Article Text

Declining trends in acute myocardial infarction attack and mortality rates, celebrating progress and ensuring future success
1. Martin O'Flaherty1,
2. Mark D Huffman2,
3. Simon Capewell1
1. 1Department of Public Health and Policy, University of Liverpool, Liverpool, UK
2. 2Department of Preventive Medicine, NorthWestern University Feinberg School of Medicine, Chicago, Illinois, USA
1. Correspondence to Dr Martin O'Flaherty, Department of Public Health and Policy, University of Liverpool, Liverpool L69 3GB, UK; moflaher{at}liv.ac.uk

## Statistics from Altmetric.com

The huge falls in coronary heart disease (CHD) mortality rates which started in Western countries in the 1960s have continued into the 21st century. They represent profoundly important and much studied epidemiological phenomena. The WHO's Multinational MONItoring of trends and determinants in CArdiovascular disease (MONICA) project, Atherosclerosis Risk in Communities (ARIC) community-wide surveillance and hospital discharge study and IMPACT model studies in diverse populations highlighted the crucial contributions of decreases in both CHD incidence and case fatality to overall trends. However, such favourable trends in CHD mortality have not been universal and absolute numbers of deaths have been generally increasing throughout the world because of population growth and ageing. In their Heart paper, Degano et al1 explore recent myocardial infarction trends in several European countries to provide a much-needed, detailed update to previous trend analyses for the new millennium.

The authors performed a well-conducted analysis, using rigorous and validated protocols in well-run population-based registries in six countries in the European Union. The study was conducted using high-quality MONICA methodology, an exemplar in cardiovascular disease epidemiology. Crucially, they paid particular attention to changing acute myocardial infarction (AMI) definitions, particularly in the era of highly sensitive cardiac biomarkers, and adjusted their estimates accordingly.

Degano et al examined the period 1985–2010, from the start of the MONICA project to a decade into the new century. This period reflects a rapidly changing era in cardiovascular medicine, which spans a period when new evidence-based treatments were progressively adopted. Over the study period, the authors found a consistent decline in AMI incidence and mortality rates, including both total and case fatality rates. However, they importantly observed substantial differences across populations. For example, annual percentage declines in incidence rates ranged from −0.6% (Girona) to −4.4% (Finland), with similar degrees of variations in mortality and case fatality rates. Given this heterogeneity, it is difficult to interpret the pooled results (even excluding Tallin, an obvious outlier) and to some extent might be attributable to surveillance bias. Nevertheless, the results are generally reassuring and suggest that progress in AMI prevention and in acute care continues.

MONICA and a variety of modelling studies suggest that this mortality fall might reasonably be primarily attributed to improving risk factors (decreasing incidence), with an increasing contribution from acute, evidence-based treatments (to reduce case fatality). The findings of this study support this hypothesis. The apparent improvements may also have been exaggerated by the progressive decrease in disease severity—the data were not ‘biomarker adjusted’. It is thus difficult to interpret the attack rates reported in this study in terms of true incident events. However, they are consistent with similar analyses in other populations. For instance, in England, Smolina et al2 reported a consistent decline in AMI incidence over the period 2002 to 2010 with 57% attributable to the decline in event rates. Likewise in the USA, a similar decline was mostly attributable to a decrease in incident events rather than recurrent ones.3 This of course is not necessarily universal. For example, Australia showed less favourable AMI incidence trends4 while many low-income and middle-income countries are experiencing increasing mortality rates, likely reflecting increases in incidence and cardiovascular risk factor levels. Furthermore, if incidence is truly declining, parallel declines in pre-hospital CHD-related deaths should be expected. This has been consistently described in other cohorts such as Framingham, ARIC and Rotterdam, where the falling trend in sudden cardiac deaths also declined in the context of incidence and total mortality declines. Similar declines were observed with most populations included by Degano et al, but not in Tallin, raising the possible explanation of less robust data in some populations.

The age ranges reported in the study reflect an eminently preventable morbidity and mortality burden. However, this decision limits the ability of these trends to help us understand the evolution of the CHD epidemic, as a substantial portion of the burden is concentrated in old age, also the age group growing fastest in size. Population growth and ageing may still lead to increased numbers of events despite falling rates but not inevitably if rates continue to decline at a fast pace.

Degano et al concluded that improvements in primary and secondary prevention are important contributors to this trend. However, they called for country-specific preventative policies, tailored to specific age and gender groups. While this template might appear rational given the country-level, gender-level and age-level heterogeneity, are ‘bespoke’ prevention policies the best, or at least the only, way forward? Although prevention strategies targeted at high-risk groups and population subsets might appear superficially attractive, they have proved consistently less powerful than population-wide policy interventions. Furthermore, targeted strategies rely mostly on healthcare-based solutions, which are often more costly and complex while increasing health inequalities.

There is substantial evidence suggesting that population-level interventions, usually as a result of relatively small changes in risk factor distributions across the entire population, deliver enormous and rapid reductions in mortality. Many natural experiments support this, including the large Finnish reduction in mortality experienced since the 1970s, and the more recent examples from Poland and Cuba.5 In all these cases, the dramatic mortality reductions can be attributed to population-level reductions in risk factors reflecting improvements in smoking prevalence and diet (with resulting falls in blood pressure and total cholesterol).5 The populations included in this analysis have experienced similar favourable changes in risk factors over the period studied. Most of these positive changes can be attributed to improvements in diet and lifestyle rather than healthcare primary prevention medications. Modelling studies suggest that the contribution of statins and hypertension primary prevention treatments to the overall CHD mortality reductions were modest: 9% in Italy, 10% in Spain and 12% in the USA.6

Reductions in incidence will reduce pressure on heavily burdened healthcare systems and should therefore result in net savings, as well as mortality and morbidity reductions. Optimising the delivery of acute treatments and primary and secondary prevention is thus essential. In the UK, uptake rates for CHD treatments have increased substantially recently and achieving optimal use of these treatments seems to be feasible.

However, far larger benefits would be delivered by population-level (structural) interventions. Thus in the UK, systematic efforts to decrease salt in processed food have achieved a 15% reduction in salt intake of about 1 g in 10 years, saving at least 6000 lives and £1.5 billion annually. Mandatory policies might achieve even bigger benefits, annually gaining 195 000 to 390 000 quality-adjusted life-years and $10 billion to$24 billion in healthcare savings in the USA.7 Similarly, large gains could come from stronger tobacco policies, banning trans fat elimination or fiscal strategies to improve diet.

The study by Degano et al suggests that the recent declines in CHD incidence and mortality are likely set to continue in these populations. However, such trends are not set in stone. For example, the cholesterol decline observed in Finland over several decades is now reverting. A similar phenomenon in Sweden has been attributed to population-level changes in diet, in spite of substantial increases in statin use.8 These unfavourable trends in a major cardiovascular risk factor will surely affect disease incidence and mortality. Likewise in Cuba, the dramatic decline in mortality commencing in 1990 was then followed by a rise shortly after economic recovery.

Investing in optimal acute care, primary prevention and secondary prevention are certainly important policy targets. However, countries also need population-level prevention policies. The massive reductions in disease burden and costs that can be achieved with improvements in diet and tobacco use could accelerate and sustain these recent downward mortality trends. Only then will this crowning glory of cardiovascular medicine continue into the 21st century. These strategies will also clearly need to be extended to low-income and middle-income countries, where resources are scarce and the burden is frequently even greater. Then the WHO 25×25 goals will become eminently achievable.

View Abstract