Article Text

Download PDFPDF

Research report
Inequalities in health expectancies at older ages in the European Union: findings from the Survey of Health and Retirement in Europe (SHARE)
  1. Carol Jagger1,2,
  2. Claire Weston2,
  3. Emmanuelle Cambois3,
  4. Herman Van Oyen4,
  5. Wilma Nusselder5,
  6. Gabriele Doblhammer6,
  7. Jitka Rychtarikova7,
  8. Jean-Marie Robine8,
  9. the EHLEIS team9
  1. 1Institute for Ageing and Health, Newcastle University, Newcastle upon Tyne, UK
  2. 2Department of Health Sciences, University of Leicester, Leicester, UK
  3. 3French Institute for Demographic Studies, INED, Paris, France
  4. 4Scientific Institute of Public Health, Brussels, Belgium
  5. 5Erasmus Medical Center, University Medical Center Rotterdam, the Netherlands
  6. 6Rostock University, Rostock, Germany
  7. 7Charles University, Prague, Czech Republic
  8. 8French Institute of Health and Medical Research, INSERM, Montpellier, France
  9. 9
  1. Correspondence to Carol Jagger, AXA Professor of Epidemiology of Ageing, Institute for Ageing and Health, Newcastle University, Campus for Ageing and Vitality, Newcastle upon Tyne NE4 5PL, UK; carol.jagger{at}


Background Life expectancy gaps between Eastern and Western Europe are well reported with even larger variations in healthy life years (HLY).

Aims To compare European countries with respect to a wide range of health expectancies based on more specific measures that cover the disablement process in order to better understand previous inequalities.

Methods Health expectancies at age 50 by gender and country using Sullivan's method were calculated from the Survey of Health and Retirement in Europe Wave 2, conducted in 2006 in 13 countries, including two from Eastern Europe (Poland, the Czech Republic). Health measures included co-morbidity, physical functional limitations (PFL), activity restriction, difficulty with instrumental and basic activities of daily living (ADL), and self-perceived health. Cluster analysis was performed to compare countries with respect to life expectancy at age 50 (LE50) and health expectancies at age 50 for men and women.

Results In 2006 the gaps in LE50 between countries were 6.1 years for men and 4.1 years for women. Poland consistently had the lowest health expectancies, however measured, and Switzerland the greatest. Polish women aged 50 could expect 7.4 years fewer free of PFL, 6.2 years fewer HLY, 5.5 years less without ADL restriction and 9.5 years less in good self-perceived health than the main group of countries (Austria, Belgium, Denmark, France, Germany, Italy, the Netherlands, Spain, Sweden).

Conclusions Substantial inequalities between countries were evident on all health expectancies. However, these differed across the disablement process which could indicate environmental, technological, healthcare or other factors that may delay progression from disease to disability.

  • Eastern Europe
  • health expectancy
  • pop health indicator

Statistics from

Request Permissions

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


Inequalities in life expectancy (LE) among European countries have been evident for some time, the major gaps being between the established EU countries (EU15) and the more recent EU members from Eastern Europe.1 However, mortality rates and LE are only indirect measures of the health of populations. With the advent of a new EU structural indicator on health, healthy life years (HLY), it has been possible to demonstrate even greater inequalities in HLY than in LE across Europe of 14.5 years for men and 13.7 years for women at age 50 between the EU25 countries.2 Furthermore countries with the longest LE were not necessarily those with the greatest HLY, suggesting that longer life and better health, indicative of compression of morbidity,3 4 may not be universal.

HLY is a health expectancy based on a global activity limitation question, known as the GALI, one of three global health measures in the EU Statistics of Income and Living Conditions survey (EU-SILC). Though global measures have a utility in terms of encapsulating differences between countries, more detailed measures are then necessary to determine where differences might lie. Since the GALI is a measure of activity limitation, the most appropriate framework for drilling down to more detailed measures is the disablement process.5 6 Models of the disablement process identify disease or pathology as the initiating event, with a resulting limitation in body functions (physical, sensory or cognitive) which combine to produce restriction in instrumental activities of daily living (IADLs) and basic personal care activities of daily living (ADLs). The presence of functional limitations is a strong predictor of future activity restrictions and may therefore indicate ‘preclinical disability’.7 Passage between these stages may vary due to factors external to the individual (including environmental factors, medical treatments and interventions) as well as inter-individual factors (lifestyle and behaviour changes, reducing the frequency of activities or adopting coping mechanisms).5

Although disability-free life expectancies based on ADLs and IADLs have been calculated for a number of countries, there has been little harmonisation, making comparisons difficult.8 An exception is the Cross-National Determinants of Quality of Life and Health Services for the Elderly (CLESA), which compared disability free life expectancy (DFLE) in five European countries, but data was harmonised post-collection and covered differing periods in each country.9 Moreover health expectancies based on stages earlier in the disablement process are rare.

