TABLE 3.
Associations between sex-specific tertiles of DII and mortality from multivariate Cox proportional hazards models stratified by the antioxidant supplementation group, SU.VI.MAX cohort, France, 1994–20071
| n (cases/noncases) | HR (95% CI) | P-trend | P-interaction | |
| All-cause mortality (n = 207) | 0.098 | |||
| Placebo group | ||||
| Continuous DII score | 106/3931 | 1.17 (1.02, 1.35) | 0.03 | |
| T1 | 25/1323 | 1.00 (referent) | 0.02 | |
| T2 | 41/1315 | 2.17 (1.21, 3.91) | ||
| T3 | 40/1293 | 2.10 (1.15, 3.84) | ||
| Antioxidant supplementation group | ||||
| Continuous DII score | 101/3951 | 1.02 (0.89, 1.16) | 0.8 | |
| T1 | 38/1310 | 1.00 (referent) | 0.8 | |
| T2 | 28/1313 | 0.76 (0.46, 1.27) | ||
| T3 | 35/1328 | 1.09 (0.67, 1.77) | ||
| Mortality by CVD or cancer (n = 164) | 0.4 | |||
| Placebo group | ||||
| Continuous DII score | 85/3981 | 1.18 (1.01, 1.39) | 0.04 | |
| T1 | 23/1323 | 1.00 (referent) | 0.06 | |
| T2 | 29/1315 | 1.65 (0.85, 3.21) | ||
| T3 | 33/1293 | 1.92 (0.99, 3.71) | ||
| Antioxidant supplementation group | ||||
| Continuous DII score | 79/3951 | 1.11 (0.96, 1.28) | 0.2 | |
| T1 | 26/1310 | 1.00 (referent) | 0.3 | |
| T2 | 23/1313 | 0.88 (0.50, 1.56) | ||
| T3 | 30/1328 | 1.32 (0.77, 2.26) | ||
| Cancer-specific mortality (n = 123) | 0.3 | |||
| Placebo group | ||||
| Continuous DII score | 66/3931 | 1.24 (1.03, 1.50) | 0.03 | |
| T1 | 16/1323 | 1.00 (referent) | 0.02 | |
| T2 | 23/1315 | 2.09 (0.92, 4.78) | ||
| T3 | 27/1293 | 2.65 (1.18, 5.98) | ||
| Antioxidant supplementation group | ||||
| Continuous DII score | 57/3951 | 1.13 (0.95, 1.34) | 0.2 | |
| T1 | 19/1310 | 1.00 (referent) | 0.3 | |
| T2 | 15/1313 | 0.79 (0.40, 1.59) | ||
| T3 | 23/1328 | 1.43 (0.76, 2.68) |
Sex-specific cutoffs for tertiles of DII adjusted for energy intake by the residual method were used. In men, lower bounds of tertiles 2 and 3 were 0.20 and 1.46. In women, lower bounds of tertiles 2 and 3 were −0.09 and 1.41. Multivariate Cox proportional hazards models were adjusted for age (timescale in the Cox model), sex, number of 24-h dietary records, BMI, physical activity, smoking status, educational level, family history of cancer in first-degree relatives, family history of CVD in first-degree relatives, energy intake without alcohol, and alcohol intake. CVD, cardiovascular disease; DII, Dietary Inflammatory Index; SU.VI.MAX, Supplémentation en Vitamines et Minéraux Antioxydants; T, tertile.