Effect of smoking cessation on mortality after myocardial infarction - Meta-analysis of cohort studies

Objective: To determine the effect of smoking cessation on mortality after myocardial infarction. Data Sources: English- and non-English-language articles published from 196 6 through 1996 retrieved using keyword searches of MEDLINE and EMBASE suppl emented by letters to authors and searching bibliographies of reviews. Study Selection: Selection of relevant abstracts and articles was performed by 2 independent reviewers. Articles were chosen that reported the results of cohort studies examining mortality in patients who quit vs continued sm oking after myocardial infarction. Data Extraction: Mortality data were extracted from the selected articles b y 2 independent reviewers. Data Synthesis: Twelve studies were included containing data on 5878 patien ts. The studies took place in 6 countries between 1949 and 1988. Duration o f follow-up ranged from 2 to 10 years. All studies showed a mortality benef it associated with smoking cessation. The combined odds ratio based on a ra ndom effects model for death after myocardial infarction in those who quit smoking was 0.54 (95% confidence interval, 0.46-0.62). Relative risk reduct ions across studies ranged from 15% to 61%. The number needed to quit smoki ng to save 1 life is 13 assuming a mortality rate of 20% in continuing smok ers. The mortality benefit was consistent regardless of sex, duration of fo llowup, study site, and time period. Conclusion: Results of several cohort studies suggest that smoking cessatio n after myocardial infarction is associated with a significant decrease in mortality.