Does increased gender equality lead to a convergence of health outcomes for men and women? A study of Swedish municipalities
Introduction
It has been proposed that the gender order, loosely defined as the structure of gender relations in a particular society (Connell, 1987; Harding, 1986), is an important determinant of gender differences in health (Chapman Walsh, Sorensen, & Leonard, 1995; Courtenay, 2000a; MacIntyre, Hunt, & Sweeting, 1996), and also that increasing gender equality may contribute to improvements in overall levels of public health (Kawachi, Kennedy, Gupta, & Prothrow-Stith, 1999). Indices of women's status in political participation, economic autonomy, employment and earnings, and reproductive rights, have been used in two US studies to examine the effect of relative gender equality on public health (Chen, Subramanian, Acevedo-Garcia, & Kawachi, 2005; Kawachi et al., 1999), and a country comparative study found that the strength of patriarchy was associated with male mortality (Stanistreet, Bambra, & Scott-Samuel, 2005). Moreover, a WHO study of school-aged children found that countries with a low gender development index score (UNDP, 1995) had a larger gender difference in health complaints (Torsheim et al., 2006). In this study, gender equality is defined as more or less similarity between women and men in every sphere of human life, including the private sphere (Moller Okin, 1989). Given this definition, few earlier studies have examined the relationship between gender equality and public health.
Sweden is ranked second according to the United Nation's Development Programme (UNDP) Gender Empowerment Measure (GEM), which consists of three dimensions: political participation and decision-making, economic participation and decision-making power, and power over economic resources (UNDP, 2003). Since the 1960s, Sweden has developed into what has variously been described as a weak breadwinner (Lewis, 1992), or individual earner-carer model (Sainsbury, 1999), characterised by the individualisation of benefits, equal access to paid work, and a general shift of the burden of domestic work from families to the state. Sweden is, however, also noted for its large occupational sex segregation characterised by a new public/private split with women being predominantly employed in the public sector and men in the private sector (Melkas & Anker, 1997). This makes women potentially doubly vulnerable to welfare state retrenchment both as claimants/service users and employees.
The present study takes as its starting point the argument that the current gender order, which is characterised by dichotomy (sexual segregation) and asymmetry (masculine domination) (Harding, 1986), has effects both on absolute levels of health, and on gender differences in health. The aim is to examine associations between dimensions of gender equality and public health. We have compared Swedish municipalities on a number of indicators in both the private and public sphere, and an additive index, and studied correlations with morbidity (indicated by compensated days from social insurance for sickness absence and disability pension) and mortality (indicated by life expectancy at birth). Women have higher levels of sickness and disability, while men are disadvantaged in terms of life expectancy. In what way does the gender order contribute to these differences, and how might greater gender equality change this pattern?
The gender order fundamentally fixes the social determinants of health along gender lines. Resources by which an individual can choose and direct his/her own life, such as participation in political and economic decision-making, work opportunities and income, as well as time constraints, are unequally distributed between men and women (European Communities (2004a), European Communities (2004b); Korpi, 2000; UNDP, 1995). Overall, women have fewer valued resources, and from this perspective gender equality has the potential to benefit women's health more than men's. A Swedish time-series analysis between 1945 and 1992 showed that economic growth had benefited women's mortality decline more than men's, and associations between the male/female wage ratio and men's excess mortality showed that a relative decline in male resources benefited women relative to men (Hemström, 1999). This finding is consistent with the theory of perception of relative social status having an impact on health through psychosocial pathways (Wilkinson, 1999). Thus, men who experience a loss of social status vis-à-vis women may feel threatened, inadequate, and humiliated—feelings that may be seen as chronic stressors, and that could also indirectly lead to ill-health through increased violence, accidents and alcohol-related deaths. However, previously presented studies show that both men and women may benefit from increased gender equality (Chen et al., 2005; Kawachi et al., 1999).
One way in which greater gender equality may affect health is through women's role expansion, from the traditionally female private sphere into the male-dominated public sphere. A number of studies have investigated the health-effects of ‘multiple roles’ based on two opposing hypotheses. The first is the stress hypothesis, which refers to the idea that individuals with many activities and responsibilities experience increased pressure, conflict and ill-health (Goode, 1960). The second, the expansion hypothesis, refers to the opposite, namely, that individuals with several life roles have health advantages compared to those with fewer roles, as they may compensate for failures in one area with positive circumstances in other areas (Thoits, 1983).
Although there could be an amount of health selection into multiple role occupancy, longitudinal studies suggest that this does not account for the association (Matthews & Power, 2002; McMunn, Bartley, Hardy, & Kuh, 2006). Beneficial health effects could be due to greater economic independence and negotiating power within the family, increased opportunities for social interaction, skill enhancement, and personal growth. Positive effects of women's expansion may, however, be negated by an increase in stress levels in a situation where the division of labour in the private sphere does not change (Blane, Berney, & Montgomery, 2001; Krantz & Östergren, 2001). The conclusion of research on multiple roles is that there is support for the expansion hypothesis over the stress hypothesis, provided that the overall stress level is not excessive (Barnett, 2004; Härenstam, Aronsson, & Hammarström, 2001). Studies examining men are rare, but there are indications that employed men also benefit from ‘high quality’ multiple roles (Barnett, Marshall, & Pleck, 1992) and that caring for children may have a buffering effect on psychosocial symptoms for men working long hours (>50 h/week) (Krantz, Berntsson, & Lundberg, 2005).
It is clear that behavioural factors play a large part in explaining men's disadvantage in some health outcomes, such as mortality through external causes and cardiovascular disease (CVD). Hemström has shown that changes in excess male mortality are heavily determined by alcohol consumption and, to a lesser extent, smoking (Hemström (1998), Hemström (1999)). Other studies also point to the importance of behavioural factors in explaining men's excess mortality and changes in men's disadvantage in life expectancy during the 20th century (Chapman Walsh et al., 1995; Waldron, 1995).
