Table 2.
RFM†, % | Hazard ratio (95% CI) | BMI‡, kg/m2 | Hazard ratio (95% CI) | WC, cm | Hazard ratio (95% CI) | |
---|---|---|---|---|---|---|
Women (n = 23,354) | ||||||
25 | 0.70 (0.33–1.48) | 11 | 0.65 (0.35–1.24) | 51 | 1.44 (0.51–4.06) | |
28 | 0.76 (0.43–1.33) | 14 | 0.73 (0.45–1.17) | 59 | 1.32 (0.61–2.86) | |
31 | 0.83 (0.57–1.20) | 17 | 0.81 (0.59–1.11) | 67 | 1.20 (0.72–2.01) | |
34 | 0.90 (0.75–1.09) | 20 | 0.90 (0.77–1.05) | 75 | 1.10 (0.85–1.42) | |
25th percentile | 37 | 1.00 (Reference) | 23 | 1.00 (Reference) | 83 | 1.00 (Reference) |
40 | 1.18 (1.03–1.37) | 26 | 1.13 (0.98–1.30) | 91 | 0.93 (0.73–1.19) | |
43 | 1.59 (1.28–1.99) | 29 | 1.31 (1.03–1.67) | 99 | 0.97 (0.63–1.48) | |
46 | 2.50 (1.92–3.27) | 32 | 1.58 (1.18–2.12) | 107 | 1.18 (0.72–1.95) | |
49 | 4.27 (2.97–6.12) | 35 | 1.97 (1.45–2.69) | 115 | 1.65 (0.99–2.73) | |
52 | 7.37 (4.47–12.16) | 38 | 2.50 (1.82–3.45) | 123 | 2.41 (1.48–3.93) | |
Men (n = 23,181) | ||||||
13 | 1.75 (0.70–4.36) | 13 | 1.05 (0.47–2.34) | 58 | 1.42 (0.51–3.95) | |
16 | 1.51 (0.76–2.97) | 16 | 1.04 (0.57–1.88) | 66 | 1.30 (0.60–2.78) | |
19 | 1.30 (0.83–2.03) | 19 | 1.02 (0.69–1.51) | 74 | 1.18 (0.71–1.96) | |
22 | 1.12 (0.90–1.39) | 22 | 1.00 (0.83–1.22) | 82 | 1.08 (0.84–1.38) | |
25th percentile | 25 | 1.00 (Reference) | 25 | 1.00 (Reference) | 90 | 1.00 (Reference) |
28 | 1.05 (0.90–1.23) | 28 | 1.07 (0.93–1.23) | 98 | 1.02 (0.84–1.23) | |
31 | 1.47 (1.17–1.84) | 31 | 1.28 (1.05–1.57) | 106 | 1.22 (0.92–1.62) | |
34 | 2.58 (1.99–3.35) | 34 | 1.67 (1.34–2.07) | 114 | 1.68 (1.24–2.26) | |
37 | 4.87 (3.48–6.82) | 37 | 2.23 (1.76–2.83) | 122 | 2.45 (1.81–3.30) | |
40 | 9.20 (5.91–14.32) | 40 | 3.00 (2.29–3.92) | 130 | 3.60 (2.62–4.94) |
BMI body mass index, CI confidence interval, RFM relative fat mass, WC waist circumference.
*Estimates represent adjusted hazard ratios obtained from Cox regression models with RFM, BMI, and WC modelled using restricted cubic splines. Hazard ratios were adjusted for age (as restricted cubic splines), ethnicity, education level, and smoking status. The rounded weighted 25th percentile of each anthropometric index was arbitrarily chosen as the reference. Index values are shown in increments of approximately 0.5 of their standard deviation to facilitate comparison. †RFM was calculated as follows: 64 − (20 × height/waist circumference) + (12 × sex); sex equals 0 for men and 1 for women; height and waist circumference were measured in the same units. ‡BMI was calculated as body weight in kilograms divided by the square of the height in meters.