The European Journal of Public Health Advance Access originally published online on October 26, 2006
The European Journal of Public Health 2007 17(1):3-4; doi:10.1093/eurpub/ckl077
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Viewpoints |
French experiences with Health Impact Assessment of disasters
Pierre Verger1, Marielle Aulagnier1, Valérie Schwoebel2 and Thierry Lang3
1 Observatoire Régional de la Santé, Provence Alpes, Côte d'Azur Marseille, France
2 Cellule Inter-régionale d'Epidémiologie Midi-Pyrénées, Institut National de Veille Sanitaire Toulouse, France
3 Institut Fédératif de Recherche 126 ‘Santé, Société’, Inserm U558 et Institut National de Veille Sanitaire Toulouse, France
Correspondence: Thierry Lang, Institut Fédératif de Recherche 126 ‘Santé, Société’, Inserm U558 et Institut National de Veille Sanitaire, Toulouse, France, e-mail: lang{at}cict.fr
Several catastrophes in France have been followed by health evaluation programmes/studies. After the AZF factory explosion (Toulouse, 2001, 30 deaths, 3000 injured persons), a large Health Impact Assessment (HIA) programme was set up which demonstrated a large impact on mental health (post-traumatic stress disorder and depressive symptoms) and hearing impairment.1 This programme has also been designed to assess indirect health effects of the disaster due to housing destruction, schooling consequences, job loss, and so on. Epidemiological studies carried out several years after the 1992 floods in South-eastern France or the 1995–96 terrorist attacks (Paris) helped to evaluate their long-lasting physical as well as mental health consequences and determine risk factors for mental health effects.2,3
These experiences lead us to consider the role of post-disaster HIA and in particular of epidemiology and the necessity to anticipate catastrophes.4 Indeed, setting up these evaluations raised countless difficulties: lack of information on pre-disaster health indicators (PTSD prevalence, etc.) or soil contamination data; lack of exposure data, and of some up-to-date demographic data bases, victims who were lost to follow-up when post-disaster HIA was launched several years after the event, difficulties to raise funds, etc. The first task we were confronted with was thus to explicitly point out what post-disaster epidemiology can provide when disaster occurs: decision-making support, public information, and scientific knowledge.4
 |
Decision-making support |
|---|
 Top Decision-making support Public informationScientific knowledgeReferences |
|---|
Decision-making during a disaster aims first of all to protect the population and to avoid its exposure, overexposure, or re-exposure to hazards. Implementation of emergency protective measures requires knowing the relevant hazards, the size and characteristics of the exposed population groups, their exposure levels, and the risks associated with these levels. After the accident at AZF, we encountered tremendous difficulties to obtain such information because of the lack of coordination between institutions.
Decisions must also be made to manage health hazards, to screen for specific health problems, and to organize medical follow-up. Early monitoring of emergencies, or health surveillance extended beyond the emergency phase to monitor long-lasting or delayed effects (cancers) will provide data useful to guide such decisions. Surveillance data have been obtained either by mobilizing and adapting existing information systems (physicians networks, hospital information systems, alert systems, cancer registries) or by setting up ad hoc surveillance systems. The value of these routine surveillance data needs to be carefully assessed in order to be used as a data source for future catastrophes and possibly avoid ad hoc studies. However, surveillance data may not always be sufficient to establish recommendations. After the AZF explosion, early surveillance systems raised the problem of post-traumatic hearing losses: lack of scientific knowledge on such effects happened to be an obstacle to rapid and effective recommendations regarding a systematic screening in the exposed population. Validation by specific epidemiological studies to assess an exposure–effect relationship was required as a first step.
|
Public information |
|---|
|
TopDecision-making supportPublic informationScientific knowledgeReferences |
|---|
Swift and transparent communication about health risks and consequences of a disaster is an institutional and political necessity. Information on health consequences of a disaster must be available to the media, public, victims, health care professionals, and various stakeholders.
Epidemiology helps to broaden the view on the health consequences of disasters (provided its results are clearly communicated): in terms of their nature (morbidity, mortality, etc.), their causes (physical, chemical, psychosocial, etc.), their importance (risk quantification), time-occurrence, and duration. The various definitions of a ‘victim’ that are employed by the epidemiologists and other institutions, the media, or the population need to be clarified, since they refer to various types of consequences (physical versus mental; direct or indirect through psychosocial consequences; on the short or long term; perceived or acknowledged by institutions, etc.). Distrust between the media and the population on one hand and the ‘experts’ on the other hand might result from the variety of definitions.
The stakes for information involve both public health and democracy: it is necessary not only to ensure transparency about the disaster's effects but also about the effectiveness of interventions, to respond to questions that may occur and concerns that will inevitably be raised, for example, if disease clusters are discovered, and to limit rumours insofar as possible. After the 1988 floods at Nîmes (South France) a rumour accused the public authorities to hide the real death toll. Mortality and population-based epidemiological studies were carried out to confirm the death toll and help to understand the exposure circumstances which explained the limited death toll.5
|
Scientific knowledge |
|---|
|
TopDecision-making supportPublic informationScientific knowledgeReferences |
|---|
Disasters have multiple consequences, which are not always documented or specified. Any disaster is therefore an opportunity to increase our knowledge and improve response to any future event. For example, psychological debriefing of victims has been used extensively in France for several years although its effectiveness has been questioned in the literature. A lot has still to be learned on mental health consequences of disasters, and risk factors of vulnerability to their effects—somatic or mental. This is all the more true if the populations affected include vulnerable socio-economic groups (AZF). Epidemiological studies conducted after the accident at AZF and the 1992 floods in South-eastern France showed that stressors associated to these disasters may vary according to the victims' socio-economic status.1,2
|
References |
|---|
|
TopDecision-making supportPublic informationScientific knowledgeReferences |
|---|
1 Lang T, Schwoebel V, Diène E, et al. Assessing post-disaster health consequences at the population level: experience from the AZF factory explosion in Toulouse. J Epidemiol Community Health In Press.
2 Verger P, Rotily M, Hunault C, et al. (2003) Assessment of exposure to a flood disaster in a mental-health study. J Expo Anal Environ Epidemiol 13:436–42.[CrossRef][Web of Science][Medline]
3 Verger P, Dab W, Lamping DL, et al. (2004) The psychological impact of terrorism: an epidemiologic study of posttraumatic stress disorder and associated factors in victims of the 1995–1996 bombings in France. Am J Psychiatry 161:1384–9.
4 Verger P, Aulagnier M, Schwoebel V, Lang T. (2005) Démarches épidémiologiques après une catastrophe. Réponses environnement, Ministère de l'Ecologie et du Développement Durable(La Documentation Française, Paris).
5 Duclos P, Vidonne O, Beuf P, Perray P, Stoebner A. (1991) Flash flood disater Nimes, france, 1988. Eur J Epidemiol 7:365–71.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||