BMI Alone Questioned for Predicting Mortality Risk in Seniors

Central adiposity was linked to a higher mortality risk in older adults, independently of body mass index (BMI), according to longitudinal data from a survey study of Medicare beneficiaries.

Over more than a decade of follow-up, older men and women with a high waist circumference (men >40 inches, women >35 inches) had a 24% higher risk of mortality compared with a reference group of adults with a low waist circumference, independent of BMI:

• Men: HR 1.24, 95% CI 1.07-1.44
• Women: HR 1.24, 95% CI 1.07-1.43

Every 5-inch increase was tied to a significant rise in the mortality hazard for both men (P=0.001) and women (P<0.001), reported Furong Xu, PhD, of the University of Rhode Island in Kingston, and colleagues in the Journal of the American Geriatrics Society.

However, analyses using BMI alone yielded opposite patterns. Lower mortality risk was reported among overweight men (HR 0.54, 95% CI 0.45-0.66), men with class I/II obesity (HR 0.49, 95% CI 0.37-0.65), and overweight women (HR 0.73, 95% CI 0.61-0.88) compared with normal-weight peers.

“BMI category alone may inadequately stratify mortality risk in older adults, particularly when central adiposity is not assessed,” the authors noted. “Instead, abdominal obesity assessed via [waist circumference] is a more critical health indicator, as it is a strong indicator of regional adiposity.”

Co-author Bryan Blissmer, PhD, also of the University of Rhode Island, told MedPage Today that the findings help clarify what is often called the “obesity paradox,” in which older adults with higher BMI appear to have better survival.

“The results suggest that this pattern should not be interpreted to mean that excess body fat is necessarily protective,” he said. “BMI does not distinguish fat mass from muscle mass or indicate where fat is stored.”

“By examining BMI and waist circumference together and allowing both to change over time, we found that higher BMI was associated with lower mortality while greater abdominal size remained associated with higher mortality,” Blissmer added. “The apparent paradox may therefore reflect, at least partly, the limitations of BMI when used by itself.”

The findings align with a recent Lancet Commission report that advocated against using BMI as a standalone diagnostic tool, favoring its use “as a surrogate measure of health risk at a population level,” noted Xu and colleagues.

Instead, clinicians should confirm obesity status using at least one additional anthropometric measure, including waist circumference, waist-to-hip ratio, or waist-to-height ratio, or direct fat mass measurement via dual-energy x-ray absorptiometry (DEXA) or bioimpedance.

“BMI should not be interpreted in isolation when evaluating older adults,” Blissmer suggested.

He also advised clinicians to pay attention to unintentional weight loss or a low or declining BMI, which may reflect loss of muscle, frailty, illness, or poor nutritional status. In this study, underweight status was associated with a higher mortality risk compared with normal weight in both men (HR 1.91, 95% CI 1.46-2.51) and women (HR 1.97, 95% CI 1.57-2.48), regardless of waist circumference.

Using 14 years of longitudinal data (2011-2024) from the National Health and Aging Trends Study, Xu and team evaluated 6,905 Medicare beneficiaries ages 65 and older (56% women, 18.1% ethnic minorities).

BMI was classified into the following categories: underweight (<18.5), normal weight (18.5-24.9), overweight (25.0-29.9), class I/II obesity (30.0 to <40), and class III obesity (≥40.0).

When BMI was combined with waist circumference, those with a normal BMI and high waist circumference had an increased mortality risk (males: HR 1.33, 95% CI 1.08-1.65; females: HR 1.23, 95% CI 1.03-1.47). Interestingly, mortality risk was lower in men who were overweight or had class I/II obesity with high waist circumference compared with men with normal BMI and low waist circumference.

“Our results should not be interpreted as recommending weight gain or suggesting that obesity is harmless in later life,” warned Blissmer. “This was an observational study, so it cannot establish that higher BMI causes longer survival. Instead, the findings support a more nuanced assessment of body size and health in older adults, including attention to both overall body mass and abdominal adiposity.”

Other limitations to the study included self-reporting of height and weight, which may have led to misclassification.

“Future research incorporating direct measures of muscle mass and fat distribution may help explain these relationships more fully,” Blissmer said.