That different measures reflect different levels of ill-health/disability is not surprising, but a comparison of a range of more specific yet comparable health expectancies across countries may further elucidate the relationship between lengthening of life and health and bring further understanding to inequalities in HLY. The GALI, on which the HLY indicator is based, was included in the Survey of Health and Retirement in Europe (SHARE) alongside a wide range of other health measures. SHARE was first conducted in 2004 on individuals aged 50 years and over in 11 European countries: Germany, Austria, Belgium, Denmark, Spain, France, Greece, Italy, the Netherlands, Sweden and Switzerland. The SHARE survey was repeated in 2006 in the original countries and additionally in Poland and the Czech Republic. This paper uses SHARE wave 2 to evaluate inequalities in a range of health expectancies, including HLY and measures that more fully cover the disablement process, at older ages in Europe, specifically investigating whether those countries with the highest life expectancies spend the extra years in better health.


Health expectancies require data on mortality in the form of life tables and the age and sex specific prevalence of health states (health self-perception, activities limitation, IADL and ADL). We used individual country life tables for 2006 (the date of the SHARE survey) from the European Health Expectancy Monitoring Unit (EHEMU website: For the health data we used SHARE wave 2, release 1.01 (downloaded 1 April 2009). The sample size for SHARE ranged from 2000 to 3000 persons per participating country, representing the non-institutionalised population aged 50 and older.10

The SHARE main questionnaire consists of 20 modules (supplemented by a self-completion questionnaire). The detailed health measures on which health expectancies were based included chronic morbidity, physical functional limitations, the global activity limitation indicator (GALI), instrumental and basic self-care activities of activities of daily living score and self-perceived health. Further details of these measures are given in appendix 1 (supplementary material).

Statistical methods

Initial comparison of the health measures between the 13 countries was by means of prevalence, standardised by age and sex to the European standard population. Health expectancies were calculated using the Sullivan method.11 12 We calculated several health expectancies covering the disablement process: life expectancy free of chronic morbidity (DisFLE), LE without physical functional limitations (LE without PFL), healthy life years (HLY) based on the GALI question, LE without IADL restriction (LE without IADL), LE without ADL restriction (LE without ADL) and LE in good self-perceived health. To take into account the population living in institutions, excluded from general population surveys such as SHARE, we assumed that the prevalence of health states outside and within institutions does not differ.11

Hierarchical cluster analysis was used to explore similar groupings of the countries on LE at age 50 and on all the health expectancies for men and women, entering the years of life spent with each health measure (transformed into Z scores).


The mean age of the study population in each country ranged from 64.3 years in the Czech Republic to 67.1 years in Austria (table 1). There was a slight majority of women in all countries, from 53.0% of the population in Denmark to 58.9% in Austria.

Table 1

Characteristics of the Survey of Health and Retirement in Europe (SHARE) population aged 50 years and over, by country

Prevalence of ill-health and disability

Switzerland had the lowest age–sex standardised (to the European standard population) prevalence of ill-health for the six measures, while Poland had the highest standardised prevalence (table 2). Most countries reported a prevalence of morbidity between 60% and 70%. The highest incidence of physical functional limitations was reported in Greece, Poland and the Czech Republic (52.2%, 63.9% and 49.0%, respectively). Most countries reported a prevalence of activity limitation of 35–40%, though the difference between countries was more marked for severe activity limitation from a low of 5% (Greece, Spain) to 24.5% (Poland). IADL restrictions were mainly reported with prevalence in the range of 11–16%, while the prevalence of ADL restriction was in the range 7–9% for most countries: lowest in Switzerland (5.4%) and highest in Poland (19.4%) (table 2).

Table 2

Standardised* prevalence (%) of health conditions in the Survey of Health and Retirement in Europe (SHARE) population aged 50 years and over, by country (95% CIs in parentheses)

Life and health expectancy

There was a 6.1 year difference in LE at age 50 for men in 2006 between the 13 countries, from 24.8 years in Poland to 30.9 years in Switzerland (table 3). For women the difference was smaller at 4.1 years, from 31.3 years in the Czech Republic to 36.0 years in France (table 4). The prevalence of ill-health and the life tables by country were combined to produce health expectancies; these are presented for age 50, together with LE, in table 3 (men) and table 4 (women) along with the between country rankings. Men in the 13 countries could expect on average 23.9 years free of severe morbidity (three or more chronic conditions) at age 50 and 9.7 years free of any morbidity, though men in Poland had only 17.9 years free of severe morbidity (table 3). For women, LE free of severe morbidity tended to be longer than for men (on average 25.5 years) but more variable, with Polish women experiencing the fewest years free of severe morbidity (19.6 years).

