This special edition of Health at a Glance focuses on health issues across the 27 European Union member states, three European Free Trade Association countries (Iceland, Norway and Switzerland) and Turkey. It gives readers a better understanding of the factors that affect the health of populations and the performance of health systems in these countries. Its 42 indicators present comparable data covering a wide range of topics, including health status, risk factors, health workforce and health expenditure.
Each indicator in the book is presented in a user-friendly format, consisting of charts illustrating variations across countries and over time, brief descriptive analyses highlighting the major findings conveyed by the data, and a methodological box on the definition of the indicators and any limitations in data comparability. An annex provides additional information on the demographic and economic context within which health systems operate.
This publication is the result of collaboration between the OECD and the European Commission, with the help of national data correspondents from the 31 countries.
Health at a Glance Europe 2010 Health at a Glance: Europe 2010 This work is published on the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of the OECD or of the governments of its member countries or those of the European Union. Please cite this publication as: OECD (2010), Health at a Glance: Europe 2010, OECD Publishing. http://dx.doi.org/10.1787/health_glance-2010-en ISBN 978-92-64-09030-9 (print) ISBN 978-92-64-09031-6 (PDF) Photo credits: Cover © Tiut Lucian/Shutterstock.com. Corrigenda to OECD publications may be found on line at: www.oecd.org/publishing/corrigenda. © OECD 2010 You can copy, download or print OECD content for your own use, and you can include excerpts from OECD publications, databases and multimedia products in your own documents, presentations, blogs, websites and teaching materials, provided that suitable acknowledgment of OECD as source and copyright owner is given. All requests for public or commercial use and translation rights should be submitted to firstname.lastname@example.org. Requests for permission to photocopy portions of this material for public or commercial use shall be addressed directly to the Copyright Clearance Center (CCC) at email@example.com or the Centre français d’exploitation du droit de copie (CFC) at firstname.lastname@example.org. FOREWORD Foreword T his first edition of Health at a Glance: Europe presents a set of key indicators on health and health systems across 31 countries – the 27 European Union member states, three European Free Trade Association countries (Iceland, Norway and Switzerland), and Turkey. The selection of indicators is based on the European Community Health Indicators (ECHI) shortlist – a set of indicators used by the European Commission to guide the development of health information systems in Europe. In addition, the publication provides detailed information on health expenditure trends across countries, building on the OECD’s established expertise in this area. This publication is a concrete example of the long and fruitful collaboration between the OECD and the European Commission in the development and reporting of health statistics. This collaboration also involves the World Health Organization (WHO). The preparation of this report has been greatly facilitated by the increased co-operation in the collection of health statistics at the international level in recent years. A joint data collection between the OECD, Eurostat (the European statistical agency) and WHO was launched at the end of 2005 to improve the availability and comparability of data on health expenditure and financing, based on the System of Health Accounts. Building on the success of the joint Health Accounts collection, a new joint data collection between the three organisations was launched in 2010 to gather data on non-monetary health care statistics. These joint data collections are improving the comparability of data across countries, while reducing the data collection burden on national administrations. Health at a Glance: Europe 2010 would not have been possible without the effort of national data correspondents from the 31 countries who have provided most of the data and the metadata presented in this report. The OECD and the European Commission would like to sincerely thank them for their contribution. This publication was prepared by a team from the OECD Health Division under the co-ordination of Gaétan Lafortune and Michael de Looper. Chapter 1 and Chapter 2 were prepared by Michael de Looper and Valerie Moran, with a contribution from Carol Jagger and Jean-Marie Robine (Network on Health Expectancy, REVES) for the indicators related to life expectancy and healthy life years. Chapter 3 was prepared by Gaétan Lafortune and Gaëlle Balestat, with a contribution from Vladimir Stevanovic and Rie Fujisawa for the two indicators related to cancer care. Chapter 4 was written by David Morgan and Rebecca Bennetts. It is important to recognise the contribution of colleagues from Eurostat (in particular Elodie Cayotte and Albane Gourdol) and WHO-European Office (in particular Ivo Rakovac), who have shared some of the data presented in this publication. This publication benefited from comments from Mark Pearson (Head of OECD Health Division) and Nick Fahy, Fabienne Lefebvre and Federico Paoli (European Commission – DG Sanco). Aart De Geus Paola Testori Coggi Deputy Secretary-General Director-General Organisation for Economic Co-operation Directorate-General for Health and Consumers and Development European Commission HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3 TABLE OF CONTENTS Table of Contents Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Résumé . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Chapter 1. Health Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.1. Life expectancy and healthy life expectancy at birth . . . . . . . . . . . . . . . . . . . . . 26 1.2. Life expectancy and healthy life expectancy at age 65 . . . . . . . . . . . . . . . . . . . 28 1.3. Mortality from all causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 1.4. Mortality from heart disease and stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 1.5. Mortality from cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 1.6. Mortality from transport accidents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 1.7. Suicide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 1.8. Infant mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 1.9. Infant health: Low birth weight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 1.10. Self-reported health and disability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 1.11. Incidence of selected communicable diseases . . . . . . . . . . . . . . . . . . . . . . . . . . 46 1.12. HIV/AIDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 1.13. Cancer incidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 1.14. Diabetes prevalence and incidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 1.15. Dementia prevalence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Chapter 2. Determinants of Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 2.1. Smoking and alcohol consumption among children . . . . . . . . . . . . . . . . . . . . . 58 2.2. Nutrition among children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 2.3. Physical activity among children. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 2.4. Overweight and obesity among children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 2.5. Supply of fruit and vegetables for consumption . . . . . . . . . . . . . . . . . . . . . . . . . 66 2.6. Tobacco consumption among adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 2.7. Alcohol consumption among adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 2.8. Overweight and obesity among adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Chapter 3. Health Care Resources, Services and Outcomes . . . . . . . . . . . . . . . . . . . . . . . . 75 3.1. Practising physicians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3.2. Practising nurses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 3.3. Childhood vaccination programmes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 3.4. Influenza vaccination for older people. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 3.5. Medical technologies: CT scanners and MRI units . . . . . . . . . . . . . . . . . . . . . . . 84 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 5 TABLE OF CONTENTS 3.6. Hospital beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 3.7. Hospital discharges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 3.8. Average length of stay in hospitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 3.9. Cardiac procedures (coronary angioplasty) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 3.10. Cataract surgeries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.11. Hip and knee replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 3.12. Screening, survival and mortality for cervical cancer . . . . . . . . . . . . . . . . . . . . 98 3.13. Screening, survival and mortality for breast cancer. . . . . . . . . . . . . . . . . . . . . . 100 Chapter 4. Health Expenditure and Financing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 4.1. Health expenditure per capita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 4.2. Health expenditure in relation to GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 4.3. Health expenditure by function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 4.4. Pharmaceutical expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 4.5. Financing of health care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 4.6. Trade in health services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Annex A. Additional Information on Demographic and Economic Context . . . . . . . . . 122 This book has... StatLinks2 A service that delivers Excel® ﬁles from the printed page! Look for the StatLinks at the bottom right-hand corner of the tables or graphs in this book. To download the matching Excel® spreadsheet, just type the link into your Internet browser, starting with the http://dx.doi.org prefix. If you’re reading the PDF e-book edition, and your PC is connected to the Internet, simply click on the link. You’ll find StatLinks appearing in more OECD books. 6 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 ACRONYMS Acronyms AIDS Acquired immunodeficiency syndrome ALOS Average length of stay AMI Acute myocardial infraction BMI Body mass index CAT (or CT) Computed axial tomography (scanner) EFTA European Free Trade Association EU European Union EU-SILC European Union Statistics on Income and Living Conditions survey GALI Global activity limitation indicator GDP Gross domestic product GP General practitioner HBSC Health Behavior in School-aged Children survey HIV Human immunodeficiency virus HLY Healthy life years IHD Ischemic heart disease ISIC International Standard Industrial Classification MRI Medical resonance imaging PPP Purchasing power parities SHA System of Health Accounts SIDS Sudden infant death syndrome HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 7 Health at a Glance: Europe 2010 © OECD 2010 Executive Summary E uropean countries have achieved major gains in population health over recent decades. Life expectancy at birth in European Union (EU) countries has increased by six years since 1980, while premature mortality has reduced dramatically. Improvements in living and working conditions and in some health-related behaviours have contributed greatly to these longevity gains, but progress in medical care also deserves much credit. Health systems are of growing size and complexity in European countries, and spending on health care has never been higher, consuming an ever-increasing share of national income. This first edition of Health at a Glance: Europe, the result of a long-standing collaboration between the OECD and the European Commission, presents a set of key indicators of health and health systems in 31 European countries – the 27 member states of the European Union, and Iceland, Norway, Switzerland and Turkey. The selection of indicators has been based on the European Community Health Indicators (ECHI) shortlist, a list of indicators that has been developed by the European Commission to guide the development and reporting of health statistics (European Commission, 2010a). However, a number of indicators in this report differ from ECHI definitions because of data availability or constraints, or in some instances because ECHI indicators are not yet ready for implementation. The publication also provides detailed information on health expenditure and its financing, building on the OECD’s established data collection and expertise in this area. The data presented in the publication come mainly from official national statistics, as gathered in OECD Health Data, the Eurostat Statistics Database and WHO-Europe’s Health for All Database. Health at a Glance: Europe 2010 presents evidence of wide variations across European countries in population health status, risk factors for health, the inputs, outputs and outcomes of health care systems, and levels of health expenditure and financing sources. It offers some explanation for these variations, providing a background to understand more fully the causes underlying such variations and to develop policy options to reduce gaps across countries. It should also be noted that while basic population breakdowns by sex and age are presented, this publication does not generally provide detail by sub-national regions, by socio-economic groups or by ethnic/racial groups. For many indicators, readers should keep in mind that there may be as much variation within a country as there is across countries. 9 EXECUTIVE SUMMARY Health status has improved dramatically in European countries, although large gaps persist ● Life expectancy at birth in EU countries has increased by six years since 1980, reaching 78 years in 2007. On average across the 27 EU countries, life expectancy at birth for the three-year period 2005-07 stood at 74.3 years for men and 80.8 years for women. France had the highest life expectancy at birth for women (84.4 years), while Sweden had the highest life expectancy for men (78.8 years). Life expectancy at birth in the European Union was lowest in Romania for women (76.2 years) and Lithuania for men (65.1 years). The gap between countries with the highest and lowest life expectancies at birth is around eight years for women and 14 years for men. ● Whether the gains in life expectancy involve additional years of life lived in good health has important implications for health and long-term care systems in Europe. Healthy life years at birth is defined as the number of years of life in which a person’s day-to-day activities are not limited by a condition or health problem. In 2005-07, healthy life years stood at 61.3 years for women and 60.1 years for men, on average, in the European Union. The gender gap is much smaller than for life expectancy, reflecting the fact that a higher proportion of women’s lives are spent with activity limitations. Healthy life years at birth in 2005-07 was greatest in Malta for both men and women, and shortest in Latvia for women and Estonia for men. ● Life expectancy at age 65 has also increased substantially over the past decades in European countries. The average in 2005-07 for the 27 EU countries was 15.9 years for men and 19.5 years for women. As for life expectancy at birth, France had the highest life expectancy at age 65 for women (22.6 years) but also for men (18.1 years). Life expectancy at age 65 was lowest in Eastern Europe – in Latvia for men (12.7 years) and in Bulgaria for women (16.3 years). ● As is the case at birth, the gender gap for healthy life years at age 65 is much narrower than for life expectancy. In 2005-07, men were slightly favoured, at 8.4 years versus 8.1 years for women. ● It is difficult to estimate the relative contribution of the numerous medical and non-medical factors that might affect variations in (healthy) life expectancy. Higher national income is generally associated with higher life expectancy across European countries, although the relationship is less pronounced at higher levels of national income, suggesting a “diminishing return” after a certain level. Other determinants of health also play an important role. Risk factors to health are changing ● Many EU countries have achieved remarkable progress in reducing tobacco consumption, although it is still a leading cause of early death. Much of this decline can be attributed to policies at national and EU level promoting public awareness campaigns, advertising bans and increased taxation. Less than 18% of adults in Sweden and Iceland now smoke daily, down from over 30% in 1980. However, almost 40% of adults in Greece continue to smoke on a daily basis. Smoking rates are also relatively high in Bulgaria, Ireland and the Netherlands. 10 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 EXECUTIVE SUMMARY ● Alcohol consumption has also fallen in many European countries over the past three decades. Curbs on advertising, sales restrictions and taxation have proven to be effective measures to reduce alcohol consumption. Traditional wine-producing countries such as Italy, France and Spain have seen their alcohol consumption per capita drop substantially since 1980. On the other hand, consumption rose significantly in a number of countries including Ireland, the United Kingdom and some Nordic countries. ● More than half of the total adult population across the European Union are now overweight or obese. This is also true in 15 of the 27 EU countries. The prevalence of obesity – which presents greater health risks than overweight – varies from less than 10% in Romania, Switzerland and Italy to over 20% in the United Kingdom, Ireland, Malta and Iceland. On average across EU countries, 15.5% of the adult population is obese. ● The rate of obesity has more than doubled over the past 20 years in most EU countries for which data are available. The rapid increase occurred regardless of what the levels of obesity were two decades ago. Obesity more than doubled in both the Netherlands and the United Kingdom between 1988 and 2008, even though the rate in the Netherlands is currently less than half that of the United Kingdom. ● Because obesity is associated with higher risks of chronic illnesses, it is linked to significant additional health care costs. A recent study in England estimated that total costs linked to overweight and obesity could increase by as much as 70% between 2007 and 2015, and be 2.4 times higher by 2025 (Foresight, 2007). Shortages of health workers is a concern in many countries ● There are concerns in many European countries about shortages of doctors. The number of doctors per capita varies greatly, and is lowest in Turkey, followed by Poland and Romania. Doctor numbers are also relatively low in the United Kingdom and Finland. ● Since 2000, the number of physicians per capita has however increased in all European countries, except the Slovak Republic. On average, the number grew from 3.0 doctors per 1 000 population in 2000 to 3.3 in 2008. It increased particularly rapidly in Ireland, rising by nearly 50%. A large part of this increase was due to the recruitment of foreign-trained physicians, with the share of foreign-trained doctors tripling during that period. Similarly, the number of doctors per capita in the United Kingdom increased by 30% between 2000 and 2008, rising from 2.0 per 1 000 population to 2.6. ● In contrast, there has been virtually no growth in the number of doctors per capita in France and Italy since 2000. Following a reduction in the number of new entrants in medical schools during the 1980s and 1990s, the number of doctors per capita in Italy peaked in 2002, and has declined since then. In France, the number peaked in 2005, and the decline is expected to continue over the next ten years. ● In nearly all countries, the balance between general practitioners and specialists has changed over past decades, with the number of specialists increasing much more rapidly. As a result, there are more specialists than generalists in most countries, except Romania and Portugal. This may be explained by a reduced attractiveness in the traditional mode of practice of general/family practitioner, as well as a growing remuneration gap. The slow growth or reduction in the number of generalists per capita HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 11 EXECUTIVE SUMMARY raises concerns about access to primary care. Many countries are considering ways to improve the attractiveness of general practice as well as developing new roles for other health care providers, such as nurses. ● There are also concerns about shortages of nurses in many European countries. Nurses play an important role in providing health care not only in traditional settings such as hospitals and long-term care institutions but increasingly in primary care, especially in offering care to the chronically ill, and in patients’ homes. In 2008, there were about 15 nurses per 1 000 population in Finland, Iceland, Ireland and Switzerland, and slightly fewer in Denmark and Norway. Turkey had the fewest nurses, followed by Greece, Bulgaria and Cyprus, at less than five per 1 000 population. ● Since 2000, the number of nurses per capita has increased in all European countries, except Lithuania and the Slovak Republic. The increase was particularly large in Portugal, Spain, France and Switzerland. Growing health expenditure puts pressure on government budgets ● Health expenditure has risen in all European countries, often increasing at a faster rate than economic growth, resulting in a rising share of GDP allocated to health. In 2008, EU countries spent, on average, 8.3% of their GDP on health, up from 7.3% in 1998. However, the share of GDP allocated to health spending varies considerably across countries, ranging from less than 6% in Cyprus and Romania to more than 10% in France, Switzerland, Germany and Austria. ● In some countries, the recent economic downturn resulted in a marked increase in the ratio of health spending to GDP. In Ireland, the percentage of GDP devoted to health increased from 7.5% in 2007 to 8.7% in 2008. In Spain, it rose from 8.4% to 9.0%. ● In 2008, Norway spent the most on health per capita among European countries, with spending of about EUR 4 300. Switzerland, Luxembourg and Austria were the next highest spending countries. Most northern and western European countries spend between EUR PPP 2 500 and 3 500 per person, that is, 10% to 60% more than the EU average. Those countries spending below the EU average are eastern and southern European countries such as Turkey, Romania, Bulgaria, Poland and Hungary. ● Health expenditure per capita tends to be positively correlated with GDP per capita, although the association is stronger among European countries with low GDP per capita. Even for countries with similar levels of GDP per capita, there can be substantial differences in health expenditure. For example, Spain and France have similar GDP per capita, but Spain spends less than 80% of the level of France on health. ● Health systems are sometimes criticised for being overly focused on “sick care”: for treating the ill, but not doing enough to prevent illness. Only around 3% of current health expenditure is spent on prevention and public health programmes on average in EU countries. ● The public sector is the main source of health financing in all European countries, except Cyprus. On average, nearly three-quarter of all health spending was publicly financed in 2008, through general taxation or social security contributions. In Luxembourg, the Czech Republic, the Nordic countries (except Finland), the United Kingdom and Romania, public financing accounted for more than 80% of all health expenditure. 12 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 EXECUTIVE SUMMARY ● The size and composition of private financing differs across countries. In most countries, it is in the form of out-of-pocket payments by patients. Private health insurance accounts for only around 3-4% of total health expenditure on average across EU countries. However, in some countries, it plays a significant role. In Germany, it provides primary coverage for certain population groups. In France, private health insurance finances 13% of overall spending, but provides complementary and supplementary coverage in a universal public system. ● Given the current need to reduce budget deficits in many countries, governments may be faced with difficult policy choices in the short-term. They may either have to curb the growth of public spending on health, cut spending in other areas, or raise taxes or social security contributions to reduce their deficits. Improving productivity within the health sector may help to reconcile these pressures, for example through more rigorous assessment of health technologies or increased used of information and communication technologies (“eHealth”). These initiatives may also have the added benefit of improving the quality of care, which is another area of collaboration between the OECD and the European Commission. HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 13 Health at a Glance: Europe 2010 © OECD 2010 Résumé L es pays européens ont accompli d’importants progrès en matière de santé au cours des dernières décennies. Dans les pays de l’Union européenne, l’espérance de vie à la naissance a augmenté de six ans depuis 1980, tandis que la mortalité précoce a fortement reculé. Si l’amélioration des conditions de vie et de travail, ainsi que de certains comportements vis-à-vis de la santé, a joué un rôle majeur dans l’augmentation de la longévité, les progrès de la médecine doivent également être salués. Les systèmes de santé dans les pays européens occupent une place de plus en plus importante et les dépenses consacrées aux soins de santé n’ont jamais été aussi élevées, représentant une part croissante du revenu national. Cette première édition de Panorama de la santé : Europe, fruit d’une collaboration de longue date entre l’OCDE et la Commission européenne, propose un ensemble d’indicateurs clés de la santé et des systèmes de santé dans 31 pays européens, à savoir les 27 États membres de l’Union européenne, l’Islande, la Norvège, la Suisse et la Turquie. La sélection d’indicateurs s’appuie sur la liste des indicateurs de santé de la Communauté européenne (European Community Health Indicators – ECHI), élaborée par la Commission européenne pour étayer la production et la publication de statistiques sur la santé (Commission européenne, 2010a). Certains des indicateurs diffèrent parfois des définitions retenues pour la liste ECHI pour des questions de disponibilité des données. Dans d’autres cas, les indicateurs ECHI ne sont pas encore prêts à être mis en œuvre. Par ailleurs, la publication fournit également des informations détaillées sur les dépenses de santé et leur financement, en s’appuyant sur l’expérience de l’OCDE en matière de collecte de données dans ce domaine. Les informations présentées dans Panorama de la santé : Europe sont essentiellement issues de sources statistiques nationales officielles, notamment d’Éco-Santé OCDE, de la base de données statistique Eurostat et de la base de données Santé pour tous de l’OMS-Europe. Panorama de la santé : Europe 2010 montre qu’il existe d’importants écarts entre les pays européens en termes d’état de santé de la population, de facteurs de risques pour la santé, d’intrants, d’extrants et de résultats des systèmes de santé, et de niveaux des dépenses de santé et des sources de financement. L’étude propose des explications à ces écarts, en fournissant le contexte nécessaire pour mieux comprendre leurs causes sous-jacentes. Il convient aussi de noter que si des disparités par sexe et par âge sont présentées, cette publication ne fournit généralement pas d’informations sur les disparités par région, par groupe socioéconomique ou par groupe ethnique. Pour de nombreux indicateurs, le lecteur doit garder à l’esprit que les variations peuvent être aussi importantes au sein d’un même pays qu’entre les pays. 15 RÉSUMÉ L’état de santé s’est amélioré de manière remarquable dans les pays européens, même si des écarts importants persistent ● Dans les pays de l’UE, l’espérance de vie à la naissance s’est allongée de six ans depuis 1980, pour atteindre 78 ans en 2007. En moyenne dans les 27 pays de l’UE, l’espérance de vie à la naissance pour la période 2005-07 s’élevait à 74.3 ans pour les hommes et à 80.8 ans pour les femmes. La France affiche l’espérance de vie à la naissance la plus longue pour les femmes (84.4 ans), tandis que l’espérance de vie la plus longue pour les hommes est observée en Suède (78.8 ans). Au sein de l’Union européenne, c’est en Roumanie que l’espérance de vie à la naissance est la plus courte pour les femmes (76.2 ans) et en Lituanie pour les hommes (65.1 ans). L’écart entre les pays à l’espérance de vie la plus longue et ceux où l’espérance de vie est la plus courte s’établit à 8 ans environ pour les femmes et à 14 ans pour les hommes. ● Il importe de savoir si l’allongement de l’espérance de vie implique des années de vie supplémentaires en bonne santé, parce que cela a des répercussions majeures sur les systèmes de santé et de soins de longue durée en Europe. L’espérance de vie en bonne santé à la naissance est définie ici comme le nombre d’années de vie au cours desquelles les activités quotidiennes de l’individu ne sont pas limitées par une maladie ou un problème de santé. En 2005-07, l’espérance de vie en bonne santé s’établissait à 61.3 ans pour les femmes et 60.1 ans pour les hommes en moyenne dans l’Union européenne. L’écart hommes-femmes est donc bien moindre qu’en ce qui concerne l’espérance de vie, ce qui tient au fait qu’une plus forte proportion de la vie des femmes est marquée par des limitations de leur activité. En 2005-07, c’est à Malte que l’espérance de vie en bonne santé était la plus longue à la fois pour les hommes et pour les femmes, tandis que la Lettonie affichait l’espérance de vie en bonne santé la plus courte pour les femmes et l’Estonie pour les hommes. ● L’espérance de vie à l’âge de 65 ans s’est aussi considérablement accrue en Europe au cours des dernières décennies. En 2005-07, elle s’élevait en moyenne dans les 27 pays de l’UE à 15.9 ans pour les hommes et 19.5 ans pour les femmes. Comme pour l’espérance de vie à la naissance, la France se distingue par l’espérance de vie à 65 ans la plus longue pour les femmes (22.6 ans) mais aussi pour les hommes (18.1 ans). Au contraire, c’est en Europe de l’Est que l’espérance de vie à 65 ans est la plus courte : en Lettonie pour les hommes (12.7 ans) et en Bulgarie pour les femmes (16.3 ans). ● Comme pour l’espérance de vie à la naissance, l’écart hommes-femmes s’agissant de l’espérance de vie en bonne santé à 65 ans est bien plus restreint que pour l’espérance de vie : en 2005-07, les hommes étaient légèrement avantagés, avec 8.4 ans contre 8.1 ans pour les femmes. ● Il est difficile d’estimer la contribution relative des multiples facteurs médicaux et non médicaux susceptibles d’influencer les écarts dans l’espérance de vie (en bonne santé). Un revenu national élevé est généralement associé à une meilleure espérance de vie dans les pays européens, quoique cette corrélation soit moins prononcée pour les niveaux de revenu élevés, ce qui suggère un « rendement décroissant » à partir d’un certain seuil. D’autres déterminants de la santé jouent également un rôle clé. 16 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 RÉSUMÉ Les facteurs de risques évoluent ● De nombreux pays de l’UE ont accompli des progrès remarquables dans la lutte contre le tabagisme, même s’il demeure l’une des principales causes de mortalité précoce. Cette réussite peut en grande partie être attribuée aux mesures mises en œuvre à l’échelle nationale et européenne pour promouvoir les campagnes de sensibilisation publique, les interdictions de publicité et la hausse des taxes. En Suède et en Islande, moins de 18 % des adultes fument désormais quotidiennement, contre plus de 30 % en 1980. En revanche, près de 40 % des adultes continuent de fumer quotidiennement en Grèce. Le taux de tabagisme demeure également élevé en Bulgarie, en Irlande et aux Pays-Bas. ● La consommation d’alcool a également diminué dans nombre de pays européens ces 30 dernières années. Les restrictions sur la publicité et les ventes et la hausse des taxes se sont avérées des outils efficaces pour réduire la consommation d’alcool. Les pays traditionnellement producteurs de vin, comme l’Italie, la France et l’Espagne, ont vu la consommation d’alcool par habitant chuter fortement depuis 1980. À l’inverse, la consommation a sensiblement augmenté dans plusieurs pays comme l’Irlande, le Royaume-Uni et certains pays nordiques. ● Plus de la moitié de la population adulte totale de l’Union européenne est désormais en situation de surpoids ou d’obésité. C’est également le cas dans 15 des 27 pays de l’UE. La prévalence de l’obésité – qui présente des risques pour la santé supérieurs à ceux du surpoids – est comprise entre moins de 10 % en Roumanie, en Suisse et en Italie à plus de 20 % au Royaume-Uni, en Irlande, à Malte et en Islande. En moyenne dans les pays de l’UE, 15.5 % de la population adulte est obèse. ● Le taux d’obésité a plus que doublé ces 20 dernières années dans la plupart des pays de l’UE pour lesquels des données sont disponibles. Cette progression rapide est intervenue indépendamment des taux d’obésité observés il y a 20 ans. L’obésité a plus que doublé aux Pays-Bas et au Royaume-Uni entre 1988 et 2008, même si le taux observé aux Pays-Bas est actuellement inférieur de plus de moitié à celui du Royaume-Uni. ● L’obésité étant associée à une augmentation des risques de maladie chronique, elle entraîne un coût supplémentaire important au niveau des soins de santé. Selon une étude récente réalisée en Angleterre, la hausse du coût représenté par le surpoids et l’obésité pourrait aller jusqu’à 70 % entre 2007 et 2015 et il pourrait être 2.4 fois plus élevé d’ici à 2025 (Foresight, 2007). La pénurie de professionnels de santé est un sujet d’inquiétude dans de nombreux pays ● De nombreux pays européens s’inquiètent d’une pénurie de médecins. Le nombre de médecins par habitant varie fortement entre les pays; il atteint son niveau le plus bas en Turquie, suivie par la Pologne et la Roumanie. Il est également relativement bas au Royaume-Uni et en Finlande. ● Depuis 2000, le nombre de médecins par habitant a néanmoins augmenté dans tous les pays européens, à l’exception de la Slovaquie. En moyenne, il est passé de 3.0 médecins pour 1 000 habitants en 2000 à 3.3 en 2008. Cette progression a été particulièrement rapide en Irlande, avec une hausse de près de 50 %. Ceci s’explique en grande partie par le recrutement de médecins formés à l’étrangers : le nombre de médecins formés à HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 17 RÉSUMÉ l’étranger a en effet triplé sur la période. De la même façon, le nombre de médecins par habitant au Royaume-Uni a progressé de 30 % entre 2000 et 2008, passant de 2.0 pour 1 000 habitants à 2.6. ● À l’inverse, le nombre de médecins par habitant est resté pratiquement inchangé en France et en Italie depuis 2000. Après une baisse du nombre de nouveaux inscrits en école de médecine dans les années 80 et 90, le nombre de médecins par habitant a atteint son point le plus haut en 2002 en Italie, pour s’orienter ensuite à la baisse. En France, il a touché son plus haut niveau en 2005 et la baisse devrait se poursuivre au cours des dix prochaines années. ● Dans la quasi-totalité des pays, le rapport entre médecins généralistes et spécialistes a évolué au cours des dernières décennies, le nombre de spécialistes ayant progressé bien plus rapidement. Par conséquent, les spécialistes sont aujourd’hui plus nombreux que les généralistes dans la plupart des pays, à l’exception de la Roumanie et du Portugal. Ce phénomène peut s’expliquer par une diminution de l’attrait offert par le mode traditionnel de la pratique du médecin généraliste/de famille, ainsi que par un écart de rémunération croissant. La hausse limitée, voire la baisse, du nombre de généralistes par habitant suscite des inquiétudes quant à l’accès aux soins primaires. De nombreux pays étudient des moyens pour renforcer l’attractivité de la médecine générale et pour concevoir de nouveaux rôles pour d’autres professionnels de santé, comme le personnel infirmier. ● Par ailleurs, de nombreux pays européens sont touchés par une pénurie de personnel infirmier. Les infirmiers jouent un rôle important dans la prestation des soins de santé non seulement dans le cadre traditionnel de l’hôpital ou des établissements de soins de longue durée mais aussi, de plus en plus, dans les soins primaires, notamment auprès des malades chroniques et dans les traitements à domicile. En 2008, on comptait environ 15 infirmières pour 1 000 habitants en Finlande, en Islande, en Irlande et en Suisse, et un peu moins au Danemark et en Norvège. La Turquie est le pays où l’on compte le moins d’infirmiers, suivie par la Grèce, la Bulgarie et Chypre, avec moins de 5 pour 1 000 habitants. ● Depuis 2000, le nombre de personnel infirmier par habitant a progressé dans tous les pays européens, à l’exception de la Lituanie et de la Slovaquie. Cette progression est particulièrement importante au Portugal, en Espagne, en France et en Suisse. L’augmentation des dépenses de santé pèse sur les budgets nationaux ● Les dépenses de santé ont augmenté dans tous les pays européens, la plupart du temps à un rythme supérieur à celui de la croissance économique, ce qui se traduit par une augmentation de la part du PIB allouée à la santé. En 2008, les pays de l’UE ont consacré en moyenne 8.3 % de leur PIB aux dépenses de santé, contre 7.3 % en 1998. Néanmoins, la part du PIB allouée aux dépenses de santé varie considérablement entre les pays, de moins de 6 % à Chypre et en Roumanie à plus de 10 % en France, en Suisse, en Allemagne et en Autriche. ● Dans certains pays, la récession récente a provoqué une hausse notable de la part des dépenses de santé dans le PIB. Ainsi, en Irlande, la part du PIB consacrée à la santé a progressé de 7.5 % en 2007 à 8.7 % en 2008. En Espagne, elle est passée de 8.4 % à 9.0 %. 