Global SouthSouthNorth Network of Developers and Users in Open Source Development for Health

1
Global (South-South-North) Network of Developers
and Users in Open Source Development for Health

• The Health Information Systems Programme
  Network -
• Mobilising Sharing of Scarce Expertise and
  Resources

Calle Hedberg WITFOR Aug 2005 1
2
HISP South Africa Partner with DEPARTMENT OF
HEALTHRepublic of South Africa

• Calle Hedberg
• National Health Information System
• Department of Health

3
Quadruple Burden of Disease

• A tiny percentage of Africans 2-3 - have
  world class
• (private!) health care
• The 97-98 majority rely on a public health
  system that
• range from functioning to non-existent
  Sangomas (80)
• Primary burden of disease related to
  backwardness,
• poverty, illiteracy (colonialism
  parasitic leadership)
• Diseases related to Western/urban lifestyles
  (drugs,
• alcohol, smoking, junk food) -gt double
  burden
• HIV/AIDS -gt triple burden (Note distorting
  donor funding)
• Crime, civil strife, traffic -gt quadruple
  burden
• There has been a strong bias towards hospital
  as well as
• curative care (repairing symptoms, not
  causes)

Calle Hedberg WITFOR Aug 2005 4
4
Centralisation Donor Dependency

• Health Care Measurement in Africa has been
  dominated
• by (donor-driven!) fragmented efforts
• Pockets of functional systems have eroded over
  time,
• or being sidelined as a result of new brooms
• Most systems are designed to serve the top
  management
• (Ministry, WHO, etc), not health workers
  and patients
• Data/info exclusively flowing upwards
  (stargazing)
• Most mgmt systems are designed to provide
  historical,
• descriptive stats not information for
  action
• Strong belief that Health IS technology
  acquisition
• Western hi-tech HIS fiascos repeated in Africa
• Very limited public access to vital health
  information,
• despite progressive legislation in several
  countries (SA)

Calle Hedberg WITFOR Aug 2005 5
5
HISP-SA Research, Development Implementation

• Network of universities, ministries, NGOs,
  companies
• Partnership with government crucial lt-gt
  independence
• Started as pilot project in Cape Town 1994
  (struggle-gt)
• HISP went national in 1998-2000, then
  international
• Currently HISP nodes/activities in South
  Africa, Malawi,
• Mozambique, Tanzania/Zanzibar, Ethiopia,
  Nigeria,
• Uganda, Swaziland, Botswana (2005), India,
  China,
• Vietnam. Potentially Ghana, Namibia,
• HISP-21 (not-for-profit) System dev and
  implementation
• HISP-UWC Research Education
• NGO partners Health Systems Trust, Valley
  Trust,
• Academic partners Univ. Pretoria, Wits,
  KwaZulu-Natal
• Private sector partners BEE, JSI, MSH, HPI,
  HSLP

Calle Hedberg (HISP) WITFOR Aug 2005 3
6
District Organisational Model
Calle Hedberg WITFOR Aug 2005 5
7
Bottom-up Top-down

• Shift perspective from workload to public health
  delivery
• Local flexibility through Essential Data Sets
  at all levels
• Shift power from IT people to health
  managers/workers

Indicators, Procedures, datasets use of info
for ACTION
International IS
National Inf. Systems
Community
District
Provincial Information Systems
Province
District Information Systems
National
International
Community Information Systems
Calle Hedberg WITFOR Aug 2005 6
8
Time Line 1996-2003

• 1997-2000 Routine data for Primary Health Care
  (public)
• 2000 Routine data for public Hospitals (still
  gaps!)
• 2000 Routine data for private hospitals
  (stalled! MAL)
• 2001 Start semi-permanent data
  (infrastructure, staff)
• 2001 Emergency Medical Services pilot
  (roll-out 2006!)
• 2002 Clinic Audits from 1998 to 2003 (ECP)
• 2002 Data from PMTCT research sites
• 2002 National Tertiary Services Grant
  (budgeting!!)
• 2002 Hospital Revitalisation Programme
  monitoring
• 2002 Client Satisfaction Surveys (picked up in
  2003/04)
• 2003 Financial data captured or imported
  (CM, ECP)
• 2003 National Facility Survey (437 facilities
  / SAHR-03)

Calle Hedberg WITFOR Aug 2005 6
9
PHC/OPD - Data Input Coverage

• Data input coverage stabilised at 98-99
  (remainder interpolated!)
• Targets for data flows (timeliness!) achieved
  for 7 of 9 provinces

Calle Hedberg WITFOR Aug 2005 7
10
Time Line 2004-2006

• 2004 STI (HTA) Sentinel Surveillance
• 2004 HIV Services Mapping project data
• 2004 Quarterly Reporting System linked to
• Provincial Strategic Planning (MTEF) 60
  DHIS data
• 2004 Mass vaccination campaign data
• 2004 Various patient data sets (DHIS_PAT
  module)
• 2005 New National Indicator / Data Set (200
  ind, 200 DE)
• 2005 Home Based Care data (pilots)
• 2005 Human Resource Development (pilots)
• 2005 ART aggregated data (some provinces)
• 2005-06 Environmental Health, School Health,
  EMS
• 2006 Correctional Services, Lab data, .

