The aim of this study was to describe cohort differences in health indicators among four birth cohorts of 70-year old men and women from Göteborg, Sweden, born in 1901/2, 1905/6, 1911/12, and 1922. With special reference to gender, education, and obesity, it is hypothesized that changes in health among elderly men and women may not be occurring in a uniform manner. The variables studied were: systolic and diastolic blood pressures, triglycerides, cholesterol, height, weight, body mass index, waist-hip ratio, physical inactivity, current smoking, and alcohol consumption, plus selected prevalent diseases. Logistic and linear regression models were used to test for secular trends and effect modification by gender. Most trends in metabolic and lifestyle indicators varied in relation to gender as well as education. For instance, later-born male cohorts were more overweight than earlier-born groups while the later-born female cohorts had similar relative weights but a more centralized fat patterning. These cohort differences further varied by education where later-born men with less education and later-born women with higher education tended to be more overweight, compared to earlier-born cohorts. Finally, significant cohort differences in previously diagnosed myocardial infarction, stroke, and diabetes mellitus at age 70 were observed in men only. Interaction terms revealed that the gender difference was statistically significant only in the case of diabetes mellitus. In conclusion, secular trends in many obesity-related health indicators among 70-year old Swedish cohorts were dependent on both gender and socio-economic factors.