Although household food security (HHFS) has been shown to affect diet, nutrition, and health of adults and also learning in children, no study has examined associations with infant feeding practices (IFP). We studied 1343 infants born between May 2002 and December 2003 in the Maternal and Infant Nutrition Intervention in Matlab study to investigate the effect of HHFS on IFP in rural Bangladesh. We measured HHFS using a previously developed 11-item scale. Cumulative and current infant feeding scales were created from monthly infant feeding data for the age groups of 1-3, 1-6, 1-9, and 1-12 mo based on comparison to infant feeding recommendations. We used lagged, dynamic, and difference longitudinal regression models adjusting for various infant and maternal variables to examine the association between HHFS and changes in IFP, and Cox proportional hazards models to examine the influence of HHFS on the duration of breast-feeding and the time of introduction of complementary foods. Better HHFS status was associated with poor IFP during 3-6 mo but was associated with better IFP during 6-9 and 9-12 mo of age. Although better HHFS was not associated with the time of introduction of complementary foods, it was associated with the type of complementary foods given to the infants. Intervention programs to support proper IFP should target mothers in food-secure households when their babies are 3-6 mo old and also mothers in food-insecure households during the 2nd half of infancy. Our results provide strong evidence that HHFS influences IFP in rural Bangladesh.