The paper by Meyers, et. al., cited in the February 6, 2010 blog, had a correction issued in the Journal of the American College of Cardiology, which had to do with a statistical calculation error. The association between smoking bans and decrease in AMI (heart attack) is still strong but less so; the rate goes down 14%, not 26%, per year after bans are in place
The formal correction published in JACC is:
Meyers DG, Neuberger JS, He J. Cardiovascular Effect of Bans on Smoking in Public Places: A Systematic Review and Meta-Analysis. J Am Coll Cardiol 2009;54:1249 –55.
In this article, the meta-analysis included data from Pueblo, Colorado, which the authors erroneously reported as incidence rate ratio (IRR): 0.30. Actually, the IRR is 0.66 (95% confidence interval [CI]: 0.58 to 0.75). This changes the meta-analysis summary IRR to 0.92 (95% CI: 0.86 to 0.99). The meta-regression of the effect of ban duration also changes. The coefficient of post-ban duration in the meta-regression model is -0.16 (95% CI: -0.20 to-0.11), meaning that the IRR decreases by 14% (95% CI: 11% to 18%) for each year of post-ban observation (e.g., IRR: 0.86 after 1 year, then 0.73, then 0.63 compared with pre-ban).
The authors regret this error.
Post a Comment