civil identity number will remain a valuable secondary unique
identifier for linking patient data and preventing duplicate registrations.
Western Cape Province patient master index (PMI) - a centralised hospital information system,
implemented in 2000, which brought all 52 hospitals and 400 clinics in the Province into a single
system. Patients in the public sector in the Western Cape have the same folder number (i.e. the
PMI number) in all facilities.
Jembi Health System patient master index – Jembi conducts software development, specialising
in system interoperability. Data from systems are merged and matched through a PMI. Their PMI
is currently being piloted in select facilities in the Western Cape Province. The majority of the
software is open source.
Western Cape Provincial Health Data Centre – merges data from multiple patient-level sources in
the Province; the data merge is in the implementation phase, developing a de-duplication
algorithm based on a complex probability match in order to track patients across services.
NHLS Central Data Warehouse (CDW) – developed an algorithm for de-duplication of all
laboratory data covering an estimated 300 million tests nationally.
Research linked to the NHLS CDW - In partnership with the NHLS, a research organisation
(HE2RO) further adapted the above de-duplication approach (based on the Jaro-Winkler string
distance method) in order to facilitate cascade analyses by geographic reporting units. The
National Institute for Communicable Diseases and the Centre for HIV & Sexually Transmitted
Infections have estimated cascade parameters based on the consolidated, de-duplicated CDW
viral load and CD4 data.
Policy
The South African government’s selection of a single patient monitoring system and ensuring that this
system was functional for all types of facilities is significant. Policies regarding case reporting were not
Page 15
fully explored but it was suggested that the government would not approve mandating HIV case
reporting.
Page 16
Table 2. Summary findings from SWOT analysis for key areas of interest in South Africa.
Overall findings
Monitoring
systems for HIV
STRENGTHS
Person-level data are already being
tracked by existing systems which
could be utilized for CBS.
Systems for collecting data from all
points of the cascade exist.
Multiple methods of de-duplication
are present.
System interoperability is being
actively pursued.
The public health system acts as
one provider; where the
electronic platforms for doing so
exist, clinicians within a Province
can see all the health information
on a patient regardless of the
facility in which they were seen.
The majority of ART information is
available in a single system
(TIER.Net) which successfully
imports data from other systems
based on a data standard. The
system has quality control
measures and uses standard tools.
To enhance retention, ART is often
distributed in clubs/groups.
CD4 and viral load are routinely
available for ART patients and are
conducted in regional laboratories
all utilising a national LIS.
Currently a moratorium on EMR
WEAKNESSES
Diagnosis and pre-ART data are
not being routinely digitised at
patient-level in current
systems.
The national civil identify
number is inadequate as it
includes personal identifying
information and is available
only to South Africa citizens.
OPPORTUNITIES
The Health Patient Record
System is in active
development as a national
PMI and needs to be tested.
Development of a system to
track rapid tests for
individual level diagnosis
data that can be linked to
treatment data.
THREATS
There is a risk that parallel
initiatives on de-duplication
could result in different
outputs which could
fragment utilization at the
Province level.
Reliable connectivity and
infrastructure remain an issue,
resulting in the majority of
facilities being at Tier 2.
Existing TIER.Net HTC and PreART modules not being
utilized.
There is a back-log of laboratory
results to be filed in medical
charts in many facilities;
incomplete laboratory data are
reported through TIER.Net as a
result.
There are insufficient resources
Develop a system to track
rapid tests for diagnosis data
either through TIER.Net or
the LIS (by sending the
results with specimens
routed to regional
laboratories).
CDW sends text and email
alerts to the National
Institute of Communicable
Diseases surveillance unit
when positive cases are
identified which may be
useful for CBS.
Staff shortages exist and
could further compromise
completeness and quality
of data.
Parallel systems send
laboratory results to
facilities which are not
always filed properly.
Tuberculosis is a notifiable
disease although data in
the disease information
system are considered
more reliable than the
notification system. The
Page 17
Unique
identifiers and
de-duplication
development.
There are examples of other
person-level data being
successfully linked with the above
data sources based on a PMI.
committed to supporting
quality assurance of primary
care clinical record keeping
and data capture.
Rapid tests used for HTC are not
reported to laboratories.
Multiple mobile applications
exist for tuberculosis
reporting; potential to create
for HIV.
Existing national civil identity
number – the majority of people
have one.
Developing a health identifier
which would cover those without
national civil identity number.
Facilities provide a patient identity
number.
CDW has capacity to de-duplicate
300 million records in 10 days.
Successful PMI over many years in
the Western Cape, for over 6
million people.
Large percentages of immigrants
are not eligible for national
civil identity number.
Laboratory staff unaware of how
data are used at the national
level.
Connectivity and infrastructure
lacking at many primary care
facilities with respect to
implementing health identity
number.
Multiple successful deduplication algorithms exist;
comparison yet to be done.
Laboratory data utilization for
cascade to care measures.
HPRS a promising opportunity
for a national PMI.
Triangulation of system data
(eg. TIER.Net and laboratory)
is a unique opportunity to
differentiate the
completeness of data.
same would likely occur for
HIV if notification were to
be considered.
Some facilities already have
an EMR system which
reports to DHIS, and feel
TEIR.Net is duplicative.
Differences in metrics from
different systems can
result in a lack of
confidence in some of the
source data.
Page 18
Interpretation of findings
Based on the assessment, it was determined that multiple data systems exist in South Africa which could
be utilised for HIV CBS.
Both TIER.Net and the NHLS collect patient-level data at a scale which is reflective of the national
treatment program. In relation to post-ART registration, these systems could be seen as best practice for
CBS. A limitation of these systems is that neither currently includes diagnosis / pre-ART data. The three
tier system provides an approach to meet the needs and capacity of a facility. Crucially, this approach
avoids pushing the adoption of an EMR/new technology in all sites irrespective of capacity.
Laboratory reporting of HIV surveillance data is feasible given that regional laboratories conduct all of
the CD4 and viral load tests which are entered into an electronic system and are reported to NHLS.
However, implementing laboratory reporting of HIV diagnosis data at the facility level could be
compromised given their limited resources.
There are national identifiers in place which assist in HIV monitoring and evaluation/surveillance,
allowing for de-duplication of cases at the national level. However, there are a number of de-duplication
efforts that may be replicating effort. Data quality assurance systems are also in place and implemented.
Recommendations





The approach to patient monitoring and reporting into monitoring and evaluation/ surveillance
systems should match the capacity of the region/facilities. The three tier system provides an
approach to meet the needs and capacity of a facility and continued use is recommended.
Linking data from HIV testing into analyses would be beneficial. This would be possible by adding
testing to the data collected by the NHLS and/or allocating resources to implement the HTC and
pre-ART modules within TIER.Net
Countries should involve the national and regional level MOH in developing data systems and
standards.
The social and political context of the country needs to be considered before implementing CBS.
In South Africa it should be possible to develop the current patient monitoring and laboratory
reporting systems to create an enhanced monitoring system that accommodates most of the
features of a CBS system without mandating HIV reporting.
Communication of stakeholders is imperative at the national and regional level. Multiple groups
are working on de-duplication of national datasets, often fully independent of each other.
Page 19
REPUBLIC OF KENYA
Background
The Kenya Ministry of Health (MOH) determined that HIV surveillance would be strengthened by moving
from aggregate reporting of persons diagnosed with HIV to a case-based surveillance (CBS) system. In
2015, a CBS pilot was performed in Kenya to collect information on the feasibility and acceptability of
CBS. The CBS pilot was developed through a technical working group comprised of representatives from
the MOH National AIDS and STD Control Programme (NASCOP), Kenya Medical Research Institute
(KEMRI), CDC/Kenya), University of California, San Francisco, Walter Reed Project United Army Medical
Research Unit Kenya, and participating counties and other stakeholders involved in either technical or
financial support for CBS. The pilot was conducted at 124 facilities in two high prevalence counties. Cases
were collected retrospectively by sub-county surveillance officers and by health care providers who were
paid to add this onto their clinical work.
A large number of facilities included in the pilot were located in rural and remote settings thereby
requiring substantial travel time by the surveillance officers. The sub-county MOH surveillance officers
were not able to complete their normal work duties during the pilot. This called into question the
sustainability of this method as a routine surveillance activity.
Scope of the assessment
The goal of our assessment was to explore reporting options that might be less demanding of human
resources than reporting by surveillance officers or paying providers to report cases. We examined the
potential for reporting from electronic medical records (EMRs), laboratories (as a stand-alone system or
in conjunction with provider/surveillance officer), and providers (clinical and HIV testing counsellors)
without additional compensation. Available personal identifiers that exist and that are collected as part
of HIV diagnosis and care were also explored.
Interviews with key stakeholders were carried out and site visits were made at select representative
health facilities and laboratories providing HIV services. During site visits, data flow / systems were
observed and interviews were conducted with various facility staff to learn about the use of EMRs, how
laboratory HIV tests were conducted, and the potential for providers to report cases. Six interviews were
conducted with stakeholders from EMR implementing partners (International Center for AIDS Care and
Treatment Programs (ICAP), International Training and Education Center for Health (I-TECH), and
Palladium), laboratory information management systems (LIMS) implementing partners (National HIV
Reference Laboratory (NHRL)), and the Clinton Health Access Initiative (CHAI). A presentation and
question and answer session was also provided by the CDC Health Systems and Evaluation Branch.
Additionally, three regional laboratories and eight facilities were visited.
Findings
The main findings from the interviews and site visits are summarized in Table 3.
EMR reporting
EMRs have been implemented in Kenya to maintain the standard MOH clinical information on HIV
patients. There are four nationally accepted EMRs: IQCare, OpenMRS (KenyaEMR), CPAD and FundSoft.
Most of the EMR installations have been KenyaEMR (n=341) and IQCare (n=324). Facilities using
OpenMRS can modify the variables collected which can impact the data available for CBS. EMR
installation is done by one partner but the facility partner must maintain the system, with little support
for staffing. The lack of support for staff and electricity costs were reported as the reasons that 30
facilities discontinued using their EMR. CPAD is only used at facilities managed by ICAP, and FundSoft is
utilized by larger facilities which offer a wide range of health services in addition to HIV and therefore
Page 20
require a billing component in their EMR. KenyaEMR is a web-based system which is, therefore,
problematic when connectivity is down or slow.
