Using behavioural theory to inform quantitative models for policy decision making

Abstract

Policy makers need to make decisions about how best to prevent people initiating smoking and facilitate quitting. Quantitative models can be used to help inform such policy decisions, by estimating the impacts of different policy options.

Several quantitative models for smoking policy appraisal have been developed, using a variety of modelling approaches. The advent of more advanced modelling techniques, improved theoretical models of behaviour, and extensive data series from detailed monthly surveys in England offer the prospect of achieving higher quality prediction of estimated outcomes than has thus far been possible.

We will present the methods and preliminary results from a collaboration between health psychologists, decision-analytic modellers and systems engineers using these advanced methods. In this first stage, causal systems mapping has been used to identify and estimate causal influences of individual level capability (e.g. knowledge of smoking harms, level of cigarette addiction), opportunity (e.g. cues to smoking from smoking peers, housing circumstances) and motivational (e.g. desire to quit) factors on: 1) onset of regular cigarette smoking, 2) initiation of a quit attempt, and 3) maintenance of abstinence for at least 4 weeks. Also modelled were causal influences of macro-level factors such as mass media spending, provision of smoking cessation support, and e-cigarette availability on the capability, opportunity and motivation factors. All variables were linked with unique identifiers of ontology classes in the Behaviour Change Intervention Ontology or Addiction Ontology.

We will then show how the systems maps can be incorporated into a simulation model that represents the dynamics of initiating smoking, making a quit attempt, and maintaining smoking abstinence. We will conclude with an overview of the remaining activities needed to develop a fully-fledged decision-analytic model for smoking policy appraisal.

Speakers

Professor Jamie Brown, Professor of Behavioural Science at University College London (UCL) and Director of the Tobacco and Alcohol Research  Group.

Jamie co-leads a CRUK programme of research to i) provide insights into population-wide influences on smoking, smoking cessation and alcohol reduction principally by management and analysis of the major population surveys, the Smoking and Alcohol Toolkit Study, and ii) advance the scientific foundation, and further the development of, potentially wide-reach digital behaviour change technologies. He is a co-author of Theory of Addiction (second edition) and ABC of Behaviour Change Theories, co-Regional Editor at the journal Addiction and an Editor of the Cochrane Tobacco Addiction Group.

Professor Robin Purshouse: Professor of Decision Sciences in the Department of Automatic Control and Systems Engineering at the University of Sheffield.

Robin’s primary research interests are in modelling and simulation methods for population health applications. He focuses on how simulation methods can be used to bring together psychosocial theory and data for model-based policy appraisal. His ESRC Future Research Leaders award (2012-14) developed a dynamic simulation model of the Theory of Planned Behaviour. He was principal investigator for an NIH R01 project ‘CASCADE’ (2016-22) that developed a formal architecture for mechanism-based modelling of social systems, with successful implementations of social norm theory, social role theory and dual process theory as dynamical simulations of alcohol use behaviours. Robin presently leads the development of an agent-based model of smoking cessation and initiation behaviours as part of the CRUK programme grant ‘Targeting multiple levels of the smoking cessation system using novel scientific approaches’ (2022-27). He also teaches a module in agent-based modelling methods to  Masters-level engineering students at the University of Sheffield.

Dr Hazel Squires: Senior Research Fellow in the Sheffield Centre for Health and Related Research (ScHARR) at the University of Sheffield.

Hazel is a healthcare decision modeller, with particular expertise in conceptual modelling, health economic modelling of public health and cancer interventions, and discrete event simulation. Within her doctoral research, Hazel developed a conceptual modelling framework for helping modellers to develop the structure of health economic models of public health interventions. She is currently building upon this within a NIHR Advanced Fellowship exploring how the influences on behaviour can be incorporated into health economic models of public health interventions.  

Professor Robert West: Professor Emeritus of Health Psychology at University College London.

Robert specialises in behaviour change. He is former Editor-in-Chief of the academic journal, Addiction, and has acted as an advisor to the English Department of Health on tobacco control and currently advises the Public Health Wales Behavioural Science Unit. He helped write the blueprint for the UK’s national network of stop-smoking clinics and is co-founder of the Capability-Opportunity-Motivation, Behaviour (COM-B) model of behaviour, the Behaviour Change Wheel framework for intervention development, and the PRIME Theory of motivation. He has published more than 900 scientific papers and is one of the world’s top cited smoking cessation experts.