Welfare state regimes and gender inequalities in the exposure to work ...

Welfare state regimes and gender inequalities in the exposure to work-related psychosocial hazards Javier Campos-Serna1, Elena Ronda-Pe´rez1,2,3, Bente E. Moen4, Lucia Artazcoz5, Fernando G. Benavides1,3 1

Center for Research in Occupational Health, Universitat Pompeu Fabra, Barcelona, Spain, 2Preventive Medicine and Public Health Area, University of Alicante, Spain, 3CIBER Epidemiologıá y Salud Pu´blica (CIBERESP), Spain, 4 Research Group for Occupational and Environmental Medicine, Department of Public Health and Primary Health Care, University of Bergen, Norway, 5Age`ncia de Salut Pu´blica de Barcelona, Spain Background: Gender inequalities in the exposure to work-related psychosocial hazards are well established. However, little is known about how welfare state regimes influence these inequalities. Objectives: To examine the relationship between welfare state regimes and gender inequalities in the exposure to work-related psychosocial hazards in Europe, considering occupational social class. Methods: We used a sample of 27, 465 workers from 28 European countries. Dependent variables were high strain, iso-strain, and effort-reward imbalance, and the independent was gender. We calculated the prevalence and prevalence ratio separately for each welfare state regime and occupational social class, using multivariate logistic regression models. Results: More female than male managers/professionals were exposed to: high strain, iso-strain, and effort–reward imbalance in Scandinavian [adjusted prevalence ratio (aPR)52.26; 95% confidence interval (95% CI): 1.87–2.75; 2.12: 1.72–2.61; 1.41: 1.15–1.74; respectively] and Continental regimes (1.43: 1.23– 1.54; 1.51: 1.23–1.84; 1.40: 1.17–1.67); and to high strain and iso-strain in Anglo-Saxon (1.92: 1.40–2.63; 1.85: 1.30–2.64; respectively), Southern (1.43: 1.14–1.79; 1.60: 1.18–2.18), and Eastern regimes (1.56: 1.35–1.81; 1.53: 1.28–1.83). Conclusion: Gender inequalities in the exposure to work-related psychosocial hazards were not lower in those welfare state regimes with higher levels of universal social protection policies. Keywords: Gender, Psychosocial work factors, Social welfare, Socioeconomic factors

Introduction Two main theoretical approaches have been used to measure the exposure to work-related psychosocial hazards in occupational health: the demand–control– support model,1,2 and the effort–reward imbalance model.3,4 The demand–control–support model postulates that job strain results from the joint effects of high job demands and low job control, while social support at work plays a buffering role in the interaction between job demands and control. The worst hypothesized scenario for exposure to work-related psychosocial hazards is called iso-strain, which is the combination of job strain and lack of social support. Alternatively, the effort–reward imbalance model puts explicit emphasis on the non-reciprocity of the

Correspondence to: J Campos-Serna, Center for Research in Occupational Health, Universitat Pompeu Fabra, Avda Dr. Aiguader, 88, 1a planta Despatx 171.01.03, Barcelona 08003, Spain. Email: javier.campos@upf. edu

ß W. S. Maney & Son Ltd 2013 DOI 10.1179/2049396713Y.0000000030

social contract (high effort combined with low reward) and defines a state of emotional distress. The prevalence of exposure to work-related psychosocial hazards seems to differ significantly by gender. Several studies indicate that employed women experience worse psychosocial working conditions than employed men, and that a higher health burden might result from these exposures.5–8 Previous research has found that men experience higher job demands, effort, and overcommitment, and lower social support at work; whereas women exhibit lower job control and lower reward.4,9,10 On the other hand, some studies have found that women experience higher emotional job demands11 and higher job reward.12 In addition, in the European Union, women’s jobs are characterized by a greater level of monotony, with lower participation in planning, higher demands, more psychological and sexual harassment, higher exposure to the public, lower salaries, less prospects for promotion, and more precariousness than those of men.13

International Journal of Occupational and Environmental Health

2013

VOL .

19

NO .

3

179

Campos-Serna et al.

Welfare state regimes and gender inequalities

The unequal gender distribution of work-related psychosocial hazards is mainly related to the horizontal segregation of the labor market, which concentrates women in occupations and economic activities (e.g. services) with high exposure to workrelated psychosocial hazards (e.g. those related to the organization of work: high strain, iso-strain, and those related to the non-reciprocity of the social contract: effort–reward imbalance).14,15 In addition, the unequal distribution of working tasks by gender within the same job title16–18 may expose women to even higher levels of work-related psychosocial hazards.19 Furthermore, vertical segregation, which places women in the lowest positions of the decisionmaking scale, reinforces this effect.14 It has been suggested that these inequalities put women at a higher risk of physical20 and mental disorders,21 sickness absence,22 disability,23 and mortality24 from work-related psychosocial hazards. In the last century, women have increasingly taken jobs outside the home, while men have not taken on proportionate responsibility for childcare and domestic duties historically assigned to women. As a result, nowadays many women now have two jobs: one in the workplace (paid) and another at home (unpaid). Thus, most employed women are doubly exposed to work-related psychosocial hazards. This phenomenon is known as the double burden.14 Welfare state modelling has long been an important strand within comparative social policy, serving as a means of reducing the complexity of crossnational welfare state comparison. In 1990, EspingAndersen25 proposed three theoretical welfare state regimes (Liberal, Conservative, and Social Democratic) based upon three principles of the labor market: decommodification (the extent to which an individual’s welfare is reliant upon the market, mainly by unemployment or sickness benefits and pensions), social stratification (the role of welfare states in maintaining or breaking down social stratification), and the private–public mix (the relative roles of the state, the family, the voluntary sector, and the market in welfare provision).25 However, Esping-Andersen’s proposed classification system was widely revised.26 The principal criticisms have focused on the range of countries and regimes used to construct his typology, primarily the misclassification of the Southern European welfare states as immature Conservative ones.27 As a result of this criticism and subsequent empirical testing,28 a range of modified and alternative typologies were proposed.27–29 A standard classification was recently proposed by Eikemo30 and Bambra31 which, among other aspects, extends the classification to a wider range of countries and includes considerations of gender and the role of

180


public services.30,31 Eikemo30 and Bambra31 introduced the ‘‘Eastern’’ welfare state typology, which has been widely used in a number of studies and recommended for public health research.32–34 A welfare state regime could act as a buffer to protect against the unequal gender distribution of work-related psychosocial hazards mentioned above.35 The particular buffer effect of each particular welfare state regime likely relates to the feminist and trade union tradition, levels of social organization, and the number of government labor market interventions in the form of regulations and social protection policies (focusing, for example in subsides for childcare and plans of employment for mothers with young children).36,37 For example, some countries such as Germany and The Netherlands have coordinated labor markets, with a high level of collective bargaining. This arguably leads to better control of work-related psychosocial hazards. These countries are also relatively redistributive with regard to social policies reaching a relatively high level of labor market decommodification. At the same time, they are characterized by rather conservative family policies, which have an impact on the segregation of men and women in the labor market (e.g. part-time employment or retrenchment from the labor market of women with young children). On the other hand, the Swedish daycare system is organized to accommodate the needs of the working women with young children. To obtain a place in the childcare center, both parents (or the single parent) must be working or studying at least 20 hours per week. This is an important fact to consider in determining how childcare subsidies increase the labor supply.38 It has been shown that welfare state regimes with more comprehensive social protection policies (e.g. that enhance child care and parental leave rights) also have labor protection policies that reduce women’s exposure to work-related psychosocial hazards.39,40 It has also been reported that welfare state regimes with the most comprehensive systems of social benefits redistribution not only improve the psychosocial work environment, but also mitigate the impact of work-related psychosocial hazards on health and health inequalities41,42 (e.g. by providing more resources to cope with stressful working events such as job insecurity and job loss).43 These arguments suggest a correlation between better social protection policies and benefits redistribution in welfare state regimes, such as occurs in the Scandinavian welfare state regime, which supposedly reduces women’s exposure to work-related psychosocial hazards. In addition, the combination of all these characteristics may lead to complex and welfare state regime-specific results in buffering the phenomenon of double exposure to work-related psychosocial hazards.38 In other words, welfare state regimes may mitigate the difference between women’s and men’s

2013

VOL .

19

NO .

3

Campos-Serna et al.

exposure to work-related psychosocial hazards by social and labor market protection policies and social benefits redistribution.4,44 For example, countries with more protective social policies, more redistributive social benefits and more regulated labor markets (e.g. the Scandinavian regime) will have a lower gap between women and men in the exposure to workrelated psychosocial hazards than those countries with more conservative and liberal welfare state regimes (e.g. the Southern regime) where the state has less tradition of implementing social protection policies and policies to redistribute social benefits, as well as less influence in regulating labor markets.41,45 Previous studies have also shown that occupational social class influences exposure to work-related psychosocial hazards. Thus, workers in lower occupational strata are more often exposed to high strain and iso-strain than workers in higher occupational strata.46,47 In addition, the distribution of low control and low support has been found to follow the social gradient (with higher exposure in workers in the lower occupational strata), although the same distribution was not shown for high job demands.46 Moreover, workers in the lower occupational strata have reported being exposed to higher job strain, greater job insecurity and lower social support over time than those in the higher occupational strata.48 For Siegrist and colleagues, the asymmetry between effort and reward may have adverse health effects that tend to disproportionately affect persons in the lowest occupational strata, who lack flexibility due to their low skill level and lack of mobility.4 Studies have also shown that patterns of gender inequality in the exposure to work-related psychosocial hazards can differ by occupational social class.49 A study carried out in Spain showed that female employees in the highest occupational strata, but not in the lowest, were more exposed than their male counterparts to job strain, iso-strain, and effort– reward imbalance.50 In addition, a study carried out in a British cohort,51 found that employed women showed a consistent trend for better social support than employed men in both the highest and the lowest occupational strata. Furthermore, a study using data from the Whitehall II Study,52 a longitudinal study of British civil servants, showed that employed women in the lowest or middle occupational strata who reported low control were at most risk for depression and anxiety. Welfare state regimes influence psychosocial work environments, mainly through shaping the labor market and social protection system.53,54 For example, Scandinavian countries have shown the lowest household income inequalities mainly because they have: (1) the lowest percentage of national income derived from capital and the highest derived from


labor; (2) the lowest wage disparities within the labor force; and (3) the highest redistributive effect of state policies.55 In addition, Scandinavian countries, such as Sweden,56 have developed policies that aim to achieve a good psychosocial working environment by trying to reduce workers’ exposure to workplace stressors. However, little is known about whether welfare state regimes influence the unequal distribution of workrelated psychosocial hazards between women and men.41,45 This is a relevant issue for policy makers57 in their efforts to reduce health inequalities, as well as for researchers and occupational health practitioners49,58 who must work to reduce the impact of these inequalities on workers’ health. The main objectives of this study were to examine whether gender inequalities in the exposure to workrelated psychosocial hazards differ by welfare state regimes, and to test whether gender patterns differ by occupational social class across five different welfare state regimes. Our hypothesis is that the more social protection or social benefits redistributive policies a welfare state regime has, the lower the likelihood of exposure to work-related psychosocial hazards and gender inequalities in exposure to work-related psychosocial hazards, independently of occupational social class.

