The article by McCrindle et al1 in the February 1996 issue of the Archives is a valuable contribution to efforts to make the referrals of infants and children to pediatric cardiologists for evaluation of murmurs more rational and efficient. I was disappointed that the authors made only passing reference to the key issue of how to improve our education of non-cardiologists in auscultation of the heart, so they may become more confident in recognizing innocent murmurs.
I was interested, also, to observe that the authors did not discuss 1 clinical feature from the history that had a P value of less than. 001, namely, the "age at encounter." Clearly, if a murmur is heard for the first time after the age of 1 year, the statistical chances of that murmur's arising from a cardiac lesion are low. Experienced clinicians have recognized this since the early days of pediatric cardiology; emphasizing