About

Summary

Logo_500x677GET ON („GETting tOwards Ninety“) is a research project that aims to assess how to reach the UNAIDS 90-90-90 targets in Lesotho. The GET ON project entails two interlinked cluster-randomized trials, HOSENG trial and VIBRA trial. The HOSENG trial evaluates the effect on HIV testing coverage with secondary distribution of oral HIV self-tests during a home-based HIV testing campaig. In control clusters, an oral HIV self-test is left behind for individuals who are absent and eligible for testing and for those who refuse testing on spot. One household member is tested and trained on the oral HIV self-test and the household can decide to return the test to the health center or the village health worker (VHW).

Individuals found HIV-positive and not taking antiretroviral therapy (ART) during the HIV testing campaign are offered participation in the VIBRA trial. The VIBRA trial assesses a new decentralized ART delivery model (“VIBRA model”) that builds on the VHW program and uses SMS technology. In control clusters, participants are offered same-day ART initiation with follow-up at the clinic. In intervention clusters, participants are offered same-day ART with the possibility of further follow-up by the nearby trained VHW. Moreover, they may receive automatically generated coded SMS with adherence reminders or viral load results. The primary endpoint of VIBRA trial is viral suppression 12 months after enrolment.

The Interventions and the campaign algorithm:

TABLE 1. Description of HOSENG trial groups

 

figure-4-description-of-procedure-in-hoseng-intervention-and-control-clusters.jpg

TABLE 2. Description of VIBRA trial groups

figure-1-description-of-procedure-in-vibra-intervention-and-control-clusters.jpg

 

FIGURE 1. HOSENG: Algorithm of the HIV and multidisease screening/prevention campaign

figure-3-algorithm-of-hiv-and-multi-disease-screeningprevention-campaign.jpg

Where will GET ON be conducted?

The study project will be conducted in the districts of Butha-Buthe and Mokhotlong, in northern Lesotho, Southern Africa, in the catchment areas of 22 health facilities:

  • Butha-Buthe: 10 nurse-led rural health centers, 1 missionary hospital, 1 governmental hospital
  • Mokhotlong: 9 nurse-led rural health centers, 1 governmental hospital.

Both districts are characterized by mostly rural settings with an estimated population of 250,000, mainly subsistence farmers and mine workers as well as construction or domestic labourers who work in neighbouring South Africa. Each district has one single mid-size town: Buthe-Buthe with ca. 25,000 inhabitants, and Mokhotlong with ca. 10,000 inhabitants. The remaining population lives in villages scattered over a very mountainous area of 5,842 km2. According to the Demographic Health Survey of 2014, the adult HIV prevalence amounts to 21.2% in Butha-Buthe and 17% in Mokhotlong.

 

Who will participate in GET ON?

GET ON is a cluster-randomized project. A list of all villages (with its corresponding VHWs) in the study districts was provided by the District Health Management Teams (DHMTs). Two local members of the SolidarMed research team cross-checked the list for its correctness and added the stratification factors by contacting all VHW coordinators of all involved health facilities. Each village was considered a cluster with the exception of smaller villages that do not have their own VHW. In that case, the neighbouring villages serviced by the same VHW form one cluster. It is not feasible to visit all villages in the two districts, thus, out of the list of all clusters, a random sampling, stratified by the pre-specified stratification variables, was performed, leaving a total of 180 clusters. Thereafter, the cluster eligibility criteria was checked.

The remaining clusters then were randomized (in block sizes of 4, in a 1:1:1:1 allocation) into the 4 groups using a computer-generated randomization list, stratified by district (Butha-Buthe vs Mokhotlong), size of village (<30 vs ≥30 households), and village access to the nearest health facility (easy to reach vs hard to reach [i.e. mountain or river to cross or/and more than 10km away from health facility]). The randomization list was prepared by a statistician not involved in the study. The figure below illustrates the cluster sampling and randomization process.

FIGURE 2. GET ON: Cluster sampling and randomization

 

figure-1-cluster-sampling-and-randomization1.jpg

We assume, based on data from previous HIV testing campaigns, that a total of approximately 100 clusters are required to reach the target sample size of VIBRA trial (N=320 HIV-positive individuals not on ART).

Every household member living in one of the study clusters will be eligible for HOSENG trial and – if HIV-positive and not on ART – screened for participation in VIBRA trial.

Why is GET ON important?

By launching the 90-90-90 strategy UNAIDS has shown a way forward in controlling and finally eradicating the AIDS epidemic. However, in Lesotho, gaps along the entire HIV care cascade are still persisting.

GET ON investigates the following questions:

1) How to reach (in a cost-effective way) 90% testing coverage during home-based, door-to-door HIV testing campaigns?

2) How to link (in a cost-effective way) 90% of the people living with HIV and not on ART after home-based HIV testing to care & ART?

3) …and how to sustainably engage these 90% on ART?

GET ON builds upon our predecessor trial (CASCADE trial), which has shown that same-day ART initiation significantly improves the second and third UNAIDS target, but is still not reaching them.

The GET ON research project addresses all 90-90-90 targets with the last 90 (viral suppression) as primary outcome. We are convinced that only a multicomponent package of differentiated care interventions can be successful, that strengthens task-shifting and decentralization of HIV care.

Findings from GET ON research project will help inform scale up of future HIV programs in the era of treat-all and help identify cost-effective interventions in a limited resource setting.

More information:

Advertisements