General of Police/Vigilance and Anti-corruption, Government of Tamil Nadu, Tamil Nadu, India.
Department of Management Studies, Indian Institute of Technology, Madras, India.
PLoS One. 2022 May 20;17(5):e0268190. doi: 10.1371/journal.pone.0268190. eCollection 2022.
This study investigates the important role of attendant factors, such as road traffic victims' access to trauma centres, the robustness of health infrastructure, and the responsiveness of police and emergency services in the incidence of Road Traffic Injuries (RTI) during the pandemic-induced COVID-19 lockdowns. The differential effects of the first and second waves of the pandemic concerning perceived health risk and legal restrictions provide us with a natural experiment that helps us differentiate between the impact of attendant factors and the standard relationship between mobility and Road Traffic Injuries. The authors use the auto-regressive recurrent neural network method on two population levels-Tamil Nadu (TN), a predominantly rural state, and Chennai, the most significant metropolitan city of the state, to draw causal inference through counterfactual predictions on daily counts of road traffic deaths and Road Traffic Injuries. During the first wave of the pandemic, which was less severe than the second wave, the traffic flow was correlated to Road Traffic Death/Road Traffic Injury. In the second wave's partial and post lockdown phases, an unprecedented fall of over 70% in Road Traffic Injury-Grievous as against Road Traffic Injury-Minor was recorded. Attendant factors, such as the ability of the victim to approach relief centres, the capability of health and other allied infrastructures, transportation and medical treatment of road traffic crash victims, and minimal access to other emergency services, including police, assumed greater significance than overall traffic flow in the incidence of Road Traffic Injury in the more severe second wave. These findings highlight the significant role these attendant factors play in producing the discrepancy between the actual road traffic incident rate and the officially registered rate. Thus, our study enables practitioners to observe the mobility-adjusted actual incidence rate devoid of factors related to reporting and registration of accidents.
本研究调查了伴随因素的重要作用，例如道路交通伤害受害者获得创伤中心的机会、卫生基础设施的稳健性，以及警察和紧急服务部门在大流行期间 COVID-19 封锁期间道路交通伤害 (RTI) 发生率中的反应能力。大流行第一波和第二波对感知健康风险和法律限制的不同影响为我们提供了一个自然实验，帮助我们区分伴随因素的影响和流动性与道路交通伤害之间的标准关系。作者在两个人口水平上使用自回归递归神经网络方法——泰米尔纳德邦 (TN)，一个以农村为主的州，以及该州最大的大都市钦奈，通过对道路交通死亡和道路交通伤害的每日计数进行反事实预测来进行因果推断。在第一波大流行期间，交通流量与道路交通死亡/道路交通伤害相关，其严重程度低于第二波。在第二波大流行的部分和封锁后阶段，道路交通伤害-严重的病例数量出人意料地下降了 70%以上，而道路交通伤害-轻微的病例数量则有所增加。在第二波大流行更为严重的情况下，伴随因素，如受害者接近救援中心的能力、卫生和其他相关基础设施的能力、道路交通碰撞受害者的运输和治疗，以及对包括警察在内的其他紧急服务的最小访问，在道路交通伤害的发生中比整体交通流量更为重要。这些发现强调了这些伴随因素在产生实际道路交通事件发生率与官方登记率之间差异方面的重要作用。因此，我们的研究使从业者能够观察到没有与事故报告和登记相关的因素的情况下的移动调整后的实际发病率。