At 335 health facilities information regarding the patient is entered into the EMR by the providers at the
time of the patient visit (also referred to as ‘point–of-care’ data entry). Eleven of the sites are paperless
and rely exclusively on the EMRs, and at the remaining facilities data entry is performed by a data clerk
after the end of the patient visit when the medical record is no longer in use. In accordance with national
policy, the majority of the facilities with point-of-care data entry also maintain a paper record. This
means that in addition to completing registers and paper-based medical records, providers and staff
must also enter data into the EMR. Additional funding in 2015 was used to support data entry clerks to
enter historical records into the EMRs.
Interoperability between EMRs is currently not available, although several groups are in the process of
developing systems. For the most part, LIMS and the electronic antiretroviral therapy (ART) dispensing
systems are not integrated with the EMRs.
All facilities reported experiencing periodic power outages and only larger facilities had back-up
generators. It is worth noting that during our SWOT two facilities were experiencing power outages and
were unable to use their generators (one was broken and the other lacked fuel). Both power outages and
non-functioning generators were reported as ongoing problems. During outages only paper records are
completed and data entry clerks enter the data retrospectively once power has returned.
Data quality assessments are rarely performed and when they are, only a small sample of records (e.g.
15 records per facility) is reviewed. The data in the EMR have not been systematically compared against
information in the paper medical record, as such, the validity of EMR data is not known.
Reporting to MOH is done aggregately from facilities directly to NASCOP using standard MOH forms.
Staff indicated that the EMRs are programmed to do this and therefore reporting was not burdensome.
With the purpose of performing cohort analyses (i.e. to measure 12 month retention), a data warehouse
was developed and is maintained by Palladium. The warehouse collects historical data and also requires
ongoing quarterly reporting from facilities with EMRs. With the exception of facilities operated by one
implementing partner (ICAP), facilities were not able to electronically submit data to the warehouse. As
such Palladium staff travelled to each facility to download the data onto thumb drives and then upload
data into the warehouse. The variables obtained from the EMRs for the data warehouse are grouped
into seven extracts: Patient Master List, ART Patient List, Patient Baselines File, Pharmacy Encounters,
Laboratory Request & Results Details, Clinical Encounters List, and Patient Status. As names are not
included the de-duplication of records relies on the Comprehensive Care Clinic (CCC) number, a unique
patient number within a facility. At the time of the SWOT, 347 (52%) of the EMR facilities were reporting
data to the warehouse. Staff at Palladium indicated that it is possible to create a separate module within
the data warehouse that could include names and other personal identifiers for CBS.
A notable system in Kenya is the Eastern Deanery AIDS Relief Program which operates 14 health facilities
in the Nairobi area. They have developed their own fully integrated, networked and paperless EMR
system. It includes an interoperable LIMS and pharmacy system. Each facility has a data manager, and
data are stored on a central server. All patients are registered in the EMR, prior to HIV testing or
receiving services. The system has many identifiers including the national identity number. They have
mobile technology for field workers to gather data. They also perform viral load testing at their central
laboratory and CD4s at three of their laboratories. It is a closed system that requires sustainable
resources, and highlights what can be achieved should resources be available.
Laboratory-reporting
CD4 testing is done at the time patients enter care and every six months until ART is initiated. A CD4 test
is obtained when ART is started, and then subsequently the patient is monitored using viral load
measurements. Viral load tests are obtained six months following ART initiation and annually thereafter.
Page 21
There are seven regional laboratories which conduct viral load and early infant diagnostic (EID) tests. All
of these use a LIMS which captures patient-level data. Six of the LIMS systems were developed by
Strathmore University in partnership with Clinton Health Access Initiative. In these six systems, the
patient CCC number, age, and facility are collected but not the patient’s name. Facilities that use these
regional laboratories maintain a cross-reference log of submitted specimens, CCC number and, at times,
patient name, to link returned test results to the patient’s files.
The NHRL functions as the seventh regional laboratory. The NHRL developed its own LIMS which
captures name, CCC number, age and facility. Although there is a national laboratory requisition form for
viral load testing that includes patient name, as well as a number of other variables relevant for CBS,
most of the facilities and laboratories visited did not use it. Test results from all of the regional
laboratories are sent back to facilities by e-mail to a personal e-mail account, or in some cases, to the
facility email account. Results are not encrypted and there did not appear to be any specific security
measures in place. Some of the small facilities use transporters to carry specimens to the facility-based
laboratories and to carry paper-based test results back to the facilities. This appeared to be a more
secure mechanism. It is worth noting that the laboratories did not have policies regarding confidentiality
and security of data.
CD4 testing is performed at many of the larger facilities. Smaller facilities that lack on-site laboratories
send their specimens to the larger hospitals for CD4 testing. Laboratories record all specimens received
for CD4 testing in the MOH registers or logbooks, including date of receipt, patient name, CCC number,
and test results. Paper copies of test results are placed in the medical records. Results are entered into
EMRs from the paper result. There are a few facilities where the LIMS is fully integrated with the EMR so
that test results are automatically recorded in the EMR.
Rapid, point-of-care HIV testing is the standard procedure for HIV diagnosis. Therefore, tests and test
results are not included within the regional LIMS. Information on the first HIV-positive test is most
reliably available from HIV testing registers (e.g. HTC, antenatal clinics, and voluntary medical male
circumcision programs).
NASCOP has developed a national web-based portal in which all viral load and EID tests are pulled from
the seven regional laboratories performing these tests. The relevant CBS variables collected include the
CCC number, the reason for the test, and test result. In consideration as a potential source of data for
CBS it should be noted that as the NASCOP system is used to monitor laboratory commodities and
workload no attempt has been made to identify and remove duplicate records.
Provider/facility staff reporting
At the voluntary HIV testing sites information for all persons testing is recorded in a register which
contains patient names, the test date, and result. Health care providers and HIV test counsellors
complete standard MOH forms which contain the individual-level information necessary for CBS
including risk behaviour. The referral form for transfers does not indicate date of diagnosis.
The number of positive tests varied by facility but in general less than one client per day tests positive at
medium and smaller facilities. When asked about the possibility of completing HIV case report forms for
each patient who tests positive, the majority of counsellors indicated that this would not be
burdensome, particularly if the number of variables required was small. In contrast, discussions with the
nurse-managers/in-charge and providers indicated that the current demands on their time for patient
management preclude the possibility of reporting patients for CBS. However, it was noted that a records
clerk based at the facility might be able to report cases.
Unique identifiers
Patient names and age are collected in the HIV testing register and at laboratories that conduct CD4
tests, while names and dates of birth are recorded in medical records. Names are not recorded at six of
the seven regional viral load/EID laboratories. At the time of initiating care, all patients are assigned a
Page 22
facility-specific unique number (the CCC number) that includes the facility number listed on a national
facility master list and a serial number. This number is unique to the patient, although if a patient selftransfers they will be given a new CCC number and the information from the two clinics will not be
linked. Kenyan adults (aged 18 and above) are able to obtain unique national identification numbers but
these have not been collected as part of routine care. There were discussions indicating that Kenya may
provide these numbers to children but it was not evident if this is currently underway. NASCOP has
modified the registers to allow for collecting national identifier as well as any additional identifiers that
may be available in the future. The HTC register is also being modified to include date of birth and the
counsellors are supposed to document the CCC number to indicate that the patient entered care.
Information on date and place of HIV diagnosis is not always obtained for patients who transfer for care.
However, all other variables needed for case reporting are collected on the CCC/blue card. Efforts are
underway to include the Integrated Population Registration System (IPRS) as a national unique personal
identifier, by having an electronic module in the EMR link to the IPRS system to search for the patient in
the system and provide identifiers.
Policies
Kenya has reporting regulations on infectious diseases, which includes case-based reporting for some
diseases and aggregate reporting for others, including HIV. Therefore, it was determined that additional
regulations for reporting HIV/AIDS in a case-based system were not needed. There are gaps in policies
and standard operating procedures regarding data security and confidentiality, and the functions,
purpose, data quality and utilization of EMRs, LIMS and the data warehouse. The existing birth and death
registration systems are paper-based.
Page 23
Table 3. Summary findings from SWOT analysis for key areas of interest in Kenya.
STRENGTHS
WEAKNESSES
OPPORTUNITIES
Overall
Person-level data are already Diagnosis data are not
Addition of names to
findings
being collected by existing
currently entered into
systems.
systems.
medical records.
Potential interoperability
System interoperability is
between EMR, regional lab,
being actively pursued.
facility lab (if LIMS),
pharmacy, and billing
systems.
EMR reporting The majority of facilities with Reliable connectivity and
EMR developers mentioned
over 500 patients have an
infrastructure remain an
creating a module for HTC
active EMR.
issue.
data.
EMRs contain the necessary
OpenMRS is web-based and
Could create a separate
variables for CBS.
data cannot be entered
DataMart module for CBS
Most EMR facilities have all
when internet connectivity
with identifiers in the data
historical data entered.
is slow or down or power is
warehouse.
Data warehouse is in
down.
development.
There are insufficient
resources committed to
supporting systems.
Laboratory
CD4 and viral load tests are
NASCOP national viral load
Viral load turn-around-time
reporting
routinely conducted.
/EID database excludes
varies by regional lab and
Viral load and EID are
patient name.
could be addressed.
conducted in regional labs
Standard viral load test form
Develop policies and
and all utilise a LIMS.
is not being utilized and it is
standard operating
Standard viral load test form
unclear if name is required.
procedures for lab
exits.
Rapid tests used for HTC are
management, LIMS and
Specimen transport from
not reported to lab.
NASCOP national viral
smaller to larger facilities
load/EID system.
works well.
Update LIMS system to
THREATS
Lack of policies for systems,
including data use and
security. When policies do
exist they are often not
being followed.
Staff shortages exist and
could further compromise
completeness and quality
of data.
Data back-ups vary between
EMRs and facilities; one
site was observed to have
lost all of their data due to
the lack of a back-up
system.
Name is excluded from six
of the seven regional lab
LIMS.
Test results for viral load
and EID are being emailed
insecurely to facilities.
Reagents stock outs lead to
a back-log.
Page 24
Provider
reporting
Standard HTC register being
utilized.
Standard forms and registers
are being utilized.
include additional data.
