In this edition of Paediatics & Child Health, Drs Leung and Sauve provide a compelling argument to change the present Canadian Paediatric Society’s recommendations for the use of iron-fortified formula in nonbreastfed infants from nine to 12 months of age to the American recommendation of 12 months (1,2). At first glance this seems to be a ‘no brainer’. Iron deficiency anemia is an important nutritional deficiency in both Canadian and American children (3–8). Iron deficiency associated with the use of whole cow’s milk in infants derives from a combination of poor dietary availability and the presence of occult blood loss in infants under six months of age (3,4). Early iron deficiency may result in lower developmental and behavioural scores in children later in early childhood (9,10). On closer scrutiny, the case for a change may be less secure. Iron deficiency is essentially a condition that reflects a poverty of body iron stores. This may result from the impoverished nutritional endowment associated with prematurity. Alternatively, iron deficiency directly reflects the economic poverty of families. Native infants and children from poor families account for the largest proportion of anemic children detected at one year of age (6). Infants from Canadian families of oriental background are also at risk (7). In most instances, anemia is due to the use of whole cow’s milk from early infancy in spite of the current recommendation that counsels against this practice. It is unlikely that a recommendation for extension in the use of infant formula to one year of age would in any way impact the early introduction of ‘store milk’ in the population at greatest risk of this problem. In many instances, parents use ‘store milk’ because they cannot afford to purchase more expensive formula. In other words, a recommendation to extend the use of formula is most likely to be adopted by a population of parents least likely to have affected children and least likely to be adopted in circumstances most likely to result in anemia. This phenomenon is not unique to iron deficiency anemia and is common to many problems in nutrition across a population. There are two rebuttals to this logic. The first is that, by adopting the recommendation for a longer period of formula feeding, Americans have lowered the prevalence of iron deficiency. The second is that, in the absence of any additional risk, extending the use of formula may help some children. However, it is most likely that the lower incidence of iron deficiency in the United States has resulted from the Special Supplemental Nutrition Program for Women, Infants and Children that provides nutritional supplements to pregnant women and their infants from lower socioeconomic groups (11). In other words, the more effective part of the policy was the targeting of poor families as the population at risk of iron deficiency owing to the early introduction of whole cow’s milk. Canadian health authorities have considered this approach but have shied away from providing free or subsidized formula to poor families because they fear being criticized for undermining breastfeeding. The second hypothesis is similarly skewed. A recommendation to extend the use of formula from the present nine to 12 months of age to 12 months of age would require families to purchase formula rather than milk for nonbreastfed infants. This would affect close to 350,000 families purchasing 1 L per day of formula at a premium of approximately $2 per day for an additional 90 days (12). Families with young children have many competing priorities for the $63,000,000 in increased expenditure. By purchasing store milk rather than formula, a family might benefit more by investing the $180 in additional vaccine coverage or a registered education savings plan for their child, especially if there is no evidence that the extra expense would provide additional health benefits. If whole cow’s milk is introduced at nine months of age with appropriate meats or iron-rich foods, the risk of iron deficiency is lessened. The risk is further decreased by limiting milk and juice intake to optimize the intake of these other foods. These phenomena are characteristic of many issues in population nutrition. It is tempting, but often ineffective, to use broad stroke national policy to deal with issues such as vitamin D deficiency or iron deficiency in a country as diverse as Canada. In these matters, we should ask, whether we would be better to invest limited funds in populations at risk, rather than adopting a recommendation that may be least likely followed in the sector where the problem exists. If we are going to change course, why not steer toward the cause of anemia and invest more in nurturing the mothers of infants vulnerable to this problem?