Men's greater economic independence and lesser responsibility for unpaid domestic and caring work provide opportunities for behaviours that put them more at risk, such as driving, excessive alcohol consumption, and irregular food habits. But health behaviours may also be seen as assets in the construction of gender (providing a motive), where men who endorse dominant or traditional norms of masculinity have greater health risks than others (Courtenay (2000a), Courtenay (2000b); Sabo & Gordon, 1996). While health care use and behaviours are socially constructed as forms of idealised femininity, unhealthy behaviours serve as cultural signifiers of ‘true’ masculinity and instruments that men use in the negotiation of social power and status (Courtenay (2000a), Courtenay (2000b)). Greater possibilities for women may also go hand in hand with ideas of the emancipated woman as embracing such risky and health-threatening ‘masculine’ practices, thus coupling opportunity with motive.
Traditional masculinity may translate into refusal to adopt a healthy lifestyle, reluctance to seek help, a desire to be seen to endure physical pain and discomfort, and lack of preventive health care (Meinecke, 1981; O’Brien, Hunt, & Hart, 2005). However, there is also research pointing to a more diverse relationship between masculinity and health (Robertson, 2006). Men's health and illness, as well as the practices of masculinity, vary greatly with socioeconomic status, race and ethnicity (Sabo & Gordon, 1996). In the transition towards gender equality, subordinate men may be threatened by female competition and the (perceived or real) loss of the little power and prestige they enjoyed. This situation might lead to a form of compensatory masculinity, which tends to be self-destructive, and which can also have a negative impact on others, for instance due to drunk driving and high levels of violence (Connell, 1995; Pyke, 1996). The more extreme ‘hypermasculine’ practices may also be seen as constructed as a response to the ‘ascendant masculinity’ of higher class men, which tends to be less overt and seen as more legitimate (Pyke, 1996). Such a development may thus also contribute to increased inequalities between social groups.
Our main hypothesis is that a higher level of gender equality is associated with a convergence of both life expectancy and sickness absence between women and men due to a convergence of the sexes’ welfare resources, roles and stress, and health behaviours. A simple theoretical model of hypothesised associations could be summarised as in Fig. 1. Due to the aggregate nature of the data, we are, however, not able to investigate the hypothesised mediating mechanisms.
For absolute levels of health, e.g., life expectancy, the evidence is conflicting and it would seem that different mechanisms work in different directions simultaneously, thus making it difficult to assess the net effects. Aggregate level studies generally favour a positive association between gender equality and health (Chen et al., 2005; Kawachi et al., 1999). Individual-level studies suggest positive effects of role expansion for women (and possibly men)—unless stress levels become excessive (Barnett, 2004; Barnett et al., 1992; McMunn et al., 2006). There seem to be beneficial effects for men who take on less masculine health beliefs and behaviours, while subgroups of men may react in ways that reinforce gender stereotypes in situations of increased gender equality, thus leading to worse male health (Courtenay (2000a), Courtenay (2000b); O’Brien et al., 2005; Sabo & Gordon, 1996). We might add that for women, leaving the traditionally health-protective female role could be associated with threats to health due to altered behaviour patterns.
Section snippets
Material
Aggregate level data for all Swedish municipalities (n=289) has been taken from Statistics Sweden's (SCB) official statistics, which is freely available on their website. Municipalities were chosen as they are the smallest administrative and political units for which data is available. Municipalities have their own decision-making and tax-levying powers. They have a significant degree of autonomy and administer local matters such as compulsory school and upper secondary education, child-care
Results
For relative health equality in compensated days for sickness and disability, four indicators are consistent with the convergence hypothesis (Table 3). Where part-time employment is more equal between men and women, the differences between men and women in sickness and disability are smaller (b=1.983). For economic resources, both indicators are significantly related to equality in sickness and disability (average income b=0.299 and relative poverty b=0.414). There is also a rather weak
Discussion
The hypothesis that gender equality would be associated with a convergence in health outcomes between men and women was supported for four indicators for sickness and disability and for temporary parental leave for life expectancy, while equality of economic resources was associated with divergence of life expectancy. This divergence was due to men being more negatively affected, not to women benefiting. Equality of economic resources was associated with convergence in sickness and disability
Acknowledgement
We would like to extend our gratitude to Sarah Wamala, Christer Hogstedt and Bo Burström for valuable comments and suggestions on earlier drafts of this paper.
References (49)
- et al.
Domestic labour, paid employment and women's health: Analysis of life course data
Social Science & Medicine
(2001) - et al.
Women's status and depressive symtoms: A multilevel analysis
Social Science & Medicine
(2005) Constructions of masculinity and their influence on men's well-being: A theory of gender and health
Social Science & Medicine
(2000)Explaining differential rates of mortality decline for Swedish men and women: A time-series analysis, 1945–1992
Social Science & Medicine
(1999)- et al.
Adjustment latitude and attendance requirements as determinants of sickness absence or attendance. Emprirical tests of the illness flexibility model
Social Science & Medicine
(2004) - et al.
Women's status and the health of women and men: A view from the States
Social Science & Medicine
(1999) - et al.
Gender differences in health: Are things really as simple as they seem?
Social Science & Medicine
(1996) Sickness absence in female- and male-dominated occupations and workplaces
Social Science & Medicine
(2005)- et al.
Socio-economic gradients in psychological distress: A focus on women, social roles and work-home characteristics
Social Science & Medicine
(2002) - et al.
It's caveman stuff, but that is to a certain extent how guys still operate: Men's accounts of masculinity and help seeking
Social Science & Medicine
(2005)