18 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 RÉSUMÉ ● En 2008, la Norvège est le pays qui affiche les dépenses de santé par habitant les plus élevées parmi les pays européens, à 4 300 EUR environ, suivie par la Suisse, le Luxembourg et l’Autriche. La plupart des pays d’Europe du Nord et de l’Ouest ont dépensé entre 2 500 et 3 500 EUR par habitant, ce qui est supérieur de 10 à 60 % à la moyenne de l’UE. Les pays où les dépenses de santé sont inférieures à la moyenne de l’UE sont les pays d’Europe de l’Est et du Sud comme la Turquie, la Roumanie, la Bulgarie, la Pologne et la Hongrie. ● Les dépenses de santé par habitant présentent généralement une corrélation positive avec le PIB par habitant, même si celle-ci est plus étroite dans les pays européens caractérisés par un PIB par habitant relativement bas. Cependant, même dans les pays au PIB par habitant équivalent, on peut observer des écarts importants en matière de dépenses de santé. Par exemple, l’Espagne et la France affichent un PIB par habitant assez proche, mais les dépenses de santé de l’Espagne représentent moins de 80 % de celles de la France. ● On déplore parfois que les systèmes de santé soient trop tournés sur les soins aux malades, c’est-à-dire qu’ils sont davantage axés sur le traitement des maladies plutôt que sur leur prévention. En moyenne dans les pays de l’UE, seulement 3 % environ des dépenses de santé sont consacrées à la prévention et aux programmes de santé publique. ● Le secteur public représente la principale source de financement de la santé dans tous les pays européens, à l’exception de Chypre. En moyenne, près de 75 % des dépenses de santé totales étaient financées par les fonds publics en 2008, au moyen des recettes fiscales ou des cotisations de sécurité sociale. Au Luxembourg, en République tchèque, dans les pays nordiques (hors Finlande), au Royaume-Uni et en Roumanie, le financement public couvre les dépenses de santé à hauteur de plus de 80 %. ● L’ampleur et la composition du financement privé varient selon les pays. Généralement, il prend la forme d’une participation financière par les patients. L’assurance maladie privée ne représente que 3-4 % seulement des dépenses de santé totales en moyenne dans les pays de l’UE. Toutefois, dans certains pays, elle a un rôle de financement important. Ainsi, elle assure une couverture primaire à certaines catégories de population en Allemagne. En France, l’assurance maladie privée finance 13 % des dépenses totales mais elle fournit une couverture complémentaire et supplémentaire dans le cadre d’un régime public universel. ● De nombreux pays étant actuellement soucieux de réduire leurs déficits budgétaires, les pouvoirs publics seront confrontés à des choix difficiles à court terme. Ils pourraient en effet être contraints soit de freiner la croissance des dépenses publiques de santé, soit de réduire les dépenses dans d’autres secteurs, ou soit d’augmenter les impôts ou les cotisations de sécurité sociale, pour réduire leurs déficits. Des gains de productivité et d’efficience dans le secteur de la santé pourraient contribuer à alléger les pressions, par exemple au moyen d’une évaluation plus rigoureuse des technologies de santé ou d’un recours accru aux technologies de l’information et de la communication. Ces initiatives pourraient en outre permettre d’améliorer la qualité des soins, ce qui constitue un autre axe important de collaboration entre l’OCDE et la Commission européenne. HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 19 Health at a Glance: Europe 2010 © OECD 2010 Introduction H ealth at a Glance: Europe 2010 presents key indicators of health and health systems in 31 European countries, including the 27 European Union member states, three EFTA countries (Iceland, Norway and Switzerland), and Turkey. It builds on the format used in the OECD’s previous editions of Health at a Glance to provide comparable information on important public health issues in Europe. The indicators have been selected on the basis of the European Community Health Indicators (ECHI) shortlist (European Commission, 2010a; ECHIM, 2010). However, in some instances, this report deviates from the formal ECHI definitions because of issues related to data availability and comparability. Detailed information is also provided in this publication on health expenditure and financing trends, based on the OECD’s long-standing data collection in this area. All indicators are presented in the form of easy-to-read figures and explanatory text. Structure of the publication The structure of Health at a Glance: Europe 2010 generally reflects the structure of the European Community Health Indicators. It is divided into four chapters: ● Chapter 1 on Health Status highlights the variations across countries in life expectancy and healthy life expectancy, and also presents other indicators of causes of mortality and morbidity, including both communicable and non-communicable diseases. ● Chapter 2 on Determinants of Health focuses on non-medical determinants of health related to modifiable lifestyles and behaviours among children and adults, such as smoking and alcohol drinking, nutrition habits, physical activity, and overweight and obesity. ● Chapter 3 on Health Care Resources, Services and Outcomes reviews some of the inputs, outputs and outcomes of health care systems, including the supply of doctors and nurses, different types of equipment used for diagnosis or treatment, and the provision of a range of services to prevent the transmission of communicable diseases or to treat acute conditions. It concludes with a review of care related to cancer, focusing on the coverage of screening programmes and survival rates for two types of cancer: breast and cervical cancer. ● Chapter 4 on Health Expenditure and Financing examines trends in health spending across European countries, both overall and for different types of health services and goods, including pharmaceuticals. It also looks at how these health services and goods are paid for and the different mix between public funding, private health insurance, and direct out-of-pocket payments by households. An annex provides some additional tables on the demographic and economic context within which different health systems operate. 21 INTRODUCTION Presentation of indicators Each of the topics covered in this publication is presented over two pages. The first provides a brief commentary highlighting the key findings conveyed by the data, defines the indicator(s) and discloses any significant national variations from that definition which might affect data comparability. On the facing page is a set of figures. These typically show current levels of the indicator and, where possible, trends over time. In some cases, an additional figure relating the indicator to another variable is included. The average in the figures includes only European Union (EU) countries, and is calculated as the unweighted average of those EU countries presented (up to 27, if there is full data coverage). Data limitations Limitations in data comparability are indicated both in the text (in the box related to “Definition and deviations”) as well as in footnotes to charts. Readers interested in using the data presented in this publication for further analysis and research are encouraged to consult the full documentation of definitions, sources and methods contained in OECD Health Data 2010 for all OECD member countries. This information is available at www.oecd.org/health/healthdata. For the six non-OECD member countries (Bulgaria, Cyprus, Latvia, Lithuania, Malta and Romania), readers should consult the Eurostat Database at http://epp.eurostat.ec.europa.eu/ portal/page/portal/statistics/search_database. Readers interested in an interactive presentation of the ECHI indicators can also consult the SANCO health indicators tool at www.ec.europa.eu/health/indicators/indicators/index_en.htm. Population figures The population figures presented in the annex and used to calculate rates per capita in this publication come from the OECD Labour Force Statistics Database (as of May 2010) for OECD member countries, and refer to mid-year estimates. For the six non-OECD member countries, the data come from the Eurostat Demographics Database (as of July 2010), and refer to estimates at the beginning of the year. Population estimates are subject to revision, so they may differ from the latest population figures released by national statistical offices. Note that some countries such as France and the United Kingdom have overseas colonies, protectorates and territories. These populations are generally excluded. The calculation of GDP per capita and other economic measures may, however, be based on a different population in these countries, depending on the data coverage. 22 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 INTRODUCTION Country codes (ISO codes) Austria AUT Lithuania LTU Belgium BEL Luxembourg LUX Bulgaria BGR Malta MLT Cyprus1 CYP Netherlands NLD Czech Republic CZE Norway NOR Denmark DNK Poland POL Estonia EST Portugal PRT Finland FIN Romania ROM France FRA Slovak Republic SVK Germany DEU Slovenia SVN Greece GRC Spain ESP Hungary HUN Sweden SWE Iceland ISL Switzerland CHE Ireland IRL Turkey TUR Italy ITA United Kingdom GBR Latvia LVA 1. Note by Turkey: The information in this document with reference to “Cyprus” relates to the Southern part of the Island. There is no single authority representing both Turkish and Greek Cypriot people on the Island. Turkey recognises the Turkish Republic of Northern Cyprus (TRNC). Until a lasting and equitable solution is found within the context of United Nations, Turkey shall preserve its position concerning the “Cyprus” issue. Note by all the European Union member states of the OECD and the European Commission: The Republic of Cyprus is recognised by all members of the United Nations with the exception of Turkey. The information in this document relates to the area under the effective control of the Government of the Republic of Cyprus. HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 23 Health at a Glance: Europe 2010 © OECD 2010 Chapter 1 Health Status 1.1. Life expectancy and healthy life expectancy at birth . . . . . . . . . . . . . 26 1.2. Life expectancy and healthy life expectancy at age 65. . . . . . . . . . . . 28 1.3. Mortality from all causes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 1.4. Mortality from heart disease and stroke. . . . . . . . . . . . . . . . . . . . . . . . 32 1.5. Mortality from cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 1.6. Mortality from transport accidents . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 1.7. Suicide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 1.8. Infant mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 1.9. Infant health: Low birth weight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 1.10. Self-reported health and disability . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 1.11. Incidence of selected communicable diseases . . . . . . . . . . . . . . . . . . 46 1.12. HIV/AIDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 1.13. Cancer incidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 1.14. Diabetes prevalence and incidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 1.15. Dementia prevalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 25 1.1. LIFE EXPECTANCY AND HEALTHY LIFE EXPECTANCY AT BIRTH Life expectancy at birth continues to increase 1.9 years more HLY for men at birth than women in remarkably in EU countries, reflecting reductions in the Netherlands. Of the remaining countries, Poland mortality rates at all ages. These gains in longevity had the largest gender gap in HLY at birth favouring can be attributed to a number of factors, including women. rising living standards, improved lifestyle and better Higher national income (as measured by GDP per education, as well as greater access to quality health capita) is generally associated with higher life expec- services. Other factors, such as better nutrition, tancy at birth, although the relationship between GDP sanitation and housing also play a role, particularly in and HLY is less obvious (Figure 1.1.2). There is a countries with developing economies (OECD, 2004). modest positive relationship, with increasing GDP per Ave ra g e l i f e e x p e c t a n cy a t b i r t h f o r t h e capita associated with increasing HLY, although it is years 2005-07 across the 27 countries of the European less pronounced at higher levels of national income. Union reached 74.3 years for men and 80.8 years for There are also notable differences in HLY between EU women (Figure 1.1.1), a rise of approximately three countries with similar income per capita. Sweden and years for men and two years for women over the the United Kingdom have higher, and Finland and decade from 1995-97. In around 70% of EU countries, Estonia lower HLY than would be predicted by their life expectancy at birth in 2005-07 exceeded 80 years GDP alone. Similarly, Figure 1.1.3 shows the relation- for women and 77 years for men. France had the high- ship between HLY at birth and health spending per est life expectancy at birth for women (84.4 years), capita. Higher health spending per capita is generally while Sweden had the highest life expectancy at birth associated with higher HLY. for men (78.8 years). At the other end of the scale, life expectancy at birth in the European Union was lowest in Romania for women (76.2 years) and Lithuania for Definition and deviations men (65.1 years). The gap between EU countries with the highest and lowest life expectancies at birth is Life expectancy measures how long, on around eight years for women and 14 years for men. average, people would live based on a given set The gender gap in life expectancy at birth of age-specific death rates. However, the actual in 2005-07 stood at 6.5 years, almost one year less age-specific death rates of any particular birth than a decade earlier. However, this average hides a cohort cannot be known in advance. If age- huge range among countries with the smallest gender specific death rates are falling (as has been the gap in life expectancy at birth in the United Kingdom case over the past decades in EU countries), and Cyprus (4.1 years) and the largest in Lithuania actual life spans will be higher than life expec- (12.1 years). The recent narrowing of the gender gap in tancy calculated with current death rates. life expectancy can be attributed at least partly to the Healthy life years (HLY) at a particular age are narrowing of differences in risk-increasing behaviours the number of years spent free of activity limita- between men and women, such as smoking, accom- tion. They are calculated by Eurostat for each EU panied by sharp reductions in mortality rates from country using the Sullivan method (Sullivan, cardio-vascular diseases among men. 1971). The underlying health measure is the On average for EU countries healthy life years Global Activity Limitation Indicator (GALI) which (HLY) at birth in 2005-07 was 61.3 years for women comes from the European Union Statistics on and 60.1 years for men. HLY at birth in 2005-07 was Income and Living Conditions (EU-SILC) survey. greatest in Malta for both men and women, and The GALI measures limitation in usual activities. shortest in Latvia for women and Estonia for men The questionnaire responses used in Denmark (Figure 1.1.1). The spread of values for HLY at birth differ slightly, resulting in an under-estimation among EU countries were much greater than for of activity limitation. Data are not available for life expectancy, being 17.0 years for women and Bulgaria, Switzerland and Turkey. 19.5 years for men, but there was a much smaller Comparing trends in HLY and life expectancy absolute difference between men and women can show whether extra years of life are healthy (2.5 years). Since the HLY indicator has only recently years. However, valid comparisons depend on been developed, there is as yet no long time series. the underlying health measure being truly In contrast to the 6.5 year gap in life expectancy comparable. While HLY is the most comparable at birth for EU countries on average, the gender gap in indicator to date, there are still problems with HLY at birth was 1.2 years in 2005-07. For life expec- translation of the GALI question, although it tancy at birth the gender gap is always in favour of does appear to satisfactorily reflect other health women. However, eight countries had a gender gap in and disability measures (Jagger et al., 2010). HLY at birth which favoured men, the greatest being 26 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.1. LIFE EXPECTANCY AND HEALTHY LIFE EXPECTANCY AT BIRTH 1.1.1. Life expectancy and healthy life years (HLY) at birth, by gender, 2005-07 HLY LE with activity limitation Females Males 84.4 France 77.2 84.2 Switzerland 79.1 84.1 Spain 77.5 84.0 Italy 78.4 83.3 Iceland 79.5 83.0 Sweden 78.8 82.9 Finland 75.8 82.8 Norway 78.1 82.7 Austria 77.1 82.3 Germany 77.1 82.2 Belgium 76.6 82.1 Netherlands 77.7 82.1 Luxembourg 76.7 82.0 Ireland 77.4 82.0 Portugal 75.5 81.9 Cyprus 77.8 81.8 Greece 77.0 81.7 Malta 77.2 81.6 Slovenia 74.3 81.4 United Kingdom 77.3 80.8 EU 74.3 80.6 Denmark 76.1 79.8 Czech Republic 73.4 79.6 Poland 70.9 78.5 Estonia 67.3 78.3 Slovak Republic 70.4 77.6 Hungary 69.1 77.2 Lithuania 65.1 76.4 Latvia 65.5 76.4 Bulgaria 69.2 76.2 Romania 69.2 75.3 Turkey 71.0 90 80 70 60 50 40 30 30 40 50 60 70 80 90 Years Years Source: European Health and Life Expectancy Information System (EHLEIS); OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932335400 1.1.2. Healthy life years (HLY) at birth, 2005-07 1.1.3. Healthy life years (HLY) at birth, 2005-07 and GDP per capita, 2007 and health spending per capita, 2007 HLY (years) HLY (years) 70 70 ISL R 2 = 0.31 R 2 = 0.33 ISL DNK DNK GRC GRC SWE GBR SWE GBR 65 ITA NLD NOR 65 ITA NLD NOR FRA IRL IRL ESP ESP FRA ROU BEL LUX ROU LUX CYP CYP BEL POL POL 60 CZE 60 CZE SVN AUT SVN AUT PRT PRT DEU DEU SVK HUN 55 HUN 55 SVK LTU FIN FIN LTU LVA LVA EST EST 50 50 0 10 000 20 000 30 000 40 000 50 000 60 000 0 1000 2000 3000 4000 5000 GDP per capita (EUR PPP) Health spending per capita (EUR PPP) Source: European Health and Life Expectancy Information System Source: European Health and Life Expectancy Information System (EHLEIS); OECD Health Data 2010; Eurostat Statistics Database; WHO. (EHLEIS); OECD Health Data 2010; Eurostat Statistics Database; WHO. 1 2 http://dx.doi.org/10.1787/888932335419 1 2 http://dx.doi.org/10.1787/888932335438 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 27 1.2. LIFE EXPECTANCY AND HEALTHY LIFE EXPECTANCY AT AGE 65 Life expectancy at age 65 has increased signifi- women than for men. Longer life expectancy at age cantly among both women and men over the past 65 does not necessarily imply more HLY. several decades in all EU countries. Some of the Contrary to life expectancy where the rankings factors explaining the gains in life expectancy at age for men and women are different, there is a close 65 include advances in medical care combined with association between HLY at age 65 for men and greater access to health care, healthier lifestyles and women. At the overall EU level, this consistency improved living conditions before and after people between the number of years spent free of activity reach age 65. limitation (HLY) between men and women at birth The average life expectancy at age 65 years and at age 65 is true also for intermediate ages. in 2005-07 for the 27 countries of the European Union Women’s longer life expectancy at all ages are more was 15.9 years for men and 19.5 years for women often years spent with activity limitation. Lower HLY (Figure 1.2.1). As for life expectancy at birth, France at age 50 across EU countries has been shown to be had the highest life expectancy at age 65 for women associated with lower GDP and with higher long-term (22.6 years) but also for men (18.1 years). Life expec- unemployment and lower life-long learning for men tancy at age 65 in the European Union was lowest in (Jagger et al., 2008). Eastern Europe – in Latvia for men (12.7 years) and in Bulgaria for women (16.3 years). The average gender gap in life expectancy at age 65 in 2005-07 stood at 3.6 years, down from the previous Definition and deviations decade by 0.4 years. Greece had the smallest gender gap Life expectancy measures how long, on of two years and Estonia the largest at 5.1 years. average, people would live based on a given set Gains in longevity at older ages in recent decades of age-specific death rates. However, the actual in EU countries, combined with the trend reduction in age-specific death rates of any particular birth fertility rates are contributing to a steady rise in the cohort cannot be known in advance. If age- proportion of older persons in EU countries (see specific death rates are falling (as has been the Annex Tables A.2 and A.4). Whether longer life expec- case over the past decades in EU countries), tancy is accompanied by good health and functional actual life spans will be higher than life expec- status among ageing populations has important tancy calculated with current death rates. implications for health and long-term care systems. Healthy life years (HLY) at a particular age are As is the case for HLY at birth, HLY at age 65 the number of years spent free of activity limita- in 2005-07 for EU countries was similar for men and tion. They are calculated by Eurostat for each women, being 8.4 years for men and 8.1 years for EU country using the Sullivan method (Sullivan, women. HLY at age 65 in 2005-07 was greatest in 1971). The underlying health measure is the Denmark and shortest in Estonia for both men and Global Activity Limitation Indicator (GALI) which women (Figure 1.2.1). It should be noted though, that comes from the European Union Statistics on the question used to measure activity limitation in Income and Living Conditions (EU-SILC) survey. Denmark differs slightly from that used in other The GALI measures limitation in usual activities. countries, resulting in an over-estimation of HLY. HLY The questionnaire responses used in Denmark is based on the Global Activity Limitation (GALI) differ slightly, resulting in an under-estimation question, which is one of three indicators included in of activity limitation. Data are not available for the Minimum European Health Module along with Bulgaria, Switzerland and Turkey. global items on self-perceived health and chronic Comparing trends in HLY and life expectancy morbidity. Health expectancies based on these alter- can show whether extra years of life are healthy native questions would rank the countries differently. years. However, valid comparisons depend on In addition, since the HLY indicator has only been the underlying health measure being truly developed relatively recently, there is as yet no long comparable. While HLY is the most comparable time series. indicator to date, there are still problems with The relationship between life expectancy and translation of the GALI question, although it HLY at age 65 is not clear-cut (Figure 1.2.2). Higher life does appear to satisfactorily reflect other health expectancy at age 65 is generally associated with and disability measures (Jagger et al., 2010). higher HLY, but the relationship is less pronounced for 28 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.2. LIFE EXPECTANCY AND HEALTHY LIFE EXPECTANCY AT AGE 65 1.2.1. Life expectancy and healthy life years (HLY) at 65, by gender, 2005-07 HLY LE with activity limitation Females Males 22.6 France 18.1 22.0 Switzerland 18.4 21.7 Spain 17.6 21.6 Italy 17.8 21.1 Finland 16.9 20.9 Iceland 18.4 20.9 Norway 17.4 20.8 Sweden 17.7 20.7 Austria 17.3 20.6 Belgium 17.0 20.4 Germany 17.1 20.4 Netherlands 16.8 20.3 Luxembourg 16.7 20.1 Ireland 16.9 19.9 Portugal 16.5 19.8 Slovenia 15.6 19.8 United Kingdom 17.2 19.5 EU 15.9 19.5 Malta 16.3 19.5 Cyprus 17.3 19.4 Greece 17.3 19.2 Denmark 16.3 18.8 Poland 14.5 18.3 Estonia 13.1 18.2 Czech Republic 14.8 17.7 Lithuania 13.0 17.6 Hungary 13.5 17.3 Slovak Republic 13.4 17.2 Latvia 12.7 16.5 Romania 13.6 16.3 Bulgaria 13.2 15.7 Turkey 13.9 25 20 15 10 5 0 0 5 10 15 20 25 Years Years Source: European Health and Life Expectancy Information System (EHLEIS); Eurostat Statistics Database; OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932335457 1.2.2. Relationship between life expectancy and healthy life years (HLY) at 65, 2005-07 Females Males Life expectancy (years) Life expectancy (years) 24 20 R 2 = 0.36 R 2 = 0.55 FRA ISL 22 18 ITA FRA ITA ESP ESP GBR SWE AUT CYP GRC NOR FIN NOR BEL SWE ISL DEU FIN NLD AUT DEU LUX BEL IRL PRT LUX SVN NLD MLT DNK 20 PRT IRL GBR 16 EU EU SVN CYP MLT DNK GRC POL CZE POL EST CZE 18 14 HUN LTU SVK ROU HUN SVK LVA EST LTU ROU LVA 16 12 2 4 6 8 10 12 14 16 2 4 6 8 10 12 14 16 Healthy life years (HLY) Healthy life years (HLY) Source: European Health and Life Expectancy Information System (EHLEIS); Eurostat Statistics Database; OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932335476 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 29 1.3. MORTALITY FROM ALL CAUSES Mortality rates are one of the most common are in the order of eight years for females and 12 years measures of population health. Statistics on deaths for males. Some important causes of mortality below remain one of the most widely available and compa- the age of 65 years that may be avoided through effec- rable sources of information on health. Registering tive evidence-based public health measures include deaths is compulsory in all European Union countries, ischemic heart disease, lung cancer, alcohol-related and the data collected through the process of registra- mortality, suicide, transport accidents, cervical cancer tion can be used by statistical and health authorities and AIDS (Cayotte and Buchow, 2009). to monitor diseases and health status, and to plan Although mortality rates in Central and Eastern health services. In order to compare levels of mortal- Europe are still comparatively high, significant declines ity across countries and over time, the data need to be have occurred in a number of these countries since 1994 aggregated in suitable ways and standardised for (Figures 1.3.2 and 1.3.3). Mortality rates in Estonia, differences in age-structure. Slovenia, the Czech Republic, Hungary and Poland have In 2008 there were large variations in age- fallen by more than 30%, a decline that is greater than standardised total mortality rates for all causes of the EU average. Ireland has also seen a fall in mortality death across European Union countries. Death rates rates of over 50%. In contrast, declines in the Slovak were lowest in Switzerland, Italy, Iceland and Spain, at Republic and Lithuania have been small. Declines in a 520 deaths per 100 000 population or less (Figure 1.3.1). number of Nordic countries (Sweden, Iceland) have also Rates in northern, western and southern European been modest, although these countries began the period countries were lower than the EU average rate of 696. with rates that were already low. They were highest in central and eastern European The leading causes of death in EU countries countries – Lithuania and Latvia, for instance, had include cardiovascular diseases (such as heart attack age-standardised rates twice those of the lowest and stroke), and cancer. Deaths from these diseases, countries at over 1 000 deaths per 100 000 population. plus selected external causes of death (transport Rates in Bulgaria, Romania, Hungary and a number accidents and suicide), are examined more closely in of other central and eastern European countries were the following four indicators. above 800. Among these countries, only Slovenia had a mortality rate that was lower than the EU average. Male mortality rates were lowest in Iceland, Switzerland and Sweden, and high in Lithuania, Latvia and Estonia. Female rates were low in France, Definition and deviations Spain and Switzerland, and high in Bulgaria, Romania Mortality rates are based on numbers of and Lithuania. A significant gender gap exists in deaths registered in a country in a year divided mortality rates (Figure 1.3.1). Across all EU countries, by the size of the corresponding population. The the male mortality rate was, on average, 70% higher rates have been directly age-standardised to the than the female rate in 2008. But large differences WHO European standard population to remove exist among countries – in Estonia, Lithuania and variations arising from differences in age struc- Latvia, male rates were more than twice those of tures across countries and over time. The source females, whereas in Iceland, the United Kingdom and is the Eurostat Statistics Database. Greece they were around 40% higher. Mathers et al. (2005) have