Need additional identifiers on
registers (e.g. date of birth
on HTC and date of
diagnosis on antenatal).
Transfer forms do not include
sufficient data for CBS.
New diagnoses at most
facilities are small enough
to have counsellor
complete CBS form.
Patients often retest; if not
enough identifiers are
collected, difficult to
match cases and account
for duplicate records.
Page 25
Interpretation of findings
Based on the assessment, all of the reporting options explored, including continued use of surveillance
officers, could be used in Kenya for HIV CBS. The choice of the reporting system will depend on the
particular circumstances.
EMR reporting is promising as it is possible to program most EMRs to produce reports. However, the
amount of work required to do this was not assessed as part of the SWOT. Although several groups are in
the process of developing interoperability between EMR systems, none currently exist. As most EMRs
were not networked, reporting would come directly from each facility or, in some situations, from a
specific implementing partner that oversees several facilities (e.g. ICAP). HTC data are not entered into
the EMRs. There appear to be discussions regarding creating a HTC module for EMRs which should
overcome this limitation. Data quality assessments are insufficient and in particular validating EMR data
against the paper records should be done prior to investing in CBS reporting from EMRs.
Given that obtaining EMR data for the data warehouse is already occurring, building in a specific CBS
module which includes identifiers may be cost-efficient. Also, reporting to the data warehouse occurs
quarterly. Facilities indicated that they are up to date with data entry which appears to have been the
result of one-time added resources. The ability to maintain complete EMRs should be reassessed
particularly in light of inconsistent electricity and internet connectivity.
For laboratory reporting, changes would be necessary to facilitate CBS as the systems have patient level
data but only limited identifiers. To be used as an additional source of data for CBS, the regional LIMS
and NASCOP viral load/EID system would need to be enhanced to include additional security measures,
names and other personal identifiers, and potentially additional information such as date of diagnosis.
Standard, national policies and procedures for managing and securing laboratory data need to be
developed. Although CD4 results are paper-based, they are entered in the EMR. Investment in
developing electronic systems for CD4 tests may be worthwhile.
For provider reporting, it was determined that health care providers would not be compliant with
reporting due to time constraints. HTC counsellors could likely report new diagnoses, as there are fewer
variables and the amount of new positives per HTC counsellor is relatively low. We did not assess the
feasibility of reporting by nurses at antenatal clinics where the work load may be greater than that of
HTC counsellors.
In addition to ensuring that patient names are included in all registers and LIMS, a national unique
identifier needs to be collected as part of HIV testing and care. The success of modifying the monitoring
and evaluation tools to include the national identifier should be evaluated after the tools have been in
use for 12 months.
Recommendations




The approach to CBS should match the capacity of the county/facilities. Having multiple
approaches (EMR reporting, lab reporting, surveillance officer reporting, HTC reporting) to obtain
the same data, which meets the needs and capacity of the facility is important.
The MOH should engage with county health and surveillance officers to determine optimal
reporting mechanism within each facility, build local surveillance capacity, and empower
counties to oversee and manage CBS activities including data use.
The MOH should learn more about Eastern Deanery AIDS Relief Program as an example of best
practices and explore how to replicate this system elsewhere.
Facilitate communication between stakeholders at the national and regional levels to avoid
duplicative efforts.
Page 26







When developing systems, policies should be developed or updated to match the new systems.
Specifically policies for the LIMS and the data warehouse should be developed and promoted.
Policies to ensure the security of patient data are essential.
Develop minimal standards for the completeness, accuracy, and timeliness of EMR data and
evaluate all EMRs to ensure they meet these standards prior to CBS.
Determine personnel and time required for programming EMRs to report out cases, including
systems to identify and report sentinel events.
Develop a plan and timeline to assist EMR facilities in moving to paperless systems to avoid
duplicate recording of medical information while ensuring all data currently available in registers
are captured and stored securely.
Ensure that all clinical specimens are submitted to laboratories with patient names and modify
LIMS to collect names, the national identifier, and other potential identifiers that may be
developed in the future. Ensure that standard MOH laboratory requisition forms are used.
Consider digitalizing CD4 registers.
Assess need for increasing the number of surveillance officers and records clerks for reporting
cases at lower volume facilities without EMRs or whose EMRs do not meet standards for
reporting.
Page 27
Republic of Haiti
Background
In order to obtain information about the scope of the HIV epidemic and better understand its impact, the
Haitian MOH mounted a decade-long effort, in collaboration with international donors and implementing
partners, to establish national HIV case-based surveillance. The resulting system has been in operation
since 2008, and features name-based case reports collected via an electronic case reporting platform and
through automated data feeds from the country’s major EMRs. The system is maintained and
implemented on behalf of the MOH by the National Alliance of State and Territorial AIDS Directors
(NASTAD).
A governmental mandate requires that health facilities report data about all persons infected with HIV.
This includes reports of new diagnoses and reports of clinical data for persons receiving care and
treatment. These data are collected via two mechanisms: 1) reports of initial HIV diagnoses submitted by
testing sites; and 2) ongoing clinical outcomes reported via EMRs. Test sites report new HIV diagnose
using a standard HIV case notification form which is completed and submitted after patient post-test
counselling. The form collects the patient’s name, address, demographic characteristics, and HIV risk
factors in HIV testing. A paper version of the case report form is filled out by the HIV test counsellor and
subsequently entered into an electronic reporting platform for submission to the case surveillance
system. The second reporting mechanism, submission of case data via EMRs, is performed through
automated feeds of longitudinal patient care data from the country’s three EMR systems. These
longitudinal data include: sentinel events, such as pregnancy, diagnosis of tuberculosis or sexually
transmitted infections, and patient death; CD4 and HIV viral load test results; prescription and pick up of
HIV-related medications; and date of last patient clinical encounter.
Data cleaning is routinely performed to eliminate duplicate reports and ensure data quality and
completeness. A matching algorithm was developed to identify duplicates and allow previously-reported
cases to be updated with new longitudinal data. The algorithm uses patient name, date of birth, sex, and
mother’s first name. Case matching and deduplication is done through an automated process followed by
manual adjudication if needed.
All notification data are stored on a secure server in a national HIV surveillance database, with access
limited to key MOH staff and authorized implementing partners. A de-identified, confidential analysis
data set is created periodically and distributed to an analytic team for creation of epidemiologic reports
and publications. Surveillance data about cases who drop out of HIV care are provided to community
health workers working at clinical sites to assist them in re-locating and re-enrolling patients.
Scope of the assessment
Because Haiti has conducted HIV CBS since 2008, the goals of this assessment were focused on the
strengths, weaknesses and gaps in the current system. Through this exercise, the assessors strove to: 1)
better understand how HIV CBS was designed, implemented and supported to serve as a potential model
for other countries; and 2) develop recommendations to further improve the system. Topics assessed
included the degree of knowledge and support for HIV CBS among stakeholders, the level to which CBS
was integrated with surveillance for other diseases, funding and human resource needs to support and
sustain the system, and gaps and weaknesses that impact data quality, timeliness and use for
programming and disease response.
Before initiating the assessment, meetings were held between the MOH and its partners, including the
CDC, the Pan American Health Organization (PAHO)/ the WHO and NASTAD, to review the SWOT
protocol and adapt it to meet Haiti-specific needs. The assessment was implemented in April 2016, and
Page 28
included both a review of relevant background and policy documents and interviews with key
stakeholders at the national, departmental (regional) and institutional levels.
Interviews were conducted in the capitol of Port-au-Prince and in two other regions. A total of 12 health
facilities were visited, including hospitals, HIV treatment centres and local health centres. Multiple visits
occurred with MOH programmes and directorates, including three regional health departments; the
National AIDS Program; the Department of Epidemiology, Laboratory and Research; and the Planning
Unit. In addition, key national partners were interviewed, including CDC, WHO and NASTAD. Finally,
interviews were conducted among the three health systems that administer EMRs: The University of
Washington/I-TECH, the Haitian Group for the Study of Kaposi's Sarcoma and Opportunistic Infections
(GHESKIO) and Partners in Health/Zanmi Lasanté (PIH/ZL).
Findings
A total of 25 interviews and site visits were conducted. Results from these interviews and site visits are
detailed in Table 4. Summary findings include the following:
Data Quality
HIV CBS has been in place in Haiti for eight years and has continued to mature and improve over time.
The system allows for presentation of data with very high granularity, including clinical cascades. Because
of the recent addition of new clinical variables to the longitudinal data reported by EMRs, the system has
the ability to track patient linkage to, and retention in care down to site level, and by neighbourhood of
residence. A pending national report, which was viewed by the assessors in draft form, demonstrates a
wealth of information the system can provide, namely census-level HIV monitoring data. Interviewees
indicated that the detailed level of available data allows for program planning and epidemic control at all
levels, and can be used to evaluate and improve the quality of care and treatment. Despite these
strengths, some system data are less than optimal or are missing. Data on key populations such as men
who have sex with men and sex workers are under-reported. There is no functioning vital statistics
registry in Haiti, and deaths among persons with HIV are infrequently reported. Finally, although CBS
participation is estimated at >90% for public institutions, there is a lack of information about
participation from private, for-profit health institutions. The Tuberculosis Control Program has developed
a method to ensure that private health facilities report patients with tuberculosis; a similar approach
could be used to compel HIV reporting from the for-profit health sector.
Reporting System
Negative perceptions exist about a lack of sustainability for this system, particularly as it costs more than
other surveillance efforts. Similarly, some perceive that HIV CBS has been “outsourced” because it is
maintained by a non-governmental organization (NGO). Respondents frequently noted that the current
HIV CBS platform in Haiti could easily be expanded to allow surveillance of other infectious and chronic
diseases. Such an expansion could help to assure system sustainability. Over the next several years the
system will transition toward greater oversight and administration from the MOH. EMR data feeds have
frequently been delayed over the past year because of updates to reporting methods and variable
requirements. It was observed that the system may be collecting too many variables; focusing more on
the key indicators/variables recommended by WHO could improve efficiency. At present, test results and
laboratory data are extracted from EMRs; the national public health laboratory is implementing a new
LIS, which offers the possibility of obtaining CBS reports directly from the laboratory in the future.
Page 29
Table 4. Summary findings from SWOT analysis for key areas of interest.
