In Europe and globally, a number of initiatives already exist to link different types of environmental data for different purposes, including efforts of the European Environment Agency (EIONET), of the WMO (IGACO), the OECD (ECOSANTE), UNECE (EMEP), Arctic Council (AMAP), the EU and European Space Agency (GMES), and the EU information system (INSPIRE). Efforts are also being made to link environmental and health information, such as the WHO ECEH indicator work, the WHO Health for All initiative, and the proposed European Integrated Environment and Health Monitoring and Response System. All these initiatives involve heterogeneous monitoring networks, and use a range of techniques (including GIS) for data harmonization and linkage. Integrated health assessment, however, is not the main purpose of any of these networks. Such assessments nevertheless necessarily require both routinely available and specialist data sets; these also need to be combined in different ways, according to the issue under consideration (e.g. the exposures of interest, the scale of analysis, or the population group). Often it is difficult to incorporate data on combined environmental pollutant loads or data on biomarkers of exposure or effect, for reasons that include non-representatives, non-availability, non-transferability, or because the effects of combined loads are not well understood and thus impossible to use. WP2.4 aims at addressing these.

Specific aims however are:

• to assess the methodological needs, and further develop methods, for integrated health assessment, in the light of the overall assessment framework (SP1) and the requirements of specific policy analyses (SP3);
• to provide methods for linking and combining data from different sources, including information on sources, the state of the environment, human population and the individual.

The work will be based on input from WPs 2.1-2.3, each of which will provide a comprehensive overview of existing information sources and methodologies, and suggest methods used to combine those within each work package. The detailed scope of this WP will thus be developed as results from these studies emerge. Important considerations, however, include the capability to provide comparable and internally consistent measures, covering different environmental agents, media and pathways, and different population groups, in the light of the often different sampling, reporting and analysis methods used. Data from monitoring and information networks, scientific studies and routine statistical sources will be used. Limited case studies will be carried out to test methods for linking, combining and complementing the different information. GIS and statistical modelling techniques will be considered, including Bayesian techniques, to help analyse the information in a coherent and logical (e.g. hierarchical) manner, allowing for their varying quality, completeness and extent. Statistical techniques for combining data sources (e.g. assimilation techniques) will also be considered. Methods for control and quantification of uncertainties will be developed in collaboration with WP 1.5, and will draw on approaches such as the NUSAP framework, used in the Dutch RIVM “Guidance for Uncertainty”.