Statistical Analysis
The meta-analysis was carried out using NCSS Software. The odds ratio
for all-cause death, CVS-caused death, MI, and stroke was evaluated in
patients taking Aspirin compared to placebo in diabetes. The
random-effects model was used with statistical significance at a p-value
less than 0.05. The odds ratio of various bleeding events was also done
to do the forest plot of studies analyzing the effect of Aspirin
compared to placebo in diabetic patients. The logarithm of the measures
was used to construct the forest plot. Again, the random-effects model
was used with statistical significance at a p-value less than 0.05.