Elsevier

Health & Place

Volume 28, July 2014, Pages 116-132
Health & Place

Cultural values and population health: a quantitative analysis of variations in cultural values, health behaviours and health outcomes among 42 European countries

https://doi.org/10.1016/j.healthplace.2014.04.004Get rights and content

Highlights

  • Inglehart, Hofstede, and Schwartz cultural values are all associated with health behaviours and health outcomes.

  • Inglehart׳s ‘self-expression’ scale has most associations with health behaviours and health outcomes.

  • Cultural values account for some of the striking variations in health behaviour between neighbouring countries.

  • Paradoxically, a shift away from ‘survival’ values promotes behaviours that increase longevity.

Abstract

Variations in ‘culture’ are often invoked to explain cross-national variations in health, but formal analyses of this relation are scarce. We studied the relation between three sets of cultural values and a wide range of health behaviours and health outcomes in Europe.

Cultural values were measured according to Inglehart׳s two, Hofstede׳s six, and Schwartz׳s seven dimensions. Data on individual and collective health behaviours (30 indicators of fertility-related behaviours, adult lifestyles, use of preventive services, prevention policies, health care policies, and environmental policies) and health outcomes (35 indicators of general health and of specific health problems relating to fertility, adult lifestyles, prevention, health care, and violence) in 42 European countries around the year 2010 were extracted from harmonized international data sources. Multivariate regression analysis was used to relate health behaviours to value orientations, controlling for socioeconomic confounders.

In univariate analyses, all scales are related to health behaviours and most scales are related to health outcomes, but in multivariate analyses Inglehart׳s ‘self-expression’ (versus ‘survival’) scale has by far the largest number of statistically significant associations. Countries with higher scores on ‘self-expression’ have better outcomes on 16 out of 30 health behaviours and on 19 out of 35 health indicators, and variations on this scale explain up to 26% of the variance in these outcomes in Europe. In mediation analyses the associations between cultural values and health outcomes are partly explained by differences in health behaviours. Variations in cultural values also appear to account for some of the striking variations in health behaviours between neighbouring countries in Europe (Sweden and Denmark, the Netherlands and Belgium, the Czech Republic and Slovakia, and Estonia and Latvia).

This study is the first to provide systematic and coherent empirical evidence that differences between European countries in health behaviours and health outcomes may partly be determined by variations in culture. Paradoxically, a shift away from traditional ‘survival’ values seems to promote behaviours that increase longevity in high income countries.

Introduction

Europe is a subcontinent of great diversity, not only in terms of language, religion and other aspects of culture, but also in terms of population health. At the start of the 21st century, life expectancy at birth in Europe is more unequal than it has been for decades (Mackenbach, 2013a), and enormous variations between countries have been documented on all available measures of population health, including mortality from specific conditions, incidence of infectious diseases and cancer, and self-reported health and disability (Mackenbach, 2013b, Marmot et al., 2012, Mladovsky et al., 2009).

The main health divide within Europe is between East and West. Over the past decades, the countries of Western Europe have experienced sustained improvements in life expectancy, with a gradual convergence of all countries towards high values. By contrast, the countries in Central and Eastern Europe have experienced stagnation and sometimes even falls in life expectancy, both before and after the fall of the Soviet empire (Leon, 2011, McMichael et al., 2004). Although many specific health indicators also vary along an East–West axis, some other patterns can be noted as well, such as the low levels of mortality from ischaemic heart disease in Southern Europe (Mackenbach and McKee, 2013b).

The explanation of these patterns of variation is undoubtedly complex, and is likely to include a wide range of factors. East–West patterns may be related to recent political history, when countries in Central and Eastern Europe lived for decades under autocratic, communist regimes (Mackenbach, 2013c, Mackenbach et al., 2013a, McKee and Nolte, 2004). More specific explanations are likely to be involved as well, such as variations in economic conditions, health-related behaviours, access to health care, and effectiveness of health policies (Bobak et al., 2007, Mackenbach and McKee, 2013a, Nolte et al., 2004, Stuckler et al., 2009). However, the generic and long-standing character of these variations in health (life expectancy was already lower in Central & Eastern Europe before the second World War (Kirk, 1946; Mackenbach, 2013c)) suggests that there may also be some historically more persistent explanations, such as cultural differences.

Despite a certain degree of cultural unity, however defined (Davies, 1996), and despite recent attempts at economic and political unification, Europe is a culturally diverse subcontinent. The concept of culture will be used here in its sociological definition of “the ways of thinking, the ways of acting, and the material objects that together shape a people׳s way of life” (Macionic and Gerber, 2011). Variations within Europe have roughly been summarized as occurring along two ‘fault-lines’. The first separates East from West, e.g., Orthodox from Catholic Christianity, and late from early industrializing societies. The second ‘fault line’, equally fuzzy, divides South from North, e.g., Romance from Germanic languages, and Catholicism from Protestantism (Arts et al., 2003).

