Identification and Estimation of Causal Effects Using a Negative Control Exposure in Time-series Studies with Applications to Environmental Epidemiology.

Identification and Estimation of Causal Effects Using a Negative Control Exposure in Time-series Studies with Applications to Environmental Epidemiology. Am J Epidemiol. 2020 Aug 24;: Authors: Yu Y, Li H, Sun X, Liu X, Yang F, Hou L, Liu L, Yan R, Yu Y, Jing M, Xue H, Cao W, Wang Q, Zhong H, Xue F Abstract The initial aim of environmental epidemiology is to estimate the causal effects of environmental exposures on health outcomes. However, due to lack of enough covariates in most environmental datasets, current methods without enough adjustments for confounders inevitably lead to residual confounding. We propose a Negative Control Exposure based on Time-series Studies (NCE-TS) model to effectively eliminate unobserved confounders using a post-outcome exposure as a negative control exposure. We show that the causal effect is identifiable and can be estimated by the NCE-TS for continuous and categorical outcomes. Simulations studies indicate the unbiased estimation by the NCE-TS model. And the potential of the NCE-TS is illustrated by two challenging applications. We find that living in areas with higher levels of surrounding greenness over six months has less risks in stroke-specific mortality based on Shandong Ecological Health Cohort during the 1st January 2010 to the 31st December 2018. In addition, the widely-established negative association between temperatures and cancer risks is actually caused by numbers of unobserved confound...
Source: Am J Epidemiol - Category: Epidemiology Authors: Tags: Am J Epidemiol Source Type: research