Body Fat Predicts Mortality Better Than BMI

Body mass index (BMI) may not accurately reflect a person's body composition or be a good indicator of health, according to a new study.

Researchers at the University of Manitoba (Winnipeg, Canada) conducted an observational study in 54,420 adults (49,476 women, 4,944 men) in their mid-60s to explore the associations of BMI and body fat percentage with mortality. All study participants underwent dual X-ray absorptiometry (DXA) scans, entered a clinical registry, and were then followed using linked administrative databases. The patients were then divided into quintiles based on BMI and DXA-derived body fat percentage, with quintile 3 used as reference.

During the study period 4,965 of the women and 984 of the men died. The results revealed that men and women with high body fat percentage—defined as greater than 38% in women and greater than 36% in men—were more likely to die over the next four to seven years. The results also showed that counterintuitively people with a lower BMI faced a greater risk of death, probably due to low muscle mass or frailty associated with chronic disease. The study was published on March 8, 2016, in Annals of Internal Medicine.

“In our society there's been this mantra that thin is ‘in’ and being heavy is ‘bad.’ Someone with a lot of muscle mass, for example, may have a high BMI and, technically, fall into the overweight category,” said lead author William Leslie, MD. “The relationship between body size and health is more nuanced than the number on your bathroom scale. It's important to be attuned to what you're made of, rather than just how much you weigh.”

BMI is a value defined as body mass divided by the square of body height, expressed in units of kg/m². It was developed as an attempt to quantify the amount of tissue mass (muscle, fat, and bone) in an individual, and then categorize people as underweight, normal weight, overweight, or obese based on that value. However, there is some debate regarding where on the BMI scale the dividing lines between categories should be placed, while recent studies claim that other measures, such as waist-to-height ratio (WHtR) and body fat, are better predictors of health.

Related Links:
University of Manitoba