Sir.—Myers et al1 concluded in a published study that formula-fed infants have more gastrointestinal tract illnesses than do breast-fed infants. Although the authors noted that bottle-feeding was associated with a greater likelihood of exposure to other infants in day-care (DC) settings, there was no attempt to deal with this confounding variable. Further analysis of the data reveals an equally plausible conclusion that the apparent adverse effects of bottle-feeding are due to other confounding variables, specifically, exposure to other infants in DC settings.
Our Table is derived from the data in Tables 1 and 3 of Myers et al.1 This was feasible only by ignoring the matching process and pooling data on patients by DC use and breast- or bottle-feeding. The Table shows the number of
Illness in Infants Classified by Day-care Center Attendance and Feeding Mode Feeding Mode No. of Infants No. of Illnesses by Type* Resp