MI is much greater than (as much as twice as high as) the rate of strokes (Lloyd-Jones et al., 2008), in this study there were more strokes than acute MIs. With respect to the analyses, this was the only study that attempted to account for previously implemented smoking bans; that is important given the large portion of the study population that was previously covered by smoking bans (New York City and several other large jurisdictions had previously implemented smoking bans). The results of the study, however, are sensitive to the assumptions used in the model and to the model choice. A sensitivity analysis showing the effect of model choice on study results might have provided more confidence in the study findings.
Saskatoon, Saskatchewan, Canada, implemented a smoking ban on July 1, 2004. The ban prohibited smoking in “any enclosed public space that is open to the public or to which the public is customarily admitted or invited.” Smoking was also prohibited in outdoor seating areas of restaurants and licensed premises. Smoking had previously been prohibited in government buildings.
Lemstra et al. (2008) conducted the only study to assess whether the smoking ban had an effect on rates of acute MI and also assessed smoking prevalence and public support of the ban. That study provides information directly related to questions about the association between smoking bans and acute coronary events. The authors measured business compliance with the ban by reviewing warnings and tickets issued by public-health inspectors to eligible businesses. Of 924 eligible establishments, 914 (98.9%) were inspected within the first 6 months of the ban. Of the 914, only 13 (1.4%) had to be issued noncompliance warnings (for not posting signs or removing ashtrays); one ticket was issued on reinspection of those 13 that were issued warnings. The committee found no exposure-assessment data.
Lemstra et al. (2008) obtained information on acute MI from the Strategic Health Information Planning Services. ICD-10 codes, rather than ICD-9 codes, were in use in Saskatoon beginning in 2000, so the analyses used data from July 2000 and later. The authors calculated age-standardized incidences of acute MI per 100,000 people in the first full year of the smoking ban (July 1, 2004–June 30, 2005) and in the previous 4 years (July 1, 2000–June 30, 2004). Data collected on smoking prevalence in 2003 and 2005 by Statistics Canada were used to evaluate changes in smoking pattern.