Since the start of the COVID-19 pandemic, the need for informed decision-making, marked by the use of massive amounts of data and increasing research on artificial intelligence, has inspired cooperation between academia and decision-makers. COLEV is a study led by an interdisciplinary research team group at Universidad de Los Andes in Colombia, which seeks to produce and communicate pertinent evidence that informs public health decision-making related to COVID-19 through the use of data science and artificial intelligence.

COLEV gathers approximately 20 experts from areas of public health, epidemiology, service logistics, artificial intelligence, data science, infectious diseases, qualitative research, data visualization, and digital storytelling. In particular, the team is working together to understand three cross-cutting topics: the effects of opening schools at a population level, COVID-19 vaccination, and mental health consequences of COVID-19 with an intersectional gender approach. 

So how have we gone about it? The team analyzes these cross-cutting topics in five lines of action:  

1. Decisions and data, identifies the priorities and needs regarding the use of artificial intelligence and data science among decision-makers.
2. Mathematical models, develops long-term and real-time mathematical models allowing to predict the behavior of the pandemic at a city level. 
3. Health management, evaluates the impact of COVID-19 on the provision of other health services and its effect on resource management policies (e.g., human, monetary, physical) in health. 
4. Social media analysis, characterizes public discourse in digital media by analyzing popular social networks such as Twitter and Facebook with artificial intelligence methods.
5. Disruptive strategy against misinformation, designs an innovative communication strategy adapted to different audiences to deal with misinformation about COVID-19. 

Discussions with decision-makers have progressed, with advocating for the reopening of schools in Colombia based on scientific evidence being one of COLEV’s key contributions. Currently, the team is working on strategies to communicate the risks of contracting COVID-19, given the results of mathematical models parameterized for the principal cities in the country. 

With regard to COVID-19 vaccination, COLEV has investigated the effects of spacing mRNA vaccine against COVID-19 from 20 days to 80 days, and the team has developed a follow-up tool based on vaccine stock and simulation models that allow vaccination centers to optimize their operations and increase COVID-19 immunization rates throughout the country. Relatedly, COLEV has been monitoring conversations regarding vaccination online to identify misinformation shared via Facebook and Twitter, whilst studying the effect of misinformation on vaccination roll-out strategies. The results will be used in the design of a disruptive communication strategy to tackle this problem. 

Now wrapping up the first year since its inception, the COLEV study has established a close relationship with a range of decision-makers and continues to contribute to discussions on mitigation strategies against COVID-19 whilst bringing scientific evidence to the table. COLEV is made possible thanks to Canada’s International Development Research Centre (IDRC) and Swedish International Development Cooperation (SIDA) efforts. It is also part of  the Global South AI4COVID Program, which is focused on leveraging AI and data science to improve responses to COVID and future pandemics in Low and Middle-Income countries. 

To learn more about our work in Colombia, visit our project page at:  https://colev.uniandes.edu.co/ and follow our work on Twitter @colev_uniandes.