SUMMARY:

The KZN 20,000 project aims to significantly reduce perinatal HIV transmission within 2 years and improve

overall child survival within 5 years in three districts of KwaZulu-Natal (KZN) through health system support

interventions that would increase the effectiveness of current Prevention of mother-to-child transmission

(PMTCT) services.

BACKGROUND:

HIV infection in children is preventable. In Europe and the United States mother-to-child transmission rates

have been reduced to less than 2% and few HIV-infected babies are born in these countries. This has been

achieved through active screening and thereby identification of HIV-infected women attending antenatal

clinics, the early initiation of highly active antiretroviral treatment (HAART) whilst women are pregnant,

delivery of infants by cesarean section and the avoidance of all breast milk.

Implementing the same interventions and achieving the same low transmission rates has not been realized

in most resource-poor settings. Whilst the relatively restricted PMTCT protocols that have been applied in

most South African provinces cannot be expected to produce the results seen in developed nations (where

perinatal HIV transmission is the exception) there is an unacceptable gap in performance of the existing

PMTCT programs. Numerous obstacles have contributed to the failure of national and district health

systems to successfully operationalize international PMTCT recommendations. While in some cases there

are genuine deficiencies in human and physical resources, as well as incomplete training, experience in

rural and urban South African PMTCT program points to a widespread failure to reliably deliver the

sequence of simple processes of care (e.g. determining a mother_HIV status, reliable dispensing of

prophylactic drugs). Additional transmissions occur due to inappropriate infant feeding choices by HIV-

infected mothers either because of poor antenatal counseling or/or lack of support from health workers.

Target population for the project includes pregnant women, their infants, and health care workers at the

district health facilities. The emphasis areas are training, strategic information and local organization

capacity building.

ACTIVITIES AND EXPECTED RESULTS:

The KZN Department of Health (DOH) and University of KwaZulu Natal agreed that the project, now known

as KZN 20,000 , would proceed in three phases. Phase I comprises of a situational analysis and planning

exercise (currently underway) to determine the level of health system intervention required to effect large

scale improvement of the PMTCT program. Phase II will focus on rapid scale up of system strengthening

and priority activities to improve effectiveness of PMTCT. Phase III will focus on infant feeding strategies

and community mobilization. FY 2008 funding will be used to implement phase II activities. Subsequent

years of funding will ensure the implementation of Phase III.

Activity 1: Implementation of KZN 20,000

KZN 20,000 will operate across three districts that were chosen by the KZN DoH, namely Ethekwini

(Durban and immediate surroundings), Ugu and Umgungundlovu. Ugu district is a Presidential nodal site,

meaning that it has been designated as a district that is exceptionally poor and under-resourced.

Umgungundlovu contains Pietermaritzburg, the second largest city in KZN. Durban and Pietermaritzburg

both have large areas of informal housing and peri-urban areas with extremely poor communities. The 3

districts contain more than half (~5m) of the entire population of KZN (~9m) and suffer high antenatal HIV

prevalence rates - Ethekwini 41.6%, Ugu 38.9% and Umgungundlovu 44.4%. Between the three districts

there are over 260 PHC clinics and 16 state hospitals delivering 82,000 babies per year. The project is

designed to reduce the number of infant infections in the three districts by 4,800 per year and improve the

health of HIV-infected mothers through strengthening of the existing health system and capacity

development at district and local level. Emphasis will be given to careful documentation of process and

monitoring of outcomes so that best practices and lessons can be rapidly extended to the other 8 districts

within the Province. The KZN DoH is committed to the project and is using it as a way of improving overall

health care management and service delivery. The health system support intervention (Phase II) will also

create a platform from which to introduce interventions to improve infant feeding practices that are critical

for preventing infant HIV infection and reducing infant mortality.

Activity 2: Health System Improvement Intervention:

KZN 20,000 will introduce health systems improvement intervention designed to improve the quality of

PMTCT services across 3 districts. The project team will train and mentor mid-level Primary Health Care

(PHC) supervisors in quality improvement methodologies and management skills that will be supported

through a data collection and monitoring system specifically designed and supported by the project.

Implementation of PMTCT services will remain the primary responsibility of health staff in clinics and

hospitals. Routine PMTCT performance indicators will be tracked as well as 3 outcome indicators namely a)

infant HIV prevalence rates at immunization clinics, b) population-based infant mortality rates and c) in-

patient child mortality.

Activity 3: Development of a Data System:

A robust system that allows for timely and accurate collection, transmission (to central data assembly

points), collation, translation and feedback of data is a critical component of an effective improvement

intervention. An IT system is in development that will install a local MS Access database on computers that

will be placed in each District Information Office. All applications will work independently but data will be

automatically uploaded to an SQL database on a remote server each day. Data security can be assured by

use of digital certificates such that data is only accepted from pre-specified machines which have valid

certificates installed. A dedicated data assistant will be located in each district office to capture and manage

data from each clinic and to produce reports for the PHC supervisors. This will initially be a system that runs

parallel to, and will derive data from the current provincial data collection system. There will be no

duplication of data collection since all data will flow to the provincial office. It is anticipated that the systems

will be merged at the end of the funding period if the potential benefits of the proposed system are realized.

The main output of the data system will be to run system performance data reports for program leadership

and site specific process performance reports (line charts and histograms) to guide activities of the nursing

supervisors and clinics staff.

Activity 4: Development of Learning Networks:

Prior experience with large scale improvement interventions indicates that change is accelerated when

successful ideas are transmitted from peer-to-peer, and when a culture of peer support can be developed.

In a traditional quality assurance environment, the front-line staff receive instructions to improve across a

Activity Narrative: broad array of indicators in what is often a pejorative context. The purpose of the learning network is to

bring together small teams (e.g. facility manager, nurse, counselor) from each health care site to set

common project aims, learn together how to map care processes, identify obstacles and solutions, learn

how to test innovations and how to collect data to track improvement. Additional support will be given to

poor performing clinics. At Learning Sessions, sites that are struggling will also be exposed to participants

from high performing sites who will share their experience and strategies for success. Between these

Learning Sessions, quality mentors will visit the hospitals and together with PHC supervisors will visit the

clinics regularly (1-2 times/month) to support the teams, and sustain the improvement process through

planning new tests of change. The concept of learning networks will apply also to the mentoring of the PHC

supervisors themselves who will be brought together each month for training, transparent review of team

progress, and peer support for successes and challenges in the field.

Activity 5: Monitoring of Infant HIV transmission rates, Infant and Child Mortality:

Perinatal transmission rates will be routinely monitored at sentinel sites in each district through surveillance

of all infants attending 6 week immunisation clinics. Dried blood samples will be collected from all infants

following informed consent from the mother or legal guardian regardless of whether the mother was part of

the PMTCT programme or not. Maternal, infant HIV prevalence rates can be determined as well as vertical

transmission rates.

The goals of project 20,000 are directly aligned with the goals and objectives of the US

President_Emergency Plan for AIDS Relief (PEPFAR). These goals include achieving primary prevention of

new HIV infections through expanding VCT programs and building programs to reduce mother-to-child

transmission. KZN 20 000 aims to improve the overall performance of PMTCT and thus reduce the

incidence of new perinatal HIV infections. Improvement of PMTCT has other desirable indirect outcomes

which include early diagnosis of HIV that leads to increased access to HAART, decrease in infant mortality

rates and overall improvement in child survival.