Skip to main content
NCBI home page
European Journal of Population = Revue Européenne de Démographie logoLink to European Journal of Population = Revue Européenne de Démographie
. 2018 Nov 26;35(4):777–793. doi: 10.1007/s10680-018-9504-2

Gender Differences in Disability and Economic Hardship in Older Europeans

Emmanuelle Cambois 1,✉, Aïda Solé-Auró 2, Jean-Marie Robine 3
PMCID: PMC6797682  PMID: 31656461

Abstract

European women live longer, but they experience more old age-related disability than men. Disability is related to social factors, among which is poverty, through various pathways. While women’s poverty has been pointed up as a challenge for Europe, our study investigates to what extent and in which countries a greater exposure to economic hardship is associated with older women’s disability disadvantage. We used the 2014 EU-SILC data in 30 European countries for men and women aged 50–79 years (N = [1179–17,474]). Disability was measured by self-reported activity limitation and economic hardship by difficulties in “making both ends meet” and “facing unexpected expenses”. Country-specific nested logistic regressions measured the women’s disability disadvantage and its association with economic hardship. We found that activity limitations and economic hardship varied substantially across Europe, being the lowest in Sweden and Norway. We found gender gaps in activity limitations in 23 countries, always to women’s disadvantage. After adjusting for age, this disadvantage was significant in 19 countries. In 11 of these countries, women’s excess disability is associated with excess economic hardship in women, especially in Iceland, France, Sweden, and Austria. Women’s excess disability and social factors such as economic hardship are linked, even in protective countries. These situations of double disadvantage for women deserve attention when designing policies to reduce health inequalities and to promote healthy ageing.

Electronic supplementary material

The online version of this article (10.1007/s10680-018-9504-2) contains supplementary material, which is available to authorized users.

Keywords: Ageing, Disability, Gender, Poverty, Health inequalities, Europe

Introduction

The challenge of the longevity revolution is to maintain high levels of good health and quality of life, along with the gains in life expectancy. There is a well-known gender gap in this respect, often seen as a paradox: women’s longevity advantage usually translates into a disadvantage in terms of health and disability (Van Oyen et al. 2013; Dahlin and Härkönen 2013; Oksuzyan et al. 2010; Luy and Minagawa 2014). Women’s disability disadvantage is largely explained by a higher prevalence of disabling conditions in women and of lethal diseases in men (Crimmins et al. 2011). Gender studies have therefore explored the factors behind these health differences, and more specifically social factors, whereas different life courses and living conditions yield different socially related exposures to health risks (Arber 2004; McMunn et al. 2006; Solé-Auró and Alcañiz 2016; Cambois et al. 2017).

Among various social factors, the lack of financial or material resources is of particular interest and is approached here through the concept of economic hardship. Besides its strong link to health, there is a general concern that women might be more exposed to economic hardship than men over their life course. Indeed, with the persistent gender division of paid and unpaid work, women have a lower average own income than men; head single parent households more frequently; and are more exposed to poverty risks (Anxo et al. 2011; OECD 2012; Van Lancker et al. 2015). Interestingly, the overall level of economic hardship and the gender gap in that matter reflect not only a country’s social organization but also its social protection system. Changes in macroeconomic characteristics modify the level of poverty risks and their association with health (anti-poverty programmes, female empowerment, the unemployment rate, etc.) (Bárcena-Martín and Moro-Egido 2013). Protective contexts may reduce the poverty level in a country and therefore reduce the consequences of poverty on health. Indeed, there is some evidence that specific protective welfare regimes with gender equity dimensions reduce gender health gaps (Borrell et al. 2014; Palencia et al. 2014; Van Oyen et al. 2010). However, while high levels of protection and anti-poverty policies are expected to reduce risks of economic hardship and poor health, they do not always equally protect all population groups (Dahlin and Härkönen 2013). As a matter of fact, protective regimes do not systematically disclose the smallest social differences in health and disability (Huijts and Eikemo 2009; Cambois et al. 2016a). Across European countries, the percentage of the population concerned by poverty and disability varies depending on the overall welfare context; so might the coincidence of these two situations. Such a double disadvantage reflects situations of vulnerability in terms of social network, labour force participation, access to services and programmes or so. While women experience a disability disadvantage in most European countries and seem to be more broadly affected by the risk of poverty, an analysis of the double disadvantage might bring some light to the literature, as both the reduction of women’s poverty and healthy ageing are considered to be major European challenges in our time (Bárcena-Martín and Moro-Egido 2013; Rechel et al. 2013).

In this paper, we explore the relation between the higher disability prevalence among women and the gender gap in economic hardship for people aged 50–79 years across 30 European countries. We aim to assess whether and where in Europe women’s disability disadvantage coincides with a disadvantage in economic hardship.

Gender Gap in Economic Hardship and Women’s Disability Disadvantage

Few resources and economic hardship are key issues in public health. They are generally linked to poor health and high mortality, and possibly in different ways for men and women (Marmot 2010). In Spain, for instance, economic hardship differs across sexes and causes of death (Benach et al. 2001). How and why are economic hardship and health linked?

First, through a causal effect, economic hardship may increase health risks due to poor living conditions and, in a life course perspective, through its long-term effects. It limits access to basic goods and services that are beneficial to health, such as nutritious food, safe housing, and healthcare (Marmot 2002; Wilkinson and Pickett 2006; Fjaer et al. 2017), and it limits opportunities to avoid or adjust to health risks (Kangas and Blomgren 2014; Okechukwu et al. 2012). Economic hardship may both increase exposure to disabling diseases and make it more difficult to limit their impact on activities of daily living (Verbrugge 1995; Cambois and Jusot 2011). Second, disability may limit access to the labour market and/or other resources. This reverse causation path was found to be moderate, compared to the causal effect (Lynch et al. 1997). Yet, it is possibly of a different magnitude across genders. Women in poor health or with disabilities may receive less social support than their male counterparts for acquiring minimum resources or obtaining a job and income: among men and women in poor health or with severe disability, women were found less likely than men to be or to remain in the labour force (Moen and Chermack 2005; Stronks et al. 1995) and more likely to have inadequate resources and housing (Reine et al. 2016). This could be partly explained by gender differences in access to and take-up of social benefits and support, even in countries where the welfare system is reputed as supportive (Hernanz et al. 2004). Third, the association between economic hardship and poor health might result from common determinants that affect both health and individual resources, such as strenuous low-paid work or adverse life events (union dissolution, job loss, conflict, etc.). Therefore, because women have lower income and are more exposed to economic hardship, they could be more exposed to disability. Also, because women report higher disability, they could be more exposed to the associated loss of resources and income. Overall, women’s excess economic hardship and disability could be associated differently depending on the country context and the level of protection against poverty, disability, and gender differences.

