Every year cardiovascular disease (CVD) causes 3.9 million deaths in Europe and over 1.8 million deaths in the European Union (EU), accounting for 37% of all deaths in the EU.1 In Portugal and in Europe as a whole, CVD is the leading cause of death.2

Despite recent decreases in mortality rates in many countries, CVD is still responsible for almost half of all deaths in Europe,2 and constitutes a major public health challenge in western Europe.3

Of all behavioral components, dietary factors pose the greatest risk for CVD mortality and reduced CVD disability-adjusted life years in the European population. High systolic blood pressure and smoking represent other major risk factors.

Several European countries have implemented measures to tackle CVD mortality, from disease prevention by tackling major risk factors to improvements in disease management and treatment.

As part of this effort, in recent years Portugal has instituted a set of health policies that directly or indirectly tackle CVD mortality, including a smoking ban in 2009, a salt reduction regulation in 2010 and the coronary fast-track system (FTS) for acute coronary syndrome (ACS). The latter policy, which was more directed linked to reducing mortality, was launched in 2000, but was not fully implemented throughout the country until 2007.

There is growing evidence-based data worldwide on the effectiveness of these health policies and public health initiatives; however, reports on the effectiveness of similar policies in Portugal is scarce. More importantly, there has been no analysis of the cumulative effect of the three major polices recently implemented in Portugal. In the context of increasingly limited resources for health care, investment in population-wide policy strategies needs to show an appropriate return, in this case a decline in CVD mortality. It is therefore important to assess the influence of these policies on observed coronary heart disease (CHD) mortality.

Our aim was thus to analyze the impact of the three health policies implemented in Portugal in reducing ACS case-fatality rates, namely the FTS for primary angioplasty, the smoking ban and the salt reduction regulation, using case-fatality rates from 2000 to 2016.

A total of 20 849 in-hospital deaths from ACS were recorded in mainland Portugal from 2000 to 2016, out of a total of 203 040 admissions for ACS.

Of the models created individually, one for each policy, the FTS showed an immediate decrease in case fatalities (beta=-1.27, p=0.003); however, it did not impact case-fatality trends after 2007, which remained steady. Similarly, the smoking ban resulted in an immediate decrease in case fatalities after its implementation (p=-0.861, p=0.05), but no significant decrease in trends was observed after 2008 (Table 1 and Figure 1).

Figure 1.

Longitudinal trends for case-fatality rates from acute coronary syndrome (percentages) for January 2000-December 2016. The vertical lines mark when the coronary fast-track system was fully implemented throughout Portugal (2007), the implementation of the smoking ban (2008), and the implementation of the salt reduction regulation (mid-2010). ACS: acute coronary syndrome.

(0.16MB).

By contrast, the salt reduction policy did not have any significant effect on case fatalities (beta=0.012, p=0.189), or any immediate impact after its implementation (beta=0.421, p=0.365; Table 1).

Using a model applying segmented regression, 2009 was identified as a year in which there was a difference in case-fatality rates. Slope trends both before (beta=-0.004; 95% CI -0.005 to -0.004) and after 2009 (p=-0.003; 95% CI -0.003 to -0.002) showed a decrease; however, after 2009 this was less pronounced (beta=0.002, p=0.004) (Table 2).

All four models were stratified for ACS case-fatality rates by gender and age. For females (beta=-1.306, p=0.020) and individuals aged ≥65 years (beta=-1.169, p=0.029), there was a significant decrease in case-fatality rates after the smoking ban. For the FTS, both sexes presented a significant decrease in case-fatality rates; however, there was no difference in trends between age groups (Figure 2).

Figure 2.

Gender- and age-stratified longitudinal trends for case-fatality rates from acute coronary syndrome (percentage) for January 2000-December 2016. (A) Males; (B) females; (C) individuals aged <65 years; (D) individuals aged ≥65 years. The vertical lines mark the years the policies were implemented. ACS: acute coronary syndrome.

(0.31MB).

In the segmented model, the year found to have a differential rate of case fatalities was consistent with the model including all the data (2009). Both trends, before and after 2009, were decreasing, though after 2009 the decrease was not as marked as for the period before 2009, as observed for the model including all the patients.

The differences in case-fatality rates before and after 2009 were higher for women (beta=0.003, p=0.029) than for men, and for individuals aged ≥65 years (beta=0.003, p=0.008; Table 2).

The seasonal pattern was consistent with that reported elsewhere,20 with higher admission rates in winter and lower rates during the summer.

In the last 11 years, a set of health policies have been implemented in Portugal to help decrease CVD mortality. The first initiative was the FTS in 2007, followed by the smoking ban in 2008 and the salt reduction regulation in mid-2010.

We created a series of statistical models, four to assess the individual impact of each of the policies, and then one to determine if there was any year during the study that could indicate a difference between trends, suggesting a possible impact of the policies in the country.

