Background. The impression that there is an excess of specialist physicians may be inappropriately applied to the pediatric endocrinology workforce. The American Board of Pediatrics has emphasized a need for more data about pediatric specialist requirements and concern about non pediatric-trained-specialists providing care to children. Objective. To describe the clinical activity of pediatric endocrinologists in university and private practice settings and the pediatric endocrinology activity of other physicians in private practice. Methods. Data were collected from surveys sent to US and Canadian members of the Lawson Wilkins Pediatric Endocrine Society (LWPES) reporting type of practice, numbers of M.D.s and extenders, hours of activity, and types of patients seen in 1994. One hundred eighteen unduplicated forms described the work of 393 physicians, 83% university associated, 65% of the estimated universe of 600 practicing pediatric endocrinologists. In addition, National Disease and Therapeutic Index (NDTI) data for 1993 were analyzed for endocrine diagnostic visits in the 0-21 year age group to general physicians, internists, and pediatricians in private practice. These data reflected 1.7 million visits for diabetes, thyroid disorders, glucose and calcium disorders, pituitary, adrenal, and sexual disorders, hyperlipidemia, and obesity. In addition to separate analyses of the LWPES survey and NDTI data, comparisons were made where possible between them (diabetes, thyroid disorders, and total endocrine visits). Results. Private M.D.s saw nearly twice the number of patients per week (mean 43, median 40) as academic M.D.s (mean 26, median 22), with similar numbers of patients per physician hour in clinic (both 1.6 +/- 0.80). The percentage distribution of diagnosis visits was virtually identical for university practice and private practice (diabetes 39/37; growth both 27; thyroid both 13; sex disorder 10/11; adrenal 7/5; and other 8/7). Half of private physician visits estimated by NDTI for endocrine diagnoses in this age group were to internists, a category including adult and pediatric endocrinologists in private practice, and one-third to pediatricians. Projection from the estimated sampling percentage for the LWPES survey indicated that pediatric endocrinologists provide 750,000 patient visits peryear, or 35%; of all such visits, and that these are distributed 45% to private and 55% to academic based pediatric endocrinologists. Private physicians ac count for 2300 visits per full time equivalent (FTE) physician per year, nearly twice that of university practice (1300 per year). Internists see a larger portion of diabetes patients (37%) than do pediatric endocrinologists (28%) and a similar proportion of thyroid patients (26% vs. 25%). Internists saw few patients in this age group for endocrine problems other than diabetes or thyroid. Conclusions. It is estimated that maintaining current activity levels requires one clinical FTE pediatric endocrinologist per 900,000 population, a total of 280 FTEs, close to the endocrine/diabetes clinical commitment of the estimated 570 practicing pediatric endocrinologists in the US. If all pediatric endocrine and diabetes diagosis and management, however, were carried out by subspecialists in this field, a substantial in crease in training resources would be needed; currently only about 30 trainees complete pediatric endocrinology fellowships annually, comparable with the attrition rate.