Methods Participants and study sample Data were obtained from the Fourth European Working Conditions Survey (EWCS).59 Briefly, the sample design was a multi-stage random sample including 31 European countries. An employed person was defined as one who was aged 15 years or older and had any paid job during the week in which the interview was held or who had a job but was temporarily absent.60 A total of 29, 680 interviews (16, 558 in men and 13, 122 in women) were conducted by trained interviewers at workers’ homes between September and November 2005. The same questionnaire, translated into 27 different languages and 15 language variants, was used in all countries covered. The final sample included 15, 063 men and 12, 402 women. The average response rate was 48% for all eligible participants, ranging from 28% in The Netherlands to 69% in the Czech Republic. The countries included in the survey were grouped into five welfare state regimes, following Eikemo30 and Bambra’s31 standard classification. We used the following typologies of welfare state regimes: Scandinavian (Finland, Norway, Sweden, and Denmark), AngloSaxon (Ireland and UK), Continental (The Netherlands, Germany, Switzerland, France, Belgium, Austria, and Luxembourg), Southern (Spain, Portugal, Italy, and Greece), and Eastern (Latvia, Lithuania, Estonia, Bulgaria, Poland, Slovenia, Croatia, Hungary, Slovakia, Romania, and the Czech Republic). Malta, Cyprus,


2013

VOL .

19

NO .

3

181

Campos-Serna et al.


and Turkey, although they were included in the EWCS, were not considered in our analysis, because they did not match any of the welfare state regime characteristics.

Table 1 summarizes the characteristics of each welfare state regime, and our hypothesis about how the social protection policies and socio-cultural traditions of each welfare state regime could be affecting

Table 1 Researchers’ hypothesis of influence on gender inequalities in the exposure to work-related psychosocial hazards based on the common features of social/work protection policies and socio-cultural context of the welfare state regime Social/work protection policies29–31 Welfare state regime Scandinavian Finland Norway Sweden Denmark

Continental Netherlands Germany Switzerland France Belgium Austria Luxembourg Anglo-Saxon Ireland UK

Southern Spain Portugal Italy Greece

Eastern Latvia Lithuania Estonia Bulgaria Poland Slovenia Croatia Hungary Slovakia Romania Czech Republic

Socio-cultural context25,39,77,90,91

H

H

The state promotes social equality of the highest standards through a redistributive social security system, providing highly decommodifying programmes, universalism and generous social transfers, a commitment to full employment and an important social protection system, and a strongly interventionist state. The well-being of their citizens is largely independent of prevailing market conditions

zz

Social-democratic. The well-being of their citizens is largely independent in social provisions from family roles. Child care and housework are well balanced between women and men. The male bread-winner model is not relevant

zz

Benefits are often earnings related, administered through the employer; and geared towards maintaining existing social patterns. The redistributive impact is minimal. However, the role of the market is minimal. Social expenditures are high and social benefits are good. Social security is tied to labor market position

z

Conservative-corporatist. The role of the family is emphasized. Child care and housework is acceptably balanced between women and men. The male bread-winner model is not relevant

z

The state provision of welfare is minimal, social protection levels are modest and often involve entitlement criteria, and recipients are usually means-tested and stigmatized. The state minimizes the decommodification effects of the welfare state regime and a rigid division exists between those, largely the poor, who rely on state aid and those who are able to provide for themselves

z

Conservative-Liberal. The role of the family is not overly emphasized. Child care and housework is acceptably balanced. The male bread-winner model is relevant

z

Characterized by a fragmented system of welfare provision which consists of diverse income maintenance schemes that range from the meagre to the generous and welfare services, particularly, the health care system, that provide only limited and partial coverage

–

Conservative. The role of the Catholic church and the family is crucial and reliance on the voluntary sector is also emphasized in social provision. Childcare and housework is quite unbalanced, with women assuming most of these responsibilities. The male breadwinner model is much more relevant

–

The formerly Communist countries of the East Europe have experienced the demise of the universalism of the Communist welfare state and a shift towards polices of marketization and decentralization. They also have limited welfare services. These countries have experienced extensive economic upheaval and have undertaken comprehensive social reforms. They have emphasized the Liberal regime approaches of marketization, decentralization, and the reform of health insurance schemes. In comparison with the other member states of the European Union, they have limited health service provision, and overall population health is relatively poor. However, these countries clearly comprise the most underdefined and understudied regions

––

Post-state-socialist countries. Difficult to categorize as conservative-corporatist, liberal, or social democratic. The role of the family is crucial. Traditional gender-roles and patriarchy within the family remains, so that childcare and housework is quite unbalanced, with women assuming most of these responsibilities. The male breadwinner model is relevant

––

Note: H: researchers’ hypothesis of influence on gender inequalities in the exposure to work-related psychosocial hazards: strongly decreasing zz, moderately decreasing z, strongly increasing – – and moderately increasing –.

182


2013

VOL .

19

NO .

3

Campos-Serna et al.

gender inequalities in the exposure to work-related psychosocial hazards.

Work-related psychosocial hazards The demand/control,1 demand/control/social support,2 and effort–reward imbalance models,3 which have been widely used in occupational health research to characterize psychosocial work environments, were followed as a guide for the measurement of the exposure to work-related psychosocial hazards. Job demands were measured through two items with seven response categories and three items with five response categories59 (Cronbach’s alpha50.61); job control was assessed with five items with five response categories59 (Cronbach’s alpha50.71) and social support with three items with five response categories59 (Cronbach’s alpha51.00). Psychosocial job effort was measured through two items with seven response categories59 and three items with five response categories59 (Cronbach’s alpha50.61); and reward was measured by seven items with five response categories59 (Cronbach’s alpha50.80) (Appendix 1). Response categories with five options were labelled from 15almost always to 55almost never, and those with seven options were labeled from 15all of the time to 75never. We calculated the total sum of the scores given for all items used to measure each work-related psychosocial factor. Work-related Psychosocial factors were dichotomized on the median. All values equal to or under to the median for each work-related psychosocial factor were classified in the lower exposure category (low-control, low-support, and low-reward). Conversely, all values over the median for each work-related psychosocial factor were classified in the higher exposure category (high-demand and high-effort). The median was used as a reference point, because there was no other objective standardized reference scale. Moreover, several previous studies that have based their analyses of work-related psychosocial hazards on working conditions surveys have also used the median in this way.22 Work-related psychosocial factors were combined to create three work-related psychosocial hazards: (1) high strain, which represents workers with a score above the median for job demands and equal to or below the median for control; (2) iso-strain, which represents workers with a score above the median for job demands and equal to or below the median for control and social support; and (3) effort–reward imbalance, which represents workers with a score above the median for effort and equal to or below the median for reward.

Covariates The adjustment variable was the age of the worker (15–24, 25–34, 35–44, 45–54, and 55 or over). Other variables used to explain gender inequalities observed were: (1) marital status; (2) family burden measured


as time invested in caring for relatives (children and elderly/disabled); (3) sector of economic activity of the employer; and (4) part-time/full-time status.59 Marital status was derived from one survey question: ‘‘Are you married or living with your partner?’’ (yes/ no). The variable family burden was assessed with the question ‘‘How many hours per day are you involved in caring for and educating your children or caring for elderly/disabled relatives?’’. The number of hours per week was calculated by multiplying each reply by 7. We assumed that the worker reported the mean number of hours worked per day considering the whole week. We generated four categories: from 1 to 14 hours a week; from 15 to 21, from 22 to 35 and 36 or more hours a week. Economic activity was determined according to the statistical classification of economic activities of the European Community (NACE11)59 and was grouped into four sectors: (1) agriculture (agriculture and fishing); (2) industry (manufacture and mining, electricity, gas and water supply, wholesale and retail trade); (3) construction (construction); and (4) services (hotels and restaurants, transport and communication, financial intermediation, real estate, public administration and defence, education and health). Finally, part-time/full-time status was derived from one question which asked ‘‘Do you work part-time or full-time?’’ with two possible answers: part-time or full-time. The analysis was stratified by occupational social class as a proxy for social class.61 Occupation was grouped into three categories: (1) managers/professional (legislators and senior officials and managers, professionals, technicians, and associate professionals); (2) clerks/service/shop workers (clerks, service workers, and shop and market sales workers); and (3) manual workers (unskilled agricultural and fishery workers, craft and related trades workers, plant and machine operators and assemblers, elementary occupations). This operationalization of the occupation into three categories of occupational social class was based on the five categories of the Registrar-General’s Social Classes classification (RGSCs): I: professionals; II: managers, III non-manual: skilled non-manual; III: manual: skilled manual; IV: semi-skilled manual; V: unskilled manual. The RGSC classification is based on occupational skill and has been widely used in previous reports and studies focusing on social class.62

Statistical analysis We calculated the proportion of men and women in each category of occupational social class overall and within each welfare state regime, and differences were analyzed using the Chi-square test with its P values. We then calculated the prevalence of high strain, isostrain, and effort–reward imbalance in men and


2013

VOL .

