BACKGROUND The impact of maternal employment on preventive child health practices has not been studied empirically. Using a household production model, we investigated the relationship between level of maternal employment and child immunization status, use of automobile seat belts, and use of bicycle helmets. METHODS Data from a longitudinal study of public school children in Pinellas County, Florida, were used to draw an analytic sample of 4,153 African American and white non-Hispanic children whose mothers responded to school surveys in 1989, 1991, and 1993. Analyses of the relationship between maternal hours worked per week and preventive child health practices were performed using multiple logistic regression procedures, controlling for maternal age and education, household income, ethnicity and gender of child, and number of siblings living at home. RESULTS No significant statistical relationships were found between maternal employment and child immunization status or use of automobile seat belts. However, maternal employment was significantly associated with bicycle helmet use, after controlling for confounders. Children of mothers who worked less than 21 h per week were 37% more likely to wear helmets compared with children of mothers who worked 21 h or more per week. CONCLUSIONS Findings suggest that children of full-time working mothers are not at greater risk for under-immunization or failure to use seat belts regularly, but they may be less likely to use bicycle helmets. These results support the hypothesis that employment does not affect episodic child health practices but may have some negative impact on preventive practices involving daily, repetitive activities. It also suggests that the effects of maternal employment may be greater on preventive practices that are nonnormative and occur in the absence of adult supervision.