Social class and self-reported health status among men and women: what is the role of work organisation, household material standards and household labour?

https://doi.org/10.1016/S0277-9536(03)00408-8Get rights and content

Abstract

Social class understood as social relations of ownership and control over productive assets taps into parts of the social variation in health that are not captured by conventional measures of social stratification. The objectives of this study are to analyse the association between self-reported health status and social class and to examine the role of work organisation, material standards and household labour as potential mediating factors in explaining this association.

We used the Barcelona Health Interview Survey, a cross-sectional survey of 10,000 residents of the city's non-institutionalised population in 2000. This was a stratified sample, strata being the 10 districts of the city. The present study was conducted on the working population, aged 16–64 years (2345 men and 1874 women). Social class position was measured with Erik Olin Wright's indicators according to ownership and control over productive assets. The dependent variable was self-reported health status. The independent variables were social class, age, psychosocial and physical working conditions, job insecurity, type of labour contract, number of hours worked per week, possession of appliances at home, as well as household labour (number of hours per week, doing the housework alone and having children, elderly or disabled at home). Several hierarchical logistic regression models were performed by adding different blocks of independent variables.

Among men the prevalence of poor reported health was higher among small employers and petit bourgeois, supervisors, semi-skilled (adjusted odds ratio—aOR: 4.92; 95% CI: 1.88–12.88) and unskilled workers (aOR: 7.69; 95%CI: 3.01–19.64). Work organisation and household material standards were associated with poor health status with the exception of number of hours worked per week. Work organisation variables were the main explanatory variables of social class inequalities in health, although material standards also contributed. Among women, only unskilled workers had poorer health status than the referent category of manager and skilled supervisors (aOR: 3.25; 95%CI: 1.37–7.74). All indicators of work organisation and household material standards reached statistical significance, excepting the number of hours worked per week. In contrast to men, among women the number of hours per week of household labour was associated with poor health status (aOR: 1.02; 95% CI: 1.01–1.03). Showing a different pattern from men in the full model, household material deprivation and hours of household labour per week were associated with poor health status among women.

Our findings suggest that among men, part of the association between social class positions and poor health can be accounted for psychosocial and physical working conditions and job insecurity. Among women, the association between the worker (non-owner, non-managerial, and un-credentiated) class positions and health is substantially explained by working conditions, material well being at home and amount of household labour.

Introduction

In spite of their usefulness in predicting health conditions, measures of social stratification (rankings of individuals along a continuum of income, occupational prestige or education) do not reveal the social mechanisms that explain how individuals reach different levels of material resources (Muntaner & Lynch, 1999). Thus, knowing that a person earns 1M dollars per year does not explain how she obtained her annual income. Social class represents an alternative approach to social stratification by maintaining that social relations of ownership and control over productive assets (physical, financial, organisational; Wright, 2000) determine social inequalities in economic resources. Social class defined in this manner has systematic and important consequences for the well being of individuals: the rights and powers over productive assets are key determinants of the strategies and practices people engage into acquire their income as well as their standards of living (Wright, 2000). Thus, the class position of “large business owner” compels occupants of this class position to hire labour and extract labour effort from them. On the other hand, the “worker” class position compels its occupants to find employment and perform labour effort for the person who hires them (Roemer, 1982).

A small number of mental health studies (Muntaner, Wolyniec, McGrath, & Pulver, 1994; Muntaner & Parsons, 1996; Wohlfarth, 1997; Wohlfarth & Van Den Brink, 1998; Muntaner, Eaton, Diala, Kessler, & Sorlie, 1998a; Muntaner et al., 1998b) have found that social stratification and social class are not empirically equivalent and that they each capture different parts of the social variation in population health. For example, Muntaner et al. (1998a) using a measure of managerial social position found that low level supervisors (who are subordinate to managers but can hire or fire frontline workers) had a higher rate of depression and anxiety disorders than both managers and workers. In addition, that study found that this association could not be accounted for occupation, education or income. Wohlfarth (1997) showed that measures of education and occupational prestige could not account for the social class association with diagnoses of depression and drug use, nor with symptom scales of demoralisation, suicidal ideation and schizoid traits.