19

NO .

3

183

184


2013

VOL .

19

NO .

3

56.2 66.8 32.2 1424 43.8 1828 665 33.2 1338 3426 67.8 1625 44.4 54.4 30.7 704 55.6 563 464 45.6 554 1174 69.3 519 46.7 70.3 16.3 384 429 101 53.3 29.7 83.7 438 181 517 44.9 64.3 29.8 1583 55.1 1289 615 35.7 1110 1467 70.2 622 52.5 65.5 24.2 Note: *All differences observed were statistically significant (P,0.001).

946 715 289 47.5 34.5 75.8 855 377 907 50.0 64.3 29.6 Managers/professionals 5004 50.0 5010 Clerks/service/shop workers 2303 35.7 4146 Manual workers 7490 70.4 3156

(%) n (%) (%) (%) (%) (%) (%) n (%) (%) (%) Occupational social class*

n

(%)

n

(%)

n

Men

n

Women

n

Men

Women

n

Men

n

Women

n

Men

n

Women

n

Men

Eastern Southern Anglo-Saxon Continental Scandinavian

Women

The proportion of women was higher than men among managers/professionals only in the Scandinavian and Eastern welfare state regimes, with an absolute difference of 5.0% and 12.4%, respectively. The proportion of women was higher than men among clerks/service/ shop workers in all welfare states regimes, with the highest absolute difference in the Anglo-Saxon regime (40.6%). Conversely, the proportion of men was higher than women among manual workers in all welfare state regimes, with the highest absolute difference also found in Anglo-Saxon regime (67.4%) (Table 2). Considering Europe as a whole, the highest prevalence of work-related psychosocial hazards was found in female managers/professionals exposed to high strain (33.1%). Conversely, the lowest prevalence was found in male managers/professionals exposed to iso-strain (13.9%) (Table 3).

Men

Results

Europe

women in the whole sample and separately in each of the five welfare state regimes and applied the Chisquare test with its P values to analyze gender differences. In addition, multivariate logistic regression models were used to estimate crude and adjusted prevalence ratios (PRs) and 95% confidence intervals (95% CIs) for all 28 countries together and separately for each of the five welfare state regimes, considering men as the reference group. Age was the common adjustment variable. After age adjustment, the following four independent adjustment models were applied to the PR to try to explain the gender inequalities observed: Model 1 (M1), PR adjusted for age and marital status; Model 2 (M2), PR adjusted for age and family burden; Model 3 (M3), PR adjusted for age and sector of economic activity; and Model 4 (M4), PR adjusted for age and pat-time/full-time status. All analyses were done separately for managers/professionals, clerks/service/shop workers, and manual workers. Respondents who did not answer the questions59 needed to characterize work-related psychosocial hazards were excluded from our analysis (a maximum of 89 men and 75 women). Analyses were performed using SPSS v15 and Stata v9. To enhance the representativity of our results, we applied two types of weighting to the data:59 selection probability weighting and non-response weighting. Both weightings were applied to the data before starting the analysis and the process of constructing the dependent variables. Selection probability weighting was applied to avoid the consequences of giving more probability of selection to respondents living in smaller households. Non-response weighting was applied to avoid a bias in the estimations caused by the different response rates. The missing values were left out of the analysis after applying both weightings.

Women


Table 2 Occupational social class structure in Europe and within each of the welfare state regimes by gender in a sample of workers (n527, 465) from the Fourth European Working Conditions Survey (2005)

Campos-Serna et al.

378 (32.2) 161 (31.0) 947 (27.6) 545 (33.5){ 181 (15.4) 56 (10.8)* 587 (17.1) 278 (17.1) 424 (36.1) 171 (32.9) 1014 (29.6) 584 (35.9){ (21.8) (12.9) (19.8) 97 (18.8) 66 (12.8) 87 (16.8) Note: *P,0.05; {P,0.01; {P,0.001.

2190 (29.2) 981 (31.1) 214 (23.6) 1309 (17.5) 481 (15.2)* 167 (18.4) 2162 (28.9) 1010 (32.0){ 195 (21.5)

85 58 70

(29.4)* 555 (37.8) 169 (20.1) 308 (21.0) 75 (24.2) 442 (30.1) 166

(27.2){ (12.1){ (26.7)

22 13 20

165 (24.8) 414 (30.9) 116 (17.4) 268 (20.0) 168 (25.3) 385 (28.8) 139 (29.9) 174 (31.4) 78 (16.8) 71 (12.8) 131 (28.2) 130 (23.5) (18.1) (11.7){ (13.3) 78 50 57 45 (24.9) 32 (17.7) 30 (16.6) 181 (29.4) 316 126 (20.5) 201 140 (22.7) 274 603 (26.2) 1204 (29.0)* 407 (17.7) 767 (18.5) 548 (23.8) 1008 (24.3)

73 (19.3) 222 (31.1){ 55 (14.6) 177 (24.8){ 79 (21.0) 163 (22.8)

(28.5) (18.1) (24.7)

293 (20.6) 585 (32.0){ 220 (15.0) 435 (23.8){ 327 (23.0) 424 (23.2) 59 (13.5) 102 (26.5){ 154 (21.9) 179 (31.8){ 49 (11.2) 80 (20.8){ 82 (11.7) 109 (19.3){ 74 (16.9) 52 (13.5) 152 (21.6) 129 (22.9) (31.5){ (20.6){ (24.5){ 340 (21.5) 406 207 (13.1) 265 276 (17.4) 315

(%) (%) (%) (%) (%) (%) (%) (%) (%) (%) n

999 (20.0) 1659 (33.1){ 153 (17.9) 388 (41.0){ 696 (13.9) 1217 (24.3){ 138 (16.2) 329 (34.7){ 987 (19.7) 1171 (23.4){ 158 (18.5) 251 (26.5){

Managers/professionals High strain Iso-strain Effort–reward imbalance Clerks/service/shop workers High strain Iso-strain Effort–reward imbalance Manual workers High strain Iso-strain Effort–reward imbalance

Men

(%) n Work-related psychosocial hazards

Women

n

Men

n

Women

n

Men

n

Women

n

Men

n

Women

n

Men

n

Women

n

Men

n

(%)

Women Eastern Southern Anglo-Saxon Continental Scandinavian Europe

Table 3 Prevalence of exposure to work-related psychosocial hazards in Europe and welfare state regimes by gender and occupational social class in a sample of workers (n527, 465) from the Fourth European Working Conditions Survey (2005)

Campos-Serna et al.


In the comparison by welfare state regime, Scandinavian women who were managers/professionals showed the highest prevalence for all types of workrelated psychosocial hazards: high strain (41.0%), isostrain (34.7%), and effort–reward imbalance (26.5%). In contrast, Anglo-Saxon men who were also managers/professionals showed the lowest prevalence of exposures to high strain (13.5%), iso-strain (11.2%), and effort–reward imbalance (16.9%) (Table 3). Among clerk/services/shop workers, women from countries in the Southern regime had the highest prevalence of exposure to high strain (31.4%), Scandinavian women had the highest prevalence of exposure to iso-strain (24.8%) and Eastern women had the highest prevalence of exposure to effort–reward imbalance (28.8%). On the other hand, Anglo-Saxon women had the lowest prevalence of exposure to any type of work-related psychosocial hazard: high-strain (18.1%), iso-strain (11.7%), and effort–reward imbalance (13.3%) (Table 3). Among manual workers, men showed the highest prevalence of exposure to work-related psychosocial hazards: high strain (37.8%) and iso-strain (21.0%) in Continental welfare states; and effort–reward imbalance (36.1%) in Southern welfare states. In contrast, among manual workers, women showed the lowest prevalence of exposure to work-related psychosocial hazards: high strain (18.8%) and effort–reward imbalance (16.8%) in Anglo-Saxon welfare states; and iso-strain (10.8%) in Southern welfare states (Table 3). Table 4 shows gender inequalities in the exposure to work-related psychosocial hazards in Europe and in each welfare state regime. Scandinavian welfare states have the highest gender inequalities in the exposure to all types of work-related psychosocial hazards, with more women than men exposed to: high strain [adjusted prevalence ratio (aPR)51.72; 95% CI: 1.52–1.95], iso-strain (1.71: 1.48–1.97), and effort–reward imbalance (1.22: 1.06–1.39). Table 5 shows gender inequalities in the exposure to work-related psychosocial hazards by occupational social class in Europe and in each welfare state regime, using different adjustment models. Using the model adjusted only for age, across Europe as a whole the prevalence of exposure to high strain (aPR51.64; 95% CI: 1.51–1.78), iso-strain (1.71: 1.55–1.90), and effort– reward imbalance (1.17: 1.07–1.29) was significantly higher in women compared to men among managers/ professionals. The prevalence of exposure to effort– reward imbalance was also higher in women than men among manual workers (1.11: 1.03–1.19). In Scandinavian and Continental welfare states (Table 5), among managers/professionals, more women than men were exposed to high strain (aPR52.26; 95% CI: 1.87–2.75; 1.43: 1.23–1.54; respectively); iso-strain (2.12: 1.72–2.61; 1.51: 1.23–1.84), and effort–reward


2013

VOL .

19

NO .

3

185

186

1.26 (1.18–1.36){ 1.22 (1.11–1.34){ 1.06 (0.99–1.14) Note: *Prevalence ratio adjusted (aPR) for age and 95% confidence intervals (95% CI). Men is the reference group in all models. { Statistically significant aPR.

