[The relationship between educational level, physical activity and mortality.]

The relationship between educational level and mortality is well known. This has been shown in the Reykjavik Study and was only partly accounted for by unequal distribution of known risk factors. The objective of the present study was to explore the relationship between educational level and physical activity and whether that relationship could partly explain differences in mortality.

This is a part of the Reykjavik Study. Presented is data from 18,912 participants, divided into four groups by educational level. Physical activity was assessed by questionnaire. The relationship between physical activity and educational level was assessed by logistic regression and between mortality and educational level by Cox regression analysis. Adjustments were made for age, year of examination, known risk factors (serum lipids, blood pressure, height, weight, smoking, use of anti-hyertensive drugs and 90 min glucose tolerance) and physical activity.

There was a positive relationship between physical activity and educational level (p<0.001). By adding adjustments for physical activity to a multiple regression analysis containing other known risk factors the relationship between total mortality and educational level was reduced. For highest versus lowest educational group hazard ratio was elevated from 0.77 to 0.80 for men and from 0.91 to 0.93 for women. Same trend existed for cardiovascular mortality and to a less extent for cancer mortality.

The association between educational level and mortality can be partly explained by differences in leisure-time physical activity. In spite of adjustments for known risk factors and physical activity there remains a statistically significant relationship between educational level and mortality.