Overall findings
Political and
organisational
support,
governance, and
human
resources
HIV care and
treatment
cascade; data
collection and
STRENGTHS
Detailed, case-specific data are
collected that provide an
epidemiologic “census” of persons
diagnosed with HIV in Haiti.
Data allow for planning, evaluation
and decision-making at all levels,
and includes variables that
characterize patient linkage to and
retention in care.
The system uses existing EMR data,
thereby reducing the cost of data
collection.
WEAKNESSES
Full potential for data use has not yet
been realized, particularly at the local
level.
CBS reports are frequently interrupted or
delayed by infrastructure problems and
by high staff turnover.
Haiti lacks vital statistics data, which
hampers reporting of HIV mortality to
the system.
HIV CBS does not encompass all private,
for-profit health facilities.
Data about key population are
insufficiently reported.
There is strong leadership and
support for the system from the
MOH National AIDS Program.
Human resources exist at the site
level, where staff have been trained
and are motivated to participate in
HIV CBS.
Consistent support from donors such
as PEPFAR has assisted in
development of strong strategic
information systems, including HIV
CBS.
There is a commitment from funders
to continue support for HIV CBS in
future years.
EMRs collect many clinical variables
and track patient visits in a variety of
ways.
Current HIV CBS supports creation of
Although there are plans to transition the
system toward full MOH administration
and oversight, this has not yet occurred.
Not all EMRs currently report data on a
regular, set schedule because of recent
changes to reporting requirements and
methods.
OPPORTUNITIES
Use of the data is improving,
and a new web-based
visualization interface will
allow for easier access to
aggregate findings at all levels.
The Tuberculosis Programme
ensures participation of
private clinics in case
surveillance of tuberculosis by
controlling distribution of
medications. A similar strategy
could be used for HIV ART to
compel HIV reporting by the
private health sector.
Haitian NGO staff implements
and administers HIV CBS on
behalf of MoH. There is
strong collaboration between
the NGO and MOH, which will
facilitate eventual transition
of the system to full MOH
oversight.
THREATS
The system is perceived by
some as too expensive, siloed,
and potentially nonsustainable over the long
term.
Future HIV CBS data will
include measuring care
cascades at all levels,
including the site level.
The system may be too
ambitious in its collection of
clinical variables. Analysis
efforts need to remain
With PEPFAR funds decreasing,
there is a fear that HIV CBS
may have less funding in the
future.
HIV CBS is not currently
supported through local
funding streams.
Page 30
reporting
clinical cascades at the national,
regional and local levels.
Data are used to assist community
outreach workers to re-link
patients to care.
Reporting
systems for HIV
and other
notifiable
diseases
There is potential for the HIV CBS
platform to be expanded to
encompass case-level surveillance of
other notifiable conditions
The system is not integrated with other
epidemiologic surveillance.
Linking clinical
data with
laboratory,
pharmacy and
other data
Pharmacy data, including prescription
information and medication pickup,
are accessed via EMR feeds to HIV
CBS.
The system employs methods to
ensure that unique cases are
identified and de-duplicated. This
process could be used for other
diseases.
EMR reporting could be expanded to
encompass additional disease and
health conditions.
Currently, there is no direct reporting of
LIS data to HIV CBS.
A national report will be
released in June that
showcases the ability of HIV
CBS to track national and
local clinical outcomes.
New WHO guidance exists that
can be used to prioritize
analysis of clinical variables
collected during HIV CBS.
There is significant interest in
assuring that HIV CBS is
integrated with other
epidemiologic surveillance.
HIV CBS platform may be
useable for surveillance of
other diseases and health
conditions.
The national public health
laboratory is implementing a
new LIS system that will
facilitate future reporting of
lab data to HIV CBS.
An agreement was developed
to begin a direct feed of HIV
RNA/DNA results to HIV CBS
from the national lab.
Collection of laboratory
RNA/DNA data will begin in
June, 2016.
focused on key indicators.
A perception exists for some
that HIV CBS has been
“outsourced” and relies too
heavily on foreign
investment and personnel.
Page 31
Interpretation of findings
Through a sustained collaborative effort, the Haitian MOH and its partners developed and implemented
an advanced health information system which complies with WHO and UNAIDS recommendations for
second generation HIV surveillance. The system has the potential to provide sufficient data for planning,
evaluation and decision-making at all levels, and includes variables that characterize patient linkage to
and retention in care. In addition, the system uses existing EMR data, thereby reducing the cost of data
collection. Assessment findings suggest that HIV CBS in Haiti has helped to reinforce and strengthen
MOH technological and epidemiologic capacities.
To assure that HIV CBS data are sufficiently leveraged, capacity for data visualization and analysis needs
to be expanded, particularly at the regional and site levels. In order to maximize system sustainability,
HIV CBS should be better integrated with other public health surveillance efforts. In addition, in order
for the system to fully transition to Haitian MOH oversight, human resource needs and gaps should be
assessed and addressed.
Recommendations
The assessment identified a number of opportunities that, if acted upon, could address weaknesses and
threats to HIV CBS in Haiti. These were used to formulate specific recommendations, grouped into nine
domains, including: 1) roles and responsibilities; 2) system implementation; 3) technical strengthening;
4) system coverage; 5) integration with other national surveillance efforts; 6) system efficiency; 7)
system effectiveness; 8) sustainability; and 9) inclusion of the private sector and system optimization for
use by other chronic disease programmes. A detailed roadmap was proposed, containing key
recommendations and next steps.
Among the recommendations in the roadmap, WHO suggests that the National AIDS Program and its
partners focus on “6 + 1” overarching priorities:
1. Define the roles and responsibilities of each partner involved in the HIV CBS.
2. Assess resource needs for HIV CBS, particularly human resources, to determine potential needs
and gaps.
3. Highlight and educate partners about the principal benefits of the system, with an emphasis on
cross-cutting benefits that might increase capacity for non-HIV related public health programs.
4. Facilitate access to aggregate HIV report data for the Directorate of Epidemiology, and
Laboratory Research (DELR), which they need for routine infectious disease surveillance.
5. Use WHO standards to consolidate and prioritize HIV CBS variables and indicators for analysis
and quality assurance.
6. Continue efforts to improve reporting feedback and epidemiologic analysis at all levels, with an
emphasis on preparation of data for sites and departments (regions). This should include access
to “dashboard” data through the developing web-based visualization interface.
The final recommendation (#7) stipulates creation of a report that describes key inputs, supports and
drivers that contributed to the development of the successful, high-functioning HIV case surveillance
system. This report will help other low- and middle-income countries to benefit from Haiti’s experience.
The National AIDS Program and its partners (including CDC and NASTAD) agreed to develop this report,
with assistance from the MeSH consortium and PAHO/WHO.
Page 32
GENERAL FINDINGS FROM THE FOUR SITUATIONAL ASSESSMENTS
In the countries assessed there was limited familiarity with disease surveillance in general and with HIV
CBS specifically. However, once the limitations of aggregate data and the benefits of CBS were
presented, sub-national and national level stakeholders indicated substantial interest in, and support
for, CBS as a way to more effectively monitor the epidemic. Although support for CBS was high, there
was concern over the resources needed. In particular, stakeholders indicated that CBS is feasible and
sustainable if it could be conducted without adding additional work for the limited number of health
care providers and surveillance officers. For the most part, stakeholders felt that having sites report data
through their existing electronic systems (EMRs and LIMS) would not require substantial additions to
facility level staff and based this on the current reporting requirements.
All of the countries assessed conduct routine HIV programme monitoring that collects information on
the number of persons tested for HIV, number of new HIV diagnoses and number of persons accessing
HIV care and treatment services. At the facilities, individual level data are collected and entered into
MOH registers and, for persons in care, into individual patient records. Current reporting of these data
from facilities to MOHs is done in aggregate. For the most part, the frequency and content of reports is
similar between countries. All of the participating countries report aggregate data for HIV and other
diseases into DHIS, although different versions are currently in use.
A CBS system must collect relevant demographic, diagnosis, care, and vital statistics information.
Therefore the program data collection tools from which CBS data originate must include information
required for a functional system. The required personal identifiers are patient first and last name (or a
reliable coding of name), date of birth and age (age can be derived from date of birth if required), sex,
and information on residence (complete or partial address such as village). Inclusion of the same
national identifier across information systems will facilitate this because these identifiers are assigned at
a national level, truly unique to an individual, and durable in contrast to names which may change over
time. It was found that most countries have a national identifier although with limitations, such as
eligibility criteria which exclude a significant proportion of the population (i.e. only for adults or only for
citizens). These national identifiers were not collected as part of HIV testing and with the exception of
South Africa, and not collected as part of care. The required Information on HIV includes the date of
diagnosis, date of first entry into care, date of ART initiation, and supporting CD4 and viral load
information and vital status. Review of systems and discussions with stakeholders revealed significant
problems with data quality.
HIV diagnosis. All of the countries have an HTC register from which the aggregate number of new
diagnoses is reported. The antenatal, PMTCT, male circumcision, and tuberculosis program registers
also conduct HIV testing and record results of tests in the registers. The personal identifying
variables that are collected in the HIV testing registers vary between countries and many do not
include all of the variables required for CBS. For example, some countries do not record name or
date of birth. Some but not all registers will add the care and treatment clinic number for those who
have entered care at the facility where tested, which facilitates locating patient medical records.
Some collect risk behaviours, although under-disclosure of risk remains an issue. All HTC registers
collect the date of diagnosis but antenatal or PMTCT registers may not include the date of diagnosis
for women who were diagnosed prior to the current pregnancy. This is because the registers record
the date of the HIV test and the result of the test but not the date of diagnosis. Because women who
are known to be infected are not retested, information on date of diagnosis for these women is not
Page 33
recorded. Although the HIV medical records are designed to record the date of diagnosis, the forms
that patients who transfer care bring to the new facility do not always include the date of diagnosis.
Care and treatment
Registers and paper medical records are standardized in all countries assessed by the MOH and the
variables collected are similar across countries. Aggregate data are collected from the standard
registers to determine the number in care and the number on ART. EMRs exist in all of the
participating countries although vary considerably, as often there are different implementing
partners. With the exception of Haiti, none of the countries use EMRs in all of their facilities. In
facilities that have EMRs, nearly all maintain both a paper and an electronic record. Most EMR data
entry occurs after the paper record is completed and as such, the paper record is the source
document for data in the EMR.
