Estimation of sub-national level population health outcomes: Applications of Small Area Estimation
The talk will illustrate three applications of the model-based small area estimation (SAE) method for estimating trends of population health outcomes at different levels of geographical detail (i.e., small domains). The first application will focus on trend estimation of sub-national level (state and territory) smoking prevalence in Australia by age and sex using the National Health Survey (NHS) data over 2001-2021.
The second application will focus on trend estimation of child stunting in Bangladesh at admin-2 (64 districts) and admin-3 (544 sub-districts) levels using the Bangladesh Demographic and Health Survey (BDHS) data collected over 2000-2018. The inclusion of remote-sensed data in the model has improved the trend prediction of stunting level, particularly in the non-survey years. Joinpoint trend analysis has also been implemented to identify the time-periods and the domains for which improvement was poor over the study time period.
In the third application, we show how the disadvantage of the local area (i.e., socio-economic indexes for areas (SEIFA)) and the remoteness (i.e., accessibility/remoteness index of Australia (ARIA)) contribute to improved prevalence estimates of child development vulnerability in statistical areas level 3 (SA3) and 4 (SA4) across Australia. These three examples will showcase how we can have improved and reliable estimates of disaggregated level population health outcomes (when compared to the direct survey-based results) through the model-based SAE approach.
Bernard Baffour is an associate professor in the School of Demography. Bernard has methodological expertise in survey methods and the analysis of complex data. He has a diverse range of experience in working across a wide spectrum of areas from education, sociology, epidemiology, public health and operational research.
Join Zoom Meeting
https://anu.zoom.us/j/82628042953?pwd=rj0XiXiE8XdoxOyeHmSti6yaFT205j.1
Meeting ID: 826 2804 2953
Password: 105940