Table 1 summarizes the sociodemographic characteristics of 2 623 females with and without T2D. Overall, 14.9% were newly diagnosed with T2D. The mean age of the participants was 44.5 years with T2D participants significantly older than their counterparts. In fact, all attributes considered (employment status, population group, and residence) were significantly different between participants with and without T2D (all p < 0.001).
Table 1
Sociodemographic characteristics of South African women by type 2 diabetes mellitus (T2D) status.
| Total (n = 2623) | T2D absent (n = 2239) | T2D present (n = 384) | P value |
---|
Age in years, Mean ± SD | 44.5 ± 17.0 | 42.3 ± 16.6 | 57.2 ± 13.6 | < 0.001 |
Employment status, n (%) | < 0.001 |
Unemployed | 870(38.85) | 796(41.41) | 74(23.34) | |
Employed | 569(25.41) | 503(26.17) | 66(20.82) | |
Other | 800(35.73) | 623(32.41) | 177(55.83) | |
Population group, n (%) | < 0.001 |
African | 1721(65.91) | 1502(67.41) | 219(57.18) | |
White | 53(2.02) | 46(2.06) | 7(1.82) |
Coloured | 700(26.80) | 600(26.92) | 100(26.10) |
Indian | 137(5.24) | 80(3.59) | 57(14.88) |
Residence, n (%) | < 0.001 |
Urban formal | 1364(52.00) | 1110(49.57) | 254(66.14) | |
Urban informal | 331(12.61) | 304(13.57) | 27(7.03) |
Rural formal (farms) | 510(19.44) | 439(19.60) | 71(18.48) |
Rural informal (tribal) | 418(15.93) | 386(17.23) | 32(8.33) |
Other: students, retired, housewife, sick, disable, unable to work |
Table 2 presents the anthropometrical indices of South African women with and without T2D; always showing significantly higher values in participant with newly diagnosed T2D compared with their counterparts (all p < 0.001).
Table 2
Anthropometric indices of South African women by type 2 diabetes mellitus (T2D) status
| Total (n = 2612) | T2D absent (n = 2148) | T2D Present (n = 364) | P value |
---|
Body mass index (kg/m2)* | 28.00(23.49–33.53) | 27.00(22.86–32.91) | 31.70(27.82–36.12) | < 0.001 |
Traditional anthropometric indices |
Waist circumference (cm)* | 89.00(77.50–100.00) | 87.00(76.00–98.00) | 98.45(89.93–108.00) | < 0.001 |
Waist-to-hip ratio# | 0.86 ± 0.09 | 0.85 ± 0.09 | 0.91 ± 0.11 | < 0.001 |
Waist-to-height ratio* | 0.57(0.49–0.64) | 0.55(0.49–0.63) | 0.63(0.58–0.69) | < 0.001 |
Novel anthropometric indices* |
A body shape index | 0.07(0.06–0.07) | 0.07(0.06–0.07) | 0.08(0.070–0.074) | < 0.001 |
Abdominal volume index | 15.92(12.21–20.11) | 15.38(11.87–19.35) | 19.49(16.40–23.40) | < 0.001 |
Body adiposity index | 34.61(29.41–40.20) | 34.12(28.85–39.79) | 37.42(32.96–42.52) | < 0.001 |
Body roundness index | 4.71(3.21–6.33) | 4.41(3.05–5.96) | 6.12(5.00-7.56) | < 0.001 |
Conicity index | 1.23(1.15–1.30) | 1.22(1.14–1.29) | 1.29(1.22–1.36) | < 0.001 |
CUN-BAE | 40.52(33.49–46.68) | 39.11(32.60-45.92) | 45.52(41.05–49.75) | < 0.001 |
Normality was tested by Kolmogorov–Smirnov test: * Skewed distribution: Median (25th -75th percentile); # Normal distribution: Mean ± Standard Deviation. CUN-BAE: Clínica Universidad de Navarra-Body Adiposity Estimator.
