Abstract:
|
Both transmission trees and pathogen phylogenies exclude all individuals who were not infected. The transmission tree itself does not generalize to future outbreaks, and excluding all uninfected individuals who were at risk of infection is not justified in any standard epidemiologic study design. Instead of bending methods to fit the data collected during outbreaks, infectious disease epidemiologists have a duty to develop and advocate better study designs for future outbreaks. Methods that incorporate pathogen genetic sequence data should be justified using standard rules of probability and causal inference, and they should be tested in simulations of epidemics, not branching processes. Pairwise survival analysis offers a general framework for study design and data analysis in infectious disease epidemiology that can incorporate pathogen genomes. Data on individuals who escaped infection is critical with or without pathogen genetic sequences. The availability of pathogen genomes is best seen as an evolutionary, not revolutionary, development in infectious disease epidemiology.
|