Four hundred seven healthy, full-term infants were divided into three groups and fed, respectively, a formula of evaporated milk and water with 5% carbohydrate; human milk; and a special modified evaporated milk designed to simulate human milk. No other foods were added to the diet. A comparison of the three groups was made, based on weight gains from birth to the end of the first 4 weeks. The conclusions refer only to weight gains; no attempt was made to determine the superiority of any particular diet.
The data show no significant differences in the 4-week weight gains among the three groups of infants, although sensitive statistical methods could be validly applied to the problem. These methods are explained.
The authors wish to emphasize the many variables which must be taken into account in a study of this sort; the necessity for careful selection of valid statistical methods; the importance of critical clinical judgement in the evaluation of the results.
Reply to Nozawa et al.: Complementary statistical methods support positive selection of a duplicated UV opsin gene in Heliconius