An article in this month’s journal presents a thorough review of the long-recognized problem of vitamin D deficiency in pregnant women and young children. The authors then present several cases of infants with multiple bony lesions. In the case presentations they imply that these children were suffering from vitamin D deficiency rickets, although the diagnosis of rickets apparently was not made in any of the children. The source of the cases was not mentioned in the article, although I suspect that these may have been cases sent to an expert by attorneys. With the exception of case 5, it also was not stated in the article if the children were diagnosed as having been abused. The problem with such a series of cases is that it might leave the impression that children with metaphyseal lesions and fractures are likely to have vitamin D deficiency rickets. A “convenience sample” can be misleading because it exhibits the logic error embodied by the availability heuristic [1]. Our perception of the frequency of events can be skewed by the examples available to the observer. It is difficult to make generalizations from a series of extreme cases. A person looking down from an airplane at the tops of mountains poking through the clouds who never sees the valleys between them cannot describe the terrain in a meaningful way. The reporting of cases collected from a forensic practice (if this is, in fact, the case) might lead to a biased sample rather than a statistically valid sample. In my practice, a child protection program in a northern climate that evaluates over 1,800 children per year for alleged abuse or neglect, we have been checking every child with multiple fractures for metabolic bone diseases for several years and have not yet identified a single child with vitamin D deficiency. One of my colleagues, however, did find one child, a solely breast-fed 9-month-old with obviously demineralized bones. Since I am not a radiologist, I cannot comment on the radiological interpretation of these cases. However, I would be quite surprised if some of these bony abnormalities were not related to maltreatment. It is unclear whether the authors of the paper are trying to redefine the previously described radiological characteristics of rickets. However, a careful correlation of radiographs and biochemical parameters in infants with proven vitamin D deficiency could be undertaken to examine the issue and address it as a valid research question. On the other hand, the careful clinical/pathological correlations of metaphyseal fractures that have been done by Dr. Paul Kleinman and his colleagues cannot be ignored and should continue to guide our practice until new discoveries are made using valid methodology [2–5]. Every case of multiple fractures or suspected child abuse should be carefully evaluated. The collaboration of pediatric radiologists and pediatricians is an important part of this evaluation. In the field of pediatrics, the development of the new subspecialty, Child Abuse Pediatrics, will set standards for pediatrics experts [6]. Three years of fellowship training (including research training), board examinations, and stringent professional standards for continuing education along with self and peer evaluation will certainly nurture experts in the field and promote excellence in clinical practice. Pediatr Radiol (2008) 38:1219–1220 DOI 10.1007/s00247-008-0995-6