Close window to return

Issue 4, July 1999, pp.4-5.

Psychosocial factors and the explanation of socio-economic gradients in common mental disorder
Stephen Stansfeld, Rebecca Fuhrer, Victoria Cattell, Jane Wardle and Jenny Head

Introduction
Socio-economic inequalities in physical illness, such as coronary heart disease, are well known. There is also increasing evidence of inequalities in common mental disorders including depressive disorders, anxiety disorders and substance abuse.(1,2) However, not all studies find consistent social class gradients in mental ill-health. This may relate to types of common mental disorder having different social class distributions, different characteristics of the populations studied and the tendency for the reporting of psychological symptoms to vary by social class.(3)

Explanations for social inequalities
Research on inequalities in physical illness suggests that health-related behaviours including smoking, an unhealthy diet and lack of exercise explain only part of the class gradient. Thus psychosocial factors, such as social relations and work characteristics, have been invoked by researchers as possible additional explanatory factors for social gradients in physical illness. In particular, low control at work ( decision authority) has been found to explain statistically the gradients in self-reported coronary heart disease among British civil servants in the Whitehall II Study.

It is not clear whether the same psychosocial factors might also explain the gradients in common mental disorder. Preliminary analyses of the Whitehall II study suggest that this may be the case: both control at work and skill discretion (variety of work and opportunity for use of skills) are the most important explanatory variables for the social gradient in depressive symptoms in men(2) (see Table 1).

Table 1:Percentage change in grade difference in depression in men by psychosocial factors,
(Whitehall II Study)

1.The percentage change in grade difference was calculated by including grade as a linear term in a regression analysis and comparing the regression coefficients for grade both before and after adjustment for each psychosocial factor. Both models were adjusted for age.

Describing psychosocial factors
The most obvious candidate mechanism for the operation of psychosocial factors on health is the Stress Hypothesis in which ill-health results from an imbalance of demands over resources. Psychosocial factors may include both stressors (risk factors) and protective factors (resources) in relation to health. Psychosocial factors may be specific to different life domains, (i.e. control at work) or more general (long-term financial difficulties). Stressors include life events (e.g. bereavement, job loss), chronic major difficulties (e.g. long-term financial strain, poor housing, depleted local facilities and resources, fear of neighbourhood crime and violence, fear of racial assault), social isolation, negative interactions in close relationships, high job demands and low job control. Conversely, psycho-social factors may be protective: for example, social support at home or at work, high levels of control over work and skill discretion.

Quantitative analysis can be used to study risk factors, including work and social support, for the whole Whitehall II cohort to explain the employment grade in depression (measured by the Composite International Diagnostic Interview); qualitative research can facilitate an emphasis on the processes involved in the associations between, for example, grade and health and support and health. These relationships are not dissociated from the wider social and historical context in which they take place. The focus on context, which is offered by qualitative analysis, provides an opportunity to look at complexity: at how and under what circumstances stressors and resources interact.

Unresolved issues for psychosocial factors and health
There are a number of issues to be addressed in researching the association between psychosocial factors and health (see Figure 1).

Figure 1: The impact of the physical environment and psychosocial factors on mental ill-health: a schematic representation


Firstly, given that psychosocial factors are embedded in people's material environments, is it possible and helpful to separate the two spheres of influence? Secondly, is it necessary for psychosocial factors (e.g. low control) to be consciously perceived for them to affect mental health. Are psychosocial effects always psychologically mediated? It seems likely that there might be social effects on behaviour and mental health that are not consciously perceived, or at least, not consciously understood.

A third issue concerns the relative importance of exposure in childhood and in adulthood to psychosocial stressors. Both Lundberg(4) and Power(5) provide evidence to suggest that exposure to hardship in childhood may be more important than adult psychosocial exposures in explaining socio-economic differences in mental health. It remains unresolved, however, how much childhood social class, and hence psychosocial stressor exposure in childhood, shapes adult exposure to psychosocial stressors (see article by Amanda Sacker, Health Variations Programme Newsletter, Issue 4, pp.8-10).

