Send to

Choose Destination
Eur J Epidemiol. 2007;22(10):685-90. Epub 2007 Aug 3.

Widening of a social gradient in obesity risk? German national health surveys 1990 and 1998.

Author information

Institute of Biometrics and Epidemiology, German Diabetes Center, Leibniz Institute at Heinrich Heine University Düsseldorf, Auf'm Hennekamp 65, 40225, Dusseldorf, Germany.



Whether differences in obesity prevalences across social status levels have widened remains controversial.


We used German national health surveys (1990-1992 and 1998, n = 7,466 and 5,583, age 25-69 years) to estimate obesity prevalences and its associations with calendar year, age (25-39, 40-60, and 61-69), and educational level (low, middle, and high), as well as an interaction term (year x educational level) in men and women. We used multiple regression models, considering the sample design.


Obesity prevalence in 1990 and 1998 was 18.1 (95% CI 16.5-19.7) and 19.9 (18.2-21.6) in men and 20.9 (19.2-22.6) and 21.6 (19.3-23.7) in women, with statistically significantly higher prevalences in higher age and lower education. A statistically significant increase of obesity prevalence was present only in men after adjustment for age and education. The increase seems to be highest in high-educated subjects. However, interaction was not statistically significant, except in middle compared to high-educated men (OR 0.67; 0.47-0.96).


Obesity prevalence increased only moderately in Germany between 1990-1992 and 1998. There was a tendency of reduction of the social gradient in obesity instead of a widening.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center