Table 2.
The association between urinary caffeine levels (μmol/L) and stroke.
| Model 1 OR (95% CI) P value | Model 2 OR (95% CI) P value | Model 3 OR (95% CI) P value | |
|---|---|---|---|
| Caffeine (μmol/L) | 1.004 (0.991, 1.018) 0.534 | 0.999 (0.984, 1.013) 0.847 | 0.996 (0.981, 1.011) 0.606 |
| Caffeine categories | |||
| Q1 | Reference | Reference | Reference |
| Q2 | 0.799 (0.515, 1.240) 0.316 | 0.780 (0.496, 1.224) 0.280 | 0.758 (0.478, 1.202) 0.239 |
| Q3 | 0.757 (0.485, 1.183) 0.222 | 0.713 (0.450, 1.129) 0.149 | 0.645 (0.402, 1.035) 0.069 |
| Q4 | 0.948 (0.622, 1.443) 0.802 | 0.816 (0.526, 1.265) 0.363 | 0.747 (0.475, 1.175) 0.207 |
| Age categories | |||
| 20–34 | 1.000 (0.900, 1.112) 0.996 | 0.993 (0.892, 1.105) 0.894 | 0.998 (0.883, 1.128) 0.973 |
| 35–49 | 0.991 (0.943, 1.042) 0.722 | 0.993 (0.943, 1.045) 0.782 | 0.994 (0.946, 1.044) 0.808 |
| 50–64 | 0.990 (0.962, 1.019) 0.495 | 0.989 (0.960, 1.019) 0.468 | 0.984 (0.953, 1.016) 0.311 |
| 65–80 | 1.000 (0.982, 1.017) 0.965 | 1.001 (0.983, 1.019) 0.941 | 1.001 (0.982, 1.020) 0.950 |
| Sex | |||
| Men | 1.000 (0.977, 1.022) 0.970 | 0.990 (0.966, 1.014) 0.417 | 0.985 (0.961, 1.010) 0.244 |
| Women | 1.008 (0.991, 1.025) 0.365 | 1.003 (0.986, 1.021) 0.717 | 1.002 (0.983, 1.022) 0.838 |
| Race | |||
| Mexican American | 0.971 (0.889, 1.060) 0.509 | 0.952 (0.872, 1.038) 0.264 | 0.886 (0.791, 0.993) 0.037 |
| Other Hispanic | 1.022 (0.971, 1.076) 0.396 | 1.007 (0.951, 1.065) 0.820 | 0.998 (0.935, 1.066) 0.959 |
| Non-Hispanic White | 1.010 (0.994, 1.025) 0.210 | 1.006 (0.991, 1.022) 0.429 | 1.005 (0.988, 1.022) 0.594 |
| Non-Hispanic Black | 0.955 (0.899, 1.014) 0.132 | 0.941 (0.885, 1.001) 0.054 | 0.944 (0.887, 1.004) 0.065 |
| Other Race | 1.017 (0.960, 1.078) 0.565 | 1.020 (0.956, 1.088) 0.556 | 1.022 (0.949, 1.099) 0.569 |
Model 1: The covariates were not adjusted.
Model 2: The age, sex, and race of the participants were adjusted.
Model 3: The age, sex, race, high blood pressure, high cholesterol level, diabetes, coronary heart disease, smoking, drinking, BMI of participants were adjusted.