Age differences in the impact of obesity on incident diabetes

Abstract

Wang TG et al [1] in Lancet Healthy Longevity examined a cohort of 93,781 participants aged ≥40 in China who were free of diabetes at baseline and were followed up for 3.1 years on average, and found that adjusted hazard ratio (HR) of incident diabetes in participants with either general (defined as body mass index - BMI ≥ 28 kg/ m2) or central obesity (as waist circumference – WC ≥ 90 cm in men and ≥ 80 cm in women) was 1·35 (95% confidence interval 1.23 - 1.48) in age of 40 ~ < 55 years, 1.20 (1.09 - 1.31) in 55 ~ < 65 years, 1.16 (1.01 - 1.32) in 65 ~ < 75 years, and 0.99 (0.72 - 1.36) in ≥ 75 years. They concluded that the impact of obesity on diabetic risk decreased with age and older people should pay more attention on other modified risk factors, e.g., reducing unhealthy sleep patterns, rather than adiposity for diabetic prevention [1]. However, we were wondering whether the definition of obesity used by the authors would be suitable for older Chinese and misled to the findings for the conclusion, since (i) BMI ≥ 28 kg/m2 may not be an appropriate cut-off point for obesity in older Chinese, and (ii) the older people could have different body shapes from their middle-aged counterparts. We have recently analysed the Anhui cohort study data [2], which included 2,809 Chinese aged ≥ 60 who were free of diabetes at baseline and had up to 10 years follow-up, and found that the BMI ≥ 26 kg/ m2 as obesity [3] better predicted the risk of incident diabetes than the BMI ≥ 28 kg/m2, while China-WC (i.e., ≥ 95 cm in men, ≥ 90 cm in women) [4] better than WC - WHO (≥ 102 cm in men, ≥ 88 cm in women) [5].