Analysis and Comparison of Different Obesity Evaluation Indices and Their Effects on Hypertension, Diabetes, and Dyslipidaemia

Background: Standard measures that dene obesity and related disorders varies widely, this study investigated the relationship between different anthropometric indices of obesity criteria and their correlation to hypertension, diabetes, and dyslipidaemia in a local adult population in China. Methods: The study participants underwent the same questionnaire survey, bio-impedance body composition analysis, and blood laboratory test. The t-test and chi-square test were used to compare the characteristics of different groups, and the receiver operating characteristic curve was used to analyse the correlation of different indicators and explore their cut-off values. Results: The study comprised 14,926 participants, of whom 39.80% (5948/14,926) were male, and the mean age of the study population was 56.75±9.74 years. The waist circumference had the greatest inuence on all factors, and BMI, AVI, and BRI were similarly correlated. WHtR had the largest AUC for predicting obesity in both sexes, and in addition, we provided a recommended cut-off value of BMI, WHR, WHtR, BAI, OBD, CI, AVI, ABSI and BRI. WHtR had the largest AUC for predicting diabetes, hypertension, and dyslipidaemia, while BMI also served as a good predictive indicator (all P<0.001). Conclusions: Among the samples in this study, WHtR may be the best indicator for predicting obesity, hypertension, and dyslipidaemia, and AVI is a good indicator in Chinese adults specically. leading to inconsistencies in result reporting (5–7). Research shows that BMI, as an index of obesity, is more dependent on height and weight, and cannot distinguish fat from thin weight (8). The waist-to-hip ratio (WHR) is often used to determine obesity, however, a high WHR indicates abdominal fat accumulation, and does not take fat accumulation in the buttocks and limbs into account (9). BF% is used to assess obesity according to the percentage of body fat content, which determines whether an increase in is due to an increase in fat or an increase in muscle and other components, serving as a more accurate measure of the degree of overall obesity (10). this study chose BF% as the gold standard of obesity indices, to explore the consistency of other standards, and investigate the cut-off value each indicator in Chinese adults and their relationship with obesity-related diseases, such as hypertension, diabetes, and dyslipidaemia.

signal sent through the hands and feet. Participants removed extra clothes, such as shoes, coats, sweaters, and metal accessories, such as earrings, rings, and watches, and stood on a balance scale with bare feet and grasped the handles of the BIA. The examination took approximately 30 s.

Statistical analyses
Data are presented as group mean ± standard deviation (SD) unless otherwise stated. Comparisons between males and females were performed using the ttest for continuous data and the chi-square test for categorical data. Receiver operating characteristic (ROC) analyses were then used to calculate the area under the ROC curves (AUC) between dyslipidaemia and anthropometric measures, adjusted for age and sex. All analyses were performed using SPSS statistical software version 26.0. All tests were 2-tailed, and P < 0.05 was considered statistically signi cant.

Results
The basic characteristics of the study population are shown in Table 1. Among the participants, 39.80% (5948/14,926) were male, the mean age of the study population was 56.75 ± 9.74 years, most of them were married (93.4%) and only had a primary school education (67.3%), and females showed less smoking and alcohol consumption, and more tea and vinegar consumption, compared to men. Males tended to have higher anthropometric measures, such as the circumference of the neck, waist, arm, and thigh (all P < 0.001). For the obesity criteria, BMI in this study showed no signi cant difference between males and females, and WHR showed little difference (P = 0.015). BF%, WHtR, BAI, and CI were higher in females than males, OBD, AVI, and BRI were higher in males than females, and ABSI was almost the same both groups (all P < 0.05). Regarding biochemical status, males tended to have higher blood pressure and FBG levels, females has higher TC, HDL-C, and LDL-C levels (all P < 0.001), and TG showed no statistically signi cant difference between the two groups. Pearson's correlation coe cients were used to measure the correlation between obesity criteria and anthropometric and biochemical variables ( Table 2).
Anthropometric measures, such as weight, height, and WC, were signi cantly correlated with obesity criteria (all P < 0.001). Among them, BMI, AVI, and BRI have more correlation factors, and waist circumference has the most in uence on all factors, notably on AVI. Hip circumference, LAC, HC, and RAC were found to correlate best with BMI, and waist circumference correlated the best with WHR, WHtR, CI, and AVI. Height and weight correlated the best with OBD and BRI.
Biochemical variables have lower correlation coe cients with obesity criteria, and only blood pressure indicators showed a low correlation. men and 24.85 in women. Men have higher values for BAI, OBD, AVI, and BRI, and lower values for BMI, WHR, and WHtR compared to females, and ABSI has the same as the cut-off point of 0.08 in both sexes. The AUC for each disease and the different obesity criteria are shown in Table 4. Overall, the accuracy of each obesity index in predicting hypertension, diabetes, and dyslipidaemia was not very high (all AUC < 0.7), and there was no signi cant difference in the use of ABSI to predict diabetes. WHtR had the largest AUC for hypertension, diabetes, and dyslipidaemia, while BMI had the same predictive level for hypertension and diabetes (all P < 0.001). diabetes and hypertension than that accumulated in the lower body. Moreover, upper body fat may directly affect fatty acids and lipid metabolism throughout the body (24,25). Although the BF% method is considered the gold standard for determining obesity, in view of the fact that the equipment for measuring body composition is generally expensive and not suitable for handling, it is inconvenient for practical application (26). The method of WHtR determining obesity is relatively simple, and is suitable for large-scale screening of the population.
This study investigated the cut-off values of different obesity indicators, which can be used as a reference for subsequent research. In order to make the evaluation objective and accurate, it is recommended to use multiple indicators for simultaneous evaluation. The use of a combination of indices is recommended when assessing obesity, together with the fat distribution, to increase the accuracy of predicting chronic cardiovascular disease. According to the status of each obesity diagnosis index, intervention to improve physical tness and the quality of life is important (27)(28)(29).
Several limitations exist in our study. Dyslipidaemia included any group of dyslipidaemia components, but the speci c dyslipidaemia components were not grouped in this study. The results are only applicable to the preliminary screening of dyslipidaemia in the community population. The data for this study is the baseline from a cohort study and contains an inner limitation of cause-and-effect analysis. Further investigation is necessary to con rm the ndings of this study.

Conclusions
Among the samples in this study, WHtR may be the best indicator for predicting obesity, hypertension, and dyslipidaemia, and AVI is a good indicator in Chinese adults speci cally.