**3.4 Threshold of independent variables for detecting HTN**

The ROC curve analyses are presented in **Table 6**. In both genders, the AUCs were significantly greater than 0.5 for both VJP and BMI (p < 0.05). The optimal threshold


### **Table 4.**

*Multiple regression analysis among Fatness, leg power, and BP among participants.*


### **Table 5.**

*Odds of risk of HTN are stratified according to gender (n = 2047).*


### **Table 6.**

*Receiver operating characteristic analysis for risk of HTN (n = 2047).*

in girls for VJP and BMI were 1501.5 W and 18.9 kg.m−2, respectively. Corresponding values for boys were 1340.7 W and 18.9 kg.m−2, respectively. Details of the results are shown in **Table 5**. The gender-specific ROC curves are presented in **Figures 2** and **3**.

In order to further evaluate the influence of fatness and LP on BP, participants were divided into four fat/power groups and the results are presented in **Table 7**. The proportion of girls within these categories was 42.1%, 53.0%, 3.6%, and 1.3% for low fat/high power, low fat/low power, high fat/high power, and high fat/low power, respectively. Corresponding values for boys were 42.2%, 53.3%, 3.0%, and 1.3%. There were significant group differences in SBP for both girls (F(3,1083) = 7.50, *p* < 0.001) and boys (F(3,956) = 4.99, *p* < 0.002). For both genders, the differences were between the fat/low power group and three other groups (low fat/high power, low fat/high power, and high fat/high power). The same group differences were documented for boys. For DBP in girls, there was a significant (F(3,1083) = 3.66, *p* = 0.012) group difference, the difference was between the two extreme groups. In boys, there were no group differences (*p* = 0.0165).

*Association of Fatness and Leg Power with Blood Pressure in Adolescents DOI: http://dx.doi.org/10.5772/intechopen.106279*

**Figure 2.** *Areas under the curve for BMI and VJP in girls.*

**Figure 3.** *Areas under the curve for BMI and VJP in boys.*


### **Table 7.**

*Differences in BP according to fat/power groups (n = 2047).*