The measures of social class used in our investigation originate from a sociological model that has gained empirical support in recent years (e.g., Wright (1985), Wright (2000); Steinmetz & Wright, 1989; Western & Wright, 1994). Thus, we used Wright's social class indicators for the assessment of relations of (1) ownership of productive assets; and (2) control and authority relations in the workplace (control over organisational assets; Wright, 2000). Ownership of productive assets is characterised by the compulsion to hire or sell labour power in private market economies. Thus, property rights over the financial or physical assets used in the production of goods and services generate three class positions: (1) employers, who are self-employed and hire labour; (2) the traditional petit bourgeoisie, who are self-employed but do not hire labour; and (3) workers who sell their labour (Wright, 2000). These social class positions capture the relational properties underlying economic inequality (Wright, 1978). Indicators of productive asset ownership gauge a relational mechanism that generates economic inequality (i.e., deriving income from owning property).

Both neo-material (Lynch, Davey Smith, Kaplan, & House, 2000) and psychosocial (Muntaner et al., 1998a) mechanisms led us to hypothesise that owners might present better overall health than workers. Property owners tend to be wealthier (Keister, 2000), and thus this class position might signal the experience of greater material well being that is conducive to better health (Lynch & Kaplan, 2000). In addition, owners enjoy the predictability and control in one's life that are predictive of better mental health (Turner & Roszell, 1994). Thus, social stratification by itself may not be adequate to capture the psychological effect of ownership. For example, annual income from salaries or wages can oscillate greatly from year to year and leave an individual who does not own assets vulnerable to economic insecurity (Wolff, 1995). Individuals who derive income from the ownership of physical or financial assets may be less subject to the unpredictability and uncontrollability of having to rely exclusively on salaries or wages for income, thus enjoying better health.

Control over organisational assets (power and control in the workplace) is determined by two kinds of relations at work: (a) influence over company policy (e.g., making decisions over number of people employed, products or services delivered, amount of work performed, size and distribution of budgets); and (b) sanctioning authority (granting or preventing pay raises or promotions, hiring, firing or temporally suspending a subordinate) (Wright, 2000). The supervisory and policy making functions of managers allow them to enjoy greater wealth than workers, for example, through income derived from shares of stock, incentives, bonuses, and hierarchical pay scales. Furthermore, workplace authority relations add another mechanism that may impact health, i.e., control over one's work and the ability to extract labour effort from others, increasing one's sense of control and predictability at work (Karasek & Theorell, 1990). The experience of control over work constitutes a second mechanism by which managers may enjoy better overall health than workers.

Wright also includes an ownership of skills/credentials relation as part of his map of class positions (see Fig. 1, the expert, semi-skilled and unskilled class positions; Wright, 2000). These class positions capture the relational nature of credentials. Experts are defined as those holding jobs that require skills, particularly credentialed skills, which are scarce relative to their demand by the market (e.g., paediatric surgeons). Experts enjoy a credential rent: their wages are usually above the cost of the reproduction of their training. And they would be worse-off economically with an equal endowment of society's pool of skills and credentials (Wright, 1985). Semi-skilled and “unskilled” class positions are defined as those holding jobs that require skills that are in large supply, particularly uncredentialled skills (e.g. construction workers).

A number of studies have addressed the pathways linking social position to health outcomes (Yen & Syme, 1999, Kaplan, 1995; Lynch, Kaplan, & Salonen, 1997). Although some authors contend that psychological and material risk factors converge among members of specific social classes (Muntaner, Lynch, & Oates, 1999), two alternative pathways, the psychosocial and the material pathway have emerged in the current literature (Lynch et al., 2000; Marmot & Wilkinson, 2001).

The psychosocial approach tends to emphasise perceptions of relative disadvantage and inequality, sense of control, anxiety, insecurity, social affiliations, culture or health behaviours (Wilkinson, 1996; Marmot & Wilkinson, 2001; Marmot, Bosma, Hemingway, Brunner, & Stansfeld, 1997) For example, Bosma, Schrijvers and Mackenbach (1999) found that low socioeconomic status was related to mortality partly because people with low socioeconomic status more often perceive low control. Bailis, Segall, Mahon, Chipperfield, and Dunn (2001), also described that perceived control over life events, underlies social inequality in health and Schrijvers, Bosma, and Mackenbach (2002) studies hostility as an intermediate factor between social position and health outcomes.

