Table 2.
Relative risk of colorectal cancer by quintiles of energy-adjusted dietary glycemic load, overall glycemic index, carbohydrate, fiber, nonfiber carbohydrate, sucrose, and fructose
| Quintile of intake | |||||||
|---|---|---|---|---|---|---|---|
| Lowest | 2 | 3 | 4 | Highest | Ptrend* | Continuous† | |
| Dietary glycemic load | |||||||
| No. of colorectal cancer cases‡ | 26 | 30 | 37 | 32 | 49 | ||
| Total No. of person-years of follow-up | 61 084 | 61 213 | 61 190 | 60 976 | 60 872 | ||
| Age-adjusted relative risk (95% CI) | 1.00 (referent) | 1.13 (0.67 to 1.91) | 1.34 (0.81 to 2.22) | 1.14 (0.68 to 1.90) | 1.68 (1.04 to 2.71) | .03 | 1.08 (1.00 to 1.15) |
| P = .04 | |||||||
| Multivariable-adjusted relative risk (95% CI)§ | 1.00 (referent) | 1.34 (0.76 to 2.34) | 1.81 (1.02 to 3.21) | 1.63 (0.86 to 3.09) | 2.85 (1.40 to 5.80) | .004 | 1.18 (1.04 to 1.33) |
| P = .007 | |||||||
| Overall glycemic index | |||||||
| No. of colorectal cancer cases‡ | 24 | 37 | 31 | 43 | 39 | ||
| Total No. of person-years of follow-up | 61 153 | 60 879 | 61 132 | 61 183 | 60 988 | ||
| Age-adjusted risk (95% CI) | 1.00 (referent) | 1.58 (0.94 to 2.63) | 1.31 (0.77 to 2.24) | 1.86 (1.13 to 3.07) | 1.73 (1.04 to 2.87) | .03 | 1.06 (1.01 to 1.11) |
| P = .02 | |||||||
| Multivariable-adjusted risk (95% CI)§ | 1.00 (referent) | 1.59 (0.95 to 2.67) | 1.24 (0.71 to 2.16) | 1.93 (1.15 to 3.24) | 1.71 (0.98 to 2.98) | .04 | 1.05 (1.00 to 1.11) |
| P = .05 | |||||||
| Carbohydrate | |||||||
| No. of colorectal cancer cases‡ | 30 | 25 | 37 | 36 | 46 | ||
| Total No. of person-years of follow-up | 61 100 | 61 241 | 61 159 | 61 079 | 60 756 | ||
| Age-adjusted risk (95% CI) | 1.00 (referent) | 0.80 (0.47 to 1.37) | 1.14 (0.71 to 1.85) | 1.07 (0.66 to 1.75) | 1.31 (0.83 to 2.08) | .12 | 1.03 (0.99 to 1.08) |
| P = .15 | |||||||
| Multivariable-adjusted risk§(95% CI) | 1.00 (referent) | 1.04 (0.59 to 1.85) | 1.58 (0.88 to 2.83) | 1.71 (0.89 to 3.29) | 2.41 (1.10 to 5.27) | .02 | 1.11 (1.02 to 1.21) |
| P = .02 | |||||||
| Fiber | |||||||
| No. of colorectal cancer cases‡ | 35 | 36 | 22 | 41 | 40 | ||
| Total No. of person-years of follow-up | 60 976 | 61 242 | 61 052 | 61 074 | 60 990 | ||
| Age-adjusted risk (95% CI) | 1.00 (referent) | 0.93 (0.58 to 1.49) | 0.53 (0.31 to 0.91) | 0.91 (0.58 to 1.44) | 0.80 (0.50 to 1.27) | .49 | 0.89 (0.68 to 1.15) |
| P = .35 | |||||||
| Multivariable-adjusted risk (95% CI)§ | 1.00 (referent) | 0.99 (0.61 to 1.62) | 0.57 (0.32 to 1.00) | 0.96 (0.58 to 1.59) | 0.79 (0.45 to 1.38) | .50 | 0.83 (0.61 to 1.14) |
| P = .25 | |||||||
| Nonfiber carbohydrate | |||||||
| No. of colorectal cancer cases‡ | 27 | 29 | 40 | 32 | 46 | ||
