Harnessing advances in computer simulation to inform policy and planning to reduce alcohol-related harms

Authored by Michael Livingston, Jo-An Atkinson, Ante Prodan, Dylan Knowles, Eloise O'Donnell, Robin Room, Devon Indig, Geoff McDonnell, John Wiggers, Sandra Jones, Paul S Haber, David Muscatello, Nadine Ezard, Nghi Phung, Louise Freebairn, Lucie Rychetnik, Jaithri Ananthapavan, Sonia Wutzke, Modelling Consortium Alcohol

Date Published: 2018

DOI: 10.1007/s00038-017-1041-y

Sponsors: Australian National Health and Medical Research Council (NHMRC)

Platforms: AnyLogic

Model Documentation: Other Narrative Flow charts

Model Code URLs: Model code not found

Abstract

Alcohol misuse is a complex systemic problem. The aim of this study was to explore the feasibility of using a transparent and participatory agent-based modelling approach to develop a robust decision support tool to test alcohol policy scenarios before they are implemented in the real world. A consortium of Australia's leading alcohol experts was engaged to collaboratively develop an agent-based model of alcohol consumption behaviour and related harms. As a case study, four policy scenarios were examined. A 19.5 +/- 2.5\% reduction in acute alcohol-related harms was estimated with the implementation of a 3 a.m. licensed venue closing time plus 1 a.m. lockout; and a 9 +/- 2.6\% reduction in incidence was estimated with expansion of treatment services to reach 20\% of heavy drinkers. Combining the two scenarios produced a 33.3 +/- 2.7\% reduction in the incidence of acute alcohol-related harms, suggesting a synergistic effect. This study demonstrates the feasibility of participatory development of a contextually relevant computer simulation model of alcohol-related harms and highlights the value of the approach in identifying potential policy responses that best leverage limited resources.

Tags

Agent-based modelling Alcohol-related harm Prevention policy Evidence synthesis