Article Text

Download PDFPDF

Minding the gap: changes in life expectancy in the Baltic States compared with Finland
  1. Marina Karanikolos1,
  2. David A Leon2,
  3. Peter C Smith3,
  4. Martin McKee1,2
  1. 1European Observatory on Health Systems and Policies, London School of Hygiene and Tropical Medicine, London, UK
  2. 2European Centre on Health of Societies in Transition, London School of Hygiene and Tropical Medicine, London, UK
  3. 3Centre for Health Policy, Institute of Global Health Innovation, Imperial College Business School and Institute for Global Health, London, UK
  1. Correspondence to Professor Martin McKee, European Centre on Health of Societies in Transition, London School of Hygiene and Tropical Medicine, 15-17 Tavistock Place, London WC1H 9SH, UK; martin.mckee{at}


Background In the 20 years since the three Baltic States, Estonia, Latvia and Lithuania, have been independent, they have converged progressively with a Western neighbours, politically economically and socially. In contrast, the health gap has widened.

Methods Trends in life expectancy in the three Baltic States were compared with Finland and were decomposed by age for the years 1994, 1999, 2004 and 2009 and by cause of death for year 2009. 1994 was when life expectancy fell to its lowest level since the three countries regained independence.

Results From the mid-1980s to the mid-1990s, the gap in life expectancy between the three Baltic States and Finland widened, especially for men. It then narrowed progressively, except Lithuania where it widened again after 1999. Decomposition by age reveals that the narrowing gap has been driven largely by reduced mortality at working ages, partly counteracted by a relative failure to improve at older ages, especially in Lithuania. Decomposition by cause of death identifies diseases of the circulatory system as the largest contributor to the gap, with the contribution largest at older ages. However, cancer deaths, especially among men, are also important as are deaths from external causes among younger men.

Conclusions Although the gaps in life expectancy between the Baltic States and Finland have reduced, improvements, especially in Latvia and Lithuania, have been fragile. There is a clear need to act on the leading causes of the persisting gap with Finland, in particular through action on hazardous drinking and other risk factors for cardiovascular disease.

  • Public health
  • public health policy
  • epidemiology
  • Eastern Europe
  • international health
  • policy
View Full Text

Statistics from


It is now 20 years since the three Baltic States, Estonia, Latvia and Lithuania, reclaimed their independence following the collapse of the Soviet Union. By 2004, they had joined the European Union (EU), and in 2011, Estonia joined the Euro zone. This period has seen huge changes in their political, social and economic fortunes. Yet while much attention has been paid to their political and economic development, rather less has been devoted to how these changes have affected population health. What work has been done in this area has largely focused on overall life expectancy. Yet, this aggregate figure may conceal different experiences at different ages. Furthermore, during a period when life expectancy has continued to improve in Western Europe, it is not sufficient simply to measure performance against the past but rather to assess whether the gap with neighbouring countries is closing.

In this paper, we describe how patterns of mortality, by age, sex and cause, have been changing in the three Baltic States. Given improvements on life expectancy in all EU countries, it was necessary to compare each country's progress with what was happening elsewhere. One possibility was the EU average, but this is problematic because the most recent data from some countries (such as Belgium and Denmark) are several years out of date. Consequently, we selected a neighbouring country, similar in terms of climate and geography. There were three options, Finland, Sweden and Poland. Finland was selected because, over the past 3 decades, its life expectancy at birth has been very close to the 27 nation EU average (it is now very slightly higher), whereas Poland is substantially lower and Sweden somewhat higher. However, Finland is also an appropriate comparator because of the extensive co-operation between it and the Baltic States in understanding health determinants, most notably in the series of FINBALT surveys.


Numbers of deaths by 5-year age groups and cause of death were obtained from WHO Mortality Database.1 Population numbers were obtained from the same source (population in 2009 for Estonia is taken from Eurostat). We used Chiang's method2 to calculate life expectancy at birth and Arriaga's decomposition method to calculate the contribution of specific ages and causes to differences life expectancy gap.3

Arriaga's life expectancy decomposition method allows us to estimate the contribution of mortality in each age group and from each cause to the total difference in life expectancy at birth between two populations. Increases in mortality make a negative contribution to life expectancy, while decreases make a positive contribution, which can then be attributed to specific age groups or causes of death. The total life expectancy gap is therefore a sum of the number of years contributed negatively or positively by deaths in each age group or cause.