The prevalence of other cardiometabolic disorders that commonly co-exist with T2D were as follows in women with and without newly diagnosed T2D: hypertension 64.94% vs 34.61%, p < 0.001, elevated triglycerides 61.31% vs 25.33%, p < 0.001, high LDL-C 54.43% vs 33.78%, p < 0.001and low HDL-C 1.20% vs 3.08%, p = 0.068 (Results table not included).
Table 3 shows the consistent increasing likelihood of T2D per standard deviation higher level of anthropometric measures in crude models (model 1); after adjustment for age (model 2), and sociodemographic factors (model 3). Notably, both WHtR and CUN-BAE showed a twofold increase in the likelihood of T2D, as indicated by the crude odds ratios. However, this estimate attenuated after adjustment for sociodemographic factors. Equivalent estimates across increasing quartiles of anthropometric indices are shown in Table 4. The odds ratios were consistently higher across high quartiles.
Table 3
Odds ratios (OR) and 95% confidence intervals (95%CIs) for type 2 diabetes mellitus (T2D) presented per standard deviation higher level of traditional and novel anthropometric indices among South African women
| Model 1 (Unadjusted) | Mode 2 (adjusted) | Model 3 (adjusted) |
---|
| Crude OR | 95%CI | P value | aOR | 95%CI | P value | aOR | 95%CI | P value |
Traditional anthropometric indices |
Body mass index | 1.61 | 1.45–1.79 | < 0.001 | 1.61 | 1.43–1.80 | < 0.001 | 1.58 | 1.38–1.81 | < 0.001 |
Waist circumference | 1.98 | 1.76–2.22 | < 0.001 | 1.85 | 1.63–2.09 | < 0.001 | 1.86 | 1.60–2.15 | < 0.001 |
Waist-to-hip ratio | 1.98 | 1.76–2.24 | < 0.001 | 1.66 | 1.46–1.89 | < 0.001 | 1.73 | 1.48–2.01 | < 0.001 |
Waist-to-height ratio | 2.04 | 1.81–2.29 | < 0.001 | 1.86 | 1.64–2.11 | < 0.001 | 1.84 | 1.59–2.13 | < 0.001 |
Novel anthropometric indices |
A body shape index | 1.39 | 1.23–1.56 | < 0.001 | 1.15 | 1.01–1.30 | 0.029 | 1.18 | 1.02–1.37 | 0.025 |
Abdominal volume index | 1.83 | 1.64–2.04 | < 0.001 | 1.73 | 1.54–1.94 | < 0.001 | 1.73 | 1.51–1.99 | < 0.001 |
Body adiposity index | 1.45 | 1.30–1.61 | < 0.001 | 1.40 | 1.25–1.57 | < 0.001 | 1.35 | 1.18–1.53 | < 0.001 |
Body roundness index | 1.88 | 1.68–2.09 | < 0.001 | 1.73 | 1.54–1.94 | < 0.001 | 1.71 | 1.49–1.96 | < 0.001 |
Conicity index | 1.94 | 1.70–2.20 | < 0.001 | 1.59 | 1.39–1.82 | < 0.001 | 1.58 | 1.34–1.85 | < 0.001 |
CUN-BAE | 2.19 | 1.92–2.50 | < 0.001 | 1.90 | 1.64–2.20 | < 0.001 | 1.86 | 1.57–2.20 | < 0.001 |
Model 1: unadjusted; Model 2: adjusted for age; Model 3: adjusted for age, population group, residence and employments. aOR: adjusted odds ratio. CUN-BAE; Clínica Universidad de Navarra-Body Adiposity Estimator. We have computed the OR per standard deviation change in the level of each anthropometric indices.