European Variation in the Gender Gaps in Disability and Economic Hardship

To contribute to the literature, we explore how the excess risk of economic hardship in women is linked to poorer functional health in 30 European countries. It is unclear from the literature whether the gender distributions of disability and of hardship are similarly unbalanced in all European countries and whether these situations of disadvantage coincide.

Several hypotheses can be formulated. Anti-poverty policies are also now considered to be drivers for reducing health inequalities (Mackenbach et al. 2013; Bergqvist et al. 2013). Studies have shown a long-term association between policy schemes and health as well as gender health gaps (Avendano et al. 2015; Berkman et al. 2015; Palencia et al. 2014; Borrell et al. 2014). We assume that the levels of disability and economic hardship are diverse across Europe, with lower overall levels in countries with high levels of protection. However, it is less clear where the largest or smallest gaps within European countries should be found. Large variations in social health gaps are found across countries within a given type of welfare regime, such as the protective Nordic regimes (Eikemo and Bambra 2008; Brennenstuhl et al. 2012; Eikemo et al. 2008; Mackenbach et al. 2008; Avendano et al. 2009). Based on the literature, we assume that high levels of protection do not always prevent gender differentials in disability and in economic hardship. We also assume that in countries with low levels of social protection, the high risk of poverty is not systematically unbalanced between men and women and therefore not strongly involved in the gender gap in disability.

To highlight how gender gaps in poverty risks and disability are articulated across Europe, we first explore whether women report more disability and more economic hardship than men. Second, in countries where women’s disadvantages are found, we estimate the extent to which their disability disadvantage was associated with an excess risk of economic hardship.

Methods

Data Set and Indicators

Data Set

The European Union Statistics on Income and Living Conditions (EU-SILC) is a database managed by national statistical offices and designed to provide comparable data across the EU. We used the 2014 EU-SILC cross-sectional data on individuals living in private households. In most countries, data are collected via ad hoc interview surveys that provide self-reported information on health and socioeconomic variables. Elsewhere, sociodemographic variables are collected through population registers, and health data are collected via a complementary survey, often using telephone interviews. EU-SILC’s response rate is generally high. However, some countries disclose a low participation at the individual level, while the information at the household level is acceptable (Table 1). This is generally the case in countries where most of EU-SILC variables are collected in population registers and a number of household and individual variables are collected by an additional ad hoc survey, conducted by telephone on a sample. Telephone surveys have higher risk of non-participation due to contact failure: this is the case of the Scandinavian countries, the Netherlands and Slovenia. Moreover, if a household member other than the intended individual provides the household information, answers to individual variables, such as health, were not allowed. This framework explains the large, missing rates in these countries. However, an exploratory study based on an earlier wave of EU-SILC found a reliable representativeness of the samples (related to age and occupational structure of the populations) when using the appropriate weighting (Cambois et al. 2016a). For this current work, we have further explored whether the missing answers on the disability question were linked to the main variables under consideration (see Sect. 4, below).

Table 1.

EU-SILC 2014 country sample size (50–79 years) and total participation rates.

Source: EU-SILC 2014; http://www.gesis.org/en/missy/materials/EU-SILC/documents/quality-reports

Countries Study sample of men and women 30–79 years old who answered the individual question (disability) Overall EU-SILC rates of participation and missing variables on disability
Men Women All Householdsa (%) Individualsb (%) Missingsc (%)
Austria AT 2245 2520 4765 77 99 0
Belgium BE 2295 2523 4818 65 99 1
Bulgaria BG 235 2925 5275 86 99 0
Cyprus CY 1906 2135 4041 91 100 0
Czech Republic CZ 2523 3632 6155 91 100 17
Germanyd DE 5881 6365 12,246 – – 1
Denmarkf DK 1538 1588 3126 73 51 47
Estonia EE 2345 2901 5246 82 87 0
Spain ES 5284 5853 11,137 79 98 0
Finlande FI 2602 2623 5225 80 100 47
France FR 4329 4966 9295 83 96 1
Greece GR 3898 4339 8237 83 99 0
Croatia HR 2789 3302 6091 – – 1
Hungary HU 3556 4899 8455 85 100 0
Ireland IE 2056 2173 4229 – – 0
Icelande IS 579 600 1179 71 100 53
Italy IT 8255 9219 17,474 79 99 4
Lithuania LT 2436 3084 552 89 98 1
Luxembourg LU 1551 159 3141 53 96 0
Latvia LV 2117 338 5497 – – 1
Malta MT 2277 2596 4873 89 98 0
Netherlande NL 2357 2744 5101 88 93 45
Norwaye NO 1695 1514 3209 82 100 47
Poland PL 5497 7039 12,536 81 99 4
Portugal PT 3126 3688 6814 – – 0
Romania RO 3433 4079 7512 97 100 0
Swedene SE 1297 1375 2672 – – 47
Sloveniad SI 1903 2206 4109 – – 58
Slovakia SK 2273 3082 5355 92 100 2
United Kingdom UK 3745 4164 7909 61 100 0
All 88,138 103,104 191,242
Open in a new tab

– No quality report available for 2014

aParticipation rate at the household level

bCoverage of the individual sample within the total household sample

cMissing on health variables

dSelf-administred collection mode

eTelephone assisted (CATI)

fSelf-administered and CATI mode (others are totally or mainly face-to-face computer or paper collection)

Once the 12% of individuals with missing values on disability are discarded, our study comprises 191,242 individuals (88,138 men and 103,104 women) aged 50–79 from 30 European countries. (Individuals aged 80 and over were excluded as they are very likely to be living in a care facility after this age, with variations by country, gender, and socioeconomic situation.)