1.09 (0.97–1.22) 0.98 (0.81–1.17) 0.87 (0.77–0.98){ 1.00 (0.92–1.10) 1.02 (0.90–1.15) 1.07 (0.97–1.18) 1.72 (1.52–1.95){ 1.71 (1.48–1.97){ 1.22 (1.06–1.39){ 1.21 (1.16–1.27){ 1.22 (1.15–1.30){ 1.03 (0.99–1.09) High strain Iso-strain Effort–reward imbalance

1.21 (0.99–1.47) 1.16 (0.91–1.48) 0.83 (0.66–1.04)

aPR (95% CI)* aPR (95% CI)* aPR (95% CI)* aPR (95% CI)* aPR (95% CI)* aPR (95% CI)* Work-related psychosocial hazards

Continental Europe

Scandinavian

Anglo-Saxon

Southern

Eastern


Table 4 Prevalence ratio (PR) of exposure to work-related psychosocial hazards in women compared to men in Europe and welfare state regimes in a sample of workers (n527, 465) from the Fourth European Working Conditions Survey (2005)

Campos-Serna et al.


imbalance (1.41: 1.15–1.74; 1.40: 1.17–1.67). The Eastern welfare state was the only one where significantly higher proportion of women than men reported were exposed to high strain across all occupational social classes: managers/professionals (1.56: 1.35–1.81), clerks/services/shop workers (1.23: 1.03–1.47), and manual workers (1.22: 1.10–1.36). Conversely, the Continental welfare state was the only one where more men than women were exposed to high strain (0.73: 0.62–0.86) and iso-strain (0.58: 0.44–0.77) among manual workers. Moreover, the higher proportion of women exposed to high strain and iso-strain among managers/professionals varied only slightly across all welfare state regimes. The Scandinavian and Eastern welfare states were the only two that showed an occupational social class gradient for gender inequalities in the exposure to high strain. Levels of gender inequalities in the exposure to high strain increased in the Scandinavian and Eastern welfare states from manual workers to managers/ professionals: manual workers (1.25: 0.96–1.63 and 1.22: 1.10–1.36, respectively); clerks/services/shop workers (1.61: 1.21–2.12 and 1.23: 1.03–1.47), and managers/ professional (2.26: 1.87–2.75 and 1.56: 1.35–1.81) (Table 5). Although no significant change was observed after adjusting the PR for marital status (M1) and parttime/full-time status (M4), a significant decrease was found after adjusting for family burden (M2) and sector of economic activity (M3). In both cases, the decrease in PR was observed in managers/professionals, and was most notable in the Scandinavian welfare states (Table 5).

Discussion Our results confirm that women are more frequently exposed to work-related psychosocial hazards than men in Europe as a whole. As found in previous studies,11,63,64 women show higher levels of high strain, iso-strain, and effort–reward imbalance than men. However, in contrast to our hypothesis, gender inequalities show only slight variation across all welfare state regimes, and are distinctly higher in managers/professionals than in clerk/service/shop and manual workers. The Scandinavian welfare state regime had the highest gender inequalities in the exposure to work-related psychosocial hazards among managers/professionals and clerk/service/shop workers. On the other hand, the Eastern welfare state was the only regime in which gender inequalities in the exposure to work-related psychosocial hazards were detected across all occupational social classes. In relation to the highest gender inequalities in the exposure to work-related psychosocial hazards (high strain, iso-strain, and effort–reward imbalance) in the Scandinavian welfare state, our results contrast with

2013

VOL .

19

NO .

3


2013

Effort–reward imbalance

Iso-strain

Manual workers High strain


Iso-strain

Clerks/service/shop workers High strain


Iso-strain

Managers/professionals High strain

aPR (95% CI)* M1 M2 M3 M4 aPR (95% CI)* M1 M2 M3 M4 aPR (95% CI)*

aPR (95% CI)* M1 M2 M3 M4 aPR (95% CI)* M1 M2 M3 M4 aPR (95% CI)* M1 M2 M3 M4

aPR (95% CI)* M1 M2 M3 M4 aPR (95% CI)* M1 M2 M3 M4 aPR (95% CI)* M1 M2 M3 M4

Work-related psychosocial hazards/welfare states

1.07 1.08 1.19 1.17 1.16 0.88 0.89 1.02 0.99 0.98 1.11

1.11 1.12 1.05 1.15 1.13 1.04 1.05 0.95 1.06 1.06 1.02 1.03 0.98 1.06 1.05

1.64 1.61 1.66 1.48 1.64 1.71 1.68 1.65 1.55 1.72 1.17 1.17 1.13 1.16 1.19

(1.00–1.16) (1.00–1.16) (1.04–1.36){ (1.08–1.26){ (1.08–1.25){ (0.78–0.99) (0.80–1.00) (0.84–1.25) (0.88–1.11) (0.87–1.09) (1.03–1.19){

(1.00–1.22) (1.01–1.24){ (0.86–1.29) (1.03–1.28){ (1.02–1.25){ (0.92–1.19) (0.93–1.20) (0.73–1.23) (0.93–1.22) (0.93–1.21) (0.92–1.14) (0.92–1.15) (0.80–1.21) (0.96–1.20) (0.94–1.18)

(1.51–1.78){ (1.49–1.75){ (1.42–1.94){ (1.37–1.62){ (1.51–1.79){ (1.55–1.90){ (1.52–1.86){ (1.36–1.99){ (1.40–1.72){ (1.55–1.90){ (1.07–1.29){ (1.07–1.28){ (0.95–1.34) (1.05–1.27){ (1.08–1.30){

Europe

1.25 1.26 1.25 1.22 1.29 1.09 1.11 1.01 1.03 1.17 1.18

1.61 1.59 1.86 … 1.63 1.68 1.66 1.51 … 1.69 1.09 1.07 1.10 1.18 1.16

2.26 2.24 1.94 1.89 2.23 2.12 2.09 1.82 1.74 2.12 1.41 1.40 1.37 1.38 1.35

(0.96–1.63) (0.97–1.65) (0.81–1.91) (0.91–1.62) (0.99–1.70) (0.78–1.52) (0.79–1.56) (0.56–1.83) (0.72–1.48) (0.82–1.64) (0.90–1.55)

(1.22–2.36){ (0.82–1.45) (0.80–1.43) (0.54–2.22) (0.84–1.67) (0.86–1.55)

(1.23–2.17){ (1.21–2.34){ (1.19–2.32){ (0.69–3.30)

(1.21–2.12){ (1.20–2.10){ (0.85–4.09)

(1.87–2.75){ (1.84–2.71){ (1.40–2.70){ (1.55–2.30){ (1.83–2.72){ (1.72–2.61){ (1.70–2.57){ (1.29–2.56){ (1.41–2.15){ (1.71–2.62){ (1.15–1.74){ (1.14–1.72){ (0.94–2.01) (1.11–1.72){ (1.09–1.67){

Scandinavian

0.73 0.74 0.86 0.83 0.86 0.58 0.60 0.74 0.77 0.69 0.88

0.96 0.99 0.80 1.01 0.96 0.88 0.90 0.68 0.90 0.89 1.09 1.12 1.04 1.11 1.09

1.43 1.38 1.35 1.32 1.40 1.51 1.42 1.45 1.37 1.43 1.40 1.41 1.01 1.39 1.44

(0.62–0.86){ (0.63–0.88){ (0.62–1.15) (0.68–1.00) (0.72–1.04) (0.44–0.77){ (0.46–0.79){ (0.48–1.18) (0.56–1.04) (0.51–0.94){ (0.74–1.05)

(0.80–1.15) (0.83–1.18) (0.55–1.15) (0.83–1.22) (0.80–1.15) (0.70–1.11) (0.71–1.15) (0.42–1.10) (0.71–1.16) (0.71–1.13) (0.89–1.34) (0.91-1.38) (0.69–1.56) (0.89–1.38) (0.87–1.36)

(1.23–1.54){ (1.19–1.61){ (0.98–1.86) (1.12–1.54){ (1.20–1.65){ (1.23–1.84){ (1.16–1.75){ (0.96–2.19) (1.11–1.70){ (1.15–1.78){ (1.17–1.67){ (1.18–1.69){ (0.69–1.46) (1.15–1.69){ (1.19–1.73){

Continental

1.18 1.18 2.70 1.22 1.17 1.06 1.06 3.04 1.03 1.27 1.21

0.72 0.73 0.49 0.75 0.79 0.64 0.66 0.44 0.65 0.77 0.80 0.80 0.78 0.88 0.85

1.92 1.91 1.94 1.72 2.02 1.85 1.87 2.24 1.64 1.91 0.80 0.79 0.77 0.81 0.88

(0.76–1.85) (0.75–1.84) (0.74–9.93) (0.72–2.09) (0.76–1.82) (0.56–1.99) (0.57–1.98) (0.58–15.94) (0.50–2.12) (0.72–2.24) (0.76–1.92)

(0.49–1.05) (0.50–1.07) (0.15–1.59) (0.51–1.11) (0.54–1.16) (0.39–1.04) (0.40–1.07) (0.10–1.97) (0.40–1.07) (0.46–1.26) (0.50–1.29) (0.50–1.29) (0.20–3.05) (0.55–1.41) (0.53–1.36)

(1.40–2.63){ (1.39–2.64){ (1.03–3.64){ (1.25–2.38){ (1.47–2.79){ (1.30–2.64){ (1.30–2.69){ (1.07–4.69){ (1.16–2.33){ (1.33–2.76){ (0.56–1.14) (0.55–1.13) (0.43–1.35) (0.55–1.19) (0.61–1.26)

Anglo-Saxon

0.98 0.98 1.04 1.03 1.00 0.72 0.72 0.99 0.86 0.73 0.91

1.03 1.04 0.85 1.12 1.01 0.75 0.76 0.49 0.85 0.70 0.82 0.82 0.79 0.90 0.80

1.43 1.43 1.67 1.26 1.46 1.60 1.60 1.47 1.50 1.64 1.04 1.04 0.96 0.99 1.05

(0.82–1.17) (0.82–1.17) (0.73–1.49) (0.85–1.25) (0.83–1.20) (0.52–1.00) (0.52–1.00) (0.49–1.99) (0.59–1.23) (0.53–1.02) (0.77–1.07)

