Article Text

other Versions

Contributions of treatment and lifestyle to declining CVD mortality: Why have CVD mortality rates declined so much since the 1960s?
  1. Martin O'Flaherty1,
  2. Iain Buchan2,
  3. Simon Capewell1
  1. 1Department of Public Health and Policy, University of Liverpool, Liverpool, UK
  2. 2North-west Institute for Bio-Health Informatics, School of Community Based Medicine, University of Manchester, Manchester, UK
  1. Correspondence to Dr Martin O'Flaherty, Division of Public Health and Policy, Whelan Building, Quadrangle, University of Liverpool, Liverpool L69 3GB, UK; moflaher{at}


Developed countries have enjoyed substantial falls in cardiovascular disease (CVD) mortality. However, low and middle income countries are drowning in a rising tide of CVD and other non-communicable diseases. Current and future trends in CVD mortality will therefore require increasing attention in the 21st century. The success of clinical cardiology in providing evidence-based cost-effective treatments should be celebrated. However, the growing understanding of CVD mortality trends highlights the crucial role of tobacco, diet, alcohol and inactivity as key drivers. Pro-active public health approaches focused on ‘upstream’ population-wide policies are increasingly recognised as being potentially powerful, rapid, equitable and cost-saving. However, the future political challenges could be substantial.

Statistics from

A number of recent papers have addressed the very important issue of the large declines in coronary heart disease (CHD) mortality since the 1960s. They dispute the relative contributions made by lifestyle and by treatments. Nabel and Braunwald recently attributed the mortality declines almost entirely to treatment effects.1 The doyen of cardiology thus made many other cardiologists proud. However, Braunwald's paper was illustrated by a time line which attracted a critical correspondence that emphasised that the declines in coronary mortality mapped quite closely to declines in smoking! Furthermore, three subsequent papers in the BMJ with an accompanying editorial by Tunstall- Pedoe all highlighted the potentially powerful contributions of decreases in other major risk factors such as cholesterol and blood pressure.2–5

Who is right? This review will briefly consider the evidence under the following headings.

  1. Recent coronary heart disease mortality time trends in different populations.

  2. What is driving these disease trends?

  3. What are the implications for clinical care?

  4. What are the implications for public policy?

Recent coronary heart disease mortality time trends in different populations

During the late 20th century, cardiovascular disease (CVD) mortality rates fell by 50–80% in most high income countries including the US, UK and Western Europe. The reduction in mortality particularly in middle and old age contributed to the continuous increase in life expectancy. Further, the declining trends in CHD have generally been mirrored by falls in ischaemic stroke,6 hence the wider narrative in this editorial deliberately spanning time trends across CVD—both coronary disease and stroke.

The changes in CVD mortality rates can be surprisingly rapid. One striking example is the sharp decline in CVD mortality observed in Poland and neighbouring countries in the 1990s following the collapse of the Soviet Union.4 ,7 ,8 All age and gender groups showed the same abrupt shoulder and decline, which have since continued through to at least 2006.4

However, these CVD mortality falls are not necessarily set to continue.9 Mortality trends neither are self perpetuating nor set in stone. Furthermore, the longstanding falls in coronary mortality in England, Scotland, Ireland, Australia and the Netherlands have recently slowed or even plateaued in young adults.10–13 This is a major wake up call for somnolent policy makers.

What is driving these disease trends?

In the last half of the 20th century, the development of safe and effective interventions to treat CVD patients has underpinned the triumph of the evidence based medicine paradigm. Effective treatments for most forms of CVD are now available and increasingly used. However, the question of their relative contribution as drivers of CVD mortality falls is still being debated.14

As Braunwald emphasised, the evolving understanding of CVD causation represents one of the great scientific achievements in the 20th century. Several landmark studies helped to define the causative role of risk factors, notably the Seven Countries and Framingham studies.15

Furthermore, it is now generally accepted that most of the disease burden from CVD (and also other non-communicable diseases -NCDs) can be attributed to just four lifestyle factors—smoking, diet, alcohol and inactivity.16 In the MONICA analysis of populations with falling CHD mortality, death rates fell by about 4% per year. Approximately three quarters of that observed fall could be attributed to the decline in event rates, (mainly reflecting risk factor improvements), while about one quarter could be attributed to a decrease in case death (mainly reflecting medical treatments).17

