Taking the results of many trials that have tested the same treatment, and looking at the results all together. This can give a more accurate picture than looking at the results of each trial separately.