The analysis examines changes in life expectancy in the Baltic States over 15 years (1994–2009), divided into four periods (1994, 1999, 2004 and 2009) in which we measure the gap in life expectancy between Finland and, respectively, Estonia, Latvia and Lithuania. We chose 1994 as the starting point for our decomposition by age and sex because that was year that the life expectancy in the three Baltic States fell to its lowest level since gaining independence, taking subsequent points at 5-year intervals (1999, 2004 and 2009) to monitor the progress in closing the gap in life expectancy. Decomposition by sex and cause of death was done for 2009 (latest year with data available for all countries), between the Baltic States and Finland. Calculations were performed in Excel 2007. As the study used routinely available data, ethical approval was not required.


Figure 1 shows the long-term trends in life expectancy in the three former Soviet Baltic States and Finland. The well-known decline in life expectancy consequent upon the break-up of the Soviet Union is clearly apparent, and the causes are well documented. Following this initial shock, there was a rebound for several years, during which life expectancy increased sharply. From this point, life expectancy in Estonia and Latvia increased at a slower but broadly similar rate. In contrast, Lithuania's progress was less consistent: from the end of the 1990s to 2007, there was a downward trend for men and little change for women. Most recently, there has been a renewed improvement in Lithuania. These contrasting trends have meant that in the post-Soviet period Lithuania's position relative to the other two Baltic States has deteriorated and, in 2009, was the lowest of the three for men.

Figure 1

Life expectancy at birth in the Baltic States and Finland, men and women, 1985–2009.

In contrast, Finland has experienced a steady but on average more gradual upward trend throughout the entire period. As a result, between 1994 and 1999, the life expectancy gaps between Estonia, Latvia, Lithuania on the one hand and Finland on the other reduced substantially by 3.3, 4.6 and 2.9 years for men and 2.2, 1.9 and 1.3 years for women, respectively (table 1). However, between 1999 and 2009, while the gap relative to Finland declined further for Estonia (1.9 male M, 1.8 female F) and Latvia (0.6M, 0.4 F), for Lithuania, it increased (−1.7M, −0.6 F).

Table 1

Contribution to life expectancy gap between Estonia, Latvia, Lithuania relative to Finland, for selected years over period 1994–2009, by age group

Age-specific changes

If these trends in the gaps between Finland and the Baltic States are decomposed by age, an interesting picture emerges (figure 2 and table 1). As might have been anticipated from other work,4 although not previously quantified, in almost all cases, there have been reductions in the gap accounted for by mortality in adults up to the age of 60 years, although this is least pronounced for Lithuania. These data, however, show a rather different pattern at older ages, with their contribution to the overall gaps in life expectancy becoming larger over time. Thus, while there is a tendency for mortality convergence in middle age, the mortality gap at older ages has widened. This is particularly apparent for women in Lithuania.

Figure 2

Decomposition of the gap in life expectancy at birth between Estonia, Latvia and Lithuania and Finland by age group, men and women, 1994–2009.

The decompositions of life expectancy described above are of course a function of the age-specific mortality rates in each country. These are summarised in table 2 and give a complementary perspective on these trends. Since 1999, in almost all age groups, there has been a consistent decline in mortality for both men and women in Finland and the three Baltic States. The only exceptions are in Lithuania in middle and older age. Most notably, there has been stagnation among men at age 60+ years and only relatively modest declines for women at the same age. However, even in the age groups where rates have declined in Lithuania, the declines have not been as steep as they have been in Estonia and Latvia. This has resulted in a reversal of Lithuania's position relative to Estonia and Latvia. Whereas among adult men, Lithuania had the lowest mortality rates of the three Baltic States in 1994, and by 2009, at each adult age, it had the highest rates. A similar although less consistent reversal is seen for adult women in Lithuania.

Table 2

Age-standardised all-cause mortality rates in Estonia, Latvia, Lithuania and Finland, for selected years over period 1994–2009, by age group

Differences by cause of death

Figure 3 and table 3 show the decomposition of the life expectancy gap between the Baltic countries and Finland by age and cause of death in 2009. Each country shows a similar pattern, although of a different magnitude. All major disease groups are contributing to the gap (table 3).

Figure 3

Decomposition of the gap in life expectancy at birth between Estonia, Latvia and Lithuania and Finland by age group and cause of death, men and women, 2009. CVD, cardiovascular disease; IHD, ischaemic heart disease.

