International efforts to eliminate HIV/AIDS by 2030 are failing high-risk populations, particularly people who inject drugs (PWID), who frequently have high HIV prevalence[1]. Limited data exists on the dynamics of these epidemics; HIV incidence trends among PWID have only been estimated in four lower- and middle-income countries (LMIC)[2], where the HIV burden is concentrated. The dearth of HIV incidence data in LMICs masks the scale of the epidemic and need for action. HIV incidence data is also crucial for evaluating the effectiveness of interventions[3] and for determining progress towards elimination, as advocated by UNAIDS[4].

Prospective cohorts, the gold-standard study design for measuring HIV incidence, are expensive and challenging to sustain[4]. A low-cost yet underutilised alternative is leveraging routinely-collected data from HIV prevention services for PWID[5], which generally undertake HIV testing and collect individual-level behavioural and intervention uptake data. This programmatic data can be used to estimate HIV incidence in real-time, characterise risk factors and estimate the impact of interventions[5]. Despite this potential, only two published studies have utilised such datasets to characterise HIV incidence among PWID in LMICs. There is a need to explore the utility of programmatic data for HIV impact assessment and monitoring HIV epidemics among PWID.

Aims and objectives
This studentship will examine the utility of using programmatic datasets for documenting trends in HIV incidence among PWID and for evaluating the impact of interventions. Specific aims (SA) are:

SA1. Use routinely-collected data to estimate HIV incidence, risk factors for HIV and the effectiveness of interventions among PWID in 2-4 LMICs;

SA2. Use mathematical modelling with other setting-specific epidemiological data to evaluate the utility and biases related to using programmatic data for monitoring epidemics/progress to elimination;

SA3. Work with UNAIDS to develop recommendations on how programmatic data can be used and optimised to inform monitoring of HIV epidemics.

Methodology
Although there is flexibility to accommodate individual interests, the focus of the PhD will be on quantitative methodologies associated with the specific aims: survival analysis (SA1), advanced epidemiological methods and policy-relevant epidemic modelling (SA2), and translating research findings into guidance (SA3). At the start of the PhD, we will complete a training needs assessment to identify skill strengths and needs, and opportunities for development for the student. Our team comprises senior and early-career researchers with expertise in quantitative methodologies including data analysis and modelling[6-8], and so the student will integrate into a rich, stimulating and supportive learning environment. We also have numerous international collaborations and links with international agencies (WHO, UNAIDS) that are crucial for this project. Specifically, the PhD will involve an active collaboration with Medicins du Monde (co-supervisor Ernst Wisse), through which we will link to specific programs, and the UNAIDS modelling reference group (co-supervisor Jeff Eaton) with whom we will work to determine how programmatic datasets can improve global projections of the HIV dynamics among key populations. Overall, the project, collaborations and learning environment will support the student in developing a well-rounded skillset in epidemiology and epidemic modelling applicable to undertaking policy-relevant studies of infectious diseases.

This project will be based in Bristol Medical School - Population Health Sciences.

References

1. Degenhardt L, Peacock A, et al. Lancet Global Health 2017; DOI: 10.1016/S2214-109X(17)30375-3
2. Artenie A, Stone J, et al. in submission
3. MacArthur G, Minozzi S, BMJ 2012; DOI: 10.1136/bmj.e5945
4. Mitchell K, Maheu-Giroux M, et al. Clinical Infectious Diseases 2022 DOI: 10.1093/cid/ciab976
5. Ompad D, Wang J, et al. International Journal of Drug Policy 2017; DOI: 10.1016/j.drugpo.2016.12.008
6. Stone J, Fraser H, et al. Lancet Infectious Diseases 2018; DOI: 10.1016/S1473-3099(18)30469-9
7. Ward Z, Stone J, et al. Lancet HIV 2022; DOI: 10.1016/S2352-3018(21)00274-5
8. Artenie AA, Minoyan N, et al. CMAJ 2019; DOI: 10.1503/cmaj.181506