Table 2.
Study Design | Quality of Evidence | Reduced Factors | Increased Factors |
---|---|---|---|
Randomized trials | High | Risk of bias | Large effect |
−1 serious | +1 large | ||
−2 very serious | +2 very large | ||
Moderate | Inconsistency | Dose response | |
−1 serious | +1 if gradient | ||
−2 very serious | |||
Indirectness | Confounding | ||
−1 serious | +1 | ||
−2 very serious | |||
Observational studies | Low | Imprecision | |
−1 serious | |||
−2 very serious | |||
Very low | Publication bias | ||
−1 likely | |||
−2 very likely |