Statistical and dynamic modelling
What is modelling?
In the drugs field, many processes are difficult to study directly as information is often imprecise, delayed or not available. By using mathematical formulae or computer programmes, a simplified model of a process can be constructed. This model is then tested for its consistency against available data, or the results are examined when one component of the model is changed. This interactive procedure often demonstrates that the model is not satisfactory. The next step is then to find what is wrong, which component has been omitted, and so on. The end result can be a useful model to investigate different outcomes under different circumstances or interventions, or a 'best estimate' of future developments in the spread of drug use or a drug-related infectious disease, such as hepatitis C. Such models can be extremely useful tools for policy makers, as they enable taking decisions on resource allocation and interventions in situations where data are scarce.
Projects
The EMCDDA and the University of York, UK, have conducted a project to explore ways of using modelling as a tool in drug epidemiology. It resulted in a scientific monograph with contributions from EU experts (click here) and proposals for specific areas of work.
In some of these areas the EMCDDA subsequently implemented pilot studies (See: Time trends and Incidence, Geographic spread and GIS, Impact and costs of drug-related infectious diseases.
With funding of the European Commission (TSER/DG Research), an European expert network has been established to stimulate further work on using mathematical and statistical models in drug research. This network has continued the work on the above themes.