The material (or neo-material approach) emphasises living conditions, economic resources and structural causes of income inequality (e.g., strength of the welfare state; Lynch et al., 2000). For example, Schrijvers, Stronks, van de Mheen and Mackenbach (1999) found that the association between educational level and mortality could be explained by material factors such as financial problems, employment status and house and car ownership. Lantz, House, Lepkowski, et al. (1998) provided evidence that health behaviours do not account for the income effects on mortality. The life-course approach has shown how material deprivation in childhood is related with inequalities in health and health behaviours in adulthood (Lynch et al., 1997, Blane, 1999, Claussen, Davey Smith, & Thelle, 2003). In addition, a recent study has also analysed the association of psychological distress in women in midlife to lifetime risk factors, finding different life course trajectories in women with a high level of distress (Kuh, Hardy, Rodgers, & Wadsworth, 2002).

Men have higher age-specific mortality rates than women, while women have poorer health than men (Verbrugge, 1989; Waldron, 1993). However, health research has frequently overlooked gender differences and even omitted gender, usually women (Macintyre & Hunt, 1997).

Until the late 1980s most research focused on inequalities in men's health according to social stratification, with less attention paid to inequalities in health among women (Macintyre & Hunt, 1997). Researchers have used different theoretical frameworks for men and women. For men, they have focused on paid work while for women they usually attempt to explain the health effects associated with the struggle to balance home and job responsibilities. This second approach has stressed the importance of examining women's health, both in terms of their structural position within society and their family roles (Arber & Khlat, 2002). However, few studies have examined the influence of working conditions on women's health or the effect of role conflict on men's health.

Epidemiologists have only recently begun to consider the processes within households and in women's daily lives that may shape their health. The study of social inequalities in health in women has to take into account the influences beyond the workplace, such as power differentials and women's subordination at home. The sexual division of labour dictates that even when women enter employment, they will typically still have the main responsibility for childcare and housework. Women do not have yet the power to oblige men to undertake an equal share of domestic labour and childcare, no matter how high the status of their employment is (Bartley, 1999). It has been shown that in developed countries women contribute more effort to household chores and childcare and less effort to the workplace than men. As a result, their total workloads appear to be somewhat greater and more diffusely distributed than those of men (Gjerdingen, McGovern, Bekker, Lundberg, & Willemsen, 2000). Additionally, the importance of household labour for women's health has already been noted in several studies (Hall, 1992; Bartley, Popay, & Plewis, 1992; Hunt & Annandale, 1993). Heavy workloads seem to have a particular negative effect on women's overall health (Frankenhaeuser et al., 1989; Lundberg, et al., 1994).

But important class differences may exist as women in capitalist or managerial class positions have the opportunity to employ poor women to care for their homes and children while unskilled workers do not have those opportunities (or cannot afford such help). Thus, household division of labour may be a mediator in the relation between social class and health status (Macintyre, Hunt, & Sweeting, 1996). Several studies have found that poor rated health among women is related to both social stratification and household labour (Lennon & Rosenfield, 1992; Artazcoz, Borrell, & Benach, 2001), but to our knowledge no studies have tried to explain social class inequalities in health through the role of household labour. In this study, we include household labour and household standards of living together with work organisation as possible mediating mechanisms of the relation between social class and health.

The goals of our study are thus to analyse the association between self-reported health status and social class and to examine the role of work organisation, material standards of living and household labour in explaining this association. Furthermore, we will examine whether the patterns are similar for men and women. Self-reported health is a valid predictor of mortality (Idler & Benyamini, 1997) and it is optimal for our purposes as it captures a population's average health (Rose, 1992) rather than the prevalence of specific conditions or high risk. Furthermore, this health indicator has been associated with multiple social risk factors and it is useful for needs assessment and public health interventions.

Section snippets

Study population, sample and data collection

The population frame was the 2000 non-institutionalised population of Barcelona city (1,500,000 inhabitants, Catalonia, Spain). Data were collected as part of the 2000 Barcelona Health Interview Survey, a cross-sectional survey carried out approximately every 5 years since 1983. We generated a representative stratified sample of the non-institutionalised population of Barcelona residents. The sample strata were the 10 Barcelona city districts. In each stratum a random sample of residents was

Description of the population

Ten per cent of men and 14.2% of women declared poor health status. Table 1 describes the population studied and also provides the percentages of poor health status for each variable. More than half of the population belonged to semi-skilled and unskilled worker social classes. We found marked gender differences in household labour (37.6% of women did the housework alone while only 4.7% of men did). The prevalence of poor self-rated health status was higher for small employers and petit

Discussion

Our study has described inequalities in the distribution of social class mediators, showing how unskilled workers are exposed to more hazardous working conditions than other social classes. Household material standards were also associated with social class: semi-skilled and unskilled workers, and also the petit bourgeoisie, had less household appliances and were less likely to hire help with household labour than other social classes. Household labour was done by women and hardly performed by

Conclusion

Overall our study sheds some light on the mechanisms that mediate the relationship between social class and health status among men and women. Among men, work organisation seems to be an important mechanism that translates managerial, high level supervisory, professional and capitalist class positions into better health. Among women, at least in a patriarchal society, the association between poor health and working class position seems to be accounted for not only by hazardous forms of work

References (74)

  • C Muntaner et al.

