Digital health services highly effective in early infant diagnosis of HIV in Kenya

A digital health system combining online services with text messages is highly effective in early infant diagnosis of HIV according to findings from a cluster-randomised controlled trial in Kenya published in The Lancet HIV.

The HIV Infant Tracking System (HITSystem), a novel integrated system, links providers of early infant diagnosis, laboratory technicians and mothers and infants to improve outcomes for HIV-exposed infants. Infants allocated to facilities with this intervention were significantly more likely to get complete early infant diagnosis services (85%) than those assigned to control facilities (60%).

Early diagnosis of HIV infection is essential for the timely initiation of life-saving antiretroviral therapy (ART) for infants born to HIV-positive mothers so preventing death and disease. Disease progression is rapid in the first few months of life. The Children with HIV (CHER) study showed a 76% reduction in disease and a 75% reduction in death, with notable short-term cost reductions, when HIV-infected infants started ART under three months of age.


polymerase chain reaction (PCR)

A method of amplifying fragments of genetic material so that they can be detected. Some viral load tests are based on this method.

deoxyribonucleic acid (DNA)

The material in the nucleus of a cell where genetic information is stored.

transmission cluster

By comparing the genetic sequence of the virus in different individuals, scientists can identify viruses that are closely related. A transmission cluster is a group of people who have similar strains of the virus, which suggests (but does not prove) HIV transmission between those individuals.


Studies aim to give information that will be applicable to a large group of people (e.g. adults with diagnosed HIV in the UK). Because it is impractical to conduct a study with such a large group, only a sub-group (a sample) takes part in a study. This isn’t a problem as long as the characteristics of the sample are similar to those of the wider group (e.g. in terms of age, gender, CD4 count and years since diagnosis).

disease progression

The worsening of a disease.

Confirmation of an HIV-exposed infant’s HIV status begins at six weeks with a succession of steps taking up to 18 months for testing to be completed. As maternal antibodies cross the placenta and persist in infant blood for up to 18 months, antibodies detected in infants usually represent exposure to maternal HIV, not true infant HIV infection. Antibody testing can only be used to exclude infant infection after 12 months of age or confirm infection after 15-18 months of age. Accurate diagnosis before this needs detection of viral components (virological testing) including DNA integrated into host cells. DNA can be detected by PCR-based tests.

In Kenya, the cascade of care includes:

  • dried blood spot sample collection for HIV DNA PCR testing at six weeks of age
  • shipment and processing of the sample at a central laboratory
  • return of the paper-based results to the health facility
  • notifying the mother of the result and
  • either starting ART for HIV-positive infants or antibody retesting at nine and 18 months for HIV-negative infants.

In Kenya, as in other high volume and remote resource-poor settings, implementation of early infant diagnosis programmes with completion of the diagnostic steps is challenging. Barriers include successfully identifying, offering and acceptance of testing among those HIV-exposed as well as unknown; accurate specimen collection, transport and laboratory processing; getting results to healthcare providers and infants’ caregivers; and difficulties in retaining mother-infant pairs in care due to transport issues as well as social and economic factors.

While early infant diagnosis increased from 36% in 2012 to 53% in 2016 it falls far short of the target of testing all HIV-exposed infants by six weeks of age and starting ART for HIV-positive infants by 12 weeks of age.

Evidence of the usefulness of digital health services for maternal and child health is increasing, including SMS text messaging to improve early infant diagnosis, completion of the initial test and turnaround time for PCR results.

The authors undertook a cluster-randomised trial in six hospitals to determine the efficacy of the HITSystem, a comprehensive eHealth initiative, to improve the quality and efficiency of early infant diagnosis services in Kenya. The HITSystem tracks infants and sends alerts to hospitals, laboratories and mothers so that interventions are completed. Hospital staff can view test results in real time; mothers are sent discreetly worded text messages inviting them to attend the clinic; and delays trigger alerts.

Quality (the primary outcome) refers to completion of all the services and efficiency (the secondary outcome) to the turnaround time of the many time-sensitive early infant diagnosis services.

The hospitals, matched on geographic region, resource level and patient volume (high, medium and low), were randomly allocated to either the HITSystem (intervention) or standard of care (control). Mothers 18 years of age or more with an infant younger than 24 weeks presenting for their first early infant diagnosis were eligible to participate.

There were 392 mother-infant pairs in the intervention sites and 298 mother-infant pairs in the control sites. There were no significant differences in education, income level, maternal age (median 29 years) or disclosure status.

Infants diagnosed as HIV-positive were followed up for a median of 2.1 months and HIV-negative infants for a median of 17 months.

Infants allocated to intervention facilities were significantly more likely to get complete early infant diagnosis services compared with those assigned to the standard of care: 85% (334 of 392) and 60% (180 of 298), respectively; adjusted odds ratio: 3.7, 95% CI: 2.5-5.5, p < 0.0001.

Initial PCR test results were received in a median of 20 days vs 38.5 days for intervention and control sites, respectively. Mothers were notified of their infant’s test results in a median of 14 days vs 23 days.

The turnaround time for ART initiation among infants diagnosed with HIV did not improve at intervention sites but was actually slower, due possibly to more maternal counselling sessions before infant ART initiation. Nonetheless, median infant age at the start of ART was 7.6 weeks younger in intervention sites (17.5 vs 25.1 weeks).

The intervention had a significant effect on quality and efficiency at sites with medium- and low-patient volume, but not at high-volume sites.

The authors conclude that the HITSystem, an innovative public health intervention, is effective and feasible in resource-poor settings. Its algorithms have been modified to reflect Kenya’s new policies to include HIV testing at birth.

Dr Richard T. Lester of the University of British Columbia, in an accompanying comment article, suggests the reason for no intervention effect in high-volume facilities is that smaller volume facilities have more to gain and/or are able to better adapt to new technologies.

A decentralised implementation approach, he adds, for innovations such as the HITSystem provides an opportunity to close the digital divide so improving access and support for those hardest to reach.

Yet the greatest challenge, Dr Lester stresses, is converting these findings into a scalable service. Too often successful pilots are not integrated into health services for decades. He cites the examples of two innovative programmes in Kenya involving SMS to support ART adherence in adults.

Funders appear to be reluctant to invest in the integration of successful pilots, which likely contributes to delays in their implementation, he concludes.


Finocchario-Kessler S et al. Evaluation of the HIV infant tracking system (HITSystem) to optimise quality and efficiency of early infant diagnosis: a cluster-randomised trial in Kenya. The Lancet HIV, online edition,, 2018.

Lester, Richard T. Digital health to support early infant diagnosis of HIV. The Lancet HIV, online edition, http://dx.doi/10.1016/S2352-3018(18)30266-2, 2018.