Abstract
The World Health Organisation's definition of public health refers to all organized measures to prevent disease, promote health, and prolong life among the population as a whole (World Health Organization, 2014). Mathematical modelling plays an increasingly important role in helping to guide the most high impact and cost-effective means of achieving these goals. Public health programmes are usually implemented over a long period of time with broad benefits to many in the community. Clinical trials are seldom large enough to capture these effects. Observational data may be used to evaluate a programme after it is underway, but have limited value in helping to predict the future impact of a proposed policy. Furthermore, public health practitioners are often required to respond to new threats, for which there is little or no previous data on which to assess the threat. Computational and mathematical models can help to assess potential threats and impacts early in the process, and later aid in interpreting data from complex and multifactorial systems. As such, these models can be critical tools in guiding public health action. However, there are a number of challenges in achieving a successful interface between modelling and public health. Here, we discuss some of these challenges.
Conclusions
Having national or international health policy change as a direct consequence of your own work can be enormously fulfilling for the analyst. However, for this to occur, and to ensure that the policy adequately represents the modelling work, proper engagement with policy makers is necessary. Productive engagement is a longterm process, and involves a deeper understanding ofthe needs and constraints of policy makers combined with a willingness to alter models inorder to try andbetter reflecttheseneeds andconstraints. At the same time, both modellers and their policy partners must work hard to appropriately interpret results, in particular appropriately communicating uncertainty. This engagement should help ensure that policy makers understand the limitations and constraints ofthe models better, while giving them more opportunities to usemodels as tools for decisions. Deeper engagement with policy makers will help modellers find the best ways of communicating clear and scientifically accurate information to best guide the development of policy.