Table 3

Contribution to life expectancy gap between Estonia, Latvia, Lithuania relative to Finland in 2009, by age group and cause of death

In all three countries, deaths from diseases of the circulatory system are the biggest contributor to the gap, with 3.3 (Estonia), 4.4 (Latvia) and 3.8 (Lithuania) years in men and 3.3 (Estonia), 3.7 (Latvia) and 4.2 (Lithuania) years in women. The age groups contributing the most are 55–74 in men and 70+ in women. It is important to note that in Estonia and Lithuania, the vast majority of causes of death contributing to the life expectancy gap in the 75+ age group were classified as cardiovascular but a compensatory positive impact of causes from the ‘other’ category implies that some disease categories are underused when recording causes of death in the older population in those countries. Close inspection of the detailed causes of death suggests, in particular, that there may be underdiagnosis of dementia and Alzheimer's disease in the Baltic States compared with Finland.

Cancer also makes an important contribution to explaining the gap, although mainly among men, accounting for 1.4 (Estonia), 1.5 (Latvia) and 1.5 (Lithuania) years of the gap; the respective figures for women are 0.6 (Estonia), 0.8 (Latvia) and 0.5 (Lithuania) years. Cancer deaths in men aged 50–74 years contribute 1.1 (Estonia), 1.1 (Latvia) and 1.2 (Lithuania) years. In women, cancer deaths between 40 and 64 contribute most to the total cancer attributable gap.

In men, external causes also make substantial input to the gap, particularly in Lithuania: 1.4 (Estonia), 1.2 (Latvia) and 2.3 (Lithuania) years, the vast majority being among working age men (15–64).


The paper has identified differences in population health, reflected in changes to life expectancy at birth, in Estonia, Latvia and Lithuania compared with Finland. In particular, we have shown that there are subtle differences in the trajectories taken by the different countries over the past 20 years that have distinctive age- and cause-specific components. However, our analyses are subject to the same limitations as any study based on routinely collected mortality data. WHO categorises the Baltic States as having good quality data.5 However, difference in certification and coding practices may potentially affect the results of cause-specific analysis, especially, as noted above, in the oldest age groups. In order to minimise this effect, and consistent with similar previous work, we used broad disease categories. The wave of emigration that affected the Baltic States once they joined the EU could potentially affect population counts. However, a recent analysis conducted by Jasilionis and colleagues6 shows that adjustment of population data to take account of migration did not have a significant impact on age-standardised mortality rates and therefore should not affect the changes in life expectancy greatly.

Despite these limitations, we think that there are some real and substantial improvements seen in Estonia and Latvia relative to Finland, with these countries managing to reduce the very high level of mortality in middle age observed in mid-1990s. Far less progress has been observed in Lithuania—which, although starting with an advantage compared with other Baltic States in this age group, has failed to sustain the improvement after 1999. There is, however, considerable scope for doing so, given, for example, how external causes overall and in younger adults now account for much more of difference in life expectancy with Finland than they do in Estonia and Latvia. Although all three countries have or are developing national policies on injury prevention,7 Lithuania has been less successful than its two neighbours in implementing effective interventions in key areas of road safety, poisoning and intentional injury.8–10 A study on road traffic injuries in Lithuania between 1998 and 2007 showed the need to address driving under the influence of alcohol, as well as to review road safety regulations.11 A very recent substantial decline in mortality from traffic injuries in Lithuania seems to be related as much to reduced driving as a consequence of the financial crisis as to any improvements in road safety.12 However, it is apparent that here is a need for a more systematic identification of the risk factors underlying mortality from external causes and subsequent revision of the national injury prevention programme.13

Cardiovascular diseases make a major contribution to the gap in life expectancy in Finland, particularly apparent in middle age men. Measures that tackled premature morality from ischaemic heart disease and stroke in this age group would help considerably in reducing the difference in life expectancy with Finland. It has been estimated that effective action against known risk factors (smoking, diet, physical activity) in the general population could postpone more than one-third of premature deaths from heart disease, while improvements in the delivery of care and lifestyle interventions with patients already diagnosed as having heart disease could postpone about 20% of premature deaths.14 It is also apparent that a much higher priority should be given to achieving blood pressure control in each country.15–17

