Studies testing the effectiveness of public health policies, like stop-smoking campaigns or exercise classes, often report outcomes at 6 or 12 months. To inform policy decisions, predicting the relative costs and benefits of these policies over the longer term is useful. Health economic models can be used to do this, but they often make simple assumptions about the impact of policies on longer-term behaviour, based on little evidence or theory. These assumptions can greatly affect model results. There has been little research on how to model behaviour within health economic models.
This project brings together experts from various disciplines including health economic modelling, complexity science, psychology, sociology and behavioural economics. The authors set out 12 methods which can be used to include influences on behaviours in public health economic models. For each method, a description is given, with key references. How the method can bring behavioural influences into the model and the minimum resources needed are also set out. The authors provide a table to help users decide which methods would be best given the type of problem and the policy context. There is also a PHEM-B toolbox website to make the toolbox easier to access.
An agenda for further research for both health economic modellers and behavioural scientists is set out because the project showed weaknesses in current behavioural theories, methods and the data available.
Dr Squires says "Policymakers have to make tough decisions about where to spend. This work provides a useful resource for modellers, behavioural scientists and policymakers to identify and use appropriate methods and behavioural theory within health economic models of public health interventions, to better inform policy decisions. It can be updated as new methods are developed."
This study was funded by the NIHR Fellowship Programme (NIHR301406). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.