Should older obese people try to lose weight? Such a simple question is more complicated than one would think.
At issue is whether obesity is harmful in older people, and whether treating it will reduce their health risks. True, obesity is an independent risk factor for cardiovascular disease and is associated with many comorbidities, including type 2 diabetes mellitus, hyperlipidemia, heart failure, and hypertension.1 An independent association also exists between obesity and all-cause mortality.2 However, there is also evidence suggesting that obesity in this age group is associated with a lower, not higher, risk of death—a finding termed the obesity paradox.3 And for that matter, what exactly constitutes obesity in elderly people, who naturally undergo changes in body composition as they age?
This article examines the literature on these controversial issues, including changes in body composition with age, the definition of obesity in older adults, the obesity paradox, and treatment of obesity in older adults.
AMERICANS ARE GETTING OLDER—AND BIGGER
Americans are living longer than ever before; life expectancy has reached a new high of 77.8 years.4,5 According to the US Census Bureau,6 about 27 million people in the United States are over age 70, and this number is expected to nearly double by 2030.
Meanwhile, the prevalence of obesity, defined as a body mass index (BMI) of 30 kg/m2 or higher, has increased in the last 25 years in all age groups in the United States, including those age 65 and older.7,8 These two trends add up to an increase in the number of obese older people. In 2000, 22.9% of people age 60 to 69 and 15.5% of those over age 70 and older were obese.9 This amounted to a 56% increase in the former group and a 36% increase in the latter group in the interval since 1991.5,9
BUT WHAT CONSTITUTES ‘OBESITY’?
Obesity is the excess accumulation of body fat, leading to a higher risk of medical illness and premature death. But measuring it is not as simple as one might think.
The body mass index can mislead
The BMI, ie, weight in kilograms divided by the square of the height in meters, correlates fairly well with body fat stores and is generally used to classify medical risk.
However, the BMI can classify some older people as overweight (BMI 30.0–34.9 kg/m2) or obese (BMI ≥ 35.0 kg/m2) who actually do not have an excess of body fat—and can fail to classify others as overweight or obese who do. For example, if a person loses height as a result of vertebral compression fractures, his or her BMI would become higher, even with no change in weight or body fat. Conversely, changes in body composition with age, including loss of muscle and an increase in fat, may not be reflected in the BMI, even if the person really does have too much body fat.10
This second limitation of the BMI is important when estimating risk in older adults, who have a particular fat distribution. Visceral, subcutaneous, intramuscular, and intrahepatic fat increase with age, and they are all risk factors for insulin resistance and type 2 diabetes mellitus.11 And in older people, having too much visceral fat is more prevalent than the BMI might predict.10
Percent body fat awaits investigation
Percent body fat is another way to assess body fat. Defined as the total weight of fat divided by total weight, it is measured in various ways.
Dual-energy x-ray absorptiometry, computed tomography, and magnetic resonance imaging can measure percent body fat, and they can differentiate visceral from subcutaneous fat (which is less metabolically active). Unfortunately, most of these tests are used for this purpose only in research, and they are relatively expensive.
Commercially available bioelectrical impedance devices send a weak electric current through the body and measure the resistance, and using this information and four other factors (height, weight, age, and sex), they calculate percent body fat. This method is fast, easy, painless, and cheap. A disadvantage is that the handheld devices measure body composition of the upper body only. Because the lower body is excluded, they do not give an accurate measurement of body fat of the abdomen and hips. Also, they cannot differentiate visceral from subcutaneous fat.
Bioelectrical impedance devices work well in healthy individuals with stable water balance. The values are only an estimate of fat-free mass, and therefore this method is not the gold standard for assessing body fat. Bioelectrical impedance is better at tracking body composition in an individual over time than at diagnosing obesity.
Percent body fat can vary by sex and race. Asians, for example, have higher percent body fats at lower BMIs, particularly when younger.12 Also, Gallagher et al12 found that percent body fat increased with age at every given BMI in both men and women (Table 1).
The traditional universal cutoffs for defining obesity by percent body fat are 25% in men and 35% in women. However, research has indicated that cutoffs of 20% to 25% in men and 30% to 38% in women may better identify those at risk of metabolic disease.13 Guidelines and evidence-based cutoffs for percent body fat must await further investigation.