|Ressource Center for Biostatistics, Epidemiology and Pharmacoepidemiology applied to Infectious Diseases - UMR 657 INSERM|
|Director : GUILLEMOT Didier (email@example.com)|
CeRBEP was created January 2004, 1st. The main activities are (i) pharmacoepidemiological research focused on the influence of population antiinfectives exposure on the infectious risk (CeRBEP is a component of the INSERM unit U657 "Pharmacoepidemiology and evaluation of the impact of health products on population"), (ii) advice, methodological support and development of new tools for epidemiological research in infectious diseases.
Our research is focused on the interaction of population exposure to antiinfectives with infectious risk by combiningepidemiological investigation, statistical and mathematical modeling (population-dynamical) approaches . Main research interest are :
1 Spatial and time analysis of the impact of antibiotic and vaccine population exposure on the infection/infectious risk in the community. We intend to quantify the impact of the decrease of the antibiotic use as well as the increase of the conjugated vaccine vaccination on Streptococcus pneumoniae antibiotic resistance. C. Bernède, R. Mansuy, J. David and D. Guillemot are involved in this research area which is developped in collaboration with the national health insurances (CNAMTS and CANAM) and the National Reference Center of Pneumococcus (L. Gutmann and E. Varon)
2 Epidemiological analysis of the selective impact of antibiotic use on bacterial resistance in context of exposure with multiple antibiotic classes and multiresistance, taking into account the environmental exposure to bacterial resistance and the influence of the dosage of antibiotic in function of mecanisms of resistance. We currently work on S. pneumoniae and on enterobacteria. C. Bernède, C. Toneatti, J. Salomon and D. Guillemot are involved in this research that is based on data from the AUBEPPIN study and the ColoRea study. The latter study will be implemented in 2005 in collaboration with B. Schlemmer, A Andremont, G. Arlet and JP Sollet (GHU nord) and with 10 intensive care units in Ile de France.
3 Mathematical modelization of population dynamics of bacterial colonization and transmission in interaction with population exposure with drugs (including vaccines). A central point of interest in this mathematical project wheither widespread antibiotic conjugate vaccines use in the human population could result in the emergence of non-vaccine resistant serotypes and the control of bacterial resistance taking into account multiresistance and multiantibiotic exposure. L. Oppatowski, R. Mansuy, J. Bullet and D. Guillemot are involved in this topic developped in collaboration with the INSERM Unit U707 (PY Boelle and L. Temime)
Further developments. We are currently to develop new research focused on Staphylococcus aureus and on the possible reemergence of bacterial diseases in the community in relation with the decrease of use of antibiotics.
We also support the epidemiological research initiated by the National Reference Centers Advice. We provide methodological advices when needed, we try to develop CNR information systems enabling linkage with other data sources related to population health in France, and we promote the application of new statistical methods.
1 Support of epidemiological research and information systems
Since January 2004, 1st, we have collaborations with the following National Reference Center (CNR): CNR des bactéries anaérobies et du botulisme, CNR des Antibiotiques, CNR de la coqueluche et autres bordetelloses, CNR des Borrelia, CNR du virus influenzae (France Nord), CNR des Méningocoques, CNR des Mycobactéries, CNR des Pneumocoques (Hôpital Européen Georges Pompidou) et le Centre d'essais vaccinaux Pasteur-Cochin.
Furthermore, we collaborate with the International Network of the Pasteur Institutes on a project intended to survey the emergence and the spread of bacterial resistance in developing countries. In this project we specifically work on the quality control process and on the consequences of bacterial resistance in terms of morbidity and mortality.
We also participate to the following French public health committees : Comité Technique des Vaccinations du Conseil Supérieur d'Hygiène Publique de France, à la Commission de Transparence et au groupe " Evaluation de l'Intérêt de Santé Publique des Médicaments " ainsi qu'au Comité Technique National des Infections Liées aux Soins
In collaboration with the "Logiciels et banques de données" team of the Pasteur Institute we have a current project which intend to develop and experiment web data recording applied to pharmacoepidemiological research.
2 - Improvement of statistical methods to assess the risk of antibiotic resistance associated with antibiotic use
The epidemiological relationship between a drug exposure and the bacterial colonization (or infection) is usually estimated by a risk ratio. It generally requires a case-control or a cohort design, with a case-control analysis which implies difficulties on the definition of control subjects and risk of biases. These drawbacks can be overcome by case-series studies including just cases. In such studies, a relative risk between the event and the exposure can be estimated from data on case exposures, which can either be retrospectively recorded during a fixed period of observation or identified by linking cases with drug consumption databases. The aim of this project is to adapt a recently developed statistical method to estimate the relative risk using only cases data. The principle is to consider each case as its own control.
Applied to bacterial resistance and antibiotic use, we propose a conditional relative risk ("RRc") for evaluating the relationship between antibiotic use and colonization by antibiotic-resistant relative to antibiotic-susceptible bacteria. The RRc has an interpretation similar to the ORc and it can be used either in cohort or case-series. The proposed method is an extension of an approach proposed recently by Farrington which allows to estimate a relative risk between a single recurrent event according to exposure status in the case-series studies. Farrington's method is identical to another procedure used in cohort studies to estimate the relative risk based on the exposure histories of only cases. This research in developed in collaboration with the INSERM Unit 471 (T. Moreau and M. Hocine)
Figure 1. Simulated changes with time in the distribution of resistance levels in the meningococcal population, starting from a situation close to that of France in 2001, under (a) constant antibiotic treatment conditions (1 treatment/3 y) and (b) a frequency of treatment reduced by half (1 treatment/6 y)
Figure 2. Evolution of antibiotic use in France since 2001, according to age
Keywords: Infectious diseases, drug, biostatistic, pharmacoepidemiology, populations
|Publications 2004 of the unit on Pasteur's references database|
|Office staff||Researchers||Scientific trainees||Other personnel|
|Roger MANSUY PhD Thesis student firstname.lastname@example.org
Jérôme SALOMON PhD Thesis student email@example.com
Yannick LE GLEAU Training fellow IUT
Julie DAVID Training fellow INAPG
Christine TONEATTI Ing. Etude INSERM firstname.lastname@example.org
Other personnel :
Claire BERNEDE CAT « A » IP email@example.com
Patricia MARTEL Ing (during pregnancy C. BERNEDE)
Lulla OPATOWSKI CAT « A » INSERM firstname.lastname@example.org
Anne DEMOND Secret. Dir. IP (50 %) email@example.com