Table 3

Life expectancy (LE) and health expectancies at age 50 for men, by country (rank between countries in parentheses, 1=lowest, 13=highest)

Table 4

Life expectancy (LE) and health expectancies at age 50 for women, by country (rank between countries in parentheses, 1=lowest, 13=highest)

LE free of physical functional limitations (PFL) showed little relationship to overall LE at age 50. Women tended to experience fewer years free of PFL than men (on average 13.8 years to men's 17.5 years), although in Poland women and men had considerably fewer years free of PFL: 7.1 (women) and 10.5 years (men) (tables 3 and 4).

LE free of activity restrictions is more commonly termed healthy life years (HLY). HLY at age 50 averaged 17.1 years for both men and women, with Greece ranking the highest for men (21.7 years) and Switzerland the highest for women (22.2 years), while Poland ranked lowest for both (men 10.5 years, women 11.1 years). LE free of IADL restrictions (average men 25.1 years; women 25.5 years) and ADL restrictions (average men 26.2 years; women 29.0 years) were longer compared to those based on other measures, reflecting the greater severity of ADL and IADL restrictions.

The more global health question of self-perceived health reflected the variability and range in the other measures of functioning and disability. Men aged 50 could expect to live around 19 years in good self-perceived health on average, while this was 1 year longer for women (20.2 years); in Poland this was only just over 10 years for both men and women.

LE at age 50 and the years of life free of each of the health measures were compared between the 13 countries. Cluster analysis identified five distinct groups: (1) Austria, Belgium, Denmark, France, Germany, Italy, the Netherlands, Spain, Sweden; (2) Greece; (3) Switzerland; (4) Czech Republic; and (5) Poland (table 5). Poland was observed to have the lowest LE for men (24.8 years) and the second lowest for women (31.5 years), and respondents reported the fewest years free of morbidity, PFL, IADL and ADL restrictions and in good self-perceived health. The Czech Republic also had a shorter than average LE by 3 years for men and 2.5 years for women, and reported fewer years free of morbidity, PFL and fewer HLY compared to the larger group of Western European countries, although years free of IADL and ADL restriction were only slightly lower.

Table 5

Mean life expectancy (LE) and years of life spent free of morbidity, impairment or disability at age 50 for men and women in each country grouping from cluster analysis

Switzerland reported the highest LE for men (30.9 years) and Swiss respondents experienced the most years free of morbidity, PFL, global activity limitation (women), IADL and ADL restriction and in good self-perceived health. Greece also had high LE at age 50 (29.8 years for men and 33.2 years for women), with above average years free of global activity restriction and IADL and ADL restriction, although fewer years free of PFL. The remaining countries (Austria, Belgium, Denmark, France, Germany, Italy, Netherlands, Spain, Sweden) had patterns of life and health expectancies at age 50 similar to the average.


Within the 13 countries included in SHARE wave 2 in 2006, LE at age 50 differed by 6 years for men and 4 years for women. When health expectancies covering the disablement process were compared, nine countries (Austria, Belgium, Denmark, France, Germany, Italy, Netherlands, Spain, Sweden) were grouped together, having similar profiles to the average on all measures. Greek individuals were also generally similar to this group of western countries but had more years spent with physical functional limitations. Poland consistently experienced the fewest healthy years, however measured, and Switzerland the most, although there seemed to be little relationship between length and quality of life over all countries. Polish women aged 50 could expect 7.4 years fewer free of PFL, 6.2 years fewer HLY, 5.5 years less without ADL restriction and 9.5 years less in good self-perceived health than the main group of countries (Austria, Belgium, Denmark, France, Germany, Italy, the Netherlands, Spain, Sweden). In the Czech Republic LE was also shorter, and years free of morbidity, activity limitation and in good or fair perceived health were substantially less than in other European countries though the gaps between years with IADL or ADL restrictions were much smaller.

The gap in LE between Eastern and Western European countries has already been noted.13 Poland and the Czech Republic therefore represent an interesting extension to the SHARE project with the major changes in the political and economic systems that have taken place in these countries. Both countries joined the European Union in 2004 but their level of economic prosperity is still much lower than the EU average. The results presented here confirm that the Czech Republic and Poland are also lagging behind in important health indicators. However, while Poland was uniformly worse on all health indicators, with respect to IADLs and ADLs, the Czech Republic was not as dissimilar to the Central European countries, suggesting that there are differences in ill-health between these countries. Previous studies have suggested that the higher mortality rates in Eastern Europe are due to a complex interaction between factors such as unfavourable socioeconomic circumstances, smoking, alcohol consumption, environmental pollution and ineffective medical care,14–18 and these no doubt contribute to the differences we found. In particular the purchasing power parity (PPP) adjusted GDP per capita in the Czech Republic is significantly higher than that of Poland; this is also related to the depth of recession experienced in these countries which happened earlier in Poland and was more pronounced than that in the Czech Republic. Moreover health insurance was introduced earlier in the Czech Republic than in Poland.19 There have been suggestions that LE in Poland could be increased by 6 months if hypertension and cerebrovascular disease were reduced.13 Whether this is due to differences in risk factors such as smoking, diet or alcohol consumption is unclear, but 6 months additional life would still result in a gap between Poland and the rest of Central Europe. This suggests that there are factors other than hypertension and cerebrovascular disease which lead to the decreased LE in Poland.