The MONICA analysis was unable to quantify how much of these falls could be attributed to particular evidence based treatments or to changes in specific risk factors.17 ,18 However, subsequent modelling studies were then able to examine this issue in more detail in a wide range of populations.19–21 For instance, the US CHD Policy Model suggested that in the US population for the period 1980–1990 risk factor changes contributed approximately 50% to the mortality decline while treatments contributed about 43%.20

IMPACT is a deterministic model developed by ourselves since in the late 1990s. Using a very large EXCEL spreadsheet, this model combines extensive data from many sources on patient numbers, treatment uptake, treatment effectiveness, risk factor trends and consequent mortality effects. The deaths prevented or postponed over a specified period are then calculated. The IMPACT model can be used to estimate the proportion of the observed change in mortality which can be attributed to specific treatments or risk factor changes. Have created a model which crudely defines a specific population, it is then possible to conduct policy analyses: estimating the future consequences of altering treatment strategies or changing specific population risk factors.

IMPACT has been used to explore the contributions of risk factors and treatments in over 15 countries where mortality rates have been declining, including England and Wales, Scotland, Finland, Sweden, Italy, Spain, Iceland, Northern Ireland, New Zealand, USA and Canada. The IMPACT model found that in different populations about 40% to 75% of the CHD mortality fall could be attributed to risk factor changes and the remaining 25% to 55% of the fall to treatments.19 In the US population for the period 1980–2000, the IMPACT Model suggested that risk factor changes explained approximately 44% of the mortality decline while treatments contributed around 47%.19 In the UK, corresponding risk factor/treatment contributions were about 58%: 42% for the period 1981–2000, decreasing to about 34%: 52% in the very recent period 2000–2007.22 ,23 This decrease mainly reflected the increasing negative effects from rising obesity and diabetes rates.23

Similar results were seen in Southern Europe24 ,25 and Central Europe.4 In Poland, the dramatic fall in CHD mortality after the collapse of the communist economy was attributed mainly to population level changes in risk factors, notably falls in total cholesterol reflecting major improvements in diet.7 ,8

Furthermore, Finland and other Nordic countries experienced even more marked declines in CHD mortality during the late 20th century, with correspondingly larger contributions from reductions in major risk factors, particularly cholesterol. Serum cholesterol declined substantially from 1970, by over 1 mmol/l in both men and women over in Finland (mainly reflecting the implementation of effective nationwide, population level policies. rather than statins).26 Blood pressure declined to 2002, but levelled thereafter. Smoking followed a more complex pattern, declining in men but modestly increasing in women until 2002, and then levelling off. BMI increased in men and in women from about 1980. Collectively, the changes in risk factors explained over 70% of the very large observed declines in CHD mortality between 1982 and 1997, while medical treatments explained 23%.27 Remarkably similar changes were reported for Iceland between 1985 and 2005.28

The same factors have also driven the recent dramatic increases in CVD mortality rates in middle and low income countries. In China, CHD mortality increased rapidly from 1984, mainly driven by persistent smoking and large increases in cholesterol levels (reflecting Westernisation of the diet).29 These adverse trends in risk factors, compounded by ageing of the population, will probably substantially increase the future CHD burden in China.30 ,31

Analysis of recent English trends using the IMPACTsec model suggests that approximately half the CHD mortality fall between 2000 and 2007 was attributable to improved uptake of treatments.23 These benefits occurred evenly across all social groups, which represent a notable achievement for the NHS However, the risk factor contributions came mostly from population falls in blood pressure. Recent decreases in other potentially powerful cardiovascular risk factors such as smoking and cholesterol were frustratingly modest, and were further negated by continuing rises in obesity and diabetes.

Our current understanding of the drivers of CHD mortality trends thus highlights the crucial role of lifestyle and dietary factors, and crucially emphasises that changes can occur surprisingly quickly.32 Further reductions in risk factors levels now seems the most appropriate target, especially if our aim is not just control CVD morbidity and mortality, but also to reduce inequalities.33

What does all this mean for future CVD treatment strategies?

What are the implications for clinical care?

Current and future trends in CVD mortality will require increasing attention in the 21st century. The rapid response of mortality trends to changes in risk factors acting at the population level emphasises the importance of preventive strategies to decrease the future cardiovascular burden. It also reminds us about the potentially high price of complacency.