Health-Related Global Activity Limitations

Disability was measured by the Global Activity Limitation Indicator (GALI) with the 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 vs Not limited). GALI was self-reported and differed across European countries, partly due to a varying propensity to report health disorders and to differences in survey design (Berger et al. 2015). However, this indicator proved to be highly correlated with other measures of functional difficulties as well as predictive of later healthcare consumption and mortality risks (Cabrero-Garcia and Julia-Sanchis 2014; Jagger et al. 2010; Van der Heyden et al. 2015).

Economic Hardship

Economic hardship was assessed by a thematic module on self-perceived situations of deprivation, as an alternative to an income-based measure in international comparative studies (Whelan and Maître 2013). In this study, we focused on the “economic stress” dimension of the thematic module. We used two items (due to a high level of non-response on the third one): people who reported that it was difficult (very or somewhat) for them “to make ends meet, namely, to pay for usual necessary expenses” and “to cope with unanticipated expenses” were considered to be in a situation of economic hardship (also referred to hereafter as “hardship”).

Analysis

First, we assessed the gender differences in GALI prevalence and the gender gap in hardship. Gender-specific logistic regressions confirmed the association between hardship and GALI, adjusting for age and age squared. Second, for countries where we found a gender gap in both hardship and GALI, we examined whether they were associated. To test this hypothesis, we used nested logistic regressions models for each country (“KHB”) (Karlson et al. 2012). With nested logistic models, one can quantify the change in the association between the explanatory variable (gender) and the outcome (GALI) before and after controlling for a potential mediator (hardship). This change corresponds to the indirect effect of the explanatory variable also considered as mediator in these models. Here, we assessed the total effect of gender (adjusted on age and age squared) and its indirect part mediated by high risk of hardship (the total effect being the sum of the direct and indirect effects) (Fig. 1). The contribution of hardship corresponds to the ratio of the indirect effect to the total effect. It expresses the extent to which gender difference in hardship is associated with the women’s excess disability.

Fig. 1.

Fig. 1

Open in a new tab

Illustration of the total and direct effect of gender on activity limitation and the indirect effects through economic hardship

Supplementary analyses were conducted to consolidate the results. The age range is large and combines different work and health situations, due to retirement and different income situations. We repeated the analysis, separating the pre- and post-retirement ages to account for the variation in situation across the life course (Figure S1 in online supplementary material). We also replicated the models, considering a possible association between economic hardship and disability mediated by the household composition (living alone; with a spouse; with or without a child; and having a partner/never partnered; separated; widowed) or by partnership status (in a union; never partnered; separated; widowed) (Figure S2 in online supplementary material).

Results

Gender Gaps in Activity Limitations or in Economic Hardship

Regarding our first question, Table 2 reports the gender-specific prevalence of GALI in the 30 countries, ranging from above 55% in Latvia, Portugal, and Slovakia to 20% or less in Malta and Sweden in women, and from about 50% in Germany, Estonia, Latvia, and Slovakia to 20% or less in Norway, Malta, Iceland, and Sweden in men. Women report more GALI than men in all countries, but the female-to-male GALI ratio is not statistically significant in six countries (≤ 10% confidence interval). The GALI W/M ratios are relatively small in Austria, Czech Republic, Latvia, and Poland. Interestingly, the largest GALI W/M ratios are found in countries with relatively low overall levels of GALI, such as Sweden. However, the ratios are also large in the Netherlands and Romania, where men’s prevalence reaches 32.2% and 39.1%, respectively.

Table 2.

Weighted prevalence of activity limitation and economic hardship.

