Health Inequalities by Education Level and Gender in Spain

How much does Economic Hardship Contribute to Differences in Health by Education Level?

Aïda Solé-Auró, Department of Political and Social Sciences,
Universitat Pompeu Fabra, Barcelona

To what extent are differences in the frequency of disability by education level and gender associated with differences in poverty? Data from the European Statistics on Income and Living Condition (EU-SILC) for Spain allows us to measure the prevalence of activity limitation (AL) and the rate of economic hardships (EH) by level of education, gender and age. We measure the association between EH and AL. We compute the increased AL disadvantage of the low-educated relative to the middle-educated, and the reduced AL advantage of the high-educated relative to the middle-educated, to get the contribution (in percentage) of EH to the association between education and AL by gender.

The distribution of EH is quite similar by gender across education groups. The extent of the AL-advantage/disadvantage varies by education level and gender. EH contributes to the AL-advantage/disadvantage but differently among men and women across educational groups. Actions to reduce poverty are needed in Spain to reduce the levels and differences in disability.

1. Introduction

Europeans are living longer than ever. In Spain, life expectancy at birth reached 83.3 years in 2014, though with significant differences by gender: life expectancy at birth for women was 86.2 years, almost six years higher than that of men (80.4 years). However, a significant part of life at advanced ages is associated with diseases and disability (Solé-Auró and Crimmins, 2013), and large variations are found across socioeconomic groups (Mackenbach et al., 2008).

Despite advances in health care, the expansion of education and increases in income, significant gaps in life expectancy and health by education level persist among Spaniards. Understanding these differences and addressing disparities is critically important to improving the population’s overall health and reducing health inequalities. Regarding the proportion of persons suffering disabilities, low-educated groups show a disadvantage, while high-educated groups have a systematic advantage (Cambois et al. 2016).

Disability results from the exposure to health events, such as chronic diseases, which deteriorate body functions and hamper the performance of activities, challenging social participation and quality of life. Disability also arises from the resources available to individuals to adjust to deteriorated functions (such as assistive devices, care giving and an adapted environment). Therefore, differences by socioeconomic status in the prevalence of disabilities stem from complex interactions between individual, household and country characteristics (Eikemo et al., 2008). 

Income, poverty and educational differentials in health

Income impacts individuals’ health status in two main ways (Marmot, 2002). First, there is the material effect of income on health related to living conditions and the amount and type of goods and services that people can afford. Second, there is the psychological effect of income on health associated with the level of control people have over their life and their environment. As a result, an increase in income might have a positive effect on health over the full range of the social gradient, although some studies show, using cross-sectional data from 56 countries, that there may be a decreasing return when a “material” ceiling is reached (Rodgers, 1979); while others, using data from the United States, show a linear relationship between income and life expectancy (Chetty et al., 2016). The income-health association partially explains the relationship between education and differences in health and disability (Montez, Hummer, and Hayward, 2012) in the same way as education partially reflects material conditions in early stages of life (Hayward and Gorman, 2004). 

Educational Expansion in Spain

The evolution of Spain’s education system since the Spanish Civil War (1936-1939) has not had a uniform impact on successive generations of students (Ballarino et al., 2009). Under Franco’s dictatorial regime, the state played what was primarily a subsidiary role in education, with schooling being dominated by the private sector, above all at the secondary level. However, the deterioration in living conditions of the Spanish population following the war made it impossible for the majority to gain access to these private institutions. This resulted in marked class- (and gender) based inequalities in educational attainment. The Spanish population had to wait until 1970 – just five years before the death of Franco – to benefit from a comprehensive system of compulsory education until the age of 14. De la Fuente and Doménech (2015) present a detailed description of the evolution in the educational attainment of the adult population between 1960 and 2010. They report a remarkable expansion in the average number of years of schooling in Spain, rising from 4.7 to 9.64 years between 1960 and 2010. At the beginning of this period, illiteracy rates were low in most developed countries; yet, Spain, along with the other Southern European countries, continued to present significant levels of illiteracy. Over these four decades, illiteracy almost disappeared, falling from 15.0 to 2.1%.

