Trends in Mortality From Ischemic Heart Disease and Cerebrovascular Disease in EuropeClinical PERSPECTIVE
1980 to 2009
Background—Trends in cardiovascular mortality across Europe demonstrate significant geographical variation, and an understanding of these trends has a central role in global public health.
Methods and Results—Ischemic heart disease and cerebrovascular disease age-standardized death rates (as per International Classification of Diseases, ninth and tenth revisions) were collated from the World Health Organization mortality database for member states of the European Union. Trends were characterized by using Joinpoint regression analysis. An overall trend for reduction in ischemic heart disease mortality was observed, most pronounced in Western Europe (>60% for the Netherlands, United Kingdom, and Ireland) for both sexes from 1980 to 2009. Eastern European states, Romania, Croatia, and Slovakia, had modest mortality reductions. Most recently (2009), Lithuania had the highest mortality for males and females (318.1/100 000 and 166.1/100 000, respectively), followed by Latvia and Slovakia. France had the lowest mortality: 39.8/100 000 for males and 14.7/100 000 for females. Analysis of cerebrovascular disease mortality revealed that Austria had the largest reduction for both sexes (76.8% males, 76.5% females) from 1980 to 2009. The smallest improvement over this period was seen in Lithuania, Poland, and Cyprus (–5% to +20% approximately). France has the lowest present-day cerebrovascular disease mortality for both males and females (23.9/100 000 and 17.3/100 000, respectively).
Conclusions—There is a growing disparity in cardiovascular mortality between Western and Eastern Europe, for which diverse explanations are discussed. The need for population-wide health promotion and primary prevention policies is emphasized.
Cardiovascular disease is the leading cause of mortality worldwide1 and trends in cardiovascular mortality have a crucial role in public health and epidemiology globally. This is an area of academic interest and of great geographical variation.2–4
Clinical Perspective on p 1926
Analysis of cardiovascular mortality data relating to the Americas has demonstrated an inequality between the United States and Canada and the Latin American nations in age-standardized decline in cardiovascular mortality.5 Previous studies on trends in Europe have highlighted a similar health gap between Western countries and Eastern, previously Communist-run, states.2–4,6–8 Moreover, a Northeast to Southwest Europe gradient was identified in the 1980s.6 Various hypotheses have been generated for this disparity, including lifestyle factors, diet, and socioeconomic influences.7
No recent analysis comparing trends in cardiovascular mortality throughout Europe has been performed. The objective of this study was to describe current mortality rates for ischemic heart disease (IHD) and cerebrovascular disease (CVD) across Europe. We also aimed to describe changes in IHD and CVD mortality between 1980 and 2009 and identify trends by using Joinpoint analysis.
Mortality data for CVD and IHD for European countries were derived from the World Health Organization (WHO) mortality database for the years 1980 to 2009.9 The WHO evaluates the quality of the data to ensure comparability and reliability, without adjustment for underreporting.10 Member states of the European Union (EU; 28 countries as of 2015) were selected to produce a defined group for analysis.11 For simplicity and clarity, mortality was restricted to CVD and IHD, excluding mortality from other cardiovascular causes. The mortality database is updated annually from deaths registered by national civil registration systems, International Classification of Diseases (ICD) coded, per 100 000 population. Data are collected according to ICD ninth and tenth revisions, and are collated according to the tenth revision, a process that has been used previously with adequate validity and data robustness.5 ICD codes were used for CVD mortality (430–438, I60–I69) and IHD mortality (410–414, I20–I25) for the ninth and tenth revisions, respectively.
The age-standardized death rate was calculated, defined as mortality weighted to the distribution of mortality per 5-year age group, according to the WHO standard population and world average age structure for 1998. This removes the effects of historical events on age structure and controls for differences in age structure in populations, producing age-specific mortality rates and more representative data.12
The estimated level of coverage for deaths with a recorded cause for death is calculated by actual reporting divided by estimated mortality rate. Population and birth recording in all countries exceeds 90%, as per the WHO standard for inclusion in the database.13 Mortality data were missing in a small subset of countries for ≥1 calendar years. When values were not available, data from previous or subsequent years were replicated in a last observation carried forward method. A total of 4.7% of data was replicated from last and subsequent observations. Sex-specific trends in cardiovascular mortality were established and analyzed for 3-year periods every decade; in 1.1%, averages were from 2-year rather than 3-year periods.