Source: EU-SILC 2014

Countries Self-reported activity limitation Self-reported economic hardship ORs of AL for EHa
Men (%) Women (%) W/M Men (%) Women (%) W/M Men Women
Austria AT 42.7 [40.6–44.7] 46.2 [44.2–48.1] 1.08§ 14.6 [13.2–16.1] 17.3 [15.8–18.8] 1.18§ 3.39 3.71
Belgium BE 27.8 [26.0–29.7] 34.3 [32.4–36.1] 1.23** 14.2 [12.8–15.6] 17.1 [15.6–18.5] 1.20§ 2.83 4.11
Bulgaria BG 24.7 [23.0–26.4] 28.4 [26.8–30.0] 1.15* 50.4 [48.4–52.5] 57.2 [55.4–59.0] 1.14** 1.53 1.48
Cyprus CY 29.5 [27.5–31.6] 33.6 [31.6–35.6] 1.14§ 51.3 [49.0–53.5] 55.0 [52.8– 57.1] 1.07§ 2.75 2.56
Czech Republic CZ 31.8 [30.0–33.6] 34.3 [32.7–35.8] 1.08§ 30.6 [28.8–32.4] 38.7 [37.2–40.3] 1.27** 2.12 2.20
Germany DE 49.2 [47.9–50.5] 50.8 [49.6–52.0] 1.03 13.2 [12.3–14.1] 14.4 [13.5–15.3] 1.09 3.45 3.79
Denmark DK 36.2 [33.8–38.6] 37.8 [35.4–40.2] 1.04 12.2 [10.5–13.8] 14.7 [13.0–16.4] 1.21 1.89 4.32
Estonia EE 50.7 [48.7–52.8] 52.5 [50.7–54.3] 1.03 35.9 [34.0–37.8] 40.2 [38.4–42.0] 1.12* 2.34 2.82
Spain ES 31.2 [30.0–32.5] 36.4 [35.1–37.6] 1.16** 35.4 [34.1–36.7] 39.0 [37.8–40.2] 1.10** 1.96 2.45
Finland FI 39.6 [37.7–41.5] 44.2 [42.3–46.1] 1.12* 13.7 [12.4–15.0] 15.6 [14.2–17.0] 1.14 2.49 2.36
France FR 30.5 [29.2–32.0] 35.0 [33.8–36.5] 1.15** 21.3 [20.0–22.5] 25.8 [24.6–27.0] 1.21** 2.08 2.50
Greece GR 34.4 [32.9–35.9] 40.1 [38.6–41.5] 1.17** 48.9 [47.4–50.5] 51.0 [49.5–52.5] 1.04 1.77 1.65
Croatia HR 42.9 [41.1–44.8] 49.3 [47.6–51.0] 1.15** 61.5 [59.7–63.3] 65.4 [63.7–67.0] 1.06* 1.45 1.50
Hungary HU 38.6 [37.0–40.2] 44.6 [43.2–46.0] 1.16** 65.6 [64.0–67.1] 67.9 [66.6–69.2] 1.04 1.81 2.25
Ireland IE 25.9 [24.0–27.8] 27.5 [25.6–29.4] 1.06 41.8 [39.7–44.0] 42.1 [40.1–44.2] 1.01* 3.30 3.08
Iceland IS 19.7 [16.5–23.0] 28.2 [24.5–31.7] 1.43* 16.8 [13.7–19.8] 28.4 [24.7–31.9] 1.69** 3.90 2.30
Italy IT 38.1 [37.1–39.2] 42.7 [41.7–43.7] 1.12** 31.8 [30.8–32.8] 35.1 [34.1–36.1] 1.10** 2.47 2.08
Lithuania LT 35.3 [33.4–37.2] 42.4 [40.7–44.2] 1.20** 51.6 [49.6–53.6] 56.4 [54.6–58.2] 1.09* 2.18 2.38
Luxembourg LU 32.2 [29.9–34.5] 37.1 [34.7–39.5] 1.15* 15.0 [13.3–16.8] 13.2 [11.6–14.9] 0.88 2.68 3.32
Latvia LV 52.9 [50.7–54.9] 57.3 [55.7–59.0] 1.08* 63.3 [61.4–65.5] 69.2 [67.7–70.8] 1.09** 1.96 1.83
Malta MT 17.1 [15.6–18.7] 20.0 [18.5–21.6] 1.17§ 17.9 [16.3–19.5] 22.0 [20.4–23.6] 1.23** 2.47 2.09
Netherland NL 32.2 [30.3–34.1] 45.7 [43.8–47.6] 1.42** 14.5 [13.1–15.9] 19.1 [17.6–20.6] 1.32** 3.88 3.35
Norway NO 18.1 [16.3–20.0] 26.0 [23.8–28.2] 1.43** 4.7 [3.6–5.7] 8.2 [6.8–9.5] 1.75** 4.87 3.03
Poland PL 36.2 [35.0–37.5] 38.7 [37.6–39.9] 1.07* 44.4 [42.9–45.5] 48.7 [47.4–49.7] 1.10** 1.90 1.73
Portugal PT 43.6 [41.8–45.3] 57.5 [55.9–59.1] 1.32** 35.9 [34.2–37.6] 41.1 [39.5–42.7] 1.14** 2.33 2.32
Romania RO 39.1 [37.4–40.7] 51.1 [49.6–52.6] 1.31** 48.5 [46.8–50.1] 51.7 [50.1–53.2] 1.07* 1.52 1.56
Sweden SE 11.7 [10.0–13.5] 17.4 [15.4–19.4] 1.48** 6.0 [4.7–7.3] 10.2 [8.6–11.8] 1.69** 15.18 3.27
Slovenia SI 43.5 [41.3–45.7] 45.8 [43.7–47.9] 1.05 44.1 [41.9–46.4] 50.7 [48.6–52.8] 1.15** 2.59 2.51
Slovakia SK 49.5 [47.4–51.5] 56.0 [54.3–57.8] 1.13** 33.0 [31.5–34.9] 34.6 [32.9–36.2] 1.05 1.65 1.52
United Kingdom UK 32.1 [30.6–33.6] 34.1 [32.7–35.6] 1.06 21.1 [19.8–22.4] 24.5 [23.2–25.9] 1.16* 3.86 2.87
Open in a new tab

Odds ratios of activity limitation associated with economic hardship (adjusted on age). Men and women aged 50–79 in 30 European countries

AL activity limitation, EH economic hardship

**p > 0.01; *p > 0.05; §p > 0.10. aStatistically significant ORs (95%)

Table 2 also shows that economic hardship’s prevalence ranges from 10% or less in Norway and Sweden, to half or more of the population in some eastern and southern countries, and up to 69% of women in Latvia. Women report significantly more hardship than men in 23 countries. The gender ratio is not statistically significant in countries with a generally high level of hardship; the largest ratios are in Iceland, as well as in Norway and Sweden, where the overall level of hardship is the lowest.

Finally, based on the odds ratios from logistic regression, we confirmed that economic hardship was significantly associated with GALI for both genders in all 30 European countries (Table 2).

The Women’s Disability Disadvantage and Its Association with Economic Hardship

Regarding our second question, nested logistic regressions modelled GALI to measure the gender gap in disability, controlling for age and age squared without, and then with, adjusting for economic hardship. The gap is not significant in UK, Bulgaria, Czech Republic, Latvia, and Malta (in addition to the above-mentioned six countries where the GALI ratio—not adjusted for age—was not significant). Yet, the GALI gender gap remains significant in 19 countries (Fig. 2). In 11 of them, the gender gap in economic hardship is significantly associated with the gender gap in GALI. In other words, in these countries, more women with disability go hand in hand with more women experiencing economic hardship. On average, 16% of the gender gap in disability relates to the gender gap in hardship (Fig. 2B). When accounting for the higher risk of hardship in women in these countries, the excess GALI prevalence associated with being a woman is reduced by 16%. Not surprisingly, gender differences in economic hardship are far from being the main factor associated with the gender disability gap on average. However, this result confirms that social factors, even as specific as economic hardship, are part of the gender health differences. Furthermore, this average hides larger associations. In Austria, the GALI gender gap is relatively small, yet up to 31% of this gap is linked to excess economic hardship in women. The association is also above average in Iceland, Sweden, and France. In contrast, the association exists but is below average in the Netherlands, Norway, and Portugal, and close to the average in the four remaining countries.

Fig. 2.

