Le but de ce travail est d'élaborer et d'évaluer une intervention de santé publique, visant à augmenter la participation au dépistage ciblé par coloscopie de la population à risque plus élevé de cancer colorectal (CCR) en raison d'antécédents familiaux en faisant appel à une méthodologie qui repose sur 3 étapes successives :1- une étude observationnelle : l'identification des déterminants associés à la réalisation de la coloscopie chez la fratrie d'un patient atteint de CCR avant 60 ans. (3 publications) 2- une expérimentation contrôlée, visant à tester l'efficacité d'une intervention personnalisée, par une infirmière de prévention, sur la participation au dépistage des fratries d'un patient atteint de CCR ou d'adénome avant 60 ans. L'intervention personnalisée qui a intégré non seulement les fratries mais également les cas index et les médecins a fait la preuve d'une efficacité jusque là jamais atteinte dans les études menées antérieurement aux USA, qui ciblaient uniquement les fratries. Cette analyse épidémiologique a été enrichie par des analyses linguistiques et sociologiques afin de mieux comprendre l'impact de cette intervention téléphonique. (1 publication)3- Cependant, les contraintes propres à la conduite d’un essai interventionnel randomisé, si elles garantissent un niveau de preuve élevé du bénéfice de l’intervention, ont aussi pour conséquence un caractère expérimental ; cet essai ne pourra être généralisé à l'identique. Une étude de recherche interventionnelle en santé des populations a été soumise à l'appel à projets 2016 de l'INCa. / The aim of this work was to develop and evaluate a public health intervention to increase participation in targeted screening colonoscopy for high-risk population of colorectal cancer (CRC) because of family history using a methodology based on three successive stages:1- an observational study to identify the determinants associated with the achievement of colonoscopy among siblings of a patient with CRC before age 60. (3 publications)2- a controlled experiment to test the effectiveness of a personalized intervention by a preventive nurse on the participation in screening of siblings of a patient with CRC or adenoma before age 60. The personalized intervention that included not only siblings but also the index case, specialists and general practitioners has demonstrated efficiency never achieved previously in earlier studies in the USA, which targeted only siblings. This epidemiological analysis was enriched by crossed linguistic and sociological analyses to better understand the impact of telephone interventions. (1 publication)3- However, the constraints on the conduct of a randomized interventional trial, if they guarantee a high level of evidence of the benefit of the intervention, also have the effect of an experimental nature that cannot be generalized. An original research study on population health intervention has been developed and submitted to the 2016 INCa call for projects.

Parce qu’il favorise l’initiation précoce du traitement qui limite la morbi-mortalité et les transmissions secondaires, le dépistage du VIH est essentiel au contrôle de l’épidémie. En France, plusieurs milliers de personnes ignorent encore leur séropositivité et un quart des diagnostics du VIH est tardif. La participation infirmière au dépistage du VIH suscite un vif intérêt car elle permet d’élargir les acteurs impliqués. Nous avons suggéré à partir d’une revue de littérature systématique que, dans les pays à épidémie concentrée du VIH comme la France, les Etats Unis ou le Royaume Uni, les infirmiers pouvaient renforcer l’offre et la réalisation du dépistage du VIH. Dans ces pays, les autorités de santé ont préconisé un dépistage généralisé (non ciblé) du VIH dans les structures de santé non spécialisées. En raison de son efficacité modeste et de la charge de travail associée, cette recommandation a progressivement été remise en cause. L’étude DICI-VIH a été conçue pour explorer un nouveau mode, ciblé, de dépistage du VIH par les infirmiers dans les services d’urgences d’Ile-de-France, une région à forte prévalence. Nous avons montré, à l’aide d’un essai à large échelle randomisé en cluster et crossover en deux périodes, que ce mode de dépistage constitue une approche intéressante. En optimisant l’utilisation des ressources consacrées au dépistage dans les services d’urgences, cette stratégie ciblant un nombre limité de patients est réalisable et efficace pour identifier les personnes qui ne connaitraient pas leur séropositivité et, de plus, son coût est limité. Ces résultats fournissent de nouvelles données permettant l’adaptation des stratégies de dépistage du VIH proposées dans les régions où l’épidémie est concentrée. Ils participent également à l’évaluation de l’implication infirmière dans les programmes de dépistage. / Screening is essential to controlling the HIV epidemic as it ensures early treatment initiation, which limits morbidity, mortality and secondary transmission. In France, thousands of individuals remain unaware of their HIV status and a quarter of diagnoses occur at a late stage of infection. Nurse participation in HIV screening is widely discussed as a means by which to get a greater number of actors involved. Through a systematic review we suggested that, in countries with concentrated epidemics such as France, the United States and the United Kingdom, nurses could increase the number of screening tests being offered and performed. In these countries, Health authorities have recommended non-targeted systematic HIV screening in non-specialized health care settings. While this recommendation has gradually been questioned due to its limited effectiveness and associated increased workload, the DICI-VIH trial was designed to explore a new targeted method of HIV screening, performed by nurses in the Emergency Departments of the Paris metropolitan area, where HIV prevalence is high. A large-scale trial with a cluster-randomized two-period crossover design showed that this screening strategy is a compelling approach. By limiting the use of resources allocated to screening in Emergency Departments and by targeting a limited number of patients, this strategy is shown to be feasible and effective in identifying individuals who are unaware of their HIV status. It is also associated with limited costs. These results contribute new data to the discussion on how to best adapt HIV screening strategies in contexts where the epidemic is concentrated and to the evaluation of the contribution of the nursing profession to screening programs.

