The Social Contouring project seeks to establish social contours during the current pandemic. The idea is to return to a normal life as much as possible while keeping the virus in check. So we developed a methodology to find social interactions we can safely undertake (and propose to call it ) social contours.
A social contour is people and places, a set of activities or social interactions you can undertake without the risk of causing a new outbreak.
There are many empty social places that could host safe economic activity and social interactions during the current work from home, shelter in place lock down. Social contours include safe people and safe places, social contours will reallow for reintegration of safe thriving communities and reignite economies.
Approach To tackle this we need a multi-level and hybrid approach in contrast to digital or biomedical responses currently being developed.
The three levels are: Macro level: social lifestyles Meso level: historical granular data about social interactions Micro level: factors putting people at risk according to the virus characteristics
The approach is anthropological, virology, and digital. The goal is maximizing social interactions within the virological boundaries. This would require fine-grain data while protecting privacy and providing a proportional public and political response.
This approach is NOT about mapping where the virus HAS BEEN, those approaches may increase our understanding of the virus dynamics which is very important but insufficient because they are not able to keep the virus in check.
Our approach maps how the virus CAN MOVE through a SPECIFIC SOCIETY by studying the structures of that society. That knowledge makes it possible to identify what MINIMAL PROTECTIVE measures are needed to keep the spread of the virus under control (R0 < 1) while disturbing that specific society as little as possible.
Both the community based approach study on Ebola (Wong et al., 2016) / Alexander et al.,2015) and the China vs Italy study (Wilder et al., 2020) show that R0 can be reduced to less than 1 with incomplete but well-placed measures. The main goal, formulated in such a way that it doesn't bash explicitly many other groups is to find out: "how to reduce R0 to less then 1 in a specific society".
Hackathon results During this hackathon, the methodology we started with was a survey to have a better understanding of social interactions in the current situation. Sociological questionnaires can establish lifestyles that allow for deep analysis of social structures to unlock social interventions and a deep epidemiological understanding of COVID19. Social factors are then analyzed for 1) disease risk 2) potential intervention 3) and the development of safe social contours.
We conducted a preliminary qualitative analysis of fictional personas that are considered at risk or socially excluded. We identified social factors according to the way they live and and that put them at risk according to the virus characteristics. This allowed us to see the value of a hybrid approach that has to be confirmed or updated with empirical data.
We created a questionnaire in five languages: French, English, Spanish, German and Italian. It contains 37 variables with different types of information related to a lifestyle: individual's views of government measures, circumstances of socialization, resources available, working conditions, alimentation, communication means, health state and demographics. Answers are anonymized and no detailed personal information was asked besides sex, age, city and country to protect the respondents identity. We have also created a logo and met new people that joined this initiative. On the first day we collaborated with team 8 developing the app co19andme. Team work was organized with trello.
We have collected nearly 1,000 answers in one week before starting the versus virus hackathon distributed in Switzerland via personal contacts. During the hackathon we have conducted a data analysis with a sample from the french version.
The analysis of 771 questionnaire responses revealed 5 clusters using Louvain community detection algorithm. Each cluster has specific characteristics that describe a lifestyle. We present them below and give examples of interventions in context that could be made.
Young student lifestyle: average age is 26, mainly students, shared room with roommate, do not own a vehicle and work or study alone in a space. Some interventions in context that can be made relate to public transportation and proximal living.
Working couple lifestyle: average age 39, shared room with partner, shared vehicle, informed via social media, shared room with a roommate. Some interventions in context that can be made relate to cross-proximal living, communication awareness.
Independent mature lifestyle: average age 41, goes outside, some are retired, does not share a room, has a vehicle, works from home. Some interventions in context that can be made relate to physical distance and medical check-ups. 30% more likely to be a man.
Female unemployed “hospitality” lifestyle: 30% more likely to be a woman, average age is 33. She does not share a room, does not go outside, owns a vehicle, cannot work anymore. Some interventions in context that can be made relate to economic subsidies, re-employment and continuing education.
Healthcare worker lifestyle: 15% more likely to be a woman, average age is 28. She doesn’t have a roommate, sleeps in a dormitory, recently visited the hospital, working in a public space. Some interventions in context that can be made relate to personal protective equipment, proximal living and physical distance.
These insights will allow us to further refine the questionnaires to adjust for application to new locations and deeper refinement of select social structures We would like to build an interactive visualization for the lifestyles identified and display the variables that characterize them. The questionnaire is translated in 5 languages, and additional data is available for analysis.
Going further We have found some limitations in our analysis and will further refine the questionnaires to better identify social contours to keep society safe during reintegration. Some limitations are small sample size, deeper clarification of social structures and risk assessment with experts.
Next steps are to partner with web application and phone application development companies to uptake questionnaires, build personas, and track outputs to inform the public, the municipality (city, county) and ministry of health and the government. This methodology could also be used as a tool in a science-policy interface to foster sound policymaking by connecting science with policy makers in a quick and effective way.
Government funding and public health awareness should result in massive population uptake to contour the societies in which we live, while finding specific solutions for vulnerable people without stigmatizing them.
The application will provide a public view detailing safe ways to reintegrate into society at a safe social distance. The web application and phone application become the digital version of the social vaccine (physical, distance, interventions necessary for planned reintegration) to break the transmission of SARSCoV2 and COVID19 in the absence of medical treatment (biomedical response).
This study represents an exploratory study grounded in society. To go further, an ethnography will provide a deepened understanding of communities, from which you can build proper e-platforms and successful applications to benefit public health, especially critical during pandemics.