Abstract and Introduction
Nonpharmaceutical interventions, such as social distancing and lockdowns, have been essential to control of the coronavirus disease 2019 (COVID-19) pandemic. In particular, localized lockdowns in small geographic areas have become an important policy intervention for preventing viral spread in cases of resurgence. These localized lockdowns can result in lower social and economic costs compared with larger-scale suppression strategies. Using an integrated data set from Chile (March 3–June 15, 2020) and a novel synthetic control approach, we estimated the effect of localized lockdowns, disentangling its direct and indirect causal effects on transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Our results showed that the effects of localized lockdowns are strongly modulated by their duration and are influenced by indirect effects from neighboring geographic areas. Our estimates suggest that extending localized lockdowns can slow down SARS-CoV-2 transmission; however, localized lockdowns on their own are insufficient to control pandemic growth in the presence of indirect effects from contiguous neighboring areas that do not have lockdowns. These results provide critical empirical evidence about the effectiveness of localized lockdowns in interconnected geographic areas.
Since the beginning of the coronavirus disease 2019 (COVID-19) pandemic, nonpharmaceutical interventions have been essential to control and prevent the transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).[1–3] Nonpharmaceutical interventions range from simple individual-level recommendations, such as wearing face masks, washing hands frequently, or maintaining physical distance, to society-level regulatory actions, such as school closures, quarantines, or lockdowns.[1,3] Efforts to control pandemic growth based on these interventions have been successful in some countries.[4–7] The effects of nonpharmaceutical interventions have been described primarily using compartmental models,[1,3,6,8,9] with the results having informed policies around the world since the start of the pandemic.[10,11] In this paper, we adopt a complementary approach from the causal inference literature.[12,13] Arguably, health-policy impact evaluations require a variety of study designs, data sets, and analytical approaches that support each other to provide stronger evidence. In general terms, this approach seeks to estimate the causal effect of localized lockdowns by approximating the hypothetical randomized experiment (trial) that would have been conducted under ideal circumstances to evaluate the policy in question.[15–18]
As the COVID-19 pandemic develops across countries, policy-makers need evidence to help them decide when and how to ease mobility restrictions or strengthen these restrictions in cases of resurgence. Even now that countries have started vaccinating their populations, large-scale nonpharmaceutical interventions continue to be important, particularly in low- and middle-income countries, to avoid large increases in the number of cases. In this context, localized lockdowns have become an increasingly relevant policy option.[20–25]
Localized lockdowns are typically implemented in transmission hot spots and can be applied to populations or areas large and small to suppress an outbreak. Localized lockdowns had not been widely used as a public health response to contain outbreaks until the current pandemic.[20–25] In principle, localized lockdowns impose fewer social and economic costs than larger-scale SARS-CoV-2 suppression strategies and are thus more sustainable. They can also provide a gradual exit from nationwide lockdowns. Early in the pandemic, for example, the Chinese government imposed a localized lockdown and other strict nonpharmaceutical interventions in the city of Wuhan, effectively suppressing SARS-CoV-2 transmission. Subsequently, governments have implemented localized lockdowns in neighborhoods (e.g., Beijing, China), suburbs (e.g., Melbourne, Australia), towns (e.g., Vo, Italy), and districts (e.g., North Rhine-Westphalia, Germany), and at the city level in Leicester, United Kingdom.[4,21] Despite the increasing importance of localized lockdowns, there is limited empirical evidence of their effectiveness.
In this study, we used data from Chile to estimate the effect of localized lockdowns on COVID-19 transmission. Our data set combined information from administrative COVID-19 surveillance records, a nationally representative household survey, and census data. We used a synthetic control approach[30,31] to build control intervention units (municipalities) with similar sociodemographic features, trajectories of contagion, and histories of lockdowns until the time of the policy intervention, taking into account the spatiotemporal structure of the data. In other words, we assessed whether the effectiveness of a localized lockdown implemented at the municipality level was affected by the lockdown status of neighboring municipalities. This indirect effect may play an important role in municipalities within cities or urban areas where social and economic interdependencies exist. Allowing for such indirect effects or interference between municipalities,[32,33] we estimated the direct effects of extending the duration of localized lockdowns and the total (sum of direct and indirect) effects of maintaining lockdowns in neighboring municipalities.
Am J Epidemiol. 2022;191(5):812-824. © 2022 Oxford University Press