Table 4
Odds ratio (OR) and 95% confidence intervals (95% CI) for type 2 diabetes mellitus (T2D) across increasing quartiles of anthropometric indices
| Model 1 | | Model 2 | | Model 3 | |
---|
Variables | OR | 95%CI | aOR | 95%CI | aOR | 95%CI |
Body mass index |
Q1 (< 23.5) | Ref | | Ref | | Ref | |
Q2(23.6–28.1) | 2.38** | 1.52–3.72 | 2.53** | 1.59–4.02 | 2.01* | 1.19–3.42 |
Q3(28.2–33.5) | 5.00** | 3.30–7.58 | 5.14** | 3.32–7.94 | 4.32** | 2.62–7.11 |
Q4(33.6+) | 5.77** | 3.82–8.71 | 5.69** | 3.70–8.76 | 4.70** | 2.86–7.72 |
Waist circumference |
Q1(< 78) | Ref | | Ref | | Ref | |
Q2(79–89) | 2.75** | 1.68–4.50 | 2.27* | 1.37–3.76 | 2.45* | 1.36–4.42 |
Q3(90–100) | 5.47** | 3.43–8.72 | 4.56** | 2.82–7.37 | 4.16** | 2.26–7.31 |
Q4(101+) | 10.23** | 6.51–16.09 | 7.73** | 4.89–12.32 | 7.54** | 4.35–13.08 |
Waist-to-hip ratio |
Q1(< 0.80) | Ref | | Ref | | Ref | |
Q2(0.81–0.86) | 2.82** | 1.77–4.49 | 2.28** | 1.41–3.67 | 2.27* | 1.31–3.93 |
Q3(0.87–0.92) | 4.38** | 2.80–6.85 | 3.20** | 2.02–5.07 | 3.27** | 1.93–5.53 |
Q4(0.93+) | 8.15** | 5.30-12.55 | 5.00** | 3.21–7.82 | 5.04** | 3.01–8.41 |
Waist-to-height ratio |
Q1(< 0.5) | Ref | | Ref | | Ref | |
Q2(0.56–0.6) | 1.79* | 1.09–2.94 | 1.44 | 0.87–2.40 | 1.50 | 0.84–2.68 |
Q3(0.61–0.69) | 4.76** | 3.05–7.43 | 3.65** | 2.31–5.77 | 3.05** | 1.80–5.17 |
Q4(0.7+) | 9.12** | 5.94–14.02 | 6.32** | 4.06–9.84 | 5.85** | 3.50–9.78 |
A body shape index |
Q1(< 0.06) | Ref | | Ref | | Ref | |
Q2 (0.061–0.069) | 1.08 | 0.74–1.58 | 0.97 | 0.66–1.45 | 0.96 | 0.61–1.50 |
Q3(0.07–0.079) | 2.10** | 1.49–2.97 | 1.59* | 1.11–2.28 | 1.60* | 1.06–2.41 |
Q4 (0.08+) | 2.61** | 1.86–3.65 | 1.63* | 1.14–2.33 | 1.51* | 1.01–2.28 |
Abdominal volume index |
Q1(< 12.21) | Ref | | Ref | | Ref | |
Q2(12.22–15.96) | 2.86** | 1.73–4.73 | 2.40** | 1.43–4.03 | 2.67** | 1.46–4.88 |
Q3(15.97–20.11) | 5.29** | 3.28–8.51 | 4.54** | 2.78–7.41 | 4.17** | 2.34–7.43 |
Q4(20.12+) | 10.70** | 6.75–16.94 | 8.27** | 5.15–13.28 | 8.46** | 4.81–14.86 |
Body adiposity index |
Q1(< 29.41) | Ref | | Ref | | Ref | |
Q2(29.42–34.62) | 2.18** | 1.47–3.24 | 2.18** | 1.44–3.30 | 1.99* | 1.23–3.20 |
Q3(34.63–40.20) | 3.05** | 2.09–4.47 | 2.84** | 1.91–4.23 | 2.43** | 1.53–3.88 |
Q4(40.21+) | 3.84** | 2.64–5.58 | 3.50** | 2.37–5.18 | 3.04** | 1.92–4.80 |
Body roundness index |
Q1(< 3.22) | Ref | | Ref | | | |
Q2(3.23–4.72) | 1.79* | 1.09–2.94 | 1.44 | 0.87–2.40 | 1.50 | 0.84–2.68 |
Q3(4.73–6.33) | 4.76** | 3.05–7.43 | 3.65** | 2.31–5.77 | 3.05** | 1.80–5.17 |
Q4(6.34+) | 9.12** | 5.94–14.02 | 6.32** | 4.06–9.84 | 5.85** | 3.50–9.78 |
Conicity index |
Q1 (< 1.15) | Ref | | Ref | | Ref | |
