Sexually transmitted infections are a major global public health problem. In many low and middle income countries the prevalence of common bacterial sexually, including chlamydia and gonorrhoea, can be as high as 20-30% amongst young adults.

Management of sexually transmitted infections includes not only treatment of the individual presenting to the health care facility (the index case), but also identification and treatment of their recent sexual partners (partner notification). In addition, expedited partner therapy can be provided via the index person.

Treatment of partners is perceived to have both benefits for the index case, by reducing the risk of re-infection, and for the broader population, by reducing the risk of transmission within the broader sexual network. Despite these perceived benefits, partner notification and therapy also carries significant risks including exposing the index case to risks of emotional and physical abuse from their partner. Because of these risks, the uptake of partner notification is very low. For example in Zimbabwe we previously demonstrated that only 5% of index cases notified their partner following an STI diagnosis. At such a low rate of uptake it is unclear whether partner notification will have either individual or wider benefit.

Mathematical models of the population dynamics of STIs allows the role of biology and behaviour in spread of STIs to be explored along with a quantification of the expected direct and indirect effects of interventions. Grounding such models in local sexual network and clinical data allows an explanation of observed infection and disease patterns and builds confidence in modeling insights. In this project we will develop a mathematical model of a sexual network of young people in Zimbabwe and calibrate it to data on sexual behaviours and the prevalence of bacterial STIs using existing data collected by our team. The model will describe the sexual network using individual based simulation, with rules on sex partner formation and dissolution that generate networks consistent with observed partnership and STI patterns in Zimbabwe. Such a model will allow a description of how partner notification and therapy propagates treatment through the sexual partner network and the effects of the index patients social, cultural, behavioural, and clinical characteristics. We will use the model to simulate varying rates of the uptake of partner notification and assess thresholds at which benefits are seen both for the index case and more broadly. The modeling will explore the role of social and cultural context, the impact of diagnostic tools, patient care pathways, and choice of therapeutics on the epidemiology of bacterial sexually transmitted infections, and in an environment with limited resources will explore the optimum intervention mix to improve STI outcomes.