High body fat percentage, low BMI both independent indicators of increased mortality risk, study finds

High body fat percentage and low body mass index (BMI) were independently associated with higher all-cause mortality in a recent study.

Researchers in Canada performed an observational study of adults 40 years of age and older who were referred for bone mineral density (BMD) testing to evaluate the association between BMI and body fat percentage and mortality. BMI and body fat percentage were examined separately and together. All study participants had dual-energy X-ray absorptiometry (DXA) and were entered in a clinical BMD registry. Follow-up was performed using linked administrative databases. In adjusted, sex-stratified Cox models, BMI and body fat percentage as determined by DXA were divided into quintiles, with 1 being the lowest, 5 being the highest, and 3 being the reference. The study results were published online March 8 by Annals of Internal Medicine.

The final study cohort included 49,476 women (mean age, 63.5 years; mean BMI, 27.0 kg/m²; mean body fat percentage, 32.1%) and 4,944 men (mean age, 65.5 years; mean BMI, 27.4 kg/m²; mean body fat percentage, 29.5%). Among women, according to the World Health Organization's BMI classification, 1,011 (2.0%) were underweight, 18,933 (38.3%) were of normal weight, 17,087 (34.5%) were overweight, 8,358 (16.9%) had class I obesity, and 4,087 (8.3%) had class II obesity. Among men, according to the World Health Organization's BMI classification, 62 (1.3%) were underweight, 1,427 (28.9%) were of normal weight, 2,246 (45.4%) were overweight, 904 (18.3%) had class I obesity, and 305 (6.3%) had class II obesity. Over a median of 6.7 years, 4,965 women died, while 984 men died over a median of 4.5 years. Higher BMI quintile was associated with lower risk for death in both women and men (P<0.001 for trend).

In models that examined BMI and body fat percentage separately, low BMI and low body fat percentage were separately associated with increased risk for death in women, while high BMI quintile or body fat percentage was not. For men, those in the lowest BMI quintile and those in the highest body fat percentage quintile had significantly higher mortality. In models that examined BMI and body fat percentage together, low BMI was significantly associated with increased mortality in women in fully adjusted models (hazard ratio [HR] for quintile 1, 1.44 [95% CI, 1.30 to 1.59]; HR for quintile 2, 1.12 [95% CI, 1.02 to 1.32]), as was high body fat percentage (HR for quintile 5, 1.19 [95% CI, 1.08 to 1.12]). Similar associations were seen in men for low BMI (HR for quintile 1, 1.45 [95% CI, 1.17 to 1.79]) and high body fat percentage (HR for quintile 5, 1.59 [95% CI, 1.28 to 1.96]).

The authors noted that the study results may not be widely generalizable, that they did not use serial measures of BMD and weight, and that data on such measures as physical activity and smoking were not available, among other limitations. However, they concluded that an independent association exists between low BMI and high body fat percentage and increased mortality and that this “may help explain the counterintuitive relationship between BMI and mortality.” Their findings suggest that direct measures of adiposity should be used in prognostic and perhaps exploratory models, they wrote.