Sylvie GADEYNE and
Patrick DEBOOSERE
1999-2006
Abstract:
Until the beginning of the 1990s, most studies on differential mortality in Belgium had to rely on geographic aggregates due to the lack of individual background variables connected to vital events in the registers. This situation has changed drastically thanks to the National Databank Mortality 1991-1996 (NDM). The NDM has been constructed in two phases. First, the 1991 census data, covering the total (legal) population in Belgium, have been linked to survival data from the National Register, indicating whether the respondent died or emigrated between the 1st March 1991 and the 1st March 1996 and the exact date of each event. In a second stage, information on the cause of death has been included for all respondents who died between the 1st March 1991 and the 31st December 1995. The construction of the NDM permits several interesting analyses. In a first stage of inequality research, inequalities have been studied in all-cause mortality among middle-aged men and women as a function of socio-economic and socio-demographic characteristics (educational level, activity status, type of income, professional class, sector of activity, housing tenure, housing quality and household position) and region of residence. In a second stage, life tables have been constructed by educational level and region, which have then been combined, at the Institute of Public Health, with health status data in order to calculate disability free life expectancy. In a third stage, inequalities in cause-specific mortality have been investigated in the population aged 12-17 and 15-29. The inclusion of cause-specific data has also allowed the participation to the European Working Group on Socio-economic determinants of Healthy Ageing. In this project, all-cause and cause-specific mortality differentials by educational level and housing tenure are compared among middle-aged and old-aged men and women in different European countries. Finally, also inequalities in all-cause and cause-specific mortality by national origin have been investigated. The doctoral research of S.Gadeyne has concentrated on socio-economic inequalities in all-cause and cause-specific mortality among Belgian youngsters and young adults (aged 15-29), adults (30-44), middle aged (45-59) and elderly persons (60-74 and 75+). Relative as well as absolute inequalities have been studied.
Most important results:
The construction of the National Databank Mortality 1991-1996 has lead to a Belgian empirical contribution to the debate concerning social inequality in mortality. Research results clearly confirm the international patterns in Belgium, with the privileged classes showing the lowest mortality risks, the deprived classes the highest mortality risks and in-between levels for the middle classes. In the middle-aged population (40-64), all socio-economic variables generate significant effects, not only in gross models, but also in net models, including age and several socio-economic variables simultaneously. Regional differences are striking too in this age interval, the French-speaking part of the country having the highest mortality risks, the Dutch-speaking part the lowest. After control for educational level, type of income, housing and household position, regional differences in mortality remain significant. Life tables by educational level and region show a difference in life expectancy of 12.4 years between a highly educated woman aged 25 in the Flemish Region and a low educated man aged 25 in the Walloon Region. Differences in healthy life expectancy are even larger. Cause-specific analyses confirm inequalities for practically all causes of death. At young age, inequalities are significant not only for exogenous mortality but also for endogenous mortality. By national origin, the low mortality of most migrant groups can be explained by man-made causes of death such as alcohol abuse, tobacco consumption, traffic accidents and suicide. In the doctoral research project of Sylvie Gadeyne, Poisson regressions confirm the internationally observed patterns with clear differences in all-cause and cause-specific mortality rates according to several dimensions of socio-economic position at all age intervals, from young age (15-29) to old age (75 and over). A decomposition of inequalities in cause-specific contributions shows that traffic accidents have the largest contribution to mortality differentials at young age, neoplasms at adult age and cardiovascular diseases and respiratory diseases at older age. Relative inequalities are significant for practically all cause groups. The patterns of course differ by age group and by socio-economic dimension. All these research results open pathways for interpretation of how social gradients in health are shaped for men and women in the different stages of life. Inequalities are not limited to the most deprived classes having the highest mortality risks, but consist of a clear gradient. The patterns suggest that health policies should focus on different dimensions of socio-economic position (education, housing, employment, profession) and that not only the most marginalised or the most deprived population groups should be targeted, but also the intermediate middle classes. Specific subgroups have shown to be vulnerable for specific causes of death, another factor which should be taken into consideration in health policies.
< BACK