Joinpoint regression analysis with annualized data (between 1979 and 2009, where available) was used to assess changes in linear slope for mortality trends over time,14 as described previously by Rodriguez et al.5 In brief, Joinpoint analysis assesses the overall trends in mortality initially with no Joinpoints and tests for significant changes in the model with the sequential addition of points where there is significant change in the slope of the line. Statistical trends were assessed by using Joinpoint software (Version 184.108.40.206) provided by the US National Cancer Institute Surveillance Research Program (http://surveillance.cancer.gov/joinpoint/). The model also computes an estimated annual percentage change for each trend by fitting a regression line to the natural logarithm of the rates.
Trends in IHD and CVD mortality are calculated as crude absolute differences between averaged 3-year data points for the earliest and most recent years available.
Differences in the change in mortality between new EU states and their older counterparts from 1980 to 2009 were assessed with the Wilcoxon rank-sum test. EU-joining nations from 2004 or later were deemed new: such countries include Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, and Slovenia.11 Of note, Cyprus was excluded from this analysis because of significantly outlying results based on few data points.
Post Hoc Investigation
From these a priori Joinpoint analyses, we observed a common trend in the decrease in mortality from IHD in the years after the end of Communism. We performed a single post hoc analysis to explore the overall effect that the end of Communism had on age-standardized death rates from IHD. For this investigation, all Communist countries were normalized to the year of the end of Communism, which is taken to be the reference year for trend analysis. We fitted a locally weighted scatterplot smoothing line to age-standardized death rates in the interval from 10 years before the reference year until 20 years after the reference year. Locally weighted scatterplot smoothing regression is a standard nonparametric smoothing procedure that makes minimal assumptions about the data and corrects for the influence of outliers from the trend to obtain a more robust trend estimate. Plots with 95% confidence bands from this analysis were inspected visually to ascertain trends in the data. Locally weighted scatterplot smoothing regression analysis was performed by using SAS software v9.4 (SAS Institute, Cary, NC).
Over the 30-year period studied, significant changes in cardiovascular mortality were observed across the EU. All countries have an estimated level of coverage of deaths that are registered with cause of death data of >95%, with the exception of Cyprus (>78.2%) and Croatia (>94.9%).13 The quality and robustness of the data collected has been reviewed previously.15 Of the 28 countries included in the analysis, 24 had quality of reporting graded as high or medium. Four of the countries (Cyprus, Greece, Poland, and Portugal) assessed in this analysis were graded to have low reporting. Values were replicated from last and subsequent observations in 4.7% of data points because of incomplete reporting.
There was a statistically significant greater mortality reduction for both IHD (P<0.01) and CVD (P<0.01) in the comparison of older and newer (2004 or later) EU countries. This is displayed in Figure 1A and 1B for IHD and CVD, respectively.
Trends in IHD Mortality
Figure 2 and Table I in the online-only Data Supplement show IHD mortality from 1980 to 2009 for each country per sex. No data were available for EU member states Malta and Luxembourg on the WHO mortality database, so Malta and Luxembourg were omitted from further analysis.
The overall trend in IHD mortality across Europe was observed to be decreasing steadily over time. Western European countries and founder EU member states showed the greatest reduction in mortality (Figure 2). The largest decrease was seen in the Netherlands (73.8% males, 72.0% females) and United Kingdom (67.3% males, 65.9% females) from 1980 to 2009. Very modest reductions in mortality were observed in previous member states of the Soviet Union and Eastern Europe from 1980 to 2009, such as Lithuania (–13.1%, –30.4%), Poland (–14.4%, +16.1%), Hungary (–13.1%, –1.9%), and Slovakia (–7.0%, +4.8%) for males and females, respectively. Furthermore, unfavorable trends in IHD mortality were observed from 1980 to 2009 in Croatia (+43.3%, +124.5%), Cyprus (+418.3%, +320.1%), and Romania (+30.2%, +14.8%) for males and females, respectively. It should be noted that the IHD mortality figures for Cyprus (+418.3% for males and +320.1% for females) are based on only 2 time points with low mortality rates, adding ambiguity to the interpretation of these data.