This study looks at the disability disadvantage that exists for low socioeconomic status groups and the disability advantage that prevails for high socioeconomic status groups in Spain. In addition, this article quantifies the extent to which the varying disability disadvantage/advantage of Spanish low-and high-educated groups compared to middle-educated groups, are explained by their respective levels of poverty. Several scenarios are plausible. A large disability disadvantage could be moderately mediated by poverty, if poverty accumulates with other important factors impacting disability (such as dangerous working conditions, limited access to efficient health care, harmful practices). In contrast, poverty could make a large contribution to the disability disadvantage of the low-educated in contexts where poverty is low but concentrated in this specific group. Regarding the high-educated groups, poverty could make an important contribution to their disability advantage relative to the middle-educated groups, where the risk of poverty is high. In contrast it could be moderate where poverty is scarce among both high- and middle-educated and differences between these groups are driven by other factors. 

2. Our study

This study uses the European Union Statistics on Income and Living Conditions (EU-SILC), a database based on information gathered by national statistical offices. We used the 2014 EU-SILC cross-sectional data for Spain, which includes a special module on material deprivation. Data is collected by ad hoc interviews in which participants provide self-reported information regarding health and socioeconomic status. We included individuals over age 30 because education is relatively well reported and for most people stays constant throughout adult life. We excluded individuals aged 80-plus due to missing information. Our study comprises 19,954 Spanish individuals (9,553 males and 10,401 females) aged 30-79 (Table 1).



We identify disability based on the Global Activity Limitation Indicator measuring health related activity limitation (AL) with a single question: “For at least the past six months, to what extent have you been limited because of a health problem in activities people usually do?” (Severely limited; Limited but not severely; Not limited)". We considered two groups; those who are severely limited and limited but not severely, versus those who are not limited.


We use education, a common but not complete proxy for socioeconomic status, since it tends to be stable after early adulthood, and relatively easy to measure, given that respondents usually report their educational attainment truthfully.  Education is strongly related to health and disability risk through a variety of pathways, specifically early life circumstances, household circumstances, job opportunities, and the development of skills to maintain health and adjust to disorders. We considered three groups based on the level of education achieved, using the 2011 International Standard Classification of Education: low (primary and lower secondary education), middle (upper secondary education) and high (tertiary education).

Gender differences between those with high and low education have narrowed in recent years.


Using the 2014 EU-SILC, poverty can be examined through a thematic module on self-perceived situations of deprivation (Whealan and Maître, 2013). In this study, we focused on the “economic stress” dimension of the thematic module, which is assessed by two items. The first item refers to “the household ability to cope with unanticipated expenses” (Yes/No). The second item (introduced by “A household may have different sources of income and more than one household member may contribute to it. Thinking of your household’s total income”) refers to “the household ability to make ends meet, namely, to pay for its usual necessary expenses”. The answer categories are “With great difficulty; with difficulty; with some difficulty; fairly easily; easily; Very easily”. We consider a situation of economic hardship (EH) to be one in which individuals report “great” or “some” difficulty making ends meet and difficulty coping with unexpected expenses.


Differentiated by gender, we first analyzed the distribution of education levels and then the prevalence of AL and EH in Spain by education level among the population from 30-79 years of age as a whole and then in two age groups (30-54 and 55-79 year-olds) to highlight changes between birth cohorts.

Secondly, using statistical techniques we looked at the degree of association between EH and AL by age groups and gender. To do so, we quantify how strongly the presence of EH is associated with AL in Spain. If the value is greater than 1, then economic hardship is considered to be positively associated with activity limitation.

Finally, we measure the contribution of EH to the AL advantage and disadvantage of the high- and low-educated groups relative to the middle-educated by gender in Spain. We explore to what extent EH mediates the disability advantage and disadvantage associated with education. As presented in graph 1, we measure the direct effect of education and an indirect effect mediated by EH. The breakdown of the total effect (direct plus indirect) allows us to estimate the percentage contribution of EH to the total effect of education (indirect effect divided by the total effect), among other social factors (such as behaviours, access to adequate care and adaptation, employment and working conditions). 

