Impact of small-area lockdowns for the control of the COVID-19 pandemic
Cuadrado, C.; Monsalves, M. J.; Gajardo, J.; Bertoglia, M. P.; Najera, M.; Alfaro, T.; Canals, M.; Kaufmann, J.; Pena, S.
Show abstract
BackgroundCountries confronting the COVID-19 pandemic are implementing different social distancing strategies. We evaluated the impact of small-area lockdowns in Chile, aimed to reduce viral transmission while minimizing the population disrupted. The effectiveness of this intervention on the outbreak control is unknown. MethodsA natural experiment assessing the impact of small-area lockdowns between February 15th and April 25th, 2020. We used mobility data and official governmental reports to compare regions with small-area lockdowns versus regions without. The primary outcome was the mean difference in the effective reproductive number (Re) of COVID-19. Secondary outcomes were changes in mobility indicators. We used quasi-experimental methods for the analysis and examined the impact of other concurrent public health interventions to disentangle their effects. ResultsSmall-area lockdown produced a sizable reduction in human mobility, equivalent to an 11.4% reduction (95%CI -14.4% to -8.38%) in public transport and similar effects in other mobility indicators. Ten days after implementation, the small-area lockdown produced a reduction of the effective reproductive number (Re) of 0.86 (95%CI -1.70 to -0.02). School and university closures, implemented earlier, led to a 40% reduction in urban mobility. Closure of educational institutions resulted in an even greater Re reduction compared with small-area lockdowns. ConclusionsSmall-area lockdowns produced a reduction in mobility and viral transmission, but the effects were smaller than the early closures of schools and universities. Small-area lockdowns may have a relevant supporting role in reducing SARS-CoV-2 transmission and could be useful for countries considering scaling-down stricter social distancing interventions.
Matching journals
The top 9 journals account for 50% of the predicted probability mass.