12-Month Infant HIV-Free Survival (HFS)

Export Indicator

The percentage of infants born to HIV-positive women in PMTCT programs who are alive at 12 months of age and HIV negative. HIV-positive women is defined as all women with a confirmed HIV-positive test result. PMTCT programs is defined as provision of a c
What it measures

This indicator measures PMTCT program effectiveness among infants
whose mothers have received PMTCT services. Measuring infant HIV status through early infant
diagnosis (DNA polymerase chain reaction [PCR] conducted at the 6-week vaccination visit)
measures HIV transmission during pregnancy and delivery. This does not reflect HIV
transmission occurring through breastfeeding during the postnatal period, which constitutes as
much as half of the total maternal-to-child transmission (MTCT). Infants born to HIV-infected
mothers are not only at risk of MTCT, but also at significant risk of mortality from other primary
causes, e.g., pneumonia, diarrhea, malaria, and malnutrition, even if they are uninfected by
HIV. The burden of non-HIV mortality in these HIV-exposed infants can be double the burden of
HIV infection and mortality. Thus, by examining HFS at 12 months, we ensure that children born
to mothers in PMTCT programs are not only successfully avoiding MTCT, but also surviving
infancy. Twelve-month HFS rates can be compared to the infant mortality rates (IMRs) in the
general population and will reflect the effectiveness of a package of interventions in the first
year to reduce MTCT and improve survival, particularly postnatal maternal ART or
maternal/infant antiretroviral (ARV) prophylaxis, postnatal care (PNC), and appropriate
breastfeeding and complementary feeding for the majority of infants. Furthermore, a 12-month
HFS measurement will capture the majority of early childhood MTCT and mortality, and will
face less loss-to-follow-up compared to measuring HFS at 18 or 24 months.

Interpretation. Infants born to HIV-infected mothers are not only at risk of MTCT, but also of
mortality from other primary causes, e.g., pneumonia, diarrhea, malaria, and malnutrition, even
if they are uninfected by HIV. The goal of PMTCT programs should not be just to reduce MTCT,
but to maximize HFS. To achieve this, PMTCT programs will be required to provide:
• Universal HIV testing of women (and male partners) in antenatal care
• Appropriate HIV treatment or prophylactic regimens for pre- and postnatal women or
postnatal prophylaxis of infants
• Obstetric and post-delivery care of mothers and infants to reduce neonatal mortality
• Infant feeding counseling and support for exclusive breastfeeding to 6 months of age and
continued breastfeeding with adequate complementary feeding to at least 12 months or
replacement feeding from birth
• Routine growth monitoring and promotion
• A full package of basic health care interventions, particularly basic immunizations,
cotrimoxazole, routine vitamin A supplementation, insecticide-treated bednets, and oral
rehydration therapy/zinc supplementation for acute diarrhea
• Appropriate clinic referral and treatment of illness

PMTCT programs that do not comprehensively address the full spectrum of needs of these
highly vulnerable HIV-exposed infants beginning in pregnancy and extending throughout the
postnatal period will have limited impact on HFS. While HIV-exposed infants are highly
vulnerable through at least the first 2 years of life, evaluating HFS at 12 months (in a cohort 12–
14 months postpartum) provides a credible and significant measure of program success
through the period of highest vulnerability.

Thus, HFS indicates the extent to which infants avoid infection from their mothers, as well as
the extent to which infants born to HIV-infected mothers avoid death despite their higher risk
of mortality within HIV-affected households. Modeling will yield a population-based estimate of
HFS in the general population, while program samples drawn from representative PMTCT sites
and mothers that have been enrolled in PMTCT services will provide a measure of program
effectiveness.

Rationale

Uses. Population-based estimates derived from modeling will help donors, national
governments, ministries of health, and other international organizations with program
planning, while program-level estimates will help them assess the success and failure of PMTCT
programs. For programs that are able to disaggregate the data by ARV intervention and infant
feeding practice, the indicator will also help describe the relative contribution of treatment and
prophylaxis regimens, infant feeding methods, location, service provision, and other factors
critical to HFS.

Numerator

The number of those infants that are alive and that test negative for HIV using a PCR test between 12 and 14 months of age. Most HIV-exposed infants in the sample can be identified and HIV-tested in conjunction with a 12- month routine clinic visit to provide maternal and child health care

Denominator

A random selection of HIV-exposed infants from PMTCT clinic registers from a representative sample of PMTCT clinics

Calculation

Numerator / Denominator

Method of measurement

Currently, this indicator is best suited for collection at the national level, using modeling techniques developed by WHO.1 The key elements required for modeling are the underlying IMR and the MTCT rate. To provide the latter, national programs must provide some information on the range of maternal CD4 counts, ARV prophylaxis regimens
given to mothers and/or infants, and infant feeding practices.

WHO currently collects this indicator through modeling at 18-months. Moving forward, as
PMTCT data collection and follow-up systems become more refined, implementers may
consider collecting this indicator at the program level. In this case, the required data elements
to calculate HFS are maternal HIV infection status at time of birth, living infant HIV infection
status, and deaths among HIV-exposed infants prior to and including 12 months of age. The
denominator is a random selection of HIV-exposed infants from PMTCT clinic registers from a
representative sample of PMTCT clinics. The numerator is the number of those infants that are
alive and that test negative for HIV using a PCR test between 12 and 14 months of age. Most
HIV-exposed infants in the sample can be identified and HIV-tested in conjunction with a 12-
month routine clinic visit to provide maternal and child health care. Mother/infant pairs or
mothers whose infants have died that do not present at the clinic between 12 and 14 months
can be contacted in the community.