Alcohol has long been recognised as an important cause of premature mortality in the former Soviet Union, including the Baltic States,18 ,19 and a particularly important contributor to the health gap between Western and Central and Eastern European countries, as well as to male–female differences in life expectancy.20 Easy access to alcohol and lax legislation21 ,22 have fostered a culture of binge drinking23 in which alcohol is a major cause of poor health in the Baltic region. While hazardous drinking has been a problem in Estonia, Latvia and Lithuania for many years and calls to reduce it have featured in many health promotion programmes, there has been no tangible progress in reducing alcohol consumption until recently. Between 1996 and 2004, affordability of alcohol in the Baltic countries increased by more than 50%.24 In both Estonia and Lithuania consumption of alcohol peaked in 2007 and in Estonia at this time, surrogate alcohols, mostly imported from Ukraine and with a strength of over 90% pure ethanol, were cheap and easily available25 and are likely to have contributed to within-country differences in mortality.26 In 2008, however, both countries increased alcohol taxes and restricted sales and advertising.27 ,28 At the same time, Latvia began to implement a programme to reduce alcohol consumption, which included strengthening retail control, combating smuggling and illegal alcohol sales, and raising awareness among public and professionals.29 Introduction of these measures coincided with a remarkably rapid improvement in life expectancy in all three countries. Conversely, in Finland, tax on alcohol reduced by a third in 2004, causing a sharp increase in alcohol-related mortality.30 However, the death rate is still much lower than that of the Baltic States.

Failure to keep up with Finland at older ages is another area contributing to differences in life expectancy, particularly in Latvia and Lithuania. Over the past 15 years, life expectancy at age 65 in Finland increased sharply in both men and women. This rapid improvement was not matched by men in any of the Baltic States, or in Latvian or Lithuanian women, while life expectancy at age 65 in Estonian women has been increasing in parallel but not converging with Finland.31 This is particularly the case for older men in Lithuania where there has been effectively stagnation in mortality rates since the late 1990s. The literature on health of older people in the Baltic States is very limited. Nevertheless, it is possible to suggest that this could be due to slow progress in improving healthcare for these age groups in the Baltic States compared with Finland, for example, in the effective detection and treatment of hypertension. However, the widening gap at older ages may also reflect the cumulative effect of adverse experiences across the life course among those living in the Baltic States, as population cohorts characterised by higher mortality rates are ageing.

The aberrant pattern in Lithuania relative to Finland can be accounted for by a failure to reduce the life expectancy gap in middle age, especially from external causes among men, coupled with a relative failure to improve older age mortality. While the overall gap with Finland at middle age in Lithuania was not as large in the immediate post-Soviet period (1994–1999) as it was in Estonia and Latvia, the slower reduction in deaths in this age group, together with a comparative failure to improve mortality at older ages, accounts for the inconsistent progress in narrowing the life expectancy gap between the two countries.


Although the gaps in life expectancy between the Baltic States and Finland have reduced, particularly in the last 2 years, the improvements seen in Lithuania and Latvia over the period analysed have been very fragile. The main causes of death contributing to the gap in life expectancy remain cardiovascular diseases, cancers and external causes (in men) in all three countries. What progress has been made has been largely by closing the gap in middle age; older people have not shared in these gains and, compared with their Finnish counterparts, are slipping further behind, especially in Lithuania. It is beyond the scope of this paper to make detailed policy recommendations, but there are some obvious pointers to where actions need to be taken. One is to address the persisting high burden of premature mortality due to hazardous alcohol consumption, especially in Lithuania. Another is to improve the management of prevention and treatment of cardiovascular diseases among older people, with better control of hypertension a potentially important goal to pursue. A third, and related measure, is to strengthen further those measures already taken to reduce smoking rates from what have been very high levels, especially among men.32

The challenge of closing the gap in life expectancy gap between the Baltic States and their Western EU neighbours remains significant, demanding action in many areas. The patchy progress achieved so far shows that there is a need for much more to be done.

What is already known on this subject

  • The gap in life expectancy between the three Baltic States and Western European countries such as Finland widened markedly in the 1990s and is only narrowing slowly.

What this study adds

  • This study quantifies the contribution made by deaths at different ages and from different causes to the gap between the three Baltic States and Finland.

  • Positive contributions to narrowing the gap have been made by falling deaths at working age, counteracted by a failure to improve at older ages.

  • The major contributors to the current gap are cardiovascular diseases and cancer, especially at older ages, and external causes among younger men.

Policy implications

  • Investment in measures to tackle alcohol-related mortality among young men and cardiovascular disease generally should be a priority.


View Abstract


  • Funding This work is supported by the programme on health system performance of the European Observatory on Health Systems and Policies (see

  • Competing interests None.

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

  • Data sharing statement All data used are publicly available.

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.