Division of Epidemiology, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada.
Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada.
PeerJ. 2024 May 31;12:e17455. doi: 10.7717/peerj.17455. eCollection 2024.
The rapid global emergence of the COVID-19 pandemic in early 2020 created urgent demand for leading indicators to track the spread of the virus and assess the consequences of public health measures designed to limit transmission. Public transit mobility, which has been shown to be responsive to previous societal disruptions such as disease outbreaks and terrorist attacks, emerged as an early candidate.
We conducted a longitudinal ecological study of the association between public transit mobility reductions and COVID-19 transmission using publicly available data from a public transit app in 40 global cities from March 16 to April 12, 2020. Multilevel linear regression models were used to estimate the association between COVID-19 transmission and the value of the mobility index 2 weeks prior using two different outcome measures: weekly case ratio and effective reproduction number.
Over the course of March 2020, median public transit mobility, measured by the volume of trips planned in the app, dropped from 100% (first quartile (Q)-third quartile (Q) = 94-108%) of typical usage to 10% (Q-Q = 6-15%). Mobility was strongly associated with COVID-19 transmission 2 weeks later: a 10% decline in mobility was associated with a 12.3% decrease in the weekly case ratio (exp() = 0.877; 95% confidence interval (CI): [0.859-0.896]) and a decrease in the effective reproduction number ( = -0.058; 95% CI: [-0.068 to -0.048]). The mobility-only models explained nearly 60% of variance in the data for both outcomes. The adjustment for epidemic timing attenuated the associations between mobility and subsequent COVID-19 transmission but only slightly increased the variance explained by the models.
Our analysis demonstrated the value of public transit mobility as a leading indicator of COVID-19 transmission during the first wave of the pandemic in 40 global cities, at a time when few such indicators were available. Factors such as persistently depressed demand for public transit since the onset of the pandemic limit the ongoing utility of a mobility index based on public transit usage. This study illustrates an innovative use of "big data" from industry to inform the response to a global pandemic, providing support for future collaborations aimed at important public health challenges.
2020 年初 COVID-19 疫情在全球迅速蔓延，迫切需要寻找可追踪病毒传播情况的领先指标，并评估旨在限制传播的公共卫生措施的后果。公共交通出行作为一种早期的候选指标，其活跃度已被证明可以对疾病爆发和恐怖袭击等先前的社会动荡做出响应。
我们对 2020 年 3 月 16 日至 4 月 12 日期间，40 个全球城市的公共交通出行应用程序中公开提供的数据进行了一项关于公共交通出行减少与 COVID-19 传播之间关联的纵向生态研究。使用两种不同的结果衡量标准：每周病例比和有效繁殖数，采用多水平线性回归模型来估计 COVID-19 传播与 2 周前的出行指数值之间的关联。
在 2020 年 3 月期间，通过应用程序中计划出行的次数来衡量的公共交通出行中位数从典型使用量的 100%（第一四分位数（Q）-第三四分位数（Q）= 94-108%）下降至 10%（Q-Q = 6-15%）。出行与 2 周后的 COVID-19 传播密切相关：出行减少 10%，与每周病例比下降 12.3%（Exp()=0.877；95%置信区间（CI）：[0.859-0.896]）和有效繁殖数下降（= -0.058；95%CI：[-0.068 至-0.048]）相关。仅出行模型对两种结果的解释率接近 60%。对流行时间的调整减弱了出行与随后 COVID-19 传播之间的关联，但仅略微增加了模型的解释率。
我们的分析表明，在 40 个全球城市 COVID-19 大流行的第一波期间，公共交通出行作为 COVID-19 传播的领先指标具有重要价值，而此时几乎没有此类指标可用。自大流行开始以来，公共交通需求持续低迷等因素限制了基于公共交通使用情况的出行指数的持续应用。本研究说明了利用行业“大数据”为全球大流行做出响应的创新方法，为旨在应对重要公共卫生挑战的未来合作提供了支持。