3. Results

Distribution of education by age groups and gender

Graph 2 shows the distribution of education for three age-groups (30-54; 55-79; All: 30-79) and by gender. Examination of the graphs clarifies that differences by gender between those with high and low education have narrowed in recent years. Low-educated men and women in Spain represent the majority of the sample in all age groups; however, in the oldest age group the proportion of low-educated women is higher compared to men and this proportion is always higher compared to low-educated Spanish men. This is partly explained by the history of the Spanish educational system. 

High-educated Spanish women increase over time as the proportion changes from 14.5% in the oldest age group to 42.4% in the youngest. Nevertheless, those proportions are lower in all age groups than for men. 

Due to the history of the educational system in Spain, there are more low-educated women than low-educated men.

High-educated men represent almost a third of the distribution (29.8%), and the proportion of low-educated people has decreased and is now lowest in the youngest generations compared to the oldest age group (55-79). The proportion of middle educated people remains quite steady across age groups. 

Distribution of activity limitation and economic hardship across age groups and education level by gender

The prevalence of AL varies across age groups and by gender and educational level (Table 1). The low-educated consistently show the highest prevalence of AL and the high-educated the lowest, although to a different extent by gender and age groups. The prevalence of AL for low-educated men and women is aproximately double in the older age group compared to the younger one; for high-educated men it is almost triple and for high-educated women it is almost double. There is no evidence of a systematic relationship between the population distribution across educational groups and the extent of differences in AL.

The average level of EH is quite similar between men and women (38.9% and 39.9%, respectively) (Table 1). Looking at the entire sample, the distribution of EH is below 21% in the high-educated group but above 51% in the low-educated group for both genders.

Although the low-educated groups are systematically more affected by EH, there are a few variations. The lowest levels of EH for the low-educated, between 42% and 44%, are observed for both males and females in the oldest age group; however, among the group between 30-54 years of age, almost two-thirds of the low-educated report suffering EH. The high-educated groups are the least affected by EH. Twenty-one point 9 percent of high-educated Spanish men 30-54 years of age say they suffer EH, while for those aged 55 to 79 the percentage suffering EH declines to 12.8%. In contrast, the differences by age are lower among high-educated Spanish women (21.8% for the younger age group and 15.4% for the older). 

Disability prevalence by gender across educational groups

Graph 3 shows the varying prevalence of AL  in the three educational groups for each gender and age-group. We represented the relative proportion of the educational groups within the population by the size of the circles: low-educated groups are generally larger than the middle or high-educated ones and are similar for both men and women. In the 30-79 age groups, low-educated Spanish women report significantly higher prevalences of AL than low-educated Spanish males. However, we do not detect significant differences by gender across the rest of the educational groups.

When we stratify by age-groups, results reveal different patterns. Lower levels of AL are visible for the younger age group for both genders and at all education levels. Low-educated groups are generally larger in the older age-group (55-79) than in the younger one (30-54). In both age-groups, low-educated Spanish men report lower levels of AL than low-educated Spanish women. There are no significant differences in the middle-educated group by gender, both men and women reporting similar levels of AL. Irrespective of gender, we identify the lowest levels of AL with the most educated. For the high-educated group, we only see significant differences by gender in the older age group (55-79), with women reporting lower levels of AL than men. 

Association of economic hardship and disability by gender in Spain

We find a positive association between EH and AL (all coefficients are higher than 1), when age and level of education are taken into account (Graph 4). EH is positively associated with AL in men and women in all age groups. The highest association between them is found among women in the older age group, and the lowest among men in the younger age group.

Contribution of economic hardship to the association between education and disability by gender

To explain the variations in activity limitation, we examine the role of poverty (using our definition of EH) across educational groups. Indeed, poverty is a major factor in poor health and income inequality correlates with health inequality.

Graph 5 shows the contribution (in percent) of EH to the AL disadvantage or advantage for the low- and high-educated (compared to the middle-educated), as well as the frequency of EH (indicated by the size of the circle for each group). EH contributes up to 23% to the differences in AL between low and middle-educated Spanish men; and 18%  to the differences in AL between low and middle-educated Spanish women (low-education increases the disability probability compared to middle-education by +0.44 for men (0.10 indirect effect) and +0.58 for women (0.10 indirect effect). 