Current IHD Mortality
Figure 3A and 3B depicts age-standardized death from IHD per 100 000 males and females, respectively, in 2009, to provide the most recent overview of mortality rates in EU member states. However, for Denmark and Belgium, 2006 data were used, and 2008 data were used for France and Italy; these were the most recent figures available. Lithuania had the highest IHD mortality for both males and females (318.1/100 000 and 166.1/100 000, respectively). France had the lowest mortality for both males (39.8/100 000) and females (14.7/100 000). There is significant disparity between EU member states with the highest and lowest IHD mortalities (France and Lithuania, for 2008 and 2009, respectively): ≈8-fold for males and 11-fold for females.
Trends in CVD Mortality
Figure 4 and Table II in the online-only Data Supplement display CVD mortality from 1980 to 2009 for both males and females. Overall, there was a significant reduction in mortality across Europe; Austria had the largest reduction for both sexes (76.8% males, 76.5% females). Western countries exhibit faster declining trends in mortality from CVD than IHD, with the exception of Denmark, Sweden, and the Netherlands. Over the 30-year period for males, the least improvement is seen in Slovakia (–4.3%), whereas Romania (+7.4%), Lithuania (+13.7%), Cyprus (+14.4%), and Poland (+16.6%) all report an increase in male CVD mortality. Similarly, for females, Lithuania (–5.4%) and Poland (–4.5%) report the smallest improvements, with Cyprus (+23.6%) having increasing CVD mortality over the study period.
With the use of the 17 countries (Austria, Belgium, Bulgaria, Estonia, France, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, the Netherlands, Poland, Portugal, Romania, Spain, and the United Kingdom) with available mortality data for 1980 to 1982, all, with the exception of Portugal, demonstrated a smaller sex gap with CVD than IHD mortality (average difference 19.4/100 000 versus 103.7/100 000, respectively). This sex disparity reduced over the course of the study period, with a reduction of 18.1% for CVD and 36.1% for IHD mortality from 1980 to 1982 to 2007 to 2009.
Current CVD Mortality
Figure 3C and 3D display the most recent calendar year data (2009) for CVD mortality, again substituting 2006 for Belgium and Denmark and 2008 for France and Italy. For both sexes, Bulgaria (154.2/100 000 for males, 112.7/100 000 for females), Romania (146.4/100 000 for males, 111.3/100 00 for females), and Latvia (121.3/100 000 for males, 84.8/100 000 for females) had the greatest age-standardized death rate. As with IHD mortality, France had the lowest CVD age-standardized death rate for both males and females (23.9/100 000 and 17.3/100 000, respectively).
Joinpoint Analysis for IHD Mortality
Tables 1 and 2 present the results of IHD mortality Joinpoint analysis for 1980 to 2009 for males and females. We report significant trend changes in the data and the estimated annual percentage change in mortality for periods covered by each trend. For both sexes, there is a general favorable trend in IHD mortality across Europe, with the Netherlands, Portugal, Bulgaria, and the United Kingdom having rapidly accelerating declines at the present day. Male mortality declined at a consistent rate across the study period, with only 1 Joinpoint in Denmark, Finland, Sweden, and Slovakia. Mortality rates in Denmark, Latvia, and Sweden also declined at a steady rate in females. Slovakia demonstrated increasing mortality rates in females with a single Joinpoint. Initially increased mortality trends, followed by a subsequent reduction, were seen for Bulgaria and Greece for both sexes. Very variable trends with 3 or 4 Joinpoints, fluctuating between increasing and decreasing mortality rates, were seen in the Eastern European countries of Croatia, Lithuania, and Poland for males and females. Similar inconsistent trends were also reported for Slovenia, Poland, Latvia, and Austria for males, and Romania for females.
