This study examined the relationship between employment status and mental and physical health in pregnancy and identified variations in this relationship that were associated with parity and single parenthood. The sample of 12,169 women was drawn from the Avon Longitudinal Study of Pregnancy and Childhood (ALSPAC). Employment status was measured at 18 and 32 weeks gestation as were the two main health outcomes: depression (using the Edinburgh Postnatal Depression Scale) and self report of nausea. A woman's parenthood status, socio-economic status, age, educational level, ethnic group membership and smoking behaviour were also recorded. Separate analyses were carried out for primiparous and for multiparous women. Within these parity groups, 2 tests were used to determine the relationship between employment status and the social, demographic and behavioural variables outlined above. Logistic regression analyses established how employment and parenthood status were associated with the health of each group of women, after controlling for potential confounding factors. These analyses showed that for depression, primiparous women were at reduced risk if they worked in pregnancy when compared with women who were not working; primiparous women who gave up work in early pregnancy were most likely to be depressed. Single parents in each employment status category were more likely to be depressed than women with partners. For nausea, primiparous women with partners who gave up work early in pregnancy were more likely to report this condition; for single parents the reverse was the case. There was no significant association between employment and either mental or physical health for multiparous women. Results were discussed in terms of whether they demonstrated the effect of work itself on pregnancy, or whether they were the consequence of health selection in and out of the labour market.