Spatial Mapping of Influenza Infection by Bayesian Approach

Abstract

Implementation and vaccination of the national policy have to need the obvious evidence based on reliable data. Geographical mapping plays an essential role to answer for these policies. However, difference of age structure effects on each area unequally. The Standardized Infection Ratio (SIR) is a risk measurement which can be able to control some confounders of interest such as age and gender among areas. A Bayesian approach is a popular mean for creating a map because of its ability in getting information from observed data and from a prior knowledge. Data on influenza infection of 506-national surveillance over the whole country, classified by age structure and by provinces, collected by Bureau of Epidemiology, Ministry of Public Health, are adopted because of occurring an epidemic of flu in Thailand 2016. Results demonstrated that there were 2.6 persons per thousand people with influenza infection in 2016. After applying the SIR of influenza infection with the national standard reference of the past four years 2012–2015 overall the country to each province, the SIR for each province broadened around the mean of 2.07 (variance = 3.93) and this outcome reported approximately increasing two times of SIR in year 2016, compared with the national reference in 2012–2015. Lastly, Bayesian estimation found that the five highest risk provinces were Bangkok (9.6117), Chiang Mai (8.0949), Phayao (7.1871), Uttaradit (6.6086) and Phitsanulok (6.2168), respectively.