In an analysis based on nearly one million hospital admissions, we found a decrease in rates of hospitalisation for acute coronary events (ACEs) among both men and women aged under 70 in the two years following the introduction of a comprehensive smoking ban in Italy. The observed reduction was stable over the study period, was similar in the different geographical areas, and was stronger among younger people.
Several pieces of evidence suggest that the Italian ban has been able to reduce the population exposure to passive smoking. The smoking legislation is almost universally respected, with less than 100 (1.5%) violations observed in about 6,000 checks by the police during the first year after the implementation of the law 
. Almost 90% of people interviewed in a large national survey in 2005 reported that the smoking ban was observed in bars and restaurants, and 70% reported the same regarding workplaces 
. Moreover, studies that measured concentrations of environmental nicotine and particulate matter with diameter <2.5 µm in hospitality premises of different Italian cities before and after the smoking ban showed reductions ranging from 70% to 97% 
Potential confounding in time series studies is limited to factors that are related to the outcome of interest and change at the time of the intervention 
. In contrast, temporal changes in the prevalence of individual risk factors (such as hypertension and cholesterol level) shape the underlying long-term trends, and are thus inherently taken into account in time-series studies. To our knowledge, with the exclusion of the national smoking regulation, no nation-wide intervention potentially able to cause a sharp decrease in rates of hospital admissions for ACEs took place in Italy during the study period. By adjusting for seasonality, we took into account most of the potential confounding effect of risk factors that change periodically during the year, including temperature and air pollution. In a previous study conducted among residents in the city of Rome, the estimates of the cardiovascular effects of the Italian smoking regulation were adjusted for a number of possible confounders including apparent temperature, flu epidemics and the air concentration of PM 10 
. Interestingly, reported crude and adjusted estimates were almost identical 
. Moreover, most of these risk factors mainly affect elderly adults 
, while the reduction in risk observed in the present study mainly occurred among younger people ().
Although a fraction of patients with acute coronary events dies before reaching a hospital, it is likely that mortality outside of the hospital has not changed substantially from 2002 to 2006. Again, the Rome study supports this assumption, as Cesaroni and colleagues found similar effects of the smoking regulation on mortality rates and hospital admissions 
. Some researchers have suggested that changes in the diagnostic criteria for acute myocardial infarction introduced by the European Society of Cardiology and the American College of Cardiology in 2000 may have induced an apparent increase in the rates of hospital admission thereafter 
. However, some studies have shown that the impact of these new diagnostic criteria was lower in Italy than in other countries, especially when acute myocardial infarction and other acute and sub-acute forms of ischemic heart disease are grouped together, as was the case in our study 
Growing evidence indicates that both active and passive smoking increases cardiac risk through both chronic (atherosclerosis) and acute (platelet activation and endothelial dysfunction) pathways 
. A reduction in the admissions for ACEs shortly after the introduction of the smoking ban is consistent with findings from laboratory and epidemiological studies, which report that the acute effects of both active and passive smoking disappear within a short time after cessation of the exposure 
Smoking regulations decrease exposure to passive smoking and may indirectly decrease active smoking. However, simulation studies 
and empirical evidence on the effect of the introduction of bans among smokers and non-smokers 
suggest that most of the observed effects of smoking regulations can be attributed to decrease in passive smoking. In a recent simulation study, we estimated that a 4% reduction in hospital admissions for ACEs could be obtained with a 15% to 60% decrease in the exposure to passive smoking from any source among the general population 
. Indeed, decreases of this magnitude have been found in population surveys that measured changes in the levels of salivary cotinine among non-smokers, following the introduction of smoking regulations 
. Moreover, the prevalence of exposure to passive smoking in public places is reported to be higher among the youngest age groups, which is consistent with our finding of a decreasing effect of the smoking ban with age 
We found a lower decrease in admissions for ACEs than some previous studies evaluating the effect of the introduction of smoking regulations in other countries 
. However, most of these studies did not fully account for underlying trends in hospital admissions; thus, they may have overestimated the real effect of smoking bans 
. Two large studies conducted in England and the New York State adjusted for long-term trends and found decreases in admissions for acute myocardial infarction of 2% and 8% respectively, consistent with our estimates 
. It should also be noted that comprehensive smoking bans were introduced in England, Italy and New York State when many public places and work environments were already smoke-free 
. For example, in Italy, smoking in public administration buildings was already banned in 1995 (even if the level of compliance with this law was not high) 
. Therefore, the low levels of exposure to passive smoking in some work environments before the introduction of comprehensive smoking regulations may have limited their potential benefits.
Assuming that the introduction of a smoking regulation changes the rates of hospital admissions for ACEs and mortality rates for coronary heart disease in the general population to a similar extent 
, the 4% decrease in events that we found can be compared with the effects of other interventions to decrease mortality by coronary heart disease and with the effects of population-wide changes in the prevalence of cardiovascular risk factors. In a recent study, Palmieri and colleagues used the IMPACT model to explain the decrease in coronary heart disease mortality in Italy between 1980 and 2000 
. Changes in prevalence of active smoking (from 31.7% in 1980 to 21.8% in 2000) were responsible for a 3.7% reduction in mortality for coronary heart disease during the study period. Similar effects were estimated for the introduction of novel therapies, such as the early medical treatment of acute myocardial infarction (−4.9%), hypertension treatments (−1.5%) and the use of statins for primary prevention (−2.7%) 
. Thus the magnitude of the effect of a smoking regulation is comparable with those of other interventions, but such effects are obtained immediately and with limited cost to the community 
The results of this study, which was carried out on a large population and over a long period of observation, suggest that smoke-free policies may result in a short-term reduction in hospital admissions for ACEs. This finding has important implications for public health. Smoking regulations represent a simple, effective and inexpensive intervention for the prevention of cardiovascular diseases and should be taken into account in prevention programmes.