Table 2.
Associations between ultra-processed food intake and risk of overall, prostate, colorectal, and breast cancer, from multivariable Cox proportional hazard models*, NutriNet-Santé cohort, France, 2009-17 (n=104 980)
| Proportion of ultra-processed food intake in the diet | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Continuous† | Sex specific quarters‡ | |||||||||
| 1 | 2 | 3 | 4 | P for trend | ||||||
| HR (95% CI) | P for trend | HR | HR (95% CI) | HR (95% CI) | HR (95% CI) | |||||
| All cancers | ||||||||||
| No of cases/non-cases | 2228/102 752 | 712/25 532 | 607/25 638 | 541/25 705 | 368/25 877 | |||||
| Model 1 | 1.12 (1.06 to 1.18) | <0.001 | 1 | 0.99 (0.89 to 1.11) | 1.10 (0.99 to 1.24) | 1.21 (1.06 to 1.38) | 0.002 | |||
| Model 2 | 1.12 (1.07 to 1.18) | <0.001 | 1 | 1.00 (0.90 to 1.11) | 1.11 (0.99 to 1.25) | 1.23 (1.08 to 1.40) | 0.001 | |||
| Model 3 | 1.12 (1.06 to 1.18) | <0.001 | 1 | 0.99 (0.89 to 1.11) | 1.01 (0.98 to 1.23) | 1.21 (1.06 to 1.38) | 0.002 | |||
| Model 4 | 1.13 (1.07 to 1.18) | <0.001 | 1 | 1.00 (0.90 to 1.11) | 1.11 (0.99 to 1.24) | 1.23 (1.08 to 1.40) | 0.001 | |||
| Prostate cancer | ||||||||||
| No of cases/non-cases | 281/22 540 | 96/5609 | 96/5609 | 59/5647 | 30/5675 | |||||
| Model 1 | 0.98 (0.83 to 1.16) | 0.8 | 1 | 1.18 (0.89 to 1.57) | 0.95 (0.69 to 1.32) | 0.93 (0.61 to 1.40) | 0.6 | |||
| Model 2 | 0.98 (0.83 to 1.16) | 0.8 | 1 | 1.18 (0.89 to 1.57) | 0.95 (0.69 to 1.32) | 0.93 (0.61 to 1.40) | 0.6 | |||
| Model 3 | 0.98 (0.83 to 1.15) | 0.8 | 1 | 1.18 (0.89 to 1.56) | 0.95 (0.68 to 1.31) | 0.92 (0.61 to 1.39) | 0.6 | |||
| Model 4 | 0.98 (0.83 to 1.16) | 0.8 | 1 | 1.18 (0.89 to 1.57) | 0.95 (0.68 to 1.32) | 0.93 (0.61 to 1.40) | 0.6 | |||
| Colorectal cancer | ||||||||||
| No of cases/non-cases | 153/104 827 | 48/26 196 | 43/26 202 | 36/26 210 | 26/26 219 | |||||
| Model 1 | 1.13 (0.92 to 1.38) | 0.2 | 1 | 1.10 (0.72 to 1.66) | 1.17 (0.76 to 1.81) | 1.49 (0.92 to 2.43) | 0.1 | |||
| Model 2 | 1.16 (0.95 to 1.42) | 0.1 | 1 | 1.12 (0.74 to 1.70) | 1.22 (0.79 to 1.90) | 1.59 (0.97 to 2.60) | 0.07 | |||
| Model 3 | 1.13 (0.92 to 1.38) | 0.2 | 1 | 1.09 (0.92 to 1.38) | 1.16 (0.75 to 1.80) | 1.48 (0.91 to 2.41) | 0.1 | |||
| Model 4 | 1.16 (0.95 to 1.42) | 0.1 | 1 | 1.12 (0.74 to 1.70) | 1.22 (0.79 to 1.89) | 1.23 (1.08 to 1.40) | 0.07 | |||
| Breast cancer | ||||||||||
| No of cases/non-cases | 739/81 420 | 247/20 292 | 202/20 338 | 179/20 361 | 111/20 429 | |||||
| Model 1 | 1.11 (1.02 to 1.22) | 0.02 | 1 | 0.97 (0.81 to 1.17) | 1.10 (0.90 to 1.34) | 1.14 (0.91 to 1.44) | 0.2 | |||
| Model 2 | 1.11 (1.01 to 1.21) | 0.03 | 1 | 0.96 (0.80 to 1.16) | 1.09 (0.89 to 1.32) | 1.12 (0.89 to 1.42) | 0.2 | |||
