|
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 #322304
|
|
|
Title:
|
Estimation of Causal Effects in Randomized Trials of Infectious Disease Prevention with General Interference
|
|
Author(s):
|
Nicole Bohme Carnegie*
|
|
Companies:
|
University of Wisconsin-Milwaukee
|
|
Keywords:
|
Contact networks ;
Epidemic models ;
Infecious disease ;
Causal inference
|
|
Abstract:
|
An issue that remains challenging in causal inference is relaxing the assumption of no interference between units. Existing methods largely depend upon an assumption of ``partial interference'' - interference within identifiable groups but not among them. There remains a considerable need for development of methods that accommodate general interference. We focus on the difference in the outcome if treatment were provided to all clusters compared to that outcome if treatment were provided to none - an overall treatment effect (OTE). In infectious disease prevention trials, the randomized treatment effect estimate will be attenuated relative to the OTE if a fraction of exposures in the treated clusters come from outside, but the proportion of contacts sufficient for transmission that are with treated clusters is potentially measurable. We leverage epidemic models to infer how a given level of interference affects the incidence of infection in clusters. This leads naturally to an estimator of the OTE that is easily implemented using existing software. We then examine under what contact network structures this estimator may fail and apply the method to a smallpox outbreak.
|
Authors who are presenting talks have a * after their name.
