The efficacy of two contrasting short term preventative interventions administered to a heterogeneous sample of new mothers during the perinatal period was evaluated. The first was infant-centered and used the Brazelton Neonatal Behavioral Assessment Scale (NBAS) as a method of highlighting newborn behavior to new mothers. The second was mother-centered and consisted of an in-depth interview focused on the mothers' concerns about parenting. Mothers were randomized into groups and were administered intervention by experienced clinicians at 3 days in the hospital and again at 14 and 30 days at home. Effects of intervention on maternal reports of parenting stress, mother-infant interactive behavior, and infant developmental quotient were evaluated at 4 months infant age. It was hypothesized that participation in the infant-centered intervention would be related to more positive maternal and infant outcomes at 4 months. It was also expected that the impact of each intervention would be moderated by differences in maternal and infant risk and parity. Hierarchical multiple regressions controlling for risk and parity yielded no significant main effects of intervention at 4 months. However, significant interactions of intervention with parity and risk were observed, indicating that intervention was beneficial for specific groups of mothers. These data suggest that early intervention should be tailored to the needs of individual groups of mothers.