The medium-term impact of non-pharmaceutical interventions. The case of the 1918 influenza in US cities
利用双重差分法估计1918年流感大流行期间美国43个城市实施的非药物干预（如关闭学校、社交隔离）对后续死亡率的中期影响，发现高强度干预可能因降低群体免疫力而导致后续年份死亡人数显著增加。
Abstract This paper uses a difference-in-differences (DID) framework to estimate the impact of non-pharmaceutical interventions (NPIs) used to fight the 1918 influenza pandemic and control the resultant mortality in 43 US cities. The results suggest that NPIs such as school closures and social distancing, as implemented in 1918, and when applied relatively intensively, might have reduced individual and herd immunity reducing the life expectancy of people with co morbidity, thereby leading to a significantly higher number of deaths in subsequent years. It would be difficult to draw any inference regarding the predicted impact of NPIs as implemented during the Covid-19 crisis as influenza and Covid-19 are two entirely different viruses and nowadays’ pharmaceutical technologies can limit these medium-term impacts.