Of 220 children referred to the Dept of Paediatrics, Royal Children’s Hospital, Parkville, Victoria, Australia, because of suspected hyperactivity, 55 were included in a 6-week open trial of the Feingold diet, 26 (47%) showed a placebo response, and 14 were identified as likely reactors. Of 8 who subsequently completed a double-blind crossover study (utilizing each child as his own control), 2 demonstrated a significant dependent relationship between the challenge and ingestion of azo dye colorings (tartrazine and carmoisine 50 mg) and behavioral change. Extreme irritability, restlessness and sleep disturbance rather than attention deficit were the common behavioral patterns associated with the ingestion of food colorings, as described by the parents in this study. The authors conclude that the inclusion of children in trials on the basis of attention deficit alone may miss some reactors, and there is little place for use of a coloring-free diet in children with ADD unless the other behavioral features of irritability, restlessness and sleep disturbance are present. 
COMMENT. The phoenix of the Feingold diet rises again with the suggestion that the treatment has been erroneously discarded because of inappropriate behavioral rating instruments and failure to identify specific reactors to food additives. In England, where the avoidance of all foods containing additives is widespread, the major problem is the level of public misinformation, occasionally leading to handicapping dietary restriction.