The contribution of working conditions and social support to the onset of depressive and anxiety disorders among male and female employees

Abstract

Poor working conditions may be an important source of stress and may therefore contribute to the development of depressive and anxiety disorders. Social support may act as a buffer and protect against the development of depression or anxiety in the face of poor working conditions. With longitudinal data from the Netherlands Mental Health Survey and Incidence Study (NEMESIS), the effect of working conditions and social support on the incidence of depressive and anxiety disorders was examined among 2646 working men and women, aged 18 through 65 years. Three dimensions of self reported working conditions were assessed: psychological demands, decision latitude and job security. Social support was assessed through validated scales for daily emotional support. About 10.5% of working women and 4.6% among working men developed an incident depressive and/or anxiety disorder over 2 years. Psychological demands predicted the incidence of depressive and anxiety disorders in both men and women (RR per score increase=2.29, 95% CI: 1.44–3.63), whereas decision latitude and job security did not. Social support protected against the incidence of depressive and anxiety disorders. This effect was stronger for men compared to women. Social support did not buffer the unfavorable mental effect of working conditions. Women were more likely to report low levels of decision latitude, whereas men reported higher psychological demands. Working conditions did not explain sex differences in the incidence of depressive and anxiety disorders.

Introduction

Depressive and anxiety disorders are both common and important in the general population (Bijl, Ravelli, & van Zessen, 1998; Ohayon & Schatzberg, 2002; Wittchen, Essau, von Zerssen, Krieg, & Zaudig, 1992). Working conditions and social support represent two domains of environmental factors that may be important to understand pathways to anxiety and depression (Garnefski, van Egmond, & Straatman, 1990; Graaf, Bijl, Ravelli, Smit, & Vollebergh, 2002; Murphy, Olivier, Monson, Sobol, Federman, & Leighton, 1991; Piccinelli & Wilkinson, 2000; Wilhelm, Kovess, Rios-Seidel, & Finch, 2004).

Working can be disadvantageous as well as beneficial for mental health depending, among other things, on the quality of working conditions (Harnois &Gabriel, 2000). The relationship between working conditions and job stress has been confirmed in cross-sectional studies (Karasek & Theorell, 1990; Pugliesi, 1999; Spielberger, Vagg, & Wasala 2003). One important aspect of working conditions is the security of the worker's position: uncertainty about the employment position causes stress (Siegrist, 1996). According to the Demand/Control Model (Karasek & Theorell, 1990), the psychological demands imposed by the job (work load) and the degree to which the worker may exert decision latitude are two other important aspects of working conditions, which are related to job stress. The decision latitude one may exert depends on the level of autonomy and the professional skills of the worker. According to the model, rising psychological demands may cause job stress. Higher degrees of decision latitude may reduce job stress and may buffer the effect of psychological demands. In several cross-sectional studies support was found for the association of psychological demands and decision latitude with anxiety and depressive symptoms (Demerouti, Bakker, de Jonge, Janssen, & Schaufeli, 2001; de Jonge & Kompier, 1997; Sanne, Mykletun, Dahl, Moen, & Tell, 2005; Stansfeld, Fuhrer, Shipley, & Marmot, 1999).

Despite the fact that various studies have demonstrated an association between poor working conditions and poor mental health, Wilhelm et al. (2004) concluded in their literature review that, due to a lack of longitudinal research, the direction of the relationship between working conditions and depressive and anxiety disorders remains unclear. Poor working conditions can result in mental health problems, but mental health problems may also contribute to poor working conditions. In addition, the assessment of mental health in most previous studies has been rather broad with a lack of discrimination between depressive and anxiety symptoms. The present study deals with these prior limitations: we use longitudinal data of a large representative cohort of working persons, and will assess the incidence of depressive and anxiety disorders among persons who start out without depressive and anxiety disorders. Our first hypothesis (‘Working conditions Hypothesis’) is that poor working conditions (low job security, low decision latitude, and high psychological demands) increase the risk of 2-year incident depressive and anxiety disorders.

Besides the knowledge of the impact of working conditions on mental health, the awareness of the importance of support from the social network is growing (Baum, 1999; McKenzie, Whitley, & Weich, 2002). A good, stable social network that provides social support can positively influence mental health (Brown & Gary, 1987; Kendler, Myers, & Prescott, 2005). In addition to a direct effect, social support may, according to the buffering hypothesis, be especially effective in stressful circumstances (Greenblatt, Becerra, & Serafetinides, 1982; Olstad, Sexton, & Sogaard, 2001), and could theoretically reduce the negative effect of poor working conditions. The buffering effect of social support from co-workers on the relationship between work stressors and mental health was not supported in studies by Loscocco and Spitze (1990) and Sanne et al. (2005). However, it could be that social support from intimate others may be a more important source of social support that provides buffering. In a prospective population study, Olstad et al. (2001) indeed found a buffering effect of overall social support on the relationship between work stress and a general indicator of mental distress. Whether this protective effect is also strong enough to reduce the incidence of mental disorders remains to be demonstrated. The present study will examine the buffering effect of social support, and hypothesizes that social support buffers the negative effect of poor working conditions on incident depressive and anxiety disorders (‘buffering effect hypothesis’).

A consistent finding in previous research is that the incidence of depressive and anxiety disorders in women is approximately twice as high as among men (Alonso et al., 2004; Bebbington, 1998; Bijl, Ravelli, & van Zessen, 1998; de Graaf et al., 2002; Weich, Sloggett, & Lewis, 1998). In general, men report better working conditions than women (Blidt & Michélsen, 2002; Pugliesi, 1999; Roxburgh, 1996). Research also shows that gender differences in social support exist: women provide and receive more social support than men receive and they appear to be more sensitive to the effect of social support (Beehr, Farmer, Glazer, Gudanowski, & Nair, 2003; Pugliesi & Shook, 1998; Reevy & Maslach, 2001; Walen & Lachman, 2000). It has been suggested that because women not only receive, but also give more social support than men do, they may be more negatively affected by emotional strain of network members (Fuhrer et al., 1999; Stansfeld et al., 1998; Walen & Lachman, 2000). And in line with this suggestion, in some studies women even appeared to profit less from social support than men (Stansfeld et al., 1998; Schwarzer & Gutiérrez-Dona, 2005). Therefore, because women may have more unfavorable working conditions and may profit less from social support than men do, the female preponderance of depressive and anxiety disorders could be partly due to the social circumstances. Consequently, the final hypothesis of the present study is that working conditions and social support contribute to the explanation of gender differences in the incidence of Depressive and Anxiety Disorders (‘Gender differences Hypothesis’).

Using 2-year longitudinal data of the Netherlands Mental Health Survey and Incidence Study (NEMESIS), we will explore whether work characteristics and social support contribute to the onset of depressive and anxiety disorders in a large general population of working men and women without baseline depressive and anxiety disorders.