Traditionally, research concerned with the relationship between paid employment and women's experience of ill-health has focused on one or other of two competing explanatory models: the role enhancement model which postulates a positive health effect of paid employment and the role overload model which postulates a negative effect. Though a synthesis of these two models is becoming apparent in the literature (Arber 1990, 1991) a number of important issues remain neglected. This paper focuses on three of these neglected issues: the extent to which the relationship between employment and health is mediated by the nature and/or extent of the workload associated with both formal and domestic work; whether the relationship differs for different types of ill-health; and the role of health related selection into different social roles.
Using data from the Health and Lifestyles survey, an innovative approach is taken to the measurement of domestic conditions, and separate measures of long-term health status, and of short-term physical and psychological health state are used. Proportional odds models are used to assess the relationship between employment status, domestic conditions and type of employment (measured in terms of the socio-economic group of a woman's own occupation) independent of underlying long-term health status. Our findings suggest that the presence of long-term illness is the single most powerful influence on present health state, but that this does not explain away the positive relationship between employment and women's health. They also suggest that this relationship is different for different types of ill health and is not the same for women in different types of occupation. In particular, our findings suggest that the association between paid work and better health is rather less apparent for physical than psychological health, and in women working full time in professional or managerial occupations. Domestic conditions appear to have an effect on women's health equal to or greater than employment status, depending on the health measure used. However, there was no evidence of an interaction between employment status and domestic conditions.
We conclude with a discussion of the ways in which our findings fit in with those of previous research. We argue in particular for more attention to be given to the nature of both the dependent (‘health’) and explanatory (‘work’) variables used in research on the relationship between employment and women's health.