Large and ever-increasing numbers of chemicals, including large quantities of a broad spectrum of organic compounds are used in modern society. More than 30 000 of the more than 100 000 chemical substances registered in the EU are estimated to be daily used, of which many will be discharged into the waste-streams handled by municipal sewage treatment plants (STPs). The main objective of the work underlying this thesis was to improve understanding of the relationships between the characteristics of sewage contaminants and their sewage treatment efficiency. Further objectives were to examine the relationships between socio-economic uses of chemicals and sludge quality, and the effects of regulatory actions on sludge quality. The quality of the sewage sludge and the levels and distribution patterns of the sludge contaminants, both within and between the STPs, seem to remain quite constant over time. The overall findings indicate that the levels of contaminants in sewage sludge seem to be largely independent of the location, size and treatment techniques applied at the STPs, and generally, of the types of human activity connected to them. The total and relative concentrations of the sludge contaminants were found to be fairly constant on a dry weight basis, with some exceptions, indicating that the pollutants originate from broad usage and diffuse dispersion rather than (industrial) point sources. The proportion of cyclic methylsiloxanes recovered in sludge seems to strongly depend on their vapour pressure, which decreases with the number of siloxane units. The higher water solubility and biodegradability of organophosphorus flame retardants and plasticizers than polybrominated diphenyl ethers (PBDEs) were also reflected in lower percentages (relative to their national use) found in sludge. Significant time-trends in levels of a-third of the sludge contaminants included in the annual national measurement program were detected over a period of seven years. The levels of compounds displaying significant time-trends generally decreased following declines in the quantities used nationally. However, a quarter of these compounds showed increasing trends, of which the linear methylsiloxanes followed the same trend as used quantities. The decaBDE was also found to be increasing in sludge, probably as a result of the phase-out of pentaBDE and octaBDE. The results indicate that the STP removal efficiency of anthropogenic substances, in Sweden, is generally good and that STPs, at least those in cold climates, do not efficiently remove certain polar contaminants. A non-targeted screening (by use of environmetrics and GCxGC-TOFMS) was performed and found to fulfil the objective to assess the STP removal efficiency, with emphasis to systematically analyse which compound classes that are not efficiently removed using the current STP technology. Many polar aromatic compounds were identified to be poorly removed. The acquired data on levels, profiles and variations in sludge contaminants (sludge quality) extend both the available information and understanding of the degree and nature of sludge contamination, which should help attempts to track changes in its contaminants and revisions, if necessary, of guideline values. This thesis also contributes to improve the knowledge base for the development of future STP technologies, and that archived sewage sludge can be used in retrospective analysis of new and emerging pollutants.

Polycyclic aromatic hydrocarbons, derivates of polycyclic aromatic hydrocarbons, chromatography