A randomized, double-blind study was conducted to determine whether continued feeding of premature infant formula after hospital discharge improve biochemical measures of bone mineral or protein status and anthropometrics during the first 8 and 12 weeks, respectively, after initial hospital discharge. Forty-three subjects were randomized to receive either a 20 kcal/ounce standard cow's milk-based formula with iron or a 20 kcal/ounce premature infant formula with iron for 8 weeks after hospital discharge. Sixteen exclusively breast-fed infants (mother's own milk) who received a multivitamin supplement with iron were compared with infants in both formula groups. There were no differences among the three groups in gender, birth weight, gestational age, or weight and age at the time of study entry. Alkaline phosphatase values were lower in infants receiving premature infant formula than in those receiving standard formula 8 weeks after discharge. Phosphorus values were lower and alkaline phosphatase values higher in the human milk-fed group than in both formula groups 8 weeks after discharge despite supplementation with calcium, phosphorus, and vitamin D before and during the study. At 8 weeks after discharge, human milk-fed infants also had lower transferrin levels than infants fed formulas. Infants in both formula groups grew similarly in weight, whereas the infants fed human milk weighed less throughout the study. The group fed premature infant formula had greater mean length and head circumference than the standard formula or human milk-fed groups. These data indicate that premature infants weighing < 1800 gm at birth may benefit from the continuation of premature infant formula during the first 8 weeks after initial hospital discharge.