Another option for programs is to collect the indicator through special studies that are limited
in scope and duration. For this approach, the study team will need to collect the same elements
as required for the program-level indicator, including the disaggregations. The validity of this
approach will hinge largely on the ability of the study team to randomly select a representative
sample of HIV-positive mothers and HIV-exposed infants.

Data collection method. At program level, the data for this indicator will be collected from
representative samples of PMTCT sites and, within those sites, from representative samples of
HIV-positive women and their infants that have participated in PMTCT programs based on
PMTCT registers. Data will be collected from clinical records to establish that these women
were HIV-positive in pregnancy, enrolled in a PMTCT program, and gave birth to a live infant.
The Three Interlinked Patient Monitoring Systems for HIV Care/ART, MCH/PMTCT and TB/HIV:
Standardized Minimum Data Set and Illustrative Tools, published by WHO in 2010, include
sample forms to collect all the necessary data elements for this indicator. WHO also has
developed a protocol for an Annual Patient Monitoring Review, which provides guidelines for
extracting data from patient records and facility registers to collect indicators that are not
routinely reported, but for which data are available, including this one. Between 12 and 14
months postpartum, mothers will be contacted within the clinic (routine maternal/well-child
12-month visit or other clinic visit) or contacted at home to determine whether the infant is
alive and to collect a blood sample from the infant to establish its HIV status by standard PCR
testing. If an infant is reported as having died, death will be confirmed and documented.

Frequency of measurement and reporting. This indicator should be measured every year
through modeling at a national level and every year at the program level for programs that can
collect information on a representative sample of HIV-positive mothers that were enrolled in a
representative sample of PMTCT clinic programs and gave birth to a live infant 12–14 months
previously.

Disaggregation. Disaggregation is optional. For programs choosing to disaggregate, data can
disaggregated by maternal treatment and maternal/infant prophylaxis regimens and infant
feeding practices.

Measurement frequency

Annual

Disaggregation
Explanation of the numerator
Explanation of the denominator
Strengths and weaknesses

Strengths. This indicator represents the first globally harmonized measure of HFS in infants and
provides a population-based estimate of HFS (through modeling) or a measure of overall
PMTCT program effectiveness (if collected at the program level). The indicator can drive
programming and provide a basis for a more balanced and rational public health approach to
address the needs of highly vulnerable HIV-exposed infants.

Weaknesses. HFS is an evaluation of PMTCT program effectiveness through the first 12 months
of life, so there will be a lag in measuring an impact on HFS when implementing new
interventions.

The indicator has several weaknesses if collected at the program level. It does not reflect the
HIV-positive women that give birth but that are unaware that they are infected or do not
attend antenatal or postnatal PMTCT and care services. For this reason, the indicator will reflect
program performance for those HIV-positive women that are enrolled and receive PMTCT and
care support. It is also dependent on 12–14 months follow-up with mothers and surviving
infants that may not come to the clinic for a routine 12-month visit. Loss-to-follow-up between
12 and 14 months may be significant if systems are not in place to track and retain
mothers/infants in clinic care. Additionally, measurement of this indicator will require the
ability to collect patient-level data at sites. Currently, most sites report aggregate-level data,
from which it is not possible to link individual mother/infant pairs across time.

Resources required. The resources required to collect this indicator will depend on the method
and level chosen for indicator collection.

If collected at national/population-level through modeling, ministries of health must provide
the underlying IMR and data elements required to estimate the MTCT rate. These include the
range of maternal CD4 counts, ARV prophylaxis/treatment regimens given to mothers and/or
infants, and infant feeding practices. Once these elements are available, ministries of health can
work with WHO and its partner organizations to run the actual modeling technique. In future, it
is possible that ministry staff themselves will acquire the capacity to determine the HFS rate
using modeling. Of the three approaches described here (see below for the other two),
modeling is the least expensive, though it provides a population-based, not program-level,
estimate of HFS.

For programs considering collecting this indicator through routine monitoring approaches, they
will require registers and patient tracking tools that provide all of the required data elements.
One example of such a tool or register is the Three Interlinked Patient Monitoring Systems for
HIV Care/ART, MCH/PMTCT and TB/HIV: Standardized Minimum Data Set and Illustrative Tools,
published by WHO in 2010. In this case, the program will need to ensure adequate provider
training on the registers and tools, as well as training on the approach to data aggregation.
Programs will be most successful collecting this indicator under a number of conditions. First,
PMTCT programs with high retention rates through the 12-month mark will have an advantage. Secondly, programs that can collect patient-level data and track individual clients across time
and potentially across facilities will be able to produce more accurate data.

For programs considering collecting this indicator through special studies, they must have the
financial resources and human resources/skills to develop and implement a rigorous study
design. This will include resources for study design, field-testing, data collection, and data
analysis.

The indicator at the population level will soon be collected by WHO for routine monitoring of
national programs. At the program level, the indicator will be field-tested by FANTA-3, the
Elisabeth Glaser Pediatric AIDS Foundation (EGPAF), and others, and revised accordingly once
the field tests are complete.