OBJECTIVE: To determine the mathematic formula for weight, height and waist and hip circumference that is most closely correlated to cardiovascular disease risk factors. DESIGN: Population-based, cross-sectional surveys. SETTING: Five Canadian provinces, between 1990 and 1992. PARTICIPANTS: A probability sample of 16,007 men and women aged 18 to 74 years was selected using health insurance registration files in each province. Anthropometry was performed on 10,054 (63%) of these adults. OUTCOME MEASURES: The power of height in the body mass index (BMI, kg/m2) and of hip circumference in the ratio of waist to hip circumference (WC/HC) was varied from 0 to 3. Simple linear regression analysis for each age-sex group was used to examine the relation of each index to systolic and diastolic blood pressure (SBP and DBP), levels of total cholesterol (TC), low-density lipoprotein cholesterol (LDL) and high-density lipoprotein cholesterol (HDL), triglycerides (TRIG) and the ratio of TC to HDL. Values for the coefficient of determination (r2) were used to compare the fits of the models. RESULTS: The r2 values were generally low (< 0.27), but were greatest in the younger age groups (18-24 and 35-54 years) and in women. Waist circumference alone (WC/HC0) showed the best fit with SBP and DBP, whereas WC/HC0.5 was most closely related to HDL, TC/HDL and TRIG. None of the indices was closely associated with TC or LDL. Whatever the power of height used, the weight-height ratios showed weaker associations with the risk factors than the waist-hip ratios. CONCLUSIONS: WC and BMI correlate most closely with blood pressure and plasma lipid and may be the best simple anthropometric indices to include in the routine clinical examination of adults.
OBJECTIVE: To determine the mathematic formula for weight, height and waist and hip circumference that is most closely correlated to cardiovascular disease risk factors. DESIGN: Population-based, cross-sectional surveys. SETTING: Five Canadian provinces, between 1990 and 1992. PARTICIPANTS: A probability sample of 16,007 men and women aged 18 to 74 years was selected using health insurance registration files in each province. Anthropometry was performed on 10,054 (63%) of these adults. OUTCOME MEASURES: The power of height in the body mass index (BMI, kg/m2) and of hip circumference in the ratio of waist to hip circumference (WC/HC) was varied from 0 to 3. Simple linear regression analysis for each age-sex group was used to examine the relation of each index to systolic and diastolic blood pressure (SBP and DBP), levels of total cholesterol (TC), low-density lipoprotein cholesterol (LDL) and high-density lipoprotein cholesterol (HDL), triglycerides (TRIG) and the ratio of TC to HDL. Values for the coefficient of determination (r2) were used to compare the fits of the models. RESULTS: The r2 values were generally low (< 0.27), but were greatest in the younger age groups (18-24 and 35-54 years) and in women. Waist circumference alone (WC/HC0) showed the best fit with SBP and DBP, whereas WC/HC0.5 was most closely related to HDL, TC/HDL and TRIG. None of the indices was closely associated with TC or LDL. Whatever the power of height used, the weight-height ratios showed weaker associations with the risk factors than the waist-hip ratios. CONCLUSIONS: WC and BMI correlate most closely with blood pressure and plasma lipid and may be the best simple anthropometric indices to include in the routine clinical examination of adults.