There are efforts in some countries to establish national data warehouses with case level data. The
systems vary by country and few include names. A few systems compile data from a single program
source, such as South Africa’s NHLS data warehouse. Kenya is developing a data warehouse for
EMRs and Tanzania compiles CTC data. The Western Cape Province is developing a data centre
which will compile data from multiple sources, health facilities, laboratories, vital registries.
Laboratories
In most countries, CD4 tests are conducted at facility-based laboratories that may test specimens
exclusively from their facility or serve as regional laboratories and test specimens submitted from
facilities. Most laboratory registers are standardized and some regional laboratories have a LIMS.
Laboratory registers are designed to collect patient names and sometimes dates of birth. However,
the personal identifier variables listed in the registers is inconsistent across countries with some but
not others collecting the patient name. For specimens originating outside of the facility where the
laboratory is based, the facility name from which the specimen is submitted is recorded in the
registers. Viral load and EID testing is expanding. At this point in time, testing for HIV RNA is done at
regional or national laboratories. In general, laboratories do not perform HIV antibody testing and as
such information on newly diagnosed persons is not available at the laboratories with the exception
of EID which is done at regional laboratories with this capacity.
Vital registries
Many resource limited countries have weak vital registries with incomplete recording of deaths and
causes of death. Currently systems are primarily paper-based. Recording of deaths does include the
decedent’s names. Vital registry data are not routinely linked to health systems to update medical
records.
Electronic reporting and linkage between systems can greatly improve CBS systems. Most of the
electronic systems are not networked which limits interoperability. For example, LIMS are not linked to
EMRs. Typically, the regional laboratory systems send results electronically to an email or printer and
from there, the results are entered manually into an EMR and the paper copy filed in the medical chart.
Kenya and Haiti have multiple EMRs, which are not interoperable. As such, submission of EMR data to a
CBS database would result in the CBS database receiving data with different variable names and
configuration. In order for the CBS database to receive and use these disparate data requires a wellfunctioning interoperability system. In Haiti, for example, this issue is addressed by variable mapping
within each EMR and naming conventions to assure that EMRs submit data in a uniform format.
Unstable power sources and internet access was a problem in many regions which adversely impacts all
electronic and networked systems.
Page 34
In the countries that have implemented or pilot tested CBS, mandates to report HIV exist although in
Kenya there were no policies regarding the data security or data release. In Haiti, a MOH mandate
requires that health systems and care sites participate in CBS. Although the MeSH assessment teams
requested documents that address issues of disease reporting policies, medical record privacy, and data
security, these were not provided, suggesting perhaps that they do not exist or that persons involved
with HIV programs were unware of these documents. To a large extent, programs rely on limiting the
use of patient names beyond the medical records as a way to protect confidentiality.
Organizationally, HIV programs are generally independent of other health care and disease surveillance
programs. This may be due to the role of donors, implementing partners, and guidelines from normative
bodies which focus on HIV testing, medical care and program monitoring and evaluation and not on
broader health or surveillance systems. This may be one factor that has contributed to countries need to
report information into both HIV program monitoring systems as well as into DHIS.
Page 35
RECOMMENDATIONS
The findings from the situational analyses support the development of CBS systems to systematically
capture routinely collected health data to describe and monitor the epidemic for program planning and
evaluation and ultimately disease control. During the situational analyses in Tanzania and South Africa, it
was determined that none of the existing regional or national programs or patient monitoring and
reporting systems examined are sufficient to provide quality data with which to accurately monitor the
indicators along the care cascade to track progress towards the UNAIDS 90-90-90 targets. In contrast,
the CBS system in Haiti and pilot in Kenya demonstrated the ability to do this. These assessments
provide valuable documentation of the possibility of establishing and maintaining CBS while at the same
time highlighting the significant challenges that must be addressed. Identifying technical and financial
resources to assist countries will go far in furthering progress towards the conduct of CBS in sub Saharan
African and possibly other resource-constrained areas.
Current HIV program monitoring and reporting systems have two significant limitations. First is that data
from persons diagnosed in HIV testing programs are not linked to data in the HIV care and treatment
program monitoring systems. Therefore, current systems cannot measure actual linkage to care,
understand populations least likely to be linked to care, or intervene to locate and engage these persons
in care. As such, it is not possible to accurately measure or identify persons who are diagnosed but have
not entered care or to measure the timeliness of entry into care. Second, beyond the facility level,
monitoring and reporting systems use aggregate data which can overestimate the number of persons in
care and the number of patients lost to follow-up. In some non-CBS data systems with individual-level
data, personally unique variables are not collected and as such, records are not de-duplicated. Although
it may be possible to adapt some of the current program and patient monitoring systems to function
effectively for CBS, there was wide variation in the systems examined and few that were open-source.
As such, we cannot recommend an existing system for widespread use. We can however recommend
approaches that may enhance developing, implementing, and conducting CBS across systems.
Approaches for developing CBS



The national Ministry of Health in collaboration with sub-national departments of health
responsible for HIV care and treatment and disease surveillance and other stakeholders should
be fully engaged in the process to ensure that the social and cultural context is considered, and
to maintain appropriate buy-in and participation from key partners.
The CBS system should match the national and sub-national ministries of health and facility
capacity to implement, conduct, and sustain CBS. Specifically, consider infrastructure,
connectivity, interoperability of electronic systems, human resource capacity, and vital statistics
in CBS system development.
When possible, leverage existing data systems and human resources for CBS. Assess all data
sources from which CBS data will be obtained and develop a system with sufficient flexibility to
include all of these with the goal of complete ascertainment of cases and reporting variables
with the ability to update a case overtime.
Policies and procedures

Develop national policies for CBS that should include:
o A mandate to report all persons diagnosed with HIV
o The surveillance case definitions
Page 36
o




Roles and responsibilities of all persons involved in CBS including outlining persons
responsible for reporting (e.g. health care providers, facility-based records clerks,
surveillance officers, HTC counsellors, laboratory personnel, staff from vital registration
systems) and managing the system
o The reporting pathway (i.e. where reports originate and where they are sent) that
assures the availability of data at sub-national and national levels
o Reporting variables and sentinel events and mechanisms for deduplication of cases.
These must include patient names, dates of birth, sex, and if available, a national
identity number, and data needed for monitoring strategic indicators
o Development of a standardized case report form
o Requirements to protect patient confidentiality and ensure data security including data
release policies
Develop standard operating procedures, technical guidelines, and performance standards for
completeness, timeliness and data quality for CBS.
Annually review and update policies and procedures to match changes in patient monitoring
data collection systems, new diagnostic methods, or roles and responsibilities.
Develop methods for ongoing communications within and outside of the MoH to understand
projects and activities that may impact or duplicate CBS efforts.
Develop regulations requiring proper identification of patients in settings that provide clinical
care. The use of CCC numbers in lieu of patient names for laboratory tests carry a substantial
risk of misidentification and incorrect information used for patient care.
Data collection and use







Develop a system for the receipt, storage, and management of CBS for use at the national and
sub-national levels that can accommodate receipt of data in paper or electronic formats (e.g.
with an interoperability layer) in an effort to minimize duplication of systems.
Improve data quality through regular, comprehensive assessments of completeness and
accuracy, quality improvement programs, and developing a stronger culture of data use.
Evaluate the completeness and accuracy of data in the electronic medical records relative to
data in the paper records prior to relying on EMR data for CBS.
Assess the degree of under-ascertainment of accurate risk behaviour data followed by plans for
improvement particularly as it pertains to CBS. For example, obtaining case reports from
programs and facilities that provide services for key populations.
Work collaboratively with disease surveillance programs to identify ways to leverage existing
systems, develop compatible systems, and integrate systems when possible.
Update patient monitoring and management, HIV testing (includes all programs where testing is
conducted), laboratory, vital statistics and other relevant program data collection tools to collect
the required personal identifiers (name, DOB, sex, and any national identify number).
Consider embedding data visualization systems to facilitate data use.
Resources

Determine the human and financial resources required to develop, implement, and sustain CBS.
This may include:
o Costs of updating current systems for surveillance and reporting of disease other than
HIV and for HIV monitoring and reporting
o CBS database costs including an interoperability component
Page 37
o
o
o
o
o
Methods used to collect case reports and the costs associated with these such as
additional surveillance officers, facility data records clerks, or adaptation of electronic
systems to permit these to report out to the CBS system
Costs related to EMR data reporting, including programming for variable mapping and
data reporting, and development of secure pathways for electronic submission the CBS
database
Costs of improving power sources and internet service
Training, supervision and performance management at the site, regional and national
levels
Data management, analysis, and dissemination
Page 38
APPENDIX
Page 39
Protocol for the situational assessment
of HIV case reporting
http://mesh.lshtm.ac.uk/
Page 40
Table of contents
Background .......................................................................................................................................................... 42
Purpose ................................................................................................................................................................ 43
Principal Collaborators ......................................................................................................................................... 44
Methods ............................................................................................................................................................... 45
Document Review Checklist................................................................................................................................. 47
Document Request Checklist ............................................................................................................................... 49
Notes .................................................................................................................................................................... 50
Interview Guide.................................................................................................................................................... 51
General information ............................................................................................................................................ 52
Situational assessment ........................................................................................................................................ 52
1.
HIV notification process: national and sub-national level.............................................................................. 52
2.
Data needs for HIV surveillance and reporting: national and sub-national level .......................................... 52
3.
Leadership: national and sub-national level .................................................................................................. 53
4.
Human resources: national, sub-national and facility level ........................................................................... 53
5.
Other case-based notifiable diseases: national and sub-national level ......................................................... 53
6.
HIV diagnosis: national and sub-national level and HIV testing facilities ...................................................... 54
7.
Clinical care: national and sub-national level and HIV care facilities ............................................................ 54
8.
Patient monitoring: national and sub-national level and HIV care facilities ................................................. 54
9.
Individual level (disaggregate) data sources: national and subnational level and relevant facilities ........... 55
10.
Unique identifiers: national and subnational level and at all facilities .......................................................... 56
11.
Ascertainment of recent HIV infections / incidence: national and subnational level .................................... 57
12.