(0.81–1.30) (0.82–1.31) (0.55–1.30) (0.88–1.42) (0.19–1.28) (0.52–1.10) (0.52–1.11) (0.24–0.96){ (0.57–1.25) (0.47–1.05) (0.63–1.06) (0.63–1.00) (0.46–1.34) (0.68–1.17) (0.62–1.04)

(1.14–1.79){ (1.14–1.79){ (1.10–2.52){ (0.99–1.59) (1.16–1.82){ (1.18–2.18){ (1.17–2.17){ (0.87–2.49) (1.09–2.77){ (1.20–2.24){ (0.80–1.35) (0.81–1.35) (0.57–1.61) (0.75–1.31) (0.81–1.36)

Southern

1.22 1.22 1.32 1.32 1.28 1.01 1.02 1.13 1.08 1.07 1.21

1.23 1.24 1.28 1.22 1.24 1.14 1.15 1.31 1.13 1.15 1.13 1.14 0.87 1.11 1.14

1.56 1.54 1.69 1.48 1.57 1.53 1.50 1.54 1.43 1.54 1.01 1.00 1.23 1.01 1.02

(1.10–1.36){ (1.10–1.36){ (1.09–1.60){ (1.18–1.46){ (1.15–1.42){ (0.86–1.17) (0.87–1.19) (0.85–1.48) (0.92–1.26) (0.92–1.25) (1.09–1.33){

(1.03–1.47){ (1.04–1.49){ (0.90–1.81) (1.01–1.47){ (1.03–1.48){ (0.91–1.43) (0.91–1.44) (0.84–2.04) (0.89–1.43) (0.92–1.45) (0.94–1.35) (0.95–1.37) (0.64–1.18) (0.92–1.34) (0.95–1.36)

(1.35–1.81){ (1.33–1.79){ (1.24–2.30){ (1.27–1.72){ (1.36–1.82){ (1.28–1.83){ (1.26–1.80){ (1.08–2.21){ (1.20–1.71){ (1.29–1.84){ (0.86–1.17) (0.86–1.17) (0.88–1.72) (0.86–1.19) (0.88–1.20)

Eastern

Table 5 Prevalence ratio (PR) of exposure to work-related psychosocial hazards in women compared to men in Europe and welfare state regimes by occupational social class in a sample of workers (n527, 465) from the Fourth European Working Conditions Survey (2005)

Campos-Serna et al. Welfare state regimes and gender inequalities

VOL .

19

NO .

3

187

188

Note: *Prevalence ratio adjusted (aPR) for age and 95% confidence intervals (95%CI); M1: aPR for age and marital status; M2: aPR for age and family burden; M3: aPR for age and sector of economic activity of the company; M4: aPR for age and part-time/full-time and each of their 95% confidence intervals (95% CI) Men is the reference group in all models. { Statistically significant aPR. …: sample too small to calculate the aPR.

1.22 1.25 1.27 1.25 (0.77–1.08) (0.61–1.30) (0.77–1.11) (0.76–1.07) 0.91 0.89 0.93 0.91 (0.77–1.95) (0.25–4.07) (0.67–2.03) (0.72–1.90) 1.23 1.01 1.17 1.17 (0.74–1.05) (0.65–1.25) (0.79–1.17) (0.82–1.20) 0.88 0.90 0.96 1.00 (0.90–1.57) (0.89–2.22) (0.89–1.63) (0.92–1.60) 1.19 1.41 1.20 1.21 M1 M2 M3 M4

Work-related psychosocial hazards/welfare states

Table 5 Continued

1.12 1.18 1.17 1.17

(1.04–1.20){ (1.03–1.35){ (1.08–1.27){ (1.09–1.26){

Southern Anglo-Saxon Continental Scandinavian Europe

(1.10–1.35){ (1.05–1.50){ (1.14–1.41){ (1.13–1.38){


Eastern

Campos-Serna et al.


those by Dragano et al.,41 who observed that the psychosocial work environment (low control and higher effort-reward imbalance) was better in those welfare state regimes with more generous systems of social and labor market protection policies. However, the data in that study41 were not analyzed separately by gender and were not stratified by occupational social class. Nevertheless, our results which show the highest gender inequalities in the exposure to workrelated psychosocial hazards in those welfare regimes with greater social protection policies are consistent with those of Bambra et al.,45 who used a similar classification of countries and which also analysed data separately by gender, although they did not include Eastern European countries and did not stratify by occupational social class. Bambra et al.45 found that women in the Scandinavian regime and in the Netherlands were more likely to report poor selfperceived health status than men. Their findings make sense in relation to our results of higher exposure to work-related psychosocial hazards in women than men (high strain, iso-strain, and effort– reward imbalance) in those regimes with more favorable social protection policies. Furthermore, it has been suggested that the mechanisms at play in terms of gender and health, and by extension, of the exposure to work-related psychosocial hazards, cannot be overcome by the traditional social democratic welfare interventions of income redistribution and extensive public service provision alone.65 One obvious explanation for the high gender inequalities in the exposure to work-related psychosocial hazards observed in the Scandinavian welfare state could be that the social and labor market protection policies promoted in this regime do not properly integrate the gender perspective; that is to say, the perceptions, experiences, knowledge and interests of women and men, and do not situate the gender equality issue at the center of analyses and policy decision. Thus, they are not adequate to tackle gender inequalities in exposure to these psychosocial hazards.66–68 It could be that Scandinavian social and work protection policies may not be sufficient to guarantee equal opportunities for employed women and men entering in a competitive androcentric labor market.14,19 Contrary to what we expected,41 the highest gender inequalities were observed both in welfare states with a long tradition of comprehensive social protection policies (e.g. Scandinavian regimes) and those welfare states with a much weaker tradition of social protection policies (e.g. Eastern regimes). But while gender inequalities in the Scandinavian welfare states were concentrated among the higher occupational strata (e.g. managers/professionals and clerks/services/shop workers), in Eastern regimes they were

2013

VOL .

19

NO .

3

Campos-Serna et al.

found among all occupational strata. Because gender inequalities in the exposure to work-related psychosocial hazards exist across all occupational strata in the Eastern regimes, but are concentrated only in managers/professionals and clerks/services/shop workers in the Scandinavian regimes, it could be argued that social protection policies have only a positive effect in reducing gender inequalities in the exposure to work-related psychosocial hazards in the lowest occupational stratum in the Scandinavian welfare states (e.g. manual workers). A distinct feature of the Scandinavian welfare states is the high proportion of women’s representation and participation in politics, their involvement in the labor market, and the extensive family and labor market policies aimed at achieving gender equality. Furthermore, the combination of political mobilization for women’s rights, the increased participation of women in both politics and work life, along with the development of public policies to support these changes, has had an important influence on the formation of this welfare state regime.69 However, given the lack of data in our study on how these policies are implemented we cannot disentangle whether social protection policies have focused primarily on the lowest occupational stratum or whether these policies have affected only the lowest occupational stratum. Future research should explore these two hypotheses. Further research among workers in the upper occupational stratum is also needed to clarify the causal pathways by which legislation on workplace organization can effectively reduce gender inequalities in the exposure to workrelated psychosocial hazards. In this regard, it would be interesting to explore whether there is a causal link between the gender inequalities shown in our study and the work-related legislation that has been implemented in each of the welfare state regimes. In addition, more efforts should be made to explain what is producing the considerable gender inequalities in the professions in the highest occupational stratum. Previous studies have shown that the macro-level characteristics of welfare state regimes are linked to micro-level working conditions through pathways such as occupational safety legislation, dismissal protection laws, or minimum wage policies,53 effectively reducing the exposure to work-related psychosocial hazards.41 However, the fact that such policies do not consider a gender perspective could lead to their being ineffective in reducing gender inequalities.70 For example, policies with a gender perspective should promote egalitarian distribution of work/home responsibilities and wages between women and men; and should also include provision of child daycare, education, housing, medical services, and other services dedicated to children, the elderly, and dependent citizens.70 Nevertheless, our study did not show the


expected pattern — that is, that increased social protection or social benefits redistributive policies in a welfare state would be associated with fewer gender inequalities in the exposure to work-related psychosocial hazards. This could reflect the fact that the typology of welfare state regime used in our study does not substantially incorporate the level of labor market protection policies or the distribution of workers’ bargaining power in the market and vis-a`-vis the government (e.g. worker’s organizations).71 Accordingly, if managers/professionals in the Scandinavian welfare state are less likely to be protected either by state protection policies or trade union involvement, this would explain the higher prevalence of exposure among women in this higher occupational stratum.55 However, this explanatory hypothesis seems implausible considering that the Scandinavian regime has the highest levels of trade union participation, social pacts, and social security expenditure.55 The fact that the prevalence of the exposure to work-related psychosocial hazards in both women and men is much higher among clerk/service/shop and manual workers than among managers/professionals, is consistent with some previous studies,46,47,50 but contrasts with others.72 It is also important to note that although the prevalence of the exposure to workrelated psychosocial hazards is much higher for both genders in the lowest occupational strata (clerk/ service/shop and manual workers), the difference between women and men is much higher among the highest occupational strata (managers/professionals) across all welfare state regimes. These findings lead us to hypothesize that other social mechanisms related to gender roles, which are determined by androcentric processes of socialization and education, could perpetuate male dominant roles and therefore act as an important determinant of the gender inequalities observed in our study in the distribution of work-related psychosocial hazards.70 Androcentricity of work organization has an important effect on the design of tasks carried out by men and women with the same job title. For example, a woman with the job title of butcher may work behind a delicatessen counter and interact with the public, while a man with the same title may work behind a meat counter cutting large pieces of meat.19 This situation would lead to an unequal distribution of working conditions and hazards between the two sexes.16,17 Some previous studies have also suggested that this mechanism is at play in terms of gender and health inequalities. It is possible that the traditional Scandinavian welfare intervention of income redistribution and extensive public service provision is not sufficient to overcome such inequalities.45,65 It has also been argued that such policies have transferred women’s economic dependency from the family to the


2013

VOL .