provided a general Lower mortality rates translate into higher life assessment of the coverage, completeness and expectancies. In 2005-07, average life expectancy reliability of data on causes of death. across all EU countries was approximately 81 years for females and 74 years for males (see Indicator 1.1). Deaths from all causes are classified to ICD-10 However, the differences in life expectancy among codes A00-Y89, excluding S00-T98. countries with the lowest and highest mortality rates 30 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.3. MORTALITY FROM ALL CAUSES 1.3.1. Mortality rates from all causes of death, 2008 (or nearest year available) Males Females Total population 394 490 Switzerland 614 401 512 Italy 658 447 516 Iceland 592 387 520 Spain 682 381 522 France 702 451 541 Sweden 652 446 549 Norway 683 451 550 Cyprus 661 440 557 Austria 710 475 566 Netherlands 688 470 568 Ireland 685 467 582 Germany 721 439 587 Finland 771 479 590 Malta 735 496 596 Greece 706 504 599 United Kingdom 714 487 603 Luxembourg 769 478 624 Portugal 808 472 634 Slovenia 849 567 683 Denmark 828 534 696 EU 912 577 747 Czech Republic 968 589 819 Poland 1 125 657 888 Slovak Republic 1 204 613 894 Estonia 1 324 686 926 Hungary 1 258 740 964 Romania 1 237 766 996 Bulgaria 1 279 707 1 007 Latvia 1 467 712 1 034 Lithuania 1 504 1 200 1 000 800 600 400 200 0 0 500 1 000 1 500 2 000 Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335495 1.3.2. Decline in all cause mortality rates, 1994-2008 1.3.3. Trends in all cause mortality rates, (or nearest year available) selected EU countries, 1994-2008 Estonia 54 Hungary Ireland Ireland 51 Slovenia 47 Italy EU Czech Republic 38 Age-standardised rates per 100 000 population Portugal 37 1 400 Hungary 36 Austria 36 Germany 34 Italy 34 Malta 32 1 200 Poland 31 Finland 31 Switzerland 30 EU 30 1 000 Norway 29 Spain 29 Netherlands 28 United Kingdom 26 Bulgaria 25 800 Denmark 24 Luxembourg 23 Latvia 22 Iceland 21 600 Greece 20 Sweden 19 Lithuania 16 Slovak Republic 15 400 0 10 20 30 40 50 60 1994 1996 1998 2000 2002 2004 2006 2008 Percentage decline Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335514 1 2 http://dx.doi.org/10.1787/888932335533 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 31 1.4. MORTALITY FROM HEART DISEASE AND STROKE Cardiovascular diseases are the main cause of of countries, however, have seen little or no decline mortality in almost all European Union countries, since 1994. In the Slovak Republic, mortality rates accounting for 40% of all deaths in the region in 2008. have increased slightly. Declines in Poland, Hungary They cover a range of diseases related to the circula- and Lithuania have been moderate, at under 20%. tory system, including ischemic heart disease (known Stroke is another important cause of mortality in as IHD, or heart attack) and cerebro-vascular disease EU countries, accounting for about 10% of all deaths (or stroke). Together, IHD and stroke comprise 60% of in 2008. It is caused by the disruption of the blood all cardiovascular deaths, and caused one-quarter of supply to the brain, and in addition to being an impor- all deaths in EU countries in 2008. tant cause of mortality, the disability burden from Ischemic heart disease is caused by the accumu- stroke is substantial (Moon et al., 2003). As with IHD, lation of fatty deposits lining the inner wall of a there are large variations in stroke mortality rates coronary artery, restricting blood flow to the heart. across countries (Figure 1.4.2). Again, the rates are IHD alone was responsible for 15% of all deaths in EU highest in central and eastern European countries, countries in 2008. Mortality from IHD varies consider- including Bulgaria, Romania, Latvia, Lithuania, the ably, however, across EU countries (Figure 1.4.1). Slovak Republic and Hungary. They are the lowest in Central and eastern European countries report the Switzerland, France, Iceland and the Netherlands. highest IHD mortality rates, Lithuania for both males Looking at trends over time, stroke mortality has and females, followed by Latvia, the Slovak Republic decreased in all EU countries (except the Slovak and Estonia. IHD mortality rates are also relatively Republic and Poland) since 1994, with a more high in Finland, Malta and Ireland, with rates several pronounced fall after 1999 (Figure 1.4.4). Rates have times higher than in France, Portugal, the Netherlands declined by one-half or more in Italy, Estonia, and Spain. There are regional patterns to the variabil- Portugal, Austria, Germany and the Czech Republic. ity in IHD mortality rates. Besides the Netherlands, As with IHD, the reduction in stroke mortality can be the countries with the lowest IHD mortality rates are attributed at least partly to a reduction in risk factors. four countries located in Southern Europe: France, Tobacco smoking and hypertension are the main Portugal, Spain and Italy, with Greece also having low modifiable risk factors for stroke. Improvements in rates. This lends support to the commonly held medical treatment for stroke have also increased hypothesis that there are underlying risk factors, such survival rates. as diet, which explain differences in IHD mortality across countries. Death rates are much higher for men than for women in all countries (Figure 1.4.1). On average across Definition and deviations EU countries, IHD mortality rates for men in 2008 were Mortality rates are based on numbers of nearly two times greater than for women. deaths registered in a country in a year divided Since the mid-1990s, IHD mortality rates have by the size of the corresponding population. The declined in nearly all countries (Figure 1.4.3). The rates have been directly age-standardised to the decline has been most remarkable in the Netherlands, WHO European standard population to remove Denmark and Norway among the Nordic countries, variations arising from differences in age Ireland, Slovenia and Estonia (although rates there are structures across countries and over time. The still high), with IHD mortality rates being cut by source is the Eurostat Statistics Database. one-half or more. A number of factors are responsible, Mathers et al. (2005) have provided a general with declines in tobacco consumption, and heavy assessment of the coverage, completeness and drinking in some countries reducing the incidence of reliability of data on causes of death. IHD, and consequently reducing IHD mortality rates. Significant improvements in medical care for treating Deaths from ischemic heart disease are classi- IHD have also played a part (Moïse et al., 2003) (see fied to ICD-10 codes I20-I25, and stroke to I60-I69. Indicator 3.9 “Cardiac procedures”). A small number 32 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.4. MORTALITY FROM HEART DISEASE AND STROKE 1.4.1. Ischemic heart disease, mortality rates, 2008 1.4.2. Stroke, mortality rates, 2008 (or nearest year available) (or nearest year available) Males Females Males Females France 21 55 Switzerland 27 31 Portugal 32 61 France 24 34 Netherlands 31 68 Iceland 34 38 Spain 30 69 Netherlands 34 38 Italy 44 86 Austria 32 40 Luxembourg 45 91 Ireland 39 40 Switzerland 45 93 Cyprus 40 41 Greece 41 96 Germany 36 42 Denmark 52 98 Spain 34 43 Norway 46 99 Norway 37 45 Slovenia 42 102 Sweden 38 46 Cyprus 44 107 Luxembourg 47 49 Germany 62 117 United Kingdom 47 50 Sweden 64 130 Finland 40 53 Austria 73 131 Italy 43 53 United Kingdom 61 132 Malta 57 61 Iceland 60 133 Denmark 47 59 Ireland 68 143 Slovenia 53 76 Poland 69 147 Greece 78 83 Malta 86 162 EU 64 82 EU 89 171 Czech Republic 71 87 Bulgaria 88 174 Portugal 70 89 Finland 88 183 Poland 64 91 Czech Republic 137 228 Estonia 63 97 Romania 151 248 Hungary 76 115 Hungary 166 289 Slovak Republic 83 124 Estonia 163 330 Lithuania 110 140 Slovak Republic 224 361 Latvia 123 174 Latvia 184 397 Romania 154 197 Lithuania 240 449 Bulgaria 163 217 0 100 200 300 400 500 0 50 100 150 200 250 Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335552 1 2 http://dx.doi.org/10.1787/888932335571 1.4.3. Trends in ischemic heart disease mortality 1.4.4. Trends in stroke mortality rates, rates, selected EU countries, 1994-2008 selected EU countries, 1994-2008 Denmark Hungary Bulgaria Italy Netherlands EU Portugal EU Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population 300 300 250 250 200 200 150 150 100 100 50 50 0 0 1994 1996 1998 2000 2002 2004 2006 2008 1994 1996 1998 2000 2002 2004 2006 2008 Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335590 1 2 http://dx.doi.org/10.1787/888932335609 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 33 1.5. MORTALITY FROM CANCER Cancer is the second leading cause of mortality in Prostate cancer has become the most commonly EU countries (after diseases of the circulatory system), occurring cancer among men in many EU countries, accounting for 26% of all deaths in 2008. Cancer particularly for those aged over 65 years of age, although mortality rates for the total population were the lowest death rates from prostate cancer remain lower than for in Cyprus, Finland, Switzerland and Sweden, at under lung cancer in all countries except Sweden. The rise in 150 deaths per 100 000 population. They were the the reported incidence of prostate cancer in many coun- highest in central and eastern European countries tries during the 1990s and 2000s was largely due to the (Hungary, Poland, the Czech and Slovak Republics, greater use of prostate-specific antigen (PSA) diagnostic Slovenia) and Denmark, above 200 deaths per tests. Death rates from prostate cancer in 2008 varied 100 000 population. from lows of less than 15 per 100 000 males in Malta and Cancer mortality rates are higher for men than Romania, to highs of more than 30 per 100 000 males in for women in all EU countries (Figure 1.5.1). In 2008, a range of central and eastern European and Nordic the gender gap in death rates from cancer was partic- countries (Figure 1.5.4). The causes of prostate cancer ularly wide in Latvia, Spain, Estonia, France, Lithuania are not well-understood. Some evidence suggests and Portugal, with mortality rates among men more that environmental and dietary factors might influ- than twice as high as for women. This gap can be ence the risk of prostate cancer (Institute of Cancer explained partly by the greater prevalence of risk Research, 2009). factors among men, as well as the lesser availability or Death rates from all types of cancer for males use of screening programmes for different types of and females have declined at least slightly in most EU cancers affecting men, leading to lower survival rates countries since 1994, although the decline has been after diagnosis. more modest than for cardiovascular diseases, Lung cancer still accounts for the greatest num- explaining why cancer accounts now for a larger share ber of cancer deaths among men in all EU countries, of all deaths. The exceptions to this declining pattern except in Sweden. Lung cancer is also one of the main are among central and eastern European countries causes of cancer mortality among women. Tobacco (Bulgaria, Romania, Latvia, Lithuania, Poland) and smoking is the most important risk factor for lung Greece, where cancer mortality has remained static or cancer. In 2008, death rates from lung cancer among increased between 1994 and 2008. men were the highest in central and eastern European countries (Hungary, Poland, Estonia, Latvia, Lithuania and others) (Figure 1.5.2). These are all countries Definition and deviations where smoking rates among men are relatively high. Mortality rates are based on numbers of Death rates from lung cancer among men are low deaths registered in a country in a year divided in Nordic countries (Sweden, Iceland, Finland by the size of the corresponding population. The and Norway) as well as in Cyprus, countries with rates have been directly age-standardised to the low smoking rates among men (see Indicator 2.6). WHO European standard population to remove Denmark and Iceland, however, have high rates of variations arising from differences in age struc- lung cancer mortality among women. tures across countries and over time. The source Breast cancer is the most common form of cancer is the Eurostat Statistics Database. among women in all EU countries (Ferlay et al., 2010). The international comparability of cancer It accounted for 31% of cancer incidence among mortality data can be affected by differences women, and 17% of cancer deaths in 2008. While there in medical training and practices as well as in has been an increase in incidence rates of breast death certification procedures across countries. cancer over the past decade, death rates have declined Mathers et al. (2005) have provided a general or remained stable, indicating increases in survival assessment of the coverage, completeness and rates due to earlier diagnosis and/or better treatments reliability of data on causes of death. (see Indicator 3.13). The lowest mortality rates from Deaths from all cancers are classified to ICD-10 breast cancer are in Spain, Norway, Finland and codes C00-C97, lung cancer to C32-C34, breast Portugal (below 20 deaths per 100 000 females), while cancer to C50 and prostate cancer to C61. the highest mortality rates are in Ireland and Denmark (above 30) (Figure 1.5.3). 34 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.5. MORTALITY FROM CANCER 1.5.1. All cancers mortality rates, males 1.5.2. Lung cancer mortality rates, males and females, 2008 (or nearest year available) and females, 2008 (or nearest year available) Males Females Males Females Cyprus 98 151 Sweden 23 29 Iceland 152 171 Iceland 38 41 Finland 114 172 Cyprus 8 39 Sweden 134 172 Finland 14 42 Switzerland 117 187 Norway 28 44 Malta 130 195 Switzerland 19 45 Norway 137 196 Portugal 8 47 Ireland 157 204 Malta 9 48 Germany 131 206 Ireland 28 49 Greece 113 210 Austria 20 51 Austria 128 211 United Kingdom 32 53 United Kingdom 154 212 Germany 20 55 Portugal 109 218 Italy 14 64 Italy 123 220 France 15 65 Luxembourg 137 221 Denmark 45 66 Spain 102 222 Luxembourg 29 66 Netherlands 155 228 EU 19 66 Bulgaria 127 230 Netherlands 33 68 EU 136 236 Spain 10 69 France 116 237 Czech Republic 20 72 Denmark 182 246 Bulgaria 11 73 Romania 128 246 Greece 13 74 Czech Republic 153 269 Slovenia 20 74 Slovenia 152 277 Slovak Republic 14 75 Poland 152 284 Romania 13 77 Slovak Republic 144 288 Lithuania 9 83 Estonia 136 290 Latvia 8 86 Lithuania 142 290 Estonia 12 89 Latvia 295 Poland 99 135 23 Hungary 177 337 Hungary 38 115 0 100 200 300 400 0 25 50 75 100 125 Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335628 1 2 http://dx.doi.org/10.1787/888932335647 1.5.3. Breast cancer mortality rates, females, 2008 1.5.4. Prostate cancer mortality rates, males, 2008 (or nearest year available) (or nearest year available) Spain 18.2 Malta 12.8 Norway 18.7 Romania 14.9 Finland 19.8 Italy 16.0 Portugal 19.8 Spain 17.0 Sweden 20.0 Greece 17.5 Luxembourg 20.5 Bulgaria 18.1 Czech Republic 21.2 Cyprus 19.0 Poland 21.2 Germany 20.6 Romania 21.6 France 21.6 Greece 21.7 Hungary 21.8 Austria 21.8 Poland 21.9 Slovak Republic 22.1 Austria 22.5 Switzerland 22.1 Slovak Republic 22.5 Estonia 22.6 Czech Republic 23.6 Cyprus 22.8 Ireland 23.6 Bulgaria 23.3 Portugal 24.0 Italy 23.6 EU 24.0 EU 23.8 Finland 24.3 France 24.0 Switzerland 24.5 Germany 24.6 Luxembourg 24.8 Latvia 24.7 United Kingdom 24.8 Lithuania 25.1 Netherlands 25.2 Hungary 26.6 Slovenia 31.2 United Kingdom 26.8 Iceland 32.1 Iceland 27.3 Sweden 34.0 Slovenia 27.4 Lithuania 34.4 Malta 27.9 Denmark 34.5 Netherlands 29.0 Norway 34.7 Denmark 31.1 Latvia 35.1 Ireland 31.1 Estonia 37.6 0 10 20 30 40 0 10 20 30 40 Age-standardised rates per 100 000 females Age-standardised rates per 100 000 males Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335666 1 2 http://dx.doi.org/10.1787/888932335685 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 35 1.6. MORTALITY FROM TRANSPORT ACCIDENTS Worldwide, an estimated 1.2 million people are EU countries since 1994 (Figures 1.6.2 and 1.6.3). killed in transport accidents each year, mostly due to Estonia has seen the largest decline in transport road traffic accidents, and as many as 50 million accident mortality of 78% between 1994 and 2008, people are injured or disabled (WHO, 2009b). In EU with most of the fall occurring in the mid-1990s countries alone, they were responsible for approxi- following independence. Reductions in Portugal, mately 48 000 deaths in 2008. In 2008, Italy, Poland, Sweden, Slovenia and Germany since 1994 are close France and Germany each experienced around 5 000- to 60%, although vehicle kilometers travelled have 6 000 transport accident deaths. increased by 2.7 times on average in European coun- Mortality from road accidents is the leading tries in the same period (OECD/ITF, 2008). Death rates cause of death among children and young people, and have also declined in Greece, but at a slower pace, and especially young men, in many countries. The fatality therefore remain above the EU average. In Bulgaria risk for motor cycles and mopeds is highest among all and Romania there have been significant increases in modes of transport, even though most fatal traffic death rates from road accidents since 1994. injuries occur in passenger vehicles (ETSC, 2003; Beck Based on an extrapolation of past trends, projec- et al., 2007). tions from the World Bank indicate that between 2000 Besides the social, physical and psychological and 2020, road traffic deaths may decline further by effects, the direct and indirect financial costs of trans- about 30% in high-income countries, but may increase port accidents are substantial; one estimate put these substantially in low- and middle-income countries if at 2% of gross national product annually in highly- no additional road safety counter-measures are put in motorised countries (Peden et al., 2004). Injury and place (Peden et al., 2004). mortality from transport accidents remains a serious public health concern. Death rates were the highest in 2008 in Lithuania, Definition and deviations Romania and Latvia, all in excess of 15 deaths per Mortality rates are based on numbers of deaths 100 000 population (Figure 1.6.1). They were the registered in a country in a year divided by the lowest in Malta, the Netherlands, Iceland, Sweden and size of the corresponding population. The rates Switzerland, at five deaths per 100 000 population or have been directly age-standardised to the WHO less. A four-fold difference exists between the coun- European standard population to remove varia- tries with the lowest and highest rates. Deaths from tions arising from differences in age structures transport accidents are much higher for males than across countries and over time. The source is the for females in all EU countries, with disparities in Eurostat Statistics Database. rates ranging from three times higher for males in Mathers et al. (2005) have provided a general Denmark, Sweden and Germany to five or more times assessment of the coverage, completeness and higher in the Slovak Republic, Slovenia and Poland. On reliability of data on causes of death. average, almost four times as many males than females die in transport accidents (Figure 1.6.1). Deaths from transport accidents are classified to ICD-10 codes V01-V99. The majority of deaths Much transport accident injury and mortality is from transport accidents are due to road traffic preventable. Road security has increased greatly over accidents. the past decades in many countries through improve- ments of road systems, education and prevention Mortality rates from transport accidents in campaigns, the adoption of new laws and regulations Luxembourg are biased upward because of the and the enforcement of these new laws through large volume of traffic in transit, resulting in a more traffic controls. As a result, death rates due to significant proportion of non-residents killed. transport accidents have been cut by around 40% in 36 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.6. MORTALITY FROM TRANSPORT ACCIDENTS 1.6.1. Transport accident mortality rates, 2008 (or nearest year available) Males Females Total population 3.6 Malta 1.4 6.0 4.1 Netherlands 2.0 6.4 4.9 Iceland 2.3 7.4 5.0 Sweden 2.5 7.5 5.0 Switzerland 1.9 8.3 5.3 United Kingdom 2.3 8.2 5.4 Germany 2.7 8.2 5.8 Denmark 3.1 8.5 6.0 Norway 2.3 9.9 6.2 Ireland 3.0 9.6 6.9 Finland 2.8 11.4 7.2 Spain 3.1 11.3 7.4 Austria 3.4 11.6 7.4 France 3.2 11.8 8.7 Luxembourg 1.7 15.6 9.1 Portugal 4.1 14.4 9.2 Italy 3.3 15.2 9.7 EU 4.1 15.6 10.3 Czech Republic 4.9 15.7 11.4 Estonia 5.6 18.2 11.5 Slovenia 3.6 19.5 11.6 Cyprus 5.4 18.2 11.7 Hungary 5.0 19.3 13.3 Bulgaria 5.7 21.4 13.3 Slovak Republic 4.6 22.7 14.1 Greece 5.2 22.9 14.6 Poland 4.1 23.4 15.9 Latvia 7.5 25.5 16.6 Romania 7.3 26.5 16.8 Lithuania 8.1 26.4 20 15 10 5 0 0 5 10 15 20 25 30 Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335704 1.6.2. Trends in transport accident mortality rates, 1.6.3. Change in transport accident mortality rates, selected EU countries, 1994-2008 1994-2008 (or nearest year available) Netherlands Poland Estonia -78 Portugal -59 Portugal EU Sweden -58 Slovenia -56 Age-standardised rates per 100 000 population Germany -55 25 Luxembourg -54 Austria -53 Switzerland -50 Netherlands -49 Spain -49 20 Ireland -47 Denmark -46 Latvia -43 Iceland -42 France -42 15 Finland -41 Czech Republic -39 EU -39 Hungary -38 Belgium -38 10 Malta -38 Lithuania -37 Poland -32 Italy -31 Greece -29 5 Norway -28 Cyprus -27 United Kingdom -18 Slovak Republic -15 Romania 5 Bulgaria 19 0 1994 1996 1998 2000 2002 2004 2006 2008 -100 -80 -60 -40 -20 0 20 Percentage change Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335723 1 2 http://dx.doi.org/10.1787/888932335742 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 37 1.7. SUICIDE The intentional killing of oneself is evidence not range of factors including rapid socio-economic tran- only of personal breakdown, but also of a deteriora- sition, increasing psychological and social insecurity tion of the social context in which an individual lives. and the absence of a national suicide prevention Suicide may be the end-point of a number of different strategy. Similarly in Hungary, societal factors includ- contributing factors. It is more likely to occur during ing employment and socio-economic circumstances, crisis periods associated with divorce, alcohol and as well as individual demographic and clinical factors drug abuse, unemployment, clinical depression and have been cited as determinants of suicide (Almasi other forms of mental illness. Because of this, suicide et al., 2009). is often used as a proxy indicator of the mental health Suicide is often linked with depression and the status of a population. However, the number of abuse of alcohol and other substances. Early detection suicides in certain countries may be under-estimated of these psycho-social problems in high-risk groups by because of the stigma that is associated with the act, families and health professionals must be part of or because of data issues associated with reporting suicide prevention campaigns, together with the criteria (see “Definitions and deviations”). provision of effective support and treatment. Many Suicide is a significant cause of death in many countries are promoting mental health and developing European Union countries, and there were approxi- national strategies for prevention, focusing on at-risk mately 55 000 such deaths in 2008. Rates of suicide groups (Hawton and van Heeringen, 2009). In Finland were low in southern European countries – Greece, and Iceland, suicide prevention programmes have been Cyprus, Italy, Malta, Spain and Portugal – as well as the based on efforts to promote strong multisectoral United Kingdom, at less than eight deaths per collaboration and networking (NOMESCO, 2007). 100 000 population (Figure 1.7.1). They were highest in the Baltic States and Central and Eastern Europe; in Lithuania, Hungary and Latvia, as well as Finland, there were more than 18 deaths per 100 000 population. Definition and deviations There is more than a ten-fold difference between Lithuania and Greece, the countries with the lowest The World Health Organization defines and high death rates. “suicide” as an act deliberately initiated and performed by a person in the full knowledge or In general, death rates from suicides are three-to- expectation of its fatal outcome. Comparability four times greater for men than for women across the of suicide data between countries is affected by European Union, except in those countries with the a number of reporting criteria, including how a highest rates, where rates are up to six times greater person’s intention of killing themselves is ascer- (Figure 1.7.1). The gender gap is narrower for attempted tained, who is responsible for completing the suicides, reflecting the fact that women tend to use less death certificate, whether a forensic investi- fatal methods than men. Suicide is also related to age, gation is carried out, and the provisions for with young people aged under 25 and elderly people confidentiality of the cause of death. Caution is especially at risk. While suicide rates among the latter required therefore in interpreting variations have generally declined over the past two decades, across countries. almost no progress has been observed among younger people. Mortality rates are based on numbers of deaths registered in a country in a year divided Since 1994, suicide rates have decreased in many by the size of the corresponding population. The EU countries, with pronounced declines of 40% or rates have been directly age-standardised to the more in Estonia, Latvia and Slovenia (Figure 1.7.2). WHO European standard population to remove Despite this progress, these three countries still have variations arising from differences in age struc- among the highest suicide rates in Europe. On the tures across countries and over time. The source other hand, death rates from suicides have increased is the Eurostat Statistics Database. since 1994 in Malta, Iceland and Portugal, though rates in Malta and Portugal still remain below the Mathers et al. (2005) have provided a general EU average. assessment of the coverage, completeness and reliability of data on causes of death. Following independence in 1990, suicide rates in Lithuania increased steadily, especially among young Deaths from suicide are classified to ICD-10 men, peaking in 1996 (Figure 1.7.3). The high suicide codes X60-X84. rates in Lithuania have been associated with a wide 38 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.7. SUICIDE 1.7.1. Suicide mortality rates, 2008 (or nearest year available) Males Females Total population 2.8 Greece 1.0 4.8 4.3 Cyprus 1.7 7.0 5.2 Italy 2.3 8.4 6.0 Malta 0.3 11.8 6.1 United Kingdom 2.7 9.7 6.5 Spain 2.8 10.4 7.9 Portugal 3.4 13.2 8.1 Netherlands 4.9 11.4 9.3 Ireland 4.2 14.5 9.4 Germany 4.4 14.9 10.1 Bulgaria 4.6 16.3 10.2 Norway 6.4 14.1 10.2 Slovak Republic 2.7 18.5 10.6 Denmark 5.7 16.0 10.6 Romania 3.5 18.4 11.4 Sweden 6.6 16.3 11.8 Czech Republic 4.2 20.2 12.0 EU 4.9 20.0 12.2 Iceland 7.5 16.7 12.7 Austria 5.8 20.7 13.9 Poland 3.7 24.9 14.6 France 7.4 22.7 15.1 Switzerland 9.1 21.8 16.0 Luxembourg 10.6 22.0 16.5 Estonia 6.2 29.1 17.3 Slovenia 6.7 29.2 18.4 Finland 8.3 29.0 20.9 Latvia 6.5 38.3 21.5 Hungary 8.6 37.1 30.7 Lithuania 9.1 55.9 35 30 25 20 15 10 5 0 0 10 20 30 40 50 60 Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335761 1.7.2. Change in suicide rates, 1994-2008 1.7.3. Trends in suicide rates, selected EU countries, (or nearest year available) 1994-2008 Estonia -60 Greece Hungary Latvia -44 Slovenia -43 Lithuania EU Denmark -39 Austria -38 Age-standardised rates per 100 000 population Hungary -36 50 Bulgaria -36 Lithuania -35 Germany -32 Czech Republic -32 Finland -30 40 Italy -25 Switzerland -23 EU -21 Slovak Republic -20 Ireland -20 30 Sweden -17 Netherlands -16 Norway -16 United Kingdom -14 Romania -14 20 Spain -12 Belgium -10 Greece -10 France -9 10 Luxembourg -9 Poland -6 Portugal 11 Iceland 24 Malta 58 0 -80 -60 -40 -20 0 20 40 60 80 1994 1996 1998 2000 2002 2004 2006 2008 Percentage change Source: Eurostat Statistics Database. Data are age-standardised to Source: Eurostat Statistics Database. Data are age-standardised to the WHO European standard population. the WHO European standard population. 1 2 http://dx.doi.org/10.1787/888932335780 1 2 http://dx.doi.org/10.1787/888932335799 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 39 1.8. INFANT MORTALITY Infant mortality, the rate at which babies of less rate among the lowest in Europe in 2008 (Figure 1.8.2). than one year of age die, reflects the effect of eco- Large reductions in infant mortality rates have also nomic and social conditions on the health of mothers been observed in Luxembourg, Slovenia and Greece. and newborns, as well as the quality of medical care On the other hand, the reduction in infant mortality and preventive services. rates has been slower in Latvia, Malta, Bulgaria and In 2008, infant mortality rates in European coun- the Netherlands. Infant mortality rates in Poland tries ranged from a low of less than three deaths per declined rapidly in the early 1990s to approach the 1 000 live births in Luxembourg, Slovenia, Nordic coun- EU average. tries (with the exception of Denmark), Greece and the Numerous studies have used infant mortality Czech Republic, up to a high of 11 and 17 deaths per rates as a health outcome to examine the effect of a 1 000 live births in Romania and Turkey respectively variety of medical and non-medical determinants of (Figure 1.8.1). Infant mortality rates were also relatively health (e.g. Joumard et al., 2008). Although most high (more than six deaths per 1 000 live births) in analyses show an overall negative relationship Latvia, Bulgaria and Malta. The average across the between infant mortality and health spending, the 27 European Union countries in 2008 was 4.6 deaths fact that some countries with a high level of health per 1 000 live births. Infant mortality rates tend to be expenditure do not necessarily exhibit low levels of higher than the EU average in central and eastern infant mortality has led some researchers to conclude European countries, with the exceptions of Slovenia that more health spending is not necessarily required and the Czech Republic, both of which have had consis- to obtain better results (Retzlaff-Roberts et al., 2004). A tently lower rates. body of research also suggests that many factors Around two-thirds of the deaths that occur dur- beyond the quality and efficiency of the health ing the first year of life are neonatal deaths (i.e. during system, such as income inequality, the social envi- the first four weeks). Birth defects, prematurity and ronment, and individual lifestyles and attitudes, other conditions arising during pregnancy are the influence infant mortality rates (Kiely et al., 1995). principal factors contributing to neonatal mortality in developed countries. With an increasing number of women deferring childbearing and the rise in multiple births linked with fertility treatments, the number Definition and deviations of pre-term births has tended to increase (see Indicator 1.9). In a number of higher-income coun- The infant mortality rate is the number of tries, this has contributed to a leveling-off of the deaths of children under one year of age in a downward trend in infant mortality rates over the given year, expressed per 1 000 live births. past few years. For deaths beyond a month (post Neonatal mortality refers to the death of neonatal mortality), there tends to be a greater range children under 28 days. of causes – the most common being SIDS (Sudden Some of the international variation in infant Infant Death Syndrome), birth defects, infections and and neonatal mortality rates may be due to accidents. Advances in neonatal care for very preterm variations among countries in registering and growth-restricted babies are also associated with practices of premature infants. Most countries a higher proportion of infant deaths occurring after have no gestational age or weight limits for the neonatal period (EURO-PERISTAT, 2008). mortality registration. Minimal limits exist for All European countries have achieved remarkable Norway (to be counted as a death following a live progress in reducing infant mortality rates from the birth, the gestational age must exceed 12 weeks) levels of 1970, when the average was 25 deaths per and in the Czech Republic, France, Malta (the 1 000 live births (Figure 1.8.1). This equates to a cumu- National Mortality Register), the Netherlands lative reduction of over 80% since 1970. Portugal has and Poland a minimum gestational age of seen its infant mortality rate reduced by over 7% per 22 weeks and/or a weight threshold of 500 grams year on average since 1970, going from the country is applied. Lithuania has a gestational age limit. with the highest rate in Europe to an infant mortality 40 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.8. INFANT MORTALITY 1.8.1. Infant mortality rates, 2008 and decline 1970-2008 2008 (or nearest year available) Decline 1970-2008 (or nearest year available) 1.8 Luxembourg 6.7 2.1 Slovenia 6.3 2.5 Iceland 4.3 2.5 Sweden 3.8 2.6 Finland 4.2 2.7 Greece 6.1 2.7 Norway 4.0 2.8 Czech Republic 5.1 3.1 Ireland 4.7 3.3 Portugal 7.2 3.4 Belgium 4.7 3.5 Germany 4.8 3.5 Spain 5.3 3.7 Austria 5.0 3.7 Italy 5.3 3.8 France 4.0 3.8 Netherlands 3.1 4.0 Denmark 3.3 4.0 Switzerland 3.4 4.6 EU 4.3 4.7 United Kingdom 3.5 4.9 Lithuania 3.5 5.0 Estonia 3.3 5.3 Cyprus 4.1 5.6 Hungary 4.8 5.6 Poland 4.8 5.9 Slovak Republic 3.8 6.7 Latvia 2.5 8.6 Bulgaria 3.0 9.9 Malta 2.7 