Educational level and longevity

Abstract

Educational level is known as a strong predictor of mortality before the age of 65. The relationships between education and early mortality from coronary heart disease, cancer and accidents consistently show an inverse pattern. More surprising is the association of mortality and education in AIDS patients. The effect of education on early mortality has several interpretations; education is considered to be an indicator of socio-economic resources and a predictor of system integrity, healthy behaviours and entry to safer environments. After the age of 65, the effect of education is less well known. With the Paquid cohort we have shown that educational level remains a risk factor for death, independently of cognitive decline or dementia. Before the age of 65, inequalities in mortality related to socio-economic status are a generalised phenomenon in the industrialised world (Leclerc et al. 1990). In each country where data are available, death rates,have been found to be higher in groups with lower occupational status, lower educational level, or lower income level (Kunst and Mackenbach 1994). However, the strength of the association between educational level and longevity is strongly related to the country of residence. Kunst and Mackenbach (1994) compared the death rates in men in several countries according to educational level (Table 1). In this analysis, inequalities were estimated by the proportional mortality increase moving from the top to the bottom of education. A total inequality estimate of 0.36 in subjects aged 55 years and older in Sweden, which is the lowest in this table, implies, according to the fitted regression equation, that death rates estimated for. those at the bottom of the Swedish educational hierarchy are 36 % higher than the death rates estimated for those at the top. Inequalities in mortality are relatively small in the Netherlands and Scandinavia and more pronounced in the United States and France, while United Kingdom occupies intermediate position. The large inequalities in mortality in the US and France were attributed in part to large inequalities in education in these countries. Indeed, in the collaborative Eurodem study (Letenneur et al. 2000), the level of education appeared to be more variable in France than in the United Kingdom or the Netherlands. Thus the inequalities were larger in France and the power of the analysis to detect differences in mortality according to educational categories would be greater.