| Variables | Coefficient | Standard Error |
|---|---|---|
| Age* | ||
| Linear component | 0.0917 | 0.0106 |
| Nonlinear component | −0.0234 | 0.0106 |
| Diabetes | 0.1102 | 0.0758 |
| Regular use of multivitamins | −0.1918 | 0.0532 |
| Family history of colon cancer | 0.2425 | 0.0850 |
| Years of education* | ||
| Linear component | 0.0721 | 0.0181 |
| Nonlinear component | −0.0734 | 0.0193 |
| Race/ethnicity | ||
| Hawaiian | 0.1609 | 0.1208 |
| Japanese | 0.2535 | 0.0820 |
| Latino | −0.1366 | 0.0925 |
| White | −0.1673 | 0.0921 |
| Body mass index* | ||
| Linear component | 0.0180 | 0.0162 |
| Nonlinear component | 0.0090 | 0.0194 |
| Alcoholic drinks per day* | ||
| Linear component | 0.2838 | 0.0752 |
| Nonlinear component | −1.7375 | 0.5356 |
| Hours of moderate activity per day* | ||
| Linear component | −0.0907 | 0.0702 |
| Nonlinear component | 0.1103 | 0.1496 |
| Regular aspirin use | ||
| Yes, not currently | −0.0323 | 0.0719 |
| Yes | −0.2096 | 0.0655 |
| Pack-years of smoking* | ||
| Linear component | 0.0002 | 0.0059 |
| Nonlinear component | 0.0177 | 0.0170 |
↵* Continuous variables were modeled using restricted cubic splines with 3 knots. Therefore, continuous variables contain one coefficient for the linear component and a nonlinear component.