11.0 Romania 3.9 17.0 Turkey 5.5 20 15 10 5 0 0 2 3 6 8 Deaths per 1 000 live births Average annual rate of decline (%) Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932335818 1.8.2. Infant mortality rates, selected European countries, 1970-2008 Portugal Slovenia Sweden EU Deaths per 1 000 live births 60 50 40 30 20 10 0 1970 1975 1980 1985 1990 1995 2000 2005 Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932335837 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 41 1.9. INFANT HEALTH: LOW BIRTH WEIGHT Low birth weight – defined here as newborns birth weight babies in these countries is now above the weighing less than 2 500 grams – is an important European average (Figure 1.9.2). Low birth weight indicator of infant health because of the close rela- proportions in Hungary, Poland and Luxembourg have tionship between birth weight and infant morbidity declined over the same time period. Little change and mortality. There are two categories of low birth occurred in Nordic countries including Iceland, Finland, weight babies: those occurring as a result of restricted Sweden and Denmark, although a rise was observed in foetal growth and those resulting from pre-term birth. Norway. Low birth weight infants have a greater risk of poor Figure 1.9.3 shows some correlation between the health or death, require a longer period of hospitalisa- percentage of low birth weight infants and infant tion after birth, and are more likely to develop signifi- mortality rates. In general, countries reporting a low cant disabilities (UNICEF and WHO, 2004). proportion of low birth weight infants also report Risk factors for low birth weight include adoles- relatively low infant mortality rates. This is the case cent motherhood, having a previous history of low for instance for the Nordic countries. Greece, however, weight births, harmful behaviours such as smoking, is an exception, reporting a high proportion of low excessive alcohol consumption and poor nutrition, a birth weight infants but one of the lowest infant low Body Mass Index, a background of low parental mortality rates. socio-economic status, and having in-vitro fertilisa- Agreed-on norms for low birth weight do not tion treatment. exist (EURO-PERISTAT, 2008). Physiological variations One-in-16 babies born in Europe in 2008 – or 6.4% in size occur among different countries and popula- of all births – weighed less than 2 500 grams at birth. A tion groups, and these need to be taken into account north-south gradient is evident in Europe for low birth when interpreting differences. Some populations may weight, in that the Nordic countries – Iceland, Sweden have lower than average birth weights than others and Finland – along with Latvia reported the smallest because of genetic differences. Comparisons of differ- proportions of low weight births, with less than 4.5% ent population groups within countries show that the of live births defined as low birth weight. Countries proportion of low birth weight infants is also be from Southern Europe including Greece, Spain and influenced by differences in education, income and Portugal, as well as Turkey, Romania, Bulgaria and associated living conditions. Hungary, are at the other end of the scale with rates of low birth weight infants above 7.5%. The proportion of low birth weight among European countries varies by a factor of more than two (Figure 1.9.1). Definition and deviations Since 1980 the prevalence of low birth weight infants has increased in a number of European coun- Low birth weight is defined by the World Health tries, most notably in Spain, Portugal, Malta and the Organization (WHO) as the weight of an infant Netherlands (Figure 1.9.1). There are several reasons at birth of less than 2 500 grams (5.5 pounds), for this rise. First, the number of multiple births, with irrespective of the gestational age of the infant. the increased risks of pre-term births and low birth This is based on epidemiological observations weight, has risen steadily, partly as a result of the rise regarding the increased risk of death to the infant in fertility treatments. Other factors which may have and serves for international comparative health influenced the rise in low birth weight are older age at statistics. The number of low weight births is childbearing and increases in the use of delivery then expressed as a percentage of total live births. management techniques such as induction of labour The majority of the data comes from birth and caesarean delivery. registers. A small number of countries supply Spain and Portugal have seen great increases in the data for selected regions or from surveys. past three decades. As a result, the proportion of low 42 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.9. INFANT HEALTH: LOW BIRTH WEIGHT 1.9.1. Low birth weight infants, 2008 and change 1980-2008 2008 (or nearest year available) Change 1980-2008 (or nearest year available) 3.8 Iceland 12 4.1 Finland 5 4.1 Sweden -2 4.3 Latvia n.a. 4.6 Estonia n.a. 4.6 Lithuania n.a. 4.6 Luxembourg -27 5.1 Norway 34 5.3 Ireland 33 5.7 Poland -25 5.8 Denmark 0 6.2 Netherlands 55 6.3 Slovenia 9 6.3 Switzerland 24 6.4 EU 15 6.7 Italy 20 6.8 Malta 62 6.8 France 31 6.8 Germany 24 7.1 Austria 25 7.1 United Kingdom 6 7.3 Czech Republic 24 7.3 Slovak Republic 24 7.6 Belgium 36 7.6 Spain 171 7.7 Portugal 67 7.9 Romania 4 8.3 Bulgaria 36 8.3 Hungary -20 8.4 Greece 42 11.0 Turkey n.a. 15 10 5 0 -50 0 50 100 150 200 Percentage of newborns weighing less than 2 500 g % change over period Source: OECD Health Data 2010; WHO HFA-DB. 1 2 http://dx.doi.org/10.1787/888932335856 1.9.2. Trends in low birth weight infants, 1.9.3. Low birth weight and infant mortality, 2008 selected European countries, 1980-2008 (or nearest year available) Finland Hungary Infant mortality (deaths per 1 000 live births) 20 Spain EU Percentage of newborns weighing less than 2 500 g R 2 = 0.32 12 TUR 15 10 ROU 8 10 MLT BGR 6 LVA SVK EST POL HUN 5 GBR LTU 4 FIN ISL GRC LUX CZE SVN 2 0 1980 1985 1990 1995 2000 2005 0 4 8 12 Low birth weight (%) Source: OECD Health Data 2010; WHO HFA-DB. Source: OECD Health Data 2010; WHO HFA-DB. 1 2 http://dx.doi.org/10.1787/888932335875 1 2 http://dx.doi.org/10.1787/888932335894 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 43 1.10. SELF-REPORTED HEALTH AND DISABILITY Most European countries conduct regular health illnesses or health problems more often than men (an surveys which allow respondents to report on differ- average of 32% versus 27% across EU countries), with ent aspects of their health. A commonly-asked the gender divide greatest in Latvia, Norway and the question relates to self-perceived health status, of the Slovak Republic. Reporting increased with age, from type: “How is your health in general?”. Despite the an average of 10% of young people aged 15-24 years, subjective nature of this question, indicators of to 70% of older persons aged 85 years or more. There perceived general health have been found to be a good is a moderate negative association between adults predictor of people’s future health care use and reporting good/very good health, and reporting a long- mortality (for instance, see Miilunpalo et al., 1997). For standing illness or health problem (R2 = –0.38). the purpose of international comparisons however, When adults were asked whether they had been cross-country differences in perceived health status limited in their usual daily activities because of a are difficult to interpret because responses may be health problem – which is one definition of disability – affected by social and cultural factors. 24% answered that they had, with 8% of respondents Keeping these limitations in mind, a majority “strongly limited” and 15% “limited to some extent” of the adult population in almost all European (Figure 1.10.2). Adults most commonly reported countries rate their health to be good or very good activity limitation in the Slovak Republic, Germany, (Figure 1.10.1). In Switzerland, Ireland, Iceland and the Latvia, Estonia and Portugal (30% or more of respon- United Kingdom, more than eight out of ten people dents), and less so in Malta, Iceland and Switzerland report good or very good health. Across the European (less than 15%). Severe activity limitation was more Union, two-thirds (67%) of all adults rated their health prevalent in Portugal, the Slovak Republic, Austria and as good or better, with Germany, Finland and France Germany (10% or more of respondents), and less so in close to this average. Adults in central and eastern Malta, Bulgaria and Switzerland (less than 5%). Adults European countries, along with Portugal, report the with activity limitations were also less likely to report lowest rates of good or very good health. In Latvia, good or very good health (R2 = 0.60). Lithuania and Portugal, less than half of all adults consider themselves to be in good health. These differences, however, do not necessarily mean that the general health of people in Switzerland or Ireland is objectively better than that of citizens in Latvia or Definition and deviations Portugal (Baert and de Norre, 2009). Self-reported health reflects people’s overall In all European countries, men are more likely perceptions of their own health, including both than women to rate their health as good or better, p hy s i c a l a n d p s y ch o l o g i c a l d i m e n s i o n s . with the largest differences in Portugal, Bulgaria and Typically ascertained through health interview the Slovak Republic. Unsurprisingly, people’s rating of surveys, respondents are asked a number of their own health tends to decline with age. In many questions on their health and functioning. The countries there is a particularly marked decline in a three questions used in the EU-SILC survey, and positive rating of one’s own health after age 45 and a some other national surveys are: i) “How is your further decline after age 65. People who are unem- health in general? Is it very good, good, fair, bad, ployed, retired or inactive more often report bad or very bad”; ii) “Do you have any longstanding very bad health (Baert and de Norre, 2009). People with illness or health problem which has lasted, or is a lower level of education or income do not rate their expected to last for six months or more?”; and health as positively as people with higher levels iii) “For at least the past six months, have you (Mackenbach et al., 2008). been hampered because of a health problem in activities people usually do? Yes, strongly Another common health interview survey ques- limited/Yes, limited/No, not limited”. tion asks whether respondents had any long-standing illnesses or health problems. Three-in-ten adults in Persons in institutions are not surveyed. EU countries reported having illnesses or health Caution is required in making cross-country problems (Figure 1.10.1). Adults in Finland, Slovenia, comparisons of perceived general health, since Hungary and Estonia were more likely to report people’s assessment of their health is subjective having illnesses or health problems, while these and can be affected by their social and cultural conditions were less commonly reported in Romania, backgrounds. Greece and Italy. Women reported long-standing 44 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.10. SELF-REPORTED HEALTH AND DISABILITY 1.10.1. Adults’ self-reported health status, 2008 Good or very good health Long-standing illness or health problem 87 Switzerland 27 84 Ireland 24 81 Iceland 26 80 United Kingdom 33 79 Sweden 33 77 Netherlands 31 77 Cyprus 26 77 Norway 32 76 Greece 22 74 Denmark 25 74 Luxembourg 24 74 Malta 25 74 Belgium 25 73 Spain 30 70 Austria 33 69 Romania 19 69 France 37 69 Finland 41 67 Turkey n.a. 67 EU 30 65 Germany 36 64 Italy 23 63 Bulgaria 24 62 Czech Republic 28 60 Slovak Republic 30 59 Slovenia 39 58 Poland 31 55 Hungary 38 55 Estonia 38 49 Portugal 33 48 Lithuania 29 45 Latvia 34 100 90 80 70 60 50 40 30 10 20 30 40 50 % of population aged 15 and over % of population aged 15 and over Source: EU-Statistics on Income and Living Conditions survey; OECD Health Data 2010; Swiss Federal Statistics Office. 1 2 http://dx.doi.org/10.1787/888932335913 1.10.2. Adults reporting a limitation in usual activities, 2008 Limited strongly Limited to some extent % of population aged 15 and over 40 30 20 10 0 ta i t z nd ic No d Bu ay Sw ria ic Re n y Cy n us EU Hu l y Sl Ge i a i t e Ir e a nd m m e Po g Re d Be n m ce k a t h ni a s Au y Fi ia Po d Es al a nd i ni xe e c ar ni ar ai n e h lan an ur bl bl an g It a tv r al iu do rw an ed ak rma pr S w ela la la a st N e ove Sp ua to r tu nm ng bo L u Gr e pu pu la M nl La lg lg m er ng Fr th Ic er Ro De Sl Ki Li d ec ov Cz Un Source: EU-Statistics on Income and Living Conditions survey; Swiss Federal Statistics Office. 1 2 http://dx.doi.org/10.1787/888932335932 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 45 1.11. INCIDENCE OF SELECTED COMMUNICABLE DISEASES Communicable diseases such as measles, pertus- Two-thirds of all pertussis cases in 2008 occurred sis and hepatitis B still pose a major threat to the among children aged under 15 years of age, although health of European citizens. Measles, a highly infec- the disease may be under-diagnosed in adolescents tious disease of the respiratory system, is caused by a and adults. The highest incidence occurred among virus. Symptoms include fever, cough, runny nose, red infants aged less than one year, many of whom are too eyes and a characteristic rash. It can lead to severe young to be vaccinated, and children aged 10-14 years, health complications, including pneumonia, enceph- who may have not had a full course of vaccination, or alitis, diarrhoea and blindness. Pertussis (or whooping who may have lost their immunity. Vaccination status cough) is also highly infectious, and is caused by the was known in only half of all reported cases, but of bacterium Bordetella pertussis. The disease derives its these 21% were unvaccinated (EUVAC.NET, 2010). name from the sound made from the intake of air Around 6 000 hepatitis B cases were reported after a cough. Hepatitis B is an infection of the liver annually in EU countries during 2006-08. The highest caused by the hepatitis B virus. The virus is trans- incidence rates occurred among six countries: mitted by contact with blood or body fluids of an Iceland (13.2 notified cases per 100 000 population), infected person. A small proportion of infections Bulgaria (9.9), Turkey (9.1), Austria (8.1), Latvia (7.3) and become chronic, and these people are at high risk of Romania (5.1) (Figure 1.11.3). The notification rate has death from cancer or cirrhosis of the liver. Protection declined in EU countries since 1991-93, when it was against each of these diseases is available through 8.3 cases per 100 000 population to 2.5 for 2006-08. vaccination (see Indicator 3.3). Hepatitis B infection is more common in the southern An average of over 5 000 measles cases were parts of Eastern and Central Europe, and low in preva- reported annually in European Union countries lence in most of Western Europe. Around twice as during 2006-08, with the highest number of cases many cases of hepatitis B occurred among males than occurring in four countries: Germany, Romania, the females in 2008, with the majority reported in the age United Kingdom and Italy. The highest crude group 25-44 years, followed by 15-24 year-olds. The incidence during 2006-08 was in Switzerland, with disease is increasingly seen as a sexually transmitted 15 cases reported per 100 000 persons (Figure 1.11.1). disease, although the disease pattern and risk groups A number of other western European countries, differ widely across Europe (ECDC, 2009). including the United Kingdom, Romania, France and Italy, also had high incidences. Across the European Union, average incidence for 2006-08 was 1.2 cases per 100 000 population. This represents a marked decline from the average rate in 1991-93, which was 27 cases per 100 000 population. In 2008, more than Definition and deviations half of all cases (53%) occurred among children and young people aged 5-19 years. Hospitalisation was National mandatory notification systems for necessary for 15% of cases. Among cases whose vacci- communicable diseases, including measles, nation status was known, the vast majority (91%) were pertussis and hepatitis B, exist in most European unvaccinated (EUVAC.NET, 2009). countries, although case definitions, laboratory confirmation requirements and reporting systems Almost 13 000 pertussis cases were reported may differ. annually among EU countries, with an overall incidence of six per 100 000 population (Figure 1.11.2). In 2008, measles notification was voluntary The highest incidences were reported in Norway in Belgium, although mandatory in schools. (113 cases per 100 000 population), Switzerland (45), Pertussis notification was mandatory only in the Netherlands (41), Estonia (26) and Slovenia (24). parts of Belgium and Germany, and Switzerland Most cases were reported from the Netherlands, and France had sentinel surveillance systems. Norway, Switzerland and Poland, which together Hepatitis B notification was voluntary in France contributed three-quarters (76%) of all cases reported and Belgium, Italy had a sentinel surveillance in 2008. Pertussis incidence has halved since 1991-93, system, and reporting was not mandatory in when the average rate among EU countries was Switzerland. 11.3 notified cases per 100 000 population. 46 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.11. INCIDENCE OF SELECTED COMMUNICABLE DISEASES 1.11.1. Incidence of measles, 2006-08 1.11.2. Incidence of pertussis, 2006-08 Switzerland 14.9 Norway 112.9 United Kingdom 7.2 Switzerland 44.7 Romania 5.5 Netherlands 41.1 France 4.0 Estonia 26.0 Italy 3.6 Slovenia 24.0 Austria 2.0 Finland 9.6 Greece 1.6 Denmark 8.0 Germany 1.5 Sweden 7.1 Ireland 1.5 EU 5.7 EU 1.2 Estonia 0.7 Czech Republic 4.7 Spain 0.7 Poland 4.6 Luxembourg 0.6 Bulgaria 3.5 Belgium 0.5 United Kingdom 2.0 Malta 0.3 Ireland 1.9 Denmark 0.3 Belgium 1.8 Netherlands 0.2 Austria 1.6 Poland 0.2 Spain 1.1 Norway 0.2 Italy 1.1 Sweden 0.2 Latvia 1.0 Latvia 0.1 Slovak Republic 0.9 Cyprus 0.04 Cyprus 0.8 Czech Republic 0.03 Lithuania 0.7 Finland 0.03 Iceland 0.6 Turkey 0.03 Luxembourg 0.4 Lithuania 0.02 Bulgaria 0.01 Hungary 0.3 Hungary 0.0 Malta 0.3 Iceland 0.0 Greece 0.2 Portugal 0.0 Portugal 0.2 Slovak Republic 0.0 Romania 0.2 Slovenia 0.0 Turkey 0.1 0 5 10 15 20 0 25 50 75 100 125 Per 100 000 population Per 100 000 population Source: OECD Health Data 2010; WHO Europe (2010). Source: OECD Health Data 2010; WHO Europe (2010). 1 2 http://dx.doi.org/10.1787/888932335951 1 2 http://dx.doi.org/10.1787/888932335970 1.11.3. Incidence of hepatitis B, 2006-08 Per 100 000 population 15 13.2 12 9.9 9.1 9 8.1 7.3 6 5.1 3.8 3.5 3.0 2.8 2.7 3 2.6 2.5 2.2 2.1 1.8 1.7 1.5 1.4 1.3 1.2 1.2 1.1 1.0 0.8 0.7 0.7 0.5 0.4 0.4 0.0 0 Ir e i c Bu nd Tu a ey ria Ro i a Be ia Li blic m Re ni a m a No r g ay EU Re a in nd er y S w ds C n us Fi e Ge and Po y Sl nd Hu ni a Gr y d Po t a De gal Fr k ce an i xe ni c al ar ar e an bl an ar tv iu al u rk rw ee an n ed S w y pr a la la st Ne It e c E s to e L u t hu a Sp r tu nm ng bo rm pu pu la el l nl M La lg lg ov m er Au Ic it z th ak h ov Cz Sl Source: OECD Health Data 2010; WHO Europe (2010). 1 2 http://dx.doi.org/10.1787/888932335989 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 47 1.12. HIV/AIDS The first cases of Acquired Immunodeficiency in tho se cou ntries wit h high A IDS incid ence Syndrome (AIDS) were diagnosed almost 30 years ago. rates – Estonia, Latvia, Portugal and Spain, along The onset of AIDS is normally caused as a result of with Switzerland – all at over 300 persons per HIV (human immunodeficiency virus) infection and 100 000 population. Over 25 000 new cases of HIV were can manifest itself as any number of different dis- diagnosed in the European Union in 2008. The eases, such as pneumonia and tuberculosis, as the predominant mode of transmission of HIV is sex immune system is no longer able to defend the body. between men, followed by heterosexual contact. There is a time lag between HIV infection, AIDS However, among ea stern European countries, diagnosis and death due to HIV infection, which can injecting drug use is still the most common mode be any number of years depending on the treatment (ECDC and WHO Europe, 2009). Approximately 75% of administered. Despite worldwide research, there is no heterosexually acquired HIV infection in Western and cure currently available. Central Europe is among migrants. In 2008, the number of newly reported cases of In recent years, the overall decline in AIDS cases AIDS was approximately 5 300 across the European has slowed down. This reversal has been accompanied Union as a whole, representing an unweighted aver- by evidence of increasing transmission of HIV in age incidence rate of 12.7 per million population several European countries, attributed to complacency (Figure 1.12.1). Following the first reporting of AIDS in regarding the effectiveness of treatment and a waning the early 1980s, the number of cases rose rapidly to of public awareness regarding drug use and sexual reach an average of more than 37 new cases per practice. Since 2000, the rate of newly diagnosed cases million population across EU countries at its peak in of HIV has more than doubled to 89 per million popula- the middle of the 1990s, almost three times current tion in 2008 (ECDC and WHO Europe, 2009). Further incidence rates (Figure 1.12.2). Public awareness inroads in AIDS incidence rates will require more campaigns contributed to steady declines in reported intensive evidence-based HIV prevention programmes cases through the second half of the 1990s. In that are focused and adapted to reach those most at addition, the development and greater availability of risk of HIV infection (UNAIDS, 2008). antiretroviral drugs, which reduce or slow down the development of the disease, led to a sharp decrease in incidence from 1996. The highest AIDS incidence rates among Definition and deviations EU countries in 2008 were reported in Estonia, Latvia, Portugal and Spain, at 25-50 new cases per million The incidence rate of acquired immunodefi- population. Spain had the highest incidence rates in ciency syndrome (AIDS) is the number of new the first decade following the outbreak, although cases per million population at year of diagnosis. there was a sharp decline from 1994 onwards. Note that data for recent years are provisional Incidence rates in Portugal peaked somewhat later, due to reporting delays, which sometimes can be towards the end of the 1990s. AIDS incidence rates in for several years depending on the country. Estonia have increased rapidly since the mid-2000s. Estimates of the number of persons living with Central European countries such as the Czech and human Immunodeficiency virus (HIV) are calcu- Slovak Republics, Hungary and Poland report the lated by the Joint United Nations Programme on lowest incidence rates of AIDS among EU countries. HIV/AIDS (UNAIDS, 2008), and are based on In the European Union, approximately national research studies. These estimates 730 000 persons were living with HIV infection in 2007 include all people (adults and children) with HIV (Figure 1.12.1). Italy, Spain and France had the greatest infection in 2007, whether or not they have number of persons, followed by the United Kingdom developed symptoms of AIDS. and Germany. HIV prevalence estimates were highest 48 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.12. HIV/AIDS 1.12.1. AIDS incidence rates in 2008, and estimated number of persons living with HIV in 2007 AIDS incidence Persons living with HIV 45.5 Estonia 9 900 43.6 Latvia 10 000 36.4 Portugal 34 000 29.1 Spain 140 000 19.5 Malta < 500 17.2 Italy 150 000 16.3 Lithuania 2 200 14.7 Switzerland 25 000 12.7 EU // 730 000 12.6 Netherlands 18 000 12.4 Luxembourg n.a. 10.4 Romania 15 000 10.0 United Kingdom 77 000 9.8 France 140 000 9.4 Belgium 15 000 8.5 Sweden 6 200 7.9 Greece 11 000 7.8 Austria 9 800 7.6 Cyprus n.a. 6.4 Ireland 5 500 6.3 Iceland < 500 5.9 Denmark 4 800 5.5 Finland 2 400 5.5 Slovenia < 500 3.8 Bulgaria n.a. 3.8 Norway 3 000 3.0 Germany 53 000 3.0 Poland 20 000 2.8 Czech Republic 1 500 2.3 Hungary 3 300 0.7 Turkey < 2 000 0.2 Slovak Republic < 500 50 40 30 20 10 0 0 50 000 100 000 150 000 200 000 New cases per million population Number of persons Source: OECD Health Data 2010; ECDC and WHO Europe (2009); UNAIDS (2008). 1 2 http://dx.doi.org/10.1787/888932336008 1.12.2. Trends in AIDS incidence rates, selected EU countries, 1981-2008 France Portugal Spain EU New cases per million population 200 150 100 50 0 1981 1986 1991 1996 2001 2006 Source: OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932336027 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 49 1.13. CANCER INCIDENCE Around 2.4 million new cases of cancer (excluding countries such as Belgium, France, the Netherlands non-melanoma skin cancers) were diagnosed in and Ireland, at over 90 cases per 100 000 population EU countries in 2008 (Ferlay et al., 2010), with 55% (Figure 1.13.3). Rates in Central, Eastern and Southern occurring among males and 45% among females. The Europe were lower, with Turkey, Greece, Romania, most common forms of the disease were prostate, Lithuania, Latvia and Poland all reporting less than colorectal, breast and lung cancer. The risk of getting 50 new cases per 100 000 population. There has been cancer before the age of 75 years is 26.5%, or around an increase in measured incidence rates of breast one in four. However, because the population of Europe cancer over the past decade, although death rates is ageing, the rate of new cases of cancer is also have declined or remained stable. Survival rates have expected to increase (European Commission, 2008b). also increased, due to earlier diagnosis and/or better Large regional inequalities exist in cancer treatment (see Indicator 3.13). incidence across the European Union. In 2008, the Prostate cancer is the most common form among incidence rate for all cancers combined was highest men in the European Union, particularly for those i n N o r th er n a n d Wes t e r n E u ro p e – D e n m a r k , aged over 65 years of age, comprising one quarter Ireland, Belgium, France, Norway and Iceland – at over (25%) of all new diagnoses in 2008. Rates were highest 290 per 100 000 population, but was lower in some in Ireland, France, Belgium and northern European Mediterranean countries such as Turkey, Greece, Cyprus countries (Norway, Sweden, Iceland and Finland). and Malta, at less than 220. Rates in Italy were above the Rates were lower in a range of central, eastern and EU average of 255 new cases per 100 000 population. southern European countries, including Turkey, Rates in central and eastern European countries varied, Greece, Romania and Bulgaria. At least part of the being highest in the Czech Republic and Hungary five-fold difference between countries with the (around 290), similar to the EU average in Slovenia and highest and lowest incidence rates is due to under- the Slovak Republic (260), and below average in registration of prostate cancer in some countries, as Romania, Bulgaria, Poland and other countries. well as the use of sensitive diagnostic tests for early Cancer incidence rates are higher for men than detection in others (Ferlay et al., 2007). for women in all EU countries (Figure 1.13.1). Here too there is great variation between countries; in Spain and Turkey, male incidence rates are 60% higher than Definition and deviations female rates, whereas in Denmark and Cyprus they are less than 10% higher. The average all cancer Cancer incidence rates are based on numbers incidence rate among EU member states was 298 per of new cases of cancer registered in a country in 100 000 males and 226 per 100 000 females. a year divided by the size of the corresponding population. The rates have been directly In 2008, lung cancer was one of the most age-standardised to the WHO World standard common cancers in EU countries, being responsible population to remove variations arising from for around 12% of all new cancer diagnoses, 16% for differences in age structures across countries males and 7% for females. Ten of the fifteen countries and over time. The source is GLOBOCAN 2008, at with male rates higher than the EU average were http://globocan.iarc.fr. located in Central and Eastern Europe (Figure 1.13.2). Rates in Hungary, Poland and Slovenia were higher Cancer registration is well established in a than 60 per 100 000 population. Male lung cancer majority of European Union member states, incidence rates in Northern Europe (Sweden, Iceland, although the quality and completeness of cancer Finland, Norway) and some southern European registry data may vary. In some countries, cancer countries (Cyprus, Portugal, Malta) were less than registries only cover subnational areas. The 40 per 100 000 population. Among females, lung international comparability of cancer incidence cancer incidence was especially high in Denmark, but data can also be affected by differences in also Hungary, Iceland and the Netherlands, at over 25. medical training and practice. Thirty per cent of all new cancer cases among The incidence of all cancers is classified to women diagnosed in 2008 were cancers of the breast ICD-10 codes C00-C97, lung cancer to C33-C34, – the most common form of cancer among women. breast cancer to C50 and prostate cancer to C61. Incidence rates were high in western European 50 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.13. CANCER INCIDENCE 1.13.1. All cancers incidence rates, 1.13.2. Lung cancer incidence rates, males and females, 2008 males and females, 2008 Males Females Males Females Turkey 113 Sweden 18 19 182 Cyprus 175 Cyprus 5 187 22 Greece 142 Portugal 6 188 29 Malta 200 Iceland 28 233 29 Romania 180 Finland 11 241 30 Bulgaria 204 Malta 6 246 33 Portugal 191 Austria 16 267 36 Finland 230 Norway 25 283 37 Poland 186 Ireland 24 285 38 Estonia 204 Switzerland 17 285 38 Sweden 241 United Kingdom 24 287 42 Austria 219 Germany 16 294 42 EU 226 Denmark 36 298 43 United Kingdom 250 Italy 11 298 45 Latvia 193 Netherlands 26 304 46 Spain 187 Luxembourg 18 310 46 Italy 252 EU 15 310 47 Switzerland 236 France 15 313 48 Iceland 273 Turkey 5 315 49 Slovenia 232 Slovak Republic 11 316 49 Lithuania 208 Estonia 9 317 51 Netherlands 271 Bulgaria 8 318 51 Slovak Republic 223 Czech Republic 16 320 53 Luxembourg 254 Greece 9 324 53 246 Spain 8 Germany 331 53 313 Romania 10 Denmark 337 55 266 Latvia 7 Norway 341 55 253 Lithuania 7 Czech Republic 344 56 285 Belgium 16 Ireland 356 58 276 Slovenia 17 Belgium 357 64 239 Poland 18 Hungary 358 71 255 Hungary 28 France 361 79 100 200 300 400 0 20 40 60 80 100 Age-standardised rates per 100 000 population Age-standardised rates per 100 000 population Source: OECD Health Data 2010; Ferlay et al. (2010). Source: OECD Health Data 2010; Ferlay et al. (2010). 1 2 http://dx.doi.org/10.1787/888932336046 1 2 http://dx.doi.org/10.1787/888932336065 1.13.3. Breast cancer incidence rates, females, 2008 1.13.4. Prostate cancer incidence rates, males, 2008 Turkey 28 Turkey 15 Greece 45 Greece 18 Romania 45 Romania 20 Lithuania 46 Bulgaria 23 Latvia 48 Hungary 33 Poland 49 Slovak Republic 40 Estonia 50 Estonia 43 Slovak Republic 53 Poland 44 Bulgaria 57 Cyprus 47 Hungary 58 Portugal 50 Portugal 60 Malta 51 Spain 61 Spain 57 Slovenia 66 Italy 58 Cyprus 68 Slovenia 62 Czech Republic 68 Czech Republic 63 Austria 70 EU 65 EU 71 Latvia 66 Malta 72 Lithuania 67 Norway 76 Denmark 73 Germany 82 Netherlands 73 Luxembourg 82 Luxembourg 75 Sweden 83 United Kingdom 80 Iceland 86 Germany 83 Italy 86 Austria 83 Finland 87 Switzerland 91 United Kingdom 88 Finland 97 Denmark 89 Iceland 102 Switzerland 89 Belgium 102 Ireland 94 Sweden 114 Netherlands 97 Norway 116 France 100 France 118 Belgium 109 Ireland 126 0 25 50 75 100 125 150 0 25 50 75 100 125 150 Age-standardised rates per 100 000 females Age-standardised rates per 100 000 males Source: OECD Health Data 2010; Ferlay et al. (2010). Source: OECD Health Data 2010; Ferlay et al. (2010). 