|
Activity Number:
|
99
- Causal Inference for Infectious Disease Outcomes: Interference, Contagion, and Networks
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 31, 2017 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #322137
|
|
|
Title:
|
Estimating Vaccine Effects in Longitudinal Studies of Interconnected Populations
|
|
Author(s):
|
Forrest W Crawford*
|
|
Companies:
|
Yale School of Public Health
|
|
Keywords:
|
vaccine ;
causal inference ;
spillover ;
interference ;
semi-parametric model ;
infectious disease
|
|
Abstract:
|
Estimating the effect of an infectious disease intervention in a single interconnected population is challenging because subjects may transmit infection to others. Preventive treatment (e.g. vaccination) may exert a direct effect on the person who receives it, and may protect others by preventing infection in the recipient and thereby transmission to someone else, or by reducing the recipient's infectiousness when infected. We outline a causal framework for estimating the direct and indirect effects of a vaccine in a single networked population. Epidemiological assumptions collapse subjects' outcomes over time into cumulative exposures experienced by susceptible subjects; conditional independence assumptions permit identification of the direct effect and partition of the indirect effect into distinct effects on susceptibility and transmissibility. I describe a semi-parametric class of infectious disease regression models motivated by a continuous-time Markov stochastic epidemic process. I describe the consequences of misspecification of the infection model, and discuss approaches to estimation when either the network or time series of infections are incompletely observed.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2017 program
|