Ascertainment of death: national and subnational level and at all facilities................................................. 57
Assessment of feasibility, barriers, and needs..................................................................................................... 59
Site Visit Checklist ................................................................................................................................................ 60
General information ............................................................................................................................................ 61
Debrief Form ........................................................................................................................................................ 63
Page 41
Background
The World Health Organization (WHO) released new guidelines in 2015 recommending consolidated
indicators to measure the reach of HIV services, and the impact achieved at both the national and global
levels. The new consolidated guidelines recommend the use of ten global indicators to collect
information along the cascade of HIV care and treatment as a principal way to track the epidemic and
response. Six of the ten WHO global indicators use data that originate from patient diagnosis, testing
and medical records. The six indicators are as follows: knowing HIV status (proportion diagnosed with
HIV); linkage to care; initiated ART; ART retention; achieved viral load suppression; deaths. Harnessing
clinical data in a format that makes them readily available for use as strategic information will allow for
routinely measuring these indicators at the subnational and national levels.
In contrast to cross-sectional surveys, which are the current standard for estimating HIV indicators,
mature, well-functioning case-based surveillance (CBS) systems provide longitudinal population-based
data, and can also provide aggregated snapshots of the service use at any given time. Furthermore, CBS
data can be collected from medical records in real time or at regular intervals (monthly, quarterly, or
annually). The CBS surveillance data can be available at the local, regional, and national levels providing
data-for-action at all levels of the public health system. While household-based cross-sectional surveys
have been able to provide national and regional estimates of indicators along the care continuum, the
costs associated with providing estimates at the local level are often prohibitive. CBS can be developed
to provide ongoing data to monitor the WHO indicators.
While the benefits of CBS are well-known and such systems serve as the backbone of HIV surveillance in
developed countries, HIV surveillance in resource limited countries has relied on anonymous HIV testing
from sentinel antenatal clinics, modelling, and surveys in the general and key populations. Although sub
Saharan African countries conduct CBS of selected priority diseases (e.g. measles and polio) and report
HIV/AIDS in aggregate, there are no national HIV CBS systems in operation in sub Saharan Africa.
Interest for developing and implementing CBS is growing in sub Saharan African countries, including
CDC, WHO and UNAIDS, which has led to efforts to understand and address country needs for
implementation of national HIV CBS systems that take into consideration current routine patient
information systems and emerging technologies. To that end, a number of situational assessments will
be conducted in selected countries that can provide a diverse spectrum of CBS and patient monitoring
systems. These assessment are being conducted by WHO and the Measurement and Surveillance of the
HIV Epidemic (MeSH) consortium which receives guidance from the WHO, CDC, and UNAIDS, and
funding from the Gates Foundation in collaboration with ministries of health (MOH) and other
stakeholders within each country. These assessments will include evaluation of current and planned
HIV/AIDS surveillance systems and patient monitoring systems that may feed into CBS systems.
Although not finalized, the proposed countries to conduct these assessments include those in which
demonstration CBS systems have been implemented (Haiti and Kenya), South Africa which has a multifacility-based integrated electronic patient monitoring system with interoperability that may serve as a
model for capturing data from electronic systems for use in CBS, and Tanzania where interest in
developing CBS is strong. If supplementary resources are available the additional countries to consider
assessing include: Cote d’Ivoire, Senegal, Botswana, Ethiopia, Malawi, Mozambique, Namibia, Rwanda,
and Zambia.
Page 42
Purpose
The situational analyses will be conducted in order to identify the strengths, weaknesses, opportunities,
and threats (SWOT) of CBS and patient monitoring systems. These SWOT assessments will be adapted to
fit within the context of each country and to meet the specific needs of the countries. Results from all of
the assessments will be used to inform the development of WHO guidelines for CBS and the develop
models for conducting or improving HIV CBS that can be pilot tested in the future.
A summary report will be prepared after completing the SWOT assessments that will be disseminated by
the WHO. The situational assessments will serve as case studies that capture unique situations that are
representative of other countries and in doing so serve as examples for other countries to learn from.
The report can be used by countries and stakeholders in establishing, evaluating or modifying CBS
systems. Countries that participated in these assessments will have the opportunity to contribute to and
review the report. Because these assessments will be adapted to the country context, the report will
describe each SWOT separately.
Although tailored for each country, the following components will be addressed in the in the report:
1. Summary of existing policies, any need to change policies for CBS, and if so, what is required
2. Summary of data sources and data collection processes
a. Data quality
b. Data accessibility
c. Case surveillance variables needed and their sources
d. Mechanisms to link data from multiple sources and across sites (e.g. from diagnosing
site to care facility, between HIV and tuberculosis (TB) care)
e. Mechanisms of integration with existing systems
3. Summary of existing data reporting mechanisms, pathways and use of the data
4. Summary of available human resources, need for additional human resources, and where CBS
would fit within the MOH’s organizational structure and the value it would add
5. Summary of sustainability issues
6. Summary of stakeholder position on establishing CBS and if feasible or if established, next steps
7. Summary of lessons learned from CBS or patient monitoring systems
8. Summary of surveillance and/or patient monitoring system attributes
9. Summary of the extent to which existing CBS and patient monitoring systems can be used to
measure the WHO strategic information indicators along the care cascade
10. Overall summary of the status of CBS opportunities and options in the various countries
11. Recommendations for next steps including demonstration projects
Page 43
Principal Collaborators
Name
Sandy Schwarcz
Organization
University of
California-San
Francisco
Andrew Boulle
University of
Cape Town
Jim Todd
London School
of Hygiene &
Tropical
Medicine
Richelle Harklerode
University of
California-San
Francisco
London School
of Hygiene &
Tropical
Medicine
World Health
Organization
Brian Rice
Serge Xueref
Role
Co-Chair MeSH
Routine Case Reporting
Working Group, Kenya
Advisor
Responsibilities
Coordinate access to desk
review documents and
scheduling interviews/site visits
for Kenya SWOT, assist with
report development
Co-Chair MeSH
Coordinate access to desk
Routine Case Reporting review documents and
Working Group, South scheduling interviews/site visits
Africa Advisor
for South Africa SWOT, assist
with report development
Co-Chair MeSH
Coordinate access to desk
Routine Case Reporting review documents and
Working Group,
scheduling interviews/site visits
Tanzania Advisor
for Tanzania SWOT, assist with
report development
Program Analyst
Conduct desk review, interviews
and site visits, draft report
Epidemiologist /
Systems Analyst
Conduct desk review, interviews
and site visits, specifically on
electronic systems, draft report
Monitoring and
Evaluation/Public
Health Advisor
Conduct desk review, interviews
and site visits (Tanzania, Haiti),
draft report
Page 44
Methods
This protocol and data collection tool was developed to ensure a standardized approach to the
assessments that takes into consideration the context and focus for each country. It is modelled from
various existing resources including: CDC’s Updated guidelines for evaluating public health surveillance
systems; WHO’s Evaluating a national surveillance system and NASTAD’s Case-Based Surveillance
Toolkit.
The assessment will begin with a meeting of stakeholders within each country to review the SWOT data
collection tool template and to adapt the tool to meet the specific country needs. The data collection
tool is intentionally designed to be a comprehensive tool that should be modified for the particular
focus of the assessment within each country. For example, in countries with established CBS systems,
the tool will be used to evaluate the system whereas in countries planning for CBS the tool will be used
as to facilitate the conduct of a situational analysis to inform the development and implementation of
CBS. The general components of these assessments will include a desk review of relevant documents,
interviews with stakeholders at the national, subnational, and facility levels, and site visits to selected
facilities and other relevant national or sub-national programs (e.g. national or regional laboratories that
conduct HIV-related tests).
Within each country the key partners to engage will likely include representatives from the MOH, the
CDC, WHO, M&E/SI technical working groups on HIV and partner agencies active in case based
surveillance and/or patient monitoring systems (e.g. National Alliance of State and Territorial AIDS
Directors [NASTAD], International Training and Education Center for Health [I-TECH]). Representatives
from the programs listed below should be considered for identifying country-specific needs of the
assessments and for contributing to the adaptation of the data collection tool, for participating in
interviews as key informants, for preparing for the in-country SWOT assessment (including providing
forms for the desk review, selecting sites and stakeholders for interviews, arranging for site visits), and
in the analysis, interpretation, and conclusions from the SWOT assessment with a primary focus on
recommendations and future plans.














HIV strategic information (SI)
HIV surveillance
HIV monitoring and evaluation (M&E)
Health information systems (HIS)
Infectious disease surveillance
HIV testing and counseling (HTC)
Prevention of mother to child transmission (PMTCT)
HIV care and treatment (C&T)
Early infant diagnosis (EID)
TB surveillance
Vital registration
Laboratories that conduct CD4 cells and viral load testing
Treatment procurement and distribution
Quality improvement
Page 45
From each country the SWOT assessment will aim to determine:









Existing data gaps and significance of the gap (e.g. Are there acceptable data at local level to
measure WHO SI indicators as they pertain to the care cascade?)
Existing disease notification systems: diseases reported (including HIV), reporting processes,
forms, data management, and dissemination and if these can be leveraged for establishing CBS
Existing policies or laws regarding HIV notification and data confidentiality or factors / practices
that could facilitate or impede CBS, including gaps in the policies
Existing and/or anticipated future HIV patient management systems for all facility levels (health
outposts, major urban hospitals) and by implementing organization (MOH, non-governmental
organizations [NGOs], private, etc.) and how they might be used for CBS
Current patient monitoring and reporting systems and practices (data collection tools, process
for reporting, M&E tools, frequency of reporting, nature of reporting [aggregate or patient-level
and if aggregate at what level does this occur], quality assurance/improvement and how these
might be used for CBS
Existing supervisory organizational structures and systems (to understand where CBS would be
placed) at national and subnational levels; within the HIV program, within the disease
notification system, and within the patient monitoring system
Technology capacity, infrastructure and systems that may facilitate or impede CBS
Human and financial resources availability and needed for CBS
How to integrate CBS with existing systems (e.g. District Health Information Software 2 (DHIS2),
other disease surveillance) and interoperability of systems
The following standardized tools have been developed for the SWOT assessment. The assessment will
include evaluation of the attributes of current patient monitoring HIV/AIDS surveillance systems.
Country representatives will be asked to provide input on adapting the tools to the meet countryspecific situational analysis and to provide guidance on the number of facilities for site visits based on
the systems currently in use.