1 2 http://dx.doi.org/10.1787/888932336084 1 2 http://dx.doi.org/10.1787/888932336103 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 51 1.14. DIABETES PREVALENCE AND INCIDENCE Diabetes is a chronic metabolic disease, charac- Italy, Bulgaria and Greece have less than ten new terised by high levels of glucose in the blood. It occurs cases per 100 000 population. Alarmingly, there is either because the pancreas stops producing the evidence that type 1 diabetes is developing at an ear- hormone insulin (type 1 diabetes), or through a lier age among children. combination of the pancreas having reduced ability to The economic impact of diabetes is substantial. produce insulin alongside the body being resistant to Health expenditure to treat and prevent diabetes and its action (type 2 diabetes). People with diabetes are at its complications is estimated to total USD 93 billion, a greater risk of developing cardiovascular diseases or approximately 10% of total health expenditure in such as heart attack and stroke if the disease is left EU countries in 2010 (IDF, 2009). Around one-quarter undiagnosed or poorly controlled. They also have of medical expenditure is spent on controlling elevated risks for sight loss, foot and leg amputation elevated blood glucose, another quarter on treating due to damage to the nerves and blood vessels, and long-term complication of diabetes, and the remain- renal failure requiring dialysis or transplantation. der on additional general medical care. Increasing Diabetes was the principal cause of death of costs reinforce the need to provide quality care for the more than 100 000 persons in EU countries in 2008, management of diabetes and its complications. and is the fourth or fifth leading cause of death in Type 2 diabetes is largely preventable. A number most developed countries. However, only a minority of risk factors, such as overweight and obesity and of persons with diabetes die from diseases uniquely physical inactivity are modifiable, and can also help related to the condition – in addition, about 50% of reduce the complications that are associated with persons with diabetes die of cardiovascular disease, diabetes. But in most countries, the prevalence of and 10-20% of renal failure (IDF, 2006). overweight and obesity also continues to increase (see Diabetes is increasing rapidly in every part of the Indicator 2.8). world, to the extent that it has now assumed epidemic proportions. Estimates suggest that more than 6% of the population aged 20-79 years in EU countries, or 33 million people, have diabetes in 2010. Almost half Definition and deviations of diabetic adults are aged less than 60 years. If left unchecked, the number of people with diabetes in EU The sources and methods used by the Inter- countries will reach more than 37 million in less than national Diabetes Federation for publishing 20 years (IDF, 2006). national prevalence estimates of diabetes are outlined in their Diabetes Atlas, 4th edition (IDF, Less than 5% of adults aged 20-79 years in 2009). Country data were derived from studies Iceland, Norway and the United Kingdom have published between 1980 and February 2009, and diabetes, according to the International Diabetes were only included if they met several criteria Federation. This contrasts with Cyprus, Germany and for reliability. Turkey, where 8% or more of the population of the same age have the disease (Figure 1.14.1). Among EU Studies from several European countries countries, abnormal glucose tolerance shows little – France, Italy, Netherlands, Norway, Slovenia association with affluence, and there was little and the United Kingdom – only provided evidence of an urban/rural divide in prevalence, self-reported data on diabetes. To account for except in a few countries (IDF, 2009). undiagnosed diabetes, the prevalence of diabetes for the United Kingdom was multiplied Type 1 diabetes accounts for only 10-15% of all by a factor of 1.5, in accordance with local rec- diabetes cases. It is the predominant form of the ommendations, and doubled for other countries, disease in younger age groups in most developed based on data from a number of countries. countries. Based on disease registers and recent studies, the annual number of new cases of type 1 Prevalence rates were adjusted to the World diabetes in children aged under 15 years is high at 25 Standard Population to facilitate cross-national or more per 100 000 population in Nordic countries comparisons. (Finland, Sweden and Norway) (Figure 1.14.2). Turkey, 52 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.14. DIABETES PREVALENCE AND INCIDENCE 1.14.1. Prevalence estimates of diabetes, adults aged 20-79 years, 2010 % 10 9.1 8.9 8.0 8 7.6 7.7 7.7 6.7 6.8 6.5 6.5 6.6 6.4 5.9 6.0 6 5.6 5.7 5.2 5.2 5.3 4 3.6 3.6 2 1.6 0 d ay ic m en nd s k d ly ce EU ia n ce nd ia l y y us ta ga nd ke an ar an ai an bl en It a ar do rw ee an ed pr la al la Sp r tu nm r pu rm la el nl lg ov Ir e Tu Po M ng Cy Gr Fr Sw No Ic er Fi Bu Po Re De Ge Sl Ki th ak Ne d i te ov Un Sl Note: The data are age-standardised to the World Standard Population. Source: IDF (2009). 1 2 http://dx.doi.org/10.1787/888932336122 1.14.2. Incidence estimates of type 1 diabetes, children aged 0-14 years, 2010 Cases per 100 000 population 60 57.4 50 41.0 40 30 27.9 24.5 22.2 20 18.0 18.0 18.8 15.6 16.3 14.7 14.9 12.9 13.0 13.2 13.6 11.1 12.2 9.4 9.9 10 8.4 3.2 0 ly ia ia y ce ce nd n l ic d us ta nd EU y s k m ay en d ga ke nd an ar ai an an en bl It a ar al do ee an rw ed pr la la Sp r tu nm r rm pu la el M nl lg ov Tu Po Ir e ng Cy Gr Fr Sw No Ic er Fi Bu Po Re De Ge Sl Ki th ak Ne d i te ov Un Sl Source: IDF (2009). 1 2 http://dx.doi.org/10.1787/888932336141 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 53 1.15. DEMENTIA PREVALENCE Dementia describes a variety of brain disorders One important cost driver is the increasing demand which progressively damage and destroy brain cells. for institutional and residential care. Affecting mainly people over the age of 60 years, The prevention and treatment of dementia has dementia results in the deterioration of mental ability been recognised as a major public health priority. The characterised by impairments in memory and cogni- European Commission has supported several projects tion. It is one of the most important causes of disability to investigate and enhance the evidence base in the elderly. The most common cause of dementia in surrounding Alzheimer’s disease and other forms the European Union is Alzheimer’s disease (around of dementia. These have included the EuroCoDe 50-70%), followed by successive strokes that lead to (European Collaboration on Dementia) project multi-infarct dementia (around 30%). There is no cure from 2006-08 which was co-ordinated by Alzheimer for dementia, but drugs exist to alleviate and temporar- Europe. The most recent initiative supports national ily delay the symptoms. efforts in the key areas of prevention, research In 2006, the number of people with dementia in coordination and best practice in treatment and care the European Union was estimated to be 7.3 million, (European Commission, 2010b). and because of their longer life expectancy, almost At a national level, various countries including 68% (4.9 million) of those affected were women. The France, Norway and the United Kingdom have put in highest prevalence rates were found in Sweden, Italy, place national plans and strategies to meet the future Switzerland and Germany, where between 1.7-1.8% of challenges posed by dementia. These plans include the population suffer from dementia. This contrasts measures to improve early diagnosis, treatment and with less than 1.0% of the population in Turkey, the the quality of care for people affected by dementia, as Slovak Republic and Ireland (Figure 1.15.1). The aver- well as providing support to carers of those afflicted age dementia prevalence rate for the 27 EU countries with dementia. was 1.4%. Age is the main risk factor for the development of dementia. Although early-onset dementia can occur before the age of 65 years, prevalence rates increase Definition and deviations steadily after that age, to reach 15% of males and 16% of females at age 80-84 years, and one-third of males Dementia prevalence rates are based on (32.4%) and half of females (48.8%) at age 95 years and estimates of the total number of persons living over (Figure 1.15.2). There is a strong positive relation- with dementia divided by the size of the corre- ship between the prevalence of dementia and the sponding population. The estimates of dementia proportion of the population aged 65 years and over in prevalence were derived by the EuroCoDe European countries (Figure 1.15.3). (European Collaboration on Dementia) project, co-ordinated by Alzheimer Europe from 2006-08. The population aged 65 years and older in the European Union is predicted to double between 1995 In order to estimate prevalence rates for and 2050, to reach 135 million. With the increased dementia across European countries, the ageing of the population, the absolute number of EuroCoDe project undertook a systematic review people with dementia will also continue to rise, of papers reporting dementia prevalence esti- placing greater demand on national health and social mates. Papers were screened according to criteria systems. Dementia places a large burden not only on which stipulated that studies be community- sufferers, but also on their carers. Patterns of care vary based, use standardised diagnostic criteria, have across EU countries, with different mixes of informal a minimum sample size of 300 and a participa- care by families and friends, and formal care either in tion rate of over 50%, and be conducted in 1990 or institutions or at home. thereafter. In addition, it was necessary that raw prevalence data be made available. Thirty-one The cost of illness associated with Alzheimer’s studies met this criteria and raw data was disease and other dementias in Europe was esti- extracted from 17 studies for use in the collabora- mated at EUR 177 billion in 2006, which is divided tive analysis. fairly evenly between costs directly attributable to the diseases responsible for dementia, and estimates Given the divergence in scale and accuracy of of costs associated with informal care. The total the sources used across countries, the preva- costs of dementia are expected to increase to over lence estimates should be used with caution. EUR 250 billion by 2030 (Alzheimer Europe, 2009). 54 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 1.15. DEMENTIA PREVALENCE 1.15.1. Prevalence of dementia, population aged 30 years and over, 2006 % 2.0 1.8 1.7 1.7 1.7 1.6 1.6 1.6 1.5 1.5 1.6 1.5 1.5 1.5 1.5 1.4 1.3 1.4 1.3 1.3 1.3 1.3 1.2 1.3 1.2 1.2 1.2 1.1 1.1 1.0 1.0 1.1 1.0 1.0 0.8 0.4 0.4 0 Ir e i c Re ke y Po d Cy d Ro r u s Ic i a m c d Li alt a R ni a S l ur g t h ni a Bu ds La a Hu v i a Es r y a Gr U Po e c e nm l Fi rk Au d No r ia K i ay B e om m Fr n Ge nc e it z any nd Sw ly en De ga xe b l i i ni ai n n an an E bl an ar It a iu a a d rw n ed la la la st t N e ove to a Sp r tu d ng p bo r e rm pu L u epu a la M el nl lg lg m e c t hu Tu er ng er Sw ak h i te ov Cz Un Sl Source: Alzheimer Europe (2009). 1 2 http://dx.doi.org/10.1787/888932336160 1.15.2. Age- and sex-specific prevalence 1.15.3. Population aged 65 years and over, of dementia in EU countries, 2006 and prevalence of dementia, 2006 Male Female Prevalence of dementia (%) 1.8 SWE Prevalence (%) R 2 = 0.89 ITA ESP CHE 60 GBR FRA DEU 1.6 NOR BEL FIN AUT DNK GRC 50 1.4 BGR PRT LVA EST HUN 1.2 LTU SVN 40 ISL CZE NLD MLT LUX CYP POL 1.0 30 SVK ROU IRL 0.8 20 0.6 10 0.4 TUR 0 0.2 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95 and 5 10 15 20 25 30 over Age group Population aged 65 years and over (%) Source: Alzheimer Europe (2009). Source: Alzheimer Europe (2009); OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336179 1 2 http://dx.doi.org/10.1787/888932336198 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 55 Health at a Glance: Europe 2010 © OECD 2010 Chapter 2 Determinants of Health 2.1. Smoking and alcohol consumption among children . . . . . . . . . . . . . . 58 2.2. Nutrition among children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 2.3. Physical activity among children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 2.4. Overweight and obesity among children . . . . . . . . . . . . . . . . . . . . . . . . 64 2.5. Supply of fruit and vegetables for consumption . . . . . . . . . . . . . . . . . . 66 2.6. Tobacco consumption among adults. . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 2.7. Alcohol consumption among adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 2.8. Overweight and obesity among adults . . . . . . . . . . . . . . . . . . . . . . . . . . 72 57 2.1. SMOKING AND ALCOHOL CONSUMPTION AMONG CHILDREN Regular smoking or excessive drinking in adoles- Figure 2.1.3. Countries above the 45 degree line have cence has both immediate and long-term health higher rates for girls, and countries below the line consequences. Children who establish smoking habits higher rates for boys. Countries with higher rates of in early adolescence increase their risk of cardio- smoking among boys also report higher rates for girls, vascular diseases, respiratory illnesses and cancer. with the same finding for drinking rates. They are also more likely to experiment with alcohol Rates of drunkenness are also available for and other drugs. Alcohol misuse is itself associated 13-year-olds (Currie et al., 2008). At this age, over one with a range of social, physical and mental health in ten children in a range of countries including problems, including depressive and anxiety disorders, Estonia, the United Kingdom, Lithuania, Latvia, obesity and accidental injury (Currie et al., 2008). Bulgaria, the Czech and Slovak Republics and Finland Results from the Health Behaviour in School- have experienced drunkenness at least twice. In aged Children (HBSC) surveys, a series of collaborative Romania, Denmark and Slovenia, high rates of cross-national studies conducted in most EU coun- repeated drunkenness at 13 are also seen for boys. tries, allow for monitoring of smoking and drinking Some of the largest increases in reported drunken- behaviours among adolescents. Generally, girls smoke ness between the ages of 13 and 15 are seen in more than boys, but more boys get drunk. Between Denmark, Finland and Lithuania. 13 and 15 years of age, the prevalence of smoking and Risk-taking behaviours among adolescents are drunkenness doubles in many EU countries. falling, with regular smoking for both boys and girls Boys and girls in central and eastern European and drunkenness rates for boys showing some decline countries (Bulgaria, Latvia, Estonia, Lithuania, the from the levels of the late 1990s (Figure 2.1.4). Levels Czech Republic, Hungary) as well as in Austria, of smoking for both sexes are at their lowest for a Finland and Italy smoke most often, with weekly rates decade with, on average, fewer than one in five around 20% or more (Figure 2.1.1). In contrast, 15% or children of either sex smoking regularly. However, less of 15-year-olds in Nordic countries (Denmark, increasing rates of smoking and drunkenness among Iceland, Norway and Sweden), Switzerland and adolescents in Baltic and other eastern European Portugal smoke weekly. Many countries report higher countries are cause for concern. rates of smoking for girls, although only Bulgaria, Austria and Spain have differences in excess of 5%. Smoking is more prevalent among boys in central and eastern European countries. Drunkenness at least twice in their lifetime is Definition and deviations reported by 40% or more of 15-year-olds in Denmark, Lithuania, the United Kingdom, Finland, Bulgaria and Estimates for smoking refer to the proportion Estonia (Figure 2.1.2). Across all surveyed countries, of 15-year-old children who self-report smoking 30% of girls and 38% of boys have been drunk on at least once a week. Estimates for drunkenness two or more occasions, with much lower rates in record the proportions of 15-year-old children Mediterranean countries such as Malta, Greece, saying they have been drunk twice or more in Portugal and Italy, as well as in Switzerland and their lives. France. Boys are more likely to report repeated drunk- Data for 26 European Union and 3 non-EU enness. Romania, Slovenia, Poland and Estonia have countries are from the Health Behaviour in the biggest differences, with rates of alcohol abuse School-aged Children (HBSC) surveys under- among boys being in excess of 15% higher than those taken between 1992-93 and 2005-06. Data are of girls. Norway, Spain and the United Kingdom are drawn from school-based samples. France, the only countries where more girls report repeated Germany and the United Kingdom report results drunkenness, and in each case rates are around 5% for certain regions only. Turkey is included in higher. the 2005-06 HBSC survey, but did not question The differences in recent smoking and drinking children on drinking and smoking. rates between 15-year-old boys and girls are shown in 58 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.1. SMOKING AND ALCOHOL CONSUMPTION AMONG CHILDREN 2.1.1. Smoking among 15-year-olds, 2005-06 2.1.2. Drunkenness among 15-year-olds, 2005-06 Smoking at least once a week Drunk at least twice in life Boys Girls Boys Girls Bulgaria 36 Denmark 56 28 59 Austria 30 Lithuania 50 24 57 Malta 24 United Kingdom 50 19 44 Czech Republic 23 Finland 44 20 47 Latvia 23 Bulgaria 42 30 51 Germany 22 Estonia 42 17 57 Finland 21 Latvia 39 23 50 France 21 Austria 36 17 41 Hungary 21 Spain 33 22 29 Luxembourg 21 Hungary 32 17 40 Netherlands 21 Iceland 32 16 31 Ireland 20 Norway 32 19 25 Italy 20 Ireland 31 20 36 Spain 20 Slovak Republic 31 14 39 Estonia 19 Czech Republic 30 27 36 EU 19 EU 30 19 38 Lithuania 18 Germany 28 26 31 United Kingdom 18 Poland 27 13 42 Belgium 17 Slovenia 27 16 43 Greece 16 Sweden 26 17 26 Slovenia 16 Belgium 22 20 32 Denmark 15 21 15 Netherlands 30 Slovak Republic 15 20 18 Luxembourg 27 Switzerland 15 19 15 Romania 45 Poland 14 18 19 France 29 Iceland 13 18 14 Italy 22 Norway 12 18 9 Portugal 25 Portugal 12 18 9 Switzerland 29 Romania 12 Greece 17 20 21 Sweden 9 Malta 15 8 18 0 10 20 30 40 0 20 40 60 80 % % Source: Currie et al. (2008). Source: Currie et al. (2008). 1 2 http://dx.doi.org/10.1787/888932336217 1 2 http://dx.doi.org/10.1787/888932336236 2.1.3. Risk behaviours of 15-year-olds by sex, 2.1.4. Trends in repeated drunkenness and regular 2005-06 smoking among 15-year-olds, EU average Regular cigarette smoking Repeated drunkenness Boys Girls Girls aged 15 (%) % Repeated drunkenness 65 45 43 41 39 35 35 DEN 35 33 55 31 GBR LTU 24 25 45 FIN BGR EST AUT SVK LVA 15 BGR 1993-94 1997-98 2001-02 2005-06 35 ISL ESP IRL NOR HUN % Regular smoking AUT DEU 45 CZE MLT POL SVN SWE 25 DEU CZE NLD BEL 35 ESP NLD GBR CHE ROU CHE FRA 27 26 15 DNK POL PRT 25 22 NOR ISL ROU 24 25 19 SWE 18 19 5 15 5 15 25 35 45 55 65 1993-94 1997-98 2001-02 2005-06 Boys aged 15 (%) Source: Currie et al. (2008). Source: Currie et al. (2000, 2004, 2008); WHO (1996). 1 2 http://dx.doi.org/10.1787/888932336255 1 2 http://dx.doi.org/10.1787/888932336274 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 59 2.2. NUTRITION AMONG CHILDREN Nutrition is important for children’s develop- The gap between the fruit consumption of boys ment and long-term health. Eating fruit during and girls is largest at age 15, for most countries. At adolescence, for example, in place of high-fat, sugar age 11, boys and girls in Lithuania as well as Italy, and salt products, can protect against health prob- France and Estonia have similar rates of fruit lems such as obesity, diabetes, and heart problems. consumption. Poland, Germany and the Netherlands Moreover, eating fruit when young can be habit form- have the biggest gaps at this age. As children reach ing, promoting healthy eating behaviours for later life. age 15, gaps in Finland, Austria and Latvia grow to a A number of factors influence the amount of fruit level where fewer than six boys for every ten girls eat consumed by adolescents, including family income, fruit regularly. the cost of alternatives, preparation time, whether Average reported rates of daily fruit consumption parents eat fruit, and the availability of fresh fruit across EU countries showed some increase which can be linked to the country or local climate between 2001-02 and 2005-06. This was most evident (Rasmussen et al., 2006). Fruit (and vegetable) among girls aged 11 (Figure 2.2.3). consumption have a high priority as indicators of Effective strategies are required in order to healthy eating in most EU countries. ensure that children are eating enough fruit to In 2005-06, only around one-third of boys and conform with recommended dietary guidelines. two-fifths of girls aged 11-15 years ate at least one Children generally hold a positive attitude toward piece of fruit daily, according to the latest Health fruit intake, and report good availability of fruit at Behaviour in School-aged Children (HBSC) survey home, but lower availability at school and during (Currie et al., 2008). Overall, boys in Italy, and girls in leisure time. Increased accessibility to fruit, combined the United Kingdom had the highest rates of daily with educational and motivational activities, can help fruit consumption. Fruit consumption was relatively in increasing fruit consumption (Sandvik et al., 2005). low among Baltic and some Nordic countries, includ- ing Latvia, Lithuania and Finland, with rates around one in four girls and one in five boys. At all ages and in most countries, girls were more likely to eat fruit daily. At age 11, girls in Norway, Portugal and Slovenia, as Definition and deviations well as boys in Portugal, Italy and Bulgaria were more likely to eat fruit daily. By age 15, girls in Italy, Nutrition is measured here in terms of the Denmark and the United Kingdom, and boys in Italy, proportions of children who report eating fruit at Portugal and Malta ate most (Figure 2.2.1). least every day or more than once a day. In addition to fruit, healthy nutrition also involves In almost all EU countries, daily fruit consump- other types of foods. tion falls between ages 11 and 15 (Figure 2.2.2). Among girls, the EU average fell from 44% at age 11, to Data for 26 European Union and four non-EU 39% at age 13 and 34% at age 15. For boys, the fall was countries are from the Health Behaviour in from 37% to 32% and then 25%. In Bulgaria and School-aged Children (HBSC) surveys under- Iceland, rates fell by up to half between ages 11 taken in 2001-02 and 2005-06. Data are drawn and 15, and severe falls were also seen in Austria from school-based samples. France, Germany (boys). Italy and Belgium are among the most success- and the United Kingdom report results for ful countries in maintaining healthy eating habits as certain regions only. children get older. 60 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.2. NUTRITION AMONG CHILDREN 2.2.1. Daily fruit eating among 11- and 15-year-olds, 2005-06 Age 15 Age 11 Girls Boys 41 32 57 Norway 43 40 36 56 Portugal 48 40 26 55 Slovenia 41 42 28 54 Switzerland 42 27 25 51 Bulgaria 44 46 29 51 Denmark 40 28 18 51 Iceland 39 40 30 51 Romania 42 35 23 50 Germany 36 44 33 50 United Kingdom 42 41 27 49 Luxembourg 38 32 18 48 Austria 38 41 25 48 Czech Republic 38 39 29 48 Ireland 38 47 37 48 Italy 44 43 34 47 Belgium 38 29 23 46 Hungary 38 34 35 46 Malta 42 30 20 45 Netherlands 33 34 24 45 Poland 31 42 26 45 Turkey 34 34 22 44 Sweden 37 34 25 44 EU 37 27 24 43 Spain 39 24 24 40 Greece 34 29 24 38 France 35 30 18 36 Estonia 33 33 22 35 Slovak Republic 30 26 15 28 Latvia 21 23 15 28 Lithuania 27 28 14 27 Finland 24 60 50 40 30 20 10 10 20 30 40 50 60 % % Source: Currie et al. (2008). 1 2 http://dx.doi.org/10.1787/888932336293 2.2.2. Regular fruit consumption 2.2.3. Average proportion of children at ages 11 and 15 years, in EU countries reporting daily fruit consumption, 2005-06 2001-02 and 2005-06 Girls Boys Age 11 Age 13 Age 15 Children aged 15 (%) % 60 60 Boys Girls 50 50 PRT ITA DNK BEL GBR CHE NOR TUR 40 SVN 40 MLT ITA PRT SVK BEL DEU GBR NOR AUT 30 FIN EST NLD 30 LVA TUR HUN ISL POL BGR BGR ESP LTU GRC NLD 20 SVK SWE 20 AUT ISL FIN EST LVA LTU 10 10 10 20 30 40 50 60 2001-02 2005-06 2001-02 2005-06 Children aged 11 (%) Source: Currie et al. (2008). Source: Currie et al. (2004, 2008). 1 2 http://dx.doi.org/10.1787/888932336312 1 2 http://dx.doi.org/10.1787/888932336331 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 61 2.3. PHYSICAL ACTIVITY AMONG CHILDREN Undertaking physical activity in adolescence is than half of those at age 11, and in Iceland, Romania, beneficial for health, and can set standards for adult Ireland and Finland, rates of physical activity among physical activity levels, thereby influencing health girls fall by over 60%. outcomes in later life. Research supports the role that To compare levels of exercise between 2001-02 physical activity in adolescence has in the prevention and 2005-06 for 15-year-old children, results are and treatment of a range of youth health issues reported in relation to the EU average (Figure 2.3.3). including asthma, mental health, bone health and In 2001-02, rates refer to 15-year-olds reporting an obesity. More direct links to adult health are found hour of moderate to vigorous exercise five days a between physical activity in adolescence and its effect week, but in 2005-06 figures refer to exercise of this on overweight and obesity and related diseases, type seven days a week. Boys’ rates were above the breast cancer rates and bone health in later life. The EU average in the Netherlands, the United Kingdom, health effects of adolescent physical activity are Greece, Spain and Switzerland in 2001-02, but fell sometimes dependent on the activity type, e.g. water below the average in 2005-06. Latvia, Belgium and physical activities in adolescence are effective in the Denmark are countries where rates of physical treatment of asthma, and exercise is recommended in activity were below the EU average in 2001-02, but the treatment of cystic fibrosis (Hallal et al., 2006; were among the higher performers in 2005-06. For Currie et al., 2008). 15-year-old boys, only the Czech Republic, Ireland and Some of the factors influencing the levels of Poland have been consistently high performers on physical activity undertaken by adolescents include measures of physical activity in both waves. For girls, the availability of space and equipment, the child’s Latvia, Belgium and Malta have moved from below present health conditions, their school curricula and average performances in 2001-02 to above average other competing pastimes. in 2005-06. In Sweden, Poland, the United Kingdom, Only one in five children in EU countries undertake Switzerland and Slovenia, rates of physical activity moderate-to-vigorous exercise regularly, according to among 15-year-old girls have fallen below the EU results from the 2005-06 HBSC survey. Children in average since 2001-02. Switzerland, Luxembourg and Italy are least likely to exercise regularly, whereas the Slovak Republic and Ireland stand out as strong performers with over 40 and 30% respectively of children aged 11 to 15 exercising for Definition and deviations a total of at least 60 minutes per day over the past week (Figure 2.3.1). The country rankings reported vary Data for physical activity considers the according to the child’s age. France appears at the lower regularity of moderate-to-vigorous physical activ- end, especially for girls, at both ages. Boys consistently ity as reported by 11- and 15-year-olds for the undertake more physical activity than girls, across all years 2001-02 and 2005-06. Moderate-to-vigorous countries and all age groups. physical activity refers to exercise undertaken for at least an hour which increases both heart rate It is of concern that physical activity tends to and respiration (and leaves the child out of breath fall between ages 11 to 15 for most EU countries sometimes) on five or more days per week (Figure 2.3.2), with boys in the Czech Republic and in 2001-02, and seven days a week in 2005-06. Luxembourg the only exceptions. In Portugal, Norway, Sweden, Austria, and Finland, the rates of exercising Indicators are taken from the Health Behaviour among boys more than halve between ages 11 and 15. in School-aged Children Survey (HBSC). Data are The rates of girls exercising to recommended levels drawn from school-based samples, but some also falls between the ages of 11 and 15 years. In countries report regional results only. many countries, rates for 15-year-old girls are less 62 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.3. PHYSICAL ACTIVITY AMONG CHILDREN 2.3.1. Children aged 11 and 15 years 2.3.2. Comparing physical activity doing moderate-to-vigorous physical activity daily of 11- and 15-year-old children by sex, in the past week, 2005-06 2005-06 15 years 11 years Boys Girls Slovak Republic 37.5 47.0 Czech Republic Ireland 20.0 44.5 Luxembourg Finland 12.0 42.5 Slovak Republic Bulgaria 20.0 32.5 Poland Denmark 18.0 28.5 Latvia Spain 15.5 28.0 Belgium Latvia 21.0 26.5 Slovenia Austria 11.5 26.0 Estonia Iceland 12.5 26.0 Lithuania EU 15.7 25.5 Malta Netherlands 16.5 25.0 Italy Turkey 14.0 25.0 Switzerland Hungary 15.0 23.5 Hungary Lithuania 16.0 23.5 United Kingdom Slovenia 14.0 23.0 EU Estonia 13.5 22.5 Denmark Germany 13.0 22.5 Germany Malta 16.0 22.5 Greece Romania 11.0 22.5 Bulgaria United Kingdom 13.5 22.5 Netherlands Belgium 18.3 22.3 Spain Czech Republic 21.5 22.0 France Norway 10.0 22.0 Iceland Poland 15.5 21.5 Romania Sweden 10.5 21.5 Turkey Portugal 10.0 21.0 Ireland Greece 11.5 20.5 Portugal France 9.5 18.0 Norway Italy 11.5 18.0 Sweden Luxembourg 15.0 15.5 Austria Switzerland 11.5 15.0 Finland 0 10 20 30 40 50 0 20 40 60 80 100 120 % 11 years-old = 100 Source: Currie et al. (2008). Source: Currie et al. (2008). 1 2 http://dx.doi.org/10.1787/888932336350 1 2 http://dx.doi.org/10.1787/888932336369 2.3.3. Standardised rates of physical activity for 15-year-old children (EU average = 1), 2001-02 and 2005-06 2005-06 2001-02 Boys Girls Slovak Republic Czech Republic Ireland Latvia Bulgaria Poland Belgium Denmark Hungary Lithuania Luxembourg Malta Slovenia Spain Estonia Netherlands United Kingdom Germany Greece Iceland Italy Romania Turkey Finland Portugal France Austria Norway Switzerland Sweden 2.5 2.0 1.5 1.0 0.5 0 0 0.5 1.0 1.5 2.0 2.5 EU average = 1.0 EU average = 1.0 Source: Currie et al. (2004, 2008). 1 2 http://dx.doi.org/10.1787/888932336388 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 63 2.4. OVERWEIGHT AND OBESITY AMONG CHILDREN Children who are overweight or obese are at obesity increased between 2001-02 and 2005-06 from greater risk of poor health in adolescence and also in 12% to 16% for 15-year-old boys, and from 7% to 9% for adulthood. Being overweight in childhood increases girls. Between 2001-02 and 2005-06, every surveyed the risk of developing cardiovascular disease or country reported an increase in overweight or obesity diabetes, as well as related social and mental health for boys aged 15. The largest increases during the four problems. Excess weight problems in childhood are year period were found in Austria, Lithuania and associated with an increased risk of being an obese Poland. A similar pattern of increases is seen for girls, adult, at which point certain forms of cancer, osteoar- with rates in Portugal and Germany almost doubling. thritis, a reduced quality of life and premature death Only Ireland, Norway and the United Kingdom report can be added to the list of health concerns (OECD, reductions in the proportion of overweight or obese 2010c; Currie et al., 2008). girls at age 15 between 2001-02 and 2005-06. However, Evidence suggests that even if excess childhood because non-response rates to questions of self- weight is lost, adults who were obese children retain reported height and weight were high in these an increased risk of cardiovascular problems. And countries, cautious interpretation is required. although dieting can combat obesity, children who Childhood is an important period for forming diet are at a greater risk of putting on weight following healthy behaviours. Schools provide an opportunity to periods of dieting. Eating disorders, symptoms of ensure that children understand the importance of stress and postponed physical development can also good nutrition and physical activity, and can benefit be products of dieting. from both. Studies show that locally focussed actions Across most EU countries, one in seven children and interventions, targeting 0-12 year-olds can be are overweight or obese (Figure 2.4.1). Aggregate effective in changing behaviours. figures for 2005-06 show that nearly one in five children in southern European countries (Malta, Greece, Portugal, Italy and Spain), are overweight or obese. Fewer than one in ten children in selected eastern European countries (Lithuania, Latvia, the Definition and deviations Slovak Republic and Estonia) as well as in the Estimates of overweight and obesity are based Netherlands, Switzerland and Denmark are on Body Mass Index (BMI) calculations using overweight or obese. child self-reported height and weight. Over- There is no clear association between weight weight and obese children are those whose BMI problems and weight reduction behaviours at the is above a set of age- and sex-specific cut-off national level. In most countries, the number of points (Cole et al., 2000). Data on weight reduc- children who say that they are trying to lose weight tion record children who report being on a diet is greater than the number with excess weight or doing something else to lose weight. problems. Generally, countries where few children Indicators are taken from the Health Behaviour report excess weight problems also report weight in School-aged Children Surveys in 2001-02 reduction behaviours close to the EU average. Many of and 2005-06. Aggregate country estimates are the countries with the highest rates of overweight and crude rates of overweight and obese 11-, 13- and obese children have similar levels of weight reduction 15-year-olds in each country. Some countries behaviour, each around the EU average of 13%. report regional data only. Data are drawn from There are important differences among children school-based samples. with excess weight problems, according to their age. Self-reported height and weight is subject to In some countries older children have more excess under-reporting and error, and requires cautious weight than younger children, for other countries the interpretation. In the 2005-06 survey, England and opposite is true. A number, including the Netherlands, Norway have missing data for over 30% of Norway, Sweden, Iceland and Switzerland, report respondents for 11-year-olds. The same is true for increases in overweight and obesity rates for both England, Ireland and Belgium for 13-years-olds, boys and girls as children get older. and in England and Ireland for 15-year-olds. Rates of overweight and obesity among boys and In 2001-02, BMI data are missing for over 30% of girls are increasing across the European Union respondents in Ireland. (Chart 2.4.2). Average reported rates of overweight and 64 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.4. OVERWEIGHT AND OBESITY AMONG CHILDREN 2.4.1. Children aged 11-15 years who are overweight or obese, and children who are involved in weight-reduction behaviour, 2005-06 % Children who are overweight or obese Children involved in weight-reduction behaviour 30 25 20 15 10 5 13.3 29.5 13.2 18.3 14.2 10.5 10.5 10.5 18.8 18.8 15.3 14.5 12.5 15.5 15.8 10.0 10.0 15.0 12.0 12.0 12.0 16.7 11.2 11.6 8.2 8.8 9.8 8.0 9.7 7.5 7.8 0 ta Po c e l ly F i in Ic i c i t z lic Hu d C z Sl r y Re i a Ir e d nd B EU m a U n G e ur g K i any Au m Ro r i a Po i a Fr d Sw ce Be en No m Tu y ey th nd Sl D e ni a Re ar k s ia a ga a nd xe ar i ni an an n bl en an It a tv a al iu do a rk rw ee an S w pub ed la N e er l a la st to Sp r tu ua nm ng bo i te rm pu M la nl el La L u ul g lg e c ov m ng Gr Es th er Li h ak d ov Source: Currie et al. (2004, 2008). 