Risk factors trends

Jun 1, 2021

I am working on the development of Bayesian models able to integrate – in a principled and meaningful way – information from heterogeneous sources (such as microdata from large scale population surveys and local studies in special populations, surveillance data, administrative data on food and alcohol sales, …) and recover temporal trends and geographical and sociodemographic distribution of major risk factors for NCDs.

Sub-themes of this research are:

Harmonisation of sampling weights across multiple surveys and integration of design-based information into a Bayesian framework
Meta-regression models integration quality weighting
Calibration of large scale surveys using local high-quality data

NCDs Modelling

Annibale Cois

Senior Specialist Scientist

Posts

SASA Conference 2022

Annibale Cois

Last updated on Dec 10, 2022 1 min read

The annual conference of the South African Statistical Association.

Publications

Bayesian modelling of population trends in alcohol consumption provides empirically based country estimates for South Africa

Annibale Cois, Richard Matzopoulos, Victoria Pillay-van Wyk, Debbie Bradshaw

January 2021 Population Health Metrics

Alcohol use has widespread effects on health and contributes to over 200 detrimental conditions. Although the pattern of heavy episodic drinking independently increases the risk for injuries and transmission of some infectious diseases, long-term average consumption is the fundamental predictor of risk for most conditions. Population surveys, which are the main source of data on alcohol exposure, suffer from bias and uncertainty. This article proposes a novel triangulation method to reduce bias by rescaling consumption estimates by sex and age to match country-level consumption from administrative data.

PDF Project Project DOI