Projects

HIV incidence and health economic modelling, for the MaxART 2017-2018

2017-2018

We will use Simpact to build a custom-made HIV transmission model, which we will them calibrate using data collected during the MaxART Early Access to ART study. Next, will use this model to reconstruct the Early Access to ART Implementation Study and contrast what was actually observed during the intervention phases against a hypothetical 'counterfactual' of what we believe would have happened during the same time periods, if the study sites had remained in the control phase (ART access only for those with CD4 counts < ??) for the entire duration of the study.
The model will reconstruct the Implementation Study not only in terms of ART uptake, adherence and viral suppression, but also in terms of age-mixing patterns and sexual risk behaviours. In addition it will evaluate the impact of Early Access to ART for All on HIV incidence and the projected long-term cost-effectiveness of the intervention.

Modelling HIV transmission dynamic networks and molecular evolution

2015-2018

Understanding the structure and the dynamics of transmission networks is a key step towards the design and implementation of effective prevention measures. Simpact provides a platform to simulate more detailed dynamic sexual and transmission networks, which are calibrated by sexual behaviour data and clinical data such as CD4 count and viral load.

Meanwhile, the field of phylodynamics is emerging with much potential to enhance our understanding of HIV transmission dynamics by using sequence data. Phylodynamics methods allow estimation of epidemiological features such as time trends of HIV prevalence and HIV incidence, reproductive number, among others.

The aim of the project is to develop a unified simulation framework that bridges these two schools of thought in order to be able to use multiple data sources. The unified framework will enable improvement in epidemiological estimates by calibration with multiple data sources (bio-behavioural and sequence data). In addition, this framework will help to assess accuracy of new phylogenetic inference methods in the study of HIV transmission dynamics.

Investigating the Role of Age-Mixing Patterns in the Sustained Spread of HIV

2014-2018

Young women in sub-Saharan Africa are at a disproportionately high risk of HIV infection. Biological and behavioural risk factors, in combination with a complex sexual age-mixing pattern, have been proposed to explain this gender discrepancy. Age-mixing patterns characterised by the frequent occurrence of large age differences between sexual partners are thought to be the result of socio-economic inequalities in society.
Informed by behavioural, clinical and socio-demographic data from South Africa and Malawi, we plan to use Simpact to investigate the role of age-mixing patterns in the sustained spread of HIV.

Directly linked to this question is that of HIV prevention: We will use Simpact to explore how changes in the age-mixing pattern affect the risk of acquiring HIV at the individual level, and alter the course of the epidemic, taking into account the biology, sociology and behavioural science behind the epidemiology of HIV in young women in sub-Saharan Africa. In 2012, the results of a cluster randomised trial in Malawi suggested that structural interventions can successfully modify the age-mixing pattern in young girls of school-going age, and that this may result in fewer new HIV infections among adolescent and young adult women. What remains unclear, however, is what the short- and long-term impact of such interventions might be on HIV incidence and prevalence in women and men of all ages.

Informing the 2013 WHO consolidated HIV treatment guidelines

2013

The HPTN 052 study showed that early antiretroviral treatment (ART) could largely reduce the heterosexual HIV transmission while a recent simulation study indicated that universal immediate ART might be able to eliminate the epidemic in the near future. However, universal and immediate ART remains unfeasible in most settings due to limited resources. Therefore alternative options of providing ART have been proposed and discussed.We used Simpact to simulate HIV transmission and the effects of HIV Treatment as Prevention (TasP) interventions in sexual networks. We considered prioritising five HIV-infected population subgroups (CD4<500, long-term serodiscordant couples, pregnant women, female sex workers, the elders, all positives) under three schemes. The expected HIV prevalence over 15 years was projected for South Africa. Results show that HIV prevalence in South Africa can be suppressed to 10% by 2023 if the current treatment access level is maintained. Expanding ART to the mentioned subgroups might not lead to a dramatic drop in HIV prevalence in the short run. Differences between the impact estimates from alternative interventions emerge after 10 years. In the scenario where 80% of HIV positives were treated immediately after diagnosis, the prevalence decreased to 5% within twenty years. Prioritisation for those with CD4 counts below 500 and serodiscordant couples might be the most effective options, spending less than 3/4 than non-prioritising strategies for per averted infection. Furthermore, the demographic structure of the HIV-positive population would be changed by shifting resources to a particular subgroup.

Our simulations suggest that the impact of TasP interventions is complex and that long-term outcomes need to be considered according to specific objectives. Expanding ART to a larger proportion of the subgroups with higher CD4 levels or in serodiscordant relationships may have the most favourable incremental cost-effectiveness ratios.

Simulation-based Method for Efficient Resource Allocation of Combination HIV Prevention

2012-2013

Over the past three decades there has been a wealth of operational research into effectively and efficiently combating human immunodeficiency virus (HIV). These interventions have had varying results. Condoms, for example, have been shown to decrease the probability of transmission per sexual act (PTSA) by 95%, but they tend to be used inconsistently. Male circumcision has been shown to reduce the PTSA by 50%, but provides consistent partial protection by design. Antiretroviral therapy (ART) is a medical treatment that slows the reproduction of HIV. ART has been associated with 96% reduction in PTSA, and has been shown to prolong the life of an infected individual.

However, it is difficult to determine how to optimally distribute limited HIV prevention resources to prevention methods due to each method's different financial costs, levels of uptake and efficiency, and potential unintuitive interactions. In this paper we implement an individual-based model that simulates HIV transmission and prevention in a complex sexual network and use it to answer the question of combination prevention. Using optimization software, we find the best combination of prevention methods for a given budget and sexual network structure.

The paper that resulted from this project can be found HERE (pdf, 300 KB).