Background: A vitamin B(1)-deficient soy-based infant formula was marketed in Israel in 2003, exposing infants to clinical or subclinical B(1) deficiency. We investigated whether subclinical B(1) deficiency in early infancy had medical, neurodevelopmental, or cognitive effects at 3-5 y of age.
Methods: A historical prospective cohort study was conducted consisting of four groups: "exposed," consuming a B(1)-deficient soy-based formula exclusively for four consecutive weeks or longer; "control," consuming no soy-based formula; "mixed," consuming the formula nonexclusively or exclusively for less than four consecutive weeks; and "other," consuming soy-based formulas other than Remedia. Participants were evaluated by medical examination, Stanford-Binet (SB) intelligence test, sensory profile evaluation, and Conners scales (attention deficit disorder/attention deficit and hyperactivity disorder (ADD/ADHD)).
Results: Following adjustment for gender, age, and maternal education, there were no significant differences among the four groups on the mean SB scores, on the verbal and nonverbal scores, or in the proportion of children in each group with scores <90. A significantly higher proportion of exposed children as compared with control children had an impaired sensory profile and scores on the Conners scales (ADD/ADHD), but these proportions were also high in the "other" and "mixed" groups.
Conclusion: The results do not support an association between subclinical B(1) deficiency in infancy and long-term development.