In pediatric practice, perhaps the most frequent response to a problem with infant feeding or a gastrointestinal irregularity is to change the baby's formula. Because there are limited data available on the frequency of formula changing and the reasons for it, we interviewed 100 parents in our pediatric outpatient clinic during April 1995. Their infants were 30-210 days old (mean 115) and had been formula fed since birth. We excluded those born at <36 weeks gestation. Ninety-seven were fed regular formula (RF) from birth and 3 soy formula. 81% were white, 12% black, 6% middle eastern, 1 east Asian; 55% were male. Parents were asked about the baby's initial formula, changes in feeding, age at change, reason for change, who directed the change, and outcome. Of the 97 infants receiving regular formula from birth, 37 (38%) had been changed to another formula - 26 to soy, 5 to lactose free, 3 to casein hydrolysate and 3 to low iron. Six (16%) were later switched back to RF. Spitting up (17/97) and colic (13/97) were the main reasons for switching formulas. Other reasons included diarrhea (4), bloody stools (3), constipation, wheezing, mother's preference and a family history of allergy. In 20/37 (54%) the decision to change formula was made by the mother and in only 13 (35%) by the pediatrician. Other decisions were made by nurses, an ER physician and a WIC employee. Following the formula change, mothers reported minimal or no improvement in 33% of infants while considerable improvement or complete resolution of symptoms was seen in 67%. Although definite or“probable” adverse reactions to cows milk actually occur in only 5% of infants (Bock, Pediatrics 1987;79:683) formulas were changed in 38% of this cohort. The fact that these changes were most often initiated by the mother suggests the need for better anticipatory guidance. On the other hand, pediatricians initiated change in about 1/3 suggesting that this practice should also be reviewed. What percentage of feeding problems was the result of true cow's milk intolerance is unknown, but inappropriate labeling of infants as “allergic” to milk products and a legacy of soy fed infants who should drink regular formula cannot be considered benign.