Sixteen trials were included (55,462 participants).
Assessment of publication bias: Kendall's tau correlation coefficient between sample size and standardised AR = 0.03 (P = 0.90). Kendall's tau correlation between sample size and standardised logarithmic RR = 0.37 (P = 0.05). Plot of sample size vs In (RR) showed significant correlation was due to 2 trials with large sample size and moderately large RR. AR per 10,000 persons associated with aspirin treatment: No significant heterogeneity among studies (P = 0.99).
All cause mortality: -120 (95% CI: -77, -162: P < 0.001). Cardiovascular mortality: -97 (95% CI: -59, -135: P < 0.001).
Total myocardial infarction: -137 (95% CI: -107, -167: P < 0.001).
Fatal myocardial infarction: -36 (95% CI: -16, -55: P < 0.001).
Total stroke: -31 (95% CI: -5, -57: P = 0.02).
Fatal stroke: 4 (95% CI: -8, 16: P < 0.5).
Haemorrhagic stroke: 12 (95% CI: 5, 20: P < 0.001).
Ischaemic stroke: -39 (95% CI: -17, -61: P < 0.001).
AR of haemorrhagic stroke in sub-groups with P value from Woolfs chi-squared test: Pre-existing disease (14 trials) AR = 16 (95% CI: 5, 26) vs no pre-existing disease (2 trials) AR = 9 (95% CI:-2, 19). P = 0.36. Age (8 trials): < 64 years AR = 9 (95% CI: 2, 16) vs age equal to or greater than 64 years AR = 34 (95% CI: 1, 66), P = 0.15. Sample size (8 trials): < 1120 subjects AR = 12 (95% CI:- 37, 62) vs equal to or greater than 1120 subjects 12 (95% CI: 5, 20), P = 0.98. Dosage of aspirin (8 trials) < 413 mg/day AR = 11 (95% CI: 4, 18) vs equal to or greater than 413 mg/day AR = 19 (95% CI: -8, 46), P = 0.59.
Study duration (8 trials): < 28 days AR = 6 (95% CI: -2, 14) vs equal to or greater than 28 days AR = 16 (95% CI: 5, 28), P = 0.14.
RR of aspirin use: No significant heterogeneity among studies (P = 0.99)
All cause mortality: 0.85 (95% CI: 0.80, 0.90: P < 0.001).
Cardiovascular mortality: 0.84 (95% CI: 0.79, 0.90: P < 0.001).
Total myocardial infarction: 0.68 (95% CI: 0.62, 0.74: P < 0.001).
Fatal myocardial infarction: 0.78 (95% CI: 0.68, 0.90: P < 0.001).
Total stroke: 0.88 (95% CI: 0.76, 1.02: P = 0.08). Fatal stroke: 1.07 (95% CI: 0.85, 1.35: P = 0.60). Date of publication (8 trials): < 1988 RR = 1.37 (95% CI: 0.72, 2.61) vs equal to or greater than 1989 AR = 2.21 (95% CI: 1.33, 3.66).