TABLE 3.
Meta-analysis of multivariable relative risks of the association between dairy product consumption and stroke using random effects1
| High vs. low intake | Dose–response (1-serving/d increase) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Dairy product | Risk estimates (n) | Participants (n) | Cases (n) | Pooled RR (95% CI) | I 2 (%) | Risk estimates (n) | Participants (n) | Cases (n) | Pooled RR (95% CI) | I 2 (%) | Quality of evidence2 |
| Total dairy | 19 | 552,573 | 27,775 | 0.90 (0.85, 0.96) | 60.8 | 16 | 431,799 | 22,677 | 0.96 (0.93, 0.99) | 74.7 | Low |
| High-fat dairy | 9 | 412,206 | 14,869 | 0.93 (0.89, 0.98) | 0.0 | 7 | 285,046 | 10,839 | 0.99 (0.97, 1.00) | 0.0 | Low |
| Low-fat dairy | 11 | 432,988 | 15,923 | 0.91 (0.87, 0.96) | 0.0 | 9 | 305,828 | 11,893 | 0.97 (0.95, 1.00) | 14.0 | Low |
| Milk | 15 | 665,044 | 27,223 | 0.96 (0.88, 1.04) | 72.1 | 12 | 343,302 | 14,713 | 0.94 (0.89, 0.99) | 76.4 | Low |
| Cheese | 10 | 335,543 | 14,799 | 0.94 (0.88, 1.00) | 23.8 | 7 | 202,562 | 10,986 | 0.96 (0.91, 1.01) | 52.1 | Low |
| Yogurt | 4 | 107,603 | 5471 | 1.07 (0.97, 1.18) | 0.0 | 3 | 73,194 | 5312 | 1.08 (0.85, 1.39) | 79.8 | Very low |
1
Pooled relative risks are from random-effects meta-analyses. I2 refers to the proportion of heterogeneity between studies.
2
The quality of evidence was graded according to the NutriGrade scoring system.