BMI cutoff for overweight needs readjustment in Asians, says study

Body mass index (BMI) levels both above and below 25.0–29.9 kg/m² are associated with increased mortality in a population without pre-existing disease, suggesting that the current cutoff of overweight (BMI ≥23 kg/m²) may need re-evaluation and that BMI alone may not be an accurate measure for indicating adiposity in Asians, according to a prospective cohort study in Korea.

“The use of waist circumference and predicted body fat mass in addition to BMI may help better predict mortality,” the researchers said. “Further, to reduce premature mortality, preventive strategies should also focus on body fat mass and waist circumference loss, rather than weight loss alone.”

A total of 44,060 participants of the Korea National Health and Nutrition Examination Survey 2007–2014 agreed to mortality follow-up through 31 December 2016. Height, weight, and waist circumference were measured at baseline.

The researchers used DXA data to derive predicted body fat and fat-free mass, and Cox proportional hazards models to estimate hazard ratios (HRs) and 95 percent confidence intervals (CIs) for the associations with mortality, adjusting for potential confounders. They also tested for linearity using the likelihood ratio test comparing nonlinear restricted cubic spline models with linear models.

A total of 1,682 deaths were recorded over ≤9.5 years of follow-up. BMI showed a J-shaped association with all-cause and cardiovascular mortality, with the nadir at BMI 25.0–29.9 kg/m² (p_nonlinearity<0.001). [Am J Clin Nutr 2021;113:142-153]

Waist circumference (≥95 vs 75.0–79.9 cm: HR, 2.10, 95 percent CI, 1.54–2.86) and predicted body fat mass (highest vs lowest sextiles: HR, 2.55, 95 percent CI, 1.60–4.06) among participants with no history of cancer or cardiovascular disease positively correlated with all-cause mortality (p_nonlinearity≤0.03 for all), as well as cancer and cardiovascular mortality.

Participants with both high predicted fat mass and low fat-free mass showed the highest mortality.

“Our observation of a J-shaped relation between BMI and mortality is consistent with findings from previous studies,” the researchers said. [Lancet 2016;388:776-786; N Engl J Med 2010;363:2211-2219; N Engl J Med 2006;355:779-787; N Engl J Med 2011;364:719-729]

A meta-analysis of 239 prospective studies from four continents found a similar J-shaped relation among health never smokers in Europe, North America, Australia, and East Asia, but the nadir of the curve occurred at a lower BMI range (20.0–24.9 kg/m²) and did not vary by geographic region. [Lancet 2016;388:776-786]

“Whereas the current study used measured height and weight, most studies included in the meta-analysis used self-reported height and weight which are more prone to measurement error (over-report of height, under-report of weight),” the researchers said. “The resulting underestimation of BMI may have contributed to the underestimation of optimal BMI.”

Other studies from East Asian countries reported the optimal BMI at 20.0–24.9 kg/m², but others found it at 25.0–34.9 kg/m². [Obesity (Silver Spring) 2008;16:1714-1717; Obesity (Silver Spring) 2008;16:2348-2355; Obesity 2010;18:362-329; Obes Res 2004;12:2031-2034; Biomed Environ Sci 2017;30:79-87; Sci Rep 2016;6:31609; Public Health Nutr 2015;18:1839-1846; N Engl J Med 2006;355:779-787]

“Studies tended to find higher optimal BMI in older adults, possibly owing to age-related biological effects and elevated reverse causation (eg, subclinical diseases),” the researchers said. [Biomed Environ Sci 2017;30:79-87; Public Health Nutr 2015;18:1839-1846; Obesity 2010;18:362-369]

“We believe our findings help further our understanding of the BMI relations with adiposity and mortality among Asians,” they added.