Causal Inference in the Time of Covid-19

TL;DR

This paper develops statistical methods for causal inference in epidemics, specifically estimating how social mobility impacts Covid-19 death rates, highlighting both findings and limitations in causal analysis during pandemics.

Contribution

It introduces a marginal structural model for epidemic data and evaluates the causal effect of mobility on Covid-19 deaths with sensitivity analyses.

Findings

Reduced mobility is associated with fewer deaths, but causality is uncertain.

Results are sensitive to model assumptions and unmeasured confounding.

Causal effect estimates should be interpreted with caution.

Abstract

In this paper we develop statistical methods for causal inference in epidemics. Our focus is in estimating the effect of social mobility on deaths in the Covid-19 pandemic. We propose a marginal structural model motivated by a modified version of a basic epidemic model. We estimate the counterfactual time series of deaths under interventions on mobility. We conduct several types of sensitivity analyses. We find that the data support the idea that reduced mobility causes reduced deaths, but the conclusion comes with caveats. There is evidence of sensitivity to model misspecification and unmeasured confounding which implies that the size of the causal effect needs to be interpreted with caution. While there is little doubt the the effect is real, our work highlights the challenges in drawing causal inferences from pandemic data.