Fig. 2

Open in a new tab

Women’s disability disadvantage. A Total and direct effect of gender and indirect effect of gender mediated by economic hardship. B Contribution of indirect effect mediated by economic hardship to the total effect of gender. Individuals aged 50–79 years in 30 European countries. NS non-statistically significant (p < 0.10). Country labels: Austria (AT), Belgium (BE), Bulgaria (BG), Cyprus (CY), Czech Republic (CZ), Germany (DE), Denmark (DK), Estonia (EE), Spain (ES), Finland (FI), France (FR), Greece (GR), Croatia (HR), Hungary (HU), Iceland (IS), Ireland (IE), Italy (IT), Lithuania (LT), Luxembourg (LU), Latvia (LV), Malta (MT), the Netherlands (NL), Norway (NO), Poland (PL), Portugal (PT), Romania (RO), Sweden (SE), Slovenia (SI), Slovakia (SK), United Kingdom (UK).

Source: EU-SILC 2014

Supplementary Analysis

We replicated the analysis in the two groups aged 50–64 years and 65–79 years, separately (online supplementary material Figure S1). Regarding the gender differentials in GALI, most results are similar in these two age groups; however, they are generally less significant in the youngest age group. Belgium and Greece tend to be in atypical situations, with a larger gender disability gap in the 50–64 age group than in older ages. Focusing on older ages, we found that older Belgians, Bulgarians, Germans, and Estonians show smaller gender differentials in GALI than the average, as do Austrians. However, the indirect effect of economic hardship in the gender gap is not significant in any country when disaggregating by age groups (online supplementary material Figure S1). We also replicated the model, accounting for household composition and partnership status. We found little change in the association between economic hardship and women’s GALI disadvantage (Supplementary material Figure S2). The only changes consisted in the association of economic hardship with gender gap in GALI becoming significant in Cyprus, Finland, and Greece (and non-significant in Hungary), although with no substantial change in the association with gender gap in GALI.

Discussion

Our study first shows that gender difference in GALI was significant in 24 out of the 30 countries and remarkably large in Iceland, the Netherlands, Norway, Portugal, Romania, and Sweden, countries with very different overall levels of GALI. After controlling for age, women remain disadvantaged in 19 countries. The gender gap in GALI is generally not found in countries with relatively high overall levels of GALI (several Eastern European and Baltic countries and Germany, but also in Denmark and Ireland). We also find that women are generally more affected by economic hardship than men, except in countries with high overall levels and in Luxembourg. These results confirm our assumptions that large gender differentials are found in some of the countries with a protective welfare regime and not found in all countries with low protection regimes.

Second, our analyses demonstrate that in 11 of these 19 countries, the gender gap in GALI is significantly associated with the gender gap in economic hardship. Women’s greater GALI partly coincides with greater economic hardship for them in these 11 countries. Our results are challenging because these 11 countries have very different levels of disability and economic hardship. We found that a variation of the percentage of the gender gap in GALI linked to excess economic hardship in women was rather low on average but reached up to 31% in Austria, a country with a small gender gap. The association is also greater than the average in Iceland, Sweden, and France, countries with very different policies and modes of social organization. Therefore, we can conclude that social factors, such as material conditions, and even as specific as economic hardship, are associated with women’s disadvantage in disability. This was found in 11 of the 30 countries under consideration, where women’s disadvantage was also shown in terms of hardship. The association does not appear to be large in most of the countries concerned; however, this can be explained by the fact that economic hardship is quite scarce in a number of countries. Other social factors might be considered in interpreting these results further. This first approach is encouraging suggesting to explore the association in greater detail, using a more precise dataset to better describe populations that experience economic hardship: for instance, in looking at the level of education, which is also known to be at the intersection of economic hardship and disability (Cambois et al. 2016b).

As suggested in Introduction, the association of gender gaps in economic hardship and disability may be due to (a) more frequent situations of hardship in women, which translates into increased risk of disability, (b) more women with disability than men being at risk of economic hardship, and (c) women being more exposed than men to determinants of both disability and hardship. The coincidence of women’s disadvantage in disability and in economic hardship might be due to insufficient protection to ensure a minimum income and/or to access basic goods and services in situations of vulnerability which women experience more often than men (single parenthood, unemployment, widowhood, etc.). It could also be due to situation of deteriorated health exposing more women than men to deprivation. This is consistent with the Swedish study showing that women with severe impairments were less likely than men to be employed (Reine et al. 2016); whereas both deprivation and disability are scarce in this country, we may consider women being more at risk of accumulating health and income difficulties. Therefore, Sweden shows an unexpected association between the gender gap in economic hardship and the gender gap in disability; it seems that a selection of women faces a double disadvantage. However, these results need to be interpreted with caution due to issues of sample size in countries where economic hardship concerns less than 10%, such as in Norway or Sweden. There are also considerations related to the survey sample (see below).

The gender gap in disability results from gender-specific exposures to disabling diseases and differential access to resources for coping with their consequences. The gender differences in exposure to economic hardship might partly result from gender-unequal access to and/or participation in the labour market, differences in individual income (Borrell et al. 2014) and household situations (separated/divorced, widowed, or single parent) (Berkman et al. 2015). However, regarding the latter aspect, the model comprising partnership status and household composition did not change the results. Finally, at the country level, the gender differences in economic hardship and disability result from the availability of healthcare and social protection systems, combined with possible gender differences in access to and take-up of their benefits (Fjaer et al. 2017).

The literature shows that public policies promoting gender equity, in terms of work and career, reduce gender differentials in poverty (Bárcena-Martín and Moro-Egido 2013), which could translate into a reduction in the gender gap in disability (Borrell et al. 2014; Palencia et al. 2014; Avendano et al. 2015; Van Oyen et al. 2010). However, our study further shows that such policies do not always limit the gender differences, as is the case in Norway or Sweden. This result is consistent with those of a study on gender differences in subjective health (Dahlin and Härkönen 2013). These countries, although generally egalitarian, still could face gender differences in resources, such as income or pension levels. Furthermore, while these countries yield relatively few situations of disability and hardship, these situations could concentrate on a selection of people, who tend to be socially vulnerable (isolated, in poor health, in the process of migration, etc.), falling through social safety nets (with unmet needs, for instance, in terms of health care) and being more exposed to disability. These situations could also correspond more frequently than in other countries to a “reverse causation” in which disability leads to adverse economic outcomes (Lynch et al. 1997). Unexpectedly, our study shows that in such contexts with low levels of hardship and disability, the gender gaps appear large and strongly associated when these situations become scarce and selected.

