This blog was written by Veronica Wignall, Assistant Editor on the BMC-Series
Quantifying the benefit of social distancing in a disease epidemic can be difficult, due to the unpredictable nature of individual behavioural responses. However, research at the University of California has demonstrated an innovative solution, using home television viewing as a proxy for behaviour.
Social distancing is a common non-pharmaceutical intervention (NPI) in the event of an epidemic: closing institutions, businesses and events can help to reduce inter-personal contact and disease transmission rates, albeit with an inevitable economic cost.
Modelling the costs and benefits of social distancing is important, since previous epidemiological research has suggested that while a highly cautious approach can suppress an epidemic, social distancing can be less effective than no control strategy if poorly implemented.
Epidemiological models include the interacting components of contact and average transmission rate, but subtleties in individual behaviour are often not taken into account. This limits estimations as such behavioural nuances have their own influence on the spread of disease.
A holistic approach
An ingenious new study led by Asst. Prof Michael Springborn at the UCLA has incorporated personal behavioural responses into existing social distancing models. The study used home television viewing data to provide an empirical analysis of the 2009 A/H1N1 influenza epidemic in Mexico. Although an ‘imperfect’ variable, TV viewing data is consistently and widely available.
Springborn’s inclusion of daily changes in behaviour (via TV viewing) produced a flexible, holistic epidemiological model with improved predictive accuracy. The swine flu outbreak that hit Mexico City in April 2009 could have been worse, but spread of the virus was reduced by people’s behavioral response of distancing themselves from each other.
Couch potatoes: who and when?According to the viewing data, the social distancing response was stronger in wealthier subgroups, and among children, but this result should be interpreted with caution. The data was also used to explore temporal change, showing attenuation in the response over time, which suggested a return to normal behaviour.
Accurate empirical analyses of NPIs are vital for future strategy design to minimize the costs of disease epidemics. A comprehensive understanding of individual behaviour in the broader context of social distancing responses, using flexible variables as in this study, will contribute to the design of schemes aimed to reduce transmission rates and suppress epidemics such as influenza.