The key public health problems of smoking, alcohol and obesity are strongly socially patterned. In recent years, Fanny Janssen says, they have contributed markedly both to average educational inequality in remaining life expectancy at age 30 (e30) in England & Wales, Finland and Italy (Turin), and to its increase.
In modern welfare states, people with a low socio-economic position (SEP) live for 3-10 fewer years, on average, than people with a high SEP (Murtin et al. 2017). In recent decades, these socio-economic inequalities in mortality and life expectancy have even widened in several European countries (Zazueta-Borboa et al. 2023), despite efforts to reduce them. To fully grasp these trends, it is essential to study the role of lifestyle factors.
Past and present effects of lifestyle factors on mortality
Smoking, alcohol misuse and behaviours resulting in obesity (unhealthy diets, insufficient physical activity) are important preventable risk factors of mortality in Europe. Their prevalence and associated mortality are currently higher among people with low SEP than those with high SEP, and they are known to contribute strongly to socio-economic mortality inequalities in high-income countries, (Petrovic et al. 2018).
Their impact on past trends in socio-economic mortality inequalities has been less widely studied, but it is expected to be large. First, these lifestyle factors contribute strongly to general mortality trends because they generally develop over time as wave-shaped epidemics, with their prevalence and associated mortality first increasing strongly, and eventually declining (Janssen et al. 2011). Second, important differences between socio-economic groups exist in the timing and impact of these lifestyle “epidemics”. Smoking, obesity, and alcohol epidemics occurred relatively late among those with a lower SEP, but with larger effects (Giskes et al. 2005; Mackenbach et al. 2015; Kagenaar et al. 2022).
In a recent article, we examined for the first time the combined contribution of smoking, alcohol and obesity on trends in socio-economic inequalities in life expectancy (Janssen et al. 2025). We did so for England & Wales (E&W), Finland and Italy (Turin), by using individually linked mortality data by educational level (low, middle, high) for individuals aged 30 and over, by single calendar year.
Combined impact of smoking, alcohol and obesity on socio-economic differences in life expectancy
In E&W, Finland and Italy (Turin), educational inequalities in life expectancy at age 30 (e30) were 4.4 years, on average, over the 1992-2017 period, of which the average combined impact of smoking, alcohol and obesity amounted to 1.9 years, representing 44% of the total (Figure 1). This impact was higher among males (2.7 out of 5.2 years; 51%) than among females (1.2 out of 3.5 years; 34%), and was highest for Italy among males (60%) and for E&W among females (43%). On average, smoking contributed 23%, alcohol 14%, and obesity 10%. In E&W, and among Italian males, smoking had the largest impact, whereas in Finland alcohol contributed the most.
Changes over time in the impact of lifestyle on educational inequalities in life expectancy
The combined impact of smoking, alcohol and obesity on educational inequalities in e30 changed over time, however (Figure 1). For Finnish males, the increasing impact up to 2008 was driven mainly by increases in educational inequalities in alcohol-attributable mortality, which can be linked to a reduction in alcohol taxes and prices in Finland before 2008. This also played a role for Finnish females, along with increasing educational inequalities in smoking-attributable mortality.
For Italian and British males (up to 2006), the changing impact was driven foremost by increases and declines, respectively, in educational inequalities in smoking-attributable mortality. These contrasting trends reflect differences in the timing of the smoking epidemic between populations and the associated socio-economic differences therein. Among British males, who were the first to be affected by the smoking epidemic, and who had already exhibited declines in smoking-attributable mortality for quite some time, inequalities narrowed due to smaller declines in smoking-attributable mortality among the high educated – who had already returned to low levels – than among the low educated. For Italian males, smoking-attributable mortality peaked more recently, and later for the low than the high educated, resulting in smaller declines for the low educated, and hence larger educational inequalities in smoking-attributable mortality. Female smoking-attributable mortality increased only recently, and more strongly for the low educated, among Finnish women especially, who saw a widening of educational inequalities in smoking-attributable mortality.
Impact of lifestyle on trends in educational inequalities in life expectancy
Without smoking, alcohol and obesity, the trends in educational inequalities in e30 are different, particularly for British males (1991-2006), Finnish males, Italian males and Finnish females (Figure 2). For Finnish males, the reversal from increasing to declining educational inequalities in e30 in 2008 even disappears when only non-lifestyle-attributable mortality is considered. This reversal can be attributed almost entirely to a similar reversal in educational inequalities in alcohol-attributable mortality following the country-specific introduction of stricter alcohol policies around 2008 that mainly resulted in declines among those with the highest levels of alcohol abuse, i.e. low-educated males.