1 2 http://dx.doi.org/10.1787/888932336407 2.4.2. Change in obesity rates between 2001-02 and 2005-06, for 15-year-old boys and girls 2005-06 2001-02 Boys Girls Greece Italy Iceland Portugal Slovenia Austria Finland Spain Bulgaria Hungary Germany Luxembourg Norway EU Ireland Sweden Czech Republic France Switzerland Turkey Belgium Denmark United Kingdom Poland Estonia Slovak Republic Netherlands Romania Latvia Lithuania 25 20 15 10 5 0 0 5 10 15 20 25 % % Source: Currie et al. (2004, 2008). 1 2 http://dx.doi.org/10.1787/888932336426 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 65 2.5. SUPPLY OF FRUIT AND VEGETABLES FOR CONSUMPTION Nutrition is an important determinant of health. little in Greece, but levels per capita vegetable remain Inadequate consumption of fruit and vegetables is the highest. In contrast, the supply of vegetables in one factor that can play a role in increased morbidity Bulgaria has declined recently to one of the lowest and premature death. A recent European Commission levels among member states. White Paper advocated increasing the consumption Many factors play a role in ensuring an adequate of fruit and vegetables as one of a number of tools supply of fruit and vegetables. In recent years, the to offset a worsening trend of poor diets and low harvested production of the main types of fruit and physical activity. Proper nutrition assists in preventing vegetables in the European Union has remained a number of obesity-related chronic conditions, relatively stable, although there was growth in including cardiovascular disease, hypertension, type 2 imports from non-EU countries (Martinez-Palou and diabetes, stroke, certain cancers, musculoskeletal Rohner-Thielen, 2008). The majority of suppliers disorders and a range of mental health conditions growing fresh vegetables are located in Romania, (European Commission, 2007). Poland and Lithuania. Most citrus farms are located in Estimates of the supply of fruit and vegetables Mediterranean countries (Spain, Greece, Italy), with available for consumption in different countries are Poland and Romania also large fruit-producing calculated by the Food and Agriculture Organization of countries. The price of fruit and vegetables varies the United Nations (FAO). In 2007, levels of the supply of considerably among member states. In 2006, for both fruit and vegetables differed substantially across example, it was almost half the EU average in Bulgaria European countries (Figure 2.5.1). The per capita fruit and a number of other eastern European countries supply in a number of central and eastern European adjusted by purchasing power parity, but more countries, including Poland, Bulgaria, Romania, Latvia, t h a n 20 % h i gh e r i n I re la n d , L u x em b o u rg a n d the Slovak Republic, the Czech Republic and Estonia, Nord ic countries (Denmark, Finland, Sweden) was below 80 kg per person, contrasting with an (Martinez-Palou and Rohner-Thielen, 2008). EU average of 105. Fruit supply was greater in Western and Southern Europe, with estimates for Luxembourg and Greece above 160 kg per person, more than twice Definition and deviations the amount of those countries reporting the lowest supply. Estimates of food available for consumption are The per capita availability of vegetables was based on annual production and trade of food highest in Mediterranean countries, including Greece, commodities figures as supplied by national Turkey, Malta, Portugal, Spain, Italy and Cyprus, all at Ministries of Agriculture and Trade to the FAO 150 kg per person or more. Supply was lower than the (Food and Agriculture Organization of the EU average in Nordic countries, as well as in some United Nations). Gross apparent consumption = central and eastern European countries (Bulgaria, (Commercial production + estimated own the Czech Republic, Slovenia). The spread between account production for self-consumption + countries with the lowest and highest per capita imports + opening stocks) – (exports + usage for supply of vegetables is three-fold. processed food + feed + non-food usage + wastage + closing stocks). The supply of fruit and vegetables for consump- tion has increased across the European Union in the Per person consumption is derived from three decades since 1980 (Figures 2.5.2 and 2.5.3). dividing the total amount of fruit and vegetables Fruit supply increased from an average of 88 kg per available for human consumption by the total capita in 1980 to 105 in 2007. Greece and Poland both population actually partaking of food supplies increased per capita fruit supply, although large abso- during the reference period. Per person figures lute differences remain. Fruit supply in Ireland represent the average supply available for the increased rapidly from the late 1990s. The supply of population as a whole, and do not necessarily vegetables increased more modestly, from an average indicate what is actually consumed by indivi- of 102 to 116 kg per capita across all EU countries duals. Actual food consumption may be lower during the years 1980 to 2007. Vegetable supply than the quantity shown, depending on wastage increased substantially in Finland, although the and losses of food in the household. amount remains relatively low. Supply has changed 66 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.5. SUPPLY OF FRUIT AND VEGETABLES FOR CONSUMPTION 2.5.1. Supply of fruit and vegetables for consumption, 2007 Fruit Vegetables 50 Poland 130 58 Bulgaria 75 59 Romania 151 61 Latvia 107 65 Slovak Republic 90 71 Czech Republic 75 78 Estonia 96 80 Switzerland 91 85 Belgium 118 85 Hungary 110 88 Germany 94 88 Spain 156 91 Lithuania 96 94 Finland 79 98 Malta 221 105 EU 116 111 Turkey 224 112 Denmark 98 117 France 98 117 Sweden 88 118 Portugal 170 119 Cyprus 150 121 Slovenia 77 127 United Kingdom 92 136 Netherlands 103 141 Ireland 79 142 Norway 78 144 Italy 152 148 Iceland 76 156 Austria 96 164 Greece 241 190 Luxembourg 87 250 200 150 100 50 0 0 50 100 150 200 250 Kg/capita Kg/capita Source: FAOSTAT Database; OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932336445 2.5.2. Trends in supply of fruit, 2.5.3. Trends in supply of vegetables, selected EU countries, 1980-2007 selected EU countries, 1980-2007 Greece Ireland Bulgaria Greece Poland EU Finland EU Kg/capita/year Kg/capita/year 200 400 150 300 100 200 50 100 0 0 1980 1985 1990 1995 2000 2005 1980 1985 1990 1995 2000 2005 Source: FAOSTAT Database; OECD Health Data 2010. Source: FAOSTAT Database; OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932336464 1 2 http://dx.doi.org/10.1787/888932336483 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 67 2.6. TOBACCO CONSUMPTION AMONG ADULTS Tobacco is directly responsible for about one in ten at an even faster pace than male rates. However, in adult deaths worldwide, equating to about 6 million seven countries, female smoking rates have been deaths each year (Shafey et al., 2009). It is a major risk increasing since the mid-1990s (Lithuania, Portugal, factor for at least two of the leading causes of premature Greece, Bulgaria, France, Germany and Austria), but mortality – circulatory diseases and a range of cancers. even in these countries women are still less likely to In addition, it is an important contributory factor for smoke than men. In 2008, the gender gap in smoking respiratory diseases, while smoking among pregnant rates was particularly large in Baltic countries (Latvia, women can lead to low birth weight and illnesses Lithuania and Estonia), as well as in Turkey and among infants. It remains the largest avoidable risk to Romania (Figure 2.6.1). health in EU countries. Several studies provide strong evidence of socio- The proportion of daily smokers among the adult economic differences in smoking and mortality population varies greatly across countries, even (Mackenbach et al., 2008). People in lower social groups between neighboring countries (Figure 2.6.1). In 2008, have a greater prevalence and intensity of smoking, rates were lowest in Sweden, Iceland, Slovenia and a higher all-cause mortality rate and lower rates of Portugal, all at less than 20% of the adult population cancer survival (Woods et al., 2006). The influence of smoking daily. On average, smoking rates have smoking as a determinant of overall health inequa- decreased by about 5 percentage points in EU coun- lities is such that, in a non-smoking population, tries since 1995, with a bigger decline in men than in mortality differences between social groups would be women. Large declines occurred in Turkey (47% to halved (Jha et al., 2006). 27%), Luxembourg (33% to 20%), Norway (33% to 21%) Figure 2.6.3 shows the correlation between tobacco and Denmark (36% to 23%). Greece maintains the consumption (as measured by grams per capita) and highest level of smoking (40%), along with Bulgaria incidence of lung cancer across EU countries for which and Ireland, with close to 30% or more of the adult data are available, with a time lag of two decades. Higher population smoking daily. tobacco consumption at the national level is also gener- In the post-war period, most EU countries tended ally associated with higher mortality rates from lung to follow a general pattern – very high smoking rates cancer one or two decades later across EU countries. among men (50% or more) through to the 1960s and 1970s, while the 1980s and the 1990s were charac- terised by a marked downturn in tobacco consumption. Much of this decline can be attributed to policies aimed Definition and deviations at reducing tobacco consumption through public awareness campaigns, advertising bans and increased The proportion of daily smokers is defined as taxation (World Bank, 1999). In addition to government the percentage of the population aged 15 years policies, actions by anti-smoking interest groups were and over reporting smoking every day. very effective in reducing smoking rates by changing International comparability is limited due to beliefs about the health effects of smoking. the lack of standardisation in the measurement Although large disparities remain, this pattern of of smoking habits in health interview surveys a decline in smoking rates is found across most across EU countries. Variations remain in the EU countries (Figure 2.6.2). Smoking prevalence among age groups surveyed, wording of questions, men continues to be higher than among women in all response categories and survey methodologies, EU countries except Sweden. Female smoking rates e.g. in a number of countries, respondents are continue to decline in most countries, and in a number asked if they smoke regularly, rather than daily. of cases (Turkey, Iceland, Belgium, Latvia and Ireland) 68 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.6. TOBACCO CONSUMPTION AMONG ADULTS 2.6.1. Daily smoking rates, 2008 (or nearest year available) Males Females All adults 14.5 Sweden 17 12 15 17.8 Iceland 20 16 18.9 Slovenia 22 11 19.6 Portugal 29 16 20.0 Belgium 24 18 20.0 Luxembourg 23 9 20.2 Romania 32 18 20.4 Finland 24 18 20.4 Switzerland 23 16 20.4 Malta 26 22 21.0 Norway 21 21 22.0 United Kingdom 22 16 22.4 Italy 29 22 23.0 Denmark 24 19 23.2 Austria 27 19 23.2 Germany 28 19 24.2 EU 30 19 24.3 Czech Republic 30 14 24.7 Cyprus 36 23 25.0 Slovak Republic 26 21 25.9 Hungary 31 22 26.2 France 31 17 26.2 Estonia 39 19 26.3 Poland 34 22 26.4 Spain 32 15 26.5 Lithuania 43 12 27.4 Turkey 44 13 27.9 Latvia 46 25 28.0 Netherlands 32 27 29.0 Ireland 31 19 29.1 Bulgaria 40 34 39.7 Greece 46 50 40 30 20 10 0 0 10 20 30 40 50 % of population aged 15 years and over % of population aged 15 years and over Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336502 2.6.2. Change in smoking rates by gender, 2.6.3. Tobacco consumption, 1990 1995-2008 (or nearest year available) and incidence of lung cancer, 2008 Males Females Lung cancer (incidence per 100 000 population) 50 -57 HUN Turkey -35 -43 Iceland -24 R 2 = 0.22 -33 Belgium -27 -33 Denmark -37 -31 Norway -36 -31 Luxembourg -43 40 -29 Sweden -44 DNK -29 Latvia -9 -28 Switzerland -30 -26 BEL Estonia -26 NLD -22 United Kingdom -24 GBR -21 Hungary -29 IRL CZE Netherlands -19 FRA -22 30 NOR -13 EU -19 GRC -12 ISL DEU Spain -27 -10 Finland -18 CHE -6 Poland -23 TUR AUT Italy -6 -16 Czech Republic -4 -10 Ireland -4 7 Austria 3 20 -9 FIN Germany 6 SWE -11 France -15 6 Bulgaria 5 13 Greece 1 20 47 Portugal -12 // 142 Lithuania 1 10 -75 -50 -25 0 25 50 1 000 1 500 2 000 2 500 3 000 3 500 % change over period Tobacco consumption (grams per capita) Source: OECD Health Data 2010; Eurostat Statistics Database; WHO Source: OECD Health Data 2010. Global Infobase. 1 2 http://dx.doi.org/10.1787/888932336521 1 2 http://dx.doi.org/10.1787/888932336540 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 69 2.7. ALCOHOL CONSUMPTION AMONG ADULTS The global health burden related to excessive while falls in consumption in France, Italy and Spain alcohol consumption, both in terms of morbidity and may be associated with the voluntary and statutory mortality, is considerable (Rehm et al., 2009; WHO, regulation of advertising, partly following a 1989 2004). It is associated with numerous harmful health European directive. In 2010, the World Health and social consequences. High alcohol intake increases Organization endorsed a global strategy to combat the the risk for heart, stroke and vascular diseases, as well harmful use of alcohol, through direct measures as liver cirrhosis and certain cancers. Foetal exposure such as medical services for alcohol-related health to alcohol increases the risk of birth defects and intel- problems, and indirect ones, such as the dissemi- lectual impairments. Alcohol also contributes to death nation of information on alcohol-related harm and disability through accidents and injuries, assault, (WHO, 2010c). violence, homicide and suicide, and is estimated to Although adult alcohol consumption per capita cause more than 2 million deaths annually. gives useful evidence of long-term trends, it does not Alcohol consumption across EU countries is identify sub-populations at risk from harmful drink- 10.8 litres per adult per year. Leaving aside Luxembourg ing patterns. The consumption of large quantities of – because of the high volume of purchases by non- alcohol at a single session, termed “binge drinking”, residents in that country – Estonia, Hungary and is a particularly dangerous pattern of consumption France reported the highest consumption of alcohol, (Institute of Alcohol Studies, 2007), which is on the with more than 12.5 litres per adult in 2007-08. At the rise in some countries and social groups, especially other end of the scale, Turkey, Malta and some of the among young males (see Indicator 2.1 “Smoking and Nordic countries (Norway, Sweden and Iceland) have alcohol consumption at age 15”). relatively low levels of alcohol consumption, ranging Figure 2.7.3 shows the relationship between from one to seven litres per adult (Figure 2.7.1). alcohol consumption in 2005 and deaths from liver Although average alcohol consumption has grad- cirrhosis in 2008. In general, countries with high levels ually fallen in many EU countries over the past three of alcohol consumption tend to experience higher decades, it has risen in some others (Figure 2.7.1). death rates from liver cirrhosis. In most EU countries, There has been a degree of convergence in drinking death rates from liver cirrhosis have fallen over the habits across the European Union, with wine past two decades, following quite closely the overall consumption increasing in many traditional beer- reduction in alcohol consumption. drinking countries and vice versa. The traditional wine-producing countries of Italy, France and Spain, as well as the Slovak Republic, Greece and Germany have seen their alcohol consumption per capita fall Definition and deviations substantially since 1980 (Figures 2.7.1 and 2.7.2). On the other hand, alcohol consumption per capita in Alcohol consumption is defined as annual Iceland, Cyprus, Finland and Ireland rose by as much sales of pure alcohol in litres per person aged as 30% or more since 1980 although, in the case of 15 years and over. The methodology to convert Iceland and Cyprus, it started from a low level and alcohol drinks to pure alcohol may differ across therefore remains relatively low. countries. Variations in alcohol consumption across coun- Italy reports consumption for the population tries and over time reflect not only changing drinking 14 years and over, and Sweden for 16 years and habits but also the policy responses to control alcohol over. In some countries (e.g. Luxembourg), use. Curbs on advertising, sales restrictions and national sales do not accurately reflect actual taxation have all proven to be effective measures to consumption by residents, since purchases by reduce alcohol consumption (Bennett, 2003). Strict non-residents may create a significant gap controls on sales and high taxation are mirrored by between national sales and consumption. overall lower consumption in most Nordic countries, 70 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.7. ALCOHOL CONSUMPTION AMONG ADULTS 2.7.1. Alcohol consumption among population aged 15 years and over 2008 (or nearest year available) Change per capita, 1980-2008 1.4 Turkey -22 5.3 Malta n.a. 6.8 Norway 28 6.9 Sweden 3 7.3 Iceland 70 8.1 Italy -50 9.0 Greece -32 9.3 Cyprus 50 9.6 Netherlands -17 9.6 Slovak Republic -34 9.7 Romania -11 9.9 Germany -30 10.2 Latvia -22 10.2 Switzerland -24 10.3 Finland 30 10.7 Belgium -21 10.8 EU -13 10.8 Poland 24 10.8 United Kingdom 15 10.9 Bulgaria -2 10.9 Denmark -7 10.9 Slovenia n.a. 11.4 Portugal -23 11.7 Spain -36 12.1 Czech Republic 3 12.4 Ireland 29 12.5 Austria -14 12.5 Lithuania n.a. 12.6 France -35 12.6 Hungary -15 14.0 Estonia n.a. 15.5 Luxembourg 16 20 15 10 5 0 -75 -50 -25 0 25 50 75 Litres per capita % change over period Source: OECD Health Data 2010; Eurostat Statistics Database; WHO (2010). 1 2 http://dx.doi.org/10.1787/888932336559 2.7.2. Trends in alcohol consumption, 2.7.3. Alcohol consumption, 2005 and chronic liver selected EU countries, 1980-2008 disease deaths, 2008 (or nearest year available) France Hungary Chronic liver disease (deaths per 100 000 population) 50 Sweden EU Alcohol consumption (litres per capita) R 2 = 0.30 ROU HUN 20 40 LTU 30 15 SVK SVN EST FIN 20 BGR POL CZE LVA DNK 10 DEU GBR AUT 10 ITA ESP FRA GRC IRL PRT SWE CHE CYP NOR NLD ISL BEL 5 0 1980 1985 1990 1995 2000 2005 0 5 10 15 20 Alcohol consumption (litres per capita) Source: OECD Health Data 2010; Eurostat Statistics Database; WHO Source: OECD Health Data 2010; Eurostat Statistics Database; WHO (2010). (2010). 1 2 http://dx.doi.org/10.1787/888932336578 1 2 http://dx.doi.org/10.1787/888932336597 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 71 2.8. OVERWEIGHT AND OBESITY AMONG ADULTS The growth in overweight and obesity rates lower rates. Again, the gradient in obesity is stronger in among adults is a major public health concern. Obesity women than in men (OECD, 2010c). is a known risk factor for numerous health problems, A number of behavioural and environmental including hypertension, high cholesterol, diabetes, factors have contributed to the rise in overweight and cardiovascular diseases, respiratory problems obesity rates in industrialised countries, including (asthma), musculoskeletal diseases (arthritis) and falling real prices of food and more time spent being some forms of cancer. Mortality also increases sharply physically inactive. Overweight and obesity has risen once the overweight threshold is crossed (OECD, 2010c). rapidly in children in recent decades, reaching double- More than half (50.1%) of the adult population in figure rates in most EU countries (see Indicator 2.4). the European Union are overweight or obese. The Because obesity is associated with higher risks of prevalence of overweight and obesity among adults chronic illnesses, it is linked to significant additional exceeds 50% in no less than 15 of 27 EU countries. In health care costs. There is a time lag between the contrast, overweight and obesity rates are much lower onset of obesity and related health problems, suggest- in France, Italy and Switzerland, although rates are ing that the rise in obesity over the past two decades also increasing in these countries. The prevalence of will mean higher health care costs in the future. A obesity – which presents greater health risks than recent study estimated that total costs linked to overweight – varies threefold among countries, from a overweight and obesity in England in 2015 could low of less than 10% in Romania, Switzerland and increase by as much as 70% relative to 2007 and could Italy to over 20% in the United Kingdom, Ireland, be 2.4 times higher in 2025 (Foresight, 2007). Malta and Iceland (Figure 2.8.1). Across the European Union, 15.5% of the adult population is obese. There is little difference in the average obesity rate of men and women in the European Union, with Definition and deviations both at around 15% (Figure 2.8.1). However, there is some variation among individual countries, with men Overweight and obesity are defined as exces- generally being more obese than women in Norway, sive weight presenting health risks because of Malta and Italy, whereas women are more obese the high proportion of body fat. The most in Latvia, Turkey and the Netherlands. The largest frequently used measure is based on the body disparities in obesity between men and women were mass index (BMI), which is a single number that in Latvia, whereas there was little, if any difference in evaluates an individual’s weight in relation to male and female obesity rates in the Czech Republic, height (weight/height2, with weight in kilograms Poland and Sweden. and height in metres). Based on the WHO The rate of obesity has more than doubled over classification (WHO, 2000), adults with a BMI the past 20 years in most EU countries for which data between 25 and 30 are defined as overweight, are available (Figure 2.8.2). The rapid increase occurred and those with a BMI over 30 as obese. This clas- regardless of what the levels of obesity were two sification may not be suitable for all ethnic decades ago. Obesity more than doubled in both the groups, who may have equivalent levels of risk Netherlands and the United Kingdom between 1988 at lower or higher BMI. The thresholds for adults and 2008, even though the rate in the Netherlands is are not suitable to measure overweight and currently less than half that of the United Kingdom. obesity among children. In most countries the rise in obesity has affected For most countries, overweight and obesity all population groups regardless of sex, age, race, rates are self-reported through estimates of income or education level, but to varying extents. height and weight from population-based health Evidence from a number of countries, including interview surveys. The exceptions are Ireland, Austria, England, France, Italy and Spain, indicates that Luxembourg, the Slovak Republic (2008) and the obesity tends to be more common among individuals United Kingdom, where estimates are derived in disadvantaged socio-economic groups, with this from health examinations. These differences relationship being particularly strong among women limit data comparability. Estimates from health (Sassi et al., 2009b). There is also a relationship between examinations are generally higher and more the number of years spent in full-time education and reliable than from health interviews. obesity, with the most educated individuals displaying 72 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 2.8. OVERWEIGHT AND OBESITY AMONG ADULTS 2.8.1. Obesity rates among adults, 2008 (or nearest year available) Males Females All adults 7.9 Romania 8 8 8.1 Switzerland 8 9 9.9 Italy 9 11 10.0 Norway 8 11 10.2 Sweden 10 10 11.1 Netherlands 12 10 11.2 France 12 11 11.4 Denmark 12 11 11.5 Bulgaria 11 12 12.4 Austria 13 12 12.5 Poland 13 13 13.6 Germany 13 14 13.8 Belgium 14 13 14.9 Spain 15 15 19 15.2 Turkey 12 16 15.4 Portugal 15 16 15.5 EU 15 15 15.6 Cyprus 17 16 15.7 Finland 15 16 16.4 Slovenia 17 17 16.9 Slovak Republic1 17 21 16.9 Latvia 12 17 17.1 Czech Republic 17 18 18.0 Estonia 18 19 18.1 Greece 18 18 18.8 Hungary 20 19 19.7 Lithuania 21 19 20.0 Luxembourg1 21 21 20.1 Iceland 19 21 22.3 Malta 24 24 23.0 Ireland1 22 25 24.5 United Kingdom1 24 30 20 10 0 0 10 20 30 % of adult population % of adult population 1. Ireland, Luxembourg, Slovak Republic and United Kingdom figures are based on health examination surveys, rather than health interview surveys. Source: OECD Health Data 2010; Eurostat Statistics Database; WHO Global Infobase. 1 2 http://dx.doi.org/10.1787/888932336616 2.8.2. Increasing obesity rates among adults in EU countries % 30 25 25 23 22 20 20 20 19 20 19 18 17 17 17 16 16 16 15 16 15 15 14 14 15 14 13 13 12 12 13 12 12 11 11 12 11 12 12 11 11 10 10 10 10 10 9 9 10 9 9 10 8 8 8 8 8 8 8 7 7 6 6 6 6 5 5 0 8) ,2 ) nd 8 9 8, 2 ) 19 99 08) ) ) 99 8) an 9 6 6 ) m 9, 2 ) ) 6) ) ) 8) ) a ( 9, 2 ) 9 9 01 5) ) a n ur g 1 8 , 2 ) ) do a l t a 0 2, 7 ) 9 8 0 2, 7 ) 98 8) 8) 7) 98 8) (1 199 05 99 04 07 (1 200 006 (1 199 008 Au lga 20 0 0 8 08 L a , 19 008 pu (19 008 to 199 008 ua 20 0 0 0 8 8 (1 (19 006 0 20 20 0 0 De ce ( , 19 200 (1 200 (1 200 19 00 Po , 19 200 19 00 0 00 rm 19 00 C y l (1 200 0 20 (2 20 20 Be y (1 , 20 0 0 20 20 2 2 ,2 a r 9 0, 8 , 2 ,2 ,2 Ge nd ( , 2 2 a ( 4, 2 , 7, 0, 8 8 8, 4, 7, 3, , 8, , ga 97, 9, 99 98 99 98 98 99 00 99 99 9 9 r tu 19 0 Sw Nor (19 0 19 s( ,1 (2 (2 ( ( ( ( Po 87, Re i c 1 7, 0, Bu 87, nd 1 8 ly ic Ne ed way ia ia an us ria 98 ni ni Sp giu bl It a tv an 9 la 19 9 nm 19 l la pr e c ub (1 (1 M er Ic bo m k( l m1 Es th C z Rep n it z ria en d Ro d m ai Li Sw an xe st h la nl el ak Lu Fr ng er Fi ov th Ki Sl d i te Un 1. Luxembourg, Slovak Republic (2008) and United Kingdom figures are based on health examination surveys, rather than health interview surveys. Source: OECD Health Data 2010; Eurostat Statistics Database; WHO Global Infobase. 1 2 http://dx.doi.org/10.1787/888932336635 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 73 Health at a Glance: Europe 2010 © OECD 2010 Chapter 3 Health Care Resources, Services and Outcomes 3.1. Practising physicians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3.2. Practising nurses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 3.3. Childhood vaccination programmes . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 3.4. Influenza vaccination for older people . . . . . . . . . . . . . . . . . . . . . . . . . 82 3.5. Medical technologies: CT scanners and MRI units . . . . . . . . . . . . . . . 84 3.6. Hospital beds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 3.7. Hospital discharges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 3.8. Average length of stay in hospitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 3.9. Cardiac procedures (coronary angioplasty) . . . . . . . . . . . . . . . . . . . . . 92 3.10. Cataract surgeries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.11. Hip and knee replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 3.12. Screening, survival and mortality for cervical cancer . . . . . . . . . . . . 98 3.13. Screening, survival and mortality for breast cancer . . . . . . . . . . . . . . 100 75 3.1. PRACTISING PHYSICIANS Access to high-quality services depends crucially the 1980s and 1990s, the number of doctors per capita on the size, skill mix, geographic distribution and in Italy peaked in 2002, and has declined since then. productivity of the health workforce. Health workers, In France, the number peaked in 2005, and the decline and in particular doctors and nurses, are the corner- is expected to continue to 2020 (DREES, 2009). stone of health systems. In nearly all countries, the balance between In 2008, Greece had, by far, the highest number of general practitioners and specialists has changed over doctors per capita, with six doctors per 1 000 population, past decades, with the number of specialists increas- nearly twice the EU average. Following Greece were ing much more rapidly than generalists. As a result, Austria, Italy and Norway, with four doctors or more per there are more specialists than generalists in all 1 000 population. The number of doctors per capita was countries, except Romania and Portugal (Figure 3.1.2). the lowest in Turkey, followed by Poland and Romania. A number of reasons explain this trend. There may be Doctor numbers are also relatively low in the less interest in the traditional mode of practice of United Kingdom and Finland (Figure 3.1.1). general/family practitioner and the workload and Since 2000, the number of physicians per capita constraints attached to it. In addition, in many coun- has increased in all European countries, except the tries, there is a growing remuneration gap between Slovak Republic. On average across EU countries, generalists and specialists (Fujisawa and Lafortune, physician density grew at a rate of 1.5% per year, rising 2008). The slow or negative growth in the number of from 3.0 doctors per 1 000 population to 3.3. The generalists per capita raises concerns about access growth rate was particularly rapid in Turkey, which to primary care for certain population groups. In started from the lowest level in 2000, thereby narrow- response to this shortage, many countries are consid- ing the gap with other countries. ering ways to improve the attractiveness of general practice as well as the development of new roles The number of doctors also increased rapidly for other health care providers, such as nurses in Ireland, rising by nearly 50% (from 2.2 per (Delamaire and Lafortune, 2010). 1 000 population in 2000 to 3.2 in 2008). A large part of this increase is due to the recruitment of foreign- trained physicians. The share of foreign-trained physicians in Ireland more than tripled during this Definition and deviations period, rising from 11% of all physicians in 2000 to 35% in 2008 (OECD and WHO, 2010). There has also been a Practising physicians are defined as doctors substantial rise in the number of students graduating who are providing care directly to patients. from medical schools in Ireland (OECD, 2010a). In some countries, the numbers also include A similar pattern has been observed in the doctors working in administration, manage- United Kingdom, where the number of doctors went ment, academic and research positions (“profes- up from 2.0 per 1 000 population in 2000 to 2.6 in 2008, sionally active” physicians), adding another an increase of 30%. The number of new registrations 5-10% of doctors. Ireland, the Netherlands and of foreign-trained doctors in the United Kingdom Portugal report all physicians entitled to increased to 2003 when it peaked at about 14 000, but practice, resulting in an over-estimation. has declined since then to just over 5 000 in 2008 Not all countries are able to report all physi- (OECD and WHO, 2010). At the same time, the number cians as generalists or specialists, and in some of new graduates from medical schools in the countries (e.g. the Netherlands), most physicians United Kingdom increased, from about 4 600 in 2003 are not reported in either of these two broad to 5 600 in 2008, gradually exceeding the number of categories. In some countries, data on medical new registrations of foreign-trained physicians. specialty may not be available for interns/ In contrast, in France and Italy there was residents (physicians in training) or for those virtually no growth. Following the reduction in the working in private practice. number of new entrants in medical schools during 76 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.1. PRACTISING PHYSICIANS 3.1.1. Practising physicians per 1 000 population 2008 (or nearest year available) Change 2000-08 (or nearest year available) 6.0 Greece1 4.2 4.6 Austria 2.2 4.2 Italy1 0.1 4.0 Norway 4.1 3.8 Switzerland n.a. 3.7 Iceland 1.0 3.7 Netherlands 2 2.7 3.7 Lithuania 0.3 3.7 Portugal 2 1.8 3.6 Czech Republic 0.8 3.6 Spain n.a. 3.6 Bulgaria 0.9 3.6 Sweden 2.5 3.6 Germany 1.1 3.4 Denmark 2.3 3.4 Estonia 0.3 3.3 EU 1.5 3.3 France 1 0.2 3.2 Ireland 2 4.8 3.1 Latvia 1.0 3.1 Hungary n.a. 3.0 Malta n.a. 3.0 Slovak Republic -1.0 3.0 Belgium 0.8 2.9 Cyprus 1.3 2.8 Luxembourg n.a. 2.7 Finland 1.1 2.6 United Kingdom 3.6 2.4 Slovenia 1.4 2.2 Romania 1.7 2.2 Poland n.a. 1.5 Turkey1 4.8 6 4 2 0 -4 0 4 8 Per 1 000 population Average annual growth rate (%) 1. Data include not only physicians providing direct care to patients, but also those working in the health sector as managers, educators, researchers, etc. (adding another 5-10% of doctors). 2. Data refer to all physicians who are licensed to practice. