2022 Volume 49 Issue 2 Pages 308-316
Purpose: This study aims to compare the risk of elevated blood pressure in relation to obesity using three of its criteria, and to determine its association with body mass index (BMI) and waist circumference (WC).
Method: This study enrolled 15,560 individuals: 7,126 men and 8,434 women, aged 20–59 years, who had undergone annual workplace health checkups at two adjacent prefectures in the Tohoku region in 2013. Some individuals were flagged because of the following reasons: missing data, use of antihypertensive medication, uncertainty about the use of antihypertensive medication, smoking habit, or clinically implausible values. These individuals were excluded. The participants were classified first by sex and then by age group: those in their 20s and 30s and those in their 40s and 50s. Four cutoff values for blood pressure (sBP/dBP: ≧120/80, ≧130/80, ≧130/85, and ≧140/90 mmHg) were compared using three obesity criteria (BMI and criteria by the Japan Society for the Study of Obesity and the World Health Organization for WC ). The areas under the receiver operating characteristic (ROC) curves (AUCs) for BMI and WC for each cutoff value were compared using the DeLong test, and further cutoff values were calculated.
Results: The odds ratios for participants in their 20s and 30s (men: 3.78–6.75, women: 3.80–8.37) were greater than those for participants in their 40s and 50s (men: 1.87–2.64, women: 2.41–4.15). Comparisons of BMI and WC by AUC of the ROC curve did not show significant differences for most criteria, except for women in their 40–50s and ≧120/80 mmHg.
Conclusion: Our results showed no significant difference between BMI and WC in determining elevated blood pressure. In addition, for many criteria, AUC<0.7 was a limitation in predicting high blood pressure using obesity indices alone.
