Both a low body mass index and a high percentage of body fat are independently associated with an increased risk of all-cause mortality in middle-aged and older adults, according to an observational study reported online in the Annals of Internal Medicine.
The finding that these somewhat contradictory measures are both associated with higher mortality helps account for the so-called obesity paradox – the fact that mortality is lower in overweight, mildly obese, and moderately obese persons than in those who are underweight or of low-normal weight, said Dr. Raj Padwal of the University of Alberta, Edmonton, and his associates.
©wragg/iStockphoto.com
The investigators assessed the relationships among body mass index, body fat, and mortality by analyzing data from a population-based cohort study of 54,420 men and women aged 40 and older. These study participants underwent assessment of body composition as part of a study of bone mineral density, then were followed for a median of 4-7 years. BMI and body fat percentage were categorized into quintiles; men and women were analyzed separately.
Among women (mean age, 63.5 years), 2% were underweight, 38% were of normal weight, 34% were overweight, 17% had class I obesity, and 8% had class II or III obesity. The mean percentage of body fat was 32%. There were 4,965 deaths during follow-up. Among men (mean age, 65.5 years), 1% were underweight, 29% were of normal weight, 45% were overweight, 18% had class I obesity, and 6% had class II or III obesity. The mean percentage of body fat was 30%. There were 984 deaths during follow-up.
Higher BMI correlated with decreased mortality in both sexes. Among women, death rates declined from 18.6/1,000 person-years in the lowest BMI quintile to 13.9/1,000 person-years in the highest BMI quintile. Among men, death rates declined from 51.5/1,000 person-years in the lowest BMI quintile to 32.7/1,000 person-years in the highest BMI quintile.
In contrast, higher body fat percentage correlated with increased mortality. Among women, a high percentage of body fat was associated with significantly higher mortality (hazard ratio, 1.19), as it was among men (HR, 1.59). “Our results suggest that BMI may be an inappropriate surrogate for adiposity, and this limitation may explain the presence of the obesity paradox in many studies,” Dr. Padwal and his associates reported (Ann Intern Med. 2016 March 8. doi: 10.7326/M15-1181).
This study was limited in that the study population was predominantly female and white, and it may have included more “health-seeking” and lower-weight individuals than there are in the general population, the investigators added.