Joinpoint Analysis for CVD Mortality
Tables 3 and 4 display similar data for CVD mortality. There is an overall downward CVD mortality trend, with the notable exceptions of Slovakia and Slovenia, which have rapidly increasing mortality for both sexes. The most recent data for Ireland revealed a modest increase in male mortality and a slowing decline in female mortality. Initially increasing mortality followed by favorable trends was seen with Lithuania, Poland, and Romania for both sexes, whereas Belgium and Estonia demonstrated this trend in males only. Mortality reductions at consistent rates, with a single Joinpoint, are seen with Bulgaria and Spain for males and with Bulgaria and Finland for females. Precipitous declining mortality for both sexes is observed in Portugal and Estonia, after an initial small increase for male mortality in Estonia from 1981 to 1994. Croatia had significantly fluctuating mortality, with multiple Joinpoints for males and females. In the United Kingdom there was a plateau, and even a moderate increase, in male mortality, in ≈2000. Denmark also experienced an increase in male mortality and a leveling out of female mortality over the same time period. Both countries then resumed mortality decline for males and females at a faster rate.
Smoothed Regression Fit for Former Communist Countries
Figure 5 displays the locally weighted scatterplot smoothing line of best fit with 95% confidence limits, using data for all of the previously Communist countries now in the EU. Time 0 represents the point of Communism ending in each of the constituent countries, with transition to capitalist rule. The final 10 years of Communism, that is, years –10 to 0 in Figure 5, show little change in IHD mortality. Mortality decline ensues 5 years after the end of Communist rule.
This study of 30-year cardiovascular mortality trends across Europe identifies significant mortality decline throughout the majority of EU member states. There is, however, an observed East-West disparity for both IHD and CVD. For IHD, the United Kingdom and the Netherlands have made the greatest improvements, with France currently having the lowest mortality for both males and females. The smallest improvements have been found in Lithuania, Hungary, and Slovakia, which are relatively new EU states. Austria has the largest reduction in CVD mortality, and France again has the lowest present-day mortality. Bulgaria, Romania, and Latvia have small reductions and currently the greatest CVD mortality.
An East-West cardiovascular disease inequality across Europe has been recognized previously.2–4,6–8 Although mortality overall is decreasing across the continent, this East-West gap appears to be widening. Furthermore, the impact of the transition from Communism can be appreciated. This suggests that there is no reduction in IHD mortality leading up to and for ≈5 years following the decline of Communism. One potential explanation for this is the turmoil associated with the changing of political landscapes. Once capitalist regimes were in place, a decreasing IHD mortality trend ensued.
An unexpected finding is the impressive CVD mortality reduction in Austria over the period studied; no such dramatic reduction is seen with IHD, which indicates the presence of CVD-specific factors. Hypertension is a plausible underlying factor, given its key involvement in stroke pathophysiology.16 The Austrian Heart Fund ran a campaign of hypertension awareness in 1978, the fruits of which could be being reaped long term. A high level of awareness of the risks of hypertension among the Austrian population has been demonstrated, although this appears to have waned recently.17,18 Another factor could be the unregulated access to all levels of health care and free provider choice delivered by the Austrian health service.19 The Austrian acute stroke service is also efficient with fast admissions, good availability of investigation modalities, and few delays.20 Conversely, Bulgaria has the highest CVD mortality for both sexes, but relatively low IHD mortality, possibly related to the prevalence of heavy alcohol use in the country.21
Lithuania has the present-day highest rate of IHD mortality for both sexes, although there is a decreasing trend for mortality in this study. A cross-sectional study of middle-aged males in Lithuania and Sweden recognized that traditional risk factors may not be entirely responsible. Dietary antioxidants and the resistance of low-density lipoproteins to oxidation was found to differ between the 2 groups.22 A prospective study found that a low-risk cardiovascular profile was present in only 1% of middle-aged urban adults in Kaunas, Lithuania (1994–2010), with a worsening trend in body mass index, fasting glucose, and healthy diet.23
Cyprus has very few mortality figures recorded on the WHO database. The few data available, however, show substantial increases in both CVD and IHD, with the exception of female CVD mortality, which demonstrates declining death rates. Conceivable reasons for this include the effects of war, erroneous and episodic health data recording, and the effect of a small population size.
