Transmission of influenza virus in households and the community : surveillance and control

Abstract

Introduction: Influenza circulates every year and is associated with a

substantial burden to society. Infrequent influenza pandemics present a

serious potential threat to public health. In the response to the 2009

influenza pandemic, one weakness of the public health response was the

lack of situational awareness about disease transmissibility, early in the

pandemic. In the next pandemic, vaccines are again likely to be delayed

while antiviral treatment could be withheld for the most severe infections.

Simple personal non-pharmaceutical interventions such as improved hand

hygiene and the use of face masks could assist in mitigating a pandemic

but their effectiveness remains uncertain. In this thesis I developed and

applied two statistical modeling approaches to aid the interpretation of

influenza surveillance data collected on population and epidemiological

data collected at the individual level. They facilitate improved situational

awareness and provide better estimates of the effectiveness of personal

non-pharmaceutical interventions.

Methods: A key quantity in infectious disease epidemiology is the Effective Reproductive Number, R, which is defined as the average number

of secondary cases generated by a single index case. I extended an existing method for estimation of R in real-time to account for reporting

delays and applied it to data on case notifications and hospitalizations

associated with pandemic (H1N1) in Hong Kong from June through October 2009. In 2008, a randomized controlled trial was conducted to

investigate whether increased hand hygiene and facemasks were effective

in reducing influenza transmission in households. Preliminary analysis of the data failed to identify significant effects of the interventions,

but did not account for the underlying transmission dynamics. I developed a stochastic transmission model, within the Bayesian framework,

to estimate the effectiveness of facemasks and improved hand hygiene in

reducing household transmission.

Results: The method developed for estimating R was demonstrated to

be useful in estimating the real-time transmissibilty of pandemic (H1N1)

virus in Hong Kong from June to October 2009, which ranged from 1.3-

1.6 through the phase of increasing incidence. The household transmission analysis was able to identify a moderate effect of improved hand

hygiene and facemasks in reducing influenza transmission in households,

where the daily interventions efficacies ranged from 18% to 27% in hand

hygiene group and ranged from 5% to 13% in the hand hygiene plus

facemasks group.

Conclusions: The extended method for estimating R provided a practical tool for the surveillance of influenza transmission and should be useful in future pandemics. Further work is needed to extend the approach

to account for age. The household transmission analysis suggested that

simple personal interventions should be taken into account in strategies

for pandemic mitigation.