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336654 3.1.2. General practitioners, specialists and other physicians as a share of total physicians, 2008 (or nearest year available) % GPs Specialists Other physicians 100 80 60 40 20 25.0 33.2 33.2 50.2 33.5 54.5 22.8 39.2 22.9 28.8 20.8 20.9 25.6 20.0 48.7 28.7 16.9 18.8 14.3 18.4 18.4 13.8 19.9 15.7 15.7 12.1 17.4 17.0 4.5 8.0 0 Es 1 n1 en 1 ic Po ni a Fr l Be ce m Au d ria a EU ly C z De ia Re ar k th ia Bu n y nd i t z nd No ic ey m m ta Ge ni a Ir e a th nd Re ds Po y Gr d ce ga g a ni i an n bl en bl It a tv ar iu al xe do ai ur rk rw an ee ak an a la S w ela N e er l a la st to a ua r tu e c nm ed rm pu pu nl M La lg lg Sp ov m bo ite Tu L u ing ov er l Ic Fi Sw Ro Sl Li K h d Un Sl Note: Specialists include paediatricians, obstetricians/gynaecologists, psychiatrists, medical specialists and surgical specialists. Other physicians include interns/residents if not reported in the field in which they are training, and doctors not elsewhere classified. 1. Data are not available for specialists. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336673 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 77 3.2. PRACTISING NURSES Nurses are usually the most numerous health particularly large in Portugal and Spain, where the profession, outnumbering physicians in most number of nurses per population increased by 45% European countries. Nurses play a critical role in and 33% respectively. In France and Switzerland, there providing health care not only in traditional settings was also a fairly large increase in the supply of nurses, such as hospitals and long-term care institutions but rising by 15-20% between 2000 and 2008. increasingly in primary care (especially in offering In 2008, the number of nurses per doctor ranged care to the chronically ill) and in patients’ homes. from about six in Ireland and Finland to under one nurse However, there are concerns in many countries per doctor in Greece and Turkey (Figure 3.2.2). The about shortages of nurses, and these concerns may average across European countries is over two-and-a- well intensify in the future as the demand for nurses half nurses per doctor, with many countries reporting continues to increase and the ageing of the “baby between two to four nurses per doctor. Beyond Greece boom” generation precipitates a wave of retirements and Turkey, the nurse-to-doctor ratio is also relatively among nurses. These concerns have prompted many low in other southern European countries, such as Italy, countries to increase the training of new nurses Spain, Portugal and Cyprus. In Greece and Italy, there is combined with efforts to increase retention rates in evidence of an over-supply of doctors and under-supply the profession (OECD, 2008b). of nurses, resulting in an inefficient allocation of This section presents data on the number of resources (OECD, 2009; Chaloff, 2008). nurses, including both “professional nurses” and “associate professional nurses” in those countries where two such levels of nurses exist. It also provides data on other lower-skilled caring personnel such as nursing aides. Definition and deviations In 2008, there were about 15 professional and associate professional nurses per 1 000 population in The data refer to nurses and other caring Finland, Iceland, Ireland and Switzerland, and slightly personnel providing direct care to patients, fewer in Denmark and Norway. Turkey had the fewest although in some countries they also include nurses, followed by Greece, Bulgaria and Cyprus, with nurses working in management, research and all these countries having fewer than five nurses per other roles. This adds another 5-10% to nursing 1 000 population. numbers. The mix between different categories of nurses “Professional nurses” are defined by ISCO-08 varies widely across European countries. In some code 2221, and include categories of nurses such countries such as France, Portugal and Poland, a as registered nurses, clinical nurses, nurse lower-level category of “associate professional nurses” anaesthetists, nurse practitioners, public health does not exist, and all nurses are reported to be at the nurses, and specialist nurses. “Associate same level. In other countries such as the United professional nurses” are defined by ISCO-08 Kingdom, Germany and Austria, the vast majority of code 3221, and include categories of nurses nurses are considered to be professional nurses, but a such as “enrolled nurses” and “practical nurses”. minority are considered to be at a lower-level. In yet “Caring personnel” includes two categories of another group of countries including the Netherlands workers defined in ISCO-08: 1) “health care and Slovenia, the number of lower-level nurses is assistants” (code 5321) who “provide direct greater than higher-level nurses (Figure 3.2.1). personal care and assistance with activities of daily living to patients and residents in a variety In addition to different categories of recognised of health care settings”; and 2) “home-based nurses, other categories of caring personnel such as personal care workers” (code 5322), including nursing aides play an important role in supporting home care aides, nursing aides at home, and nurses in providing care in some countries. However, personal care providers. because these personnel are usually not part of a registered profession, the availability and coverage Midwives are usually excluded from nurses. of data is more limited. Based on the available data, However, about half of European countries the number of such additional caring personnel is the report midwives together with nurses, as they highest in the Netherlands, Norway and Denmark. In are considered specialist nurses. the Netherlands and France, there are in fact more Austria reports only nurses working in hospi- caring personnel than nurses. tals. The data for Germany does not include Since 2000, the number of nurses per capita nurses who have three years of education and has increased in all European countries, except in are providing services for the elderly. Lithuania and the Slovak Republic. The increase was 78 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.2. PRACTISING NURSES 3.2.1. Professional nurses, associate professional nurses and caring personnel per 1 000 population, 2008 (or nearest year available) Professional nurses Associate professional nurses Caring personnel Per 1 000 population 25 23.0 24.6 24.5 21.4 20 16.5 16.2 15.5 14.8 15 10.9 10.8 10.7 10.4 9.8 9.7 9.5 10 9.3 8.7 8.7 8.0 7.7 7.1 6.5 6.3 6.3 5.7 5.5 5.2 5 4.7 4.3 4.2 1.3 0 Ir e c e 1 4 Ge en 4 K i lic 4 2 L i l y 3, 4 4 Ro l 1, 4 Gr 4 ,4 ay ds Es ia Hu ni a Fr d d m d Au r y S w ur g y n Un Re E U Sl om a ta pu i a Po i a C y nd Bu c e Tu ria No r k 1, Po lic 1, ey 1 us ria an ni ai n an xe a n en an Re a t v al a a nd Ne r w ee S w r lan la la a ga to Sp ua an b nm ed d ng pr bo st rm M nl L u Ic el lg It a ov m er i te pu rk ng b la L r tu th Fi e it z De th d h ec ak Cz ov Sl 1. Data include not only nurses providing direct care to patients, but also those working in the health sector as managers, educators, researchers, etc. (adding another 5-10% of nurses). 2. Austria reports only nurses employed in hospitals. 3. In Italy, data refer to all nurses who are licensed to practice. 4. The breakdown between professional and associate professional nurses is not available. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336692 3.2.2. Ratio of nurses to physicians, 2008 (or nearest year available) 7 6.4 6 5.7 5 4.2 4.0 4.0 4.0 3.9 4 3.7 3.5 3.3 3.0 3.0 3.0 3 2.6 2.5 2.4 2.4 2.2 2.0 2.0 1.9 1.9 1.9 2 1.7 1.6 1.5 1.3 1.2 1.0 0.9 1 0.6 0 Sl ay ic nd De and m k Ic g th nd Un w i t nds K i and No m Sw ia Ge en Au y ria Ro E U Po i a nd Re e Hu t a L i ar y ov E s i a ic Re i a C y ia Po r us l Bu n ia ly Gr y ce ga an e nc xe ar ai ur bl en bl an n ak ton ar It a tv al do rk rw ee ed la Ne ela la st ua Sp r tu L u nm ng p bo rm e c Fr a pu pu S rla nl l M La lg ov m Tu Ir e i t e z er ng th Fi e h d Cz Sl Note: Nurses only include professional and associate professional nurses and exclude other caring personnel. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336711 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 79 3.3. CHILDHOOD VACCINATION PROGRAMMES Childhood vaccination continues to be one of the Figure 3.3.3 shows that the average percentage of most cost-effective public health interventions. All children aged 2 years vaccinated for hepatitis B across European countries, or in some cases sub-national countries with national programmes is also over 95%. jurisdictions, have established vaccination pro- However, some European countries do not currently grammes based on their interpretation of the risks require children to be vaccinated by age 2, or do not and benefits of each vaccine. have routine programmes, and consequently the rates Vaccination against pertussis (often adminis- for these countries are significantly lower. For example, tered in connection with vaccination against diph- in Denmark and Sweden, vaccination against theria and tetanus) and measles is part of almost all hepatitis B is not an obligatory part of vaccination programmes. Reviews of the evidence supporting the programmes, and is only recommended to specific risk efficacy of vaccines against these diseases have groups. In France, hepatitis B vaccination remains concluded that the respective vaccines are safe and controversial, given ongoing speculation over possible highly effective. In the European Union, the gradual side effects. take-up of the measles vaccine has meant that Figure 1.11.3 in Chapter 1 indicates that the inci- measles incidence is around twenty times lower dence of hepatitis B is low in the majority of European than the rate of the early 1990s (see Indicator 1.11), countries, at less than 2 per 100 000 population. although outbreak can still occur. However, in Iceland, Bulgaria, Turkey, Austria, Latvia A vaccination for hepatitis B has been available and Romania, the rates are more than two times the since 1982 and is considered to be 95% effective in EU average. preventing infection and its chronic consequences, such as cirrhosis and liver cancer. In 2004, it was esti- mated that over 350 million people were chronically infected with the hepatitis B virus worldwide and at risk of serious illness and death (WHO, 2009a). In 2007, Definition and deviation more than 170 countries had already begun to follow Vaccination rates reflect the percentage of the WHO recommendation to incorporate hepatitis B children at age 1 or 2 receiving the respective vaccine as an integral part of their national infant vaccination. Childhood vaccination policies differ immunisation programme. across countries. Some countries administer Figures 3.3.1 and 3.3.2 demonstrate that the overall combination vaccines (e.g. DTP for diphtheria, vaccination of children against pertussis (including tetanus and pertussis) while others administer diphtheria and tetanus) and measles is high in most the vaccinations separately. Schedules for admin- European countries. On average, about 95% of 2-year-old istering vaccines also differ. children receive the recommended pertussis and Some countries ascertain vaccinations based measles vaccination, a level that is high enough to on surveys and others based on encounter data, provide effective immunity. Vaccination rates are the which may influence the results. lowest in Malta and Austria, with less than 85% of children vaccinated against these diseases. 80 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.3. CHILDHOOD VACCINATION PROGRAMMES 3.3.1. Vaccination rates for pertussis, children aged 2, 2008 (or nearest year available) % 100 98.0 98.0 97.4 97.3 97.0 97.0 97.0 97.0 97.0 96.8 96.7 96.7 96.5 96.1 96.0 96.0 96.0 99.9 99.6 99.4 99.0 98.7 95.0 95.0 94.6 94.5 95 94.0 93.0 90.0 90 89.0 85 83.0 80 75 72.0 70 Re b l i c Po ic Au r k ta Re ar y Be nd m Sw ce Fi n Ge d Ic y ov d C y ia La s Ro i a Po ni a l ly m n Es rg i t z ni a Ir e y nd De ece ria Tu d th ey a Un B E U K i ar i a t h om No ds ga u a an ni xe a i e an an n bl en tv It a iu al a u rk rw an ed n pr la la la st S w to a Lu Sp r tu ua nm N e ngd ng bo e rm pu pu la M nl el lg i t e ul g m er Gr Fr er Hu Sl Li h ak d ec ov Cz Sl Source: OECD Health Data 2010; WHO vaccine-preventable diseases: monitoring system 2010 global summary. 1 2 http://dx.doi.org/10.1787/888932336730 3.3.2. Vaccination rates for measles, children aged 2, 2008 (or nearest year available) % 100 97.6 97.0 97.0 97.0 97.0 96.6 99.9 99.5 96.2 96.0 96.0 96.0 96.0 98.9 98.7 95.4 95.4 98.0 97.9 94.0 93.7 93.0 93.0 95 90.0 89.5 89.0 89.0 90 87.0 87.0 85.9 85 83.0 80 78.0 75 70 Fi ce Gr i c ic it z ay Ir e k Fr d Re r y Po d Re nd Tu n Li t via P ni a l Sw ia Bu en Ic y d N m De l y i te C y e K i us ria ta t h ur g Ro n i a Ge ni a La y Sl ds Es ia Be EU nd Au m xe g a an e ar c ai n an an bl bl en It a ar al iu do ak ga rk ee S w or w an ed n d pr la la la st to Sp a ua L u or t u nm Ne bo rm pu pu M nl la el lg lg ov m er ng ov Hun th er m h ec Cz Un Sl Source: OECD Health Data 2010; WHO vaccine-preventable diseases: monitoring system 2010 global summary. 1 2 http://dx.doi.org/10.1787/888932336749 3.3.3. Vaccination rates for hepatitis B, children aged 2, 2008 (or nearest year available) % 100 95.2 94.5 93.0 92.0 90.2 99.8 99.4 99.4 99.3 99.0 97.5 97.0 96.8 96.5 96.2 96.1 96.0 96.0 96.0 95.3 83.0 82.0 80 60 40 33.1 20 15.7 0.3 0 1 ce 2 en 2 k2 ic ic n nd y ia m ia l a ly ia ia a ce g us y y ria ta ga EU ni ke an ni ar ai ur bl bl an en It a ar tv iu al ar ee pr la st to Sp ua r tu an ed ng bo r rm pu pu M La lg lg nm ov m Tu Po Au Cy Gr Es th Fr Sw Bu Be Hu m Po Ro Re Re Ge Sl De Li xe ak h Lu ec ov Cz Sl 1. EU average only includes countries with required or routine immunisation. 2. In France, Sweden and Denmark, vaccination for hepatitis B is not required or routinely provided. Source: OECD Health Data 2010; WHO vaccine-preventable diseases: monitoring system 2010 global summary. 1 2 http://dx.doi.org/10.1787/888932336768 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 81 3.4. INFLUENZA VACCINATION FOR OLDER PEOPLE Influenza is a common infectious disease world- care providers other than physicians. However, a wide and affects persons of all ages. Most people with number of barriers need to be overcome in some the illness recover quickly, but elderly people and those countries if they wish to increase their coverage rates with chronic medical conditions are at higher risk further. For example, possible reasons put forward for for complications and even death. For example, the relatively low vaccination rates in Austria include between 2000 and 2008, influenza along with other poor public awareness, inadequate insurance cover- acute upper respiratory infections accounted for about age of related costs, and lack of consensus within the 44 000 hospitalisations per year in France and 77 000 in Austrian medical profession about the importance of Germany. The impact of influenza on the employed vaccination (Kunze et al., 2007). population is substantial, even though most influenza New types of influenza, such as the H1N1 “swine morbidity and mortality occurs among the elderly and flu”, have emerged in recent years and prompted those with chronic conditions (e.g. 85-90% of people rapid responses to contain the pandemic. While who die from influenza in France and Germany are symptoms of the H1N1 influenza are mild in most over 65 years of age). people, a minority have suffered severe disease with Immunisation against seasonal influenza (or flu) some dying from it. The majority of those people who for older people has become increasingly widespread have suffered severely from the disease have other in many European countries over the past decade. chronic medical conditions such as asthma or heart Influenza vaccination for patients with chronic disease. But there have also been cases of people who conditions and other at-risk groups is also strongly became severely ill without any underlying condition recommended in many countries. (European Commission, 2010c). A series of public In 2008, more than half of the population health measures used to combat seasonal flu have aged 65 years and over were vaccinated for influenza been used to combat new strains of influenza in in 14 European countries (Figure 3.4.1). There is a wide Europe, including massive vaccination campaigns for variation in vaccination rates, ranging from lows of risk groups (European Commission, 2010c). 21% in the Czech Republic and 26% in Slovenia, to over 75% in the Netherlands and the United Kingdom. Figure 3.4.2 indicates that while the European average increased markedly between 1998 and 2003, it Definitions and deviations remained relatively stable between 2003 and 2008. From 2003, some countries marginally increased their Influenza vaccination rate refers to the coverage whereas others reduced it, most notably in number of people aged 65 and older who have countries which were already below the EU average, received an annual influenza vaccination, such as Slovenia, the Slovak Republic and Hungary. divided by the total number of people over A number of factors contributed to the rise in 65 years of age. The main limitation in terms of influenza immunisation rates in most European data comparability arises from the use of differ- countries over the past decade, including greater ent data sources, whether survey or programme, acceptance of preventive health services by patients which are susceptible to different types of errors and practitioners, improved public health insurance and biases. coverage for vaccines, and wider delivery by health 82 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.4. INFLUENZA VACCINATION FOR OLDER PEOPLE 3.4.1. Influenza vaccination coverage, population aged 65 and over, 2008 (or nearest year available) % 80 77.0 75.1 70.1 70.0 66.2 65.4 65.0 62.5 60 57.0 56.0 56.0 54.2 54.1 51.0 50.4 40 37.8 36.1 35.5 26.0 21.2 20 0 y1 1 ly ic ic ia s m nd ce n 4) k en nd ) g d l y ga ria es nd ar ar ai an ur bl bl en It a an do 00 an ed la la Sp r tu nm ng tri st bo pu pu la nl ov Ir e rm er ng Fr (2 Sw Au er Fi un Hu m Po Re Re it z De Sl Ki Ge th m xe co Sw ak h iu Ne d Lu ec i te 8 lg ov (1 Cz Be Un Sl EU 1. Population aged 60 and over. Source: OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932336787 3.4.2. Vaccination rates for influenza, population aged 65 and over, 1998-2008 (or nearest year available) % 1998 2003 2008 80 70.0 70.1 77.0 66.2 75.1 65.0 62.5 57.0 57.0 65.4 56.0 60 54.1 51.0 50.4 56.0 36.1 40 37.8 35.5 21.2 26.0 20 0 y1 1 ic ic s m nd ce ly n m en ) nd g l ia d y k ga ria es nd ar ar ai an ur bl bl en It a an iu do an ed la la Sp r tu nm ng tri st bo pu pu la nl lg ov Ir e rm er ng Fr Sw Au er Fi Be un Hu m Po Re Re it z De Sl Ki Ge th xe co Sw ak h Ne d Lu ec i te 1 ov (1 Cz Un Sl EU 1. Population aged 60 and over. Source: OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932336806 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 83 3.5. MEDICAL TECHNOLOGIES: CT SCANNERS AND MRI UNITS New medical technologies are improving diagno- has the highest number of these two types of sis and treatment, but they are also increasing health scanners (Figures 3.5.3 and 3.5.4). The number of CT spending. This section presents data on the availability and MRI exams per capita is also above average in and use of two diagnostic technologies: computed Belgium, Luxembourg and Iceland. It is the lowest in tomography (CT) scanners and magnetic resonance the Slovak Republic and Czech Republic, as well as in imaging (MRI) units. CT scanners and MRI units help the Netherlands for CT exams. physicians diagnose a range of conditions by producing In Greece, most CT and MRI scanners are installed images of internal organs and structures of the body. in the growing number of private diagnostic centres, Unlike conventional radiography and CT scanning, and only a minority are found in public hospitals. newer imaging technology used in MRI units do not There is no regulation concerning the purchase of MRI expose patients to ionising radiation. units in Greece, while the purchase of CT scanners The availability of CT scanners and MRI units has requires a licence that is granted following a review increased rapidly in most European countries over the based on a criteria of population density. There are also past two decades. For example, in the Netherlands, the no guidelines concerning the use of CT and MRI number of MRI units per capita multiplied by ten scanners (Paris et al., 2010). The current situation has between 1990 and 2008, while the number of CT scan- led the Greek Ministry of Health and Social Solidarity to ners also increased. Similarly, in Italy, the number of MRI establish an expert’s committee to review regulations scanners per capita multiplied by five between 1997 and propose new criteria for the purchase of CT and and 2007, and the number of CT scanners doubled. MRI scanners. In 2008, Greece had the highest number of MRI Many other European countries are also examin- and CT scanners per capita (together with Cyprus for ing ways to promote more rational purchase and use CT scanners). Switzerland, Iceland, Italy and Austria of such diagnostic technologies (OECD, 2010b). In the also had significantly more MRI and CT scanners than United Kingdom, the National Institute for Health and the EU average (Figures 3.5.1 and 3.5.2). However, the Clinical Excellence has recently set up a Diagnostics number of MRI and CT scanners in all European Advisory Committee to evaluate and make recom- countries remains much lower than in Japan and the mendations for the appropriate use of diagnostic United States (OECD, 2010a). The number of MRI units technologies within the NHS in England (NICE, 2009). and CT scanners per population were the lowest in Romania and Hungary. There is no general guideline regarding an ideal Definition and deviations number of CT scanners or MRI units per population. However, if there are too few such items of equip- The figures relate to the number of CT and MRI ment, this may lead to access problems, either in scanners per million population. terms of geographic proximity or waiting times. On The data generally cover the equipment the other hand, if there are too many, this may result installed in hospitals and ambulatory settings, in an overuse of these costly diagnostic procedures, with the exception of Belgium, Germany and with little if any benefits to patients. Spain where the equipment outside hospitals is Data on the use of these diagnostic equipment not included, and France where only a small are available only for a smaller group of countries. number of equipment in ambulatory settings is Based on this more limited country coverage, the included. In the United Kingdom, the data refer number of CT and MRI exams per capita is the highest only to scanners in the public sector. in Greece, consistent with the fact that Greece also 84 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.5. MEDICAL TECHNOLOGIES: CT SCANNERS AND MRI UNITS 3.5.1. Number of MRI units, 2008 3.5.2. Number of CT scanners, 2008 (or nearest year available) (or nearest year available) Hospital Outside hospital Total Hospital Outside hospital Total Greece 21.8 Cyprus 35.5 Italy 20.0 Greece 33.9 Iceland 18.8 Switzerland 32.0 Austria 18.0 Malta 31.7 Cyprus 16.5 Iceland 31.3 Finland 16.2 Italy 31.0 Denmark 15.4 Austria 29.9 Switzerland1 14.0 Luxembourg 27.6 Luxembourg 12.7 Portugal 26.0 Ireland 12.3 Latvia 22.9 Belgium1 10.6 Bulgaria 22.4 Netherlands 10.4 Denmark 21.5 Spain1 9.9 EU 19.2 EU 9.5 Lithuania 16.6 Portugal 8.9 Finland 16.5 Germany1 8.6 Germany1 16.4 Estonia 8.2 Spain1 15.3 Malta 7.3 Ireland 15.1 Turkey 6.9 Estonia 14.9 Latvia 6.6 Slovak Republic 13.7 France1 6.1 Belgium1 13.6 Slovak Republic 6.1 Czech Republic 13.5 United Kingdom 2 5.6 France1 11.0 Czech Republic 5.1 Poland 10.9 Lithuania 4.2 Slovenia 10.9 Slovenia 3.5 Netherlands 10.3 Bulgaria 3.1 Turkey 10.2 Poland 2.9 United Kingdom 2 7.4 Hungary 2.8 Hungary 7.1 Romania 1.3 Romania 4.0 0 5 10 15 20 25 0 10 20 30 40 Scanners per million population Scanners per million population Note: The EU average does not include countries which only Note: The EU average does not include countries which only report equipment in hospital. report equipment in hospital. 1. Data for equipment outside hospital are not available. 1. Data for equipment outside hospital are not available. 2. In the United Kingdom, any equipment in the private sector is 2. In the United Kingdom, any equipment in the private sector is not included in the data. not included in the data. Source: OECD Health Data 2010; Eurostat Statistics Database. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336825 1 2 http://dx.doi.org/10.1787/888932336844 3.5.3. Number of MRI exams, 2008 3.5.4. Number of CT exams, 2008 (or nearest year available) (or nearest year available) Exams per 1 000 population Exams per 1 000 population 100 350 98.1 320.9 300 80 72.3 250 62.8 60 53.7 200 182.6 175.5 48.5 47.6 164.0 38.8 37.8 150 139.8 139.4 40 36.5 130.0 27.8 24.2 100 83.8 83.3 82.7 20 60.3 50 0 0 ic ic ) ic ic ce d g m ce ) s a ce m g d a ce s k k es es nd nd ni ni ar ar an an ur ur bl bl bl bl iu iu an ee an ee to to nm nm tri tri bo bo pu pu pu pu la la el el lg lg Gr Gr Fr Fr Es Es er Ic er Ic Be Be un un m m Re Re Re Re De De th th xe xe co co h ak h ak Ne Ne Lu Lu ec ec (9 (9 ov ov Cz Cz EU EU Sl Sl Source: OECD Health Data 2010. Source: OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932336863 1 2 http://dx.doi.org/10.1787/888932336882 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 85 3.6. HOSPITAL BEDS The number of hospital beds provides an indica- the share of beds in public hospitals from 44% to 41%. tion of the resources available for delivering services The remaining 30% were beds in private not-for-profit to in-patients in hospitals. This section presents data hospitals, whose share also declined slightly. In on the total number of hospital beds, including those Austria also, the share of beds in private for-profit allocated for curative, psychiatric, long-term and hospitals has increased over the past decade, from 7% other types of care. It does not capture the capacity of in 1995 to just over 10% in 2008, although the vast hospitals to furnish same-day emergency or elective majority of beds continue to be in publically-owned interventions. hospitals. In France, the allocation of beds in public Over the past 15 years, the number of hospital and private hospitals has remained fairly stable beds per population has decreased in all European since 1997, with about 65% of beds located in public countries. On average across EU countries, the hospitals, 15% in private not-for-profit hospitals, and number fell from 7.3 beds per 1 000 population in 1995 the remaining 20% in private for-profit hospitals to 5.7 in 2008 (Figure 3.6.1). This reduction in the (OECD, 2010a). number of hospital beds has been accompanied by a In several countries, the reduction in the overall reduction in average length of stays in hospitals number of hospital beds has been accompanied by an (Indicator 3.8) and, in some countries, a reduction in increase in their occupancy rates. Since 1995, the hospital admissions (Indicator 3.7). The reduction in occupancy rate of curative care beds increased signif- the number of hospital beds per population has been icantly in Ireland, Italy, Norway and Switzerland particularly pronounced in Latvia, Lithuania and (OECD, 2010a). Bulgaria. In all countries, progress in medical technologies has enabled a move to day-surgery and a reduced Definition and deviations need for long hospitalisation. In addition, in many countries, cost-containment policies over the past Hospital beds are defined as all beds that are decade or so have targeted the hospital sector, as it regularly maintained and staffed and are imme- remains the largest health spending category in most diately available for use. They include beds in European countries. general hospitals, mental health and substance abuse hospitals, and other specialty hospitals. In 2008, Germany and Austria had the highest Beds in nursing and residential care facilities are number of hospital beds per capita, with about 8 beds excluded. per 1 000 population (Figure 3.6.1). The high supply of hospital beds in these two countries is associated with Curative care beds are beds accommodating a large number of hospital admissions/discharges, as patients where the principal intent is to do well as long average length of stays in Germany. one or more of the following: manage labour Turkey had the lowest number of beds per capita, (obstetric), cure non-mental illness or provide followed by Spain, the United Kingdom and Portugal. definitive treatment of injury, perform surgery, relieve symptoms of non-mental illness or injury Two-thirds of hospital beds are allocated for cura- (excluding palliative care), reduce severity of tive care on average across EU countries. The rest of non-mental illness or injury, protect against the beds are allocated for psychiatric (14%), long-term exacerbation and/or complication of non- (10%) and other types of care (8%). In some countries, mental illness and/or injury which could the share of beds allocated for psychiatric care and threaten life or normal functions, perform long-term care is much greater than the average. In diagnostic or therapeutic procedures. Finland, a greater proportion of hospital beds are allocated for long-term care (35%) than for curative Psychiatric care beds are beds accommodating care (30%). This is because local governments in patients with mental health problems. They Finland are responsible for managing both health and include beds in psychiatric departments of long-term care services, and use hospitals to provide general hospitals, and all beds in mental health at least some of the institution-based long-term care and substance abuse hospitals. (OECD, 2005). In Ireland, just over half of hospital beds Long-term care beds are hospital beds accom- are allocated for acute care, with 30% devoted to modating patients requiring long-term care due long-term care (Figure 3.6.2). to chronic impairments and a reduced degree of The share of beds in private for-profit hospitals independence in activities of daily living. They has increased in some countries over the past few include beds in long-term care departments of years, while it has remained stable in others. In general hospitals, beds for long-term care in Germany, the share increased from 23% of all beds specialty hospitals, and beds for palliative care. in 2002 to 29% in 2008, accompanied by a decline in 86 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 0 2 4 6 8 10 12 Ge 0 20 40 60 80 100 % Ge rm rm an 2.3 an 9.7 Au y Au y 8.2 st st ria ria 8.5 La 7.7 La tv tv ia 11.1 Per 1 000 population ia Cz Cz ec M 7.4 ec M h a h a Re l t a n.a. Re l t a pu 7.4 pu b b Hu l i c 8.5 Hu l i c ng 7.3 ng ar n.a. ar Fr y an 7.0 Fr y an Li c 8.5 Li c th e 6.9 Curative care beds th e ua ua n 10.9 Be ia Be a ni lg 6.8 iu lg Sl m n.a. HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 iu 6.7 Sl m ov Po ak la ov Po ak la Re nd n.a. Re nd pu 6.6 pu Ro b l i c 8.3 b m 6.6 Ro l i c an m 7.6 an Fi ia 6.6 Source: OECD Health Data 2010; Eurostat Statistics Database. Source: OECD Health Data 2010; Eurostat Statistics Database. nl Fi ia a nl Bu nd 8.1 a lg 6.5 Bu nd ar 1995 Lu lg ia 10.3 xe ar i L u Ic el 6.5 m a xe a n n.a. Psychiatric care beds bo m d ur bo 5.8 g ur g n.a. EU 5.8 Es EU 7.3 to Es 5.7 ni to I a ni 8.3 I a S w r ela S w r ela 5.7 i t z nd i t z nd 7.0 er la er 5.2 n la n 6.6 2008 Gr d ee Gr d 5.2 ee Sl c e Sl c e 4.9 Ne ove N e ove 4.8 t h ni a t h ni a 5.8 er er la Long-term care beds la 4.8 nd nd s s 5.3 It a 4.3 It a ly 6.3 Countries ranked by declining order of hospital beds per 1 000 population 3.8 Cy C y ly p p De r us De r us 5.1 nm nm 3.8 a a 4.6 No r k 3.6 No r k rw rw 4.0 3.6.2. Hospital beds by function of health care, 2008 (or nearest year available) Un P ay Un P ay i te or t i t e or t 3.5 u d u 3.6.1. Hospital beds per 1 000 population, 1995 and 2008 (or nearest year available) d K i gal 3.9 K i gal ng ng 3.4 do do m m n.a. Sp 3.4 Other hospital beds Sp ai ai 3.9 n Tu n 3.3 Tu rk rk ey n.a. ey 2.3 3.6. HOSPITAL BEDS 87 1 2 http://dx.doi.org/10.1787/888932336920 1 2 http://dx.doi.org/10.1787/888932336901 3.7. HOSPITAL DISCHARGES Hospital discharges measure the number of peo- overnight stays in hospitals to same-day procedures. In ple who need to stay overnight in a hospital each year. the group of countries where discharge rates have Together with the average length of stay, they are decreased over the past decade, the reduction can be important indicators of hospital activities. Hospital explained at least partly by a strong rise in the number activities are affected by a number of factors, includ- of day surgeries (see Indicator 3.10, for example, for ing the demand for hospital services, the capacity of evidence on the rise in day surgeries for cataracts). hospitals to treat patients, the ability of the primary Lithuania has the highest discharge rate for care sector to prevent avoidable hospital admissions, circulatory diseases, followed by Latvia, Bulgaria, and the availability of post-acute care settings to Germany and Austria (Figure 3.7.2). The high rates in provide rehabilitative and long-term care services. Lithuania, Latvia and Bulgaria are associated with In 2008, hospital discharge rates were the highest high mortality rates from circulatory diseases, which in Austria and France, although the high rate in France may also be used as a proxy indicator for the occur- is partly explained by the inclusion of some same-day rence of these diseases (see Indicator 1.4). This is not separations (Figure 3.7.1). Discharge rates were also the case however for Germany and Austria, suggesting high in Bulgaria, Germany and Romania. They were that different clinical practices may play a role. the lowest in Cyprus, Malta and Turkey. Austria and Germany have the highest discharge In general, countries that have a greater number rates for cancer, followed by Hungary (Figure 3.7.3). of hospital beds also tend to have higher discharge While the high rate in Hungary is associated with a rates. For example, the number of hospital beds per high mortality rate from cancer (which may also be capita in Austria and Germany is more than twice used as a proxy for the occurrence of the disease; see than Spain and the United Kingdom, and discharge Indicator 1.5), this is not the case for Austria and rates are also twice as large (see Indicator 3.6). Germany. In Austria, the high rate is associated with a Trends in hospital discharge rates vary widely high rate of hospital readmissions for further investi- across European countries. In about one-third of gation and treatment of cancer patients (European EU countries (including Austria, Germany and Greece), Commission, 2008a). discharge rates have increased over the past ten years. In a second group of countries (including Belgium, France, Spain, Sweden and the United Kingdom), they have remained stable, while in the third group (includ- Definition and deviations ing Denmark, Finland and Italy), discharge rates fell Discharge is defined as the release of a patient between 1998 and 2008. who has stayed at least one night in hospital. Trends in hospital discharges may reflect several It includes deaths in hospital following in- factors that are not easily disentangled. Demand for patient care. Same-day separations are usually hospitalisation may grow as populations age, since excluded, with the exception of France and the older population groups account for a disproportion- Slovak Republic which include some same-day ately high percentage of hospital discharges in all separations. countries. For example, in Austria and Germany, 42% of Healthy babies born in hospitals are excluded all hospital discharges in 2008 were for people aged 65 completely (or almost completely) from hospital and over, more than twice their share of the population discharge rates in several countries (e.g. Austria, (17% and 20% respectively). However, population Cyprus, Estonia, Finland, Greece, Ireland, Latvia, ageing alone may be a less important factor in explain- Luxembourg, Malta, Norway, Spain, Sweden, ing trends in hospitalisation rates than changes in Turkey), resulting in an under-estimation of medical technologies and clinical practices. A signifi- 3-6% of all discharges. cant body of research shows that the diffusion of new medical interventions gradually extends to older Some countries do not cover all hospitals. For population groups, as interventions become safer and instance, data for Denmark, Ireland and the more effective for people at older ages (e.g. Dormont United Kingdom are restricted to public or and Huber, 2006). The diffusion of new medical publicly-funded hospitals only. Data for Portugal technologies may also involve a reduction in hospitali- relate only to public hospitals on the mainland. sation if it entails a shift from procedures requiring 88 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.7. HOSPITAL DISCHARGES 3.7.1. Hospital discharges per 1 000 population, 2008 (or nearest year available) Per 1 000 population 300 267 264 250 239 232 228 222 208 203 202 200 191 189 184 178 175 173 172 172 169 165 165 160 159 155 150 140 136 133 120 113 109 106 100 95 75 50 0 Fr i a 1 Bu c e 2 Re v i a 1 Fi e 1 Hu d 1 a1 P y1 m n1 Sl r g 1 K i nd 1 Tu n 1 ey 1 Cy ta 1 us 1 ic 2 Ge ar i a Ro a n y Gr l i c ov L i t ni a pu i a Es r y Be EU N o um i t z nd Lu we d D e ni a Ic r k i t e Ir e l d m Po l y er l Sp s th ga nd S an an an It a a ni c do a ai an xe d e a b r S w ol a al u e a Ne r tu rk rw ee an bl nm pr i ng a st rm pu la l el ec L at to lg lg ov m a k hu bo er M nl ng Au Re h d Cz Un Sl 1. Excludes discharges of healthy babies born in hospital (between 3-6% of all discharges). 