Due to the small sample size in some of the countries considered, and due to possible selection in the respondents, additional analyses should be conducted to confirm these results and provide further explanations. As previously mentioned there are large missing values on GALI in the Scandinavian countries and in the Netherlands (associated with the telephone collection mode). The large gender gap found in Scandinavian countries and its strong association with economic hardship might be exaggerated by the selected sample of respondents (for instance, if the missing values are preferably men with disability). We explored the association of the probability of missing values on GALI with age, sex, economic hardship, and working status (multivariate logistic models provided on request). The probability of missing GALI is negatively associated with economic hardship in all these countries and with age in Denmark and the Netherlands; the likelihood is greater for men in Denmark and Finland. We also find a positive association with being employed in all countries, except Norway (even though employment is usually negatively associated with disability and economic hardship risks). With this, we assume that the selected sample of respondents is not likely to have exaggerated the patterns we found. Furthermore, our results for the Scandinavian countries are congruent with other findings in Sweden.

At this stage, our study cannot disentangle the causation relationship between hardship and disability to interpret these results and the role of the social context. Pathways need to be further studied. Our study also has several limitations regarding the measures used. Hardship and disability are self-reported measures, and there are some variations in how men and women detect and report their health, disability, and economic problems. In this study, the measurements are not based on the exact health situation (diseases) and economic situation (income), which are sensitive to a certain knowledge which can differ across genders. The measurements are based on difficulties in performing normal activities due to health issues and difficulties in meeting both ends meet due to financial constraints. A number of studies have demonstrated that these measures are highly correlated with more objective indicators (see Sect. 2). Also, some studies tend to demonstrate that gender differences in poor health are generally due to poorer health profiles in women rather than differences in reporting propensity (Oksuzyan et al. 2009, 2010). So even though more precise measurements could usefully complete these analyses, our findings are still congruent with other studies. We believe they highlight patterns of gender inequality that deserve attention.

There are also variations across European countries, which might be due to the data. There are some differences in the survey design, and comparison of results across countries requires caution; yet our study more specifically concentrates on how men and women’s situations differed within each country, and we focused on these gender-specific patterns, although we compare the general patterns across countries.

Regarding interpretation, while studies usually report significant gender differences in disability, we found no gender differences in six countries (Crimmins et al. 2011). This could be due to the nature of the GALI indicator, which tends to increase men’s propensity to report disability, especially in working ages: it was found that people were more likely to report GALI when they were in difficulty at work, and situations of work disability were more frequent in men (Tubeuf et al. 2008).

The indicator of economic hardship also reflects different social situations in different countries. More research is needed to describe these situations across Europe. A step further could be to explore other situations of distress linked to isolation or exclusion. At this stage, our results do not claim to explain the underlying mechanisms that produce the gender gap in disability, especially those linked to morbidity differences; they nonetheless demonstrate the gender-sensitive intersection between economic hardship and disability in various European contexts.

Despite these limitations, we found that a part of the gender disability gap is associated with excess economic hardship in women in a number of countries with very different social contexts. Whatever the causal relationship, our results contribute to the existing literature, indicating that women might be exposed to a double disadvantage in a number of European countries. It may be that women with health problems are less likely than men to get the support that limits the risk of poverty: through a lower probability of keeping/getting a job or to take up protective social benefits. Or, it may be that women in adverse economic situations are less likely than men to access health protection programmes; for instance, access to job-related programmes when women are on average more distant from the labour market. While poverty has been presented as a challenge throughout Europe, our results tend to indicate that situations of economic hardship are often more frequent in women—which is the case even when overall levels are low—and are linked to their poorer functional health in 11 countries. This suggests that policies to combat poverty as well as health differentials need to concentrate on these associations of disability and economic hardship and to pay specific attention to gender differences.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (PDF 583 kb) (583.7KB, pdf)

Acknowledgements

This research was supported by Department of International Affairs of the French Institute for Demographic Studies (INED) and the Department of Political and Social Science at University Pompeu Fabra.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest. The data used in this research come from anonymized datasets of population surveys run by European Statistical Institutes and delivered to researchers upon request by EUROSTAT.