Calle Hedberg WITFOR Aug 2005 8
11
Client Satisfaction Surveys

• Batho Pele (patients first!)
• Simple questionnaire
• Done by independents
• Integrated with DHIS
• DoH 50-150 hosp
• Provinces/Mun PHC

• SA needs balance between routine, sentinel, and
  survey data sets
• Routine Info System / DHIS must bind disparate
  sets together

Calle Hedberg WITFOR Aug 2005 9
12
Emergency Medical Services

• Developed with minimal HISP support by Siyanda
  HD, NC
• Main factor in making EMS service more efficient
  (80c/km)

Calle Hedberg WITFOR Aug 2005 10
13
Human Resource/Skills Module

• Web-based system complementing/interfacing to
  other HR systems
• Focus on local managers need to track staff,
  staff cost, HRD
• Tracks courses/training, requests for training,
  outcome
• WAP/GPRS enabled?

Calle Hedberg RHINO Oct 1 2003 11
14
Education/Training 80 of total effort

• Sensitise ALL health workers on data
  information use
• Training in data collection lt-gt streamline
  collection tools
• Training on-the-job support in analysis use
  crucial
• Training in building cases, presentation,
  lobbying
• Management training Feedback versus
  stargazing
• Range of 5-day training courses (Mx, EDP, DHIS,
  etc)
• 2-6 weeks courses with certificates (Pret, Dar,
  Moz)
• Certificate and diploma courses at African
  institutions
• M.Sc. Programmes in public health / health
  informatics
• Ph.D. programmes as South-North partnerships
• Education/research linked to DHIS activities in
  the field
• 60 M.Sc. 20 Ph.D. students involved in DHIS
  dev

Calle Hedberg WITFOR Aug 2005 11
15
Bohlabelo
South Africa RDS nodes
Thabo Ofutsanyane
Kgalagadi
Sekhukhune
Umkhanyanakude
Zululand
Umzinyathi
Ugu
Alfred Nzo
OR Thambo
Ukhahlamba
Chris Hani
Central Karoo
16
A Good HMIS drives Delivery - TB
- South Peninsula used DHIS to increase cure rate
from 61-85 - Same approach replicated in whole
City 66-75 and climbing
Calle Hedberg WITFOR Aug 2005 7
17
Steady Expansion Nation/International

• Main projects 2004-2006
• Eastern Cape Hospital Ward Info Systems
• HIV/AIDS information system/use training (SA
  DoH)
• Country-wide management support/training DHIS
  (DoH)
• WITFOR.. Training/support to NI, TZ, SZ, MW,
  MZ, NA
• Waiting time surveys (Rockefeller)
• EQUITY Gauge (Atlantic Philanthropies)
• District Health Expenditure Reviews (with
  Valley Trust)
• DHIS version 1.4 (MS Office)
• National Data Dictionary (Java)
• National Web-based Data/information Repository
  tools
• Notification system / surveillance / sentinel
  systems
• EPR system pilot for PHC using Tablet PCs
  Biometrics

Calle Hedberg (HISP) WITFOR Aug 2005 17
18
Sharing and Support WITFOR/BEANISH

• BEANISH comprises nine countries (7 south, 2
  north)
• Funded by European Union over 30 months
• Strengthening the HISP network
• Sharing and collaboration in Open Source HMIS
• Implement HISP/DHIS in Botswana (see separate
  pres)
• Starting NOW (although some activities already)

Calle Hedberg (HISP) WITFOR Aug 2005 18
19
DHIS v 1.4 Core Specs

• Flexible OrgHierarchy with anything from 2 to n
  levels
• Routine Data of any collection frequency (daily
  -gt yearly)
• Semi-permanent Data (staff, infrastructure,
  pop-estimates)
• Survey Data (often captured elsewhere linked
  to DHIS)
• Indicators spanning all three types of data
• Data elements, indicators, validation rules
  user-defined
• Module for capturing patient-data (with
  auto-aggregation)
• Basic system in MS Access and Excel (pivot
  tables, graph)
• Easy transfer of data to e.g. MS SQL, MySQL,
  ORACLE
• Web Pivot Reporter Web Portal Access via
  browsers
• GIS tools for easy creation of thematic maps
• 1.4 basis for DHIS v 2.0 (in development, Java,
  web-based)

Calle Hedberg WITFOR Aug 2005 13
20
Policy Vision is Crucial!

• Equity Quality in Health is fundamental for
  development
• Our quadruple burden of disease are
  increasingly integrated in
• our people integrated information systems
  crucial to good service
• HISP workers At least 80 from the South,
  80-20 national/regional
• Sharing collaboration key strategy to balance
  donor dominance

Routine Health Information Systems Are a Core
Vehicle for Integration Political Action!
BUT RHIS must be integrated with or interfaced
to Semi-permanent survey data, sentinel
systems, Community-based IS, Information sharing
systems Bottom Line HMIS -gt impact on health
Calle Hedberg WITFOR Aug 2005 17