| Model 3 | 1.11 (1.02 to 1.22) | 0.02 | 1 | 0.97 (0.80 to 1.17) | 1.09 (0.90 to 1.33) | 1.14 (0.91 to 1.44) | 0.2 | |||
| Model 4 | 1.11 (1.01 to 1.21) | 0.03 | 1 | 0.96 (0.80 to 1.16) | 1.08 (0.89 to 1.32) | 1.13 (0.89 to 1.42) | 0.2 | |||
| Premenopausal breast cancer | ||||||||||
| No of cases/non-cases | 264/57 151 | 90/14 263 | 70/14 284 | 55/14 299 | 49/14 305 | |||||
| Model 1 | 1.09 (0.95 to 1.25) | 0.2 | 1 | 0.91 (0.67 to 1.25) | 0.92 (0.65 to 1.29) | 1.30 (0.90 to 1.86) | 0.3 | |||
| Model 2 | 1.07 (0.93 to 1.23) | 0.4 | 1 | 0.90 (0.66 to 1.24) | 0.90 (0.64 to 1.27) | 1.25 (0.87 to 1.80) | 0.4 | |||
| Model 3 | 1.09 (0.95 to 1.26) | 0.2 | 1 | 0.91 (0.67 to 1.25) | 0.92 (0.66 to 1.30) | 1.30 (0.91 to 1.88) | 0.3 | |||
| Model 4 | 1.08 (0.94 to 1.24) | 0.3 | 1 | 0.91 (0.66 to 1.24) | 0.91 (0.64 to 1.28) | 1.27 (0.88 to 1.83) | 0.4 | |||
| Postmenopausal breast cancer | ||||||||||
| No of cases/non-cases | 475/29 191 | 107/7309 | 128/7289 | 123/7294 | 117/7299 | |||||
| Model 1 | 1.13 (1.01 to 1.27) | 0.04 | 1 | 1.23 (0.95 to 1.60) | 1.28 (0.98 to 1.66) | 1.39 (1.07 to 1.82) | 0.02 | |||
| Model 2 | 1.13 (1.00 to 1.27) | 0.05 | 1 | 1.23 (0.95 to 1.60) | 1.27 (0.98 to 1.65) | 1.39 (1.05 to 1.81) | 0.02 | |||
| Model 3 | 1.13 (1.00 to 1.27) | 0.04 | 1 | 1.23 (0.95 to 1.59) | 1.27 (0.98 to 1.65) | 1.38 (1.06 to 1.81) | 0.02 | |||
| Model 4 | 1.13 (1.00 to 1.27) | 0.05 | 1 | 1.23 (0.95 to 1.59) | 1.27 (0.97 to 1.65) | 1.38 (1.05 to 1.81) | 0.02 | |||
HR=hazard ratio.
Model 1=multivariable Cox proportional hazard model adjusted for age (timescale), sex, energy intake without alcohol, number of 24 hour dietary records, smoking status, educational level, physical activity, height, body mass index, alcohol intake, and family history of cancers; breast cancer models were additionally adjusted for menopausal status, hormonal treatment for menopause, oral contraception, and number of children. Model 2=model 1 plus intakes of lipids, sodium, and carbohydrates. Model 3=model 1 plus Western dietary pattern (derived by factor analysis). Model 4=model 1 plus intakes of lipids, sodium, and carbohydrates and Western dietary pattern (derived by factor analysis). Pearson correlation coefficients with Western dietary pattern were 0.5 for dietary lipids, 0.6 for sodium, and 0.40 for carbohydrates.
Hazard ratio for increase of 10% in proportion of ultra-processed food intake in diet.
Sex specific cut-offs for quarters of ultra-processed proportions were 11.8%, 16.8%, and 23.3% in men and 11.8%, 16.8%, and 23.4% in women. In premenopausal women, cut-offs were 12.8%, 18.1%, and 25.0%. In postmenopausal women, cut-offs were 10.1%, 14.3%, and 19.5%.