Variations in culture may have a profound impact on health, for example through variations in health-related behaviours (Payer, 1996). Individuals in different European societies differ in, among other things, their fertility patterns, lifestyles, and rates of participation in preventive programs, and some of these variations may well be due to variations in attitudes, norms or other elements of culture. The same applies to variations in what will be denoted here as “collective health behaviour”, in the form of national policies in the areas of prevention, health care, and the environment (Mackenbach, 2013b). However, although ‘culture’ is an implicit or explicit part of many theories of the determinants of health behaviour (Glanz et al., 2002), it is often treated as a background variable that eludes measurement, so that quantitative evidence of the impact of culture on between-country variations in health is extremely scarce.

One area of cross-cultural research where measurement issues have been tackled effectively, and where quantitative data on between-country variations in culture have become available, even abundantly, is that of cultural values. ‘Values’ are broad preferences concerning appropriate courses of action that the members of a society share, and that underlie their norms of behaviour in specific situations (Nolan and Lenski, 2004). Over the past decades, several approaches to operationalizing cultural values have evolved, each with its own theoretical rationale and its own set of survey-based indicators.

One widely known approach has been developed by Ronald Inglehart, a political scientist from the United States who developed the theory of ‘post-materialism’ (Inglehart, 1977). In respondents’ answers to survey questions on their beliefs and attitudes, he discovered an intergenerational shift in the cultural values of the populations of advanced industrial societies, from religious to secular, and from survival-oriented to ‘well-being’ or ‘self-expression’-oriented values (Inglehart, 1990). Inglehart has proposed to measure countries’ cultural value orientation with two indicators, each based on five survey questions, capturing a ‘traditional’ to ‘secular-rational’ and a ‘survival’ versus ‘well-being’ or ‘self‐expression’ axis, respectively (Inglehart, 1997, Inglehart and Baker, 2000, Inglehart and Welzel, 2005). These two dimensions have since been confirmed, with small variations, in several other analyses (Hagenaars et al., 2003), and have allowed large-scale measurement of variations and changes in cultural values in many countries, particularly in the context of the World Values Study and the European Values Survey.

Another approach has been developed by Geert Hofstede, a Dutch sociologist who, on the basis of factor analysis of attitude surveys among employees of the IBM company in 50 countries, initially identified 4 dimensions of national cultures: ׳power distance׳ (extent to which the less powerful accept that power is distributed unequally), ׳individualism׳ (emphasis on personal achievements and individual rights), ׳uncertainty avoidance׳ (tendency to cope with anxiety by minimizing uncertainty and ambiguity), and ׳masculinity׳ (strict division of emotional roles between the genders and emphasis on competitiveness and power) (Hofstede, 1980, Hofstede, 2001). These results were later confirmed in other populations, but analyses of data from national values surveys have revealed two additional dimensions: ׳long-term orientation׳ (importance attached to the future and emphasis on persistence and saving) (Hofstede, 2001, Minkov, 2007) and ׳indulgence׳ (emphasis on gratification of natural human drives related to enjoying life) (Hofstede et al., 2010). The latter scale emerged from an analysis of Inglehart׳s ‘self-expression’ dimension, which appeared to contain two different subdimensions, one corresponding to Hofstede׳s ‘individualism’, the other to this newly coined ‘indulgence’ (Minkov, 2009). Important variations between countries, also within Europe, on Hofstede׳s dimensions have been documented (Hofstede, 1980, Hofstede, 2001, Hofstede et al., 2010).

A third approach is that developed by Shalom Schwartz, an American-Israeli social psychologist. On the basis of a theory of basic human needs, and of surveys measuring the priority attached to items within each of these needs domains, he proposed seven cultural orientations. These are ‘affective autonomy’ (emphasis on the desirability of individuals independently pursuing pleasure and other positive experiences), ‘intellectual autonomy’ (desirability of individuals independently pursuing their own ideas), ‘embeddedness’ (or conservatism; importance of maintaining the social order), ‘egalitarianism’ (importance of transcending self-interest and promoting the welfare of others), ‘hierarchy’ (legitimacy of an unequal distribution of power and resources), ‘harmony’ (fitting harmoniously into the environment), and ‘mastery’ (getting ahead through active self-assertion) (Schwartz, 1994, Schwartz, 1999, Schwartz, 2006). Originally measured in relatively small and restricted samples, particularly students and teachers, questions capturing these dimensions are now also included in the European Social Survey (Davidov et al., 2008). Applications have documented important cross-national variations, also within Europe (Schwartz, 1999, Schwartz, 2006, Schwartz and Bardi, 1997).