Various hypotheses have been postulated for the disparity in European CVD mortality, ranging from the prevalence of socioeconomic inequality, differing diets, proportional health budget expenditure, and access to health care. Rising levels of smoking, alcoholism, and poor diet and exercise participation, as well, were recorded in Russia after the Communist decline.24 With the use of Russia as a postsocialist model, several legislative factors have been faced in health improvement, including lack of a formal health policy, scanty health promotion, and the absence of epidemiological surveillance systems.25 The trend of IHD mortality was examined in Germany 10 years after reunification, and demonstrated a clear East-West gradient with mortality ≈50% greater in former Communist East Germany.26 This study found differences in classical cardiovascular risk factors (high cholesterol, hypertension, obesity, and diabetes mellitus), and lifestyle factors (smoking, levels of exercise and diet), between East and West Germany.
The Eastern EU members have lower gross domestic product per capita than their Western counterparts, with resultant lower socioeconomic class, shorter life expectancy, and lower healthcare expenditure.27–29 Nonetheless, Eastern European countries have greater mortality than levels of wealth would predict.30 All-cause mortality has been found to be inversely proportional to educational attainment.31 Furthermore, education is inversely related to blood pressure, smoking, and obesity, and correlated with health knowledge and physical activity.32 Smoking is more prevalent in Eastern European countries, and is thought to be a considerable causal factor in cardiovascular health inequalities.33,34 Moreover, EU smoke-free legislation in public places has further reduced mortality and hospital admissions, a regulation not imposed stringently in many ex-socialist countries.35–37
Although not the focus of this study, there is an observed narrowing of sex inequalities in cardiovascular mortality, in keeping with a previous WHO report.38 This WHO article notes that not only is the gap closing, but also the gap is smaller still when healthy life expectancy is considered. There are many potential contributory factors for this trend, and the role of behavioral factors – smoking prevalence, in particular – may be crucial. Male smoking rates have peaked and are declining in many countries across Europe, whereas female smoking rates are increasing, both for daily smokers and current smokers.39 In addition, females may be more sensitive to the harmful effects of cigarette smoking, perhaps because of hormonal factors.40
This study follows a heterogeneous group of countries temporally, allowing the observation of significant mortality developments over a 30-year period. With an epidemiological data set as vast and diverse as the WHO database, issues of data reliability and accuracy of recording arise. There is reliance on retrospective data collection and death certification, and the robustness of these data needs to be considered. A potential contributing factor to the results of this study includes the variation in assignment of causes of death, between and within countries, which may fluctuate with changing coding practices over time. Nevertheless, the WHO scrutinizes death certification data for accurate levels of reporting, and all the countries included in this analysis had satisfactory coverage. Conversely, Mathers et al15 found that 4 countries in this study had low levels of reporting. Further potential loss of data reliability in this study arises because of changes in international death certification coding over the studied period (ICD-9 to ICD-10). No data were available for several time points, allowing for the potential of undocumented variation. As a descriptive analysis, little statistical evaluation has been performed; moreover, all explanations for mortality trends are speculative at present.
There is a growing disparity in cardiovascular mortality between Western Europe and the former Soviet states of Eastern Europe. Diverse fundamental explanations exist for this disparity, and the need for population-wide health promotion and primary prevention policies is emphasized to ensure that these differences do not expand further.
Sources of Funding
Dr Sikkel is supported by a National Institute of Health Research Clinical Lecturer award (#2670).
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.115.018931/-/DC1.
- Received August 9, 2015.
- Accepted March 18, 2016.
- © 2016 American Heart Association, Inc.
- 1.↵World Health Organization. 2014. Top 10 Causes of Death. http://www.who.int/mediacentre/factsheets/fs310/en. Accessed April 11, 2015.
- Levi F,
- Lucchini F,
- Negri E,
- La Vecchia C
- Moran AE,
- Forouzanfar MH,
- Roth G,
- Mensah G,
- Ezzati M,
- Murray CJ,
- Naghavi M
- Nichols M,
- Townsend N,
- Scarborough P,
- Rayner M
- Rodríguez T,
- Malvezzi M,
- Chatenoud L,
- Bosetti C,
- Levi F,
- Negri E,
- La Vecchia C
- Sans S,
- Kesteloot H,
- Kromhout D
- Bobak M,
- Marmot M
- 9.↵World Health Organization. 2014. WHO Mortality Database. http://apps.who.int/healthinfo/statistics/mortality/whodpms/. Accessed April 11, 2015.
