OBJECTIVE To determine if there is a relationship between center-based child care attendance from ages 3 to 5 y and future overweight at ages 6-12 y. METHODS Longitudinal, observational study of child experience and future body mass index. METHODS A total of 1244 US children aged 6-12 y included in the 1997 Panel Study of Income Dynamics Child Development Supplement. METHODS Parent-reported child care attendance from ages 3 to 5 y, trichotomized as 'none', 'limited' (>0 but <15 h/week), and 'extensive' (> or =15 h/week). Overweight defined as a body mass index > or =95th percentile for age and gender. Candidate covariates (selected a priori): gender, race, age, poverty status, birth weight, hours of television per day, Behavior Problems Index score >90th percentile, and Home Observation for Measurement of the Environment-Short Form (HOME-SF) cognitive stimulation score. RESULTS Of the potential confounding variables, race, HOME-SF cognitive stimulation score, and age significantly altered the relationship between child care attendance and overweight in the multiple logistic regression model. With these covariates in the final model, limited center-based child care attendance from ages 3 to 5 y was independently associated with a decreased risk of overweight at ages 6-12 y (adjusted odds ratio=0.56, 95% confidence interval 0.34, 0.93) relative to no child care attendance. Extensive center-based child care attendance was not associated with future overweight. CONCLUSIONS Limited center-based child care attendance during the preschool years was independently associated with a decreased risk of future overweight relative to no child care attendance. Additional studies are needed to clarify these findings.