References

  1. Anxo D, Flood L, Mencarini L, Pailhé A, Solaz A, Tanturri M-L. Gender differences in time use over the life-cycle: A comparative gender analysis of Italy, France, Sweden and the United States. Feminist Economics. 2011;17(3):159–195. doi: 10.1080/13545701.2011.582822. [DOI] [Google Scholar]
  2. Arber S. Gender, marital status, and ageing: Linking material, health, and social resources. Journal of Aging Studies. 2004;18(1):91–108. doi: 10.1016/j.jaging.2003.09.007. [DOI] [Google Scholar]
  3. Avendano M, Berkman LF, Brugiavini A, Pasini G. The long-run effect of maternity leave benefits on mental health: Evidence from European countries. Social Science and Medicine. 2015;132:45–53. doi: 10.1016/j.socscimed.2015.02.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Avendano M, Jürges H, Mackenbach J. Educational level and changes in health across Europe: Longitudinal results from SHARE. Journal of European Social Policy. 2009;19(4):301–316. doi: 10.1177/1350506809341512. [DOI] [Google Scholar]
  5. Bárcena-Martín E, Moro-Egido AI. Gender and poverty risk in Europe. Feminist Economics. 2013;19(2):69–99. doi: 10.1080/13545701.2013.771815. [DOI] [Google Scholar]
  6. Benach J, Yasui Y, Borrell C, Saez M, Pasarin MI. Material deprivation and leading causes of death by gender: Evidence from a nationwide small area study. Journal of Epidemiology and Community Health. 2001;55(4):239–245. doi: 10.1136/jech.55.4.239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Berger N, Van Oyen H, Cambois E, Fouweather T, Jagger C, Nusselder W, et al. Assessing the validity of the Global Activity Limitation Indicator in fourteen European countries. BMC Medical Research Methodology. 2015;15:1. doi: 10.1186/1471-2288-15-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bergqvist K, Yngwe MA, Lundberg O. Understanding the role of welfare state characteristics for health and inequalities—An analytical review. BMC Public Health. 2013;13:1234. doi: 10.1186/1471-2458-13-1234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Berkman L, Zheng Y, Glymour MM, Avendano M, Borsch-Supan A, Sabbath EL. Mothering alone: Cross-national comparisons of later-life disability and health among women who were single mothers. Journal of Epidemiology and Community Health. 2015;69:865–872. doi: 10.1136/jech-2014-205149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Borrell C, Palencia L, Muntaner C, Urquia M, Malmusi D, O’Campo P. Influence of macrosocial policies on women’s health and gender inequalities in health. Epidemiologic Reviews. 2014;36:31–48. doi: 10.1093/epirev/mxt002. [DOI] [PubMed] [Google Scholar]
  11. Brennenstuhl S, Quesnel-Vallee A, McDonough P. Welfare regimes, population health and health inequalities: A research synthesis. Journal of Epidemiology and Community Health. 2012;66(5):397–409. doi: 10.1136/jech-2011-200277. [DOI] [PubMed] [Google Scholar]
  12. Cabrero-Garcia J, Julia-Sanchis R. The Global Activity Limitation Index mainly measured functional disability, whereas self-rated health measured physical morbidity. Journal of Clinical Epidemiology. 2014;67(4):468–476. doi: 10.1016/j.jclinepi.2013.10.005. [DOI] [PubMed] [Google Scholar]
  13. Cambois E, Garrouste C, Pailhé A. Gender career divide and women’s disadvantage in depressive symptoms and physical limitations in France. SSM: Population Health. 2017;3:81–88. doi: 10.1016/j.ssmph.2016.12.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cambois E, Jusot F. Contribution of lifelong adverse experiences to health inequalities in France: Evidence from a population survey. European Journal of Public Health. 2011;21(5):667–673. doi: 10.1093/eurpub/ckq119. [DOI] [PubMed] [Google Scholar]
  15. Cambois E, Solé-Auró A, Brønnum-Hansen H, Egidi V, Jagger C, Jeune B, et al. Educational differentials in disability vary across and within welfare regimes: A comparison of 26 European countries in 2009. Journal of Epidemiology and Community Health. 2016;70:331–338. doi: 10.1136/jech-2015-205978. [DOI] [PubMed] [Google Scholar]
  16. Cambois E, Solé-Auró A, Robine J-M. Economic hardship and educational differentials in disability in 26 European countries. Journal of Aging and Health. 2016;28(7):1214–1238. doi: 10.1177/0898264316656503. [DOI] [PubMed] [Google Scholar]
  17. Crimmins EM, Kim JK, Solé-Auró A. Gender differences in health: Results from SHARE, ELSA and HRS. European Journal of Public Health. 2011;21(1):81–91. doi: 10.1093/eurpub/ckq022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Dahlin J, Härkönen J. Cross-national differences in the gender gap in subjective health in Europe: Does country-level gender equality matter? Social Science and Medicine. 2013;98:24–28. doi: 10.1016/j.socscimed.2013.08.028. [DOI] [PubMed] [Google Scholar]
  19. Eikemo TA, Bambra C. The welfare state: A glossary for public health. Journal of Epidemiology and Community Health. 2008;62(1):3–6. doi: 10.1136/jech.2007.066787. [DOI] [PubMed] [Google Scholar]
  20. Eikemo TA, Huisman M, Bambra C, Kunst AE. Health inequalities according to educational level in different welfare regimes: A comparison of 23 European countries. Sociology of Health & Illness. 2008;30(4):565–582. doi: 10.1111/j.1467-9566.2007.01073.x. [DOI] [PubMed] [Google Scholar]
  21. Fjaer EL, Stornes P, Borisova LV, McNamara CL, Eikemo TA. Subjective perceptions of unmet need for health care in Europe among social groups: Findings from the European social survey (2014) special module on the social determinants of health. European Journal of Public Health. 2017;27:82–89. doi: 10.1093/eurpub/ckw219. [DOI] [PubMed] [Google Scholar]
  22. Hernanz, V., Malherbet, F., & Pellizzari, M. (2004). Take-up of welfare benefits in OECD countries: A review of the evidence. OECD social, employment and migration working papers (Vol. 17, pp. 1–47).
  23. Huijts T, Eikemo TA. Causality, social selectivity or artefacts? Why socioeconomic inequalities in health are not smallest in the Nordic countries. European Journal of Public Health. 2009;19(5):452–453. doi: 10.1093/eurpub/ckp103. [DOI] [PubMed] [Google Scholar]
  24. Jagger C, Gillies C, Cambois E, Van Oyen H, Nusselder W, Robine J-M. The Global Activity Limitation Indicator (GALI) measured function and disability similarly across European countries. Journal of Clinical Epidemiology. 2010;63(8):892–899. doi: 10.1016/j.jclinepi.2009.11.002. [DOI] [PubMed] [Google Scholar]
  25. Kangas O, Blomgren J. Socio-economic differences in health, income inequality, unequal access to care and spending on health: A country-level comparison of Finland and 16 other European countries. Research on Finnish Society. 2014;7:51–63. [Google Scholar]