- 10.↵World Health Organization. 2014. Description and Sources of Data. http://apps.who.int/healthinfo/statistics/mortality/whodpms/help/desc.htm. Accessed April 11, 2015.
- 11.↵European Commission. 2015. EU member states. http://europa.eu/about-eu/countries/member-countries/index_en.htm Accessed July 29, 2015.
- Ahmad OB,
- Boschi-Pinto C,
- Lopez AD,
- Murray CJ,
- Lozano R,
- Inoue M
- 13.↵World Health Organization. 2014. Census and Civil Registration Coverage Data by Country. http://apps.who.int/gho/data/node.main.121?lang=enWHO.%202014.%20WHO. Accessed April 11, 2015.
- Steiner S,
- Dorner TE,
- Fodor JG,
- Kunze M,
- Rieder A
- Pichlhöfer O,
- Maier M
- Balabanova D,
- McKee M
- Kristenson M,
- Ziedén B,
- Kucinskienë Z,
- Elinder LS,
- Bergdahl B,
- Elwing B,
- Abaravicius A,
- Razinkovienë L,
- Calkauskas H,
- Olsson AG
- Tamosiunas A,
- Luksiene D,
- Baceviciene M,
- Bernotiene G,
- Radisauskas R,
- Malinauskiene V,
- Kranciukaite-Butylkiniene D,
- Virviciute D,
- Peasey A,
- Bobak M
- Müller-Nordhorn J,
- Rossnagel K,
- Mey W,
- Willich SN
- Le Gallo J,
- Ertur C
- Kitagawa EM
- Luepker RV,
- Rosamond WD,
- Murphy R,
- Sprafka JM,
- Folsom AR,
- McGovern PG,
- Blackburn H
- 34.↵World Health Organization. 2014. Prevalence of Tobacco Use Among Adults and Adolescents. http://gamapserver.who.int/gho/interactive_charts/tobacco/use/atlas.html. Accessed April 11, 2015.
- Cox B,
- Martens E,
- Nemery B,
- Vangronsveld J,
- Nawrot TS
- Sargent RP,
- Shepard RM,
- Glantz SA
- 37.↵European Commission. Overview of Smoke-Free Legislation. 2015. http://ec.europa.eu/health/tobacco/docs/smoke-free_legislation_table_en.pdf. Accessed April 11, 2015.
- Prescott E,
- Hippe M,
- Schnohr P,
- Hein HO,
- Vestbo J
Cardiovascular disease is the leading cause of mortality worldwide and trends in cardiovascular mortality have a central role in public health and epidemiology globally. Previous studies on trends in Europe have highlighted a health inequality between Western countries and Eastern, previously Communist-run, states. This study provides an up-to-date analysis of age-standardized death rates from ischemic heart disease (IHD) and cerebrovascular disease (CVD) across member states of the European Union between 1980 and 2010. The effect of changing political ideologies is explored in previously Communist-run countries. Data are collated from the World Health Organization mortality database with the use of the International Classification of Diseases, ninth and tenth revisions, and are analyzed by using Joinpoint regression analysis. An overall trend for reduction in mortality from IHD and CVD is seen across Europe, although very modest mortality reductions are observed in Eastern European states, Romania, Poland, and Slovakia, for example. Lithuania and Bulgaria have the present-day highest mortality for both sexes for IHD and CVD, respectively. Conversely, significant reductions are seen in cardiovascular mortality in Western countries; the lowest mortality from IHD and CVD for both males and females has been observed in France. In pooled data for formerly Communist-run countries, a trend for no reduction in IHD mortality leading up to, and for a number of years following, the decline of Communism is noted. Reasons for the growing East-West disparity are explored, and topics highlighted that may address this inequality. The need for population-wide health promotion and primary prevention policies are emphasized.