EH contributes up to 32% and 57% to the differences in AL between low and middle-educated Spanish men and women, respectively (high education decreases the disability probability compared to middle-education by -0.29 for men (-0.09 indirect effect) and -0.28 for women (-0.16 indirect effect)).

Economic hardship contributes up to 57% to the differences in activity limitation between middle and high-educated Spanish women.

Even though the distribution of EH is quite similar between men and women across education groups, the  contribution of economic hardship to the differences in AL found between the high- vs middle educated  is greater for Spanish women (57%) than Spanish men (32%). This may be explained by the slightly higher difference in EH between middle and high educated women (41.8% vs 20.7%) compared to middle and high educated men (36.6% vs 19.4%) and by different social factors determining disability in men and women. 

In addition, the contribution of EH to differences in activity limitation for the low- and middle educated is more similar among genders (23% for Spanish men and 18% for Spanish women) and could be considered moderate. The EH difference between low- vs middle educated men is 15%, and 10% between low- vs middle educated women. The higher level of EH for men than women might be explained by the slightly different contribution for low-educated men than low-educated women vs middle educated. There are other factors that contribute more than EH to the differences in AL for the low educated Spanish population.

4. Conclusions

The level of education, the level of economic hardship and gender all influence health differences in Spain. The number of low-educated women is usually greater across all age groups than the number of low-educated men; and low-educated groups consistently show the highest prevalence of activity limitation and high-educated groups the lowest. By gender, low-educated Spanish women report significantly higher prevalences of AL than low-educated Spanish men. However, we do not find significant differences by gender across the rest of the educational groups, which report similar levels of AL.

The distribution of economic hardship does not change much by gender across educational groups, but it does across age groups. Spanish women in the younger age group with low education report the highest incidence of EH. In fact, irrespective of education level, Spanish adults in the younger age group report higher levels of EH than those in the older age group. 

Irrespective of education level, young Spanish adults report higher levels of economic hardship than older adults. Young, low-educated Spanish women report the highest levels of economic hardship.

Overall, material deprivation contributes to the association between education and disability but the contribution differs. The lowest EH contribution to the differences in AL is between low and middle-educated Spaniards, both men and women. This suggests that other social factors not studied here, such as access to care, health-related behaviors, employment and working conditions, may also contribute to reported differences in AL in the most disadvantaged group (the low educated). The largest contribution from economic hardship to differences in AL is found between high and middle-educated women (57%), where we also find a difference in the prevalence of EH between high- and middle educated women of 21.1 percentage points. 

Irrespective of gender, we identify the lowest levels of activity limitation with the most educated.

This finding suggests that even though EH and disability are less frequent for the most educated (persons that report these problems have very specific features), poverty also matters in assessing differences between high-educated and middle-educated Spanish women, although different social factors determining disability in men and women might contribute to this high contribution.

Economic hardship might be rare or common depending on both the economic situation of the country and the level of protection from material deprivation. The reverse causality in the association between material deprivation and disability might be found where the risk of poverty is largely reduced by policy schemes; the contribution to social differences might be important as being strongly and directly linked to disability. Also, its contribution might be small in percentage because other factors (variation in health-related behaviors, different chances to cope with health problems) might contribute to produce differences.

Our findings for Spain suggest that schemes for reducing poverty might play an important role all along the educational gradient; however, the return would be moderate as long as the part of the differences in the frequency of disability explained by economic hardship is small. Actions to reduce poverty will help to reduce the differences in disability between socioeconomic groups and  increase healthy life expectancy.

Aïda Solé-Auró, Department of Political and Social Sciences
Universitat Pompeu Fabra, Barcelona

5. References

Ballarino, G., F. Bernardi, M. Requena and H. Schadee (2009): “Persistent inequalities? Expansion of education and class inequality in Italy and Spain”, European Sociological Review, 25(1).

Cambois, E., A. Solé-Auró, H. Brønnum-Hansen, V. Egidi, C. Jagger, B. Jeunee, W.J. Nusselder, H. Van Oyen, C. White and J.-M. Robine (2016): “Educational differentials in disability vary across and within welfare regimes: a comparison of 26 European countries in 2009”, Journal of Epidemiology and Community Health, 70(4).