Document Review Checklist
Document Request Checklist
Interview Guide
Site Visit Checklist
Debrief Form
Page 46
Document Review Checklist
Reviewer(s):
Country:
Date:
Person(s) Providing Documents:
Obtain the following documents for review, at least two weeks prior to visiting the country. Email the
‘Document Checklist’ form to country contact. The team will split up the workload, drafting a paragraph
for each item below to be compiled into a summary for the team. This will be compiled at least one
week prior to the visit and be reviewed by the team as preparation for the visit.
☐ Existing patient level/clinical data collection systems from C&T, HTC, PMTCT, EID
 These sources should include systems and tools used by the MOH and other implementing
partners. These are likely to include:
☐ Registers (C&T, HTC, PMTCT, home-based testing and counseling, EID, pharmacies,
laboratories), which can be both paper-based and electronic
☐ Clinic cards/standardized medical records (both paper-based and electronic)
☐ Patient-held medical records/booklets
☐ EMR systems and patient-held electronic medical cards (e.g. SmartCare in Zambia)

For each data source determine the following:
☐ Variables collected and how they differ by partner
☐ Inclusion of patient identifiers (names, dates of birth) and individual level
identification methods (national level identification numbers, clinic-assigned patient
numbers, regionally assigned patient number, biometrics)
☐ Flow of data
☐ If sources are being linked/integrated

Assess availability of following CBS measures:
☐ HIV diagnosis date
☐ Entry to care date
☐ CD4 results and date (first and follow-up)
☐ Viral load results and date (first and follow-up)
☐ Any additional (to CD4 and viral load) follow-up information highlighting retention
in care
☐ Initiation of ART date and regimen
☐ Advanced HIV diagnosis date (WHO stage III or IV)
☐ Death date and cause of death
Page 47
☐ Vital registration and statistics forms to determine
 Determine the following
☐ Format of vital records
☐ Variables collected
☐ Death registration requirements and methods
☐ Existing policies: HIV reporting regulations, other disease notification regulations, data release
from patient information systems (paper-based and electronic including EMRs) and vital registries
 Determine the following
☐ Variables required to be reported and the format
☐ If individual level identifiers are required to be reported
☐ Gaps in reporting policies
☐ Gaps in data release and confidentiality
☐ Strategic plans: disease notification strategic plans including but not limited to HIV, HIV
surveillance strategic plans, national HIV strategic plans
 Determine the following
☐ Nature of the epidemic
☐ Future plans for HIV surveillance
☐ Strengths of the strategic plans for HIV surveillance
☐ Uses and dissemination of data
☐ Reports: disease notification, patient information systems, data quality assessments/evaluations of
paper and electronic patient management systems
 Determine the following
☐ Quality and functionality of current systems
☐ Quality and completeness of surveillance data
☐ Types of electronic systems being utilized
Notes (e.g. reasons documents not provided, additional contacts):
Page 48
Document Request Checklist
Country:
Person(s) Providing Documents:
Please complete the checklist and provide the following documents by [date] to assist us in preparing for
the visit. Thank you for your assistance.
Documents provided
Yes
No
Not Available
A. Registers and other data collection forms or list of variables in electronic database or system
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
B. Clinical/medical records
☐
☐
☐
☐
☐
☐
☐
☐
☐
C. Vital statistics
☐
☐
☐
☐
☐
☐
☐
D.
☐
☐
☐
☐
☐
E.
☐
C&T
HTC
PMTCT
Home-based testing and counseling
EID
Pharmacy
Laboratory
Clinic cards/standardized medical records used by MOH and
implementing partners (both paper-based and electronic)
Patient-held medical records/booklets used by MOH and implementing
partners
EMR systems and patient-held electronic medical cards
Methods for recording and maintain vital statistics
Vital registration and statistics forms and documents outlining
requirements for completing forms
☐
☐
Vital registration and statistics forms and documents outlining
requirements for completing forms
Disease notification and data release policies and forms
(note if reported in aggregate or individual level and if names are included)
☐
☐
HIV reporting regulations
☐
☐
HIV notification forms
☐
☐
Other disease notification regulations
☐
☐
Data release from patient information systems (paper and EMR)
☐
☐
Vital registries
Plans and reports
☐
☐
Strategic plans: disease notification strategic plans including but not
limited to HIV, HIV surveillance strategic plans, national HIV strategic
plans
Page 49
☐
☐
☐
Reports: disease notification, patient information systems, data quality
assessments/evaluations of paper and electronic patient management
systems
NOTES (e.g. additional documents reviewed; reasons documents not provided):
Page 50
Interview Guide
Schedule interviews with the following stakeholders at least one week prior to visiting the country.
Some of the interviews may occur during the site visits.
If scheduling the interviews yourself, provide a brief overview of the SWOT assessment and why it is
being conducted. For example:
”We are interested in learning more about HIV surveillance in [country] so that
we can develop recommendations on how to implement HIV case based
surveillance in resource limited countries. Case based surveillance is longitudinal
and tracks individuals through the cascade of care. This evaluation is to assess
the current surveillance system attributes (patient monitoring, case surveillance)
and their utility for case based surveillance.”
1. National level stakeholders will be representatives from the following programs:
☐ HIV strategic information (SI)
☐ HIV surveillance
☐ HIV monitoring and evaluation (M&E)
☐ Health information systems (HIS)
☐ Infectious disease surveillance
☐ HIV testing and counseling (HTC)
☐ Prevention of mother to child transmission (PMTCT)
☐ HIV care and treatment (C&T)
☐ Early infant diagnosis (EID)
☐ TB surveillance
☐ Vital registration
☐ Laboratories that conduct CD4 cells and/or viral load testing
☐ Treatment procurement and distribution
☐ Quality improvement
2. Sub-national personnel involved with:
☐ Disease notification (HIV and other diseases)
☐ M&E reporting
☐ Other relevant programs
3. Facility level staff (include all levels of care and range of implementing partners):
☐ Clinic managers
☐ Health care providers
☐ Pharmacy manager (or pharmacy technicians if no manager)
☐ Laboratory manager (or lab technicians if no manager)
Page 51
General information
Interviewer:
Date of interview:
Respondent organization:
 Greet the key informant. Introduce yourself if you have not had previous personal contact with
him/her.
 Thank the key informant for taking the time to talk with you about HIV surveillance. Provide a brief
overview of the SWOT assessment.
 State to the key informant: All of your answers will be kept confidential and only used to inform
planning activities. If you are uncomfortable with any of the questions, you don’t have to answer
them or can choose to end the interview at any time.
Situational assessment
1. HIV notification process: national and sub-national level (from testing, diagnostic and care /
treatment providing facilities)
a. What are the policies for HIV disease notification in the country? (Prompts: Who is
required to report? What entities are involved?)
b. What is the reporting process/pathway? (Prompts: How frequently are HIV data
reported?)
c. Are disaggregate or aggregate data reported? If aggregate data are reported, at what
level are data aggregated?
d. What are the reportable events (HIV diagnosis? Sentinel events?)?
e. What variables are reported? (Prompt: are unique identifiers reported? If so, which
ones? (e.g. sex, date of birth, full or part name))
f.
How much time is spent entering, editing and storing data?
g. Has there been any evaluation of the surveillance system and if so, what were the
findings?
2. Data needs for HIV surveillance and reporting: national and sub-national level
a. What data are used for HIV strategic planning? What data are currently used for
measuring HIV prevalence and the care cascade (such as the number diagnosed,
number in care, on ART, virally suppressed)?
b. How could case-based surveillance assist in your surveillance activities?
Page 52
3. Leadership: national and sub-national level
a. Is there an identified person responsible for disease notification/HIV notification?
b. Is there investment/interest from any leaders to support case-based surveillance? Who
are the appropriate leaders to engage?
c. Is there a technical working group (HIV, disease surveillance)?
d. Is there a national strategy for collecting and using HIV data? (Prompt: is case-based
surveillance part of the HIV national strategy?)
4. Human resources: national, sub-national and facility level
(in relation to levels below respondents position)
a. National level
i. What surveillance staff report to either the HIV SI director or infectious disease
notification director? (Prompt: surveillance coordinator, data manager,
epidemiologist/data analyst?)
b. Sub-national level
i. Is there a surveillance director or someone who can serve as the director (e.g. a
HIV program manager, surveillance director for other notifiable diseases)?
ii. Is there a data manager? Who is responsible for checking data quality?
c. Facility level
i. Is there a site director (e.g. director of VCT or PMTCT)?
ii. Are there persons on site who can systematically report cases or does the
system need additional surveillance officers to report?
iii. Is there a data manager? Who is responsible for checking data quality?
iv. What training is required to collect, clean and/or interpret data?
5. Other case-based notifiable diseases: national and sub-national level
a. Is case-based notification used for any other disease reporting?
If answer to question above is yes then please proceed to Q4.b. below; if the answer is no then proceed
to Q5
b. What diseases are reported on a case basis and why are they case-based?
Page 53
c. Does notification include patient name?
d. What is the reporting process (by whom; frequency; format; variables; data use)?
e. Has there been any evaluation of the surveillance system and if so, what were the
findings?
6. HIV diagnosis: national and sub-national level and HIV testing facilities
a. Where are HIV tests conducted? (e.g. within health care facilities that provide HIV care,
facilities that do not provide HIV care, home based or self-testing, as part of surveys)
What types of tests are used? (e.g. point of care, rapid, PCR)
b. Are HIV tests conducted outside of facilities recorded? If so, how?
c. Following a positive HIV test result, are people referred into care and tracked to see
whether they access HIV care / treatment facilities? If yes, then what is the process for
referral and tracking, and what information is collected?
7. Clinical care: national and sub-national level and HIV care facilities
a. At enrolment into care, and initiation of ART, are CD4 and viral load testing conducted?
Is it routine for patient monitoring or only to confirm the need to change medications?
b. Are there plans to expand viral load testing? If so, what are these and what is the time
line for implementation? Where is VIRAL LOAD testing done (local or regional labs)?
Where are VIRAL LOAD results recorded? How are these results reported to district and
national levels and what is the turn-around time?
c. How are lab specimens collected / requested? Do clinical laboratories currently provide
results to MOH and if so, in what format (e.g. line listed, aggregate)?
d. Within laboratories are Quality Control / Quality Assessment practices systematically
run to ensure Good Laboratory Practice? (Prompts: conducted as per SOPs; participation
in External Quality Assessment; validation of results by additional staff members)
e. What records/registers are kept by the pharmacy? Does the pharmacy also report
patient level / related results upwards to the district and national levels?
