Infectious Disease Analytics
Prepare and respond to epidemics and pandemics
Population growth and globalisation have accelerated the spread of infectious diseases. We have developed epidemic simulation models that predict how control measures such as frequent testing, social distancing, mask wearing and vaccination reduce the risk of workplace transmission and subsequent severe disease outcomes. Not only do such predictions enable evidence-based decisions that protect the health of employees, they also ensure business continuity.
We may have the worst of the Covid-19 pandemic behind us, but history suggests that pandemics are likely to occur at an increasing rate, due to high population density and mobility. Covid-19 could pose less danger over time but will still need to be managed as a cyclical epidemic within the work and public context. In addition, several other infectious diseases continue to threaten public health, including tuberculosis (TB), Influenza, Malaria, HIV, Ebola and multidrug-resistant bacteria.
We integrate into your surveillance and response strategy by providing information and advice based on our models and medical expertise. We curate and sythesise the latest data on infectious disease outbreaks: their infectivity and pathogenicity (health impact), as well as the effectiveness of preventative and curative measures. We analyse and visualise the data from your organisation’s monitoring and mitigation programme, and we deploy our epidemic models to assess the effectiveness of critical controls.
provide alerts and analyse data on emerging pathogens
dashboards for tracking of controls and crisis management
scenario modelling for epidemics and pandemics
and mortality of infectious diseases
expert advice on management of infectious diseases
A case study
Can we predict the size of COVID-19 wAVES?
From the beginning of the COVID-19 pandemic, epidemiological models have been used in a number of ways to aid governments and organisations in efficient planning of resources and decision making. These models have shed light on important epidemiological transmission considerations, in addition to making short-term projections.
We were requested by a multinational mining company with operations in multiple countries around the world to develop epidemiological models to guide operational and strategic planning for business continuity and COVID-19 responses. We constructed separate models for countries where the business had most of their mining operations, and additionally built finer-grain tools at a province/state level for more targeted projections. The algorithms were calibrated to publicly available data on laboratory-confirmed SARS-CoV-2 infections and deaths in these regions, resulting in four-week and six-month projections.
Comparing short-term projections to actual confirmed cases (retrospectively), the model performed very well. We obtained an overall forecast error of below 8% on average for four-week-ahead projections in all the countries and regions. As the epidemic evolved and more data became available, the model’s parameter estimates became more accurate. Building on the lessons learnt, we moved on to a second generation model that allows for explicit modelling of close contact patterns, vaccination and booster campaigns, and other mitigation strategies. For example, we used this new agent-based model to predict the impact of alternative testing strategies among office staff and the residual risk of allowing for “testing holidays” after an immunity boost (which may be vaccine-induced or the result of natural infection).
Thanks to this data-driven, model-guided approach, the company leadership was able to make informed decisions around vaccination policies, mask mandates, testing schedules and office occupancy rates, such that operational efficiency was maximised without compromising the health risk to employees. We also created dashboards on the company’s intranet that visualised real-time indicators of Covid-19 risk, enabling employees to make rational decisions about working from home versus working from the office.