Smoking, alcohol and obesity combined contributed, respectively, 12%, 57%, 63%, 43% to the observed increases in educational inequalities in e30 among British males (2006-2017), Finnish males (1987-2008), Finnish females (1987-2017) and Italian males (1990-2018). In addition, educational differences in mortality from smoking, alcohol and obesity combined contributed 24% to the decline in educational inequalities in e30 among British males (1991-2006); without lifestyle-attributable mortality the decline among British females (1992-2017) would have been 38% larger.
The largely similar increasing trends for Finnish males and females, and – to a lesser extent – for Italian males and females, after the effect of lifestyle has been removed, could well indicate the importance of gradual increases in general inequalities in material or other social resources, potentially resulting from the lower educated becoming a more homogeneous group with worse health because of educational expansion, or from the increasing inflow of people from low-income countries.
For E&W, the trend break in 2008 for males that remains after removing the effect of lifestyle suggests that factors beyond smoking, alcohol and obesity, with a larger effect on educational inequalities in e30 for males than for females, are mainly responsible for the recent increase in educational inequalities in e30 among males in E&W. This might be the result of austerity measures introduced in 2008 which have affected British males more than British females, but this requires further research.
Conclusion
The results imply that targeting socio-economic differences in health behaviours –smoking and alcohol misuse in particular – may not only substantially reduce socio-economic inequalities in e30, but may also slow the increase over time in these inequalities. The observed country-level differences in individual lifestyle factors that contributed most to the observed educational inequalities in e30 suggest the need for context-specific strategies.
Funding
This work is financed by the Netherlands Organisation for Scientific Research (NWO) as part of the research programme “Forecasting future socio-economic inequalities in longevity: the impact of lifestyle ‘epidemics’”, under grant no. VI.C.191.019. See: www.futurelongevitybyeducation.com.
Bibliography
Giskes K, Kunst AE, Benach J, Borrell C, Costa G, Dahl E, et al. (2005). Trends in smoking behaviour between 1985 and 2000 in nine European countries by education. Journal of Epidemiology & Community Health 59(5):395-401.
Janssen, F., Martikainen, P., Zengarini, N., Sizer, A. & A.E. Kunst (2025). The combined impact of smoking, alcohol and obesity on past trends in educational inequalities in life expectancy in England & Wales, Finland and Italy, 1990-2017. European Journal of Public Health, doi: 10.1093/eurpub/ckaf181.
Janssen, F., Trias-Llimós, S., and Kunst, A. E. (2021). The combined impact of smoking, obesity, and alcohol on life expectancy trends in Europe. International Journal of Epidemiology 50(3): 931-941. doi: 10.1096/ije/dyaa273.
Kagenaar, E., Van Hemelrijck, W. M. J., Kunst, A. E. & F. Janssen (2022). Long-Term Trends in Obesity Prevalence by Socio-Economic Group in Five European Countries and the USA: The Relevance of the Diffusion of Innovations Theory. Obesity Facts 15(6): 753-761. doi: 10.1159/000527070.
Mackenbach, J. P., Kulhanova, I., Bopp, M., Borrell, C., Deboosere, P., Kovacs, K., . . . de Gelder, R. (2015). Inequalities in alcohol-related mortality in 17 European countries: A retrospective analysis of mortality registers. PLoS Medicine, 12(12), e1001909.
Murtin, F., Mackenbach, J., Jasilionis, D., & d’Ercole, M. M. (2017). Inequalities in longevity by education in OECD countries: Insights from new OECD estimates. OECD Statistics Working Papers, 2017/02, OECD Publishing, Paris.
Petrovic, D., de Mestral, C., Bochud, M., Bartley, M., Kivimäki, M., Vineis, P., Mackenbach, J. & Stringhini, S. (2018). The contribution of health behaviors to socioeconomic inequalities in health: A systematic review. Preventive Medicine, 113, 15-31.
Zazueta-Borboa, J.D., Martikainen, P., Aburto, J.M., Costa, G., Peltonen, R., Zengarini, N., Sizer, A., Kunst, A.E. & Janssen, F. (2023). Reversals in past long-term trends in educational inequalities in life expectancy for selected European countries. Journal of Epidemiology & Community Health 77: 421-429. doi: 10.1136/jech2023-220385.