Although these three approaches have different theoretical rationales, the dimensions overlap both conceptually and empirically (Gouveia and Ros, 2000, Inglehart and Oyserman, 2004, Schwartz, 2006). This also emerges from Fig. 1 which summarizes the geographical distribution of countries׳ scores on these variables within Europe, on the basis of their associations with longitude and latitude. Some dimensions, such as Inglehart׳s ‘self-expression’ and Schwartz׳s ‘intellectual’ and ‘affective autonomy’ cluster closely together, all having higher values in the West, with Hofstede׳s ‘individualism’ and ‘indulgence’ not far away.

On the other hand, many others have more distinct geographical patterns, implying that there may be scope for a comparative analysis of their impact on health behaviours. Several cultural values display either a clear North–South pattern (such as the ‘secular-rational’, ‘normative/religious’, and ‘uncertainty avoidance’ scales), or a clear West–East pattern (such as ‘embeddedness’ and ‘egalitarianism’). Some other values escape a simple geographical pattern, as in the case of ‘masculinity’, ‘long-term orientation’, and ‘mastery’.

Empirical studies of the relation between cultural values and health-related variables are scarce, but national scores on ‘self-expression’ and related scales like ‘individualism’, ‘indulgence’ and ‘autonomy’ have on occasion been found to be associated with better self-assessed health, higher life expectancy, greater happiness, lower cause-specific mortality, extraversion, modern consumer behaviour, sports participation and other positive health outcomes (Hofstede et al., 2010, Hofstede and McCrae, 2004, Mackenbach and McKee, 2013a, Matsumoto and Fletcher, 1996, Minkov, 2009), perhaps through an effect on citizens’ tendency to invest in their personal well-being, or through an effect on levels of democracy, health care spending and other aspects of the functioning of public institutions (Hofstede et al., 2010, Inglehart and Welzel, 2005). We therefore expect higher average scores on these scales to be associated, over-all, with better health behaviours and better health outcomes.

We also expect average scores on the ‘secular-rational’ and ‘autonomy’ scales to be associated with better health behaviours and better health outcomes, through greater individual and collective reliance on modern scientific thought instead of tradition (and a reverse effect for ‘embeddedness’). On the other hand, ‘uncertainty avoidance’ and ‘power distance’ have previously been found to be associated with more unfavourable health outcomes, such as worse self-assessed health, higher rates of suicide, higher use of antibiotics, and less willingness to donate blood (Arrindell et al., 1997, de Kort et al., 2010, Deschepper et al., 2008, Hofstede et al., 2010), perhaps through higher levels of individual anxiety (in the case of ‘uncertainty avoidance’) or less effective functioning of health care professionals and other public institutions (in the case of ‘power distance’) (Hofstede et al., 2010). The same may, by analogy, apply to the ‘hierarchy’ and ‘egalitarianism’ scales. For the other scales it is more difficult to derive testable predictions.

The main objective of the study reported in this paper then is to assess the explanatory power of these three sets of cultural values for variations between European countries in a wide range of health behaviours and health outcomes. We will also evaluate whether the associations of cultural values and health outcomes are mediated by variations in specific health behaviours.

Finally, after having determined the role of cultural values in the full range of countries with available data, we will further test their explanatory power by comparing the value orientations of neighbouring countries, which, despite being similar in many respects, including national income, have very different levels of health behaviour: Sweden and Denmark, the Netherlands and Belgium, the Czech Republic and Slovakia, and Estonia and Latvia (Mackenbach, 2013b, Mackenbach and McKee, 2013a).

Section snippets

Data (independent variables)

Data on Inglehart׳s ‘self-expression’ and ‘secular-rational’ scales come from the 2006 or most recent waves of the World Values Study, and were extracted from the Quality of Government Dataset (Teorell et al., 2011) (http://www.qog.pol.gu.se/data/). We also used two alternative operationalizations. The first is a set of two indices calculated from respondent answers on World Values Study questions that predate the two dimensions currently in use, and that were designed to directly measure

Health behaviours

Large variations in aggregate individual as well as collective health behaviours are found between European countries. Fig. 2a summarizes the geographical distribution of health behaviours in Europe; for full details see Web Appendix Table A1b. There are no individual health behaviours with strong North–South gradients, as shown by the rather low correlations with latitude. There are, however, several with strong East–West gradients, as shown by positive or negative correlations with longitude:

Summary of main findings

In univariate analyses, all scales are related to aggregate individual and collective health behaviours and most scales are related to health outcomes, but in multivariate analyses Inglehart׳s ‘self-expression’ (versus ‘survival’) scale has by far the largest number of statistically significant associations. Countries with higher scores on ‘self-expression’ have better outcomes on 16 out of 30 health behaviours and on 19 out of 35 health indicators, and variations on this scale explain up to

Conclusions

This study is the first to provide systematic and coherent empirical evidence that differences between European countries in health behaviours and health outcomes may partly be determined by variations in culture. Paradoxically, a shift away from traditional ‘survival’ values seems to promote behaviours that increase longevity in high income countries.

