Latent class analyses can be used early in the postpartum period to identify mothers of preterm infants experiencing similar patterns of psychological distress symptoms, but whether these classes of mothers also differ in parental responses to their infants or in their infants' development is largely unknown. In this longitudinal multisite-repeated measures study, we evaluated the usefulness of three psychological distress classes (low distress, high depressive and anxiety symptoms, and extreme distress) in predicting mother–infant interactions, quality of home environment, and infant development in 229 mother-preterm infant pairs. Mothers completed psychological distress questionnaires at study entry; parent-infant interaction was recorded at 2 and 6 months of age corrected for prematurity; and infant developmental data were collected 12 months corrected age. Mothers in the extreme distress class engaged in more developmental stimulation at 2 months (β = .99, p < 0.01) and at 6 months (β = 1.38, p < .01) than mothers in the other classes and had better quality of home environment at 2 months (β = 2.52, p = .03). When not controlling for neurological insult, infants of mothers in the extreme distress class had poorer cognitive (β = −10.28, p = .01) and motor (β = −15.12, p < .01) development scores at 12 months corrected age than infants of mothers in the other distress classes, but after controlling for infant neurological insult, there were no differences in cognitive, motor, and language development based on maternal psychological distress class. © 2016 Wiley Periodicals, Inc.