Table 6.

Overall quality assessment of nine cohort studies (796,069 participants in total) examining the impact of vegetable consumption on anthropometric outcomes using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system.

Category	Rating with Reasoning
Limitations	−1 quality levels due to limitations related to measurement of exposure
Inconsistency	No subtraction of levels, as inconsistency does not affect confidence in results
Directness of evidence	−1 level due to indirect measure of exposure over time
Precision	No subtraction of levels as the total sample size of included studies was large
Publication bias	No subtraction of levels, as studies with both significant and insignificant outcomes included and grey literature adequately searched
Upgrading factors: Dose response	+1 as 3 studies clearly indicated a dose response whereby higher vegetable intakes were associated with the lowest risks of weight gain
Overall quality	Moderate: our confidence in the overall evidence is moderate, as the true effect is likely to be close to the estimate of the effect but there is possibility that it is different