Final Report Summary - SARSTRANS (Control policy optimisation for SARS and other emerging infections: characterising transmission dynamics and estimating key epidemiological parameters)
The project builds on past work with collaborators in Hong Kong which facilitated the creation of a structured database for all SARS cases in those regions and Taiwan, and the analysis day by day of the course of the epidemic and the impact of different interventions (Riley et al., Science 300, 1961-1966, 2003, published online 23 May 2003; Donnelly et al., Lancet, 361, 17611766, 2003, published online 7 May 2003).
Clinical aspects addressed in this project include: the typical course of infection in patients of different ages and with various co-morbidities; the infectiousness (level of viral shedding) of patients through time following acquisition of infection; and the distributions of incubation and infectious periods as a function of modifying factors such as age. Epidemiological aspects include the routes of transmission of the virus, and the types of contacts that resulted in transmission.
Although much of the research is focused on SARS, the overall aims included the development of a mathematical and statistical framework that will facilitate general analyses. The framework allows assessments of how best to control the spread of infectious agents of a given type (i.e. given incubation and infectious period distributions) in defined local, national, or global settings.
In previous deliverables, we concentrated on outlining recommendations on the construction and real-time analysis of databases for the monitoring and evaluation of epidemic outbreaks of infectious diseases that integrate epidemiological, clinical and treatment information. We not only provided insight into the type of information that needs to be collected and stored, but also, equally importantly, described efficient data-capture systems and central data-base management and work protocols to allow optimal use of data collected.
We now add to this the development of the tools of analysis necessary for robust and coherent evaluation of the possible effects of control options. During this second reporting period, the consortium has concentrated on developing a generic set of tools to guide data collection and analysis, as well as public health policy formulation, in situations where new infectious agents emerge and transmit within communities and countries. Examples of tools developed include statistical methods for parameter estimation and the fitting of models to daily case reports, and mathematical models to test the potential impact of different interventions, acting alone or in combination.
The consortium has also addressed risk and cost benefit analysis for intervention strategies, both in terms of the magnitude of the change in transmission and the length of time the intervention needs to be in place. The majority of this information is available as peer-reviewed scientific articles.
We have also developed user-friendly software to simulate simple model predictions. The online application (please see http://www.u707.jussieu.fr/periodic_regression/ online) allows easy detection of special events in an epidemiologic time series and quantification of excess mortality/morbidity as a change from baseline. A stochastic general epidemics model has also been developed and can be accessed at http://cran.r-project.org/src/contrib/Descriptions/stochasticGEM.html.
Clinical aspects addressed in this project include: the typical course of infection in patients of different ages and with various co-morbidities; the infectiousness (level of viral shedding) of patients through time following acquisition of infection; and the distributions of incubation and infectious periods as a function of modifying factors such as age. Epidemiological aspects include the routes of transmission of the virus, and the types of contacts that resulted in transmission.
Although much of the research is focused on SARS, the overall aims included the development of a mathematical and statistical framework that will facilitate general analyses. The framework allows assessments of how best to control the spread of infectious agents of a given type (i.e. given incubation and infectious period distributions) in defined local, national, or global settings.
In previous deliverables, we concentrated on outlining recommendations on the construction and real-time analysis of databases for the monitoring and evaluation of epidemic outbreaks of infectious diseases that integrate epidemiological, clinical and treatment information. We not only provided insight into the type of information that needs to be collected and stored, but also, equally importantly, described efficient data-capture systems and central data-base management and work protocols to allow optimal use of data collected.
We now add to this the development of the tools of analysis necessary for robust and coherent evaluation of the possible effects of control options. During this second reporting period, the consortium has concentrated on developing a generic set of tools to guide data collection and analysis, as well as public health policy formulation, in situations where new infectious agents emerge and transmit within communities and countries. Examples of tools developed include statistical methods for parameter estimation and the fitting of models to daily case reports, and mathematical models to test the potential impact of different interventions, acting alone or in combination.
The consortium has also addressed risk and cost benefit analysis for intervention strategies, both in terms of the magnitude of the change in transmission and the length of time the intervention needs to be in place. The majority of this information is available as peer-reviewed scientific articles.
We have also developed user-friendly software to simulate simple model predictions. The online application (please see http://www.u707.jussieu.fr/periodic_regression/ online) allows easy detection of special events in an epidemiologic time series and quantification of excess mortality/morbidity as a change from baseline. A stochastic general epidemics model has also been developed and can be accessed at http://cran.r-project.org/src/contrib/Descriptions/stochasticGEM.html.