The success of clinical cardiology in providing cost-effective treatments that are solidly based on scientific evidence needs to be celebrated. However, treatment uptake is still less than optimal and the real world efficacy of treatments is further decreased by variable sustainability with medications and often mediocre adherence.19 Substantial additional gains could be achieved by providing appropriate treatments for a larger proportion of existing patients.34 However, this will require further resources and may therefore be considered unaffordable for many low and middle income countries (or even in more affluent societies during the current recession).

What are the implications for public policy

Many recent analyses have focused on which strategy might better serve to prevent or control future CVD and other NCDs.35 ,36 As debate continues, ‘upstream’ population wide policies are increasingly emerging as both powerful and rapid.32 A pro-active public health approach also makes strong scientific sense.

Such ‘upstream’ measures might include controlling tobacco and increasing physical activity. Promoting healthier diets could include re-formulation (improving the contents of processed food products); restricted marketing of junk food (energy dense, nutrient poor, processed foods high in fat, salt and sugar); food package labelling, taxation of sugary drinks and junk food, and subsidies making healthier foods more affordable, such as fruit vegetables and wholefoods.37 Furthermore, such options could be substantially cost saving.35 Crucially, when these strategies are implemented across the entire population, they tend to decrease inequalities, thus offering additional gains for deprived groups.33

However, some of the most powerful policy-based prevention strategies are politically challenging, such as implementing plain packaging of cigarettes, or banning industrial transfats. Many politicians would prefer to emphasise individual responsibility, or rely on less contentious but weak voluntary agreements with the food industry.38 However, commercial partnerships defy both common sense and the emerging evidence.39 The food industry aims to make money, not promote public health. And public awareness is growing about the powerful roles that the processed food manufacturing, distribution and marketing industries are playing in maintaining the CVD and obesity epidemics. The emerging tactics of ‘Big Food’ multinationals are proving surprisingly similar to the harmful roles historically played by Big Tobacco and Big Alcohol.40 The last decade witnessed impressive successes in tobacco control.41 Thus, a more critical societal engagement with the food industry could powerfully shape future policy decisions.

Policy makers may thus need to overcome their natural reluctance to legislate. Indeed, this reluctance is eroding as the situation grows worse. The ‘epidemics’ of childhood obesity and adult diabetes continue, in spite of numerous well meaning but mostly ineffective initiatives. And the previous fall in CVD mortality rates is now flattening. Some leaders have already realised that more decisive actions could bring large and rapid benefits. In the US for instance, nationwide labelling targeting industrial transfats has now been synergised by food reformulation and pressure on several fast food outlets.42 Blood levels of transfats have halved in the last decade.43

Globally, most middle income and low income countries are still experiencing the rapid increase phase of their cardiovascular epidemics. In 2011, this finally led to the first UN High level summit on the prevention and control of Non-Communicable Diseases.44 The aim was to engage member countries in a concerted and accountable effort to tackle one of the key challenges of the upcoming decades. The participant countries have now agreed the ‘WHO 25×25 goals’—a 25% reduction in non-communicable disease deaths by 2025.44

However, decisive action will now be required to achieve these goals. The global burden of non-communicable diseases is rapidly increasing, and the control of the CVD epidemic across communities and populations is vulnerable to ‘adoption delays’. For example, the CHD mortality decline in the US commenced at the beginning of the 1960s in California, Maryland and the District of Columbia; but only started 5 years later in the more deprived Southeast states.45 Likewise, mortality deaths in Western European countries started to decline in the 1970s–1980s, but falls only occurred in Central Europe after 1990.32 Meanwhile mortality rates are still rising in Eastern Europe, Central Asia and beyond.

In conclusion, high income countries have enjoyed substantial falls in CVD mortality, while low and middle income countries are drowning in a rising tide of CVD and other non-communicable diseases. The scientific knowledge of what we need to do is already substantial—the evidence of the effectiveness of policy interventions is considerable. And the examples of the ‘early adopters’ are now more widely known. More effort should therefore be directed to supporting the adoption of effective strategies by more countries.46 Population-wide interventions aimed at the major tobacco, diet and lifestyle determinants could produce large, rapid and cost saving benefits. But the political challenges to implement these policies are not small.

Fortunately, we know what works. Action is now essential, because any further delays would cost lives and money.


View Abstract


  • Funding MOF was supported by the UK Medical Research Council and the MedCHAMPS project and funded by EC FP7 grant no. 223705. SC and IB are supported by the UK Higher Education Funding Council.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.