eBooks View all | By AtoZ | By category

Likelihood-Based Estimation of Dynamic Transmission Model Parameters for Seasonal Influenza by Fitting to Age and Season Specific ILI Data

By Michael Waithaka

University of Hasselt, Belgium

Biometrics & Biostatistics

Published: Mar 24, 2017 | pg. no: 1-15

Abstract: Mathematical models provide important practical insights into the epidemiology of infectious diseases, and concepts derived from such models are widely used in the design of infection control programmes. This project was aimed at directly estimating the parameters of a dynamic transmission model using likelihood-based estimation methods, by fitting the model to age-specific influenza-like-illness (ILI) incidence over multiple influenza seasons. In an attempt to achieve the goal of this project, the dynamic transmission model for seasonal influenza of Vynnycky et al. [1] was adopted and the various model parameters estimated. Weighted Least Squares and Maximum Likelihood estimation methods were applied for the model parameters estimation. From the obtained estimates of these parameters, estimates for the average basic reproduction numbers, which is an important measure used in infectious disease control, immunization and eradication programmes, were also derived. This modelling approach is an improvement to the previous approaches where the parameter values of seasonal influenza models were commonly chosen ad hoc though projections based on such models heavily rely on the assumed input parameter values. Moreover, there exists considerable uncertainty over the most appropriate values for parameters for such models. The importance of parameter estimation and accounting for uncertainty when using dynamic transmission model outcomes as input for economic evaluations related to infectious diseases have already been highlighted by several previous studies [2,3].

View eBook

Contact Us:

MedCrave Group, Danforth Rd., Edmond, OK 73003
Tel: +1-918-208-0288, Fax: +1-918-514-7616,
Email: ebooks@medcraveonline.org

Follow us

Indexing

google scholar