8. Patient monitoring: national and sub-national level and HIV care facilities
a. Are patients in care able to be tracked overtime (i.e. is the registry longitudinal)? If so,
what is the follow-up process? What updated information is collected? Are data sources
linked (e.g. pharmacy, laboratory, and clinic records/EMRs)?
Page 54
b. Do clinic staff have specific activities that they do to track patients who have missed
appointments or who are lost to follow-up to determine whether they are no longer
accessing care, are receiving care elsewhere, have migrated out of area, or if they have
died? If so, what criteria are used to determine if a patient should be tracked, who does
the tracking, what is the tracking process, and how is the information recorded?
c. Is there a set protocol for following up patients? If so, is the protocol strictly adhered to?
If not, what are the reasons for not following the protocol? Are follow-up efforts
documented (e.g. the number of attempts to call a patient, or number of visits to the
patient’s home)
d. Among HIV diagnosed patients, how are deaths ascertained? If a death occurs in the
hospital is this information recorded in the clinic record? Is there a process for reporting
and recording deaths in the community?
9. Individual level (disaggregate) data sources: national and subnational level and relevant
facilities
a. What data systems are in place that could be utilized for case-based surveillance (i.e.
they collect and can report individual level / line listed data)? Do these systems include
EMRs?
If EMR systems are in place then please proceed to Q9.b. below; if the answer is no then proceed to Q9.i
b. Where are EMRs in use and how many different EMRs are in use?
c. Are the EMRs open source (what are approximate costs)? Is there interoperability
between EMRs?
d. Are EMRs only used for HIV or are other infectious diseases recorded? If yes, which
ones?
e. What types of EMR hardware and software are being used?
f.
Is EMR information completed in real-time without the need of paper records or are
data initially collected on paper forms and later entered into the EMR? What is the time
interval between collection of the data on the paper form and entry of these data into
the EMR?
g. What features of the EMRs work well and what are the limitations? Are there issues
relating to data quality and timeliness? If so, what can be done to improve the system?
h. Have the EMRs been evaluated? What were the findings? Was any corrective action
taken?
i.
Are there other digital systems for recording HIV clinical data such as SmartCare?
Page 55
j.
What data are collected and reported at the individual level? Are behavioural risks
coded and reported? What patient identifiers are collected?
k. How comprehensive are the systems (i.e. what proportion of all people diagnosed with
HIV is captured)? Does this vary geographically or by key populations? Has this been
evaluated?
l.
Have the data collection systems been evaluated for completeness and accuracy? If so,
what were the results and when was the evaluation conducted? What were the main
findings?
m. Is there ongoing data quality improvement / data quality assessment? If so, what were
the findings? How are data cleaned? What data quality controls are in place?
n. Are there data security systems and policies for patient-level data? How are records
archived and disposed?
o. Have changes been made to the system (including EMRs) in the past (new variables,
new technology)? Was the system easy to change or did problems arise? When did
these changes occur?
p. Have there been times when the system has been down (i.e. was not functioning either
because of a problem with the system, electricity, or lack of staff)? How often did this
occur in the past 12 months? For how long was the system not functioning?
q. What reports are generated from the data and at what frequency and for whom? What
is the timeliness of the data reported?
r.
Are the reports individual level or aggregate? If aggregate data, where does aggregation
occur and have there been data queries / questions that have been difficult to answer
due to reporting data in aggregate?
s. How are data currently used (cohort analysis, cascade, etc.)? Is there a process to
systematically use the information to inform relevant policies, strategies, update
HIV/AIDS control and treatment programs?
t.
Is there interest in, or a vision of how, to use patient monitoring data for case-based
surveillance?
10. Unique identifiers: national and subnational level and at all facilities
a. Are there national identifiers (either for health programs or external national identifiers)
that are, or that could be, used as part of case-based surveillance?
b. Are identifiers currently utilized to de-duplicate reporting of cases (at diagnosis or in
care) and link information relating to a single case over time? If so, which ones and
how?
Page 56
c. What quality assurance has been done on the identifiers?
d. Are any identifiers collected but not used? If yes, what can be done to have them
included in case-based surveillance?
e. Are names and dates of birth collected at HIV testing sites/services or HIV
care/treatment sites? If names are not currently collected will it be possible to do so in
the future? How would patients and communities currently view the collection of
identifiers among people diagnosed with HIV? Do people generally know their complete
date of birth (i.e. day, month, year)?
f.
Have biometrics been used in relation to HIV? Have biometrics been used in any other
setting? If so, describe these.
11. Ascertainment of recent HIV infections / incidence: national and subnational level
a. Do you / have you ever / do you have plans to apply a test to blood (liquid serum or
plasma) specimens to distinguish recent HIV infections from those which are longstanding (for example, using the LAg-Avidity EIA)? If yes, then please provide further
information (for example, how samples are taken and stored, which test for new
infection was / will be used, whether the incidence assay was conducted on the same
specimen as viral load and CD4 cell count, and what standard testing algorithm / cut off
point was used for the assay), where the test was/will be conducted, how specimens
can be transported).
b. To calculate the false recent rate of tests (applied now or in the future) certain
information on the person diagnosed with HIV is required. Which of the following
information is collected on people newly diagnosed with HIV (within 3 months of an HIV
positive test) –
i.
ii.
iii.
iv.
v.
CD4 cell count:
Viral load:
Previous or current antiretroviral therapy:
HIV sub-type:
Date of most recent negative HIV test
YES/NO
YES/NO
YES/NO
YES/NO
YES/NO
Comments:
12. Ascertainment of death: national and subnational level and at all facilities
a. Are deaths routinely recorded in the country? (Prompt: Are there laws or regulatory
issues that result in complete recording of deaths (e.g. are death certificates required
for burial)?)
Page 57
b. How are deaths recorded? (Prompts: Who is responsible for recording deaths? Is there a
death certificate? What information is included on the death certificate? If a cause of
death is recorded how is that determined?)
c. Where and in what format are death records stored? Which personal identifiers are
stored?
d. Are there any Health and Demographic Surveillance Sites?
e. Are there death records available for linking with HIV case records? If yes, then have
death records been linked to HIV case records and how successful was this?
Page 58
Assessment of feasibility, barriers, and needs
1. What are the resources/costs of operating current systems?
2. What are the funding sources for surveillance, including EMRs? What is the potential for further
funding?
3. Can staff at clinical facilities report HIV cases? Do they have the time and motivation?
4. Is there someone within clinical settings who reports other infectious diseases? Can that system
be expanded to include HIV case-based surveillance? How might that be done?
5. How could implementing partners assist in developing case-based surveillance? Which partners?
6. Are data security and patient privacy measures of sufficient strength to support and instill
confidence in a HIV case reporting system?
7. Does MOH have the capacity to receive and process electronic reports?
8. Does the capacity exist for electronic reporting from sites and clinical laboratories?
9. What are the technological limitations within the country that might affect case-based
surveillance?
10. What barriers do you foresee to developing case-based surveillance?
11. Can the system evaluated be expanded or adapted for national HIV case-based surveillance?
What would this require? What steps can be taken to move this system forward? Are these
elements that can be pilot tested?
12. Is there anything else you would like to add?
Thank you for your participation!
Page 59
Site Visit Checklist
Travel logistics:
☐ At least three weeks prior to the visit, coordinate with MOH/CDC places for potential of site visits
☐ Sites in the capital area and a high prevalence area, which is geographically different
☐ Sites varying in size (e.g. hospital, health center, HIV testing only)
☐ Sites varying by partner support (e.g. MOH, USAID, NGOs)
☐ Sites varying by patient monitoring system and EMR type
☐ Laboratory sites performing viral load/CD4/EID
☐ Sites with innovative systems or surveillance methods
☐ Plan agenda for the visit and review with MOH/CDC contact
☐ Schedule date and time to allow someone from MOH/CDC to attend
☐ MOH/partner lets facility know ahead of time that we will be visiting
☐ Obtain point of contact information
☐ Schedule transport/travel if overnight
Sample agenda:
 Welcome and introductions (meet in charge)
 Overview of the project
 Facility tour
 Observe HIV surveillance process
 Interview a few key surveillance staff, as necessary
Sites:
Site Name & Location
1
Visit Date & Time
Reason Site Selected
2
3
4
5
6
Page 60
General information
Site Visitor:
Date of visit:
Site:
Type of site:
 Greet the in charge. Introduce yourself if you have not had previous personal contact with him/her.
 Thank them for taking the time to let you allow you to visit their facility to observe HIV surveillance.
Provide a brief overview of the SWOT assessment.
Assess:
☐ Existing infrastructure
☐ Human resources and need for training
☐ Reporting mechanisms
☐ Technology capacity and interoperability of systems
☐ Data standards/MOH and implementing partners collect same information
☐ MOH and implementing partner’s use of the information produced by CBS or PMS
☐ Feasibility in capturing pharmaceutical data
☐ Feasibility in capturing lab data
☐ Feasibility of linking vital stats data to patient monitoring/CBS
Page 61
Observe:
☐ Ability to uniquely identify patients
☐ Demographic variables collected
☐ Care continuum measures collected
☐ Flow and transfer of data
☐ Vital stats availability and completeness (especially cause of death)
☐ Data quality and use of data, including data from laboratories
☐ Innovative ways to capture diagnosis to reduce paper-based
☐ Patient flow/experience (how and when data are recorded, reasons why data might not be
recorded)
Page 62
Debrief Form
The last day in-country the team will debrief with the stakeholders.
Highlights of the SWOT for case based surveillance and patient monitoring systems
Strengths identified:
Weaknesses identified:
Opportunities identified:
Threats identified:
Summary of Data Available for CBS Measures:
Yes
No
☐
☐
HIV diagnosis date
☐
☐
Entry to care date
☐
☐
CD4 results and date (first and follow-up)
☐
☐
Viral load results and date (first and follow-up)
☐
☐
Initiation of ART date and regimen
☐
☐
Advanced HIV diagnosis date (WHO stage III or IV)
☐
☐
Death date
Additional Comments:
Page 63