19

NO .

3

189

Campos-Serna et al.


state, from private to public patriarchy.73,74 Thus, some studies posit that these social protection policies alone cannot adequately overcome gender-based inequities in occupational health without accompanying changes at the cultural and societal levels.45 This transfer from ‘‘the private to the public patriarchy’’ is mainly seen in the Continental welfare states, where generous ‘‘out-of-work-subsidies’’ are keeping women out of work (e.g. long maternity leave in Germany). In contrast, Scandinavian welfare states are keeping women in the labor market which could result in less favorable outcomes on work-related psychosocial hazards. It is precisely in countries that promote policies to keep a large proportion of women (young mothers) in the labor market — women who at the same time have heavy family responsibilities — that the double burden may be expected to be very strong. This argument is consistent with the reduced PRs found in the Scandinavian welfare states among managers/professionals for job strain, iso-strain, and effort–reward imbalance after adjusting for the time spent in caring for children and the elderly (family burden), a task which is traditionally assumed by women.14,39 Although it has been argued that in cases of double burden, women are more tempted than men to exchange full-time jobs for part-time jobs (with lower quality: less job control, similar job demands, and reduced wages),40 this may be only part of the explanation, since we noticed no differences in the PRs after adjustment for part-time/full-time status. Thus, in the Scandinavian welfare state, the larger proportion of women among manager/professionals that do not change working hours or job contents, while at the same time are assuming traditional family responsibilities in caring for children and the elderly could be the main explanation for women’s higher exposure to work-related psychosocial hazards. Another explanation — not mutually exclusive — could be that the double burden issue added to the pressure of the work sphere could could affect female managers/professionals, clerk/service/shop, and manual in different ways. For example, the effect on managers/professionals could be more independent of the welfare state regime in place, while women working as clerk/service/shop and manual may generally assume a more traditional role as wife, mother, and worker, while being ‘‘protected’’ under the ‘‘male breadwinner’’ model.45 It has been previously shown that the welfare state regime influences the time men and women dedicate to household tasks.39,40,75 In those countries where public policy provides for child daycare and where men are eligible to take parental leave, women tend to spend less of their time on this unpaid work.40 Nevertheless, these measures may still require women to stay at home or in part-time jobs. Therefore, we should expect to find greater differences

190


among welfare state regimes, unless these social protection policies are insufficient to reduce gender inequalities among the higher occupational strata, a possibility that would partly explain our results, but should also be investigated in future research. Our study showed that gender inequalities among managers/professionals remained wider than in the lower occupational strata not only in the Scandinavian welfare states, but also across all welfare state regimes. A possible explanation could be a difference in self-perceived exposure to work-related psychosocial hazards, and of what exactly is understood by ‘‘a good job,’’ since standards for job quality may be different between countries, based on social identity aspects related to the socio-culture context in the welfare state regimes.76 In addition, the androcentric models of education could influence women’s perceptions of their work. Cultural and educational values may also shape structures of social inequality, the division of labor, the labor market, and the family, mitigating and/or reinforcing the effect of social welfare policies. Ideas and perceptions may vary according to material interests of social groups, but can also be shared by a majority of the population independent of their material interests.77 These cultural and educational values could facilitate or hinder the worker’s report of being exposed to work-related psychosocial hazards in the different socio-cultural contexts of the welfare state regimes. This could mean that because of the differences in the educational process and the different socio-cultural contexts of the welfare state, women in the Scandinavian welfare state are more sensitive to the perception of work-related psychosocial hazards than in the Southern regime.45,76 Whereas in Norway and Sweden, the protective laws for equal education, equal opportunity for men and women to access the workplace, and against discrimination due to gender or sexual role have been in force for more than 20 years, such legislation has only recently been promoted in Spain.78 Although the results of our study could not support this assertion, there is enough empirical evidence77,79–81 to reasonably hypothesize that working women in Scandinavian countries may be more educated regarding workplace policies and perhaps also regarding feminist critiques of traditional gender roles than women in Southern welfare regimes. This could make women more sensitive to the perception of work-related psychosocial hazards. Nevertheless, more specific studies should be developed to test this explanatory hypothesis. Another plausible hypothesis explaining the higher gender inequalities found among managers/professionals in the Scandinavian welfare states could be that in those welfare states with a longer tradition of egalitarian working policies, women are increasingly

2013

VOL .

19

NO .

3

Campos-Serna et al.

moving to the top of the hierarchy within the workplace. However, while gender equality may be growing, women could be occupying intermediate positions involving important responsibilities but lacking the level of control needed to mitigate job demands without interfering with their care-giving duties at home. More specific studies are required to confirm this hypothesis. It is possible that gender egalitarian policies have major unexpected side effects. Mandel and Semyonov82 claim that employing women in the public sector in countries with highly developed gender egalitarian policies is likely to increase rather than decrease the gender gap in rewards, based on an argument that public sector jobs could be linked to worse work-related psychosocial hazards, such as job strain and effort– reward imbalance. In addition, Mandel and Semyonov83 argue that in highly developed welfare states the ‘‘glass ceiling’’ has become lower and wider resulting in reduce access to positions of power, authority, and with high economic rewards for women in higher occupational groups. In general, state policies do not enhance women’s occupational and economic achievements, since none of them seriously challenge the traditional distribution between men and women of market–family responsibilities. The Scandinavian gender egalitarian care policies have shown that the availability of a publicly-funded social care system unintentionally depresses women’s earnings by intensifying their concentration in feminized service jobs and lowering their representation in highly-paid, maledominated positions.84 In Scandinavian welfare states only about 10 percent of women workers belong to the highest earnings quintile compared to about 30 percent of men. Conversely, In the conservative welfare states, especially the familistic Southern welfare states, women would are equally represented due to a relatively selective female labor force that is under strong pressure to adopt the male model of commitment to work.85 On the other hand, the fact that no gender inequalities were found among manual workers in Scandinavia, Anglo-Saxon and Southern regimes and that more men than women reported high strain and iso-strain in the Continental welfare states, could be a possible consequence of the higher efforts made by these welfare state regimes to reduce gender inequalities in the exposure to work-related psychosocial hazards in the lowest occupational strata, in which there was a higher prevalence of exposure for both, women and men across all welfare state regimes. The Eastern regime was the only one in which the exposure to work-related psychosocial hazards was higher in women than in men across all occupational social classes. This regime has experienced extensive economic upheavals, undertaken comprehensive social


reforms,30,86 and taken on policy characteristics associated with the more liberal welfare state regimes. In comparison with the other welfare state regimes, the Eastern regime has a limited health service provision and overall population health is relatively poor.30 The Eastern typology comprises the most underdefined and understudied region and represents a new challenge in the welfare state modelling debate.25 Among clerk/service/shop workers across all welfare state regimes, gender inequalities were not higher when compared to the traditionally male-dominated managers/professionals and manual worker strata, despite the fact that clerk/service/shop work mostly represents feminized ‘‘pink collar’’ occupations with higher levels of psychosocial hazard exposure. In the same way that women working in traditionally maledominated occupations experience greater discrimination than those working in female-dominated occupations, women may be more exposed to greater work-related psychosocial when working in traditionally male-dominated occupations. Part of the gender inequalities observed in the higher occupational social strata in our study could also be explained by economic sector. After adjusting for sectors, PRs decreased an average of 15% among managers/professionals. This could be explained by the fact that women mainly work in sectors with greater levels of exposure to work-related psychosocial hazards, for example, the service sector.15 Although the cross-sectional design represents a study limitation, a major strength is that it was carried out with data from a large population, which were collected and controlled through a rigorous quality-protocol for the EWCS.59 Another limitation may be due to our choice of welfare state typology, which may have obscured or highlighted some differences in employment organization and work design. It is also possible that neither socio-cultural elements nor the state approaches to reducing gender inequalities were captured by the classification typology followed in our study.31,87 The fact that researchers do not yet completely agree on which countries should be included in the Eastern welfare state regime must also be considered.31,34 However, the rest of the welfare state typologies used in our study (Scandinavian, Continental, Anglo-Saxon, and Southern) have been described as one of the most empirically accurate one.33,34,45 Another important limitation is the possibility that the models used to characterize work-related psychosocial hazards were not properly represented in the questionnaire items of the EWCS.59 A further limitation is the fact that only the summary scales (job strain, iso-strain, and effort–reward imbalance) were considered, rather than their constituent subdimensions (high demand, low control, low support,


2013

VOL .

19

NO .

3

191

Campos-Serna et al.


high effort, and low reward). Although the internal consistency and reliability of the summary scales (Cronbach’s alpha) was quite acceptable,88,89 doing similar analyses with the separate sub-dimensions instead of the summary scales could have shown different results. Nevertheless, exposure to workrelated psychosocial hazards as reflected in the summary scales has been found to be correlated with physical20 and mental disorders.21 Although the response rate varied widely across all countries, the average was quite acceptable for this type of study. In addition, a specific weight was applied to control the non-response rate.59 Another possible limitation could stem from the difficulties of translating the survey questions into different languages, cultures, and contexts, which could involve a certain degree of misunderstanding. However, the survey used validated questions and scales. Furthermore, both the linguistic issue and the cultural and contextual aspects were taken into account in the translation process.59 Thus, misunderstanding of the questionnaire should not introduce a limitation per se. In addition, trained and experienced interviewers participated in the survey process and the interviewers were subject to random quality controls. A final limitation could be researchers’ decision to operationalize the occupational social class into three categories instead of the five proposed by the RGSC classification.62 However, this decision was necessary so that the occupational groups would be sufficiently large to maintain the statistical power in the analysis. In conclusion, taking into account the aforementioned limitations, our results suggest that the unequal distribution of work-related psychosocial hazards among women and men persists in the highest occupational strata across all welfare state regimes, with only slight variation. Contrary to what we hypothesized, gender inequalities in the exposure to work-related psychosocial hazards were not lower in those welfare

state regimes with greater levels of wealth redistribution and more universal social protection policies (e.g. Scandinavian regime) and more specifically among the highest occupational strata. Nevertheless, in the lowest occupational strata, gender inequalities were lower among those welfare state regimes with more comprehensive social protection policies compared to much less comprehensive welfare state regimes (e.g. Eastern). Transcultural differences in gender roles and perceptions of exposure to hazards in the workplace and at home, as well as the different impact of social protection policies on reducing gender inequalities by occupational social class could help explain the gender inequalities in the exposure to work-related psychosocial hazards observed in our study. A challenge for future research in this field will be to analyze how exposure to work-related psychosocial hazards could affect differently employed women’s and men’s health across all welfare state regimes and occupational social classes.

