Table 2 Results of the receiver operating characteristics (ROC) curve analyses for identifying optimal anthropometric cut-points for detecting the presence of at least two components of MS (excluding WC) in the derivation data set (N 19 880: men 8055, women 11 825)

From: Deriving an optimal threshold of waist circumference for detecting cardiometabolic risk in sub-Saharan Africa

Anthropometric variable

AUC

P-valuea for difference in AUC between index and WC

Cut-point

Sensitivity (%)

Specificity (%)

Youden index

Overall

 WC

0.66 (0.65–0.67)

—

81.1 (80.4–81.6)

61 (60–62)

65 (64–66)

0.260

 BMI

0.63 (0.62–0.64)

<0.001

24.8 (23.9–25.6)

51 (50–52)

70 (70–71)

0.214

 WHR

0.56 (0.55–0.57)

<0.001

0.87 (0.85–0.89)

51 (50–53)

59 (58–60)

0.103

 WHtR

0.65 (0.64–0.66)

0.277

0.51 (0.49–0.53)

53 (52–55)

70 (69–71)

0.234

Men

 WC

0.66 (0.65–0.68)

—

81.2 (78.5–83.8)

53 (51–55)

73 (72–74)

0.258

 BMI

0.64 (0.62–0.65)

<0.001

23.2 (22.1–24.3)

48 (46–50)

74 (73–76)

(0.220

 WHR

0.59 (0.58–0.60)

<0.001

0.88 (0.85–0.90)

62 (60–64)

53 (51–54)

0.156

 WHtR

0.65 (0.64–0.67)

0.276

0.48 (0.47–0.49)

58 (56–60)

66 (65–68)

0.242

Women

 WC

0.66 (0.65–0.67)

—

81.0 (79.2–82.8)

64 (63–65)

61 (60–62)

0.251

 BMI

0.62 (0.61–0.63)

<0.001

25.1 (23.6–26.6)

57 (55–58)

65 (64–66)

0.193

 WHR

0.56 (0.55–0.57)

<0.001

0.83 (0.81–0.85)

62 (60–63)

49 (48–51)

0.114

 WHtR

0.65 (0.64–0.66)

<0.001

0.54 (0.53–0.56)

54 (53–56)

69 (68–70)

0.229

  1. Abbreviations: AUC, area under the curve; BMI, body mass index (kg m−2); N, number of participants; WC, waist circumference (cm); WHR, waist hip ratio; WHtR, waist-to-height ratio.
  2. aP-values reported for tests restricted to individuals in which both WC and the anthropometric measure assessed are determined. Data in brackets are 95% confidence intervals.
Back to article page