Q2 (1.16–1.23) | 2.23** | 1.44–3.45 | 1.88* | 1.20–2.95 | 1.76* | 1.06–2.92 |
Q3 (1.24–1.30) | 3.94** | 1.99–4.63 | 2.22** | 1.44–3.43 | 2.11* | 1.31–3.42 |
Q4 (1.31+) | 6.37** | 4.27–9.50 | 3.89** | 2.67–5.89 | 3.47** | 2.17–5.53 |
Clínica Universidad de Navarra-Body Adiposity Estimator |
Q1(< 33.50) | Ref | | Ref | | Ref | |
Q2(33.51–40.53) | 5.15** | 2.86–9.27 | 3.19** | 1.75–5.82 | 3.23** | 1.63–6.41 |
Q3(40.54–46.68) | 11.78** | 6.71–20.70 | 6.89** | 3.88–12.23 | 6.59** | 3.40-12.76 |
Q4(46.69+) | 13.88** | 7.93–24.32 | 7.76** | 4.38–13.75 | 7.35** | 3.80-14.23 |
Model 1: unadjusted; Model 2: adjusted for age; Model 3: adjusted for age, employment, residence and population group; Q1, Q2, Q3 &Q4: quartiles; aOR: adjusted odds ratio; Ref: reference group. **p < 0.001, *p < 0.05 indicates the significant ability of each index to measure the risk of T2D.
Finally, based on the ROC analysis, we observed that none of the anthropometrical indices (both traditional and novel) exhibited the AUC values above 0.8 (Table 5 and Fig. 2). The only indices that performed acceptably to predict T2D were WC and WHtR (traditional indices, yielding the AUC above 0.7 and cutoff points of > 86.65 cm and 0.57, respectively); and the AVI, BRI, and CUN-BAE (novel indices, yielding the AUC above 0.7 and the cutoff points of 15.52, 93.83 and 38.35, respectively).
Table 5
Area under the curve (AUC), optimal cut-off points (using Youden Index), sensitivity and specificity for novel anthropometric indices in receiver operating characteristic (ROC) analysis to predict type 2 diabetes mellitus (T2D).
| AUC | 95%CI | P value | Cut-off point | sensitivity | 1-specificity | Youden’s index |
---|
Traditional anthropometric indices |
Body mass index | 0.67 | 0.64–0.69 | < 0.001 | 24.53 | 0.88 | 0.65 | 0.54 |
Waist circumference | 0.71 | 0.68–0.74 | < 0.001 | 86.65 | 0.83 | 0.52 | 0.35 |
Waist-to-hip ratio | 0.69 | 0.66–0.72 | < 0.001 | 0.84 | 0.80 | 0.54 | 0.34 |
Waist-to-height ratio | 0.72 | 0.69–0.75 | < 0.001 | 0.57 | 0.79 | 0.43 | 0.22 |
Novel anthropometric indices |
A body shape index | 0.61 | 0.58–0.64 | < 0.001 | 0.07 | 0.80 | 0.67 | 0.46 |
Abdominal volume index | 0.71 | 0.68–0.73 | < 0.001 | 15.52 | 0.80 | 0.49 | 0.29 |
Body adiposity index | 0.63 | 0.60–0.66 | < 0.001 | 32.00 | 0.79 | 0.60 | 0.39 |
Body roundness index | 0.72 | 0.69–0.75 | < 0.001 | 3.83 | 0.89 | 0.61 | 0.50 |
Conicity index | 0.69 | 0.66–0.71 | < 0.001 | 1.18 | 0.85 | 0.63 | 0.48 |
Clínica Universidad de Navarra-Body Adiposity Estimator | 0.70 | 0.68–0.73 | < 0.001 | 38.35 | 0.86 | 0.53 | 0.39 |
AUC: area under the curve; ROC: receiver operating characteristic |