Body mass index and old-age survival: a comparative study between the Union Army Records and the NHANES-I Epidemiological Follow-Up Sample.

On the basis of a historical review of the BMI trend among white male Americans since the late 19th century, this paper investigates the association between BMI measured at age 50-59 years and survival in the subsequent 18 years of follow-up through a comparative study between the Union Army Records (n = 1,238) and the 1971-1975 National Health and Nutritional Examination Survey Epidemiological Follow-up Sample (n = 861). A descriptive analysis of the trend in BMI suggests that the age-adjusted average BMI among adult white males has increased from 22.8 in the late 19th century to 28.0 in the year of 2000. The corresponding increase in the age-adjusted percentage of obesity is from 2.5% to 28.2%. Given the significant increase in BMI, it becomes important to evaluate how the BMI-mortality association has changed since the 19th century, and what implications these changes might have to mortality in the future. The results from the Cox proportional hazard analysis indicate that as the average BMI at population level increases, the optimal BMI in terms of survival increases, rather than stable. With the whole American population moving into a higher BMI regime, the optimal BMI in terms of old-age survival for white male Americans has experienced a substantial upward shift, from 20.6-23.6 in the late 19th century to 22.7-27.3 in the early 1970s. Mortality differentials across BMI quintiles have become more salient since the late 19th century. There is still good potential for old-age mortality to further decline. However, to what extent this potential can be developed will partially depend on the future trend in the prevalence of obesity.