Article Text

Download PDFPDF

Declining cardiovascular mortality masks unpalatable inequalities
  1. Alastair H Leyland,
  2. Ruth Dundas
  1. MRC/CSO Social and Public Health Sciences Unit, University of Glasgow, Glasgow, UK
  1. Correspondence to Professor Alastair H Leyland, MRC/CSO Social and Public Health Sciences Unit, University of Glasgow, Glasgow G2 3AX, UK; alastair.leyland{at}

Statistics from

Mortality from cardiovascular disease is declining in many European countries, but they remain socially patterned. Di Girolamo and colleagues investigate how inequalities in cardiovascular mortality have changed in 12 countries since the 1990s.1 In general, they appear cautiously optimistic, describing trends in such inequalities as ‘favourable overall’ while noting that further improvement is an important aspiration.

The paper raises an interesting (and old) question: which are more important, relative or absolute inequalities? The authors present both, which is good practice. It is not possible to summarise the distribution of mortality across social groups in a single number, no matter how much we might wish this to be the case. This fact is recognised by some policymakers; in Scotland, for example, the long-term monitoring of health inequalities2 includes publication of the relative index of inequality (detailing the magnitude of the inequality gradient), the absolute gap (the difference between groups at the extremes of the social spectrum) and the scale (indicating the magnitude of the problem).

To gauge the extent to which current trends should be regarded as favourable, we can examine future mortality. Di Girolamo et al provide the means to project cardiovascular disease mortality rates by occupational class (based on their Supplementary Tables S7 and S10). In table 1, we present various scenarios enabling us to look at potential future inequalities for upper non-manual employees and manual workers.

Table 1

Projected scenarios for upper non-manual employees and manual workers under assumptions that absolute or relative changes in cardiovascular disease mortality are maintained

First, if we assume that mortality rates continue with the absolute declines experienced between 1990–1994 and 2010–2014, based on the most recent reported rates the mortality rate in upper non-manual employees will fall to zero in most countries between 4 and 16 years. (The exception is Lithuania where the very slow decline seen between 2000–2004 and 2010–2014 of under 1 per 100 000 means that it would take nearly 400 years to fall to zero.) In Italy (Turin), the mortality rate for manual workers will fall to zero at about the same time as for upper non-manual employees; in other countries, it will take 2–6 years longer for the rate in manual workers to fall to zero than for upper non-manual employees. (The exception is again Lithuania, where an increasing mortality rate among manual workers means that its projection will never reach zero.) Although these are not big differences, it means that there are periods during which cardiovascular disease between 35 and 64 years becomes a disease of the poor and therefore relative inequalities—measured using a simple ratio of the mortality rates in the two groups—are infinite.

Under a potentially more realistic scenario, we assume that the relative declines in mortality experienced between 1990—1994 and 2010—2014 will continue. By 2020—2024, absolute inequalities (the difference between the two groups) will have decreased in every country apart from Lithuania, the difference falling to as little as 10 or 15 per 100 000 in Italy (Turin) and Austria respectively. Relative inequalities, however, will increase in every country apart from Italy (Turin). In Finland, for example, the mortality rate in manual workers will be more than three times that in upper non-manual employees.

What do these future scenarios tell us about the importance of absolute and relative inequalities, and of progress towards the reduction of inequalities? Whatever the absolute level of reduction, a disease becoming the sole preserve of the more disadvantaged is not a desirable situation and does not suggest progress. On the other hand, faster relative reductions in the more advantaged groups mean that it will take longer for the mortality rate of the disadvantaged group to fall by a fixed proportion. Again, if the mortality rate among upper non-manual employees in Finland will halve in 14 years while it will take 19 years for the mortality rate to halve among manual workers—and that from a considerably higher starting point—then it is difficult to describe this as progress in reducing inequalities.

So continuing down the same path is unlikely to lead to a favourable situation in most countries. Something needs to change if relative inequalities are to decrease. Recent declines in mortality from coronary heart disease in Scotland are almost equally attributable to improvements in medical treatment and reductions in risk factors.3 It is not enough for these to be experienced equally across social groups if we are to see inequalities fall; future declines in mortality must be more pronounced among the more disadvantaged. An effective and fair health system will be responsive to the needs of its population regardless of social circumstances; perhaps the greatest opportunity for medical care to reduce inequalities is if more disadvantaged groups are encouraged to seek healthcare earlier in the progress of the disease. The greatest opportunity to reduce inequalities, however, must be through the modification of lifestyle risk factors including smoking, alcohol consumption, diet and physical activity. These can bring about rapid change,4 show strong social patterning and are amenable to population-wide intervention such as through policy.5 6

We know that the most successful preventive measures (and those most likely to reduce inequalities) are those that change environments at the whole population level rather than individual behaviours.7 Using tobacco reduction as an example, it was known for many years that smoking was detrimental to health but, despite many individual behaviour change interventions delivered through the health service, smoking prevalence remained high, and higher among more disadvantaged groups. It was not until smoking bans in public places were introduced, which impacted at a population level and on everyone in the population equally, that larger reductions in smoking prevalence and subsequent mortality from smoking-related causes were seen.8

But even risk factor reductions have the potential to disappoint. We have seen, for example, the levels of alcohol-related harms to be greater in more disadvantaged groups at given levels of alcohol consumption.9 Such effect modification means that an even greater reduction in risk factors in the more disadvantaged groups will be needed to produce the larger reduction in mortality rate needed to decrease inequalities.

Cardiovascular disease contributes substantially to overall inequalities in mortality, and as such reducing inequalities in cardiovascular mortality is an important part of the fight against inequalities. Di Girolamo et al show that cardiovascular mortality has reduced substantially over the past 20 years, but remains strongly patterned such that faster declines are seen among the better off. If the current trends continue, without any additional interventions in the more disadvantaged groups, it will take between 8 and 28 years in most countries for cardiovascular disease mortality among manual workers to fall to the levels currently experienced by upper non-manual employees. Is this fair?


The Social and Public Health Sciences Unit is funded by the Medical Research Council (MC_UU_12017/13) and the Scottish Government Chief Scientist Office (SPHSU13).



  • Twitter @AlastairLeyland

  • Contributors AL and RD collaborated on all stages and drafts of this editorial.

  • Funding Dr Leyland reports grants from Medical Research Council (MC_UU_12017/13), grants from Chief Scientist Office (SPHSU13) during the conduct of the study.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • 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.

Linked Articles

  • Health care delivery, economics and global health
    Chiara Di Girolamo Wilma J Nusselder Matthias Bopp Henrik Brønnum-Hansen Giuseppe Costa Katalin Kovács Mall Leinsalu Pekka Martikainen Barbara Pacelli José Rubio Valverde Johan P Mackenbach