Chetty, R., M. Stepner, S. Abraham, L. Shelby, B. Scuderi, N. Turner, A. Bergeron and D. Cutler (2016): “The association between income and life expectancy in the United States, 2001-2014”, The Journal of the American Medical Association – JAMA, 315(16).

Eikemo, T.A., M. Huisman, C. Bambra and A.E. Kunst (2008): “Health inequalities according to educational level in different welfare regimes: a comparison of 23 European countries”, Sociology of Health & Illness, 4.

Fuente, A. de la, and R. Doménech (2015): “Educational attainment in the OECD, 1960-2010. Updated series and a comparison with other sources”, Economics of Education Review, 48.

Hayward, M.D., and B.K. Gorman (2004): “The long arm of childhood: the influence of early-life social conditions on men’s mortality”, Demography, 41(1).

Mackenbach, J.P., I. Stirbu, A.J. Roskam, M.M. Schaap, G. Menvielle, M. Leinsaly and A.E. Kunst (2008): “Socioeconomic inequalities in health in 22 European countries”, The New England Journal of Medicine, 23.

Marmot, M. (2002): “The influence of income on health: views of an epidemiologist”, Health Affairs (Millwood), 21(2).

Montez, J.K., R.A. Hummer and M.D. Hayward (2012): “Educational attainment and adult mortality in the United States: a systematic analysis of functional form”, Demography, 49(1).

Rodgers, G.B. (1979): “Income and inequality as determinants of mortality: an international cross-section analysis”, Population Studies, 33(2).

Solé-Auró, A., and E.M. Crimmins (2013): “The oldest old: health in Europe and the United States”, in J.-M. Robine, C. Jagger and E.M. Crimmins (eds.): Healthy longevity. A global approach, Annual Review of Gerontology and Geriatrics, New York: Springer.

Whealan, C., and B. Maître (2013): “Material deprivation, economic stress, and reference groups in Europe: an analysis of the EU-SILC 2009”, European Sociological Review, 29(6).



Aïda Solé-Auró, Department of Political and Social Sciences,
Universitat Pompeu Fabra, Barcelona



Related contents


Una propuesta de renta fiscal universal para España

La propuesta de una renta fiscal universal busca convertir el mínimo vital definido en el IRPF en una auténtica renta garantizada para asegurar a todos los ciudadanos un mínimo nivel de ingresos.


Cambio tecnológico y renta básica

El cambio tecnológico parece favorecer un aumento de la polarización salarial y de la desigualdad. En este contexto, la renta básica universal se vislumbra como una medida para compensar a los más desfavorecidos por los cambios en los modos de producción.


Los complementos salariales y la garantía de ingresos: posibilidades y límites

¿Se puede garantizar una renta a toda la sociedad asignando algún tipo de complemento a los salarios? Repasamos las principales características de los complementos salariales a partir de distintas experiencias, especialmente anglosajonas.


Problemas de incentivos: renta básica universal versus prestación de ingresos mínimos

Análisis comparado de la renta básica y la prestación de ingresos mínimos. Se analizan los efectos de ambas sobre la oferta de trabajo, los costes de financiación o los problemas de englobar estas medidas en el contexto internacional. 


Renta básica (incondicional). Por qué es importante para la libertad y cómo se puede financiar

En este capítulo se realiza un análisis de la renta básica atendiendo a dos cuestiones esenciales: ¿es justa? y ¿es viable? Se aborda principalmente la relación de la renta básica con la libertad del individuo. 

You may also like


Reforzar el bienestar social: del ingreso mínimo a la renta básica

Reforzar el bienestar social: del ingreso mínimo a la renta básica

Social Inclusion

Este informe recoge el análisis sobre la situación social en España y sobre la capacidad de las ayudas y subsidios existentes para garantizar unos ingresos mínimos a todos los ciudadanos, con el objetivo de reducir la pobreza y la desigualdad.




Social Inclusion

This report analyses four challenges relating to health in Spain: improving the population’s health, promoting healthy lifestyle habits, guaranteeing access to healthcare and the viability of services for dependency.




Social Inclusion

What social challenges does decent housing represent in Spain? This report analyses three challenges in this field: access, conditions and energy needs.