Background: Obesity and its health consequences will dominate health care systems in many countries during the next decades. However, the body mass index (BMI) optimum in relation to all-cause mortality is still a matter of debate.
Material and methods: Data of the Vorarlberg Health Monitoring & Prevention Program (VHM&PP, 1985-2005) and data provided by the Main Association of Austrian Social Security Institutions (MAASSI, 2005-2015) were analyzed. Information was available on age, sex, smoking status, measured height and weight, and mortality. Generalized additive models were used to model mortality as a function of BMI, calendar time, age, and follow-up.
Results: In MAASSI (N = 282,216, 46.0% men), men and women were on average 2.7 years older than in VHM&PP (N = 185,361, 46.1% men). Average BMI was slightly higher in men (26.1 vs 25.7 kg/m2) but not in women (24.6 vs 24.7 kg/m2). We found an interactive effect of age and follow-up on the BMI optimum. Over age 35 years in men and 55 years in women, the BMI optimum decreased with length of follow-up. While keeping covariates fixed, BMI optimum increased slightly between 1985 and 2015 in men and women, 24.9 (95% CI, 23.9-25.9) to 26.4 (95% CI, 25.3-27.3), and 22.4 (95% CI, 21.7-23.1) to 23.3 (95% CI, 22.6-24.5) kg/m2, respectively.
Conclusion: Age and length of follow-up have a pronounced effect on the BMI associated with the lowest all-cause mortality. After controlling for age and length of follow-up, the BMI optimum increased slightly over 30 years in this large study sample.
Keywords: BMI; age; length of follow-up; mortality; secular trend.