Objective  To construct a body shape index (ABSI) for Chinese adults and explore the index's association with blood pressure and to provide references for constructing an obesity index for Chinese adult population.

  Methods  Totally 9 592 adults aged 20 – 69 years were recruited with stratified cluster random sampling at three of the National Physical Fitness Surveillance Monitoring Cities (Shenyang, Dandong and Chaoyang) in Liaoning province and then measurements of body weight, height and waist circumference (WC) were performed among the residents during May – June 2010 and June – July 2014. The collected anthropometric data were used to derive an ABSI formula. According to the allometric model, linear regression analysis was performed with natural logarithm function In (WC) as a dependent variable and with In (body weight) and In (height) as the independent variables to derive ABSI formulas for male and female adults, respectively.

  Results  The established linear regression equations were as following: for male: In (WC) = – 2.641 + 0.700 × In (weight) – 0.945 × In (height) (R² = 0.748, P < 0.001) and \rmABSI\left( \rmk\rmg^ - 0.7\rmm^1.945 \right) = \dfrac\rmWC\left( \rmm \right)\rmbody\;\rmmass\;\rmindex\left( \rmBMI,kg/\rmm^2 \right)^0.7 \times \rmheight\left( \rmm \right)^0.455, for female: In (WC) = – 2.750 + 0.734 × In (weight) – 1.060 × In (height)(R² = 0.688, P < 0.001) and \rmABSI\left( \rmk\rmg^ - 0.734\rmm^2.06 \right) = \dfrac\rmWC\left( \rmm \right)\rmBMI\left( \rmkg/\rmm^2 \right)^0.734 \times \rmheight\left( \rmm \right)^0.408. The ABSI of the male adults (0.072 ± 0.004 kg^–0.7m^1.945) was significantly higher than that of female adults (0.064 ± 0.004 kg^–0.734m^2.06) (t = 73.09, P < 0.001). Correlation analysis revealed that the ABSI was positively correlated with WC in both the male and female adults (r = 0.487, 0.550; all P < 0.01), but the ABSI was not correlated with height, weight, and BMI (all P>0.05). Binary logistic regression analysis demonstrated that compared to the males with the ABSI of < 0.070 kg^–0.7m^1.945, the males with the ABSI of 0.070 – 0.072 kg^–0.7m^1.945 and ≥ 0.0732 kg^–0.7m^1.945 were at a significantly increased risk of hypertension (odds ratio OR=1.45, 95% confidence interval 95% CI: 1.23 – 1.71; OR = 1.63, 95% CI: 1.38 – 1.92); while, the females with the ABSI of 0.062 – 0.066 kg^–0.734m^2.06 and ≥ 0.066 kg^–0.734m^2.06 were more likely to have normal high blood pressure (OR = 1.37, 95% CI: 1.10 – 1.71 and OR = 1.69, 95% CI: 1.36 – 2.11) and hypertension (OR = 1.66, 95% CI: 1.35 – 2.04 and OR = 2.49, 95% CI: 2.04 – 3.03) in comparison with the females with the ABSI < 0.062 kg^–0.734m^2.06 (P < 0.01 for all).

  Conclusion  The study constructs a body shape index based on waist circumference standardized by weight and height for Chinese adults. As a central obesity index, a body shape index is not correlated with body mass index but positively associated with the risk of hypertension in the population.