Cardiorespiratory fitness and the paradoxical BMI-mortality risk association in male veterans

Mayo Clin Proc. 2014 Jun;89(6):754-62. doi: 10.1016/j.mayocp.2014.01.029.


Objective: To assess the effect of fitness status on the paradoxical body mass index (BMI)-mortality risk association.

Patients and methods: From February 1, 1986, through December 30, 2011, we assessed fitness and BMI in 18,033 male veterans (mean age, 58.4 ± 11.4 years) in 2 Veterans Affairs Medical centers. We established 3 fitness categories on the basis of peak metabolic equivalents achieved during an exercise test as well as 5 BMI categories. The primary outcome was all-cause mortality.

Results: During the follow-up period (median, 10.8 years, comprising a total of 207,168 person-years), 5070 participants (28%) died. After adjusting for age, risk factors, muscle-wasting diseases, medications, and year of entry, mortality risk was higher for individuals with a BMI of 20.1 to 23.9 kg/m(2) (hazard ratio [HR], 1.21; 95% CI, 1.12-1.30) and 18.5 to 20.0 kg/m(2) (HR, 1.56; 95% CI, 1.37-1.77) than for those with a BMI of 24.0 to 27.9 kg/m(2); mortality risk was not increased for those with a BMI of 28.0 kg/m(2) or greater. When stratified by fitness, the trend was similar for low-fit and moderate-fit individuals. However, mortality risk was not increased for high-fit individuals across BMI categories. When fitness status was considered within each BMI category, mortality risk increased progressively with decreased fitness and was more pronounced for moderate-fit (HR, 2.52; 95% CI, 2.06-3.08) and low-fit (HR, 2.48; 95% CI, 2.0-3.06) individuals with a BMI of 18.5-20.0 kg/m(2). Mortality risk was not significantly increased for high-fit individuals (HR, 1.17; 95% CI, 0.78-1.78; P=.45).

Conclusion: A high mortality risk associated with low BMI levels was observed only in moderate-fit and low-fit individuals, and not in high-fit individuals. Thus, fitness greatly affects the paradoxical BMI-mortality risk association. Furthermore, our findings indicate that lower BMI levels do not increase the risk for premature death as long as they are associated with high fitness. Thus, the paradoxically higher mortality risk observed with lower body weight as represented by lower BMI is likely the result of unhealthy reduction in body weight and, perhaps most importantly, considerable loss of lean body mass.

MeSH terms

  • Body Mass Index*
  • Exercise Test / statistics & numerical data
  • Humans
  • Male
  • Middle Aged
  • Mortality*
  • Obesity / mortality
  • Physical Fitness*
  • Proportional Hazards Models
  • Risk Factors
  • United States / epidemiology
  • Veterans / statistics & numerical data*