Seniors jogging on a forest road

There is a staggering amount of scientific evidence linking excess body weight and mortality risk in various populations around the globe. Most of these studies have relied on the use of the body mass index (BMI, the body weight for a given height) to assess obesity associated mortality risk. Although total body fat is well correlated with BMI, a high BMI is not always a reflect of a high fat mass as BMI also takes into account the amount of fat-free mass, which is mostly driven by muscle and bone mass. As aging is associated with a steady decline in fat-free mass, the relationship between BMI and mortality risk may be influenced by body composition, and perhaps even more so in the elderly. In opposition to the number of scientific studies linking BMI to mortality risk, few high-quality studies have documented the impact of body composition on mortality risk in the elderly.

In the latest issue of the American Journal of Clinical Nutrition, researchers from the Geneva University Hospital published the results of a thorough investigation of the mortality risk associated with an elevated BMI, fat mass or fat-free mass in subjects older than 65 years ( The study included close to 3200 men and women who were recruited between 1990 and 2011. Their body composition was assessed using bioelectrical impedance analysis. During the course of the study, 1007 men and 766 women died at a average age of 79 and 83 years, respectively in men and women, which corresponds to normal life expectancy in Europe. Results of the study suggest that only men in the very low BMI category (BMI<18,5 kg/m2) appeared to be at increased mortality risk compared to the other subgroups. There was also no relationship between fat mass and mortality risk. However, men who had the highest amount of fat-free mass were at considerably reduced mortality risk, more than 20% compared to those who had the lowest amount. With regard to women, there was no relationship whatsoever between BMI, fat mass or fat-free mass with mortality risk. Whether this finding is attributable to the fact that fewer women and fewer women cases were included in the study or because the relationship between body composition and mortality risk is different in men vs. women needs to be confirmed in additional studies. This prospective study has some limitations. For instance, physical activity levels were not recorded in participants of this study. There is good evidence suggesting that individuals who are physically active are most likely to have a higher muscle mass or at least a slower decline in muscle mass as they age. They are also likely to have healthier bones. Therefore, physical activity levels may represent a major confounding factor between a healthy body composition and a reduced mortality risk. In other words, people with elevated fat-free mass could be at reduced mortality risk because they are physically active. Waist circumference could also represent another factor in this equation. Indeed, studies have shown that in the elderly, waist circumference, but not BMI, predicts cardiovascular risk. Altogether, results of this study strongly support that interventions aiming at decreasing body weight loss should not be recommended to individuals >65 years. Interventions aiming at increasing fat-free mass such as strength training should therefore be considered for this population. Whether this would lead to a reduction in mortality risk still needs to be determined by future studies but in the meantime, older individuals could obtain more benefits if they are advised to perform more physical activity rather than simply being told to lose weight.