Fourthly, there are methodological issues to be addressed. If both stressors and mental health outcomes are measured by self-reported questionnaire, there is a risk that they might be subject to negative affectivity, that is by a tendency to describe both one's exposure to psychosocial stressors and one's mental health, in negative terms. Studies should measure negative affectivity and take it into account in analysis. Another concern is that psychosocial factors, such as control at work, may be confounded with measures of social class, such as employment grade in the Whitehall II Study. This may be the case to some extent, but the same explanatory association, although less strong, is found between control and social class differences in depression when social class is measured by access to a car and housing tenure.(2) Furthermore, it may be incorrect to view control as a confounding factor when it may be a mediating factor between class and health.

Project Aims
Our project (Determinants of Social Class Differences in Mental and Physical Ill-health) aims to shed further light on the role of psychosocial factors in explaining the socio-economic gradient in depression, through two linked investigations based on the Whitehall II Study of civil servants. The first aim of the project is a qualitative study, examining the dynamics between stressors, supports and coping mechanisms in three specific contexts: home, neighbourhood and work; and exploring the nature of these psychosocial factors in a subsample to see if they differ qualitatively by social class and to understand possible mechanisms.

Insights derived from the in-depth interviews will inform the second aim of the study, based on a diagnostic interview measure of depression carried out on the total cohort population of the Whitehall II Study.

Drawing on both the qualitative and quantitative studies, our project will examine how constructs such as control at work or demands may differ by employment grade (as our previous quantitative analyses suggest that they may do ). It seems very likely that control may be interpreted differently according to social class and gender(6) and further understanding of this through qualitative analysis may throw new light on existing quantitative findings. It will also look at differences in supportive resources. Although social networks can provide many benefits, including a protective or direct influence on mental health, these functions may vary according to certain characteristics of the network.(7) Our study will explore structural aspects of an individual's network, and will consider for example, the role of weaker as well as stronger ties in either influencing or moderating inequalities in mental health by grade or gender.

Conclusions
Understanding the role of psychosocial factors and how they differ by socio-economic status may help to focus on which psychosocial interventions are likely to be effective in improving population mental health. Determining whether psychosocial risk factors are the same for both inequalities in mental and physical ill-health will be informative in deciding whether or not separate interventions to improve mental and physical health are needed.

Stephen Stansfeld, Rebecca Fuhrer, Victoria Cattell, Jane Wardle and Jenny Head are based at the Department of Epidemiology and Public Health, Royal Free and University College Medical School, University College London.

References:
1. Lewis, G., Bebbington, P., Brugha, T., Farrell, M., Gill, B., Jenkins, R. and Meltzer, H. (1998) 'Socio-economic status, standard of living, and neurotic disorder' The Lancet, 352 pp.605-609.
2. Stansfeld, S. A., Head, J. and Marmot, M. G. (1998) 'Explaining social class differences in depression and well-being'. Soc. Psychiatry. Psychiatr. Epidemiol., 33 pp.1-9.
3. Stansfeld, S. A. and Marmot, M. G. (1992) 'Social class and minor psychiatric disorder in British Civil Servants: a validated screening survey using the General Health Questionnaire' Psychological Medicine, 22 pp.739-749.
4. Lundberg, O. (1991) 'Causal explanations for class inequality in health - an empirical analysis' Social Science and Medicine, 32 pp.385-393.
5. Power, C. and Manor, O. (1992) 'Explaining social class differences in psychological health among young adults: a longitudinal perspective' Soc. Psychiatry. Psychiatr. Epidemiol., 27 pp.284- 291.
6. Ostergren, P. O., Lindbladh, E., Isacsson, S. O., Odeberg, H. and Svensson, S. E. (1995) 'Social network, social support and the concept of control - a qualitative study concerning the validity of certain stressor measures used in quantitative social within epidemiology'. Scandinavian Journal of Social Medicine, 23 pp.95-102.
7. Cattell, V. and Evans, M. (1999) Neighbourhood images in East London. Social capital and social networks on two East London estates, York : Joseph Rowntree Foundation pp.1-61.