    Social class, assets, organizational control and the prevalence common groups of psychiatric disorders

    Social Science & Medicine

    (1998)
  • C Muntaner et al.

    A critical appraisal of the demand/control model of the psychosocial work environmentEpistemological, social, behavioral and class considerations

    Social Science & Medicine

    (1993)
  • I Waldron

    Recent trends in sex mortality ratios for adults in developed countries

    Social Science & Medicine

    (1993)
  • T Wohlfarth

    Socioeconomic inequality and psychopathologyAre socioeconomic status and social class interchangeable?

    Social Science & Medicine

    (1997)
  • T Wohlfarth et al.

    Social class and substance use disordersThe value of social class as distinct from socioeconomic status

    Social Science & Medicine

    (1998)
  • P Armitage et al.

    Statistical methods in medical research

    (1987)
  • L Artazcoz et al.

    Gender inequalities in health among workersThe relation with family demands

    Journal of Epidemiology and Community Health

    (2001)
  • Barnett, R.C. (1998). Women, job stress and illness. In S. L. Sauter et al. (Eds.), Encyclopaedia of occupational...
  • M Bartley

    Measuring women's social positionThe importance of theory

    Journal of Epidemiology and Community Health

    (1999)
  • M Bartley et al.

    Domestic conditions paid employment and women's experience of ill-health

    Sociology of Health and Illness

    (1992)
  • Benach, J., Gimeno, D., & Benavides, F. G. (2002). Employment status and health. Dublin: European foundation for the...
  • Berkman, L.f., & Kawachi, I. (Eds.) (2000). Social epidemiology. New York: Oxford University...
  • D Blane

    The life course, the social gradient and health

  • Borrell, C., Baranda, L., & Rodrı́guez, M. (2002). Manual de l’enquesta de salut de Barcelona 2000. Barcelona, Institut...
  • C Borrell et al.

    Social inequalities in perceived health and utilization of health services in a south European urban area

    International Journal of Health Services

    (2000)
  • H Bosma et al.

    Socioeconomic inequalities in mortality and importance of perceived controlCohort study

    British Medical Journal

    (1999)
  • Carabaña, J., & González, J. (1992). Clases sociales: estudio comparativo de España y la Comunidad de Madrid. Comunidad...
  • B Claussen et al.

    Impact of childhood and adulthood socioeconomic position on cause specific mortalityThe Oslo mortality study

    Journal of Epidemiology and Community Health

    (2003)
  • J.R Dunn et al.

    Identifying social pathways for health inequalitiesThe role of housing

    Annals of the New York Academy of Sciencies

    (1999)
  • M Frankenhaeuser et al.

    Stress on and off the job as related to sex and occupational status in white-collar workers

    Journal of Organizational Behaviour

    (1989)
  • D Gjerdingen et al.

    Women's work roles and their impact on health, well-being, and careerComparisons between the United States, Sweden, and The Netherlands

    Women Health

    (2000)
  • E.M Hall

    Double exposureThe combined impact of the home and work environments on psychosomatic strain in Swedish women and men

    International Journal of Health Services

    (1992)
  • D.W Hosmer et al.

    Applied logistic regression

    (1989)
  • K Hunt et al.

    Just the job? Is the relationship between health and domestic and paid work gender specific?

    Sociology of Health and Illness

    (1993)
  • E.L Idler et al.

    Self-rated health and mortalityA review of twenty-seven community studies

    Journal of Health and Social Behaviour

    (1997)
  • J.V Johnson et al.

    Job strain, work place social support, and cardiovascular diseaseA cross-sectional study of a random sample of the Swedish working population

    American Journal of Public Health

    (1988)
  • G.A Kaplan

    Where do shared pathways lead? Some reflections on a research agenda

    Psychosomatic Medicine

    (1995)
  • Cited by (257)

    View all citing articles on Scopus
    View full text