Mothers with infants are one of the fastest growing segments of the labor force. National estimates reveal that about half of all American women with infants were in the labor force in 2002 (Downs, 2003). While it is well established that women face physical and psychological changes in pregnancy, childbirth and the postpartum, relatively little evidence exists to document the effects of employment and work organization characteristics on women's health at periods of heightened vulnerability such as after childbirth. With many new mothers returning to work relatively soon after childbirth, this study was designed to examine the time course for women's recovery from pregnancy and childbirth and identify factors that facilitate the merger of working and parenting roles during the first 18 months after childbirth. This was a prospective longitudinal study of employed mothers designed to identify the effects of personal and employment characteristics, job stress and role conflicts on maternal postpartum health. We enrolled 817 women (71 % response rate) at the time of childbirth from selected Minnesota hospitals in 2001 who met sample selection criteria including: giving birth to a singleton, healthy infant; English-speaking; 18 years of age or older, employed during pregnancy, and planning to return to work after childbirth. At enrollment, demographic and medical information was abstracted from mothers' charts and eligible women were interviewed. Women were again interviewed at approximately six weeks (N=716, 88% of enrollees), three months (N=661; 81 %), six months (N=625; 76%), twelve months (N=575; 71 %) and eighteen months (N=559; 68%) after delivery and asked about demographic and employment characteristics, health status, health services and family leave benefits used and work-family conflict. Findings of initial multivariate analyses estimating maternal health at approximately six weeks after delivery, when 7% of women were back at work, revealed that, on average, these postpartum women scored slightly worse on general physical health (Physical Component Summary or PCS score of 51.4) than the population norm for non-postpartum women (PCS: 52.7), but slightly better on general mental health (Mental Component Summary or MCS score of 49.4) relative to the population norm (MCS: 47.2). Physical and mental health was measured with the SF-12 v-2 created by Ware and associates (2002). However, these mothers continue to experience an average of six childbirth-related symptoms (e.g., fatigue, breast symptoms, sexual symptoms, hemorrhoids, constipation, and excessive sweating) indicating a need for rest and recovery beyond the traditional medical perspective of a postpartum period of four to six weeks. Additionally, approximately 6% of the sample had symptoms of postpartum depression as measured with the Edinburgh Postnatal Depression Scale (Cox, Holden, Sagovsky, 1987). By approximately twelve weeks after delivery, 48% of these mothers had returned to work. On average, these mothers still experienced an average of 4 childbirth-related symptoms, and 5% of the sample had symptoms of postpartum depression. Multivariate models were also estimated to assess the effects of work-family conflict on women's postpartum health. For example, a measure of "job spillover" evaluated the extent to which employment obligations interfered with home responsibilities or time spent with family, while "home spillover" evaluated the extent to which home life interfered with work responsibilities and time ideally spent on the job. Study findings revealed that both job spillover to home and home spillover to job had a statistically significant, negative impact on maternal mental health outcomes, but little impact on their physical health. Co-worker support resulted in a strong, positive impact on maternal physical health when compared to women who reported little or no coworker support. Multivariate models were also used to estimate the effects of work-related stresses and strain on maternal postpartum depression at twelve week after delivery. Higher depression scores were associated with being non-white, having higher job stress, lower job flexibility, poorer mental health at six weeks postpartum and more prenatal mood problems. Measures of psychological job demands and decision latitude at work were nonsignficant in the model. By approximately six months after delivery, women's return to work status and health status appears to have stabilized with average physical health (PCS) scores ranged from 55.1 at six months to 54.5 at eighteen months postpartum, and average mental health (MCS) scores ranged from 50.4 at six months to 51.4 at eighteen months postpartum. The proportion of women with symptoms of postpartum depression ranged between 4% and 6% during the same period. Longitudinal analyses were conducted using data from all interview time periods-six weeks, twelve weeks, six months, twelve months and eighteen months after childbirth using a subsample of 668 women. The preliminary findings revealed factors that were significantly associated with maternal health, on average, across all five time periods. Factors associated with poorer postpartum physical health included being poor, having lower levels of preconception general health, having had a cesarean delivery, being on leave (vs. working), and having lower levels of perceived control over home and work activities. Factors associated with poorer postpartum mental health included having lower levels of preconception general health, prenatal mood problems of depression and anxiety, higher total workload (hours of paid and unpaid work), more perceived job stress, lower levels of perceived control over home and work activities, less available social support, less coworker support, a baby with colic or an irritable temperament, a baby with sleep problems, and a male infant. Psychological job demands and decision latitude were nonsignificant in the model. Additional analyses are in progress to evaluate the potential endogeneity in the model. The primary limitation of the study is that the findings may only be generalized to women of the same demographic and delivery characteristics. The study population included mothers 18 years and older, who were, on average, 29 years (SD: 5.7 years). These Minnesota mothers were less likely to be married (77%) than new mothers, ages 25-29 years, nationwide (83%) (Bachu, O'Connell, 2001), and were less likely to deliver by cesarean-section relative to national data (17% vs. 24%) (Martin, Hamilton, Ventura et al, 2002). However, these Minnesota mothers were more likely to have completed high school (94%) relative to national data on new mothers, ages 25 to 34 years (84%-90%) (NCHS, 2005). The proportion of Caucasian and Native American mothers in the study population was comparable to national data (78% and 1 %, respectively), but there were relatively fewer African American mothers (9% vs. 15%) and more Asian mothers (11 % vs. 5%) than national data (NCHS, 2005). Study findings suggest the importance of health care providers evaluating women during the pregnancy and the postpartum to discuss strategies to promote health, expected childbirth-related symptoms and duration of disability leave from work in relation to delivery type to promote maternal well-being. Postpartum evaluations should include an assessment of fatigue, the most prevalent postpartum symptom, as well as other symptoms that may indicate a need for time for rest and recovery. Health care providers should be particularly alert to symptoms related to depression and anxiety to identify women in need of referral to mental health specialists. Primary care and occupational health providers should discuss with expectant mothers their plans to return to work and possible need for medical certification under the Family and Medical Leave Act, including intermittent leave. Leave from work provides the time for rest and recovery from pregnancy, labor and delivery, and intermittent leave allows mothers the opportunity to return to work on a gradual, part-time basis. Women meeting state poverty guidelines may need particular assistance from health care providers or human resource personnel to identify financial and other resources, because the leave benefits are unpaid under both the federal Family and Medical Leave Act and the state's parental leave law. Findings also suggest a role for primary care and occupational health providers to discuss with women the potential for adverse health effects associated with high total workloads (hours of paid and unpaid work) and high levels of job stress, and to explore the flexibility of women's work arrangements to identify potentially modifiable factors that could decrease stress and workload and enhance women's sense of control. Providers should also help women to identify and ask for social support from relatives, friends and coworkers to relieve the burden women may feel trying to balance work roles and family obligations. Study findings were inconsistent with Karasek and Theorell' s (1990) job demand and control model raising questions about its application to postpartum women. Research is needed to further investigate modification of the job demand and control model for this population and to identify the determinants of job stress and workload, and the nature of flexible work arrangements, that could serve as the basis for intervention studies.