Acknowledgements

I am grateful to Geert Hofstede for his critical comments on a previous version of this paper, to Shalom Schwartz, Jacques Hagenaars and Loek Halman for sharing their data with me, and to Ruut Veenhoven for giving useful suggestions for additional analyses.

References (75)

  • Abramson, P.R. (2011). Critiques and Counter-Critiques of the Postmaterialism Thesis: Thirty-four Years of Debate. CSD...
  • P. Anderson et al.

    Alcohol in Europe

    (2006)
  • W. Arts et al.

    Europe and its values in an historical perspective

  • S.O. Becker et al.

    Was Weber wrong? A human capital theory of protestant economic history

    Q. J. Econ.

    (2009)
  • M. Bobak et al.

    Societal characteristics and health in the former communist countries of Central and Eastern Europe and the former Soviet Union: a multilevel analysis

    J. Epidemiol. Community Health

    (2007)
  • E. Davidov et al.

    Bringing values back in the adequacy of the European Social Survey to measure values in 20 countries

    Public. Opin. Q.

    (2008)
  • N. Davies

    Europe. A history

    (1996)
  • W. de Kort et al.

    Blood product collection and supply: a matter of money?

    Vox Sang.

    (2010)
  • P. Dekker et al.

    Divers Europa. De Europese Unie in de publieke opinie & Verscheidenheid in cultuur, economie en beleid. Bijlage bij Staat van de Europese Unie 2007

    (2007)
  • R. Deschepper et al.

    Are cultural dimensions relevant for explaining cross-national differences in antibiotic use in Europe?

    BMC Health Serv. Res.

    (2008)
  • C.L. Fincher et al.

    Pathogen prevalence predicts human cross-cultural variability in individualism/collectivism. [Research Support, Non-U.S. Gov’t]

    Proc. Biol. Sci.

    (2008)
  • K.L. Frohlich et al.

    A theoretical proposal for the relationship between context and disease

    Sociol. Health Illn.

    (2001)
  • M.J. Gelfand et al.

    Differences between tight and loose cultures: a 33-nation study

    Science

    (2011)
  • V.V. Gouveia et al.

    Hofstede and Schwartz׳s models for classifying individualism at the cultural level: their relation to macro-social and macro-economic variables

    Psicothema

    (2000)
  • P. Grigoriev et al.

    Mortality in Belarus, Lithuania, and Russia: divergence in recent trends and possible explanations

    Eur. J. Popul.

    (2010)
  • J. Hagenaars et al.

    Exploring Europe׳s basic values map

  • M. Haller

    Theory and method in the comparative study of values: critique and alternative to Inglehart

    Eur. Sociol. Rev.

    (2002)
  • Hofstede, G.,1980. Culture׳s Consequences: International Differences in Work-Related Values (First ed.). Beverly Hills:...
  • G. Hofstede

    Culture׳s Consequences. Comparing values, Behaviors, Institutions, and Organizations Across Nations,

    (2001)
  • G. Hofstede et al.

    Cultures and Organizations: Software of the Mind,

    (2010)
  • G. Hofstede et al.

    Personality and culture revisited: linking traits and dimensions of culture

    Cross-Cult. Res.

    (2004)
  • S.P. Huntington

    The Third Wave: Democratization in the Late 20th Century

    (1991)
  • R. Inglehart

    The Silent Revolution

    (1977)
  • R. Inglehart

    Culture Shift In Advanced Industrial Society

    (1990)
  • R. Inglehart

    Public support for environmental protection: Objective problems and subjective values in 43 societies

    PS: Politi. Sci. Polit.

    (1995)
  • R. Inglehart

    Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies

    (1997)
  • Cited by (44)

    • Political interest across cultures: The role of uncertainty avoidance and trust

      2022, International Journal of Intercultural Relations
      Citation Excerpt :

      In time, this mental programming differentiates groups of people from others. Researchers emphasized that culture has a determinant role in individuals' attitudes, beliefs, norms, and values towards a wide range of phenomena (Lim et al., 2009; Mackenbach, 2014), meaning it works as a precursor of different variables. To identify cultural dimensions, Hofstede collected data from IBM workers in different countries; and he suggested six dimensions depending on this study.

    View all citing articles on Scopus
    View full text