Scientific activities
Flux RSS Glossary   Faq   
Scientific activities > Assessments methods > Information System

Survey on data linkage in the environmental health field - - Action 35 of the French National Environment and Health Action Plan (NEHAP) "Improve the performance and integration of environmental health information systems"

In order to answer the questions over the health effects on short and medium term, of exposure to some of their environmental pollutants, a French National Environment and Health Action Plan (NEHAP) was drawn up for the period 2004-2008. The NEHAP is split up into 45 actions organised under 8 key points covering the whole field of the relationship between the environment and health through a consideration of all pollutants and living environments. Key point no. 7 concerns the improvement of systems for monitoring, surveillance and alert which in part rely on information systems. An information system is a set of resources (human, equipment, data etc) enabling information to be managed, stored, processed and disseminated. In an effort to strengthen and better coordinate environmental health information systems, two projects were undertaken within the framework of action 35 of the NEHAP: "Improving the performance and integration of environmental health information systems". The first project involved finding out what already existed by making an inventory of existing databases in the field of health and the environment. This inventory was then used as a basis for a survey on the linkage of data in the field of environmental health. The French agency for Environmental and Occupational Health & Safety (Afsset) and the French institute for the Environment (Ifen) carried out this work within multi-institutional coordination and steering groups.



In order to improve information on exposures and health impacts, a feasibility survey of the linkage of environmental, health and demographic data was undertaken. Health and environmental information systems have very often been designed and developed according to different models and for different purposes, in response to specific restrictions and needs (monitoring of environmental quality, application of regulations, alerts etc). Nevertheless, they are often subsequently used in conceptual frameworks and with different objectives from those for which they were created. Such a process often implies a reprocessing of the data for each research project.


Working method:

The methodology chosen consisted of making a feedback survey of French scientific teams having, during their research, carried out the linkage of data from existing environmental, health, demographic and/or geographical databases. This was achieved through semi-directive interviews undertaken with these teams about one of their past or ongoing studies, based on a common interview grid used for all of the studies. The type of information gathered by the interview grid concerned the survey itself (objective, statistical approaches, results, boundaries etc), data and linking data methods (raw data, linked data, transformation prior to data linkage), the advantages, limits and difficulties encountered depending on the scale, ways of accessing the data etc, developments desired for the databases used and the expectations of the team concerning the survey.


On the basis of the information gathered during the interviews, a interview report was systematically sent to the team concerned for additional information and validation. This survey was carried out with assistance from the consultants PricewaterhouseCoopers. From these results, a final report was drawn up aimed at showing what lessons could be learned about the feasibility of linking data and about the methods used by the teams enabling this to be done. This report reveals the perceptions of the participants with no intention of being exhaustive or representative of all the research conducted in the environmental health field.



In all, 19 scientific teams took part in the survey, mainly research teams. Nearly all the studies included in the survey aim at statistically comparing the description of the state of health of a population with various environmental and/or socio-demographic indicators.


To meet the objective defined in its study protocol, the scientific team sets up a analytical database in which a large number of variables (dependent, independent and confounding) are included to link in a statistical model or by a geographical information system. These variables are constituted using a wide diversity and consequently variety of data sources that were usually set up with a different objective (administrative, management, regulatory etc) from the study's objective. For the scientific teams this means undertaking a complex process of identifying the data sources, analysing and selecting adequate data, acquiring data and transforming them into variables for linking.


Even if each study is a specific case in itself, the report highlights general aspects applicable to all these stages prior to data linkage: the time spent in identifying the data sources and appraising the data they contain, ways of accessing data based on relationships and negotiation, sometime complex ways of optimising or get round the limits of available data, etc.
The information from this survey now allow us to make recommendations with the purpose of better use of environmental, health demographic and geographic data in the environmental health field through the improvement of the connected information systems.




 Environmental health information systems: survey on data linkage in the environmental health field (Action 35 of the French National Environment and Health Action Plan)


Copy the link to this page