Organization: Health Economics and Epidemiology Research Office (HE2RO) and Audere
Study objectives: This study conducted by Indlela@HE2RO in partnership with Audere, and funded by the South African Medical Research Council (SAMRC), tested whether Your Path, an AI-powered tool could support health decision-making after participants conducted a simulated HIV self-test (HIVST).
The specific objectives of the study were:
- Primary objective: Assess community members’ (CMs) intention to engage in HIV care after completing a simulated HIVST and interacting with Your Path AI-powered tool.
- Secondary objectives: Explore CMs’ acceptability and perceptions of Your Path features following a simulated HIVST, identify behavioural barriers and motivators influencing engagement with Your Path, and explore healthcare providers’ (HCPs) perceptions of the clinical appropriateness of Your Path with CMs.
Intervention description: CMs were randomly assigned a mock HIV-negative or HIV-positive test result. They completed a pre-test survey on intentions to engage in HIV care, followed by a simulated HIVST session guided by Your Path. Afterwards, CMs completed a post-test survey which assessed changes in intention to seek HIV care and the usability of Your Path using the System Usability Scale (SUS). Intention to seek care was used as a proxy for uptake of appropriate HIV service, since actual uptake was not measured in the study. HCPs reviewed CM-AI chat interactions and summaries to assess clinical appropriateness of responses and whether generated summaries were complete and relevant to r support care interactions.
Evaluation description: We used an exploratory mixed-methods design, combining quantitative and qualitative approaches to assess the study outcomes.
Study outcomes:
Primary outcome:
- CMs’ intention to access HIV care after a simulated HIV self-test
Secondary outcomes:
- CMs’ insights on the usability, acceptability, and feasibility of Your Path for HIVST.
- HCPs’ perspectives on the clinical appropriateness and usefulness of Your Path for post HIVST counselling.
Study site: The study was conducted in Johannesburg, South Africa.
Target population: We recruited community members and healthcare providers from the Indlela’s B-Hub who met the following inclusion criteria:
Inclusion criteria for community members and care recipients
- ≥18 years old
- Willing to provide written informed consent
- Comfortable consenting and speaking in English
Exclusion criteria
- Participants who are not willing or available to travel to the study site
- Self-reported HIV positive status
Inclusion criteria for the healthcare providers
- ≥18 years old
- Willing to provide written informed consent
- Comfortable consenting and speaking in English
- Had experience in provision of HIV services
Exclusion criteria:
- Participants who were not willing or available to take part in this sub-study
Sample size & analysis: We enrolled 100 CMs and 25 HCPs. Descriptive statistics including (percentages, means, medians and standard deviations) were generated in Stata to summarise socio-demographic characteristics and survey responses. Qualitative audio recordings were transcribed and analysed thematically in NVivo using deductive and inductive approaches.
Results: 59.0% of CMs enrolled were female and 51.0% received a mock HIV-negative result. After interacting with Your Path, 91.0% of CMs reported that it positively influenced their intention to access HIV services. Your Path demonstrated high usability, with a mean SUS score of 81.6 (SD 17.5). Most participants (83.7%) found Your Path easy to use, and 94.9% expressed willingness to use it again. However, 15.3% noted that it did not always provide consistent responses, with some variability or lack of clarity in certain answers. In addition, 33.6% preferred receiving initial assistance before independently using the tool.
Qualitative findings indicated that CMs described Your Path as a helpful guide with a user-friendly design. Many appreciated its private and non-judgmental tone, which reduced feelings of stigma. At the same time, participants expressed mixed views on privacy and security: while some valued the sense of confidentiality, others raised concerns about data protection. HCPs found the conversational summaries generated by Your Path to be clinically appropriate, informative, and useful for facilitating linkage to HIV care. They also highlighted limitations, including reduced emotional responsiveness compared to human counselling and potential barriers posed by the English-only interface. Recommendations included integrating local languages, enabling data-free access, and aligning the tool with existing healthcare systems to improve accessibility and coordination of care.
Duration: 12 months