  26. Karlson KB, Holm A, Breen R. Comparing regression coefficients between same-sample nested models using logit and probit: A new method. Sociological Methodology. 2012;42:286–313. doi: 10.1177/0081175012444861. [DOI] [Google Scholar]
  27. Luy M, Minagawa Y. Gender gaps—Life expectancy and proportion of life in poor health. Health Reports. 2014;25:12–19. [PubMed] [Google Scholar]
  28. Lynch JW, Kaplan GA, Shema SJ. Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. New England Journal of Medicine. 1997;337(26):1889–1895. doi: 10.1056/NEJM199712253372606. [DOI] [PubMed] [Google Scholar]
  29. Mackenbach JP, Karanikolos M, McKee M. The unequal health of Europeans: Successes and failures of policies. Lancet. 2013;381(9872):1125–1134. doi: 10.1016/S0140-6736(12)62082-0. [DOI] [PubMed] [Google Scholar]
  30. Mackenbach JP, Stirbu I, Roskam AJ, Schaap MM, Menvielle G, Leinsalu M, et al. Socioeconomic inequalities in health in 22 European countries. New England Journal of Medicine. 2008;358(23):2468–2481. doi: 10.1056/NEJMsa0707519. [DOI] [PubMed] [Google Scholar]
  31. Marmot M. The influence of income on health: Views of an epidemiologist. Health Affairs (Millwood) 2002;21(2):31–46. doi: 10.1377/hlthaff.21.2.31. [DOI] [PubMed] [Google Scholar]
  32. Marmot, M. (2010). Fair society, healthy lives (Strategic review of health inequalities in England post-2010): The Marmot Review (pp. 238) ISBN 978–0–9564870–0–1.
  33. McMunn A, Bartley M, Kuh D. Women’s health in mid-life: Life course social roles and agency as quality. Social Science and Medicine. 2006;63:1561–1572. doi: 10.1016/j.socscimed.2006.03.039. [DOI] [PubMed] [Google Scholar]
  34. Moen P, Chermack K. Gender disparities in health: Strategic selection, careers, and cycles of control. Journals of Gerontology. Series B, Psychological Sciences and Social Sciences. 2005;60(2):99–108. doi: 10.1093/geronb/60.Special_Issue_2.S99. [DOI] [PubMed] [Google Scholar]
  35. OECD . Closing the gender gap: Act now. Paris: OECD; 2012. [Google Scholar]
  36. Okechukwu CA, El Ayadi AM, Tamers SL, Sabbath EL, Berkman L. Household food insufficiency, financial strain, work-family spillover, and depressive symptoms in the working class: The work, family, and health network study. American Journal of Public Health. 2012;102(1):126–133. doi: 10.2105/AJPH.2011.300323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Oksuzyan A, Crimmins E, Saito Y, O’Rand A, Vaupel JW, Christensen K. Cross-national comparison of sex differences in health and mortality in Denmark, Japan and the US. European Journal of Epidemiology. 2010;25(7):471–480. doi: 10.1007/s10654-010-9460-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Oksuzyan A, Petersen I, Stovring H, Bingley P, Vaupel JW, Christensen K. The male-female health-survival paradox: A survey and register study of the impact of sex-specific selection and information bias. Annals of Epidemiology. 2009;19(7):504–511. doi: 10.1016/j.annepidem.2009.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Palencia L, Malmusi D, De Moortel D, Artazcoz L, Backhans M, Vanroelen C, et al. The influence of gender equality policies on gender inequalities in health in Europe. Social Science and Medicine. 2014;117:25–33. doi: 10.1016/j.socscimed.2014.07.018. [DOI] [PubMed] [Google Scholar]
  40. Rechel B, Grundy E, Robine JM, Cylus J, Mackenbach JP, Knai C, et al. Ageing in the European Union. Lancet. 2013;381(9874):1312–1322. doi: 10.1016/S0140-6736(12)62087-X. [DOI] [PubMed] [Google Scholar]
  41. Reine I, Palmer E, Sonnander K. Are there gender differences in wellbeing related to work status among persons with severe impairments? Scandinavian Journal of Public Health. 2016;44(44):772–783. doi: 10.1177/1403494816669638. [DOI] [PubMed] [Google Scholar]
  42. Solé-Auró A, Alcañiz M. Educational attainment, gender and health inequalities among older adults in Catalonia (Spain) International Journal for Equity in Health. 2016;15:126. doi: 10.1186/s12939-016-0414-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Stronks K, van de Mheen H, van den Bos J, Mackenbach JP. Smaller socioeconomic inequalities in health among women: The role of employment status. International Journal of Epidemiology. 1995;24(3):559–568. doi: 10.1093/ije/24.3.559. [DOI] [PubMed] [Google Scholar]
  44. Tubeuf S, Jusot F, Devaux M, Sermet C. Social heterogeneity in self-reported health status and measurement of inequalities in health. Paris: IRDES; 2008. p. 24. [Google Scholar]
  45. Van der Heyden J, Berger N, Van Oyen H. Comparison of self-rated health and activity limitation as predictors of short term mortality in the older population. Public Health. 2015;129:283–285. doi: 10.1016/j.puhe.2014.12.005. [DOI] [PubMed] [Google Scholar]
  46. Van Lancker W, Corluy V, Horemans J, Marchal S, Vinck J. Effects of poverty on the living and working conditions of women and their children. Brussels: European Union, Policy Department C: Citizens’ Rights and Constitutional Affairs: Gender Equality; 2015. p. 91. [Google Scholar]
  47. Van Oyen H, Cox B, Jagger C, Cambois E, Nusselder W, Gilles C, et al. Gender gaps in life expectancy and expected years with activity limitations at age 50 in the European Union: Associations with macro-level structural indicators. European Journal of Ageing. 2010;7(4):229–237. doi: 10.1007/s10433-010-0172-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Van Oyen H, Nusselder W, Jagger C, Kolip P, Cambois E, Robine J-M. Gender differences in healthy life years within the EU: An exploration of the “health-survival” paradox. International Journal of Public Health. 2013;58(1):143–155. doi: 10.1007/s00038-012-0361-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Verbrugge L. New thinking and science on disability in mid-and late life. European Journal of Public Health. 1995;5:20–28. doi: 10.1093/eurpub/5.1.20. [DOI] [Google Scholar]
  50. Whelan C, Maître B. Material deprivation, economic stress, and reference groups in Europe: An analysis of the EU-SILC 2009. European Sociological Review. 2013;29(6):1162–1174. doi: 10.1093/esr/jct006. [DOI] [Google Scholar]
  51. Wilkinson RG, Pickett KE. Income inequality and population health: A review and explanation of the evidence. Social Science and Medicine. 2006;62:1768–1784. doi: 10.1016/j.socscimed.2005.08.036. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material 1 (PDF 583 kb) (583.7KB, pdf)

Articles from European Journal of Population = Revue Européenne de Démographie are provided here courtesy of Springer

RESOURCES