Non parametric approaches to understanding disease transmissions dynamics in the context of policy information

In my DPhil thesis "non parametric approaches to understanding disease transmissions dynamics in the context of policy information", supervised by Dr. Oliver Ratmann, I am aiming to develop epidemiological models to better understand the spread of dangerous infective diseases.

In particular, we will focus on HIV in southcentral Uganda, where HIV prevalence ranges up to 40% in some communities. The introduction of the Universal Test and Treatment rollout in the last years lead to improvements of the treatment of the disease, however prevalence has decreased modestly.

Simultaneously, increases in HIV drug resistance have been observed, and the COVID-19 pandemic has caused severe disruption in the healthcare facilities. In this context, the large, population-level, prospective study RCSS is providing data to document transmissions and viremia over time, as well as providing phylogenetic data to analyse. One of our main goals will be to develop Bayesian non-parametric models to compare HIV viremia and drug resistance before and after the COVID-19 epidemic, in order to assess the effectiveness of policies targeted at epidemic control. Further, we aim to quantify the disruption caused by the COVID-19 pandemic on risk of HIV transmission, as well as on healthcare facilities, treatment and prevention.

The study will be significant in understanding whether the UNAIDS goals for the 'end AIDS by 2030' - fewer than 200,000 new infections among adults- are feasible with the current policies.Further, it will provide insight on the evolution of HIV drug resistance. Although the study is limited to Ugandan communities, we expect the results to be applicable in other African countries.

The main project is still in its early stages, as I spent the latest half of my first year on another epidemiological problem.
In particular, I took part to a multi-disciplinary study aiming to understand the observed wide changes of COVID-19 in-hospital fatality rates in Brazil.
We used public available data from various Brazilian institutes to assess which factor was more strongly associated with higher fatality: differences among cities, dates of emergence of the Gamma variant, or increases in hospital resource demand.

The various effect were hard to disentangle, but we used a Bayesian multistrain model to estimate the variant dynamics to, in turn, estimate the proportion of hospitalisations with a given variant. We found that Gamma's consequences on fatality were mainly mediated through resurgence in COVID-19 transmission rather than an increase in disease severity attributable to Gamma. This meant that locations with effective non-pharmaceutical intervention and more equipped healthcare systems did not observe as dramatic consequences as other cities.We hope that the above findings and our conclusion that a significant proportion of hospitalisations could have been saved will soon be published, as we just submitted a manuscript.

In my HIV project, I will use the knowledge and experience gained in the last month to better model and quantify the different factors coming into play.The study of epidemic outbreaks and policies to tackle infectious diseases makes the project fall within the EPSRC Global uncertainties research area.

The primary CDT impact will be training 75 PhD graduates as the next generation of leaders in statistics and statistical machine learning. These graduates will lead in industry, government, health care, and academic research. They will bridge the gap between academia and industry, resulting in significant knowledge transfer to both established and start-up companies. Because this cohort will also learn to mentor other researchers, the CDT will ultimately address a UK-wide skills gap. The students will also be crucial in keeping the UK at the forefront of methodological research in statistics and machine learning.
After graduating, students will act as multipliers, educating others in advanced methodology throughout their career. There are a range of further impacts:
- The CDT has a large number of high calibre external partners in government, health care, industry and science. These partnerships will catalyse immediate knowledge transfer, bringing cutting edge methodology to a large number of areas. Knowledge transfer will also be achieved through internships/placements of our students with users of statistics and machine learning.
- Our Women in Mathematics and Statistics summer programme is aimed at students who could go on to apply for a PhD. This programme will inspire the next generation of statisticians and also provide excellent leadership training for the CDT students.
- The students will develop new methodology and theory in the domains of statistics and statistical machine learning. It will be relevant research, addressing the key questions behind real world problems. The research will be published in the best possible statistics journals and machine learning conferences and will be made available online. To maximize reproducibility and replicability, source code and replication files will be made available as open source software or, when relevant to an industrial collaboration, held as a patent or software copyright.