The FTS and the smoking ban led to an immediate decrease in case-fatality rates. Previous national and international studies have shown that an FTS, by substantially shortening the time between symptom onset and treatment, directly influences survival. It has been demonstrated that a symptom onset-treatment time of more than 120 min leads to increased in-hospital mortality in an almost linear fashion.21–23 However, the FTS in Portugal is mainly aimed at ST-elevation MI, which accounts for around 40% of all ACS, and so it would not be expected to affect non-ST-elevation MI, meaning that around 60% of ACS patients are not covered by this service.

The relationship between smoking and CVD mortality has been extensively demonstrated. Smoking worsens atherosclerotic alterations, including narrowing the vascular lumen and inducing a hypercoagulable state. Such changes increase the risk of acute thrombosis and thus mortality.24 The results of the current study were very encouraging regarding the smoking ban, as an immediate decrease in case fatalities was found after the ban was implemented. Since the ban targets both smokers and those exposed to second-hand smoke, case-fatality rates should decrease for both groups. Smoking cessation, and even small reductions in consumption, have been linked to decreases in CVD events and in mortality.25

Few studies have analyzed the effect of tobacco control on CVD mortality. Of those available, most focus on how regulations affect hospital admissions26; however, the impact on mortality is equally important. Nevertheless, we are not the first to analyze the impact on mortality. A recent study similarly found a reduction of up to 11% in MI mortality one year after a smoking ban was implemented.27 Another study found a 13% decrease in all-cause mortality, with a decrease of up to 26% in ischemic heart disease (IHD) mortality.

In the current study, we found that 2009 was a breakpoint in our observed trends, which supported the results from the individual models. Up to 2009 the trends were significantly impacted by the policies, leading to a significant decrease in case-fatality rates. After that year, although the trend continued to decrease it was less marked. The impact of the salt regulation in mid-2010 showed a similar effect, with no change in trends after that year. A similar pattern was observed when the impact of this policy on ACS admissions was analyzed.28

We hypothesised that although the salt regulation was an important initiative, it was highly dependent on individual adherence. The way the regulation was implemented may not have produced the desired effect on the population, perhaps due to the low level of health literacy observed in Portugal, and the fact that the policy envisioned a maximum salt content of bread of 1.4 g of salt per 100 g of bread, while other countries set lower limits.

When stratified by gender and age, our results showed that the smoking ban led to an immediate decrease in case-fatality rates for women and individuals aged ≥65 years. Although the number of women who smoke has increased in recent years, the number of men smoking was even higher, and their length of time smoking was also greater. As a consequence, men will have increased mortality from CVD, even if they quit smoking or reduce their tobacco consumption. Older people are more vulnerable to exposure to second-hand smoke, and so reducing it could lead to a decrease in case-fatalities in this group. Our findings were also consistent with another study in which post-ban reductions in IHD mortality were seen in those aged ≥65 years, but not in younger subjects.29

Analyzing the impact of the FTS by groups, we found similarly decreasing case-fatality rates for men and women, an effect previously observed in other studies in which the rate of decrease in CHD mortality was similar in both sexes.30 In addition, no immediate change in case-fatality rates was observed after implementation of the FTS for either younger or older patients. For both groups, there were significant changes in trend; however, the trends observed were increasing. The prevalence of obesity and diabetes31 is rising in Portugal, which could have weakened any reduction in the younger group. This effect has been observed in other studies30,32 in which increases in obesity lead to lower reductions in mortality for younger groups. A similar pattern was observed for the older group, which has previously been found in other datasets.32

Our results for age stratification when using segmented regression showed significant differences between trends before and after 2009, for females and older subjects. As for the non-stratified model, the decrease in trends observed before 2009 was steeper than the decrease observed afterward. These results were consistent with those observed when studying the smoking ban with age stratification, in which a significant decrease in case-fatality rates was observed for older but not for younger individuals. For the gender stratification, although the FTS analysis showed a significant decrease after 2007 for men, this was an immediate effect and not a trend.

As in any ecological study, it is not possible to directly prove any association between policy implementation and the reduction in case-fatality rates from ACS.

One of the strengths of our study is the use of a well-validated and standardized database, enabling easy comparison with other studies in different countries, particularly in other European countries. In addition to the availability of information on gender and age, this enabled us to assess the robustness of our findings among different subgroups. The time-series method is preferred over the simpler pre- and post-proportion comparison method, because it does not take pre-intervention trends into account and also enables corrections to be made for autocorrelation.33

This study extends the existing literature on the patterns of ACS mortality over time, and assesses whether public health interventions to reduce mortality by ACS were successful. Furthermore, it indicates that to decrease case-fatality rates a multifactorial strategy was needed, rather than a single approach.

Strategies such as the smoking ban and the FTS led to an immediate decrease in case-fatality rates; however, after 2009 no major decreases in ACS trends were found.

Considering that CHD constitutes an immense public health problem,34 it is crucial for different stakeholders, including decision-makers, public health authorities, medical societies and the cardiology community, to keep working together to reduce ACS mortality rates.

The authors have no conflicts of interest to declare.