Low cigarette consumption and risk of coronary heart disease and stroke: meta-analysis of 141 cohort studies in 55 study reports.

Although we had only summary data (hence limited ability to show trends reliably), we saw some suggestion of a positive trend between the size of the relative risk for smoking one cigarette per day and the last calendar year of follow-up for each study: Spearman’s correlation 0.51 (P=0.008) for men and 0.21 (P=0.42) for women when we used studies of coronary heart disease, and 0.23 (P=0.39) and 0.56 (P=0.11) among men and women for studies of stroke.

also in 11,005 other papers

Also, we found no evidence of a negative trend between size of relative risk for smoking 20 cigarettes per day and last calendar year of follow-up (which might suggest many heavy smokers cutting down, and whether this increases over time): Spearman’s correlations were positive: 0.30 (P=0.15) for men and 0.33 (P=0.20) for women (coronary heart disease studies).

also in 4,201 other papers