Acknowledgements We thank the Research Group for Occupational and Environmental Medicine in the Department of Public Health and Primary Health Care from the University of Bergen, Norway, for their valuable comments on the manuscript. We also thank Emily Felt for her contribution in reviewing the English and writing style of the manuscript. This paper will be used as part of Javier Campos-Serna’s PhD training programme and dissertation at the Universitat Pompeu Fabra, Barcelona, Spain.

Disclosure Fundings This work received no specific funding from any funding agency in the public, commercial, or not-forprofit sector.

Conflicts of interest None declared.

Appendix 1: Definition of the work-related psychosocial hazards from the questionnaire of the Fourth European Working Conditions Survey (2005) Work-related psychosocial hazard

Work-related psychosocial factors

Item*

Definition

High strain (high demand and low control) High demand Q20B_a Q20B_b Q25F Q25L Q25M

192


2013

The The The The The

worker worker worker worker worker

VOL .

19

is working at very high speed is working to tight deadlines has not enough time to get the job done finds the job intellectually demanding finds the job emotionally demanding

NO .

3

Campos-Serna et al.


Appendix 1 Continued Work-related psychosocial hazard

Work-related psychosocial factors

Item*

Definition

Low control Q25_d Q25_e Q25_g Q25_h Q25_j

The worker has influence over the choice of his or her working partners The worker can take his break when he or she wishes The worker is free to decide when to take holidays or days off The worker has the opportunity to do what he or she does best The worker is able to apply his or her own ideas in the workplace

Iso-strain (high demand, low control, and low social support) Low support Q25_a Q25_b Q25_c

The worker cannot get assistance from his or her colleagues when he or she asks for it The worker cannot get assistance from his or her supervisors/boss when he or she asks for it The worker cannot get external assistance when he or she asks for it

Effort–reward imbalance (high effort and low reward) High effort Q20B_a Q20B_b Q25F Q25L Q25M

The The The The The

worker worker worker worker worker

is working at very high speed is working to tight deadlines has not enough time to get the job done finds the job intellectually demanding finds the job emotionally demanding

Q25_i

The worker’s job does not gives him or her the feeling of work well done The worker has not the feeling of doing useful work The worker disagrees or strongly disagrees about the perception of being well paid for the work he or she does The worker disagrees or strongly disagrees about the perception that his or her job offers him or her good prospects for career advancement The worker disagrees or strongly disagrees about the feeling of being like ‘at home’ in his or her organization The worker disagrees or strongly disagrees about the perception that at work, he or she has opportunities to learn and grow The worker disagrees or strongly disagrees about the perception that at work, he or she has very good friends

Low reward

Q25_K Q37_b Q37_c

Q37_d Q37_e

Q37_f

Note: *The coding showed for the items is the one which is used in the EWCS (2005).

References 1 2

3 4

5

6

7

Karasek R. Job demands, job decision latitude and mental strain: implications for job design. Admin Sci Q. 1979;24:285–308. Johnson JV, Hall EM. Job strain, work place social support, and cardiovascular disease: a cross-sectional study of a random sample of the Swedish working population. Am J Public Health. 1988;78:1336–42. Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol. 1996;1:27–41. Siegrist J, Starke D, Chandola T, Godin I, Marmot M, Niedhammer I, et al. The measurement of effort–reward imbalance at work: European comparisons. Soc Sci Med. 2004;58:1483–99. Bildt C, Michelsen H. Gender differences in the effects from working conditions on mental health: a 4-year follow-up. Int Arch Occup Environ Health. 2002;75:252–8. Bond MA, Punnett L, Pyle JL, Cazeca D, Cooperman M. Gendered work conditions, health, and work outcomes. J Occup Health Psychol. 2004;9:28–45. Peter R, Siegrist J, Hallqvist J, Reuterwall C, Theorell T. Psychosocial work environment and myocardial infarction:

8 9

10 11

12 13

improving risk estimation by combining two complementary job stress models in the SHEEP Study. J Epidemiol Community Health. 2002;56:294–300. Vermeulen M, Mustard C. Gender differences in job strain, social support at work, and psychological distress. J Occup Health Psychol. 2000;5:428–40. Karasek R, Brisson C, Kawakami N, Houtman I, Bongers P, Amick B. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol. 1998;3:322–55. Nelson D, Burke R. Gender, work stress, and health. Washington (DC): American Psychological Association; 2002. Magnusson Hanson LL, Theorell T, Oxenstierna G, Hyde M, Westerlund H. Demand, control and social climate as predictors of emotional exhaustion symptoms in working Swedish men and women. Scand J Public Health. 2008;36:737–43. Li J, Yang W, Cho SI. Gender differences in job strain, effort– reward imbalance, and health functioning among Chinese physicians. Soc Sci Med. 2006;62:1066–77. Paoli P, Merllie´ D. Third European survey on working conditions 2000 [document on the Internet]. Luxembourg:


2013

VOL .

19

NO .

3

193

Campos-Serna et al.

14

15

16*

17*

18

19 20

21

22

23

24

25 26 27*

28 29 30* 31

32

33

34*

35

36

37

194


Office for Official Publications of the European Communities; 2001 [cited 2012 Dec 30]. Available from: http://www.euro found.europa.eu/publications/htmlfiles/ef0121.htm Chodorow N. Glass ceilings, sticky floors, and concrete walls: internal and external barriers to women’s work and achievement. In: Seelig B, Paul R, Levy C, editors. Constructing and deconstructing woman’s power. London: Karnac; 2002. p. 18– 28. Benach J, Muntaner C, Santana V. Employment conditions and health inequalities. Final Report to the WHO Commission on Social Determinants of Health [document on the Internet]. Geneva: WHO; 2007 [cited 2012 Dec 30]. Available from: http://www.who.int/social_determinants/ resources/articles/emconet_who_report.pdf Messing K, Mager Stellman J. Sex, gender and women’s occupational health: the importance of considering mechanism. Environ Res. 2006;101:149–62. Krieger N. Genders, sexes, and health: what are the connections–and why does it matter? Int J Epidemiol. 2003;32:652–7. Messing K, Dumais L, Courville J, Seifert AM, Boucher M. Evaluation of exposure data from men and women with the same job title. J Occup Med. 1994;36:913–7. McDiarmid MA, Gucer PW. The ‘GRAS’ status of women’s work. J Occup Environ Med. 2001;43:665–9. Kuper H, Marmot M. Job strain, job demands, decision latitude, and risk of coronary heart disease within the Whitehall II study. J Epidemiol Community Health. 2003;57:147–53. Stansfeld SA, Fuhrer R, Shipley MJ, Marmot MG. Work characteristics predict psychiatric disorder: prospective results from the Whitehall II Study. Occup Environ Med. 1999;56:302–7. Gimeno D, Benavides FG, Amick BC, 3rd, Benach J, Martinez JM. Psychosocial factors and work related sickness absence among permanent and non-permanent employees. J Epidemiol Community Health. 2004;58:870–6. Krause N, Lynch J, Kaplan GA, Cohen RD, Goldberg DE, Salonen JT. Predictors of disability retirement. Scand J Work Environ Health. 1997;23:403–13. Kivimaki M, Virtanen M, Elovainio M, Kouvonen A, Vaananen A, Vahtera J. Work stress in the etiology of coronary heart disease — a meta-analysis. Scand J Work Environ Health. 2006;32:431–42. Esping-Andersen G. The three worlds of welfare capitalism. London: Polity; 1990. Arts W. Three worlds of welfare or more? J Eur Soc Policy. 2002;12:137–58. Bambra C. Sifting the wheat from the chaff: a twodimensional discriminant analysis of welfare state regime theory. Soc Policy Adm. 2007;41:1–28. Bambra C. Decommodification and the worlds of welfare: revisited. J Eur Soc Policy. 2006;16:73–80. Ferrera M. The southern model of welfare in social Europe. J Eur Soc Policy. 1996;6:17–37. Eikemo TA, Bambra C. The welfare state: a glossary for public health. J Epidemiol Community Health. 2008;62:3–6. Bambra C. Going beyond The three worlds of welfare capitalism: regime theory and public health research. J Epidemiol Community Health. 2007;61:1098–102. Bambra C, Eikemo TA. Welfare state regimes, unemployment and health: a comparative study of the relationship between unemployment and self-reported health in 23 European countries. J Epidemiol Community Health. 2009;63:92–8. Eikemo TA, Bambra C, Judge K, Ringdal K. Welfare state regimes and differences in self-perceived health in Europe: a multilevel analysis. Soc Sci Med. 2008;66:2281–95. Eikemo TA, Huisman M, Bambra C, Kunst AE. Health inequalities according to educational level in different welfare regimes: a comparison of 23 European countries. Sociol Health Illn. 2008;30:565–82. Borchorst A, Siim B. Women and the advanced welfare state — a new kind of patriarchal power? London: Hutchinson; 1987. Kremer M. How welfare states care: culture, gender and parenting in Europe. Amsterdam: Amsterdam University Press; 2007. Siaroff A. Work, welfare and gender equality: a new typology. In: Sainsbury D, editor. Gendering welfare states.London: Sage Publication; 1994.


