Personal tools
You are here: Home Recherche Nos projets de recherche Projet CENSUR
Document Actions

Projet CENSUR

by Roch GIORGI last modified 2012-11-17 13:30

CENSUR: Challenges in the Estimation of Net SURvival

Ce travail a bénéficié d'une aide de l'Agence Nationale de la Recherche dans le cadre de l'appel à projets programme Blanc 2012 (ANR-12-BSV1-0028).

Contact : Roch GIORGI  

Context

Estimation of survival is used in many medical studies that aimed to estimate the prognostic of patient, to estimate the impact of some variables on the disease under study. More generally, estimation of survival is a valuable indicator of progress in disease control. For chronic diseases, and more especially for cancer, the creation of registries has permitted to increase the knowledge in the epidemiology of the diseases under study. Over the past decade, population-based cancer registry data have been used increasingly worldwide to evaluate and improve the quality of cancer care. Depending on the country, they are organised at a local geographical level (an administrative area) or at a national level. For example, in 2009 in France there is a network of 22 registries, which are grouped in the FRANCIM network (Association of the French Cancer Registries). A lot of efforts have been made to organize and to produce a set of indicators from population-based registries, allowing for national and international comparisons (one can cited, FRANCIM, EUROCARE project, CONCORD project, the US SEER program,...). It is difficult to quantify the number of articles published in peer-review journals, reports, and books, presenting results of analyses based on registry data. More and more, these results are used by the appropriate national health institutes for decision making in public health.

In the context of international comparisons, net survival represents a key indicator. Net survival is the survival that would be observed in absence of mortality due to other causes. Since it is unaffected by changes in mortality of other diseases, net survival is the only survival indicator that can be used for cross countries comparisons or trends analysis.

The statistical methods used to estimate net survival from population-based registries have been developed for population-based cancer registries, and rely mainly on the relative survival approach. However, the models used to estimate relative survival at different times since diagnosis and to incorporate prognostic covariates may vary between research centres. The conclusions from such studies may rely on the methods used to analyse the data and observed differences could be attributed to methodological and statistical approaches or to real effects. This fact is more significant since recent works, in which we have contributed with our previous project, MESURE, have noted that the standard relative survival ratio estimators do not provide information on cancer mortality that is independent of the national general population mortality. Thus they are not suitable for comparison between countries.

Thus, it is still of great interest, for both the scientist and the public health decision maker, to assess and to explore the influence of different methodological options, and to develop new approaches to answer some important problems that are still waiting for appropriate solutions (e.g. evaluation of model adequacy, measurement bias, competing event’s,...).
 

Rationale

The overall aim of this project is to improve the current methods for estimating net survival and to broaden their field of application in order to obtain i) tools to model complex data, and ii) more accurate estimates that enable to have information on survival for a studied disease and on its public health impact. More precisely, there are three main research axes devoted to: (1) propose new methodological developments to answer questions that are the result of our works during our previous project (MESURE); (2) extend and assess new statistical methods; (3) transfer net survival methods used in cancer to some other specific applications.

These themes correspond to some of the current challenges in the estimation of net survival. Following our previous project, CENSUR project is more ambitious, considering the scope of the methods investigated and the new development that are envisaged. While the focus is on methodological aspects, the network implies also members that have skills in epidemiology and in population-based data analyzes with the objective to produce survival statistics useful in Public Health. At the end of this project, in order to optimize the use of methods to estimate net survival, we will organize a course. Furthermore, free-licensed statistical programs derived from our work will be available for the scientific community.
 

Scientific program

To achieve our objectives, the scientific program is divided into four work packages (WP):
  • WP1: New developments directly related to the results obtained in the previous project MESURE
  • WP2: Extension and assessment of statistical methods to estimate net survival
  • WP3: Methodological developments for specific applications
  • WP4: Diffusion
 

Partners of the project

  • Sciences Economiques et Sociales de la Santé et Traitement de l'Information Médicale, UMR 912 INSERM/IRD/Aix-Marseille Université, Marseille, France - site web
  • Service de Biostatistique des Hospices Civils de Lyon, Lyon – UMR CNRS 5558 et Université Claude Bernard Lyon I, Lyon, France
  • Service de Biostatistique et Informatique Médicale, Dijon, France – INSERM Equipe 5 U866
  • Département d'Epidémiologie en Entreprise, Institut national de recherche et de sécurité; Nancy, France

External members

  • Réseau FRANCIM : Réseau des registres français des cancers, France
  • Non communicable Disease Epidemiology Unit, London School of Hygiene and Tropical Medicine, London, United-Kingdom -  site web 
  • Centro Nazionale di Epidemiologia, Instituto Superior di Sanita, Roma, Italy - site web 
  • Department of Epidemiology and Biostatistics, McGill university, Montreal, Canada - site web 
  • Institute for Biostatistics and Medical Informatics, Faculty of Medicine, Ljubljana, Slovenia - site web 
Powered by Plone, the Open Source Content Management System

Ce site utilise le système de gestion de contenu Plone