Séance comparaisons européennes

14H : The Washington Group on disability statistics

Proposing a worldwide measure of disability par Emmanuelle Cambois (INED)  

Improving comparability with cognitive tests par Claire Scodellaro, Elena de Palma (ISTAT, Rome) - Discutante : Chantal Cases (Ined)

Producing health statistics at the national level and comparable data at the international level has become a challenge for most states, in order to identify both people with disabilities and risk factors for disabilities and impairments at the individual and group levels. One major issue is the comparability of indicators over time and space. One solution is to build indicators thanks to harmonised sets of questions that can be used in every country. The purpose of the Washington Group on Disability Statistics (UN) is to elaborate such sets of questions through cognitive tests and field tests all over the world.

The aim of this paper is to present the experience of the Granada Group in carrying out a cognitive test in Europe and USA on a set of questions on health conditions/disabilities, to be used in population surveys. This working group was created in February 2010 for this specific task, in cooperation between the Washington Group on Disability Statistics (UN) and the Budapest Initiative on Health Status Measurement (Eurostat-WHO-UNECE).

The cognitive test is a pre-test technique aimed at improving the survey instrument, and consequently, data quality. If applied to cross-national surveys, it can be a useful means to improve international comparability. Cognitive interviewing consists in administering draft survey questions while collecting additional verbal information about the responses, which is used to evaluate the quality of the response and help to understand whether questions are actually generating the information expected.

Through the analysis of cognitive interviews it is possible to identify potential response errors and interpretation patterns, and provide an insight into social-cultural factors having an impact on the response process.

This paper will consist of three main sections: 1) problems in comparing data in health interview survey; 2) overview of main characteristics of cognitive testing methods, 3) description of the Cognitive test conducted by the Granada Group for Budapest Initiative/Washington Group.

15H15 : International comparability of long-term cause-of-death statistics: experience from France, West Germany and the Czech Republic par Marketa Pechholdova (Ined)

The use of long-term reliable data on mortality by cause of death is the only way to fully understand mortality trends and bring more clarification to the theory of health transition. When dealing with cause-of-death statistics, researchers are often discouraged by periodic revisions of the classification of diseases, which disables comparability in time, and by the issues of different interpretation of the classification between countries, which affects comparability in space. My contribution tries to deal with both of these by using existing reconstructed time series for all the three countries, updating them to the most recent tenth revision of the classification, and analysing the observed problems with comparability. Finally, a solution to avoid as much bias as possible is proposed in a form of analytical shortlist and compare