SPREAD PATTERN FORMATION OF H5N1-AVIAN INFLUENZA AND ITS IMPLICATIONS FOR CONTROL STRATEGIES
Author(s)
Venkata Rama Satya Kumar Duvvuri, MSc, PhD, Postdoctoral Fellow1, Rongsong Liu, MS, PhD, Assistant Research Professor2, Jianhong Wu, MSc, PhD, Professor11York University, Toronto, ON, Canada; 2 Purdue University, West Lafayette, IN, USA
To develop deterministic models for comparing the relative effectiveness of various control measures, estimation of the spread speed of the virus in the environment and the interactions between spatial diffusion of birds and virus. Deterministic models were developed based on the H5N1 transmission cycle by partitioning the birds into three classes based on their epidemiological characteristics for the disease under consideration: poultry (c), wild birds - susceptible and die after H5N1 infection (w), and wild birds - susceptible but survive after H5N1 infection without apparent disease symptoms (d). Model parameters were obtained from existing literatures. Ordinary differential equations (ODE) and partial differential equations (PDE) analogues were utilized for the construction of models as described below. ODE was used to access the effects of control measures, one-dimensional PDE to study the spread speed of the disease, and two-dimensional PDE to examine how the epicenter moves in space and time under different assumptions about the speed and direction of poultry. ODE Simulations showed that by reducing 95% of the initial susceptible poultry population or by culling all infected poultry birds within one day disease outbreak could control lead in a local setting. Results further elucidated that cleaning the environment is also a feasible and useful control measure, but culling wild birds and destroying their habitat are ineffective control measures. We noticed from the PDE models that the diffusion rate of the (w) has very little impact on the spread speed (1.69-1.74 km/day) where as (d) has shown substantial raise of spread speed (2-7.8 km/day) depending on the transmission direction indicating significant role of migration. Finally, we assumed that epicenter progresses dominantly along the convection direction of the domestic poultry and the disease spread to other direction via random diffusion. Mathematical modeling could prove effective in answering epidemiological issues.
Conference/Value in Health Info
2008-05, ISPOR 2008, Toronto, Ontario, Canada
Value in Health, Vol. 11, No. 3 (May/June 2008)
Code
PIN49
Topic
Health Service Delivery & Process of Care
Topic Subcategory
Hospital and Clinical Practices
Disease
Infectious Disease (non-vaccine)