38 Bussemaker J, van Kersbergen K. Gender and welfare states: some theoretical reflections. In: Sainsbury D, editor. Gendering welfare states. London: SAGE Publications; 1994, 8–25. 39 Craig L, Mullan K. How mothers and fathers share childcare: a cross-national time-use comparison. Am Sociol Rev. 2011;76:834–61. 40 Hook JL. Gender inequality in the welfare state: sex segregation in housework, 1965–2003. AJS. 2010;115:1480– 523. 41* Dragano N, Siegrist J, Wahrendorf M. Welfare regimes, labour policies and unhealthy psychosocial working conditions: a comparative study with 9917 older employees from 12 European countries. J Epidemiol Community Health. 2010;65: 793–9. 42 Sekine M, Chandola T, Martikainen P, Marmot M, Kagamimori S. Socioeconomic inequalities in physical and mental functioning of British, Finnish, and Japanese civil servants: role of job demand, control, and work hours. Soc Sci Med. 2009;69:1417–25. 43 Bartley M, Blane D, Montgomery S. Health and the life course: why safety nets matter. BMJ. 1997;314:1194–6. 44 Hasselhorn HM, Tackenberg P, Peter R. Effort–reward imbalance among nurses in stable countries and in countries in transition. Int J Occup Environ Health. 2004;10:401–8. 45* Bambra C, Pope D, Swami V, Stanistreet D, Roskam A, Kunst A, et al. Gender, health inequalities and welfare state regimes: a cross-national study of 13 European countries. J Epidemiol Community Health. 2009;63:38–44. 46 Bosma H, Marmot MG, Hemingway H, Nicholson AC, Brunner E, Stansfeld SA. Low job control and risk of coronary heart disease in Whitehall II (prospective cohort) study. BMJ. 1997;314:558–65. 47 Siegrist J, Benach J, McKinght A, Goldblatt P, Muntaner C. Employment arrangements, work conditions and health inequalities: report on new evidence on health inequality reduction, produced by Task Group 2 for the Strategic Review of Health Inequalities post 2010. London: Marmot Review; 2009. 48 Ibrahim S, Smith P, Muntaner C. A multi-group cross-lagged analyses of work stressors and health using Canadian National sample. Soc Sci Med. 2009;68:49–59. 49 Artazcoz L, Borrell C, Cortes I, Escriba-Aguir V, Cascant L. Occupational epidemiology and work related inequalities in health: a gender perspective for two complementary approaches to work and health research. J Epidemiol Community Health. 2007;61(Suppl 2):S39–45. 50 Campos-Serna J, Ronda-Perez E, Artazcoz L, Benavides FG. [Gender inequalities in occupational health in Spain]. Gac Sanit. 2012;26:343–51. Spanish. 51 Matthews S, Stansfeld S, Power C. Social support at age 33: the influence of gender, employment status and social class. Soc Sci Med. 1999;49:133–42. 52 Griffin JM, Fuhrer R, Stansfeld SA, Marmot M. The importance of low control at work and home on depression and anxiety: do these effects vary by gender and social class? Soc Sci Med. 2002;54:783–98. 53 Fenwick R, Tausig M. The macroeconomic context of job stress. J Health Soc Behav. 1994;35:266–82. 54* Muntaner C, Benach J, Chung H, Edwin NG, Schrecker T. Welfare state, labour market inequalities and health. In a global context: an integrated framework. SESPAS report 2010. Gac Sanit. 2010;24(Suppl 1):S56–61. 55 Navarro V, Shi L. The political context of social inequalities and health. Soc Sci Med. 2001;52:481–91. 56 Swedish Work Environment Authority. The Work Environment Act [document on the Internet]. Solna: Swedish Work Environment Authority; 2010 [cited 2012 Dec 30]. Available from: http://www.av.se/inenglish/lawandjustice/ workact/ 57 Messing K, Silverstein BA. Gender and occupational health. Scand J Work Environ Health. 2009;35:81–3. 58 Marmot M, Bell R. Challenging health inequalities — implications for the workplace. Occup Med (Chic Ill). 2010;60:162–4. 59 European Foundation for the Improvement of Living and Working Conditions. Fourth European Working Condition Survey 2005 [document on the Internet]. Dublin: Eurofound; 2005 [cited 2012 Jan 23]. Available from: http://www.euro found.europa.eu/surveys/ewcs/2005/index.htm 60 Eurostat. The European Union labour force survey — Methods and definitions — 2001–2003 [document on the

2013

VOL .

19

NO .

3

Campos-Serna et al.

61

62

63

64

65

66

67

68

69

70* 71

72

73

74*

Internet]. Luxemburg: Office for Official Publications of the European Communities; 2003 [cited 2012 Dec 30]. Available from: http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/ KS-BF-03-002/EN/KS-BF-03-002-EN.PDF Krieger N, Williams DR, Moss NE. Measuring social class in US public health research: concepts, methodologies, and guidelines. Annu Rev Public Health. 1997;18:341–78. Bartley M. Measuring socio-economic position. In: Bartley M, editor. Health inequality: an introduction to theories, concepts and methods. Cambridge: Polity Press; 2004. p. 22–34. Walters V, McDonough P, Strohschein L. The influence of work, household structure, and social, personal and material resources on gender differences in health: an analysis of the 1994 Canadian National Population Health Survey. Soc Sci Med. 2002;54:677–92. Artazcoz L, Escriba-Aguir V, Cortes I. [Gender, paid work, domestic chores and health in Spain]. Gac Sanit. 2004;18(Suppl 2):S24–35. Spanish. Navarro V, Muntaner C, Borrell C, Benach J, Quiroga A, Rodriguez-Sanz M, et al. Politics and health outcomes. Lancet. 2006;368:1033–7. Crespi I. Gender mainstreaming and family policy in Europe: perspectives, researchers and debates. Macerata: Edizioni Universita` di Macerate; 2007. Skjeie H, Teigen M. Political constructions of gender equality: travelling towards a gender balanced society? Nord J Feminist Gender Res. 2005;13:187–97. Kjeldstad R. Gender policies and gender equality. In: Fritzell J, Hvinden B, Kautto M, Kvist J, Uusitalo H, editors. Nordic welfare states in the European context. London: Routledge; 2001, 55–78. Christensen A, Raaum N. Models of political mobilisation. In: Bergqvist C, editor. Equal Democracies? Gender and Politics in the Nordic Countries. Oslo: Scandinavian University Press; 1999. p. 350. Orloff A. Gender in the welfare state. Annu Rev Sociol. 1996;22:51–78. Muntaner C, Borrell C, Ng E, Chung H, Espelt A, RodriguezSanz M, et al. Politics, welfare regimes, and population health: controversies and evidence. Sociol Health Illn. 2011;33:946– 64. Cortes I, Artazcoz L, Rodriguez-Sanz M, Borrell C. [Inequalities in mental health in the working population]. Gac Sanit. 2004;18:351–9. Spanish. Showstack-Sassoon A, Borchost A, Siim B. Women and the advanced welfare state: a new kind of patriarchal power? In: Showstack-Sassoon A, editor. Women and the state: the shifting boundaries of public and private. London: Routledge; 1987, 128–157. Sum B. The Scandinavian welfare states — towards sexual equality or a new kind of male domination? Acta Sociol. 1987;30:255–70.


75 Garcıá-Mainar I, Molina J, Montuenga V. Gender differences in childcare: time allocation in five European countries. Fem Econ. 2011;17:119–50. 76 Bambra C. Health inequalities and welfare state regimes: theoretical insights on a public health ‘puzzle’. J Epidemiol Community Health. 2011;65:740–5. 77 Pfau-effinger B. Culture and welfare state policies: reflections on a complex interrelation. J Soc Policy. 2005;34:3–20. 78 Ministry of Health Social Services and Equality. [Organic Law 3/2007 of March 22 for effective equality of women and men] [document on the Internet]. Madrid: Government of Spain; 2007 [cited 2013 Mar 25]. Available from: http://www.boe.es/ buscar/doc.php?id5BOE-A-2007-6115 79 Pfau-Effinger B. Socio-historical paths of the male breadwinner model — an explanation of cross-national differences. Br J Sociol. 2004;55:377–99. 80 Pfau-Effinger B. Change of family policies in the sociocultural context of European studies. Comp Soc Res. 1999;18:135–59. 81 Pfau-effinger B. Development of culture, welfare states and women’s employment in Europe. Hants: Ashgate Publishing Limited; 2004. 82 Mandel H, Semyonov M. Family policy, wage structures, and gender gaps: sources of earnings inequality in 20 countries. Am Sociol Rev. 2005;70:949–67. 83 Mandel H, Semyonov M. A welfare state paradox: state interventions and women’s employment opportunities in 22 countries. AJS. 2006;111:1910–49. 84 Korpi W, Ferrarini T, Englund S. Egalitarian gender paradise lost? re-examining gender inequalities in different types of welfare states. Stockholm: Swedish Institute for Social Research, Stockholm University; 2009. 85 Mandel H, Shalev M. How welfare states shape the gender pay gap: a theoretical and comparative analysis. Social Forces. 2009;87:1873–912. 86 Kovacs JM. Approaching the EU and reaching the US? Rival narratives on transforming welfare regimes in East-Central Europe. West Eur Polit. 2002;25:175–204. 87 Bambra C. The worlds of welfare: illusory and gender-blind? Soc Policy Soc [Internet]. 2004 [cited 2012 Dec 12];3(3):201–12. Available from: http://journals.cambridge.org/action/display Abstract?fromPage5online&aid5228839 88 Aday L, Cornelius L. Designing and conducting health surveys: a comprehensive guide. 3rd ed. San Francisco (CA): Jossey-Bass; 2006. 89 Frost MH, Reeve BB, Liepa AM, Stauffer JW, Hays RD. What is sufficient evidence for the reliability and validity of patient-reported outcome measures? Value Health. 2007;10(Suppl 2):S94–105. 90 Arter D. Scandinavian politics today. Manchester: Manchester University Press; 1999. 91 Klenner C, Leiber S. Welfare states and gender inequality in Central and Eastern Europe. Brussels: ETUI; 2012.


2013

VOL .

19

NO .

3

195