In contrast, LE at age 50 in Switzerland was 1.2 years greater for men and 0.9 years greater for women compared to the main group of Central European countries, and Switzerland experienced a substantially higher number of years free of most of the measures of ill-health and disability. In general, the Swiss healthcare system is regarded as excellent. Living in mountainous areas has been found to have a protective effect on total and coronary mortality,20 suggesting that environment is partly responsible for length of life and health. However, increased physical activity from walking on mountain terrains and moderate hypoxia may also explain these findings, and therefore current guidelines to improve levels of physical activity may be of wider benefit.

Ideas such as health or morbidity are difficult to define and since all the health measures we compared were self-reported, inequalities between countries may be due to cultural differences in reporting, particularly with the more subjective self-perceived health. For example, a Polish respondent may have a lower threshold of poor perceived health compared to a Swiss respondent. In many Mediterranean cultures it is the norm for older adults to expect their children to provide help with ADLs,21 and this can lead to under reporting of problems with activities of daily living while others, particularly older men, may withhold information about their difficulties in order to maintain their independence. Evidence also suggests that the availability of help and support influences self-perceived health.22 23 Though these cultural differences may account for some of the variation observed in this study, the activity limitation question has been validated against other subjective and more objective measures of disability and ill-health.24 However, further research is necessary to evaluate differences in self-reported measures of health and disability, and the influence of cultural norms.

The pattern of DFLE differed in Greece compared to the major group of European countries in that, though Greek respondents had average levels of morbidity and IADLs and a low prevalence of ADLs and poor perceived health, they were more likely to have experienced physical functional limitations. It is interesting in that PFLs, which occur earlier in the disablement process, did not appear to be translated to problems with activities of daily living or poor self-perceived health. Although this again may be due to cultural differences, it is worthy of further investigation as, if real, it may provide clues on delaying progression of disability. Nevertheless our choice of measures across the disablement process was not only to examine any trade-offs between measures; each indicator is relevant for different policy audiences. For example, years free of chronic morbidity are useful for public health prevention and for health service providers. Years free of ADL restriction, on the other hand, are important for social policy as ADLs, reflecting dependency, are the main drivers of the costs of long-term care.25

The continued increases in LE and the demand for healthy active life in old age are changing the focus from length of life to quality of life in old age. Health expectancies, which combine information on mortality and morbidity, have thus become essential indicators of the health of ageing populations. Comparing health expectancies across countries can be helpful for setting health priorities and, moreover, a range of health expectancies can provides indicators of relevance for different policy audiences. This is in contrast to quality adjusted life expectancy (QALE), which has also demonstrated inequalities between countries and between social groups within countries,26 although it is a less transparent measure since different health states are weighted and then combined into a single indicator. The results presented here suggest that countries with the longest life expectancies were not necessarily those with the most years of health. Substantial inequalities between countries were evident on all health expectancies but the patterns were not the same across the disablement process. Further work is required to assess how much of the differences in patterns of health expectancies between countries indicate environmental, technological, healthcare or other factors that could potentially delay progression from disease to disability.

What is already known on this subject

  • Although life expectancy (LE) is increasing in all European countries, substantial inequalities still exist and do not appear to be reducing.

  • Inequalities in healthy life years (HLY) at age 50 years have been observed to be greater than inequalities in LE.

What this study adds

  • This study shows that inequalities in LE and HLY at age 50 between European countries are evident in a range of health expectancies that cover the disablement process.

  • Although Eastern European countries experienced the fewest healthy years whatever the health measures used, patterns were not consistent throughout the disablement process.

  • The differences in patterns of health expectancies between countries may indicate environmental, technological, healthcare or other factors that could potentially delay progression from disease to disability.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Funding This work was supported by the European Public Health Programme (EHLEIS Project Grant Number 2006 109). SHARE data collection in 2004–07 was primarily funded by the European Commission through its 5th and 6th framework programmes (project numbers QLK6-CT-2001- 00360; RII-CT- 2006-062193; CIT5-CT-2005-028857). Additional funding by the US National Institute on Ageing (grant numbers U01 AG09740-13S2; P01 AG005842; P01 AG08291; P30 AG12815; Y1-AG-4553-01; OGHA 04-064; R21 AG025169) as well as by various national sources is gratefully acknowledged (see for a full list of funding institutions). These funding organisations did not participate in the analysis, interpretation of the data, or preparation, review and approval of the manuscript.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.