2. Includes same-day separations. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336939 3.7.2. Hospital discharges for circulatory diseases 3.7.3. Hospital discharges for cancers per 1 000 population, 2008 (or nearest year available) per 1 000 population, 2008 (or nearest year available) Lithuania 42 Austria 29 Latvia 38 Germany 24 Bulgaria 35 Hungary 23 Germany 35 Latvia 19 Austria 34 Poland 19 Estonia 33 France 19 Slovak Republic 32 Greece 19 Hungary 31 Slovak Republic 18 Romania 31 Romania 18 Czech Republic 29 Slovenia 18 Finland 28 Finland 17 Greece 28 Estonia 17 Sweden 26 Lithuania 17 Poland 26 Czech Republic 17 Norway 25 Norway 17 EU 24 Luxembourg 16 Luxembourg 22 EU 15 France 21 Bulgaria 15 Italy 21 Sweden 14 Belgium 21 Denmark 13 Slovenia 19 Iceland 13 Denmark 19 Italy 12 Switzerland 17 Belgium 12 Netherlands 16 Switzerland 11 Iceland 15 Netherlands 10 Portugal 14 Portugal 10 Spain 14 Spain 10 United Kingdom 13 United Kingdom 9 Ireland 12 Ireland 9 Turkey 11 Turkey 6 Malta 9 Cyprus 5 Cyprus 9 Malta 4 0 15 30 45 0 10 20 30 Per 1 000 population Per 1 000 population Source: OECD Health Data 2010; Eurostat Statistics Database. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336958 1 2 http://dx.doi.org/10.1787/888932336977 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 89 3.8. AVERAGE LENGTH OF STAY IN HOSPITALS The average length of stay in hospitals is often Focusing on average length of stay for specific regarded as an indicator of efficiency, since a shorter diseases or conditions can remove some of the hetero- stay may reduce the cost per discharge and shift care geneity arising from different mix and severity from in-patient to less expensive post-acute settings. of conditions treated in hospitals across countries. However, shorter stays tend to be more service inten- Figure 3.8.3 shows that the average length of stay for a sive and more costly per day. Too short a length of stay normal delivery ranges from less than two days in Tur- could also have adverse effects on health outcomes, key and the United Kingdom, to over five days in the Slo- or reduce the comfort and recovery of the patient. vak Republic, Romania, Hungary and Switzerland. The If this leads to a rising readmission rate, costs per length of stay for a normal delivery has become shorter episode of illness may fall little, or even rise. in nearly all countries over the past decade, dropping In all European countries, the average length of from five days in 1995 to less than four days in 2008 on stay in hospitals has decreased over the past decade, average in EU countries. falling from 8.3 days in 2000 to 7.2 days in 2008 on Lengths of stay following acute myocardial infarc- average (Figure 3.8.1). Several factors explain this tion (AMI, or heart attack) also declined over the general decline, including the use of less invasive past fifteen years. In 2008, it was the lowest in Turkey surgical procedures, changes in hospital payment and some Nordic countries (Norway, Denmark and methods, and the expansion of early discharge Sweden). At the other end of the scale, it was highest in programmes enabling patients to return to their home Germany, Lithuania, Finland and Estonia (Figure 3.8.2). to receive follow-up care. The reduction in average In this latter group of countries, long average length length of stay was particularly marked in Switzerland of stays may be due to the fact that some patients (which had the highest length of stay in 2000), originally admitted for AMI are no longer receiving Bulgaria and the Netherlands. In Switzerland, the pro- acute care, but nonetheless stay in hospitals for a gressive move from bed-day payments to DRG-based certain period to receive post-acute care. payments has contributed to the reduction in average length of stay in those cantons that have modified their payment system (OECD and WHO, 2006). In 2008, the average length of stay in hospitals was the lowest in Turkey, Malta, and in several Nordic Definition and deviations countries (Norway, Denmark, Iceland, Sweden). It was the highest in Finland, followed by Switzerland and Average length of stay (ALOS) refers to the Germany. The high average length of stay in Finland is average number of days that patients spend in due to a large proportion of beds allocated for conva- hospital. It is generally measured by dividing the lescent patients and long-term care (see Indicator 3.6). total number of days stayed by all in-patients Focusing only on stays in acute care units, the average during a year by the number of admissions or length of stay in Finland is not greater, indeed is even discharges. Day cases are excluded. lower than in most other European countries. 90 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.8. AVERAGE LENGTH OF STAY IN HOSPITALS 3.8.1. Average length of stay in hospital for all causes, 2000 and 2008 (or nearest year available) Days 2000 2008 14 12.6 13.7 12.4 11.8 12 10.7 10.5 10 9.9 9.8 9.3 9.3 8.9 8.8 8.7 8 8.4 8.4 8.4 8.3 8.0 7.8 7.7 7.7 7.7 7.7 7.6 7.5 7.5 7.5 7.3 7.3 7.2 7.1 7.0 7.0 7.0 6 6.8 6.8 6.8 6.4 6.4 6.4 6.3 6.2 6.2 6.2 6.1 6.1 6.0 6.0 6.0 5.9 5.7 5.7 5.6 5.6 5.4 5.2 4 4.9 4.8 4.3 0 i t z nd Ge and K i lic ta Tu y ey La y Li t via B e ni a Un k R ium C z Ro o m E lic Re ni a ( 2 em ni a un r g Gr ) ce Au n ria Bu l y Sl r i a Ir e i a nd th r us Hu ds Fr y Sw ce Ic n Po d Po nd nm l k De ga es a an ar ar ai e an en It a al c o ou rk rw ee an d ub b n ed S w inl a la la st a E U Lu x s to ua a Sp r tu d ng Ne C yp tri rm pu l la M el a lg lg ov ec m er ng b i te ep No th er F h 2 ov Sl Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932336996 3.8.2. Average length of stay following acute 3.8.3. Average length of stay myocardial infarction (AMI), 2008 for normal delivery, 2008 (or nearest year available) (or nearest year available) Germany 10.9 Slovak Republic 5.5 Lithuania 10.8 Romania 5.3 Finland 10.2 Hungary 5.2 Estonia 10.1 Switzerland 5.2 Greece 10.0 Cyprus 5.0 Latvia 10.0 Czech Republic 4.8 Ireland 9.0 France 4.4 United Kingdom 8.8 Poland 4.4 Romania 8.6 Belgium 4.3 Spain 8.3 Bulgaria 4.3 Portugal 8.2 Austria 4.2 Switzerland 8.1 Luxembourg 4.1 EU 7.9 Greece 4.0 Italy 7.9 Slovenia 3.9 Belgium 7.8 Latvia 3.9 Slovenia 7.8 Lithuania 3.8 Cyprus 7.8 EU 3.7 Austria 7.6 Italy 3.5 Malta 7.6 Finland 3.4 Iceland 7.0 Germany 3.3 Hungary 6.9 Norway 3.2 Czech Republic 6.8 Denmark 2.8 Luxembourg 6.7 Portugal 2.7 Netherlands 6.7 Spain 2.6 Poland 6.4 Malta 2.6 France 6.2 Sweden 2.3 Slovak Republic 5.6 Ireland 2.1 Sweden 4.8 Netherlands 2.1 Denmark 4.7 Iceland 2.0 Norway 4.3 United Kingdom 1.8 Turkey 4.2 Turkey 1.4 4 6 8 10 12 1 2 3 4 5 6 Days Days Source: OECD Health Data 2010; Eurostat Statistics Database. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932337015 1 2 http://dx.doi.org/10.1787/888932337034 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 91 3.9. CARDIAC PROCEDURES (CORONARY ANGIOPLASTY) Heart diseases are a leading cause of hospitalisa- ischemic heart diseases; ii) differences in the capacity tion and death in OECD countries (see Indicator 1.4). to deliver and pay for these procedures; iii) differences Coronary angioplasty is a revascularisation procedure in clinical treatment guidelines and practices; and that has revolutionised the treatment of ischemic iv) coding and reporting practices. heart diseases over the past twenty years. It involves The large variations in the number of revascular- the threading of a catheter with a balloon attached to isation procedures across countries do not seem to be the tip through the arterial system, usually started closely related to the incidence of ischemic heart in the femoral artery in the leg, into the diseased disease (IHD), as measured by IHD mortality (see coronary artery. The balloon is inflated to distend the Figure 1.4.1). IHD mortality in Germany is lower than coronary artery at the point of obstruction. The place- the average across EU countries, but Germany has the ment of a stent to keep the artery open accompanies highest rate of revascularisation procedures. On the the majority of angioplasties. Drug-eluting stents (a other hand, IHD mortality in Finland is above the EU stent that gradually releases drugs) are increasingly average, while revascularisation rates are below being used to stem the growth of scar-like tissue average. surrounding the stent. Coronary angioplasty is an expensive interven- There is considerable variation across European tion, although it is much less costly than a coronary countries in the us e of c oronary ang ioplasty bypass because it is less intrusive. In 2007, the average (Figure 3.9.1). Germany and Belgium have the highest estimated price of an angioplasty was about EUR 6 000 rates of angioplasty in 2008, followed by Italy and in France, EUR 8 000 in Sweden and EUR 8 600 in Italy. Norway. In Belgium, the high rate of coronary Nonetheless, the estimated price of an angioplasty in angioplasty can only be partly attributed to patient Italy remains 30% lower than in the United States mobility. In 2006, only 2.5% of people who received an (Koechlin et al., 2010). angioplasty on an in-patient basis in Belgium were non-residents (European Commission, 2008a). The rate of use of angioplasty is the lowest in the Netherlands and Switzerland, although these two countries report only the main procedure (not Definition and deviations all procedures), resulting in a significant under- The data relate to in-patient procedures, estimation (see box on definition). normally counting all procedures. However, The use of angioplasty has increased rapidly classification systems and registration practices since 1990 in most OECD countries, overtaking vary across countries, and the same procedure coronary bypass surgery as the preferred method of can be recorded differently (e.g. an angioplasty revascularisation around the mid-1990s – about the with the placement of a stent can be counted as same time that the first published trials of the efficacy one or two procedures). Some countries report of coronary stenting began to appear (Moïse, 2003). In only the main procedure (or the number of most European countries, angioplasty now accounts patients receiving one or more procedures), for at least 70% of all revascularisations (Figure 3.9.2). resulting in a significant under-estimation of Although angioplasty has replaced in many cases the total number. This is the case for the bypass surgery, it is not a perfect substitute since Netherlands and Switzerland. In Ireland, the bypass surgery is still the preferred method for data only include activities in publicly-funded treating patients with multiple-vessel obstructions, hospitals (it is estimated that over 10% of all diabetes and other conditions (Taggart, 2009). hospital activity in Ireland is undertaken in A number of reasons can explain cross-country private hospitals). For all countries, the data do variations in the number of revascularisation proce- not include coronary angioplasties performed dures in general and angioplasty in particular, includ- on an ambulatory basis. ing: i) differences in the incidence and prevalence of 92 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.9. CARDIAC PROCEDURES (CORONARY ANGIOPLASTY) 3.9.1. Coronary angioplasty per 100 000 population 2008 (or nearest year available) Change 1998-2008 (or nearest year available) 140 Netherlands 9.8 141 Switzerland n.a. 141 Portugal 15.8 142 Finland 12.9 143 Luxembourg 5.6 147 Estonia n.a. 166 Denmark 10.9 169 Greece n.a. 170 Hungary n.a. 173 Sweden 16.5 178 Ireland 10.6 185 Slovenia n.a. 189 France 4.8 212 Poland 24.4 224 EU 12.2 232 Austria n.a. 235 Spain 13.3 236 Iceland 3.6 248 Czech Republic n.a. 287 Norway 13.7 384 Italy 13.4 427 Belgium 8.7 568 Germany n.a. 600 400 200 0 0 10 20 30 Per 100 000 population Average annual growth rate (%) Note: Some of the variations across countries are due to different classification systems and recording practices. Source: OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932337053 3.9.2. Coronary angioplasty as a percentage of total revascularisation procedures, 1998-2008 Belgium France Denmark Finland Italy Netherlands Norway Sweden % of total revascularisation procedures % of total revascularisation procedures 90 90 80 80 70 70 60 60 50 50 40 40 30 30 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 Note: Revascularisation procedures include coronary bypass and angioplasty. Source: OECD Health Data 2010. 1 2 http://dx.doi.org/10.1787/888932337072 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 93 3.10. CATARACT SURGERIES In the past 20 years, the number of surgical pro- thetists (Castoro et al., 2007), together with limitations cedures carried out on a day care basis has steadily in data coverage. In France, the share of cataract grown in European countries. Advances in medical surgeries carried out on a same-day basis has technologies, particularly the diffusion of less inva- increased rapidly over the past decade, from 23% sive surgical interventions, and better anaesthetics in 1998 to 70% in 2007, but it still remains below that have made this development possible. These innova- of many other European countries. tions have improved effectiveness and patient safety. In Sweden, there is evidence that cataract surger- They also help to reduce the unit cost of interventions ies are now being performed on patients suffering by shortening the length of stay. However, the overall from less severe vision problems compared to five or impact on cost depends on the extent to which any ten years ago. This raises the question of how the greater use of these procedures may be offset by a needs of these patients should be prioritised relative reduction in unit cost, taking into account the cost of to other patient populations (Swedish Association of post-acute care and community health services. Local Authorities and Regions and National Board of Cataract surgery provides a good example of a Health and Welfare, 2008). high volume surgery which is now carried out predominantly on a day care basis in most European countries. It has become the most frequent surgical procedure in many European countries. Definition and deviations The number of cataract surgeries per capita ranges Cataract surgeries consist of removing the lens from a low of about 200 surgeries per 100 000 population of the eye because of the presence of cataracts in Cyprus to a high of 1 848 per 100 000 population in which are partially or completely clouding the Belgium (Figure 3.10.1). Both demand factors (including lens, and replacing it with an artificial lens. The an older population structure) and supply factors (such surgery may be carried out as day cases or as as the capacity to perform the intervention in hospital in-patient cases (involving an overnight stay in and outside hospital) provide explanations for these hospital). Same-day interventions may either be cross-country variations. However, the comparability of performed in a hospital or in a clinic. However, data is also limited by registration procedures, parti- the data for many countries (e.g. Ireland, cularly the lack of registration of day surgeries carried Hungary, Poland) only include interventions outside hospitals in some countries, which explain the carried out in hospitals. Caution is therefore low rates in Ireland, Poland and Denmark. The very high required in making cross-country comparisons rate in countries such as Belgium may be explained of available data, given the incomplete coverage partly by the registration of more than one procedure of day surgeries in several countries. per surgery. Denmark only includes cataract surgeries Cataract surgeries are now predominantly carried out in public hospitals, excluding proce- performed on a day care basis in many European dures carried out in the ambulatory sector and countries. Day surgery accounts for 90% or more of all in private hospitals. In Ireland too, the data cataract surgeries in about half of the countries for cover only procedures in public hospitals (it is which data are available, including in the Nordic estimated that over 10% of all hospital activity countries, the Netherlands, the United Kingdom and in Ireland is undertaken in private hospitals). Spain (Figure 3.10.2). However, the diffusion of day The data for Spain only partially include the surgery is still relatively low in some countries, activities in private hospitals. such as Cyprus, Lithuania, Poland and Hungary. In Luxembourg, only 35% of all cataract surgeries were Classification systems and registration prac- carried out as day cases in 2007, a modest increase tices for cataract surgeries vary across countries, compared with the rate in the late 1990s. The small for instance whether they are counted as one share of day surgeries in these countries may be intervention involving at least two steps (removal explained by more advantageous reimbursement for or the lens and replacement with an artificial in-patient stays, national regulations, and obstacles to lens) or as two separate interventions. changing individual practices of surgeons and anaes- 94 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.10. CATARACT SURGERIES 3.10.1. Number of cataract surgeries, in-patient and day cases, per 100 000 population, 1998 and 2008 (or nearest year available) Day cases in 2008 In-patient cases in 2008 In-patient and day cases in 1998 Per 100 000 population 2 000 1 848 1 500 1 205 1 141 1 000 944 896 864 807 790 786 777 773 765 676 674 651 563 475 500 450 421 410 393 351 2.3 193 0 ce ly ic ay k nd nd a nd us m n l g s en d a EU y d ia ga m nd ni ni ar ar ai an an ur bl en It a iu rw an ed pr la la la do to Sp ua r tu nm ng bo pu la nl el lg ov Ir e Po er Cy Fr Sw No Es ng th er Ic Fi Be Hu m Po Re it z De Sl th Li Ki xe Sw h Ne Lu d ec i te Cz Un Note: Some of the variations across countries are due to different classification systems and recording practices. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932337091 3.10.2. Share of cataract surgeries carried out as day cases, 1998 and 2008 (or nearest year available) 1998 2008 % of total cataract surgeries 100 95 95 94 93 99 99 96 98 97 97 97 87 84 84 84 80 79 80 71 71 70 67 66 65 65 63 60 40 35 34 28 23 21 20 18 20 11 7 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1 0 0 a d ic s en k ay m n d m l ly nd nd ce EU g ia y nd a us ga nd ni ni ar ar ai an an ur bl en It a iu do rw an ed pr la la la to Sp ua r tu nm ng bo pu la nl el lg ov Ir e Po er ng Cy Fr Sw No Es th er Ic Fi Be Hu m Po Re it z De Sl Ki th Li xe Sw h Ne d Lu ec i te Cz Un Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932337110 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 95 3.11. HIP AND KNEE REPLACEMENT Significant advancements in surgical treatment growing number of people over 60 and 65 years with a have provided effective options to reduce the pain and greater risk of suffering from osteoarthritis (even if disability associated with certain musculoskeletal the age and sex specific rate does not increase); and conditions. Joint replacement surgery (hip and knee 2) the growing prevalence of obesity, which is the replacement) is considered the most effective inter- main risk factor for osteoarthritis beyond age and sex vention for severe osteoarthritis, reducing pain and (European Commission, 2008b). disability and restoring some patients to near normal The number of hip and knee replacement has function. increased rapidly over the past ten years in most Ostheoarthritis is one of the ten most disabling European countries (Figures 3.11.3 and 3.11.4). On diseases in developed countries. Worldwide estimates average, the number of hip replacement increased are that 9.6% of men and 18.0% of women aged over by one-third between 1998 and 2008. The growth 60 years have symptomatic osteoarthritis, including rate was even higher for knee replacement, which moderate and severe forms (WHO, 2010a). Age is the more than doubled during this ten-year period. For strongest predictor of the development and progres- example, in the United Kingdom, hip replacement rate sion of osteoarthritis. It is more common in women, increased by 40% since 2000, while knee replacement increasing after the age of 50 especially in the hand increased by 112%. and knee. Other risk factors include obesity, physical A hip or knee replacement is an expensive inter- inactivity, smoking, excess alcohol and injuries vention, although the cost varies across countries. (European Commission, 2008b). While joint replace- In 2007, the average estimated price of a knee replace- ment surgery is mainly carried out among people ment in France was EUR 10 600, about 20-25% more aged 60 and over, it can also be performed among than in Finland, Germany, Portugal and Sweden. people at younger ages. Nonetheless, the estimated price of a knee replace- There is considerable variation across countries ment in France remained 15-20% lower than in the in the rate of hip and knee replacement (Figures 3.11.1 United States (Koechlin et al., 2010). and 3.11.2). Germany, Austria, Belgium, Norway and Switzerland have the highest rates of hip replace- ment. These countries are also amongst those that have the highest rates of knee replacement. A number Definition and deviations of reasons can explain these cross-country variations Hip replacement is a surgical procedure in in the rate of hip and knee replacement, including: which the hip joint is replaced by a prosthetic i) differences in the prevalence of osteoarthritis implant. It is generally conducted to relieve problems; ii) differences in the capacity to deliver and arthritis pain or treat severe physical joint pay for these expensive procedures; iii) differences in damage following hip fracture. clinical treatment guidelines and practices; and iv) international mobility of patients across borders Knee replacement is a surgical procedure to (e.g. in Belgium, about 2% of knee replacement are replace the weight-bearing surfaces of the knee performed on people who are not residing in the joint to relieve the pain and disability of osteo- country; European Commission, 2008a). arthritis. It may be performed for other knee diseases such as rheumatoid arthritis. There are too few comparable studies on the prevalence of osteoarthritis in Europe to draw any Classification systems and registration conclusions on cross-country variations. Nor is there practices vary across countries, which may any evidence as to whether the age- and sex-specific affect the comparability of the data. In Ireland, incidence of osteoarthritis has changed in recent the data only include activities in publicly- decades. However, the number of people suffering funded hospitals (it is estimated that over 10% of from osteoarthritis has increased, and is expected to all hospital activity in Ireland is undertaken in continue to increase in the coming years, for two private hospitals). reasons: 1) population ageing, which is resulting in a 96 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.11. HIP AND KNEE REPLACEMENT 3.11.1. Hip replacement surgery, 3.11.2. Knee replacement surgery, per 100 000 population, 2008 per 100 000 population, 2008 (or nearest year available) (or nearest year available) Germany 289 Germany 206 Austria 243 Austria 187 Belgium 240 Finland 184 Norway 231 Switzerland 179 Switzerland 226 France 220 Belgium 168 Luxembourg 217 Luxembourg 155 Sweden 207 United Kingdom 146 Netherlands 205 Netherlands 119 United Kingdom 195 Iceland 119 Finland 195 Slovenia 189 France 114 Denmark 174 Sweden 110 Iceland 165 EU 107 Italy 154 Spain 106 EU 153 Denmark 106 Greece 140 Italy 97 Ireland 126 Latvia 107 Slovenia 79 Spain 96 Latvia 61 Hungary 96 Portugal 54 Estonia 91 Hungary 47 Portugal 85 Cyprus 46 Romania 46 Poland 39 Ireland 45 Cyprus 15 Romania 5 0 100 200 300 0 100 200 300 Per 100 000 population Per 100 000 population Source: OECD Health Data 2010; Eurostat Statistics Database. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932337129 1 2 http://dx.doi.org/10.1787/888932337148 3.11.3. Trend in hip replacement surgery, 3.11.4. Trend in knee replacement surgery, 1998 to 2008 (or nearest year available), 1998 to 2008 (or nearest year available), selected countries selected countries Belgium Portugal Spain Ireland Italy Netherlands United Kingdom EU (14 countries) Switzerland EU (12 countries) Per 100 000 population Per 100 000 population 250 200 200 150 150 100 100 50 50 0 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 Source: OECD Health Data 2010; Eurostat Statistics Database. Source: OECD Health Data 2010; Eurostat Statistics Database. 1 2 http://dx.doi.org/10.1787/888932337167 1 2 http://dx.doi.org/10.1787/888932337186 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 97 3.12. SCREENING, SURVIVAL AND MORTALITY FOR CERVICAL CANCER Cervical cancer is largely preventable. Screening cancer mortality rates reflect the effect of care in past by regular pelvic exam and pap smears can identify years and changes in incidence. Mortality rates premalignant lesions, which can be effectively treated for cervical cancer are higher in eastern European before the occurrence of the cancer. Regular screening countries (Figure 3.12.3). Between 1998 and 2008 the also increases the probability of diagnosing early rates declined for most European countries, with stages of the cancer and improving survival. Conse- larger improvements for Iceland, Denmark, Slovenia, quently, the Council of the European Union and the the Czech Republic and Norway. European Commission promote population based cancer screening programmes among member states (European Union, 2003; European Commission, 2008c) and European countries have instituted screening programmes with specific periodicity and target Definitions and deviations groups. In addition, promising cancer preventing Screening rates for cervical cancer reflect the vaccines have been developed based on the discovery proportion of women who are eligible for a that cervical cancer is caused by sexual transmission screening test and actually receive the test. As of certain forms of the Human Papilloma Virus. The policies regarding screening periodicity differ efficacy and safety of those vaccines is now well across countries, the rates are based on each established, but debates about cost-effectiveness country’s specific policy. An important consider- and the implications of vaccination programmes for ation is that some countries ascertain screening teenagers for a sexually transmitted disease continue based on surveys and other based on encounter in a number of countries (Huang, 2008). data, which may influence the results. If a Screening rates vary widely across countries with country has an organised screening programme, Austria, Norway, the United Kingdom and Sweden but women receive care outside the programme, achieving coverage of around 80% of the target popula- rates may be underreported. Survey-based tion (Figure 3.12.1). Some countries with very low results may also underestimate the rates due to screening rates, like Turkey and Latvia, did not have recall bias. uniform national screening programme as of 2008; Relative cancer survival rates reflect the the low rates reflect local programmes or opportu- proportion of patients with a certain type of nistic screening. Screening rates in several countries cancer who are still alive after a specified time declined slightly between 2000 and 2008. period (commonly five years) compared to those Relative survival rates are commonly used to track still alive in absence of the disease. Relative progress in treating cancer over time as they reflect survival rates capture the excess mortality that both how early the cancer was detected and the can be attributed to the diagnosis. For example, effectiveness of the treatment provided. Survival rates a relative survival rate of 80% does not mean have been used to compare European countries in the that 80% of the cancer patients are still alive EUROCARE study, in comparisons between European after five years, but that 80% of the patients that countries and the United States (Gatta et al., 2000), and were expected to be alive after five years, given in national reporting activities in many countries. their age at diagnosis, are in fact still alive. All Nearly all countries recorded five-year relative survival the survival rates presented here have been age- rates above 60% for the period 2002-07. The rates standardised using the International Cancer rang ed from 71% in Iceland to 50% in Poland Survival Standard (ICSS) population. The (Figure 3.12.2). Over the periods 1997-2002 and 2002-07, survival rates are not adjusted for tumor stage at the five-year relative rates improved in most countries, diagnosis, hampering assessment of the relative although in all instances the increase is not statistically impact of early detection and better treatment. significant. The definition of cancer mortality rates is Mortality rates alone are not sufficient to draw provided under Indicator 1.5. timely inferences about quality of care, but current 98 HEALTH AT A GLANCE: EUROPE 2010 © OECD 2010 3.12. SCREENING, SURVIVAL AND MORTALITY FOR CERVICAL CANCER 3.12.1. Cervical cancer screening, 3.12.2. Cervical cancer five-year relative percentage of women screened aged 20-69, survival rate, 1997-2002 and 2002-07 2000 to 2008 (or nearest year) (or nearest period) 2000 2004 2008 1997-2002 2002-07 Austria 2 81.5 71.0