| Total No. of person-years of follow-up | 61 129 | 61 198 | 61 131 | 61 069 | 60 809 | ||
| Age-adjusted risk (95% CI) | 1.00 (referent) | 1.01 (0.60 to 1.72) | 1.42 (0.87 to 2.30) | 1.09 (0.65 to 1.82) | 1.52 (0.94 to 2.45) | .08 | 1.04 (1.00 to 1.09) |
| P = .08 | |||||||
| Multivariable-adjusted risk (95% CI)§ | 1.00 (referent) | 1.30 (0.73 to 2.29) | 1.97 (1.11 to 3.52) | 1.62 (0.83 to 3.13) | 2.60 (1.22 to 5.54) | .02 | 1.11 (1.02 to 1.21) |
| P = .02 | |||||||
| Sucrose | |||||||
| No. of colorectal cancer cases‡ | 28 | 29 | 33 | 39 | 45 | ||
| Total No. of person-years of follow-up | 60 883 | 60 871 | 61 169 | 61 126 | 61 286 | ||
| Age-adjusted risk (95% CI) | 1.00 (referent) | 1.00 (0.60 to 1.69) | 1.11 (0.67 to 1.84) | 1.27 (0.78 to 2.07) | 1.41 (0.88 to 2.27) | .08 | 1.07 (0.96 to 1.19) |
| P = .21 | |||||||
| Multivariable-adjusted risk (95% CI)§ | 1.00 (referent) | 1.05 (0.62 to 1.80) | 1.19 (0.70 to 2.02) | 1.41 (0.84 to 2.35) | 1.51 (0.90 to 2.54) | .06 | 1.08 (0.96 to 1.21) |
| P = .19 | |||||||
| Fructose‖ | |||||||
| No. of colorectal cancer cases‡ | 21 | 38 | 38 | 34 | 43 | ||
| Total No. of person-years of follow-up | 61 002 | 61 117 | 61 065 | 61 107 | 61 044 | ||
| Age-adjusted risk (95% CI) | 1.00 (referent) | 1.66 (0.97 to 2.83) | 1.59 (0.93 to 2.71) | 1.36 (0.79 to 2.35) | 1.69 (1.00 to 2.85) | .18 | 1.02 (0.92 to 1.14) |
| P = .73 | |||||||
| Multivariable-adjusted risk (95% CI)§ | 1.00 (referent) | 1.87 (1.07 to 3.28) | 1.88 (1.06 to 3.33) | 1.68 (0.93 to 3.06) | 2.09 (1.13 to 3.87) | .08 | 1.04 (0.91 to 1.18) |
| P = .60 | |||||||
P values were obtained from two-sided Wald tests.
Continuous variables are 10 U/day, except for overall glycemic index, which is 1 U/day. CI = confidence interval.
The 174 cases of colorectal cancer and the number of person-years are shown for age-adjusted models; five cases were dropped from the multivariable models (n = 169) because of missing covariate information.
Multivariable model was adjusted for age (y), body mass index (kg/m2), history of oral contraceptive use (yes, no), postmenopausal hormone use (never, past, current), family history of colorectal cancer (yes, no), smoking (never, past, current), alcohol use (never/rarely, 1–3 drinks per mo, 1–6 drinks per wk, ≥1 drink per day), physical activity (tertiles of kJ expended weekly in recreational activity and stair climbing), nonsteroidal anti-inflammatory use (never/rarely, >1 time per wk), total energy intake (natural logarithm-transformed kJ), energy-adjusted total fiber (g) (except for fiber model), energy-adjusted total fat (g), energy-adjusted folate (µg), energy-adjusted calcium (mg), and energy-adjusted vitamin D (mg).
Fructose includes the fructose component of sucrose.